Dirk.R.Gently wrote:And the divide between users and developers become more obvious.
An incomplete 2 month old project package.gentoo, dilapidated front-page and website, no release schedule and dwindling releases.
Err, the lack of 2007.1 has been explained - please see the announcement on http://www.gentoo.org/ ). I also believe it's also been at least hinted that there will be a 2008.0 and relatively soon from what I can gather.
Mirrors can disappear for any number of reasons. This particular one was hosted by a university chemistry department - maybe the academic maintaining it left or the machine was needed for other purposes. Mirrors do occasionally disappear or get created. Just because one of about several thousand mirrors in existence disappears, it doesn't mean the rest are shortly to follow. To incite that it does is pure FUD-mongering.
Problem: The social climate into the Gentoo community has been degrading during last years Description of the problem: see the excellent analogy made by c0d3g33k Why this problem exists:
Finding the causes of a social problem is not easy especially in our case since there are both psychological and technical aspects. However:
- 1) Lack of communication between developers and users
- 2) A slightly unfriendly tone in _some_ devs' answers to the users' questions. "It's really amazing how easy it is to exclude and discourage people (consciously or not) by simply being passively uninviting and slightly exclusionary."
- 3) Probably some users' deception caused by: a) their requests or dreams being ignored (like the alternative of having binary packages officially available, for instance) ; b) major technical mistakes like the expat upgrade which came without warning and couldn't be fixed with a simple revdep-rebuild on many systems. Proposed solutions:
- 1) Probably the new GWN/GMN should include information about new developer innovations . It seems that not all the users pay attention to Planet Gentoo.
- 2) Probably GWN/GMN should include information about decisions regarding the general direction of Gentoo - maybe not all the users are subscribed to gentoo-project or gentoo-dev
- 3) Probably the developers should put more care into the style of their replies on the forums or mailing lists because users are easy to confuse and often take a dev's answer as an official position.
- FIXME: More suggestions needed.
This is Just a short digest of some posts I've seen around. I don't remember a dev being unfriendly to me ever (so there's no need to prove me I'm wrong), but I do feel that the overall atmosphere got colder during last years. Also do note that I didn't mention Robbins' return as a way of improving the social climate because I don't want useless flames (I'm not saying it wouldn't be a solution). And please help with finding solutions.
Last edited by omnio on Sat Jan 19, 2008 11:44 am, edited 2 times in total.
AllenJB wrote:
Err, the lack of 2007.1 has been explained - please see the announcement on http://www.gentoo.org/ ).
And what an explanation it is:
Work began on the release as planned, but vast numbers of security vulnerabilities between September 2007 and December 2007 and reduced manpower left the Release Engineering team constantly overworked...
Sounds like a healthy project.
AllenJB wrote:I also believe it's also been at least hinted that there will be a 2008.0 and relatively soon from what I can gather.
Why only hints? Why do you have to "gather" anything? Healthy projects have definite plans and transparent scheduling.
Edit: And we wouldn't even have that little tidbit if DRobbins hadn't made a stink.
would you rather have too much work or not enough work?
besides the JMICRON support for C2Ds I see absolutely NO reason to continue to beat the dead horse that is 'no 2007.1' system profiles. Yes, we should have installs that support JMICRON but who the hell really cares if their gentoo version is '2007.0' or '2007.1'? if you care you aren't paying attention.
jonnevers wrote:if you care you aren't paying attention.
If you don't care that "emerge -u" after an install takes forever, your time is not as valuable as mine. You also don't work with a bunch of Debian fanbois who'll use it as an excuse to whine about how Gentoo is so hard to install, and it takes so long...
jonnevers wrote:if you care you aren't paying attention.
If you don't care that "emerge -u" after an install takes forever, your time is not as valuable as mine. You also don't work with a bunch of Debian fanbois who'll use it as an excuse to whine about how Gentoo is so hard to install, and it takes so long...
BTW, this means that if I make a "fresh" install now I'll get the old expat kicking and rocking? Sorry to interrupt.
Last edited by omnio on Fri Jan 18, 2008 6:13 pm, edited 1 time in total.
jonnevers wrote:if you care you aren't paying attention.
If you don't care that "emerge -u" after an install takes forever, your time is not as valuable as mine. You also don't work with a bunch of Debian fanbois who'll use it as an excuse to whine about how Gentoo is so hard to install, and it takes so long...
BTW, this means that if I make a "fresh" install now I'll get the old expat kicking and rocking?
Sorry to interrupt.
yes - unless you're using the stage tarballs provided by Daniel
berferd wrote:If you don't care that "emerge -u" after an install takes forever, your time is not as valuable as mine. You also don't work with a bunch of Debian fanbois who'll use it as an excuse to whine about how Gentoo is so hard to install, and it takes so long...
I don't work with people that whine about anything. I work with professionals.
AND daily stage tarballs are not equivalent to a 2007.1 profile! why are you people deliberately misunderstanding what you are talking about?
berferd wrote:How is this possible?! Making stage tarballs is such a lengthy and arduous process that there's no way a single person can do it in their spare time.
1) you love looking up links huh? even though they don't really support what you are saying.
2) you're again deliberately misrepresenting what we are talking about.
it's one thing to pop off daily builds of the stable portage tree, AGAIN THE S T A B L E PORTAGE TREE. get it? it is something ENTIRELY different to create a *NEW* profile that changes what constitutes THE STABLE TREE. This would be more like including a major gcc version bump. Two completely different tasks....
does the 2007.0 profile still bring in the old expat to update to the new version? this is a serious question as I'm not sure.
, which gets you a stage 2.
Using the stage 2, you add in the odds and ends, which gets you a stage 3
Other than making the stage 1 using another system, this used to be the only way to install Gentoo.
Even on a 233MHz P1, the process takes about 24 CPU hours and about 30 minutes of user input.
With proper planning and a few scripts, little user input is required.
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
NeddySeagoon wrote:You normally start with a stage 1...
I thought stage 1 installs were not supported?
NeddySeagoon wrote:....Other than making the stage 1 using another system, this used to be the only way to install Gentoo.
Even on a 233MHz P1, the process takes about 24 CPU hours and about 30 minutes of user input.
With proper planning and a few scripts, little user input is required.
Are you saying this is a reasonable way to install an OS on a large number of machines?
Stage 1 installs are not supported any more if there is a stage 3 for your arch.
However, the way to make a stage 3 from nothing is to build yourself a stage 1 and work your way up.
Thats how releng do it and why all the stages are still or the mirrors.
If you want to install on a large number of *identical* machines make a stage 4 image and deploy that, or set up a BINHOST and have them all pull from that, either way, packages are only made once ans used by all the installs.
If the machines are not quite identical, can you find some CFLAGS and USE settings that suit them all ?
The icing on the cake is distcc across all the machines so your machines work together on the common install.
A stage 1 on each system is a waste.
Feel free to try a stage 1. It takes longer, is more error prone and has no advantages over a stage 3.
Support for stage 1 was dropped as it makes no sense for users to do it, other than the educational experiance
Regards,
NeddySeagoon
Computer users fall into two groups:-
those that do backups
those that have never had a hard drive fail.
omnio wrote:BTW, this means that if I make a "fresh" install now I'll get the old expat kicking and rocking? :)
Sorry to interrupt.
yes - unless you're using the stage tarballs provided by Daniel :)
Or if you use [topic=546828]update[/topic] from the beginning, it'll pick up on the expat upgrade and handle it (and other revdep stuff) for you automatically.
omnio wrote:BTW, this means that if I make a "fresh" install now I'll get the old expat kicking and rocking?
Sorry to interrupt.
yes - unless you're using the stage tarballs provided by Daniel
Or if you use [topic=546828]update[/topic] from the beginning, it'll pick up on the expat upgrade and handle it (and other revdep stuff) for you automatically.
Thank you. I appreciate for sure your attitude and your efforts (here, on #bash, on #friendly-coders). Anyway what worries me is that even now in 2008 someone who makes a so-called "fresh" install hits this problem. While I believe that update works well and it's worth trying it, I'm still anxious because of the lack of an "official" fix.
The Unknown wrote:You know, its funny I did a fresh install about a month ago on new hard drive and never hit the expat issue, actually kind of forgot about it.
That is probably because there were both gcc and glibc updates waiting at the end of your install. So you had to "emerge -e system" and "emerge -e world" anyway. (which would resolve the expat problem)
That is probably because there were both gcc and glibc updates waiting at the end of your install. So you had to "emerge -e system" and "emerge -e world" anyway. (which would resolve the expat problem)
As I'm recalling, your right. I actually ended up doing it twice because I made a mistake in the step where you link to your profile, I ran the command
"ls -snf" instead of "ln -snf" lol (well... I laugh now anyways)
Yeah you were lucky
Yeah you could say that. But the expat issue the first time was only a 15 minute set back for me anyways.
AllenJB wrote:
Err, the lack of 2007.1 has been explained - please see the announcement on http://www.gentoo.org/ ).
And what an explanation it is:
Work began on the release as planned, but vast numbers of security vulnerabilities between September 2007 and December 2007 and reduced manpower left the Release Engineering team constantly overworked...
Sounds like a healthy project.
AllenJB wrote:I also believe it's also been at least hinted that there will be a 2008.0 and relatively soon from what I can gather.
Why only hints? Why do you have to "gather" anything? Healthy projects have definite plans and transparent scheduling.
Edit: And we wouldn't even have that little tidbit if DRobbins hadn't made a stink.
guys, i ve been off for a while.
i m just astounded of all this.
it s a hard and complicated problem.
Not engouh devs because it s not "that easy" to be a dev.
handbooks are great and everyone is able to learn a lot incomputers, that will always be the main contribution from gentoo to me. You know what you re doing...
it s really easy to create a revolution by words, by nicknames and creating bad moods in a community.
The purposes can be complex and there are many.
If you feel something is going wrong ... there s already a problem.
the problem is money guys. it s always been. A nice project such as gentoo is s o beauutiful and so charming that everyone is happy and blablabla
then difficulties arise, devs are saturated because too many things to sort on, linux growing crazily, some other distros being really invasive and tempting, gentoo becoming maybe too popular means a lot of new packages want to be added and maybe the main lines of the distro should be different... divergences begin to appear, and mass manipulators arise... and that what happened IMO.
Why Ubuntu worked: millions to professional guys.
Let s suppose you re a dev: everyone asking you to solve many problems you don t even think they re prioritary...It takes your time, you begin to become annoyed by some guys, and as this is hard work and BENEVOLE (non profitable) you become pissed, slowly but constantly. And maybe some clever guys give you the forbidden fruit: just do the same, but for another distro, and I ll pay you juge tons of money and you won t have to deal with 2 or 3 lifes in one.
the community grows crazily and their demands too. the devs are highly sollicited ad they lose their patience, SO HUMAN.
It s easy to get there might be a problem.
Now, let s assume i m a really interested guy: I could use Gentoo projects, Ideas, and great community for other purposes. But even if I don t do this, come on, just reading what is happening here, or having a look at gentoo, frums, it was SO EASY to crack the match and make everything fluctuate... in such a scenario of being interested, I could try to be a saver, or I could try to make devs depart to another distro (mine?), I could try to make things worse animating the discussions from time to time, and AS WE ARE HUMANS, let the things get worse in communication, cohesion, and all the dev that reached a non return point could fefinitely leave.
Then, I ve reached my objective as Big666: revolution, anarchy, division, and seeing that things were not perfectly stable, this small deormation leads easily to a possible catastrophe.
Now, being the creator of gentoo, I might feel this, and it hurts, it is a lot of pain. I want to change things, and I might be convinced that something has to change.
I reallly think the problem is money.
I also consider gentoo desserves recognition in the society standards:
I would create an annual fee to access the forums. Symbolic. Like $10 a year, dunna. Just keeop the idea. The distro is free. "our support" isn t.
This money could pay the devs, the core devs, to gather other talented guys. And what about paying a fee for the HOWTOs, too? A symbolic tax to access the How tos? Something everyone could afford and without feeling schoked, like 5$ a year?
This money would help Gentoo.
The money is CRUCIAL and is the essence of the problem: I could easily imagine some people thinking "OK I HAVE TO BEAR ALL YOUR "STUPID QUESTIONS" AND I DON T HAVE TIME TO WORK ON IMPORTANT THINGS but at least i m paid for it, so it s ok, I ll cope with that, because somehow it helps me in my life too. I have to work like a dog day and night and my wife is just getting bored but this will pay our 3 weeks travel next October and I ll make her happy so it s fine... "
The main problem here is money. We don t have too much time in our lifes, and it s part of a sacrifice to contribute here. SORRY: it s not a sacrifice provided I LOVE AND ENJOY what I m doing. Many devs might reach a critical point, because gentoo is a great distro, and linux increased a lot so a big mass of requests has been done... I m REALLY SORRY FOR MY ENGLISH, I m sad and I want to say a lot of things and words don t pop up into my mind as i wish.
About Daniel Robbins: I have no clue of what his conditions are... please enlighten me. guys. One thing I know: he created it, he loves it, he wants the best for it. Maybe he is wrong in some aspcts, but maybe trying to find a concensius could be cool. We might not need a leader, but a Council of leaders. I don t know the names of them, but I trust comptenet guys to decide for them. And the devs shouldn t be worried: if such a thing happens, that will flop...
He desserves recognition for what he did, everyone desserves recognition. And a lot of us need money.
Apple and Ubuntu have money. They don t hesitate on spending money. Then they are strong.
We can work for free but then we have personal problems, and these may alterate our mood if we see a lot of guys asking "stupid" things (from a dev point of view)
Points of view DIFFER, AND? It s OK. I mean, we need to differ. But we also need GREAT PROFESSIONALS to state strong ideas that other users will follow.
I never felt annoyed by devs, nor by users here. Everyone is just willing to help. The howtos are THE great database over internet.
There s a problem. But it s also a strategy for personal purposes to magnify that problem; maybe Daniel wants to come back and having great recognition, ok... maybe another distro plans to grab great talented devs in their new world giving them money and stability and time.
My english leads to confusion, many things aren t clear, I m sure of that, so jsut say it and I ll try to explain myself better. Soory again.
And remeber the idea of paying 15$ a year for FORUMS AND HOWTOs.
How much money would that bring? Would the users stop going to the forums? Defintiely not. It s just one click a year. And you give it to help the community. To help you, to help me, to help us. And stability could come again.
NAming leaderships could be great too, to discuss about the future lines of the distro: why do we love Gentoo? Because it was focused differently, it is state of the art, it s different. The basic ideas came from a guy, or some guys. Let us allow these guys give their opinion and maybe some freedom to act and make gentoo even better.
Daniel, I don t know the conditions you set, I wish I knew them, but could you think again of them and maybe rerach an accpetable concensius?
You want Gentoo to be what it was again, I guess. Devs fear this. Maybe reassuring them would help?
Sorry to interrupt.
(well... I laugh now anyways)
