On Fri, Jan 19, 2018 at 09:00:17PM +0000, Stephen Gallagher wrote: Clearly, the default needs to be plymouth-theme-hot-dog. -- John.

On Mon, Jan 22, 2018 at 1:50 PM Bastien Nocera <bnocera(a)redhat.com&gt; wrote: I think you completely missed the part where I pointed out that during offline updates, this splash screen is visible for *minutes*. I can buy your argument for initial boot-up. But not when a user is seeing something for what is guaranteed to be a longer period of time.

On Thu, Jan 25, 2018 at 05:18:17PM +0100, Mathieu Bridon wrote: I don't think that's what's meant at all, no. Fedora is not a generic, white-box operating system. People see our logo and have positive, happy thoughts. Let's capitalize on that. I generally hate being advertised at, but when my Android phone boots, I don't think "oh %#$@%@, the capitalist overlords have me trapped". I think "hey, that's a pretty nifty logo representing the device I chose to purchase". We've been talking about establishing some visual cues to help build Fedora Workstation identity for a long time. Maybe Stephen's suggestion isn't the best thing we can come up with, but c'mon, everybody. Please come up with *something*. -- Matthew Miller <mattdm(a)fedoraproject.org&gt; Fedora Project Leader

On Thu, Feb 01, 2018 at 10:51:21AM -0500, Bastien Nocera wrote: Yes, I see a Google logo during bootup. It used to be an Android logo but apparently now it's just "Powered by Android" at the bottom. Wait, what? I have seen no changes at all, except the lighter top bar, and that doesn't really say "Fedora" to me in any way. The details panel? Are you kidding? I think this is a good idea, but it's hardly sufficent. I'm not sure it really fits with the direction Workstation is going — I notice Terminal isn't even in the default Dash. -- Matthew Miller <mattdm(a)fedoraproject.org&gt; Fedora Project Leader

Ryan Lerch has been working on a comprehensive Fedora Workstation visual identity proposal for a while. I will ping him about it, as maybe that will let us have a more concrete discussion about how we differentiate on the visual side. Yes, we also need to find ways to differentiate on the functionality, but as we all know that is quite challenging in an open source setting, but I have gotten some suggestions for possible projects we could look at in that direction. Christian On Thu, Feb 1, 2018 at 12:14 PM, Matthias Clasen <mclasen(a)redhat.com&gt; wrote:

----- Original Message ----- There were 2 things in this list. That's fine, we're frustrated as well. Matthias, Rishi and myself are the only people who added any information to the "Downstream Branding" page on the GNOME wiki, and nobody asked us to add anything more. The discussions usually go something with you proposing something, a lot of folks working upstream thinking it's not the best option, with more or less vehemence, and the functionality getting in anyway. It's not. It just makes things worse. It's like we never had those discussions in the past. Keep asking the same thing, and that one time when we're not looking, you'll slip this change past us. I'm pretty sure that the omni-presence of a logo isn't something that got explicitly requested by designers at Endless. Or the Ubuntu designers requesting the UI to be mostly orange. That's probably fine for not everyone using terminals all the time, that means we're doing something right. But I'd rather we took a considered and holistic approach to branding rather than slapping logos where there's some blank space.

----- Original Message ----- The list isn't a checklist of all the downstream branding we want to implement, it's a list of what different OSes use. Which is exactly why I don't think we want to implement poorly thought out branding, or not look into the existing branding to see how well it integrates in our goals of being recognisable as Fedora, with or without branding. Then again, I think this is our goals, but they were never explicitly defined, and we would gain a lot in this discussion by defining them before making decisions to accomplish them. Important doesn't mean necessary, or thought out.

On Thu, Feb 01, 2018 at 09:29:54PM +0000, Peter Robinson wrote: I wouldn't say that anybody, or at least most people, flat out thinks that a distributor shouldn't be able to differentiate by having their own visual identity. However, there might be certain things that some people might like less, just as there might be things that might be liked more. The boot splash, default wallpaper, network hostname, log in screen, the Settings panel, a logo on the desktop, and all those things listed on https://wiki.gnome.org/Design/OS/DownstreamBranding - they are all different things that can evoke different responses from different people. I mean, isn't Fedora shipping with the word "fedora" overlaid on the default wallpaper?

On Thu, Feb 01, 2018 at 12:43:40PM -0500, Matthew Miller wrote: This sounds a lot like how we got the overlaid Fedora logo on the wallpaper. :) Are we going to withdraw that one? I thought people stopped liking it. I just installed a fresh new Fedora 27 VM and it was there. Or am I just confused?

On Thu, Feb 1, 2018 at 11:43 AM, Matthew Miller <mattdm(a)fedoraproject.org&gt; wrote: Let's start with plymouth. The current charge theme just does not look very good, and putting our logo on a boot process that does not look very good is not in our best interest. I don't think the spinfinity theme is what we are looking for, either. So unless a new, better-looking Fedora plymouth theme materializes, we should definitely switch to the upstream spinner theme. What would a Fedora plymouth theme look like? Let's take some inspiration from RHEL's boot theme. Watch the first 10 seconds: https://www.youtube.com/watch?v=AsIqlwjNf4M Note that the plymouth theme transitions seamlessly into the noise background of the login screen. It matches the upstream designs quite closely [1][2], which should keep GNOME people like me happy, but with a downstream flair (the artsy 7 in the background) that should be immeditately recognizable by anyone who has seen RHEL 7 boot up before. It's not recognizable to RHEL newcomers, since there's no logo, but it leaves you thinking "wow! that looks good" up until you reach the login screen with the RHEL logo. Then you think "wow! Red Hat looks good." The entire this is extremely well-executed. IMO this is exactly the effect we should be going for. Can we get something like that for Fedora? An infinity etching on the noise background, perhaps? A few different mockups along this line? Michael [1] https://wiki.gnome.org/Design/OS/BootOptions [2] https://wiki.gnome.org/Design/OS/LoginScreen

On Thu, Feb 1, 2018 at 9:49 PM Chris Murphy <lists(a)colorremedies.com&gt; wrote: Just to interject, as I mentioned in the original message, the boot splash is also visible for the entire time that offline updates are running (which may be for *minutes*), so that *is* an opportunity for an improved experience. I'm not trying to dictate what experience that should be, but as near as I can tell, no one is arguing that the current one is perfect. So let's hear some ideas on how to fix it.

On Thu, Feb 01, 2018 at 08:53:13PM +0000, Peter Robinson wrote: It seemed to me that Matthew was hinting that Terminal should be more prominently placed, and the response to that was whether there are actual users who wanted to use Terminal but couldn't find it because it wasn't in the default dash. But now everybody is saying that there are users who either rarely use Terminal or don't know what it is. However, that doesn't answer the question whether someone actually wanted to use it but couldn't get to it because it isn't prominently placed. :)

My mobile email clients threading is awful so I dont actually know who said the below quoted phrase, nor can I reply below it.... but the below philosophy I think is absolutely the right approach. I think it is already working as you can see in the media with the past few releases Fedora's general perception is more and more positive. where I want to know more, is how is this possible for an open source project? is this the *first* fedora f, where we get the awesome functionality first? while this is good, i do worry that folks using fedora for casual use rather than hardcore linux fans dont update so frequently for fedora having a feature first being a big deal. eg flatpak has nade it easier than ever to get the newest inkscape or gimp or whatever (sorry this is just pertinent to my little subcommunity but can be extrapolated) without updating the entire OS. are there other ways to differentiate as an open source project? is it the integration and default configs? is there a focus on that now? (i dont know.) is it in the variety and quality of apps on top of the platform? i dont know. what do you have in mind?

On Thu, Feb 1, 2018 at 3:51 PM, Chris Murphy <lists(a)colorremedies.com&gt; wrote: It's simply not true. Upstream's designs call for a well-integrated user experience where the noise background begins on the bootloader theme, continues through plymouth, and arrives at the login screen. I assume there must be technical problems that have prevented the bootloader theme from panning out, but we already have the login screen in Fedora, and the spinner plymouth theme is available just waiting to be enabled should we fail to design a quality Fedora-branded replacement for charge. https://wiki.gnome.org/Design/OS/BootOptions https://wiki.gnome.org/Design/OS/LoginScreen Michael

On Thu, Feb 01, 2018 at 07:56:28PM -0700, Chris Murphy wrote: The part where you assume that just because something doesn't originate from the *.gnome.org infrastructure or is not explicitly branded as GNOME, is somethng that GNOME didn't put any effort into building. See: https://www.gnome.org/technologies/ There are a lot of things on that page that are deliberately not kept on the *.gnome.org infrastructure and not branded as GNOME to encourage broader adoption, but are/were very much conceived, promoted, developed and moved forward by GNOME contributors to solve problems faced by GNOME. All that amounts to some very serious effort. Regardless of whether Plymouth falls into that category or not, the fact that GNOME contributors went out to work on solutions that go beyond benefitting the immediate GNOME community, does give GNOME the moral authority to have a say in how the whole user experience should be like. Also, Fedora Workstation should "take responsibility for the user's experience". It is not a matter of theming the boot splash in a certain way, and the log in screen in another way, and so on. All the pieces are, and should be treated as, part of a single coherent entity.

On Thu, Feb 1, 2018 at 9:57 PM Chris Murphy <lists(a)colorremedies.com&gt; wrote: My memory on the subject suggests that the issue was that when we first attempted this, we were still supporting a LOT of hardware that was incapable of running in modeset at early boot. As a result, the fancy-pants GRUB2 wouldn't display on those systems. Given that we are no longer shipping install media for Workstation on anything but x86_64 systems, I think that constraint is lifted and we could reconsider doing it again.

On Fri, Feb 2, 2018 at 5:29 AM, Andreas Nilsson <lists(a)andreasn.se&gt; wrote: Other distros have done this work on their own and have a better boot experience. It has everything to do with committed distro level work, and nothing to do with GNOME having some imaginary moral authority over that experience. And near as I can tell none of those distros had to go through some upstream GNOME bureaucratic process to arrive at their customized boot experience, and plainly this moral authority as applied apparently to just Fedora hasn't resulted in Fedora magically getting a better boot experience. So the moral authority argument I find unconvincing. These are compelling arguments. -- Chris Murphy

On Fri, Feb 02, 2018 at 05:42:20AM -0500, Bastien Nocera wrote: This says to me "no GNOME designer is invested in Fedora success". (Additionally, the sole entry under "Related Links" on that page loudly signals that engagement is unlikely to be productive.) That's fine. It's not their job. That's why I bring this up here, rather than focusing on the upstream page. The high level goal is: It is important to increase Fedora brand reach and recognition. A strong visual identity is an important part of this. There are several things which follow from that, but that's all there is to the basic request. -- Matthew Miller <mattdm(a)fedoraproject.org&gt; Fedora Project Leader

On Fri, Feb 02, 2018 at 08:22:44AM -0500, Bastien Nocera wrote: Sure. I'm actually working on a Fedora Marketing language document, and that's probably relevant here. Brand isn't the logo — it's the emotions and ideas associated with Fedora. The Four Foundations are part of this, and the Fedora Infinity mark nicely encapsulates the "freedom, infinity, voice" that was our earlier motto and still part of our project DNA. (See https://fedoraproject.org/wiki/Logo/History) The goals of distinguishing between Fedora and Debian and Arch in specific situations are a step or two zoomed-in, I think. We want people's daily use of their Fedora systems to tie in with the positive associations of being a Fedora user, and more than that, being part of the Fedora community. Visual cues which are special to Fedora are a powerful way to build and emphasize this connection. We want people who are not yet Fedora users to see that and be part of it. And, again, recognizable visual cues are important here. We're doing an excellent job with technology (and in putting that technology together in a way that users can consume easily with little frustration or difficulty). This is awesome. (And GNOME and the Workstation team are a large part of that. Again, awesome.) Each happy Fedora user, every time someone does something cool with Fedora, every problem someone has that we solve — these things together build up our brand. Logos and other recognizable cues help tie that all together into a package that reinforces people's sense of belonging, trust, connection, pride. I'm sorry for being terse earlier, but this is why I balked at the comments about the Details panel. That's somewhere you go when you're looking to solve a problem. It doesn't address day-to-day interaction at all. The wallpaper overlay is more... omnipresent (when it's not behind windows), but I think it feels kind of tacked-on. I do like your terminal prompt idea, especially combined with other terminal enhancements we can provide — that's great, really. I think we need some GUI equivalents to that as well. I think it's worth looking at the process Canonical just went through in adopting aspects of the Unity desktop to their new GNOME-based environment. I don't agree with all of the specific decisions and don't think we should copy them, but I think the *process* is important: as they transitioned the underlying technology, they also worked to keep the visual branding consistent. This helps them retain the trust and goodwill they've built. (And this is the same reason I've asked the design team for a consistent color scheme and overall sense for the wallpaper from release to release.) -- Matthew Miller <mattdm(a)fedoraproject.org&gt; Fedora Project Leader

----- Original Message ----- <snip> Those are helpful, but just about as high level as the existing reference. And while they're interesting, they don't disprove the original article in the "we don't need logos on everything" sense, and further show the need for a good logo, but don't explain how to use it. I guess sparsely. Anyway, I'll add that to the Wiki page. I'm tempted to just add links to Muji and Uniqlo, just to show that the products don't need branding to be recognisable, or for the brand to be loved by its users (disclosure, I'm a happy Muji customer ;)

On 02/06/2018 09:51 AM, Bastien Nocera wrote: Green is considered a cool color. [1] Some things to consider: 1) Cool colors have a more soothing and calming influence than warm colors, which tend to draw attention and arouse. Across many studies blue is cited as being the highest ranked color in terms of favor, and also as a soothing color in contrast to red which was perceived as "stimulating." [2] 2) Cool colors fade into the background; warm colors pop and are often use to signal warning or error states. Two of the main principles of the original GNOME Shell design document [3] are "Less is More" referring to a reduction of load on the user, and "The technology should act as a mediator", in part referring to the UI being the vehicle, not the destination. A desktop background shouldn't draw attention to it; it's meant to serve as a background and fade into the background and the user's work is meant to take center stage. E.g., the red default background of RHEL 5 to my knowledge elicited complaints along these lines (e.g. distracting.) May be related to findings that usage of the color red impairs task performance [4]; in my grad-level HCI program one of our professors called it the "bear detector color", eg the color of blood signals danger, as research appears to support [5, 6]. 3) Green as a color for digital display is somewhat problematic as it is the color with the poorest coverage in the sRGB color gamut, meaning across various displays it can appear very different, making it an inconsistent color for branding. (see [7], figure 4 for a diagram of the gamut) [1] http://www.handprint.com/HP/WCL/color12.html#warmcool [2] https://pdfs.semanticscholar.org/4711/624c0f72d8c85ea6813b8ec5e8abeedfb61... [3] https://people.gnome.org/~mccann/shell/design/GNOME_Shell-20091114.pdf [4] https://scholar.google.com/scholar?cluster=9535436793991630289&hl=en&... [5] http://www.tandfonline.com/doi/abs/10.1080/00140139.2014.889220 [6] https://www.sciencedirect.com/science/article/pii/S0022103109000821 [7] http://www.displaymate.com/Display_Color_Gamuts_1.htm ~m

On Tue, Feb 6, 2018 at 9:17 AM, Máirín Duffy <duffy(a)fedoraproject.org&gt; wrote: There are different reasons why each color can fail to render as expected, but the number one reason is the lack of display compensation, i.e. transform original content from sRGB into custom RGB values based on the currently selected display profile. Second guessing how a color will render wrong in a world without display compensation, is folly. We're more sensitive to hue error in blue than red, and more sensitive to hue error in red than green - and even CIELUV underestimates the problem in particular in blue. And more sensitive to hue error in achromatic color than chromatic color. So there's really all kinds of ways for color matching to fail and just avoiding greens isn't going to improve the chance of a color match. To improve color matching, UI elements need to be assumed to be sRGB, and those RGB values converted on the fly to RGB values for the display in use. *shrug* Everything else is a crap shoot. Apple's been doing this for a long time. Windows can do it as an opt in. And while a fair chunk of infrastructure work is present for this to be possible on Linux, right now GNOME would have to do this itself on its own just like any other application does, because there isn't a standard API or compositor that an application can use to opt into display compensation more automatically, or for free. -- Chris Murphy

Hi Chris, On 02/06/2018 06:09 PM, Chris Murphy wrote: LOL I was actually hoping you might jump in with a more informed background on the issue :) I'm not saying green should be avoided entirely, but if it was the main color of a background.... I'm just not sure it works out the best. One of the Fedora "Four Fs" is a bright green and we've had some issues with it. (Maybe more with printing than display tho) Maybe incorporating the "four f's" colors into one of the backgrounds could help make the background more colorful though. We've tried to stay pretty blue because the few times we tried to veer off we got a lot of negative feedback, although the latest background (jellyfish) has a blue-purple gradient and I've honestly not seen a bad thing written about it. ~m

----- Original Message ----- Right. <snip> This is all fine, but I wasn't really saying we should replace our majority white, blue and black background with one that's essentially red, or green. And I certainly wouldn't use a solid red or blue background. The crux of the problem here is that the default background image looks like it was an illustration out of a dystopian sci-fi novel, not so much whether it's a colour that'll be easily printable ;) Note that I think the current default background is really nice, it's just not appropriate I don't think.

----- Original Message ----- It might not be, but we're talking about making Fedora "warmer" and about how the background stamp/watermark isn't the best solution to adding branding to the background. I think it's a conversation to have, even as it's relevant to the bigger picture. The dichotomy isn't that one, it's about what Fedora, or other distros, wants or needs to be able to implement that differentiation, and whether it's something that upstream thinks is a good idea, and therefore makes it easier for downstreams to differentiate. For example, we went back and forth over the years as to whether the Details panel in the Settings should reflect the GNOME version and logo, or the distribution one. The WIP changes make it straight forward to brand that part of the system, and will mean less work for downstream to adapt to their needs. <snip> I thought I made that clear. It's to my taste, just not as a default background that gives me warm feelings of belonging. Which is what we were discussing in the thread.

----- Original Message ----- It's earlier in the thread, in a post from Matt.

----- Original Message ----- Matt talking about the logo said it "reinforces people's sense of belonging, trust, connection, pride." Both the quote you took out and this one make me think that a "warmer" experience was called for, though I probably took a few shortcuts in my head to get to that conclusion. Should have "shown my work" there.

----- Original Message ----- But if the front grill is the differentiator, upstream can provide a way to easily swap out this front grill, no? Say you'd want to change the default grain texture in the lock screens, to something different (maybe better, maybe more recognisable), making that easy to do could be done upstream. Completely agreed. That's out of my scope (technically), but yes, a better and more polished out-of-the-box experience (boot-up or otherwise) is definitely something we need to work on. However Matt mentioned that this ("awesome functionality", which I equate to "technical excellence", or "attention to details") wasn't enough of a differentiator. I totally understand. You can't have known about those goals before they were even set. I'm glad that you're going to be the first ones to be able to take action on something mentioned in this thread, because the actionable items are few and far between so far. If we figure out _what_ we need to change to achieve our goals, then we can have a conversation about whether changes need to happen upstream or downstream. But we're still at the point when we set down goals. Hopefully the Fedora design team can translate some of those discussions into more concrete ideas which we can discuss upstream and downstream.

Debarshi Ray wrote: I think it's best to assume good intentions here. -- Rex

Hi everyone, apologies for not chiming in on this thread sooner. I don't particularly see the need for a disagreement about high-level principles here - I think we can probably all agree that: a. Promoting awareness of Fedora [1] is a good thing b. Branding isn’t about simply slapping a logo on something - the first job is to create a fantastic product. c. Brand visuals should be applied with sensitivity - they shouldn’t be used too much, in the wrong contexts, or in ways that undermines the product they’re applied to. From my perspective, the challenge with regards to Fedora's product identity is not so much the goals and principles, but how to achieve them. There are technical and design constraints which limit what can be done, including the fact that large parts of the user experience are defined within the context of an upstream project [2]. With that said, some comments on the branding options that have been put forward in this thread: - Adding a logo to the top bar - in my view this just doesn't fit within the GNOME 3 top bar design, and a redesign of significant parts of the shell would be required to include a logo in a meaningful and coherent manner. - Boot loading and OS update installation screens - I would absolutely love to update the current theme to something more modern. While I’m fond of the existing designs for a simple spinner, I’m sure that we can work out a design that is consistent with the GNOME 3 look and feel and still promotes downstream brand awareness. My understanding is that there’s going to be some technical work happening in this area soon, so we have a good opportunity to take this forward. - Theme colours - this isn't going to be enough to clearly differentiate Fedora or on its own, but it could perhaps help to reinforce a sense of Fedora-ness. I’m not sure how it would work in terms of a) maintenance b) coherence with other UI colours c) custom application theming - so from my perspective it would require some research and discussion. Of course, there might be other options worth pursuing [3]. We're currently working on new designs for login/unlock and I can certainly commit to investigate branding opportunities as a part of that. If there are any other ideas out there, I’d be happy to look into those. One thing that would help from my perspective would be some indication of what kind of logo treatments might be acceptable. I'm particularly interested in flat logos and stencil treatments, as these make it easier to subtly apply a brand mark. Thanks, Allan -- [1] While this thread is about Fedora branding, the same arguments apply to RHEL and CentOS. You can assume that I’m talking about all three. [2] This apparent weakness could be turned into a strength, of course. If Fedora/RHEL/CentOS were to become the defacto GNOME distributions, when someone seeing GNOME 3 would assume that it's a Red Hat product. This seems like an excellent goal! [3] Some others are listed on the GNOME wiki page that was referred to earlier in this thread, which I’ve now filled out - https://wiki.gnome.org/Design/OS/DownstreamBranding

Bastien Nocera <bnocera(a)redhat.com&gt; wrote: ... We've also considered allowing applications to set the colour, or deriving it from their icon, in order to differentiate the windows of different applications. ... ... ... Apologies for losing those - I've added them to the list. Allan

Hi Allan, On 02/19/2018 08:44 AM, Allan Day wrote: This last point, #2, is the thing I can't wrap my head around - all it takes is for one non-Red Hat distribution to ship vanilla GNOME 3 and this kind of falls apart, doesn't it? Or am I missing something? ~m

On 02/20/2018 09:39 AM, Allan Day wrote: We don't have design patents on it, though, and I don't think we would want them. So even if that's the perception, it puts us in a vulnerable place. I think that's why trademarks are used to signify authenticity, no? ~m

On Fri, Feb 02, 2018 at 11:03:13PM +0000, Debarshi Ray wrote: I didn't say or even imply that anyone is an enemy. I don't mean it to be "toxic". I don't really know how to phrase it differently. This is important to Fedora. I have been saying so for years. In the part you have snipped that I was responding to, it's said that GNOME designers are not interested in helping us. I'm not sure of another way to read that. If upstream designers don't want to "touch" what Fedora needs, we will need to figure it out here instead. That wasn't intended to be hostile. If you have a different way to get past this, please let me know. -- Matthew Miller <mattdm(a)fedoraproject.org&gt; Fedora Project Leader

Do we have a single page anywhere that discusses the boot/restart/updates experience listing what Fedora wants and what any issues are? While Grub/Plymouth and the Gnome login screen can be themed with similar designs to show seemless transitions, using a black/no background in grub and plymouth also does this. The boot process needs to be fast, but as the same process is used for updates, it does not necessarily mean that it will be fast. With the current plymouth theme, the issues I see are as follows: 1. For a majority of the sequence, the Fedora logo is not present. There is an outline that is filling in. Whichever side of the branding wars you stand on this is sub optimal 2. The changes in animation are jarring. It is not smooth. 3. When there is no change in the animation there is no feedback that something is happening. 4. A filling logo does not really give adequate feedback on the boot process as it may stop for a longer period at some point (not sure if this one still happens). Without going into a branding war, the simple issues can probably be addressed by creating a new plymouth plugin - with a dark background and with an element that animates without breaks or suggestion of "how long" it is left for the boot process to complete. On 1 February 2018 at 18:47, Chris Murphy <lists(a)colorremedies.com&gt; wrote: