Hello all, I wanted to bring a few things up and I wanted to bring them up on devel@lists.fp.o because this is where most people spend their time.
First off: "Does letting thousands of contributors do what they want have a negative impact on our OS? (Mike)"[0] - I would prefer that this be rephrased to a quote I read that originated from John Rose (inode0) "isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" and I would prefer a focus be turned towards something like "Why was that the result of doing something that is essentially chaotic?" .... I guess my main question is: "Why are we fixing something that isn't broken?"
Second: "The Board has been working on defining a target audience for Fedora. In response to this, some people feel that Fedora should allow sub-groups to define their own target audience"[1] - I don't entirely understand this, don't SIGs or (sub groups) essentially exist purely because there is some target audience? Clarification on this not would be appreciated.
Now, we come to the part that I feel is going to be viewed as a touchy subject by many..... Why are there words like "letting" and "allow" being thrown around so often? I understand there are guidelines and policies for certain things of technical or legal nature in Fedora, but it feels a little like there is an attempt here to dictate how myself, as well as all others, spend their time contributing to The Fedora Project.
[0] https://fedoraproject.org/wiki/Unfinished_Board_issues [1] https://fedoraproject.org/wiki/User:Walters/SpinsSigsRemixes_TargetAudience
I would just like to know other contributors thoughts on these topics.
Thank you for your time, -AdamM
On Tue, Feb 02, 2010 at 10:28:37AM -0600, Adam Miller wrote:
Hello all, I wanted to bring a few things up and I wanted to bring them up on devel@lists.fp.o because this is where most people spend their time.
First off: "Does letting thousands of contributors do what they
want have a negative impact on our OS? (Mike)"[0]
- I would prefer that this be rephrased to a quote I read that
originated from John Rose (inode0) "isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" and I would prefer a focus be turned towards something like "Why was that the result of doing something that is essentially chaotic?" .... I guess my main question is: "Why are we fixing something that isn't broken?"
Second: "The Board has been working on defining a target audience
for Fedora. In response to this, some people feel that Fedora should allow sub-groups to define their own target audience"[1]
- I don't entirely understand this, don't SIGs or (sub groups)
essentially exist purely because there is some target audience? Clarification on this not would be appreciated.
Now, we come to the part that I feel is going to be viewed as a
touchy subject by many..... Why are there words like "letting" and "allow" being thrown around so often? I understand there are guidelines and policies for certain things of technical or legal nature in Fedora, but it feels a little like there is an attempt here to dictate how myself, as well as all others, spend their time contributing to The Fedora Project.
[0] https://fedoraproject.org/wiki/Unfinished_Board_issues [1] https://fedoraproject.org/wiki/User:Walters/SpinsSigsRemixes_TargetAudience
I would just like to know other contributors thoughts on these topics.
Thank you for your time,
Thyank you Adam, I cannot say anything to this but, I agree with you and inode0 100% on these points.
-Toshio
On Tue, 2010-02-02 at 10:28 -0600, Adam Miller wrote:
Now, we come to the part that I feel is going to be viewed as a
touchy subject by many..... Why are there words like "letting" and "allow" being thrown around so often? I understand there are guidelines and policies for certain things of technical or legal nature in Fedora, but it feels a little like there is an attempt here to dictate how myself, as well as all others, spend their time contributing to The Fedora Project.
In my mind this comes into play where the interests of one sig and another conflict as to how a package or configuration should perform. When there are more than one group trying to modify the behavior of a package, or asking the maintainer to change it in some way, and these modifications conflict with eachother, how is the maintainer to decide? How is FESCo to decide if it is brought to them for moderation? How is the board to decide if it gets escalated up to that level?
When viewing these discussions what strikes me is that none of it really applies until you have conflict. Without conflict there is no reason to worry, but when there is conflict, some general help in deciding who gets the change and who doesn't is welcome.
On Tue, Feb 2, 2010 at 10:41 AM, Jesse Keating jkeating@redhat.com wrote: <snip>
In my mind this comes into play where the interests of one sig and another conflict as to how a package or configuration should perform. When there are more than one group trying to modify the behavior of a package, or asking the maintainer to change it in some way, and these modifications conflict with eachother, how is the maintainer to decide? How is FESCo to decide if it is brought to them for moderation? How is the board to decide if it gets escalated up to that level?
When viewing these discussions what strikes me is that none of it really applies until you have conflict. Without conflict there is no reason to worry, but when there is conflict, some general help in deciding who gets the change and who doesn't is welcome.
<snip>
I will agree with that, I can see an application space for certain decisions when presented with conflict, but how often does this happen and how is it currently, as well as how has it been in the past, handled and resolved?
-AdamM
On Tue, 2010-02-02 at 11:02 -0600, Adam Miller wrote:
I will agree with that, I can see an application space for certain decisions when presented with conflict, but how often does this happen and how is it currently, as well as how has it been in the past, handled and resolved?
As sigs and spins grow in size or numbers the potential for this to happen is great. There have been cases of the Desktop sig wanting to bring in newer versions of some software, which the KDE sig was not ready for, there are conflicts between the traditional Unix folks and the future looking Desktop folks, there are conflicts between the "everything should work to it's fullest extent out of the box" folks and "our installs should be as slim and trim as possible with all optional functionality in separate and not installed by default packages" folks, and so on. Because we decree that our spins cannot make material changes to the packages and only some minor config changes, we put them in an awkward situation if they want to target an audience that is vastly different than the other audiences. To me, that's why it's important to define the overall target audience who "trumps" other audiences when there is a conflict. So that we can say "You are free to do whatever meets your needs, so long as it doesn't disrupt the needs of our target audience".
On Tue, Feb 02, 2010 at 09:15:15AM -0800, Jesse Keating wrote:
On Tue, 2010-02-02 at 11:02 -0600, Adam Miller wrote:
I will agree with that, I can see an application space for certain decisions when presented with conflict, but how often does this happen and how is it currently, as well as how has it been in the past, handled and resolved?
As sigs and spins grow in size or numbers the potential for this to happen is great. There have been cases of the Desktop sig wanting to bring in newer versions of some software, which the KDE sig was not ready for, there are conflicts between the traditional Unix folks and the future looking Desktop folks, there are conflicts between the "everything should work to it's fullest extent out of the box" folks and "our installs should be as slim and trim as possible with all optional functionality in separate and not installed by default packages" folks, and so on. Because we decree that our spins cannot make material changes to the packages and only some minor config changes, we put them in an awkward situation if they want to target an audience that is vastly different than the other audiences. To me, that's why it's important to define the overall target audience who "trumps" other audiences when there is a conflict. So that we can say "You are free to do whatever meets your needs, so long as it doesn't disrupt the needs of our target audience".
I agree with the idea that conflict resolution is something we need to work on but disagree vehemently with the idea that target audience should be what we throw into the mix to help decide that. Fedora the Project needs to be a good environment for contributors to join and try to create their visions of an open source operating system. Defining a trumping target audience means that there are potential roadblocks to that. Instead of defining a target audience so that conflicts can be decided with one side winning, a better way is to figure out ways that both sides get what they want.
Perhaps that means that we need to revisit things like our Conflict Guidelines. yum install '*' is the goal that the current Conflict Guidelines aim at. But if it limits the ability for people to make both a minimal-dependency version and a maximal-feature version simultaneously coexist perhaps that goal needs to be discarded rather than deciding that we must choose one or the other.
The leadership of Fedora needs to be servant leaders. People who work hard to help the people they lead to achieve their goals. We need mediators who work to help parties reach compromises not juries that render outside decisions.
-Toshio
On Tue, 2 Feb 2010, Adam Miller wrote:
Hello all, I wanted to bring a few things up and I wanted to bring them up on devel@lists.fp.o because this is where most people spend their time.
First off: "Does letting thousands of contributors do what they
want have a negative impact on our OS? (Mike)"[0]
- I would prefer that this be rephrased to a quote I read that
originated from John Rose (inode0) "isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" and I would prefer a focus be turned towards something like "Why was that the result of doing something that is essentially chaotic?" .... I guess my main question is: "Why are we fixing something that isn't broken?"
citation needed. I've worked hard on this question to find data and so far I've not been able to find any. I'm not about to answer this question with an opinion and I'd expect the same from everyone else. I'm sure the question makes contributors feel uneasy, after all it's putting our work into question. But if people are unwilling to ask it then there's no limit to what we can't accomplish.
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
-Mike
-----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1
On Tue, Feb 2, 2010 at 11:45 AM, Mike McGrath wrote:
<snip>
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
Is that how we measure success?? I am not suggesting it should or shouldn't be, but what is the measure of success for Fedora? I'd likely argue that's probably about as varied as the goals of contributors.
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
John
On Tue, 2 Feb 2010, inode0 wrote:
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
This is the fundamental difference between the two of us I think. I'm asking the questions[1] and trying to find the answers. You seem to think we don't need to ask the questions. That's why "does X cause Y" is a good question while "Isn't it great how?" isn't.
This particular question has already been answered, I've not yet put it on the wiki yet. The notes from our last meeting yesterday hasn't gone to the list, I'll update the wiki today though.
-Mike
[1] I'm asking these questions because I'm not happy with the state of our operating system. It's almost the entire reason I ran for the board.
On Tue, Feb 2, 2010 at 11:01 AM, Mike McGrath mmcgrath@redhat.com wrote:
On Tue, 2 Feb 2010, inode0 wrote:
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
This is the fundamental difference between the two of us I think. I'm asking the questions[1] and trying to find the answers. You seem to think we don't need to ask the questions. That's why "does X cause Y" is a good question while "Isn't it great how?" isn't.
If I seem to think we don't need to ask questions why did I just ask one that you are going to answer later?
The question I asked curiously is really of the "does X cause Y" variety. Restated it is does the state of the OS cause the lack of growth you cite? Or another way does the marketing effort cause the lack of growth you cite?
This particular question has already been answered, I've not yet put it on the wiki yet. The notes from our last meeting yesterday hasn't gone to the list, I'll update the wiki today though.
-Mike
[1] I'm asking these questions because I'm not happy with the state of our operating system. It's almost the entire reason I ran for the board.
Good for you. The rest of us get to ask questions too.
John
On Tue, 2 Feb 2010, inode0 wrote:
On Tue, Feb 2, 2010 at 11:01 AM, Mike McGrath mmcgrath@redhat.com wrote:
On Tue, 2 Feb 2010, inode0 wrote:
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
This is the fundamental difference between the two of us I think. I'm asking the questions[1] and trying to find the answers. You seem to think we don't need to ask the questions. That's why "does X cause Y" is a good question while "Isn't it great how?" isn't.
If I seem to think we don't need to ask questions why did I just ask one that you are going to answer later?
The question I asked curiously is really of the "does X cause Y" variety. Restated it is does the state of the OS cause the lack of growth you cite? Or another way does the marketing effort cause the lack of growth you cite?
Unless you were misquoted the question you asked was:
"Isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" [1]
As far as your question about the lack of growth, I was under the impression that to you our growth didn't matter[2]. My apologies.
[1] I'm asking these questions because I'm not happy with the state of our operating system. It's almost the entire reason I ran for the board.
Good for you. The rest of us get to ask questions too.
I encourage that. FAB is a good place for it.
-Mike
[1] http://lists.fedoraproject.org/pipermail/devel/2010-February/130157.html [2] http://lists.fedoraproject.org/pipermail/devel/2010-February/130167.html
On Tue, Feb 2, 2010 at 11:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
Unless you were misquoted the question you asked was:
"Isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" [1]
That was a rhetorical answer to the question, "Does letting thousands of contributors do what they want have a negative impact on our OS?" It translates more succinctly as "no." And was followed by a suggestion that asking questions about what in the unorchestrated stew that is the Fedora Project caused that result was something worth investigating.
As far as your question about the lack of growth, I was under the impression that to you our growth didn't matter[2]. My apologies.
I don't consider growth for the sake of growth important or part of the Fedora Project's mission. Targeted and sustainable growth where that growth furthers the Fedora Project's mission is what I care about and I don't think that is reflected in download statistics.
But my question about growth was sincere. Identifying lack of growth as a problem to me suggests on the surface a marketing issue, not an OS issue so I wanted to know why we were addressing it as an OS problem. There could be reasons it is, I'm not denying that possibility.
John
Mike McGrath said the following on 02/02/2010 09:01 AM Pacific Time:
This particular question has already been answered, I've not yet put it on the wiki yet. The notes from our last meeting yesterday hasn't gone to the list, I'll update the wiki today though.
The notes from our last meeting are here: http://lists.fedoraproject.org/pipermail/advisory-board/2010-February/007899...
On Tue, Feb 2, 2010 at 9:54 AM, inode0 inode0@gmail.com wrote:
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
The marketing "problem" is this: Who are we marketing to? Defining the target audience - as broad as it may be - helps here.
I'd also speculate that part of the reason that Fedora is not seeing as much grown in terms of downloads is that a lot of people don't like to fix what isn't broken. When things -just work-, the average end-user doesn't necessarily want to rock the boat. It could be a good thing. :) Especially when you consider that - although growth in downloads may not be consistent - contributor account growth seems to be very healthy. http://fedoraproject.org/wiki/File:Accounts_2009-10.png
John
devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
On Tue, Feb 2, 2010 at 11:28 AM, Robyn Bergeron robyn.bergeron@gmail.com wrote:
On Tue, Feb 2, 2010 at 9:54 AM, inode0 inode0@gmail.com wrote:
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
The marketing "problem" is this: Who are we marketing to? Defining the target audience - as broad as it may be - helps here.
To be clear I wasn't suggesting there actually was a marketing problem, although there is probably always a marketing problem in the absence of a monopoly.
I can imagine other approaches though. What are the characteristics of good contributors? Market to that segment of the population. What the desktop spin is or isn't probably doesn't matter in that case to the marketing effort.
Why hasn't marketing defined *its* target audience(s)? Why can't marketing identify the characteristics of groups they wish to market Fedora to and do it?
I'd also speculate that part of the reason that Fedora is not seeing as much grown in terms of downloads is that a lot of people don't like to fix what isn't broken. When things -just work-, the average end-user doesn't necessarily want to rock the boat. It could be a good thing. :) Especially when you consider that - although growth in downloads may not be consistent - contributor account growth seems to be very healthy. http://fedoraproject.org/wiki/File:Accounts_2009-10.png
I agree the one metric cited tells only a small part of the story.
John
On Tue, Feb 02, 2010 at 10:28:01AM -0700, Robyn Bergeron wrote:
On Tue, Feb 2, 2010 at 9:54 AM, inode0 inode0@gmail.com wrote:
On Tue, Feb 2, 2010 at 10:45 AM, Mike McGrath mmcgrath@redhat.com wrote:
And to answer your question about what "isnt' broken". I suggest you look at our http://fedoraproject.org/wiki/Statistics page. We've only seen growth in 2 of our last 6 releases. Think about that.
While I don't see that as directly relating to the mission of the Fedora Project I understand it is important to many people and I understand there is an indirect link with the mission. But what indicates that is a problem with the distribution as opposed to a marketing problem?
The marketing "problem" is this: Who are we marketing to? Defining the target audience - as broad as it may be - helps here.
I read your blog post from last month and liked its basic idea but had problems once it developed to the conclusion :-)
I think that the Fedora Project's target audience needs to be people who want to work on open source operating systems. If you want to market the Fedora Project, that's the audience that needs to be addressed.
If you want to market a physical product, like the Fedora Desktop Spin, then that should be a decision made below the Board level. Making a decision about the target audience of the various distributions that we have limits the choices of the people who want to work on open source operating systems. Making a target audience decision at the SIG level widens the choices as marketing/artistic/documentation/etc people can choose which audiences they want to address via which medium.
-Toshio
Toshio Kuratomi (a.badger@gmail.com) said:
I think that the Fedora Project's target audience needs to be people who want to work on open source operating systems. If you want to market the Fedora Project, that's the audience that needs to be addressed.
If you want to market a physical product, like the Fedora Desktop Spin, then that should be a decision made below the Board level. Making a decision about the target audience of the various distributions that we have limits the choices of the people who want to work on open source operating systems. Making a target audience decision at the SIG level widens the choices as marketing/artistic/documentation/etc people can choose which audiences they want to address via which medium.
I don't find this completely workable. By decreeing this sort of non-target for the project, and limiting Spin maintainers to only changes in the package set, but not the packages themselves (as we do), you're essentially telling them they'll never be able to attack their target audience *well*.
Bill
On Tue, Feb 02, 2010 at 01:54:37PM -0500, Bill Nottingham wrote:
Toshio Kuratomi (a.badger@gmail.com) said:
I think that the Fedora Project's target audience needs to be people who want to work on open source operating systems. If you want to market the Fedora Project, that's the audience that needs to be addressed.
If you want to market a physical product, like the Fedora Desktop Spin, then that should be a decision made below the Board level. Making a decision about the target audience of the various distributions that we have limits the choices of the people who want to work on open source operating systems. Making a target audience decision at the SIG level widens the choices as marketing/artistic/documentation/etc people can choose which audiences they want to address via which medium.
I don't find this completely workable. By decreeing this sort of non-target for the project, and limiting Spin maintainers to only changes in the package set, but not the packages themselves (as we do), you're essentially telling them they'll never be able to attack their target audience *well*.
In my other message, I go into more detail on this. Instead of defining a target audience and thereby limiting some contirubtors to working on second class products, change the other limitations that we have in place to allow them to do the work they need. Be an enabler, not a judge. Figure out ways that the packageset can be made to accomodate multiple uses rather than saying that the packageset is only optimized for one usecase and any contributor that wants to work on a conflicting use case needs to find a new project to work on.
My other mail suggests that one way to work with this is to create new conflicting packages that are optimized for the different usages. There's other ways as well but the general theme is that we need to be looking at ways to open up what people can do with the raw material of the Fedora Project to create their vision of a free software operating system rather than closing off what Fedora can be good for and making it so that certain visions are second class citizens that can only advance as long as they don't conflict with a different, specific vision.
-Toshio
Toshio Kuratomi (a.badger@gmail.com) said:
My other mail suggests that one way to work with this is to create new conflicting packages that are optimized for the different usages. There's other ways as well but the general theme is that we need to be looking at ways to open up what people can do with the raw material of the Fedora Project to create their vision of a free software operating system rather than closing off what Fedora can be good for and making it so that certain visions are second class citizens that can only advance as long as they don't conflict with a different, specific vision.
Would that mean that users who don't start with one of these 'products' get to magically try and choose which implementation of which they want? Perhaps even mix and match, leaving QA and the developers to sort out the results.
Furthermore, you then leave 'downstream' higher-level packages and applications having to, for example, code to PolicyKit0, PolicyKit1, or consolehelper, depending on what each 'product' use case might use. Or, having to build their python extensions simultaneously for python2.4, python2.6, and python3.0. These sorts of things would be extremely painful for developers, and would bloat the QA matrix excessively.
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
Bill
On Tue, Feb 2, 2010 at 2:17 PM, Bill Nottingham notting@redhat.com wrote: <Snip>
Would that mean that users who don't start with one of these 'products' get to magically try and choose which implementation of which they want? Perhaps even mix and match, leaving QA and the developers to sort out the results.
Furthermore, you then leave 'downstream' higher-level packages and applications having to, for example, code to PolicyKit0, PolicyKit1, or consolehelper, depending on what each 'product' use case might use. Or, having to build their python extensions simultaneously for python2.4, python2.6, and python3.0. These sorts of things would be extremely painful for developers, and would bloat the QA matrix excessively.
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
<Snip>
I think the responsibility of these things should be placed upon the SIG members who perform the functions from within these different groups. Why not have a QA person from each SIG work together with the larger QA efforts instead of potentially against them?
-AdamM
On Tue, Feb 02, 2010 at 02:22:37PM -0600, Adam Miller wrote:
I think the responsibility of these things should be placed upon the SIG members who perform the functions from within these different groups. Why not have a QA person from each SIG work together with the larger QA efforts instead of potentially against them?
If a spin wants to use a modified kernel package, what's the procedure for ensuring that it receives the same level of QA as the normal kernel?
On Tue, Feb 2, 2010 at 2:27 PM, Matthew Garrett mjg59@srcf.ucam.org wrote:
On Tue, Feb 02, 2010 at 02:22:37PM -0600, Adam Miller wrote:
I think the responsibility of these things should be placed upon the SIG members who perform the functions from within these different groups. Why not have a QA person from each SIG work together with the larger QA efforts instead of potentially against them?
If a spin wants to use a modified kernel package, what's the procedure for ensuring that it receives the same level of QA as the normal kernel?
<snip>
That's not something I think would be in the scope of a SIG, nor do I think something like that would make it past Spin review. This would also take the current SIG/Spin outside the scope of being part of the Fedora Project as it is no longer using Fedora packages, this (in my opinion at least) would be a situation where a fork would be needed.
-AdamM
On Tue, Feb 02, 2010 at 02:32:19PM -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:27 PM, Matthew Garrett mjg59@srcf.ucam.org wrote:
If a spin wants to use a modified kernel package, what's the procedure for ensuring that it receives the same level of QA as the normal kernel?
<snip>
That's not something I think would be in the scope of a SIG, nor do I think something like that would make it past Spin review. This would also take the current SIG/Spin outside the scope of being part of the Fedora Project as it is no longer using Fedora packages, this (in my opinion at least) would be a situation where a fork would be needed.
But beyond that, it's a matter of degree rather than principle. If we refuse to allow conflicting kernels to be included in the distribution, we're preventing some people from producing the spins that they want to work on. By only supporting a single kernel, we're implicitly stating that the focus of Fedora is limited to the people catered for by that kernel.
On Tue, 2 Feb 2010, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:17 PM, Bill Nottingham notting@redhat.com wrote:
<Snip> > Would that mean that users who don't start with one of these 'products' > get to magically try and choose which implementation of which they want? > Perhaps even mix and match, leaving QA and the developers to sort out > the results. > > Furthermore, you then leave 'downstream' higher-level packages and > applications having to, for example, code to PolicyKit0, PolicyKit1, or > consolehelper, depending on what each 'product' use case might use. Or, > having to build their python extensions simultaneously for python2.4, python2.6, > and python3.0. These sorts of things would be extremely painful for > developers, and would bloat the QA matrix excessively. > > Not to reduce the debate to too much of a soundbite, but it almost > seems like attempting to decide whether we want Fedora to be Debian, > or to be something useful for users of it. I'd always pick the latter... <Snip>
I think the responsibility of these things should be placed upon the SIG members who perform the functions from within these different groups. Why not have a QA person from each SIG work together with the larger QA efforts instead of potentially against them?
QA is a particular skill set, not every sig has a QA member and requiring it wouldn't work either. I feel it's like assuming that just because I've done turbogears apps that someone would ask me to do CSS as well. I don't think it's safe to assume that because someone can put a spin together that they have the tools and knowledge to do proper QA on it.
-Mike
On Tue, Feb 2, 2010 at 2:29 PM, Mike McGrath mmcgrath@redhat.com wrote: <snip>
QA is a particular skill set, not every sig has a QA member and requiring it wouldn't work either. I feel it's like assuming that just because I've done turbogears apps that someone would ask me to do CSS as well. I don't think it's safe to assume that because someone can put a spin together that they have the tools and knowledge to do proper QA on it.
-Mike
devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
I'm not saying that it should be a requirement of a SIG, but if it truly hits a point where it is not manageable from a QA standpoint to support all ends of the Fedora project then things will have to take a more granular approach and the QA project themselves will have to scope out what they will and won't work on and I imagine the SIGs/Spins will be first on the chopping block (and rightfully so).
-AdamM
Adam Miller (maxamillion@fedoraproject.org) said:
Furthermore, you then leave 'downstream' higher-level packages and applications having to, for example, code to PolicyKit0, PolicyKit1, or consolehelper, depending on what each 'product' use case might use. Or, having to build their python extensions simultaneously for python2.4, python2.6, and python3.0. These sorts of things would be extremely painful for developers, and would bloat the QA matrix excessively.
I think the responsibility of these things should be placed upon the SIG members who perform the functions from within these different groups. Why not have a QA person from each SIG work together with the larger QA efforts instead of potentially against them?
Take a random downstream app. (Firefox is an example, but there are many others.) Right now, it only needs to track a single version of python, or a single auth framework, even if it may be used on any desktop or any spin. The implication is that in some sort of future with SIG-specific conflicting frameworks, this downstream app maintainer now must be familiar with, and handle *all* of the frameworks, even though they're not specifcally a part of any SIG. That's sort of a rotten thing to do to Joe Random Maintainer.
You could say that the SIG needs to then supply people to handle every potential downstream app, but that's also not nice, and is going to lead to fun coordination with updates.
Bill
On Tue, Feb 2, 2010 at 2:29 PM, Bill Nottingham notting@redhat.com wrote: <snip>
Take a random downstream app. (Firefox is an example, but there are many others.) Right now, it only needs to track a single version of python, or a single auth framework, even if it may be used on any desktop or any spin. The implication is that in some sort of future with SIG-specific conflicting frameworks, this downstream app maintainer now must be familiar with, and handle *all* of the frameworks, even though they're not specifcally a part of any SIG. That's sort of a rotten thing to do to Joe Random Maintainer.
You could say that the SIG needs to then supply people to handle every potential downstream app, but that's also not nice, and is going to lead to fun coordination with updates.
<snip>
I don't think that's an issue either, I'm not proposing we change anything such that it could cause problems. I'm saying the way things are now works and I don't understand the desire to change it.
-AdamM
Adam Miller (maxamillion@fedoraproject.org) said:
On Tue, Feb 2, 2010 at 2:29 PM, Bill Nottingham notting@redhat.com wrote:
<snip> > Take a random downstream app. (Firefox is an example, but there are many > others.) Right now, it only needs to track a single version of python, > or a single auth framework, even if it may be used on any desktop or any > spin. The implication is that in some sort of future with SIG-specific > conflicting frameworks, this downstream app maintainer now must be familiar > with, and handle *all* of the frameworks, even though they're not > specifcally a part of any SIG. That's sort of a rotten thing to do to > Joe Random Maintainer. > > You could say that the SIG needs to then supply people to handle every > potential downstream app, but that's also not nice, and is going to lead > to fun coordination with updates. <snip>
I don't think that's an issue either, I'm not proposing we change anything such that it could cause problems.
Toshio was, and that's what I'm responding to.
Bill
On Tue, 2010-02-02 at 14:36 -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:29 PM, Bill Nottingham notting@redhat.com wrote:
<snip> > Take a random downstream app. (Firefox is an example, but there are many > others.) Right now, it only needs to track a single version of python, > or a single auth framework, even if it may be used on any desktop or any > spin. The implication is that in some sort of future with SIG-specific > conflicting frameworks, this downstream app maintainer now must be familiar > with, and handle *all* of the frameworks, even though they're not > specifcally a part of any SIG. That's sort of a rotten thing to do to > Joe Random Maintainer. > > You could say that the SIG needs to then supply people to handle every > potential downstream app, but that's also not nice, and is going to lead > to fun coordination with updates. <snip>
I don't think that's an issue either, I'm not proposing we change anything such that it could cause problems. I'm saying the way things are now works and I don't understand the desire to change it.
The way things are now "works" because of status quo. We tell anybody who wants to change status quo to go start a fork and do it there. Status quo is hard to define though, and hard to measure. So instead of trying to capture what the status quo is, where every package maintainer is essentially free to design their package however they see fit (within the package guidelines), instead we'd like a general idea of what the status quo should be. A project level idea of what type of things our packages should target. That will help people decide whether or not their spin can be done within those constraints, or if they have to go the way of a remix to accomplish their goal. Right now they don't have any guideline and are left to discover things on a package by package basis creating conflict as they go.
On Tue, Feb 02, 2010 at 01:11:47PM -0800, Jesse Keating wrote:
On Tue, 2010-02-02 at 14:36 -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:29 PM, Bill Nottingham notting@redhat.com wrote:
<snip> > Take a random downstream app. (Firefox is an example, but there are many > others.) Right now, it only needs to track a single version of python, > or a single auth framework, even if it may be used on any desktop or any > spin. The implication is that in some sort of future with SIG-specific > conflicting frameworks, this downstream app maintainer now must be familiar > with, and handle *all* of the frameworks, even though they're not > specifcally a part of any SIG. That's sort of a rotten thing to do to > Joe Random Maintainer. > > You could say that the SIG needs to then supply people to handle every > potential downstream app, but that's also not nice, and is going to lead > to fun coordination with updates. <snip>
I don't think that's an issue either, I'm not proposing we change anything such that it could cause problems. I'm saying the way things are now works and I don't understand the desire to change it.
The way things are now "works" because of status quo. We tell anybody who wants to change status quo to go start a fork and do it there.
Wait... The entire list of times I can remember someone being encouraged to take their contributions elsewhere are:
1) Kernel modules 2) Non-free software 3) Free software with legal issues 4) I think something to do with packaging content may have resulted in something but I don't know anything about the outcome there.
Who's been told to fork Fedora because of the status-quo-target-audience?
-Toshio
On Tue, Feb 02, 2010 at 17:16:14 -0500, Toshio Kuratomi a.badger@gmail.com wrote:
Who's been told to fork Fedora because of the status-quo-target-audience?
The guy who was complaining about nonfree firmware. He actually made a forked distribution for at least a while.
On Tue, Feb 02, 2010 at 05:16:14PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 01:11:47PM -0800, Jesse Keating wrote:
On Tue, 2010-02-02 at 14:36 -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:29 PM, Bill Nottingham notting@redhat.com wrote:
<snip> > Take a random downstream app. (Firefox is an example, but there are many > others.) Right now, it only needs to track a single version of python, > or a single auth framework, even if it may be used on any desktop or any > spin. The implication is that in some sort of future with SIG-specific > conflicting frameworks, this downstream app maintainer now must be familiar > with, and handle *all* of the frameworks, even though they're not > specifcally a part of any SIG. That's sort of a rotten thing to do to > Joe Random Maintainer. > > You could say that the SIG needs to then supply people to handle every > potential downstream app, but that's also not nice, and is going to lead > to fun coordination with updates. <snip>
I don't think that's an issue either, I'm not proposing we change anything such that it could cause problems. I'm saying the way things are now works and I don't understand the desire to change it.
The way things are now "works" because of status quo. We tell anybody who wants to change status quo to go start a fork and do it there.
Wait... The entire list of times I can remember someone being encouraged to take their contributions elsewhere are:
- Kernel modules
- Non-free software
- Free software with legal issues
- I think something to do with packaging content may have resulted in
something but I don't know anything about the outcome there.
Who's been told to fork Fedora because of the status-quo-target-audience?
Not in so many words, but the whole Zope/Plone fiasco from a few releases ago seems a prime example here. Fedora moved on with python, and we didn't allow a compat-python package for Zope and Plone to continue working. The reasons were varied, but they boiled down to python being a framework and having two frameworks providing almost identical things was not deemed to be something Fedora was going to do [1].
Those are the kinds of headaches Bill is talking about.
josh
[1] I realize Fedora is now doing python 2.6 and python3 side-by-side. I guess we'll find out how manageable that really is now ;)
On Tue, Feb 02, 2010 at 07:52:55PM -0500, Josh Boyer wrote:
On Tue, Feb 02, 2010 at 05:16:14PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 01:11:47PM -0800, Jesse Keating wrote:
The way things are now "works" because of status quo. We tell anybody who wants to change status quo to go start a fork and do it there.
Wait... The entire list of times I can remember someone being encouraged to take their contributions elsewhere are:
- Kernel modules
- Non-free software
- Free software with legal issues
- I think something to do with packaging content may have resulted in
something but I don't know anything about the outcome there.
Who's been told to fork Fedora because of the status-quo-target-audience?
Not in so many words, but the whole Zope/Plone fiasco from a few releases ago seems a prime example here. Fedora moved on with python, and we didn't allow a compat-python package for Zope and Plone to continue working. The reasons were varied, but they boiled down to python being a framework and having two frameworks providing almost identical things was not deemed to be something Fedora was going to do [1].
Once again, not a target audience decision. We didn't say, "Fedora is not for web developers, therefore we don't care enough to support zope and plone". We said, the python maintainer thinks that supporting multiple python stacks is infeasible therefore we aren't going to support this. It was a contributor and technical decision. Not a target-audience decision.
(Also note, I fell on the losing side of that discussion as well :-)
Those are the kinds of headaches Bill is talking about.
And I agree there are headaches there. But I think if something is valuable enough to a contributor, they'll step up to solve the headaches if they're requisites to being able to fulfill their vision. Instead of forbidding things we should be identifying the headaches and allowing them
After all, everything we do now is one big headache. Yet we have contributors willing to deal with every aspect of that.
-Toshio
On Tue, Feb 02, 2010 at 09:38:38PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 07:52:55PM -0500, Josh Boyer wrote:
On Tue, Feb 02, 2010 at 05:16:14PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 01:11:47PM -0800, Jesse Keating wrote:
The way things are now "works" because of status quo. We tell anybody who wants to change status quo to go start a fork and do it there.
Wait... The entire list of times I can remember someone being encouraged to take their contributions elsewhere are:
- Kernel modules
- Non-free software
- Free software with legal issues
- I think something to do with packaging content may have resulted in
something but I don't know anything about the outcome there.
Who's been told to fork Fedora because of the status-quo-target-audience?
Not in so many words, but the whole Zope/Plone fiasco from a few releases ago seems a prime example here. Fedora moved on with python, and we didn't allow a compat-python package for Zope and Plone to continue working. The reasons were varied, but they boiled down to python being a framework and having two frameworks providing almost identical things was not deemed to be something Fedora was going to do [1].
Once again, not a target audience decision. We didn't say, "Fedora is not for web developers, therefore we don't care enough to support zope and plone". We said, the python maintainer thinks that supporting multiple python stacks is infeasible therefore we aren't going to support this. It was a contributor and technical decision. Not a target-audience decision.
It is. It's one step removed. There were people actively wanting to make Zope/Plone work via a compat-python stack. It went all the way to FESCo and got voted down. The zope/plone users were the target audience there. There were people willing to do the work, all they needed was a yes from FESCo. We told them no. As Jesse has mentioned, 'status quo' won out.
Those are the kinds of headaches Bill is talking about.
And I agree there are headaches there. But I think if something is valuable enough to a contributor, they'll step up to solve the headaches if they're requisites to being able to fulfill their vision. Instead of forbidding things we should be identifying the headaches and allowing them
Not sure if you truncated that last sentence, but this whole paragraph sounds counter to your one above.
After all, everything we do now is one big headache. Yet we have contributors willing to deal with every aspect of that.
Everything we do is a big headache? I'm prone to hyperbole myself, but that's a bit over the top. If everything was a headache, nobody would volunteer for it.
josh
Josh Boyer wrote:
It is. It's one step removed. There were people actively wanting to make Zope/Plone work via a compat-python stack. It went all the way to FESCo and got voted down. The zope/plone users were the target audience there. There were people willing to do the work, all they needed was a yes from FESCo. We told them no. As Jesse has mentioned, 'status quo' won out.
I think this was just a bad decision. I complained back then and I still think we did the wrong thing. We should be as encompassing as legally possible within our Free Software ideals. Those packages eventually ended up in RPM Fusion anyway, like most of the stuff we refuse, so what was the point of preventing them from going into Fedora? Supportability concerns aren't going to vanish just because the package ends up in a third-party repository, and we have no way to prevent that.
I also think for the same reasons that we should allow acceptably-licensed (GPLv2 or compatible) kernel modules as external packages in Fedora, banning them gains us nothing and loses us hardware support we could gain without any moral (software freedom) compromises or legal risks.
Kevin Kofler
On Wed, 2010-02-03 at 19:08 +0100, Kevin Kofler wrote:
Josh Boyer wrote:
It is. It's one step removed. There were people actively wanting to make Zope/Plone work via a compat-python stack. It went all the way to FESCo and got voted down. The zope/plone users were the target audience there. There were people willing to do the work, all they needed was a yes from FESCo. We told them no. As Jesse has mentioned, 'status quo' won out.
I think this was just a bad decision. I complained back then and I still think we did the wrong thing. We should be as encompassing as legally possible within our Free Software ideals. Those packages eventually ended up in RPM Fusion anyway, like most of the stuff we refuse, so what was the point of preventing them from going into Fedora? Supportability concerns aren't going to vanish just because the package ends up in a third-party repository, and we have no way to prevent that.
I also think for the same reasons that we should allow acceptably-licensed (GPLv2 or compatible) kernel modules as external packages in Fedora, banning them gains us nothing and loses us hardware support we could gain without any moral (software freedom) compromises or legal risks.
What happens if we rebuild the kernel and one of the sub-modules doesn't get rebuilt and the maintainer goes awol? or it needs major rework to get built. Clearly you've never actually read any of the reasoning behind why we do this.
Dave.
Dave Airlie wrote:
What happens if we rebuild the kernel and one of the sub-modules doesn't get rebuilt and the maintainer goes awol? or it needs major rework to get built. Clearly you've never actually read any of the reasoning behind why we do this.
This can also happen with any other API change.
Ideally, someone would just get the module to build. (While sometimes major rework is indeed required, in most cases it's just a matter of finding out what changed and monkeying the change in the third-party module, which is something most provenpackagers should be able to do. I'd certainly help getting things to build if I'm not extremely busy with other stuff.) If not, I guess we have no other choice than dropping support for the module with the new kernel.
Kevin Kofler
Kevin Kofler wrote on 03.02.2010 19:08:
Josh Boyer wrote:
It is. It's one step removed. There were people actively wanting to make Zope/Plone work via a compat-python stack. It went all the way to FESCo and got voted down. The zope/plone users were the target audience there. There were people willing to do the work, all they needed was a yes from FESCo. We told them no. As Jesse has mentioned, 'status quo' won out.
I think this was just a bad decision. I complained back then and I still think we did the wrong thing.
Strong +1 to this
We should be as encompassing as legally possible within our Free Software ideals. Those packages eventually ended up in RPM Fusion anyway, like most of the stuff we refuse, so what was the point of preventing them from going into Fedora? Supportability concerns aren't going to vanish just because the package ends up in a third-party repository, and we have no way to prevent that.
I also think for the same reasons that we should allow acceptably-licensed (GPLv2 or compatible) kernel modules as external packages in Fedora, banning them gains us nothing and loses us hardware support we could gain without any moral (software freedom) compromises or legal risks.
As one of those behind the kmod stuff and someone interested in the topic for years: I think having the kernel modules outside of Fedora is very good thing, as that makes it quite clear to everyone: "This is unsupported by Fedora and its upstream source and hence might be crap and vanish at any time; don't rely on it and don't buy (or suggest others buying) hardware that is supported by this crap like this".
Further: In the end it afaics all boils does to the "if it's not good for the official kernel, why should it be good enough for Fedora".
By shipping that stuff we also might cannibalize upstream, which is not good for everyone in the long term most of the time and afaics. Or, IOW: The more changes and out-of-tree stuff Fedora integrates into its kernel the closer we get to a mess like the one with Android that is currently discussed quite a lot on the net ( http://lwn.net/Articles/372419/ and the comments there ). Sure, it is unlikely that it becomes that bad for us, but I for one would prefer if Fedora would not even go down that route at all and works so close with upstream that we ideally can ship vanilla kernels.
Cu knurd
On Tue, Feb 02, 2010 at 03:17:30PM -0500, Bill Nottingham wrote:
Toshio Kuratomi (a.badger@gmail.com) said:
My other mail suggests that one way to work with this is to create new conflicting packages that are optimized for the different usages. There's other ways as well but the general theme is that we need to be looking at ways to open up what people can do with the raw material of the Fedora Project to create their vision of a free software operating system rather than closing off what Fedora can be good for and making it so that certain visions are second class citizens that can only advance as long as they don't conflict with a different, specific vision.
Would that mean that users who don't start with one of these 'products' get to magically try and choose which implementation of which they want? Perhaps even mix and match, leaving QA and the developers to sort out the results.
Nope.
Users get a Product. That product has made choices about what packageset they receive. Mixing and matching of implementations is done at the level before the end-user. The Project can find ways to make this saner without going all the way to "if you conflict with the target audience your vision is not valuable here."
Furthermore, you then leave 'downstream' higher-level packages and applications having to, for example, code to PolicyKit0, PolicyKit1, or consolehelper, depending on what each 'product' use case might use. Or, having to build their python extensions simultaneously for python2.4, python2.6, and python3.0. These sorts of things would be extremely painful for developers, and would bloat the QA matrix excessively.
Also no.
You think that you can make people work on things they don't have an interest in? I certainly don't. Let's look at PolicyKit0 and PolicyKit1. KDE has one or two apps that uses PolicyKit0, Gnome has many apps that use PolicyKit1. People concerned with Gnome are packaging PolicyKit1. KDE SIG volunteers to package PolicyKit0 for their apps' consumption. Do the gnome apps have to support building with PolicyKit0? no. Do the KDE apps have to support building with PolicyKit1? no. You have people doing the work they need to in order to realize their vision.
For Fedora 13 we are going to be supporting both one python-2.x and python-3.x release. People who care about the respective stacks are going to start building extensions simultaneously for both. People who care about it are doing the work to see their vision fulfilled. OTOH, there aren't people who care about launching a similar effort to build and package for python-1.x (or even 2.4 unless you count EPEL). No work is being done there -- and no one is wasting their time doing it.
If there's a need, people will do the work to support their needs. If there's no need, nobody will. The job of Fedora's Leaders is to balance the individual needs of people who are working to create, sometimes conflicting visions, with the needs of each of those visions to get along in the same playground. Rather than classifying some of the kids as second class and siding against them when there's a dispute, it's better to find ways of expanding the playground to give more people the opportunity to form communities working on what they want.
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
-Toshio
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
If our products and brand suffer as a result...
and If that means our user base starts trending down (people don't like crap no matter how many people made it)...
Why would we be an attractive place for people wanting to come and experiment with their vision if so few are here to see it and so many other options are available?
-Mike
On Tue, Feb 02, 2010 at 04:16:30PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
I agree with things like Robin's statement of how having a target audience helps to market a product. What I think is wrong is to have the Fedora Board define the target audience that then constrains all of the products that Fedora produces.
I know we're all sick of car analogies but let's try one on for size. If Ford mandated that the target audience for the Ford brand should be soccer moms and anytime a conflict arose the decision that was friendlier to soccer moms was chosen we'd have a lot of people who weren't satisfied with the next generation of trucks, muscle cars, or economy cars.
-Toshio
On Tue, Feb 2, 2010 at 4:33 PM, Toshio Kuratomi a.badger@gmail.com wrote:
On Tue, Feb 02, 2010 at 04:16:30PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
I agree with things like Robin's statement of how having a target audience helps to market a product. What I think is wrong is to have the Fedora Board define the target audience that then constrains all of the products that Fedora produces.
I know we're all sick of car analogies but let's try one on for size. If Ford mandated that the target audience for the Ford brand should be soccer moms and anytime a conflict arose the decision that was friendlier to soccer moms was chosen we'd have a lot of people who weren't satisfied with the next generation of trucks, muscle cars, or economy cars.
-Toshio
-- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
+1 for the last ... 3? .. 4? ... how every many posts from Toshio, each well stated and I agree on the points stated.
-AdamM
On Tue, Feb 02, 2010 at 16:35:42 -0600, Adam Miller maxamillion@fedoraproject.org wrote:
+1 for the last ... 3? .. 4? ... how every many posts from Toshio, each well stated and I agree on the points stated.
I think he is putting up strawmen. Just because there is a target audience doesn't mean that anyone not directly producing stuff for that target audience is not really wanted by Fedora. Nor does it mean that the first step in resolving conflicts is to say the people targeting the stated market do whatever they want and other people have to change to accommodate them. I expect that everyone would try to accommodate others to the extent practical and only where accommodation is not practical would people be told that they have to do things the way needed to support the target audience.
I think deciding who Fedora is most interested in satisfying is important for the project. It provides us with a way to make consistent decisions when resolving conflicts. It provides us with a way to better optimize scarce resources. (Despite claims to the contrary, people donating time can be told where their efforts are most needed and thereby affect how they allocate their time.)
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 04:16:30PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
Because there's a thousand of us, you and I alone would come up with different ideas of what we're producing. I suspect the same for everyone.
I agree with things like Robin's statement of how having a target audience helps to market a product. What I think is wrong is to have the Fedora Board define the target audience that then constrains all of the products that Fedora produces.
So who says what that audience should be? We as a group are failing, I hope I'm wrong, but I really don't think I am. We need leadership and don't have it.
I know we're all sick of car analogies but let's try one on for size. If Ford mandated that the target audience for the Ford brand should be soccer moms and anytime a conflict arose the decision that was friendlier to soccer moms was chosen we'd have a lot of people who weren't satisfied with the next generation of trucks, muscle cars, or economy cars.
This is an excellent example. Ford has several audiences and several products. Now, imagine a world where Ford is forced to only produce one product. That's the world we're in right now. Lots of different people, lots of different needs yet we pretend we can produce something for all of them.
-Mike
On Tue, Feb 02, 2010 at 04:42:28PM -0600, Mike McGrath wrote:
This is an excellent example. Ford has several audiences and several products. Now, imagine a world where Ford is forced to only produce one product. That's the world we're in right now. Lots of different people, lots of different needs yet we pretend we can produce something for all of them.
Bullshit. Each of the SIGs is producing different products. They pull from the same resources but they produce different things. Just look at the KDE SIG's spin vs the desktop spin vs the Electronics Lab Spin vs the Games spin.
* A user who downloads any one of these products gets a different experience than someone who downloads one of the others. * Switching from one product to another is not an easy task of merely installing one package group and removing another. You have to know what packages to install and what packages to uninstall and sometimes you also need to know what configuration switches to hit. * Each of these products has a different target audience and a different use. * The major packages that each of these products is showcasing up front is different as well. Even if a minivan and a pickup truck have the same engine, drivetrain, and transmission under the hood they're still different products.
-Toshio
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 04:42:28PM -0600, Mike McGrath wrote:
This is an excellent example. Ford has several audiences and several products. Now, imagine a world where Ford is forced to only produce one product. That's the world we're in right now. Lots of different people, lots of different needs yet we pretend we can produce something for all of them.
Bullshit. Each of the SIGs is producing different products. They pull from the same resources but they produce different things. Just look at the KDE SIG's spin vs the desktop spin vs the Electronics Lab Spin vs the Games spin.
Then no one is actually using our products. People don't use spins after they install them. After install they're all pointed at the same thing. I'm a KDE user but I'm not using a KDE spin right now.
- A user who downloads any one of these products gets a different experience than someone who downloads one of the others.
- Switching from one product to another is not an easy task of merely installing one package group and removing another. You have to know what packages to install and what packages to uninstall and sometimes you also need to know what configuration switches to hit.
spins don't help this situation.
- Each of these products has a different target audience and a different use.
Citation needed. Both on what their target audience is *and* if they're succeeding.
- The major packages that each of these products is showcasing up front is different as well. Even if a minivan and a pickup truck have the same engine, drivetrain, and transmission under the hood they're still different products.
A pickup truck with a minivan engine sucks bad. Real bad. That's why they don't make them.
The above bullets say to me more then ever the spins are harming Fedora and not helping it. They're a place for us to focus, spend time, QA, hosting, etc, and at the end of the day gain absolutely nothing. Let the KDE sig focus on the KDE related packages and experience and not some crappy spin that no one who is reading this email right now is actually using.
-Mike
On Tue, 2010-02-02 at 18:33 -0600, Mike McGrath wrote:
The above bullets say to me more then ever the spins are harming Fedora and not helping it. They're a place for us to focus, spend time, QA, hosting, etc, and at the end of the day gain absolutely nothing. Let the KDE sig focus on the KDE related packages and experience and not some crappy spin that no one who is reading this email right now is actually using.
Mike, think I agree with you here, but care to elaborate just a little on what you mean to make sure? Are you saying focus on packages related to your sig, but not creating a spin for everyone of them or something to that extent?
On Tue, Feb 02, 2010 at 06:33:52PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 04:42:28PM -0600, Mike McGrath wrote:
This is an excellent example. Ford has several audiences and several products. Now, imagine a world where Ford is forced to only produce one product. That's the world we're in right now. Lots of different people, lots of different needs yet we pretend we can produce something for all of them.
Bullshit. Each of the SIGs is producing different products. They pull from the same resources but they produce different things. Just look at the KDE SIG's spin vs the desktop spin vs the Electronics Lab Spin vs the Games spin.
Then no one is actually using our products. People don't use spins after they install them. After install they're all pointed at the same thing. I'm a KDE user but I'm not using a KDE spin right now.
I disagree. Even though you all point at the same package set, you do not have the same product installed.
- A user who downloads any one of these products gets a different experience than someone who downloads one of the others.
- Switching from one product to another is not an easy task of merely installing one package group and removing another. You have to know what packages to install and what packages to uninstall and sometimes you also need to know what configuration switches to hit.
spins don't help this situation.
They do. I tried to switch from the Desktop spin to the KDE spin in F10 and ended up without a usable desktop environment. Reinstalled from the KDE spin and it worked. So "how do you get KDE on your computer?" "Install the Fedora KDE spin." Easy answer.
- Each of these products has a different target audience and a different use.
Citation needed. Both on what their target audience is *and* if they're succeeding.
Don't know if they're succeeding and didn't claim that they were or weren't. OTOH, target audience is a bit of a fudge -- The desktop and KDE spins seem to be aiming at similar audiences.
"Fedora remains the sole linux distribution distributing FEL methodologies for hardware design, simulation and verification. Thereby users tend to have the one-stop linux distribution for advanced electronic design. " https://fedoraproject.org/wiki/ElectronicLab_Spin
OTOH, I wasn't able to find anything written for either the KDE or desktop SIGs that can be construed as a statement of target audience so even if they appear to be aiming at the same targets, we don't know for sure. This appears to be a definite lack that would help marketing and other people trying to work at the product level instead of the project level. So, why doesn't the Board ask that a statement of target audience be added to the checklist of things that each spin produces?
- The major packages that each of these products is showcasing up front is different as well. Even if a minivan and a pickup truck have the same
engine, drivetrain, and transmission under the hood they're still different products.
A pickup truck with a minivan engine sucks bad. Real bad. That's why they don't make them.
Are you drawing the conclusion here that it's time for us to split the package collection so each SIG can work from their own packageset or are you just nit picking an irrelevant point in mine?
The above bullets say to me more then ever the spins are harming Fedora and not helping it. They're a place for us to focus, spend time, QA, hosting, etc, and at the end of the day gain absolutely nothing. Let the KDE sig focus on the KDE related packages and experience and not some crappy spin that no one who is reading this email right now is actually using.
Well, you're gaining nothing from them but other people are.
-Toshio
On Tue, Feb 02, 2010 at 21:07:23 -0500, Toshio Kuratomi a.badger@gmail.com wrote:
OTOH, I wasn't able to find anything written for either the KDE or desktop SIGs that can be construed as a statement of target audience so even if they appear to be aiming at the same targets, we don't know for sure. This appears to be a definite lack that would help marketing and other people trying to work at the product level instead of the project level. So, why doesn't the Board ask that a statement of target audience be added to the checklist of things that each spin produces?
Some of that is probably on the spin pages for the rest of the spins. I don't know that the default spins have them though.
I can tell you the games spin is constructed as a marketting tool. I don't think it was viewed that way by the previous games spin maintainer. This affected which packages were kept in the spin. Specifically I wanted the spin to work as is and dropped all of the games that were usable without downloaded content. (I probably should have also dropped the 3d games, but by the time I figured that out we had working 3d support with free drivers on some cards.)
While not quite agreeing with McGrath, I do feel that once someone has Fedora installed they can pull in any or all games without too much trouble, so I don't focus on the games one might install, but rather what ones I think make good showcases on a live (read only) image.
And note I ended up doing this because I wanted to make custom games spins and asked some questions about problems with the orphaned game spin and got asked to maintain it.
Bruno Wolff III wrote:
(I probably should have also dropped the 3d games, but by the time I figured that out we had working 3d support with free drivers on some cards.)
We do. The ATI Radeon 9200 SE on my desktop and the Intel GM965 on my notebook both work just fine for 3D with the Free drivers included in Fedora.
Kevin Kofler
On Wed, Feb 03, 2010 at 04:13:40 +0100, Kevin Kofler kevin.kofler@chello.at wrote:
Bruno Wolff III wrote:
(I probably should have also dropped the 3d games, but by the time I figured that out we had working 3d support with free drivers on some cards.)
We do. The ATI Radeon 9200 SE on my desktop and the Intel GM965 on my notebook both work just fine for 3D with the Free drivers included in Fedora.
The 9200 does 3d now, but for a while it didn't. OpenGL stuff just started working again shortly before the F12 release. I have one on my primary desktop machine.
Bruno Wolff III wrote:
The 9200 does 3d now, but for a while it didn't. OpenGL stuff just started working again shortly before the F12 release. I have one on my primary desktop machine.
For mine, it was perfect in F10, quite buggy in F11, mostly working in the F12 release (but Extreme Tux Racer triggered an assertion failure :-( ), now with the current updates it basically all works again. That said, there's some sluggishness in 3D games (sometimes the image just stops moving for a short instant) which seems to point at some remaining issue. I haven't tried Rawhide (neither before nor after F12).
Kevin Kofler
On Wed, Feb 03, 2010 at 04:30:11 +0100, Kevin Kofler kevin.kofler@chello.at wrote:
Bruno Wolff III wrote:
The 9200 does 3d now, but for a while it didn't. OpenGL stuff just started working again shortly before the F12 release. I have one on my primary desktop machine.
For mine, it was perfect in F10, quite buggy in F11, mostly working in the F12 release (but Extreme Tux Racer triggered an assertion failure :-( ), now with the current updates it basically all works again. That said, there's some sluggishness in 3D games (sometimes the image just stops moving for a short instant) which seems to point at some remaining issue. I haven't tried Rawhide (neither before nor after F12).
I use rawhide a lot on my main desktop now. So I probably see more breakage than you do. I also have monitors that don't do EDID and that caused me a lot of grief in the past.
On a related note, it looks like my nv20s might get nouveau support very soon. Phoronix posted an article today that some new code for mesa that supports older nvidia cards (through some nv20s) that don't use shaders is about to get merged to mesa master. (And I'm hoping shows up in rawhide shortly afterwards.)
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 06:33:52PM -0600, Mike McGrath wrote:
spins don't help this situation.
They do. I tried to switch from the Desktop spin to the KDE spin in F10 and ended up without a usable desktop environment. Reinstalled from the KDE spin and it worked. So "how do you get KDE on your computer?" "Install the Fedora KDE spin." Easy answer.
Spins didn't help, reinstalling did. You could have done the same thing without a spin and the live or dvd media.
A pickup truck with a minivan engine sucks bad. Real bad. That's why they don't make them.
Are you drawing the conclusion here that it's time for us to split the package collection so each SIG can work from their own packageset or are you just nit picking an irrelevant point in mine?
That's it exactly. I'm not picking nits. If we're going to have different target audiences we need to be able to build specifically for them. That includes letting the spins decide on their own if their audience needs a quick release cycle. Or if they really need the newer kernel, or a support cycle of 4 years, etc.
The alternative is picking one audience and targeting just for them.
So why do I argue for the latter? Because to do the former correctly would be a massive massive undertaking. Right now we're doing the former in a half assed manner. So people show up thinking Fedora is for them, it's for the server, it's for the desktop, it's for OLPC's, it's for kiosks, etc. Then when they find out it's really not that, they have a bad experience with Fedora and go elsewhere.
I really don't know what our users are a measure of. I don't think it's marketing as inode0 suggests, because the people using Fedora already know about it. But if we step back and take our users seriously. We'll find that since Fedora Core 6 released in 2006-10-24 to today, we've experienced a net growth of negative 3%. Yup, a 3% loss of users.
Our own users are moving _AWAY_ from Fedora. For whatever reason more users have chosen to not use Fedora then who have chosen to use Fedora. I suspect many have moved downsteam to Enterprise Linux. Which is ok but it's an indication that people came, tried Fedora, and moved on.
-Mike
Mike McGrath (mmcgrath@redhat.com) said:
I really don't know what our users are a measure of. I don't think it's marketing as inode0 suggests, because the people using Fedora already know about it. But if we step back and take our users seriously. We'll find that since Fedora Core 6 released in 2006-10-24 to today, we've experienced a net growth of negative 3%. Yup, a 3% loss of users.
Our own users are moving _AWAY_ from Fedora. For whatever reason more users have chosen to not use Fedora then who have chosen to use Fedora. I suspect many have moved downsteam to Enterprise Linux. Which is ok but it's an indication that people came, tried Fedora, and moved on.
This is, IMO, the crux of the problem. While I'm all for increasing our contributor community, if all we're doing is creating our own thing for *just* this contributing community, I'd be rather said. After all, if we're not going to try to do something big, why bother? :)
Bill
Mike McGrath wrote:
Spins didn't help, reinstalling did.
No. His problem was with switching desktop environment. It was solved by reinstalling with the spin for the target environment, getting the exact package selection optimized for that target environment.
(That said, adding KDE to a system installed from the GNOME spin is *supposed* to work! Removing GNOME, on the other hand, is near-impossible. Hint: "yum groupremove" does not and cannot work for this purpose.)
Kevin Kofler
On Wed, Feb 03, 2010 at 07:12:24PM +0100, Kevin Kofler wrote:
Mike McGrath wrote:
Spins didn't help, reinstalling did.
No. His problem was with switching desktop environment. It was solved by reinstalling with the spin for the target environment, getting the exact package selection optimized for that target environment.
(That said, adding KDE to a system installed from the GNOME spin is *supposed* to work! Removing GNOME, on the other hand, is near-impossible. Hint: "yum groupremove" does not and cannot work for this purpose.)
I am happilly switching KDE/Gnome and a few others all the time without many problems.
Richard
On Wed, 3 Feb 2010, Mike McGrath wrote:
I really don't know what our users are a measure of. I don't think it's marketing as inode0 suggests, because the people using Fedora already know about it. But if we step back and take our users seriously. We'll find that since Fedora Core 6 released in 2006-10-24 to today, we've experienced a net growth of negative 3%. Yup, a 3% loss of users.
Our own users are moving _AWAY_ from Fedora. For whatever reason more users have chosen to not use Fedora then who have chosen to use Fedora. I suspect many have moved downsteam to Enterprise Linux. Which is ok but it's an indication that people came, tried Fedora, and moved on.
Along with the above... If we're going to be the best at something don't we need to pick something to be the best at?
http://www.linux.com/learn/docs/ldp/282996-choosing-the-best-linux-distribut...
I particularly like this:
"Ubuntu edges out its closest contenders, Fedora and openSUSE, because its development team is constantly focused on the end-user experience."
What is it we're focused on? Do I need to just ask everyone individually and hope we all say the same thing?
-Mike
On Wed, Feb 3, 2010 at 8:28 PM, Mike McGrath mmcgrath@redhat.com wrote:
On Wed, 3 Feb 2010, Mike McGrath wrote:
I really don't know what our users are a measure of. I don't think it's marketing as inode0 suggests, because the people using Fedora already know about it. But if we step back and take our users seriously. We'll find that since Fedora Core 6 released in 2006-10-24 to today, we've experienced a net growth of negative 3%. Yup, a 3% loss of users.
Our own users are moving _AWAY_ from Fedora. For whatever reason more users have chosen to not use Fedora then who have chosen to use Fedora. I suspect many have moved downsteam to Enterprise Linux. Which is ok but it's an indication that people came, tried Fedora, and moved on.
Along with the above... If we're going to be the best at something don't we need to pick something to be the best at?
http://www.linux.com/learn/docs/ldp/282996-choosing-the-best-linux-distribut...
I particularly like this:
"Ubuntu edges out its closest contenders, Fedora and openSUSE, because its development team is constantly focused on the end-user experience."
What is it we're focused on? Do I need to just ask everyone individually and hope we all say the same thing?
Sadly they don't have categories like the best linux distribution for developers there.
John
On Wed, 3 Feb 2010, inode0 wrote:
On Wed, Feb 3, 2010 at 8:28 PM, Mike McGrath mmcgrath@redhat.com wrote:
On Wed, 3 Feb 2010, Mike McGrath wrote:
I really don't know what our users are a measure of. I don't think it's marketing as inode0 suggests, because the people using Fedora already know about it. But if we step back and take our users seriously. We'll find that since Fedora Core 6 released in 2006-10-24 to today, we've experienced a net growth of negative 3%. Yup, a 3% loss of users.
Our own users are moving _AWAY_ from Fedora. For whatever reason more users have chosen to not use Fedora then who have chosen to use Fedora. I suspect many have moved downsteam to Enterprise Linux. Which is ok but it's an indication that people came, tried Fedora, and moved on.
Along with the above... If we're going to be the best at something don't we need to pick something to be the best at?
http://www.linux.com/learn/docs/ldp/282996-choosing-the-best-linux-distribut...
I particularly like this:
"Ubuntu edges out its closest contenders, Fedora and openSUSE, because its development team is constantly focused on the end-user experience."
What is it we're focused on? Do I need to just ask everyone individually and hope we all say the same thing?
Sadly they don't have categories like the best linux distribution for developers there.
Is that what we're doing? If so would we win it?
-Mike
On Wed, Feb 3, 2010 at 8:51 PM, Mike McGrath mmcgrath@redhat.com wrote:
On Wed, 3 Feb 2010, inode0 wrote:
Sadly they don't have categories like the best linux distribution for developers there.
Is that what we're doing? If so would we win it?
One thing I know that I am not doing is competing with Ubuntu for the market it appeals to. Another thing I know that I am not doing is trying to win anything.
I am trying to support a community that works in a variety of ways to promote freedom, whether that be in terms of software or in terms of content or in terms of culture.
It is well known for being an engine of innovative, cutting-edge technology largely accomplished by working closely with upstream projects. I suspect that is something that appeals to a healthy segment of the developer pool and that distinguishes us from other distributions. I don't need to win a prize or see Fedora in a poll finish ahead of Ubuntu to view this as a success.
... omission of about 50 other things we stand for and promote ...
If we foster the sort of community described on the overview page of the wiki, we are winning what matters - we are living the mission we defined.
John
+1 inode0
-AdamM from Android(CM)
On Feb 3, 2010 9:29 PM, "inode0" inode0@gmail.com wrote:
On Wed, Feb 3, 2010 at 8:51 PM, Mike McGrath mmcgrath@redhat.com wrote:
On Wed, 3 Feb 2010, inod...
One thing I know that I am not doing is competing with Ubuntu for the market it appeals to. Another thing I know that I am not doing is trying to win anything.
I am trying to support a community that works in a variety of ways to promote freedom, whether that be in terms of software or in terms of content or in terms of culture.
It is well known for being an engine of innovative, cutting-edge technology largely accomplished by working closely with upstream projects. I suspect that is something that appeals to a healthy segment of the developer pool and that distinguishes us from other distributions. I don't need to win a prize or see Fedora in a poll finish ahead of Ubuntu to view this as a success.
... omission of about 50 other things we stand for and promote ...
If we foster the sort of community described on the overview page of the wiki, we are winning what matters - we are living the mission we defined.
John
On Wed, 2010-02-03 at 20:51 -0600, Mike McGrath wrote:
On Wed, 3 Feb 2010, inode0 wrote:
On Wed, Feb 3, 2010 at 8:28 PM, Mike McGrath mmcgrath@redhat.com wrote:
On Wed, 3 Feb 2010, Mike McGrath wrote:
I really don't know what our users are a measure of. I don't think it's marketing as inode0 suggests, because the people using Fedora already know about it. But if we step back and take our users seriously. We'll find that since Fedora Core 6 released in 2006-10-24 to today, we've experienced a net growth of negative 3%. Yup, a 3% loss of users.
Our own users are moving _AWAY_ from Fedora. For whatever reason more users have chosen to not use Fedora then who have chosen to use Fedora. I suspect many have moved downsteam to Enterprise Linux. Which is ok but it's an indication that people came, tried Fedora, and moved on.
Along with the above... If we're going to be the best at something don't we need to pick something to be the best at?
http://www.linux.com/learn/docs/ldp/282996-choosing-the-best-linux-distribut...
I particularly like this:
"Ubuntu edges out its closest contenders, Fedora and openSUSE, because its development team is constantly focused on the end-user experience."
What is it we're focused on? Do I need to just ask everyone individually and hope we all say the same thing?
Sadly they don't have categories like the best linux distribution for developers there.
Is that what we're doing? If so would we win it?
(Apologies for diving into this thread, this got me thinking)
"best linux distribution for developer" seems too vague to me to be achievable. I think there are different categories of developer.
Here's an attempt at a concrete and realistic (though fictitious) persona: - Gillian is one of 6 employees at "stelthix.com", a startup based in Cambridge MA. She is a graduate of MIT. - The startup is in in "stealth" mode, building a web-based service that will be the next Google, or at least, they hope, acquired by Google (they're not yet saying what the service does). - They hope to launch the site in 3 months time; they are working every waking hour building the site and the backend, talking to investors, signing up service providers etc - All of the employees do at least some "development", even if it's just editing HTML templates, and tweaking of Python scripts. - Their web site is implemented in Django, and they're heavily using Python throughout the backend, though they have some optimized C code which one of the other developers wrote for a compute-intensive task. - They have an internal Trac instance which they're using as a private wiki, an issue tracker, and for SVN. The SVN instance stores all of their code (for both the web site, the scraping/data mining tool that feeds the data, their custom scripts that leverage Google's APIs etc). - They're happy to use FLOSS, but their code is going to be proprietary (alas). They have written an API which customers of the site can use for some purposes, but those customers will never see the implementation. - They are renting time on Amazon EC2 for the compute-heavy parts of the backend, and the beta instance of the site is hosted on Linux. - They have a buildbot that is running the full test suite after every check-in; this is running on a Linux box somewhere. - Most of the team use Mac laptops running OS X (alas), but the deployment environment is Linux, and some of the team have Linux boxes which they use for development as well. - They try to stick to the standard Python libraries plus Django because it's fiddly tracking additional dependencies in their (mixture of Mac + Linux) world.
I think this is a realistic story [1], and is more concrete than "best linux distribution for developers". It leads to these questions: why will Gillian choose to use Fedora on her laptop? Why will Gillian choose to use Fedora on the backend servers? Why will Gillian recommend Fedora to the new hire after the company gets more VC funding?
I'm somewhat biased towards Python here; you could rewrite this somewhat and change Python and Django to Ruby and Rails, and it's probably important to do both cases well; we want a great Rails story as well as a great Python story - "Ray was in the same class as Gillian, and now works at wearemorepragmaticthanyou.com", perhaps.
Another developer persona might be: - Fred is a sysadmin and postdoc at example.ac.uk - he manages a variety of servers and workstations on the campus as a job, whilst working towards finishing his thesis - in his spare time he is working directly on a re-implementation of an encumbered piece of software - He cares deeply about software freedom, and needs a decent build of the tools he needs (gcc, GNU make, gdb, perl). - He worries about software patents, and has tried to avoid MP3 for some years, but doesn't always succeed.
I hope this is useful and realistic, and not too much of a caricature.
I believe we currently do a good job of appealing to Fred (though we could always do better), but less so at appealing to Gillian or Ray.
This hits the "web 2.0 startup" cases and the "enthusiastic FLOSS volunteer" case; there'd probably need to be a persona for a Java developer within a large company too, and probably stuff I've not thought of.
Hope this is helpful Dave
[1] though the last time I worked at a company of that size was some years ago; caveat lector
On Thu, Feb 4, 2010 at 8:41 AM, David Malcolm dmalcolm@redhat.com wrote:
On Wed, 2010-02-03 at 20:51 -0600, Mike McGrath wrote:
On Wed, 3 Feb 2010, inode0 wrote:
Sadly they don't have categories like the best linux distribution for developers there.
Is that what we're doing? If so would we win it?
(Apologies for diving into this thread, this got me thinking)
"best linux distribution for developer" seems too vague to me to be achievable. I think there are different categories of developer.
Well, according to the above analysis it is achievable if linux.com says we are. But even if we assume we are in fact the best linux distribution for developers or for engineers or for graduate students in scientific fields or for whatever it doesn't follow that we want to have that group in mind for a target audience for any particular product of the distribution (at least I find it inconceivable those would be target audiences of the default desktop).
Even if we change the focus to identifying a target audience for the project, which is where I think "developers" would rank very high on the list your analysis is valid. We don't appeal to all developers as a project either.
Do we focus on a narrow achievable target audience that it is realistic for us to be the best for now? I bet that would result in a worsening of the perceived crisis. Or should we focus on a group with broad appeal that while perhaps not ever being achievable will lessen the indicators of the crisis? Or do we go about our business attracting, say, recreational FOSS python developers who as a side-effect of adding cool feature X to the Fedora distribution also add less visible things Y and Z to make the life of a python developer using Fedora better?
I always, perhaps mistakenly, thought the point of the default spin was to showcase the work of the developers, artists, documentation writers, and others who are contributing so much to the Fedora Project. Who is supposed to find that sort of showcase interesting? Or is that just a quaint old notion of the output of a project in its infancy? It probably is ...
John
On Wed, Feb 03, 2010 at 08:43:45PM -0600, inode0 wrote:
http://www.linux.com/learn/docs/ldp/282996-choosing-the-best-linux-distribut...
I particularly like this:
"Ubuntu edges out its closest contenders, Fedora and openSUSE, because its development team is constantly focused on the end-user experience."
What is it we're focused on? Do I need to just ask everyone individually and hope we all say the same thing?
Sadly they don't have categories like the best linux distribution for developers there.
I would prefer the category don't annoy experienced users. In this category there is imho some room for improvement:
* get away from the mantra "clean new install" is better than system upgrade * not everyone likes to have everything in just one big partition
Richard
On 02/02/2010 09:07 PM, Toshio Kuratomi wrote:
- A user who downloads any one of these products gets a different experience than someone who downloads one of the others.
- Switching from one product to another is not an easy task of merely installing one package group and removing another. You have to know what packages to install and what packages to uninstall and sometimes you also need to know what configuration switches to hit.
spins don't help this situation.
They do. I tried to switch from the Desktop spin to the KDE spin in F10 and ended up without a usable desktop environment. Reinstalled from the KDE spin and it worked. So "how do you get KDE on your computer?" "Install the Fedora KDE spin." Easy answer.
Spins make sense when there is a deep-reaching feature that touches a majority of packages on the system. Examples include:
- the desktop environment with all the supporting runtime libs
- I would say 32 and 64-bit environments are two 'spins'
- a hypothetical major version of glibc-based 'spin'
I don't understand why 'Electronic Design Lab' is a separate spin: if I install all the EDA-related packages that it contains, would I not get an equivalent capability?
The only reason I can think of is the media capacity limitation, which forces dropping some packages to make space for someone's desired set which is not already part of the mainstream collection.
On Wed, 2010-02-03 at 11:23 -0500, Przemek Klosowski wrote:
I don't understand why 'Electronic Design Lab' is a separate spin: if I install all the EDA-related packages that it contains, would I not get an equivalent capability?
The only reason I can think of is the media capacity limitation, which forces dropping some packages to make space for someone's desired set which is not already part of the mainstream collection.
There is also the issue of multiple providers of a given functionality. When all are present, an algorithm tries to pick the "best" provider, which may not make sense to a human, but every human is different. By breaking up the large package set into a smaller subset, one can short circuit that "best" selection by only having one provider for that given functionality.
FEL exists for the reason you stated above, but also as a marketing tool, as it is very easy to install from a Live image, you had somebody a disk and say "install this, you'll have an electronics lab". Much easier than handing them a stack of DVDs and saying "Start this install, select this package here, this group there, remove this package here, format accordingly, and hopefully you got all the right selections done".
On 02/03/2010 11:46 AM, Jesse Keating wrote:
On Wed, 2010-02-03 at 11:23 -0500, Przemek Klosowski wrote:
I don't understand why 'Electronic Design Lab' is a separate spin: if I install all the EDA-related packages that it contains, would I not get an equivalent capability?
The only reason I can think of is the media capacity limitation, which forces dropping some packages to make space for someone's desired set which is not already part of the mainstream collection.
There is also the issue of multiple providers of a given functionality. When all are present, an algorithm tries to pick the "best" provider, which may not make sense to a human, but every human is different. By breaking up the large package set into a smaller subset, one can short circuit that "best" selection by only having one provider for that given functionality.
What do you mean by 'functionality'? Is it what's provided by an RPM package? This would suggest that packages in spins would be functionally different---which is a little uncomfortable to me, because how can I ever know that I have the best version of every tool? To make sure I would have to try all the tools from all the spins, in principle.
FEL exists for the reason you stated above, but also as a marketing tool, as it is very easy to install from a Live image, you had somebody a disk and say "install this, you'll have an electronics lab". Much easier than handing them a stack of DVDs and saying "Start this install, select this package here, this group there, remove this package here, format accordingly, and hopefully you got all the right selections done".
I have a generic Fedora install with an 'Electronics' tab in the 'Applications' menu. I got it after I selected it from the Engineering section in the yumex GUI, I believe. This is preferable to me, as compared to installing a separate spin.
I can see a psychological difference here. I am used to having one computer on which I do everything, from watching youtube videos to designing PCBs, rather than several computers for specialized tasks. Maybe the kids today see it differently :).
Przemek Klosowski wrote:
Spins make sense when there is a deep-reaching feature that touches a majority of packages on the system. Examples include:
- the desktop environment with all the supporting runtime libs
… and applications!
Our spins also select core applications (file manager, text editor, web browser, word processor etc.) which are part of the desktop environment.
I don't understand why 'Electronic Design Lab' is a separate spin: if I install all the EDA-related packages that it contains, would I not get an equivalent capability?
Yes, but having a spin with them already on it is much simpler for its target audience. (That said, I wouldn't use it since they moved away from KDE to GNOME. :-/ If I needed FEL, I'd rather either groupinstall their comps group on a KDE spin install or install individual apps.)
Kevin Kofler
Once upon a time, Kevin Kofler kevin.kofler@chello.at said:
Yes, but having a spin with them already on it is much simpler for its target audience. (That said, I wouldn't use it since they moved away from KDE to GNOME. :-/ If I needed FEL, I'd rather either groupinstall their comps group on a KDE spin install or install individual apps.)
Would it be possible to put spin kickstarts on the common install DVD, with an option in anaconda to choose them (and notes that network access may be required for some packages)? This would give an easier way to install alternate spins, without having to download and burn lots of CDs, boot, and then transfer to the hard drive.
On Wed, Feb 3, 2010 at 12:35 PM, Chris Adams cmadams@hiwaay.net wrote:
Once upon a time, Kevin Kofler kevin.kofler@chello.at said:
Yes, but having a spin with them already on it is much simpler for its target audience. (That said, I wouldn't use it since they moved away from KDE to GNOME. :-/ If I needed FEL, I'd rather either groupinstall their comps group on a KDE spin install or install individual apps.)
Would it be possible to put spin kickstarts on the common install DVD, with an option in anaconda to choose them (and notes that network access may be required for some packages)? This would give an easier way to install alternate spins, without having to download and burn lots of CDs, boot, and then transfer to the hard drive.
<snip>
That's actually a really cool idea and I'd be curious to know if it was possible as well.
-AdamM
Would it be possible to put spin kickstarts on the common install DVD, with an option in anaconda to choose them (and notes that network access may be required for some packages)? This would give an easier way to install alternate spins, without having to download and burn lots of CDs, boot, and then transfer to the hard drive.
We talked about something along these lines at the last FUDCon, but other pressures have ensured I've had no time to spend working on it. I'd still like to, or at least sit down and type up what we hashed out so other people can take a stab at it.
- Chris
Chris Lumens (clumens@redhat.com) said:
Would it be possible to put spin kickstarts on the common install DVD, with an option in anaconda to choose them (and notes that network access may be required for some packages)? This would give an easier way to install alternate spins, without having to download and burn lots of CDs, boot, and then transfer to the hard drive.
We talked about something along these lines at the last FUDCon, but other pressures have ensured I've had no time to spend working on it. I'd still like to, or at least sit down and type up what we hashed out so other people can take a stab at it.
Not to hijack a completely different bug report/thread, but I suspect that product.img could be used for this (or multiple product.img files)?
Bill
Mike McGrath wrote:
Then no one is actually using our products. People don't use spins after they install them. After install they're all pointed at the same thing. I'm a KDE user but I'm not using a KDE spin right now.
Then you're missing out on some of the integration work we do. That said, I'm fairly confident *you* know what stuff to install and what to remove to get a decent KDE after a DVD install / netinstall / "some other spin" install, but the average user will end up with a suboptimal KDE if they install from anything other than the KDE spin.
The above bullets say to me more then ever the spins are harming Fedora and not helping it. They're a place for us to focus, spend time, QA, hosting, etc, and at the end of the day gain absolutely nothing. Let the KDE sig focus on the KDE related packages and experience and not some crappy spin that no one who is reading this email right now is actually using.
Both my machines were installed from the KDE spin. And I take offense at you calling it "crappy", we work a lot on it (especially our live image maintainer, Sebastian Vahl, does, but we all participate). It's also what we primarily test. A live image is perfect for testing, just burn it and boot from it, or boot it in a VM, and you immediately see what works and what doesn't work. It's also great for users to install, you get a known-working packageset with the basic desktop environment (e.g. KDE Plasma Desktop) and all the basic apps on it and then you add the additional apps you want.
And in addition, if you want us to focus more on the installer-based spins (DVD, multi-CD set, netinstall), you need to solve (or get somebody else to solve) the "comps is biased towards GNOME" problem: all the groups of desktop applications, e.g. "Multimedia / Sound and Video", "Internet applications" etc. all default to GNOME apps, sometimes even as mandatory entries. To make the installer useful for our purposes, we'd need Anaconda to show a desktop selection screen like the one in the openSUSE installer before the package selection screen, then have comps defaults conditionalized on the selection made there. Until/unless that happens, we cannot recommend the installer-based spins to KDE users.
Kevin Kofler
On Wed, 3 Feb 2010, Kevin Kofler wrote:
Mike McGrath wrote:
Then no one is actually using our products. People don't use spins after they install them. After install they're all pointed at the same thing. I'm a KDE user but I'm not using a KDE spin right now.
Then you're missing out on some of the integration work we do. That said, I'm fairly confident *you* know what stuff to install and what to remove to get a decent KDE after a DVD install / netinstall / "some other spin" install, but the average user will end up with a suboptimal KDE if they install from anything other than the KDE spin.
And the reason you install from a spin, as history will show, was a technical limitation in the live CD that doesn't allow package selection. We never designed it that way. It just happened. There was a time where, when you wanted KDE, you clicked the checkbox next to KDE at install time. With our default and spin media we've actually _LOST_ functionality that AFAIK is still not back after years of work. If it is please correct me.
-Mike
On Tue, 2010-02-02 at 23:03 -0600, Mike McGrath wrote:
There was a time where, when you wanted KDE, you clicked the checkbox next to KDE at install time. With our default and spin media we've actually _LOST_ functionality that AFAIK is still not back after years of work.
pardon? Our DVD distribution still includes KDE on it, and you can still click KDE to install it. What functionality has been lost here?
On Tue 2 February 2010 9:10:13 pm Jesse Keating wrote:
What functionality has been lost here?
Working KDM, for one... Installing from the live DVD (as Kevin Kofler mentioned earlier) is essentially broken if you want KDE as the primary DE but choose to install any other comps.
On Wed, Feb 03, 2010 at 12:44:08AM -0800, Ryan Rix wrote:
On Tue 2 February 2010 9:10:13 pm Jesse Keating wrote:
What functionality has been lost here?
Working KDM, for one... Installing from the live DVD (as Kevin Kofler mentioned earlier) is essentially broken if you want KDE as the primary DE but choose to install any other comps.
We didn't have live media previously. It's not lost (or regressed) function. It's simply not working on the newer media type.
josh
Josh Boyer wrote:
On Wed, Feb 03, 2010 at 12:44:08AM -0800, Ryan Rix wrote:
On Tue 2 February 2010 9:10:13 pm Jesse Keating wrote:
What functionality has been lost here?
Working KDM, for one... Installing from the live DVD (as Kevin Kofler mentioned earlier) is essentially broken if you want KDE as the primary DE but choose to install any other comps.
We didn't have live media previously. It's not lost (or regressed) function. It's simply not working on the newer media type.
I think he actually means the NON-live DVD.
Kevin Kofler
On Tue, Feb 02, 2010 at 05:33:02PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 04:16:30PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
I agree with things like Robin's statement of how having a target audience helps to market a product. What I think is wrong is to have the Fedora Board define the target audience that then constrains all of the products that Fedora produces.
What? No. The Board has defined a default spin, and is working on a target audience for the default Spin. The Board has explicitly declared that SPINS are ALLOWED to define their OWN target audience.
josh
On Tue, Feb 02, 2010 at 07:56:53PM -0500, Josh Boyer wrote:
On Tue, Feb 02, 2010 at 05:33:02PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 04:16:30PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
I agree with things like Robin's statement of how having a target audience helps to market a product. What I think is wrong is to have the Fedora Board define the target audience that then constrains all of the products that Fedora produces.
What? No. The Board has defined a default spin, and is working on a target audience for the default Spin. The Board has explicitly declared that SPINS are ALLOWED to define their OWN target audience.
Does the Board create the default Spin? No? So why shouldn't the Board just ask the people who create the spin to clearly state their target audience?
-Toshio
On Tue, Feb 02, 2010 at 09:41:11PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 07:56:53PM -0500, Josh Boyer wrote:
On Tue, Feb 02, 2010 at 05:33:02PM -0500, Toshio Kuratomi wrote:
On Tue, Feb 02, 2010 at 04:16:30PM -0600, Mike McGrath wrote:
On Tue, 2 Feb 2010, Toshio Kuratomi wrote:
Not to reduce the debate to too much of a soundbite, but it almost seems like attempting to decide whether we want Fedora to be Debian, or to be something useful for users of it. I'd always pick the latter...
The problem with this sound bite is that Fedora Project and Fedora product get mixed up. Users use a Fedora product. The Fedora Project attracts the contributors who make various Fedora products. You can't continue to be an attractive place for people wanting to experiment with creating different visions that don't necessarily appeal to the target audience if they're always going to be a second class citizen.
These are 3 if's and they're impossible to say for sure right now but over time we'll know:
If we don't have a coherent vision for what our products are and who they are for..
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
I agree with things like Robin's statement of how having a target audience helps to market a product. What I think is wrong is to have the Fedora Board define the target audience that then constrains all of the products that Fedora produces.
What? No. The Board has defined a default spin, and is working on a target audience for the default Spin. The Board has explicitly declared that SPINS are ALLOWED to define their OWN target audience.
Does the Board create the default Spin? No? So why shouldn't the Board just ask the people who create the spin to clearly state their target audience?
The Board is responsible for Fedora overall. They are concerned with Fedora uptake and ways of increasing contribution. Based on that, they are trying to come up with personas that seem a likely candidate to use and eventually contribute to Fedora. Based on that, they are trying to come up with a target audience for the DEFAULT spin.
The current default spin is the Desktop spin, yes. It was chosen partly because of 'status quo' and partly because it was offered up as such. If the Board comes up with a target audience for the default spin that does not match with what the Desktop spin is trying to accomplish, then perhaps some other spin will eventually fill that void.
However, if you think the Board hasn't heard and taken into account what the target audience of the current Desktop spin is then you are just being silly. The proposals for the default spin being Desktop came from the Desktop team. The discussions on target audience involve members of the Desktop team. They're more than aware of what is going on, and at the moment the target audience seems to fit perfectly there. If it eventually doesn't, then something else might fill the void.
If you'd like to next argue that the Board is simply crafting a target audience that fits the pre-defined Desktop spin, please don't. There is real thought going into this from a broader Fedora stance and simplifying this topic into a pissing match between spins is not productive. I might not agree with everything the Board has identified for the target audience, but I do at least appreciate the efforts and discussions that have taken place to get to that point.
josh
Josh Boyer wrote:
The Board is responsible for Fedora overall. They are concerned with Fedora uptake and ways of increasing contribution. Based on that, they are trying to come up with personas that seem a likely candidate to use and eventually contribute to Fedora. Based on that, they are trying to come up with a target audience for the DEFAULT spin.
The whole concept of a default spin is what I and a few others here object to in the first place. There should be no one default! There should be a set of 2 or 3 primary spins (GNOME, KDE and possibly some third option, probably something lightweight and/or netbook-oriented) to choose from as equal first-class citizens.
(And FWIW, I really don't see why the Fedora Project insists on abusing the word "Desktop" to mean "GNOME".)
Kevin Kofler
On Wed, Feb 03, 2010 at 07:23:22PM +0100, Kevin Kofler wrote:
to in the first place. There should be no one default! There should be a set of 2 or 3 primary spins (GNOME, KDE and possibly some third option, probably something lightweight and/or netbook-oriented) to choose from as equal first-class citizens.
Cloud appliance baseline.
On Tue, 2010-02-02 at 17:33 -0500, Toshio Kuratomi wrote:
Let's cut this off right at the top :-) If a vision for what our products are is a problem why don't we have the people producing the products explain their vision? I keep saying that vision for products needs to come from the people producing those products, not from the Board or FESCo.
I'm the one creating the "Fedora" product, the DVD that gets so many downloads, and the thing that gets network installed. I have no idea what the vision is for it, because I have no guidance. I'm just making it up as I go, continuing to take what was Red Hat Linux and Fedora Core and trimming it as necessary to fit on a DVD.
On Tue, 2010-02-02 at 18:20 -0800, Jesse Keating wrote:
I'm the one creating the "Fedora" product, the DVD that gets so many downloads, and the thing that gets network installed. I have no idea what the vision is for it, because I have no guidance. I'm just making it up as I go, continuing to take what was Red Hat Linux and Fedora Core and trimming it as necessary to fit on a DVD.
I would call what your actually doing, to be a "fedora" vision as a whole or in general. That is what I would consider to be the basic overall vision of the fedora project and the packages that are currently included (not the DVD you create, but the rawhide or F-x directory that exists as an official release).
And what I believe everyone else (guess spelling this out for either you, or whomever might not understand yet) is talking about, is taking all those packages that you have had built (via koji, moch, pungi, etc..) and putting them into spins and having them work like that. But to do that, and to know exactly what that spin needs to have, is the argument of who needs to decide what the audience is for that spin so they know how to build that spin. And the further argument (or the one that started it all) is if one spin has a package that is built one way, but needs to be changed so that spin works, but another spin also needs that package but needs it built this way (and the two ways collide) and therein lies the problem.
My 2 cents, is choices can be good, but too many and they become confusing and/or problems and you spread yourself way too thin. You work on so many, that you have lots of OK things. But if you limit your choices to just a few, then you have a few *great* things because you can concentrate more on the fewer choices.
Anyway, prolly screwed this up so just delete it if it's terrible haha.
Toshio Kuratomi (a.badger@gmail.com) said:
Would that mean that users who don't start with one of these 'products' get to magically try and choose which implementation of which they want? Perhaps even mix and match, leaving QA and the developers to sort out the results.
Nope.
Users get a Product. That product has made choices about what packageset they receive. Mixing and matching of implementations is done at the level before the end-user. The Project can find ways to make this saner without going all the way to "if you conflict with the target audience your vision is not valuable here."
... and the people who chose the net install get what, exactly?
Furthermore, you then leave 'downstream' higher-level packages and applications having to, for example, code to PolicyKit0, PolicyKit1, or consolehelper, depending on what each 'product' use case might use. Or, having to build their python extensions simultaneously for python2.4, python2.6, and python3.0. These sorts of things would be extremely painful for developers, and would bloat the QA matrix excessively.
Also no.
You think that you can make people work on things they don't have an interest in? I certainly don't. Let's look at PolicyKit0 and PolicyKit1. KDE has one or two apps that uses PolicyKit0, Gnome has many apps that use PolicyKit1. People concerned with Gnome are packaging PolicyKit1. KDE SIG volunteers to package PolicyKit0 for their apps' consumption. Do the gnome apps have to support building with PolicyKit0? no. Do the KDE apps have to support building with PolicyKit1? no. You have people doing the work they need to in order to realize their vision.
Sure, and then if you run a GNOME app on KDE, you get what, exactly? If you have a non-GNOME, non-KDE app, which do you choose to support? By letting each desktop choose their own environment, you make things worse for anyone that has to support both.
Bill
On Tue, Feb 2, 2010 at 12:05 PM, Toshio Kuratomi a.badger@gmail.com wrote: <snip>
I think that the Fedora Project's target audience needs to be people who want to work on open source operating systems. If you want to market the Fedora Project, that's the audience that needs to be addressed.
If you want to market a physical product, like the Fedora Desktop Spin, then that should be a decision made below the Board level. Making a decision about the target audience of the various distributions that we have limits the choices of the people who want to work on open source operating systems. Making a target audience decision at the SIG level widens the choices as marketing/artistic/documentation/etc people can choose which audiences they want to address via which medium.
-Toshio
<snip>
+1 Toshio, well said.
-AdamM
On Tue, Feb 02, 2010 at 01:05:22PM -0500, Toshio Kuratomi wrote:
I think that the Fedora Project's target audience needs to be people who want to work on open source operating systems. If you want to market the Fedora Project, that's the audience that needs to be addressed.
I don't think this works. Treating Fedora in this way effectively means that the Fedora project exists in order to facilitate derivatives. This is clearly workable (see the relationship between Debian and Ubuntu, for instance), but you then go on to say:
If you want to market a physical product, like the Fedora Desktop Spin, then that should be a decision made below the Board level. Making a decision about the target audience of the various distributions that we have limits the choices of the people who want to work on open source operating systems. Making a target audience decision at the SIG level widens the choices as marketing/artistic/documentation/etc people can choose which audiences they want to address via which medium.
And this doesn't work. Spins don't have the freedom to customise packages in the way that full-scale derivatives do, which means that they need to work directly on Fedora. And, obviously, what it's appropriate to do to the Fedora package set depends on who we want Fedora to be for.
On Tue, Feb 2, 2010 at 1:04 PM, Matthew Garrett mjg59@srcf.ucam.org wrote:
On Tue, Feb 02, 2010 at 01:05:22PM -0500, Toshio Kuratomi wrote:
I think that the Fedora Project's target audience needs to be people who want to work on open source operating systems. If you want to market the Fedora Project, that's the audience that needs to be addressed.
I don't think this works. Treating Fedora in this way effectively means that the Fedora project exists in order to facilitate derivatives. This is clearly workable (see the relationship between Debian and Ubuntu, for instance), but you then go on to say:
If you want to market a physical product, like the Fedora Desktop Spin, then that should be a decision made below the Board level. Making a decision about the target audience of the various distributions that we have limits the choices of the people who want to work on open source operating systems. Making a target audience decision at the SIG level widens the choices as marketing/artistic/documentation/etc people can choose which audiences they want to address via which medium.
And this doesn't work. Spins don't have the freedom to customise packages in the way that full-scale derivatives do, which means that they need to work directly on Fedora. And, obviously, what it's appropriate to do to the Fedora package set depends on who we want Fedora to be for.
<snip>
Your example doesn't work, Xubuntu is still bound to the package set in the Ubuntu repositories in the same sense that the Xfce Spin is bound to the package set in the Fedora repositories. The difference is that we understand that the Xfce Spin isn't a fork and shouldn't be presented as a completely separate project. The "silos of community" that have spawned out of the "Lets make a ${foo}buntu for every possible value of $foo" in my opinion has divided the Ubuntu project from a contributor stand point.
-AdamM
On Tue, 2010-02-02 at 13:15 -0600, Adam Miller wrote:
Your example doesn't work, Xubuntu is still bound to the package set in the Ubuntu repositories in the same sense that the Xfce Spin is bound to the package set in the Fedora repositories. The difference is that we understand that the Xfce Spin isn't a fork and shouldn't be presented as a completely separate project.
This only works if your special interest groups are completely segregated, and that they agree on how the shared packages work. But what if you don't? What if the Desktop (gnome) set wants the newest versions of PolicyKit, of NetworkManager, of DeviceKit, etc.. but the KDE group doesn't have any software that works with those, and instead wants the older versions they do have software to work with. How do you resolve this conflict of interest? Who wins?
Once the differences within the special interest groups moves beyond just different subsets of packages, and actually starts to focus on the behavior of the shared packages things break down.
On Tue, Feb 2, 2010 at 1:43 PM, Jesse Keating jkeating@redhat.com wrote:
On Tue, 2010-02-02 at 13:15 -0600, Adam Miller wrote:
Your example doesn't work, Xubuntu is still bound to the package set in the Ubuntu repositories in the same sense that the Xfce Spin is bound to the package set in the Fedora repositories. The difference is that we understand that the Xfce Spin isn't a fork and shouldn't be presented as a completely separate project.
This only works if your special interest groups are completely segregated, and that they agree on how the shared packages work. But what if you don't? What if the Desktop (gnome) set wants the newest versions of PolicyKit, of NetworkManager, of DeviceKit, etc.. but the KDE group doesn't have any software that works with those, and instead wants the older versions they do have software to work with. How do you resolve this conflict of interest? Who wins?
<snip>
I thought the whole point of having a "Default" is saying that it is what would "win" and I was under the impression that had all been sorted out by this point.
-AdamM
On Tue, 2010-02-02 at 14:16 -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 1:43 PM, Jesse Keating jkeating@redhat.com wrote:
On Tue, 2010-02-02 at 13:15 -0600, Adam Miller wrote:
Your example doesn't work, Xubuntu is still bound to the package set in the Ubuntu repositories in the same sense that the Xfce Spin is bound to the package set in the Fedora repositories. The difference is that we understand that the Xfce Spin isn't a fork and shouldn't be presented as a completely separate project.
This only works if your special interest groups are completely segregated, and that they agree on how the shared packages work. But what if you don't? What if the Desktop (gnome) set wants the newest versions of PolicyKit, of NetworkManager, of DeviceKit, etc.. but the KDE group doesn't have any software that works with those, and instead wants the older versions they do have software to work with. How do you resolve this conflict of interest? Who wins?
<snip>
I thought the whole point of having a "Default" is saying that it is what would "win" and I was under the impression that had all been sorted out by this point.
Not at all. That's why the target audience discussion continues. The "Default" we have now is a product of inertia from Red Hat Linux, status quo of continuing that inertia, and the chaos of every packager having their own target audience in mind and designing packages within their influence for it, regardless of other consumers of their package.
What we also have is two distinct arguments:
1) What is our target audience?
and
2) Do we need a target audience
These distinct arguments are often rolled into the same discussion and go nowhere.
On Tue, Feb 02, 2010 at 11:43:32AM -0800, Jesse Keating wrote:
On Tue, 2010-02-02 at 13:15 -0600, Adam Miller wrote:
Your example doesn't work, Xubuntu is still bound to the package set in the Ubuntu repositories in the same sense that the Xfce Spin is bound to the package set in the Fedora repositories. The difference is that we understand that the Xfce Spin isn't a fork and shouldn't be presented as a completely separate project.
This only works if your special interest groups are completely segregated, and that they agree on how the shared packages work. But what if you don't? What if the Desktop (gnome) set wants the newest versions of PolicyKit, of NetworkManager, of DeviceKit, etc.. but the KDE group doesn't have any software that works with those, and instead wants the older versions they do have software to work with. How do you resolve this conflict of interest? Who wins?
An interesting note here is that target audience is of no use in deciding this. KDE and GNOME aim for the same target audiences but have different ideas of how to reach them. The details that moving forward or staying back with these libraries and services would entail is not about target audience but more about Fedora being being on the leading edge of technology. Even that isn't a good fit for making a decision as there's no demand that everything move forward -- any app might be on the leading edge in one technology even when some other pieces of its underpinings are more stodgy.
-Toshio
On Tue, 2010-02-02 at 17:07 -0500, Toshio Kuratomi wrote:
An interesting note here is that target audience is of no use in deciding this. KDE and GNOME aim for the same target audiences but have different ideas of how to reach them. The details that moving forward or staying back with these libraries and services would entail is not about target audience but more about Fedora being being on the leading edge of technology. Even that isn't a good fit for making a decision as there's no demand that everything move forward -- any app might be on the leading edge in one technology even when some other pieces of its underpinings are more stodgy.
I think if you look closer at KDE vs Gnome you will find a difference in the target audience. One one hand you have people who want to use their computer to do tasks, and have the operating system stay out of the way, and on the other hand you have people who want to be able to configure their computer to work in a very specific to them way and have the operating system allow them to make these configuration choices.
I'm blowing up a subtle difference, but it's that subtle difference that is very clear in Gnome vs KDE, and it becomes an interesting point. Do we, the project wish to default to a product that targets the people who just want to use their computers easily without tweaking every last detail, or do we wish to default to a product that caters to the tinkerers and the tweakers and those that wish to have total control over how their system works?
But you are right, not only are we picking a target audience, but we're also picking a route to that target audience.
On Tue, Feb 02, 2010 at 06:29:33PM -0800, Jesse Keating wrote:
On Tue, 2010-02-02 at 17:07 -0500, Toshio Kuratomi wrote:
An interesting note here is that target audience is of no use in deciding this. KDE and GNOME aim for the same target audiences but have different ideas of how to reach them. The details that moving forward or staying back with these libraries and services would entail is not about target audience but more about Fedora being being on the leading edge of technology. Even that isn't a good fit for making a decision as there's no demand that everything move forward -- any app might be on the leading edge in one technology even when some other pieces of its underpinings are more stodgy.
I think if you look closer at KDE vs Gnome you will find a difference in the target audience. One one hand you have people who want to use their computer to do tasks, and have the operating system stay out of the way, and on the other hand you have people who want to be able to configure their computer to work in a very specific to them way and have the operating system allow them to make these configuration choices.
This may be true on its own but we need to be careful of setting it up as a dichotomy because it becomes false when put in that context. I want my computer to stay out of my way and let me do things. Yet I use KDE because KDE stays out of my way much better than Gnome. So no matter what the desktop environment targeted, KDE gave me the option to have the OS stay out of my way whereas gnome forced me to fight the OS when I just wanted to get my work done.
I'm blowing up a subtle difference, but it's that subtle difference that is very clear in Gnome vs KDE, and it becomes an interesting point. Do we, the project wish to default to a product that targets the people who just want to use their computers easily without tweaking every last detail, or do we wish to default to a product that caters to the tinkerers and the tweakers and those that wish to have total control over how their system works?
So there's a false thought in here. Just because you have a configuration option doesn't mean you have to change it. It just means you can change it. If I install KDE and don't touch any configuration options I have a usable, general purpose desktop that probably fits me as well as gnome. If I were the kind of person that hated touching configuration, I'd be in the same boat whether I used KDE or GNOME.
But you are right, not only are we picking a target audience, but we're also picking a route to that target audience.
<nod>
-Toshio
Toshio Kuratomi wrote:
This may be true on its own but we need to be careful of setting it up as a dichotomy because it becomes false when put in that context. I want my computer to stay out of my way and let me do things. Yet I use KDE because KDE stays out of my way much better than Gnome. So no matter what the desktop environment targeted, KDE gave me the option to have the OS stay out of my way whereas gnome forced me to fight the OS when I just wanted to get my work done.
What you found is an inherent problem with lack of options: if what you need to do is not covered by the unchangeable defaults, you computer stands in your way in the most annoying possible way.
So there's a false thought in here. Just because you have a configuration option doesn't mean you have to change it. It just means you can change it. If I install KDE and don't touch any configuration options I have a usable, general purpose desktop that probably fits me as well as gnome. If I were the kind of person that hated touching configuration, I'd be in the same boat whether I used KDE or GNOME.
Right, and that's exactly why not offering options doesn't make sense. Nobody forces anybody to use the KDE options. Despite rumors to the contrary, KDE carefully choses sane defaults. But you still get the option to override them if they don't match what you like or what you need.
Kevin Kofler
On Tue, Feb 02, 2010 at 06:29:33PM -0800, Jesse Keating wrote:
On Tue, 2010-02-02 at 17:07 -0500, Toshio Kuratomi wrote:
An interesting note here is that target audience is of no use in deciding this. KDE and GNOME aim for the same target audiences but have different ideas of how to reach them. The details that moving forward or staying back with these libraries and services would entail is not about target audience but more about Fedora being being on the leading edge of technology. Even that isn't a good fit for making a decision as there's no demand that everything move forward -- any app might be on the leading edge in one technology even when some other pieces of its underpinings are more stodgy.
I think if you look closer at KDE vs Gnome you will find a difference in the target audience. One one hand you have people who want to use their computer to do tasks, and have the operating system stay out of the way, and on the other hand you have people who want to be able to configure their computer to work in a very specific to them way and have the operating system allow them to make these configuration choices.
There are many more reasons to choose a particular desktop system. Netbook users will use whatever is usable with the screen resolution - I found old fashioned window managers, eg Enlightenment that work with virtual screens bigger than monitor resolution far superior to anything else.
Btw it stinks that such useful features that used to be easy 10 years ago no longer work or are hard to achieve even for experienced users. Anyone running Gnome programs over ssh sessions? Much fun.
Richard
On Tue, Feb 02, 2010 at 01:15:29PM -0600, Adam Miller wrote:
Your example doesn't work, Xubuntu is still bound to the package set in the Ubuntu repositories in the same sense that the Xfce Spin is bound to the package set in the Fedora repositories. The difference is that we understand that the Xfce Spin isn't a fork and shouldn't be presented as a completely separate project. The "silos of community" that have spawned out of the "Lets make a ${foo}buntu for every possible value of $foo" in my opinion has divided the Ubuntu project from a contributor stand point.
And Xubuntu and Kubuntu are both of woeful quality because the entire focus of the core Ubuntu developer base is on Ubuntu, with no concern about the effects of that on derivatives. Ubuntu is better than Debian because they were able to focus their development efforts on a specific goal. Xubuntu is worse than Debian because they're forced to cope with architectural changes made without reference to them.
On Tue, Feb 2, 2010 at 2:01 PM, Matthew Garrett mjg59@srcf.ucam.org wrote: <snip>
Ubuntu is better than Debian
<snip>
If you honestly believe that, I have pitty on you.
-AdamM
On Tue, Feb 02, 2010 at 02:19:35PM -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:01 PM, Matthew Garrett mjg59@srcf.ucam.org wrote:
<snip> >Ubuntu is better than Debian <snip>
If you honestly believe that, I have pitty on you.
For the market they're aiming at? I don't think there's any doubt at all.
On Tue, 2010-02-02 at 20:30 +0000, Matthew Garrett wrote:
On Tue, Feb 02, 2010 at 02:19:35PM -0600, Adam Miller wrote:
On Tue, Feb 2, 2010 at 2:01 PM, Matthew Garrett mjg59@srcf.ucam.org wrote:
<snip> >Ubuntu is better than Debian <snip>
If you honestly believe that, I have pitty on you.
For the market they're aiming at? I don't think there's any doubt at all.
-- Matthew Garrett | mjg59@srcf.ucam.org
From the rapid spread of not just users, but site installs, and even people buying products, Ubuntu has been a very strong success. Book publishers don't like to publish anything Linux unless it has Ubuntu in the title. Why? Ubuntu books outsell Fedora / Red Hat and SuSE counterparts combined, by a long shot. They have been very successful in determining their focus and attacking it with disregard for anything else and achieving massive penetration into their target. We cannot say the same.
On Tue, Feb 2, 2010 at 9:28 AM, Adam Miller maxamillion@fedoraproject.org wrote:
Hello all, I wanted to bring a few things up and I wanted to bring them up on devel@lists.fp.o because this is where most people spend their time.
First off: "Does letting thousands of contributors do what they want have a negative impact on our OS? (Mike)"[0]
- I would prefer that this be rephrased to a quote I read that
originated from John Rose (inode0) "isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" and I would prefer a focus be turned towards something like "Why was that the result of doing something that is essentially chaotic?" .... I guess my main question is: "Why are we fixing something that isn't broken?"
Second: "The Board has been working on defining a target audience for Fedora. In response to this, some people feel that Fedora should allow sub-groups to define their own target audience"[1]
- I don't entirely understand this, don't SIGs or (sub groups)
essentially exist purely because there is some target audience? Clarification on this not would be appreciated.
Now, we come to the part that I feel is going to be viewed as a touchy subject by many..... Why are there words like "letting" and "allow" being thrown around so often? I understand there are guidelines and policies for certain things of technical or legal nature in Fedora, but it feels a little like there is an attempt here to dictate how myself, as well as all others, spend their time contributing to The Fedora Project.
I don't think the goal is to dictate how the community spends their time. I'd like to think of it as more of a starting point for attracting new contributors; "here are our broader goals, how you can help, and why you should invest your time in our community, rather than elsewhere."
The open-source concept isn't exactly niche anymore; there is an increasing number of people out there who want to contribute, but unlike days long since past where people would contribute because something was broken or non-existent, they're just here to contribute, learn how community works. Finding those potential contributors and giving them a pathway to success is essential to Fedora's long-term growth and reaching our goals (whatever those may be).
[0] https://fedoraproject.org/wiki/Unfinished_Board_issues [1] https://fedoraproject.org/wiki/User:Walters/SpinsSigsRemixes_TargetAudience
I would just like to know other contributors thoughts on these topics.
Thank you for your time, -AdamM
-- http://maxamillion.googlepages.com
() ascii ribbon campaign - against html e-mail /\ www.asciiribbon.org - against proprietary attachments -- devel mailing list devel@lists.fedoraproject.org https://admin.fedoraproject.org/mailman/listinfo/devel
Adam Miller said the following on 02/02/2010 08:28 AM Pacific Time:
Hello all, I wanted to bring a few things up and I wanted to bring them up on devel@lists.fp.o because this is where most people spend their time.
Please consider that most of the things you are referring to have not been finalized.
First off: "Does letting thousands of contributors do what they
want have a negative impact on our OS? (Mike)"[0]
- I would prefer that this be rephrased to a quote I read that
originated from John Rose (inode0) "isn't it amazing how thousands of contributors doing whatever they want created such a spectacular OS?" and I would prefer a focus be turned towards something like "Why was that the result of doing something that is essentially chaotic?" .... I guess my main question is: "Why are we fixing something that isn't broken?"
We are attempting to address that question too: https://fedoraproject.org/wiki/User:Poelstra/Importance_of_strategy
Second: "The Board has been working on defining a target audience
for Fedora. In response to this, some people feel that Fedora should allow sub-groups to define their own target audience"[1]
- I don't entirely understand this, don't SIGs or (sub groups)
essentially exist purely because there is some target audience? Clarification on this not would be appreciated.
These are *working drafts and in process documents* all the in spirit of transparency. It would be more helpful to these discussions to get clarification on advisory-board first rather than conclude that the board has run off the rails by using words like "letting" and "allowing" in documents that are brainstorming and unfinished.
I guess what I'm trying to get at here is that we've posted all meeting recaps to advisory-board list and there has been ZERO discussion or inquiries there. We specifically asked for feedback to the original list of "unanswered questions" on advisory-board. Is there a particular reason you did not respond there?
http://lists.fedoraproject.org/pipermail/advisory-board/2010-January/007886....
I feel like you are discounting the board's efforts and misdirecting a lot energy by launching a new thread here with your "concerns" before first getting clarification on advisory-board.
Now, we come to the part that I feel is going to be viewed as a
touchy subject by many..... Why are there words like "letting" and "allow" being thrown around so often? I understand there are guidelines and policies for certain things of technical or legal nature in Fedora, but it feels a little like there is an attempt here to dictate how myself, as well as all others, spend their time contributing to The Fedora Project.
Great questions. Why not ask the original authors first what their core motivations and intentions are and if these are their final conclusions in the original forum they were presented? This would be more constructive for everyone involved.
John
On Wed, Feb 3, 2010 at 9:36 AM, John Poelstra poelstra@redhat.com wrote: <snip>
These are *working drafts and in process documents* all the in spirit of transparency. It would be more helpful to these discussions to get clarification on advisory-board first rather than conclude that the board has run off the rails by using words like "letting" and "allowing" in documents that are brainstorming and unfinished.
I guess what I'm trying to get at here is that we've posted all meeting recaps to advisory-board list and there has been ZERO discussion or inquiries there. We specifically asked for feedback to the original list of "unanswered questions" on advisory-board. Is there a particular reason you did not respond there?
http://lists.fedoraproject.org/pipermail/advisory-board/2010-January/007886....
I feel like you are discounting the board's efforts and misdirecting a lot energy by launching a new thread here with your "concerns" before first getting clarification on advisory-board.
I'm not on some crusade to undermine the Board if that's what you think, I'm honestly looking for clarification but not only from those involved in the Board but the community as well and both are located here on this list. I don't see why it matters where the questions are asked, just so long as they are asked.
As far as replying to the advisory-board mailing list first, I will be sure to do so in the future. I apparently forgot my place in the hierarchy for a moment. Apologies for not following protocol.
-AdamM
Adam Miller said the following on 02/03/2010 08:02 AM Pacific Time:
On Wed, Feb 3, 2010 at 9:36 AM, John Poelstrapoelstra@redhat.com wrote:
<snip> > These are *working drafts and in process documents* all the in spirit of > transparency. It would be more helpful to these discussions to get > clarification on advisory-board first rather than conclude that the > board has run off the rails by using words like "letting" and "allowing" > in documents that are brainstorming and unfinished. > > I guess what I'm trying to get at here is that we've posted all meeting > recaps to advisory-board list and there has been ZERO discussion or > inquiries there. We specifically asked for feedback to the original > list of "unanswered questions" on advisory-board. Is there a particular > reason you did not respond there? > > http://lists.fedoraproject.org/pipermail/advisory-board/2010-January/007886.html > > I feel like you are discounting the board's efforts and misdirecting a > lot energy by launching a new thread here with your "concerns" before > first getting clarification on advisory-board. >
I'm not on some crusade to undermine the Board if that's what you think, I'm honestly looking for clarification but not only from those involved in the Board but the community as well and both are located here on this list. I don't see why it matters where the questions are asked, just so long as they are asked.
Thanks for your clarification. I think it is great to ask questions, I ask a lot of them myself. I question how productive it is to all of us though, to ask questions if the starting point of those questions is incorrect.
My sense here was that a few words on a wiki page struck you the wrong way so instead of going to the people that wrote them by asking, "Hey, what do you guys mean? These _______ things concern me for these reasons." It was first asked instead to a mailing list that didn't write them :).
I specifically requested feedback on advisory-board for this very purpose and received no responses. Is there something I could have done better on advisory-board list to engage the people that have participated so freely here?
As far as replying to the advisory-board mailing list first, I will be sure to do so in the future. I apparently forgot my place in the hierarchy for a moment. Apologies for not following protocol.
I didn't mean to imply that you'd broken any rules. I thought we might be able to have a more productive discussion if we had an accurate starting point.
John
On Wed, Feb 3, 2010 at 10:26 AM, John Poelstra poelstra@redhat.com wrote: <snip>
Thanks for your clarification. I think it is great to ask questions, I ask a lot of them myself. I question how productive it is to all of us though, to ask questions if the starting point of those questions is incorrect.
My sense here was that a few words on a wiki page struck you the wrong way so instead of going to the people that wrote them by asking, "Hey, what do you guys mean? These _______ things concern me for these reasons." It was first asked instead to a mailing list that didn't write them :).
I specifically requested feedback on advisory-board for this very purpose and received no responses. Is there something I could have done better on advisory-board list to engage the people that have participated so freely here?
<snip>
Ah, ok. Makes sense. Thanks for clarification on that.
<snip>
I didn't mean to imply that you'd broken any rules. I thought we might be able to have a more productive discussion if we had an accurate starting point.
<snip>
My mistake, I must have taken it out of context or incorrectly. Apologies.
-AdamM
On Wed, Feb 3, 2010 at 10:26 AM, John Poelstra poelstra@redhat.com wrote:
Adam Miller said the following on 02/03/2010 08:02 AM Pacific Time:
I'm not on some crusade to undermine the Board if that's what you think, I'm honestly looking for clarification but not only from those involved in the Board but the community as well and both are located here on this list. I don't see why it matters where the questions are asked, just so long as they are asked.
Thanks for your clarification. I think it is great to ask questions, I ask a lot of them myself. I question how productive it is to all of us though, to ask questions if the starting point of those questions is incorrect.
While I understand your point I think (reading too much into draft remarks with possibly not the full context of the surrounding discussions) I do think after all this time there are still a number of people in the community (I am one of them) who aren't convinced that the board isn't going down an unproductive path founded in assumptions of a community structure that doesn't really exist.
I believe that what fundamentally makes the Fedora Project a great place to be is that it is an open community where the participants share a group of core values that guide them both individually and collectively toward an unwritten end that is worth pursuing and I see danger ahead in trying to write that ending in advance because that short-circuits the evolving direction the project gets from the collective wisdom of its contributors.
I wonder how widely that belief is held in the community?!
My sense here was that a few words on a wiki page struck you the wrong way so instead of going to the people that wrote them by asking, "Hey, what do you guys mean? These _______ things concern me for these reasons." It was first asked instead to a mailing list that didn't write them :).
I can't speak for Adam here, but to me it isn't a few words on a wiki page causing the concern, those words reinforce the concern. The board has a really difficult task when it comes to its leadership role. Since it doesn't have much structural authority to impose its will on contributors it requires that the board make a case that is compelling to the contributors so that they internalize and adopt it as part of what they do. If contributors won't do that, then stating our target audience is X will fall on deaf ears.
While I've not been convinced that defining a target audience is remotely a good idea, I know from talking to a lot of people in the community that *they* do think it is. So don't be too discouraged, the folks with doubts are more likely to jump up and down than the folks who agree.
I specifically requested feedback on advisory-board for this very purpose and received no responses. Is there something I could have done better on advisory-board list to engage the people that have participated so freely here?
Perhaps that indicates that the advisory-board list wasn't the best place to ask.
John
On Wed, 2010-02-03 at 12:54 -0600, inode0 wrote:
I believe that what fundamentally makes the Fedora Project a great place to be is that it is an open community where the participants share a group of core values that guide them both individually and collectively toward an unwritten end that is worth pursuing
Perhaps the problem is we don't all agree on those core sets of values, or how those values should guide us to what unwritten end. Or we suspect we don't agree because so much of it is unwritten.
If the assumption is that we all share these values, what are they? The four F's? Those are just vague enough to be practically meaningless in this context.
On Wed, Feb 3, 2010 at 3:26 PM, Jesse Keating jkeating@redhat.com wrote:
On Wed, 2010-02-03 at 12:54 -0600, inode0 wrote:
I believe that what fundamentally makes the Fedora Project a great place to be is that it is an open community where the participants share a group of core values that guide them both individually and collectively toward an unwritten end that is worth pursuing
Perhaps the problem is we don't all agree on those core sets of values, or how those values should guide us to what unwritten end. Or we suspect we don't agree because so much of it is unwritten.
We are about to fall off the edge of the philosophical cliff now. I really don't analyze how my values guide my actions. I approach the check-out counter behind a little old lady. I could speed up and cut in front of her, I could slow down and let her go first. I make a decision which I believe is formed in large part by my values without thinking about them.
If the assumption is that we all share these values, what are they? The four F's? Those are just vague enough to be practically meaningless in this context.
Enumerating the values with surgical precision is meaningless too if you want it to lead to an idea of what the Fedora distribution will look like in 5 years. It just doesn't work that way.
John
On Wed, 2010-02-03 at 16:25 -0600, inode0 wrote:
On Wed, Feb 3, 2010 at 3:26 PM, Jesse Keating jkeating@redhat.com wrote:
On Wed, 2010-02-03 at 12:54 -0600, inode0 wrote:
I believe that what fundamentally makes the Fedora Project a great place to be is that it is an open community where the participants share a group of core values that guide them both individually and collectively toward an unwritten end that is worth pursuing
Perhaps the problem is we don't all agree on those core sets of values, or how those values should guide us to what unwritten end. Or we suspect we don't agree because so much of it is unwritten.
We are about to fall off the edge of the philosophical cliff now. I really don't analyze how my values guide my actions. I approach the check-out counter behind a little old lady. I could speed up and cut in front of her, I could slow down and let her go first. I make a decision which I believe is formed in large part by my values without thinking about them.
If the assumption is that we all share these values, what are they? The four F's? Those are just vague enough to be practically meaningless in this context.
Enumerating the values with surgical precision is meaningless too if you want it to lead to an idea of what the Fedora distribution will look like in 5 years. It just doesn't work that way.
John
Since we can't act as a single hive mind, we have to come to some sort of agreement, and to do so, we need guidelines rather than "whatever I feel like today". You seem to be sidestepping any point that has to do with a conflict within the project.
On Wed, Feb 3, 2010 at 4:35 PM, Jesse Keating jkeating@redhat.com wrote:
On Wed, 2010-02-03 at 16:25 -0600, inode0 wrote: Since we can't act as a single hive mind, we have to come to some sort of agreement, and to do so, we need guidelines rather than "whatever I feel like today". You seem to be sidestepping any point that has to do with a conflict within the project.
Guilty as charged. The Board, Steering Committees, various guidelines exist and have been used to resolve conflicts for years, right?
This is about more than conflict resolution, isn't it? This is about giving direction to the efforts of those working on the distribution, isn't it? If it isn't, someone should make that very clear now.
John
On Wed, 2010-02-03 at 17:05 -0600, inode0 wrote:
Guilty as charged. The Board, Steering Committees, various guidelines exist and have been used to resolve conflicts for years, right?
This is about more than conflict resolution, isn't it? This is about giving direction to the efforts of those working on the distribution, isn't it? If it isn't, someone should make that very clear now.
Outside of a very very few people, we can only suggest what people work on. We can't dictate what people volunteer their time for. We can however say what kind of changes and work would be seen as favorable and likely to find other like minded people to help out with, vs not. We can say what we'd /like/ to see marketing target, and what we'd /like/ to see QA focus efforts on.
I see the target audience discussions as both conflict resolution and as charting a course for where we'd /like/ to see the project go.
John Poelstra wrote:
I guess what I'm trying to get at here is that we've posted all meeting recaps to advisory-board list and there has been ZERO discussion or inquiries there. We specifically asked for feedback to the original list of "unanswered questions" on advisory-board. Is there a particular reason you did not respond there?
Probably because it's yet another mailing list most maintainers don't read?
Kevin Kofler
Kevin Kofler (kevin.kofler@chello.at) said:
John Poelstra wrote:
I guess what I'm trying to get at here is that we've posted all meeting recaps to advisory-board list and there has been ZERO discussion or inquiries there. We specifically asked for feedback to the original list of "unanswered questions" on advisory-board. Is there a particular reason you did not respond there?
Probably because it's yet another mailing list most maintainers don't read?
The devel list is for development of the distribution; advisory-board is for project-wide direction. While there's certainly overlap with the devel list for this, advisory-board is the far more appropriate place for issues of project-wide direction.
Bill