On Tue, 2004-01-06 at 19:56, George Shapovalov wrote: > Gentoo is growing and we are gonna be faced with larger and larger number of > new submissions.
Why is that, there are only so many packages, the main ones are covered. There will be new things, others will disappear, but i don't think the submission rate necessarily increases (i don't have figures, but i doubt it does). > So, we can lock the tree and only accept a handfull of new packages now and > then. Well, I do not think this will work in the long run: > 1. This puts a lot of stress on user-developer relations, and it shows in a > regular outbursts of this nature. Plus the locked distro is effectively a > dead one - people will start leaving it eventually.. As long as you can make clear why it shouldn't be in the tree it should give no problems, we're all reasonable individuals here. There are no rules for package rejection at this time, but those could be devised and should be anyway. > 2. Its too late anyway (actually being like that for a long time already). We > are at 100-200 devs (realistically ~100 "maintainers" as there are many > doc/infrastructure/other people) but we have 4000+ packages and 7000+ ebuilds > (may be even more by now). The ratio is already unhealthy and has been like > that for a long time. It did not grow too fast lately because we were > stalling somewhat on new submissions, but continuing to do so will increase > strain and user unhappiness :(. The ratio may not be be the best, but ebuilds are pretty simple and Gentoo's vanilla nature allows for a lot more packages per dev than other distros have. > Another approach: grow the developer base. Not good either - we would have to > get like 1000 more devs onboard and, eventually, more close to 10000. Plus, > if we would want to match the speed of ebuild submissions (we are taking > people in, what I am referring to here is accepting them in "quickly enough") > we would not be able to do proper trining. So either forget QA or this will > persist for a bit more. But then growing into 10000 arear has a good chances > of turning Gentoo into something slow and not very responsive (perhaps other > that to the maintained ebuilds needs). In my opinion this is all based on assumptions not necessarily true. For example, more devs is not necessarily the best solution to more packages. I don't think a distros responsiveness is based on the rate it's package number increases. It's about being responsive to current packages, their quality and not lingering versions behind for no-good reasons that makes a distro worthwhile to me. > So, here I would like to stress the importance of user involvment once again > and point out that effectively we do rely on it. With IRC, mailinglists & bugzilla i feel we're pretty much up to speed on 'user involvement'. Improvements could be made of course, but the far-fetched utopian suggestions here are not the way to go in my opinion. > That description is based around the idea of "splitting" the tree (via the > means of KEYWORDS for example, but lots have changed since, we might want > another way now) into "official" (with its further stable/testing) and "user" > areas (considered less stable by portage. This makes these submissions > automatically visible and easy to install for those who care, but retains > them invisible (and perhaps even unfetchable) for those who dont). > > While there was support behind it, there was an opposition as well. One real > and I think most important objection is along the lines 'do we really want to > stress our servers by all these "unsupported" ebuilds?' I think that's a non-issue and certainly not the main argument against it. It's more about ensuring quality of the distro as a whole, where do you put the line of what is Gentoo and what is not, what is supported and what isn't. It all becomes a fuzzy area, maybe clear to our users or not even all of them, certainly not the outside linux world. I don't expect newcomers to Gentoo/Linux that now happily use ~arch because someone on IRC recommended it -while it really is meant as a testing ground for experienced users, to help out the distro- to know the difference between the different levels of Gentoo-ness or make a conscience choice on what they want. They probably go for 'hey that's a cool new alpha quality app on that screenie. Hey more cool, someone on IRC says it's in the Gentoo user submitted ebuilds level, i'll make that my default level from now on.', getting an unreliable distro in return. This may be a bleak picture, but in a sense these things are already happening. I also fear that devs will miss important changes in ebuilds. For example, say you are a dev and only do reviews, you probably take the easy way out sometimes and just check for syntactical correctness, say ok and grant it in on a higher level, while for all you know all deps may have changed, options may have been added, etc. This already happens in devland for all i know, i see ebuilds where i'm pretty sure the actual package source never got checked on a bump. This isn't too problematic if it happens on occasion, but if it is done on a regular basis the quality of the ebuild decreases. Besides that i see ebuild submissions as a way to educate users on how to write proper ebuilds as-is. Hardly ever i get ebuilds submitted that i do not have to work on, this is nothing more than logical : I do a lot more review/ebuilding than the general user. Submitting a user ebuild without review is a bad idea. And the whole QA deal in the tree itself is already a problem and a lot of ebuild mistakes are made by devs also, so lets start there and make sure all devs do it correctly. - foser -- [EMAIL PROTECTED] mailing list
