Save Gentoo Thread

[96140]

Users are often referred tothe gentoo wiki but with a sensible health warning about if/when it breaks, they can keep the pieces.

This seems somewhat at odds with the fairly recent comment I linked to:

Gentoo-wiki does not now nor will it ever get linked to from official Gentoo media, documentation, or anything else within the www.gentoo.org namespace...

If that dev was talking out of his arse, fair enough, but could you possibly provide something showing that that's the case if so?

I don't think he meant that devs refer users to the wiki, because we don't. Users refer other users to the wiki.

Keep in mind that binary distros deprive their users of many of the choices that gentoo insists users make, so binary distros can afford a lower standard of documentation and to take more risks with it.

I personally don't believe this is true, but even assuming it were, it's not much good if the standard for documentation is set so high that very few people bother trying to meet it. Even the GLDP itself admits many contributors are put off.

Cite your sources for that comment, please. I'm in the GDP.

I don't really see people "put off" by the GDP -- that is, in order to be put off, first users have to take the initial step even of contacting us on IRC to query interest in a doc, or to just submit a patch or a new doc via Bugzilla. That happens rarely, so it's not as much of "No thanks from the GDP for some reason or another," as that will happen even more rarely, and only after someone first actually tries to contribute.

What do we turn down right away, if anything? Off the top of my head, the infrequent suggestions by folks who just say "look, here's a link to something on the wiki." Thanks, but no thanks. Other times, we'll ask users to put their requests/patches into coherent order or for the right code (from the source XML, not rendered html!), things like that. Sometimes users can't be bothered to understandably state what they want or to provide a truly helpful bit of text. In that case, the bug often won't go anywhere.

We really don't see a whole lot of contributors, period. There's not really any one to "put off" when they don't help out in the first place. But in the interest of making the community aware of what's going on, if anyone wants to help, our most pressing issue right now is properly fixing bug 106301.

[NeddySeagoon]

Syntaxis,

Your quote from dostrow is correct. If you want to see developers referring users to the wiki, lurk in #gentoo for a while.
Links to unconfigured unmanaged documents are one thing, individual references are another. You will also see the caveat I mentioned earlier.

There are far too many documents knocked up by complete amatuers at the relevent tasks.

Again, this assumes either the existence of a better page written by someone more knowledgeable (and that people can find), or that no documentation is preferable to suboptimal documentation...

Unfortunately, bad documentation is often worse than no documentation. Bad documents are misleading and give a false sense of security. No documentation forces some trial and error and even reading of man and info pages, neithier of which tell you how to achieve a given end. Only the capabilities provided by the package. Some original thought is still required.
Writing good documentation is difficult and it cannot be done by those who have lots of enthusiasm but almost no understanding of the topic they are addressing.

[AllenJB]

I can name atleast one page on the wiki that developers regularly refer the users to on IRC: http://gentoo-wiki.com/Safe_Cflags

I have to agree with Neddy on the "some documentation can be worse than no documentation" issue.

With regards to the comment regarding the utilisation of the wiki, I don't believe that's true. There are probably 30 odd pages on the wiki that are regularly edited and referred to and of half decent quality. The rest is generally garbage created and then forgotten about.

I'm sure I said it above, but I'll say it again, I believe the wiki's biggest problem is lack of moderation. A quality wiki with 50 really good maintained articles is way better than what we have now, which is a thousand unmoderated, unmaintained articles.

If I had the time to dedicate to such a project, I'd consider starting a new wiki project from scratch that would be properly moderated and maintained because I believe that's the best solution to the current wiki - trash it all and start again. As it is, I make do with maintaining 3-4 articles and trying to apply some moderation to a few more when I can.

[Naib]

I can name atleast one page on the wiki that developers regularly refer the users to on IRC: http://gentoo-wiki.com/Safe_Cflags

yer, maybe its worth that getting pulled into gentoo.org to minimise dev's sending ppl to the wiki if it isn't offical or well maintained?

[AllenJB]

I can name atleast one page on the wiki that developers regularly refer the users to on IRC: http://gentoo-wiki.com/Safe_Cflags

yer, maybe its worth that getting pulled into gentoo.org to minimise dev's sending ppl to the wiki if it isn't offical or well maintained?

I really don't know why they haven't to date to be honest. They did produce http://www.gentoo.org/doc/en/gcc-optimization.xml - While it's a good page to point the wannabe ricers to, I believe most people just want to be given a set of safe CFLAGS they can use and stick with that.

[NeddySeagoon]

AllenJB,

There is no "they" there is only the Gentoo community and we are all a part of it. So there is only a "we"

[mikecore]

I left gentoo when things started going down hill. I went through most of this post. I see Admins and devs posting to it. One a made a point that I already guessed. Gentoo doesn't have a road map. How are dev's and users supposed to know where to go when they don't even know where they have been. IMHO the last time Gentoo had any kind of focus it was early 2004. Things are just done as another dev pointed out. nobody in control everybody just doing whatever they want. As a result you have the current state of Gentoo! no leadership no focus no direction nothing but broken ebuilds and broken systems after a update. A user here stated the other distro's just don't do it for him well I feel like he does. however I not willing to keep putting in the time to make a complete custom gentoo install including building my own kernel ( as I did many times ) hand writing my own FVWM config script. when after updating my system I have to spend a week fixing my installation with work arounds becuase the ebuilds are screwed. And as another dev pointed out the ebuild that broke my system may not even be of interest to a dev unless it affects him/her.
So I just deal with this sorry Fedora install. In the end a computer and OS are here so people can use them, constantly having to fix your box gets old real fast so does
trying to keep up with the changes that are made by devs just doing what they want. Changes are made first then after wards documentation follows in the mean time users are left in the wind. I would love to come back to gentoo! as soon as there is a road map and better organization. Until then I use fedora and keep an eye on the gentoo fourms. Don't wait too long this fedora crap is killing me.

[desultory]

I went through most of this post. I see Admins and devs posting to it. One a made a point that I already guessed. Gentoo doesn't have a road map. How are dev's and users supposed to know where to go when they don't even know where they have been. IMHO the last time Gentoo had any kind of focus it was early 2004.

There is a reason for not having a project wide road map to focus the project on a single particular target: Gentoo is a light bulb, not a laser.

[G2k]

I went through most of this post. I see Admins and devs posting to it. One a made a point that I already guessed. Gentoo doesn't have a road map. How are dev's and users supposed to know where to go when they don't even know where they have been. IMHO the last time Gentoo had any kind of focus it was early 2004.

There is a reason for not having a project wide road map to focus the project on a single particular target: Gentoo is a light bulb, not a laser.

Touche. It's light is definitely not as strong.

[desultory]

The functional difference between a light bulb and a laser is a matter of spectra and directionality, not intensity.

[welp]

Lightsaber!

[legine]

Gentoo doesn't have a road map. How are dev's and users supposed to know where to go when they don't even know where they have been. IMHO the last time Gentoo had any kind of focus it was early 2004.

Hard words.

First of all. We are living on the edge. So what should a roadmap express? I cant imagine how a roadmap for gentoo should look like.
Usually Distros try to have a development and release model. In this modell you need a Roadmap to incorparate changes and track position in order to get release dates. (which you all publish)
Gentoo does imho not work that way. We do constantly release things and test things. There is nothing like "the Breezy" version.
Lets see our last release of Gentoo Linux 2007.0, code named "Secret Sauce".
Which is only a Installer and a basic installation. But after installing once, Gentoo is dynamicly changeing as the Linux Community does.

Things are just done as another dev pointed out. nobody in control everybody just doing whatever they want. As a result you have the current state of Gentoo! no leadership no focus no direction nothing but broken ebuilds and broken systems after a update.

Brakes usually happen out of unsave use of gentoo. We do have a QA Project which should work and point Devs in the right direction to make things right.
I think currently we have a problem with ebuilds that are marked stable, but due changes on dependencies, brake. I am unsure if we have a workfolow for this. At least you won't see if a stable package breaks due Dependecy.
However you are the one to fix it and report fixes. Thats your job as Linux Community user. And that wont change much in other Distros. Well just the Problem area is changeing.

ncluding building my own kernel ( as I did many times )

I switched to genkernel, and it works wonderfull. (exept after I had an accident a while ago. But thats my fault not gentoos.)

And as another dev pointed out the ebuild that broke my system may not even be of interest to a dev unless it affects him/her.

We have some ebuilds as far I know (marked generally unstable) that even do not have an maintainer but are within the system. So you need to be more detailed about the ebuild
I know someone that had an issue with CUPS on SUSE. There was an error within CUPS fixed in a newer version of CUPS. SUSE said they will fix this as soon as their Labor has tested the fix. (which was with next release of SUSE a half year later). If you ask me that is also kinda crazy thing. You are forced into certain versions just because a patch cannot be backward developed and a Version jump is not possible because you have freezed the repository.
I want to say that other Distros have other Problems which are as nasty as Gentoos. the point is which way you like more? I like Gentoos way because i am free to find a workaround myself easily enough without beeing fixed on one version and stay within the system. (Which is very comfortable.)

writing my own FVWM config script.

I dont use that wm, far to complicated. But dont you have to do that anyway if you customize FVWM to your likins? I guess you should look for another wm if thats to time consumeing

changes are made first then after wards documentation follows in the mean time users are left in the wind.

Important Changes are always very well documented. The only Problem is to know of these changes.
In the past GWN was a good ressource to learn about these changes. But due to the Problems of the writer of the GWN we had a lack of some information. But the GWN is a free service and if only one is in charge, you have to count on misstakes.

To me Linux is Chaos. There is no order at all to it. And everyone which says something else is hideing the fact from you that no one is in charge on Linuxwide development. This fact reflects within Gentoo. And if you hate this I guss you should take a peek on MAC Os or something more closed stuff. There youllbe fine

Does anyone know were Linux as complete OS is next year? I mean if you can answer this question you will get the Roadmap for Gentoo. (a bit naive yes, but basicly that is our roadmap ) We are there were the "Linux-Wave" is going and we stay at the edge. Right guys?

Another Idea. Maybe it would be a nice thing to have a Devloper Newsletter like the GWN (or incorperated into the GWN) which tells us a bit more what different Devs of Gentoo think were they want to go. I have the impression that maybe reporting about what devs are concerned with would give a bit more feeling what moves gentoo between the dev-user Gap. And it would fill GWN again.
what are you guys thinking?

For example. A little Interview about the Gnome Project answering what are they trying to do at the moment, what they think are the most serious Problems within Getnoo what is the status about it and what they would like to see icoming up soon. And a little View over the edge, how the Gentoo Gnome Project sees the development of Gneome in general.

Wouldnt that give us the user a bit more feeling about what is gentoo about and were we are going and were we are?

[Codo]

Ok, the last 2 pages of this thread I just bypassed and not read. I've just realised that people are drowning in a pond.

I've just come back from IRC, and as always, #gentoo is one of the busiest channels on freenode, attracting people from all around, from all distros, to ask questions to the gentoo people because they rock. And we do, C'MON.

1.- Get the bloody GWN running again please, post anything, even the bugs ranking, I don't care. Put an article about people complaining about GWN, copy paste from IRC, I don't care, just make sure is weekly. Just make sure there is a new post every week. Post photos from the london meeting, anything! Who's in charge? Send me an E-mail and I'll do it next week if you want. Just tell me what to do.

2.- Regarding the state of the devs and becoming a dev and all that malarkey,... A few words for many of you. Having a @gentoo.org E-mail address doesn't make you anything. In my case, I go to IRC to give help any time I can. And almost always I am helping a user install gentoo. When I do that, I feel I am bringing someone new to the community and that's very rewarding. For others, it may be posting an Ebuild, or maybe bringing Gentoo to a client. I used to help in the forums (I now live in a place where I have no easy internet access). **CONTRIBUTE**. If you're looking to make your nickname have an elevated status because you're maintaining baselayout or some other core package, I'm telling you know, **PLAIN ENGLISH** is not about that. JUST CONTRIBUTE. Help. Answer questions in the forums. Patch a missing comma in an ebuild. CONTRIBUTE.

3.- Regarding the coordination and authority, I agree. It is a bit of a mess at the moment. This thread demonstrates it. In the old days, the heads of the herds will just do, and things go ahead. Releases were in time. No bull****ing. Now everybody is sinking in pointless discussions. Can everybody stop bitching around and get their hand to work? I'm helping at IRC. If you have bandwidth to download the source and to look at why KDE complains at not being able to eject my USB stick... would you do it? And just send the patch. If the dev doesn't reply, then pressure him/her to tell you your patch is crap or to say thank you. Make it happen. MAKE IT HAPPEN.

4.- Regarding the devs.... I agree again. I have been sometimes in contact with devs (indirectly by bugs.gentoo.org), or reading around, and the devs have been sometimes despective (or just not taking seriously, ignoring) of contributions from the community, either by ebuilds, patches, etc... this is another topic but the heads of the herds must understand that for a newbie for wanting to dig into the source code of kwin or dbus is a BIG thing. I wanted at some point to implement aiglx composite for kwin, but the kwin mailing list got a bit not nice (and the maintainer ended up doing it). In that case I just wanted some coaching, but once a message appeared from the maintainer saying "So what's going on! Is it happening or not!" or the like. I mean, I didn't need that. I just unsubscribed from kwin at the time. Let's keep those things away from the gentoo mailing lists. And for the devs, there is an emotional bit on people wanting to contribute time an effort. Please appreciate that. Many of us have full time jobs (I am writing this and is around 1am. I was working today).

5.- Can someone organise a meeting at IRC so we can choose a few people to make things happen? Something like the "Gentoo Make it Happen of 2007" that have the responsibility of coordinating efforts to make sure the bits missing / urgents are dealt with before July ends? Like the GWN, Sorting out the lack of morale and the like?

Thanks guys, going to sleep. See you tomorrow.

[rambam]

Ok What the hell is going on within the Dev's of Gentoo

Thats: Bryan Østergaard
http://archives.gentoo.org/gentoo-dev/msg_145740.xml

And: Alexander Færøy
http://archives.gentoo.org/gentoo-dev/msg_145764.xml


Sure dev's come and go, the problem is as anyone know you need some people around with experience to help bring upto speed the new blood.
Quite a few experienced dev's have been leaving so what state does that leave the foundation?

Who cares ?
The dwindling of Gentoo is pretty irrelevant except to the fanatics.
Most of have used many Linux distributions and Gentoo will not be the last one we use.

Gentoo has been ailing for a while, and its market share has eroded severely.
But that is ok, there are plenty of other Linux distributions around.

[mark_alec]

Gentoo has been ailing for a while, and its market share has eroded severely.
But that is ok, there are plenty of other Linux distributions around.

Do you have any figures to back that up? I tried looking on Netcraft, but all I could find was an article from 2004.

[tld]

Do you have any figures to back that up? I tried looking on Netcraft, but all I could find was an article from 2004.

I'd like to see that as well. So what...all of a sudden those of us who have no use for binary distributions are suddenly going to go with linux from scratch or something because of some rumors that Gentoo is on the way down? If it is, you'd never know it based on how busy these forums are.

Tom

[tld]

There is nothing like "the Breezy" version.
Lets see our last release of Gentoo Linux 2007.0, code named "Secret Sauce".
Which is only a Installer and a basic installation. But after installing once, Gentoo is dynamicly changeing as the Linux Community does.

This may be one of the best observations in this whole thread. One of my favorite things about Gentoo is that there really aren't distinct versions like other distros...and it doesn't have them because it doesn't need them. Binary distros need major upgrades because there's no other way to handle major upgrades of core components. And why does Gentoo need a "Road-map" beyond continuing to do what it's been doing? The closest thing to a new Gentoo "version" is a new default profile. Who gives a crap about defaults when the main goal of the distro is customization? Beyond that, as legine pointed out, it just evolves along with the upstream projects. That is the roadmap the way I see it.

Some people gripe about things like the gcc upgrade from 3 to 4 and recompiling the system. After my experiences with major binary upgrades in the past, I was thrilled that I could actually do this...especially given I recompiled the whole system in an aterm window while I continued using it...and this is probably one of the most painful sorts of upgrades you're going to encounter with Gentoo. I'm sure not complaining.

Tom

[aidanjt]

I went through most of this post. I see Admins and devs posting to it. One a made a point that I already guessed. Gentoo doesn't have a road map. How are dev's and users supposed to know where to go when they don't even know where they have been. IMHO the last time Gentoo had any kind of focus it was early 2004.

There is a reason for not having a project wide road map to focus the project on a single particular target: Gentoo is a light bulb, not a laser.

Laser light can be omni-directional as well

[desultory]

Laser light can be omni-directional as well

Show me a spread spectrum laser, and I will retract my analogy as completely broken.

[NeddySeagoon]

desultory,

There is an excellent article on white light lasers in a recent Scientific American, (Dec 2006 ?) complete with images of coloured light going into one end of a fibre optic, from a conventional laser and white light emerging from the other.
Its still not omni-directional but it is a broadband laser.

[glide87th]

Ok, the last 2 pages of this thread I just bypassed and not read. I've just realised that people are drowning in a pond.

I've just come back from IRC, and as always, #gentoo is one of the busiest channels on freenode, attracting people from all around, from all distros, to ask questions to the gentoo people because they rock. And we do, C'MON.

...snip

Codo, your attitude is what's missing from the Gentoo community. I was an arch tester for a couple of years until the sniping on the mailing lists and abuse in the IRC channel just got to me one day. I was truly embarrassed when Daniel showed up on the lists again this year and they just flamed him right out the door. You can sound off on the lists and hang out in IRC to show everyone how Leet your are, but none of that matters if everyone walks.

I spend my time now helping a python based project because the creators and community there still remember what it means to build a strong open source community. They are thoughtful and respectful, and they approach the task with a sense of joy and humor that makes helping them a rewarding experience. it's not about the software; it's about the people.

Mike 'glide' Bonar
ex-amd64 arch tester

[cokey]

It's definitely not hard, it's you. When you're not able or willing to go through the documentation to be able to "master" the quiz, it's very likely that you're either not good enough or won't sustain as a Gentoo dev for long, as it is not always fun to work on annyoing bugs or requests you personally don't give a flying fart about, even though they're reasonable. So, if you don't invest a minimum of time, you're not worth it, wasting anyones time to mentor you.

Maybe you are right. I don't know that since I never walked that way.

put it this way... I've passed it

[steveL]

put it this way... I've passed it

Wahey! Does this mean you're going to be a dev now, cokehabit? Go for it man!

[cokey]

put it this way... I've passed it

Wahey! Does this mean you're going to be a dev now, cokehabit? Go for it man!

no, i just took it and passed it. I made some idiotic small mistakes which were obvious so i was corrected on them...

Anyway, i have no clue of bash or patching so i couldn't be a proper dev, they'd have to teach me...

[Codo]

I was an arch tester for a couple of years until the sniping on the mailing lists and abuse in the IRC channel just got to me one day. I was truly embarrassed when Daniel showed up on the lists again this year and they just flamed him right out the door. You can sound off on the lists and hang out in IRC to show everyone how Leet your are, but none of that matters if everyone walks.

I spend my time now helping a python based project because the creators and community there still remember what it means to build a strong open source community. They are thoughtful and respectful, and they approach the task with a sense of joy and humor that makes helping them a rewarding experience. it's not about the software; it's about the people.

Well, that's a real shame that you're not putting your efforts on Gentoo anymore... Would it matter if I tell you 'please come back'!

I think patience is your friend sometimes, and everyone should kindly ignore the sniping/attack/nasty sentences... Two days ago a guy in #handhelds spoke to me with swearing and the like and it really got me.... I got a bit nasty... But of course I'm not going to let that nasty <<noname>> take me away from getting involved and hacking my PDA?!?! Deep breath, and off we go. So my message would be: Deep Breath my friend, and keep sailing forward...