New Council – Expectations?

Ok, we got a new council, I’m still there so thank you for renewing the trust on me =)

Looks like that less people found me or what I did that compelling to make me into the council, so surely I did something wrong. Solar got the first place so his cleanly cut ways are perceived better.

I started polling people about what they feel about Gentoo and what they’d like. The first thing I noticed is that people are sick of endless discussions on marginal stuff and even more sick of outside projects trying to push it’s agenda on Gentoo using the shovel-in-throat way.

Second item is about trying to make the place nicer for everybody and better involve our large userbase. We used to be the nicest distribution regarding attitude towards newcomers and slow learner, now other distributions are better. We could re-learn from them.

That’s what I perceived so far. As I said before I see the council just as the last resort to get something decided if we, developers, cannot find a large agreement. Solar likes more to be proactive in my opinion. You liked him so I guess we as council should try to push people express themselves and get new&interesting stuff done instead of discussing which is the new way to define a quantity next to infinity or why embedding information somewhere is right or wrong in theory.

That said, how wrong I am so far and how we could get Gentoo to improve even more?

9 thoughts on “New Council – Expectations?”

  1. > even more sick of outside projects trying to push it’s agenda on Gentoo using the shovel-in-throat way

    > Second item is about trying to make the place nicer for everybody

    Yeah, making malicious accusations about people is a great way to do that.

    And for someone who so loudly insists that GLEPs 54 and 55 are such minor issues, you’ve put a hell of a lot of effort into saying “no”.

  2. I think the problem of discussions are not discussions by themselves but more how people in Gentoo handle it. To be concrete we need to find a way to be more

    – to the point
    – solution-oriented
    – listening, not telling

    A few more people acting like mediators maybe could help. If only it wasn’t that hard to learn.

  3. It would be nice if the community was a little bit more friendly. That being said, any questions new users pose are normally answered, just often in a backhanded way.

    Personally, I think the main issue with getting new users is improving the resources on the net for learning how to use Gentoo. The website has a number of documents on it that are out of date, and the newsletter stops in november of last year.

    When I volunteered to try and help out with the above, i was told to learn Guide XML and start looking at the documentation bugs in bugzilla. Now I’m starting to do that, but it still represents a very high barrier to contributing; there are also sections, like the website and newsletter, that need more of a project approach than a bug patching approach.

    Conversely, I’ve been able to add information to the Gentoo wiki with ease, and the information on there isn’t covered in labels saying it isn’t maintained (whether or not that’s the case).

    There seems to be a school of thought that suggests new users shouldn’t be welcomed into Gentoo till they have enough knowledge to make their way in themselves. I can see the point behind that, but I tend to believe that by enticing as many people as possible into using the distribution you have more potential contributors and a better end product. I’d like to see the council putting someone in charge of lowering the entry barriers into using Gentoo and helping with new users.

  4. I think one of the problems with gentoo is that it has become complicated.

    At one time, an ebuild basically pulled the sources and built them.

    Then Use flags where added
    then ~x86 and the like where added,
    then /etc/portage/stuff was added
    then layman overlays were added
    then sets were added

    It’s impossible to keep track of this stuff, and also impossible to keep a running system with the newer software on it. kde4 has been a mess and just the other day I unmased the /usr/portage/ kde4.2 whatever it was, and now it won’t start, something about a config problem.

    I have no problem using bash, or reading a bit to figure things out, but gentoo has just gone way overboard, adding warts to the build system.

    Now I’m not saying that the warts aren’t necessary or even that they aren’t useful. I’m saying they are confusing to track and understand.

    So in order for gentoo to fix this problem as perceived by me, it needs a nice package manager that handles all that in the background.

    For example, I happen to like kde3.5 for some projects and 4.2 for others. Do I use kdeprefix, somethings I read says no, others say yes but not recommended.
    Or how about sets. A new feature that I’ve used once successfully by following directions in the forum. Where do I find out what set’s are available. Does emerge tell me? (Not that I can tell).

    When I first started using gentoo, I had a pretty good understanding of how the system worked. Now, I feel that I have to put my blind faith in the developers and often that faith is rewarded, and often that faith is punished. My fear is that the system is too complicated even for developers to handle, and they are often missing the tricks they should be using.

    If you want to improve gentoo, I think usablity is key.

    Sincerely,
    Chris Bruner

  5. Yeah, get things *done* – even if you’re just covering 95% of the users. Don’t be afraid to break old stuff. Provide users with up-to-date information.

    It got better during the last months or so, but still, I think the devs should still be more communicative on the web.

    Gentoo needs an User-Evangelist. 😉 Someone that writes a column regularly (!) on what’s going on in the gentoo-dev-world (yes, the controversial issues are interesting too).

    For a lot of users, nothing seems to happen “behind the curtains”, but in the end I believe, it’s just lack of communication.

  6. Congrats for your reelection!

    About communication: at least get an automated GWN/GMN out. With stats generated from bugzilla (like in the old GWN), stats from the forums, stats on ebuilds, last rites, some graphs, etc. This way, even if the Newsletter folks are busy elsewhere, we get some info and we see that things are happening in Gentoo land.

    About getting things done: it seems there are some decisions to be made. Most Gentoo devs seem to be working on ebuilds. There isn’t enough manpower to perfectly maintain and create all packages we’d like, and there never will be. It’s been like that for a long time now, and from a user’s perspective it seems Gentoo reached status quo on this and nothing is being done to solve the problem. Hence a sense that quality and reactivity are dropping, compared to Fedora (!) for instance.

    As a user, I’d really like to see essential Gentoo stuff being worked on. Portage for instance: fixing the -t option, adding the ACCEPT_LICENSES features… stuff that has been broken or missing for like forever. Focus on the toolchain, on @system, on default configs for X or for networking.
    But sooner or later, you’ll have to trust the community with simple, non-critical application packages. (And no, Sunrise is not the answer we’re waiting for – cut this IRC requirement!). It’s obvious that devs can’t keep up with all ebuild requests in bugzilla; it’s probably time to accept it, to stop pretending, and to find a solution.

    Good luck 🙂

  7. I’v been using Gentoo as my main OS since the launch of C2D. I tried it before, but since I like to experiment, just used binary distro (Debian–>Ubuntu–>Debian) until I got enough power to recompile the whole system in less then 24 hours.

    Somehow in last year, there is a little annoying smell of death. It is at least hard to find any plans. You don’t see any posts about Gentoo on any IT site. Packages released long after other distros. IMHO this drives users and developers away. And it certainly does not attract newcomers.
    It just seems like headless herd without vision, plans, direction and future.

    I’d like to help, but the only use i found was translation (take example of Ubuntu translations) and posting some humble howto or answering in forums.

    I trully hope that all said above is ill hallucination. But if it is, than it shared by large number of ex/potential/current users.

    Sincerely,

    P.S.
    Please fix the email validation to accept something like first.last@somthing.com

  8. Well you find plans per project so the python people are thinking how to support various python implementations and let you select which one to use with lots of granularity, Diego and the ruby people trying to get ruby, jruby and yarv to work seamless, those may or may not interest you and usually it’s better if you won’t notice such changes till you need to since they should give you more power but not require interaction if you don’t want it.

    Probably would be nice having some way to convey those project information in an easy way (like having a team blog post) but in the end I guess most of the gentoo people enjoys more doing stuff than talk about them ^^.

    If somebody wants to join the pr project to help extracting juicy news from our developers I think he/she would be quite welcomed =)

  9. Congratulations and well done.

    As we can all see Gentoo is still not perfect. *sigh* Never will be. It should always be a work in progress.

    As such I should like to from a users perspective suggest some things that might help.

    There should be more AI available to users from a default system. Package managers have been discussed of course. Having run recently a Ubuntu system I was impresssed with some things like synaptics and the janitor. It would be nice to see some cli scripts that help make some good default choices.

    A supervisor not a daemon but something that could parse a few logs and make recommendations if needed. Choose a default profile and tweak it for the users intended use. Herds could be utilised to add a recommended install extending perhaps the concept of profiles. In this way a default LAMP install for a VPS Host or server environment could be configured with one extended profile will a kde education server could be another perhaps incorporating some features from overlays?

    I also wonder at the “system” definition. Seems somewhere along the line that was determined to be a build environment?

    Always been a peeve of mine that emerge system does not include kernel and security updates for all system files including the logger. If you include GLSA support though that would mean anything gui or web based on the system. but not difficult since the glsa is available.

    perhaps emerge uptodate

Leave a Reply to Dima Weissman Cancel reply

Your email address will not be published.