Ask Fedora is one of our primary user-facing support resources. It's reasonably successful, with over 18,000 questions, with almost 80% of those having at least some answer. It has a small but active and passionate community. On the other hand, I've heard from the infrastructure team that it's taking an increasing amount of time just to keep the service running, and making improvements is out of the question. We're having trouble keeping up with things like keeping the software on the recent version, let alone advancing new ideas. Since this is such an important thing for users, I think we need an improvement. I asked the community what to do <https://ask.fedoraproject.org/en/question/95681/ask-fedora-is-it-working-... and the top-voted response is from the upstream author, who has offered to host the site for $250/month, including 6 hours of development time. Assuming we can make the money work out, what do you think about this?

I think if the big question becomes *what to do with Ask Fedora*, we should give special consideration to choosing a route forward that will provide the best user experience for people who are (1) asking questions, and (2) answering the questions too. I'm not sure of the process to become integrated with the

However, I'm concerned about setting the precedent of paying to host a service. The cost to Fedora to host it is not in machine time or bandwidth, but in sysadmin and maintenance. Ideally, some of those

same passionate Fedora Ask community members would step forward to help run the site by becoming part of FI, even if only to service Ask.

Okay, so.... needs here are: Sysadminy/packager stuff: - some more hands on deck for routine administration - updating to newest codebase - package as RPM? - update ansible playbooks - reconfiguration on site itself to better match high-load site Developery stuff (in my rough order of priority) - Fix the "closed: answered" reason (as I've discussed before) - Make duplicate handling useful — include voting system - Creation of a real "meta" subsite - A flagging system for issues beyond "offensive" - Insight into things like most viewed, most duplicated, and most upvoted questions, so we can consider creating Fedora Magazine articles and more formal documentation for those things. - Rework notifications so they more useful (broken down by categories, not punishing users for helping copy-edit questions) - Ability to put questions in need of help (for example, unclear with not enough information) on hold rather than closing outright - Community review queues a la Stack Exchange - Tie in better to Fedora Badges (possibly completely disable integrated badges) This doesn't all need to be instant, but I'd like to be moving towards it....

On Wed, Dec 14, 2016 at 03:42:37PM -0500, Matthew Miller wrote: > > Developery stuff (in my rough order of priority) > > - Fix the "closed: answered" reason (as I've discussed before) > > - Make duplicate handling useful — include voting system > > - Creation of a real "meta" subsite > > - A flagging system for issues beyond "offensive" > > - Insight into things like most viewed, most duplicated, and most > upvoted questions, so we can consider creating Fedora Magazine > articles and more formal documentation for those things. > > - Rework notifications so they more useful (broken down by > categories, not punishing users for helping copy-edit questions) > > - Ability to put questions in need of help (for example, unclear > with not enough information) on hold rather than closing outright > > - Community review queues a la Stack Exchange > > - Tie in better to Fedora Badges (possibly completely disable > integrated badges) > > This doesn't all need to be instant, but I'd like to be moving towards > it.... Why not adding it to the next GSoC then? It really seems like we need a few very specific features. Also, it would possibly bring some potential maintainer for ask on board. Of course it would have to wait until march/may, when students start looking for projects, unless we add it to the gsoc page already and maybe people would start working on it sooner.

Which loops us back to the original. The upstream is trying to make this an open source product where they are paid for their time to work on this by organizations who are using said product.

The best would be if upstream was to accept GSoC student for these questions, but then they would have to have the time and will to do so.

Assuming we can make the money work out, what do you think about this?

On Wed, Dec 14, 2016 at 06:00:58PM -0000, Josh Boyer wrote: > However, I'm concerned about setting the precedent of paying to host > a service. The cost to Fedora to host it is not in machine time or > bandwidth, but in sysadmin and maintenance. Ideally, some of those This is a good point that I hadn't considered. It wasn't really the hosting per se that I'm interested in, but getting the service overall into a better place, and this seemed like a possible way to do that. But, if we were to try to directly sponsor every upstream project on which we depend in some way, we'd use up not just our budget but pretty much all the money in the world. > same passionate Fedora Ask community members would step forward to > help run the site by becoming part of FI, even if only to service > Ask. Okay, so.... needs here are: Sysadminy/packager stuff: - some more hands on deck for routine administration - updating to newest codebase - package as RPM? - update ansible playbooks - reconfiguration on site itself to better match high-load site Developery stuff (in my rough order of priority) - Fix the "closed: answered" reason (as I've discussed before) - Make duplicate handling useful — include voting system - Creation of a real "meta" subsite - A flagging system for issues beyond "offensive" - Insight into things like most viewed, most duplicated, and most upvoted questions, so we can consider creating Fedora Magazine articles and more formal documentation for those things. - Rework notifications so they more useful (broken down by categories, not punishing users for helping copy-edit questions) - Ability to put questions in need of help (for example, unclear with not enough information) on hold rather than closing outright - Community review queues a la Stack Exchange - Tie in better to Fedora Badges (possibly completely disable integrated badges) This doesn't all need to be instant, but I'd like to be moving towards it....

Ask Fedora is one of our primary user-facing support resources. It's reasonably successful, with over 18,000 questions, with almost 80% of those having at least some answer. It has a small but active and passionate community.

On the other hand, I've heard from the infrastructure team that it's taking an increasing amount of time just to keep the service running, and making improvements is out of the question. We're having trouble keeping up with things like keeping the software on the recent version, let alone advancing new ideas.

Since this is such an important thing for users, I think we need an improvement. I asked the community what to do <https://ask.fedoraproject.org/en/question/95681/ask-fedora-is-it-wor king-and-whats-the-future/> and the top-voted response is from the upstream author, who has offered to host the site for $250/month, including 6 hours of development time. Assuming we can make the money work out, what do you think about this?