Fedora 15 Retrospective

From FedoraProject

Revision as of 17:49, 7 June 2011 by Gholms (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

Contents

Introduction

This page is intended to gather feedback from the Fedora community on things that worked well and things that could have been better during the Fedora 15 release cycle, including (shifting the feature freeze until after FUDCon, NetworkManager 0.9, and so on). The feedback will be used as a basis for identifying areas for improvement for Fedora 16. Any thoughts, big or small, are valuable. If someone already provided feedback similar to what you'd like to add, don't worry ... add your thoughts regardless.

The QA team has their own retrospective page. If you have specific feedback related to testing and/or QA, please visit the Fedora 15 QA Retrospective page.

Providing feedback

  • Gwjasu - I like ____ about the new ____ process

Adding feedback is fairly straight forward. If you already have a Fedora account ...

  1. Login to the wiki
  2. Select [Edit] for the appropriate section below.
  3. Add your feedback using the format:
    * ~~~ - I like ____ about the new ____ process
  4. When done, Submit your changes

Feedback

Things that went well

  1. Jared Smith - BoxGrinder - I was very happy to see the BoxGrinder feature work well in Fedora 15, and even happier to see the efforts put into starting to get the packaging for parts of JBoss to a point where they can be included in Fedora.
  2. Bruno Wolff - slips - In sort of a mixed bag category, I thought the extra time provided by the two one week slips ended up with us releasing a better product. We had several ambitious features and the extra time seemed to be helpful for polishing them.
  3. Bruno Wolff - packaging - This was the first release that used git for packaging all the way through. I have found it makes backporting updates easier than when we were using CVS.
  4. Bruno Wolff - spins - The timing of XZ support for squashfs getting into the kernel was fortuitous as we would have had to either make hard decisions about cuts to keep several live spins CD sized or decide to use images that wouldn't fit on CDs.

Could have been better

  1. Jared Smith -- NetworkManager bump - The move to NetworkManager 0.9 came after the feature freeze, and the Sugar desktop wasn't accounted for in the scramble to get alternate desktop environments working with it.
  2. Jared Smith -- artwork - There seemed to be a lack of consensus from the Design team on whether or not the artwork was considered ready or not
  3. Jared Smith -- systemd - The Fedora Packaging Committee had a hard time getting specifics from the systemd developers in a timely manner, which caused difficulties in creating packaging guidelines with respect to systemd
  4. Jared Smith -- marketing - The Fedora Project Leader left the Red Hat press release and web copy until the last minute, and it should have been done much earlier.
  5. Paul Frields -- marketing - There didn't seem to be the usual number of interviews and feature profiles coming out around the release, which was even more surprising since the look and feel of the release with GNOME 3 was so visibly different. This was a lost opportunity to partner well with a major upstream where we have a lot of participants producing new code.
  6. Jared Smith -- translation - We switched from a self-hosted version of Transifex 0.8 to a hosted instance at transifex.net mid-way through the F15 release cycle. This caused many difficulties for the translation team, and I'm not sure how well the developers and packagers in Fedora understood the impact of the changes.
  7. Bruno Wolff -- spins - The Spins SIG leader got busy (with work I think) and we ended up making rushed (though reasonable) decisions (after a number of meetings didn't happen) on which spins should be produced.
  8. Bruno Wolff -- hardware requirements - The amount of memory needed to install F15 increased significantly. There is a plan to reduce this for F16, but there wasn't time to do it for F15.
  9. Máirín Duffy -- artwork - not having separate wallpapers per desktop environment would have been better, caused a lot of confusion in what artwork to use in producing collateral materials for the release and a burden on team
  10. Máirín Duffy -- artwork - some kind of metrics for artwork quality as part of QA would be good
  11. Jared Smith -- Amazon EC2 images were not prepared for release time. (Still trying to learn exact details of what held them up.)
  12. Garrett Holmstrom -- release notes - At release time the release notes had not yet been updated to reflect the greater amount of memory needed to install with install media.

Wishlist

References