Your Choices with Gentoo
There is a large interest among the Gentoo and larger Linux user community for me to come back as Chief Architect. And recently, I found out that in theory - maybe with a bit of fighting - the position was still available to me, as I was still registered as the President of the Gentoo Foundation up until recently.
However, there's lots of reasons why trying to jump back into the Chief Architect role this isn't the right thing to do. First, for better or worse, the trustees have been overseeing the organization of Gentoo for over three years. And the developers have set up a council to provide technical direction for Gentoo (the council is elected by the developers.) The trustees and council haven't asked me to return to the project as Chief Architect, and I do not expect them to any time soon.
So, if I were to return to the project, I would need to return by force, by claiming that the trustees are not the legitimate leaders of the Gentoo Foundation since they did not file paperwork to remove me as President. Then, it would not be over - you see, the trustees have no control over the technical direction of Gentoo. So I would then need to take over the council.
Sure, it would be exciting. A lot of you think that this epic battle is long overdue. In some sense, you may be right. But this undertaking would also be incredibly political, an unfortunate waste of time, and disruptive to everything. And that's just the kind of thing that most people think Gentoo development has too much of these days. So in trying to do good, I'd just be adding fuel to the fire.
The solution, I think, is to look for positive ways to improve the Gentoo community without trying to personally go where I'm not welcome, and not to try to do it all by myself. The Gentoo community can be improved in many ways without challenging the authority of or even involving the trustees, council or official Gentoo developers. There are many things that can be done to help Gentoo that don't involve me putting myself, my family, and the Gentoo organization itself through a tremendous amount of turmoil.
So, am I doing stuff behind the scenes to help Gentoo? You betcha. But I'm not biting off more than I can chew, and I'm not doing anything subversive. This open source thing is a marathon, and the first and most important lesson is to not burn out and look for positive, sustainable ways to make a difference.
But, OK, some of you don't want to wait. You want immediate action. You want to see some blood. OK. Let's start by outlining some things that you (yes, you ) can do to help support the larger Gentoo community. This is a team effort, right? So let me contribute to the cause by giving you some suggestions:
1) If you don't like the way Gentoo is going and are feeling the urge to switch distros, consider Sabayon Linux. Sabayon is a desktop-oriented Gentoo with great hardware support. Too bloated, you say? Geared towards desktop users, you say? Well, did you know that Fabio added a base "stage3+" gui-less install image to the latest 3.4 build of Sabayon, which is basically an enhanced vanilla Gentoo? And this basic install is even a more recent build than 2007.0. If you're sick of Gentoo the organization but not comfortable leaving Gentoo the distro, make a statement by using the Sabayon base install and help to make it a long-term viable alternative to Gentoo proper. Let people know that you are using Sabayon, not Gentoo. Sure, Sabayon is a derivative of Gentoo, but it's also separated enough that there is an opportunity to do something new. You can work towards building a solid user community around this new Sabayon offering. This new Sabayon base install is an opportunity to innovate and fix some things that you may not like about Gentoo - both the technology and community. With sufficient user involvement, it can and will reach critical mass.
2) Personally, I think it sucks that the Gentoo trustees and council do not have any mechanism to get feedback from Gentoo users. Yes, users are by far the large majority in the Gentoo community, but they are woefully under-represented and the Gentoo organization is horribly out of touch with its users. Developers have quite a bit of prestige and power, but is it OK to leave out the users? I don't think so. So, let the trustees and council know that you want true user representation regarding the technical and organizational direction of Gentoo. It's something that is long overdue. And you have the power to make it happen. Let the trustees and council know that you want your voice to be heard, and that you want accountability.
3) Are you really motivated to fix Gentoo? Then become a Gentoo developer. Gentoo developers as of late don't have a good reputation in the open source community, but this doesn't mean that all Gentoo devs are rude and unpleasant. There are quite a few devs that are extremely competent and polite and quietly get things done while the unpleasant ones are busy flaming users on bugzilla and getting into arguments with each other on mailing lists or irc. If you want Gentoo to be better, then let's get a critical mass of genuinely friendly and helpful people developing Gentoo so that the bickering and obnoxious devs get frustrated and leave. And if you are already a developer who wants the Gentoo ecosystem to improve, consider getting involved with the council, as a trustee, or work to give regular users more of a voice on the project. Gentoo the distro begins and ends with its users.
If you have any other positive ideas, please feel free to post them here. And let's get to work, shall we?
49 comments:
Go to war? To arms! Anyway, it is way to war,another civil cold nuclear war.
Sorry Daniel, but you have lost me here. I find your 'solutions' very immature, simplistic and riddled with assumptions without backing them up with facts. (e.g. Gentoo developers as of late don't have a good reputation in the open source community)
Oh and now that suddenly 'Sabayon is also separated enough (from Gentoo) that there is an opportunity to do something new.'
Do you still contest -as you put in your previous post- that it should receive the same support from Gentoo devs (e.g. in bugzilla)?
Daniel, thank you for all your great work. I wish you luck with your future endeavours.
P.S.
How can you ask people to become a dev and "if you want Gentoo to be better,
then let's get a critical mass of genuinely friendly and helpful people
developing Gentoo so that the bickering and obnoxious devs get
frustrated and leave."
Whilst you threw in the towel yourself this year after being a dev for one week?!
Aniruddha,
Heh. thank you for wishing me luck with my future endeavors. Isn't a bit like saying "please don't hit your head on the way out?" :)
In any case, if you don't agree with my opinions, that is totally fine, but in that case then the post wasn't really meant for you.
This post was for those people who are unhappy with the direction Gentoo has been going. I am suggesting (I think) positive approaches for people who want to see Gentoo ecosystem improve.
Obviously, if you are totally happy with the direction of Gentoo, then you are going to be a bit lost when people talk about how to fix things.
The purpose of my blog post was not to prove to you or anyone else that Gentoo has problems. There are a lot of people who firmly believe that Gentoo is going in the wrong direction. This post was intended those who have this opinion.
Also, if we're going to critique, I am not sure what is "immature" and "simplistic" about suggesting that Gentoo users have a voice in the future of their distribution. Care to explain in more detail?
-Daniel
Aniruddha,
To answer your second flame, I left because it was emotionally upsetting to witness the state that Gentoo was in when I returned. It literally made me depressed and upset. I created Gentoo and care about the project - I have invested a lot of myself into the project over the years and it is difficult for me to see it degenerate. Others will likely not have the personal connection to Gentoo that I have and may be in a better position to contribute to the project without being emotionally affected by its current state.
-Daniel
Since I use both vanilla Gentoo and Sabayon (for my wife's laptop), I thought I'd chime in here. I think all three of these are very positive and constructive suggestions. If I had time, I would work on all three (becoming a Gentoo Dev isn't feasible for me right now in terms of time, so I just lurk around bugzilla fixing minor issues for folks).
I do disagree a little bit about Sabayon being separated from Gentoo. While the Sabayon community IS very distinct from the Gentoo communit, they are tied together by the portage tree in such a manner that Sabayon depends very heavily on Gentoo.
In other words, if Gentoo were to go up and smoke and blow away the portage tree, Sabayon would break too.
In some ways, it might be nice if Sabayon could branch away and set up its own fork of the portage tree. That would be very expensive, but would give Sabayon more independence and shelter it from the rabid political (religious?) battles that seem to plague Gentoo.
Aniruddha -- if you don't think Gentoo Devs have a bad reputation, you've been burying your head in the sand. The slashdot article here illustrates this very well. Or at least the links in it do.
Robert, thanks for your comment and I do totally agree that Sabayon is dependent on Gentoo in a very, very big way. At the same time, the fact that it has some "distance" from the official project does allow for some new, independent development to happen on the Sabayon side, and I think that it could be pretty cool to see what could happen from that.
Eh - 1:37 AM - time for bed. Have a good weekend everyone :)
Off course I care to explain ;) With immature and simplistic is mean:
I find your recommendation for Sabayon is simplistic, because it leaves out other Gentoo based distro's and the possibility to fork Gentoo. I find it immature because it will only deepen the 'bad blood' between Gentoo and Sabayon. Therefor not providing a solution but causing people to 'trench in' further.
Furthermore I miss the 'Next Action part'. Point 2 sounds like a nice daydream. To get somewhere -and to mobilize people- however you need to point out a clear destination and the 'Next Action' necessary to get there 'You have the power to make it happen' just isn't enough. And to be honest it sounds childish to me. I haven't been addressed like this since I stopped watching cartoons.
In closing about point 3. How can you ask others to do something that you yourself hadn't the stamina to complete?
Daniel,
I have no intention of flaming. I try to remain with the facts and let you know I disagree, that's all.
@Robert
The slashdot link you gave says nothing. To knows if something is wrong you need to know more than circumstantial evidence. For example:
-How many Gentoo devs are retiring yearly?
-How many Gentoo dev are joining yearly?
Furthermore opinions from DW and Slashdot is one thing. On the inside Gentoo devs say 'there is no mass exodus of devs' .
http://forums.gentoo.org/viewtopic-t-541247-postdays-0-postorder-asc-start-300.html
Gentoo past months has lost many skilled and experienced devs for sure and a recruit can't fill the gap so easily.
What i see is that new and old devs are in some kind of hiatus.
They mainly working on bugs and just "get things working".
Nothing big, nothing amazing.
It has the feeling of rottenness in raining days. Everything is moving slow.
It's really strange situation to talk in first place.
Yes, Sabayon seems better place to be if you are looking for more creative spirit but their main target is desktop, no?
I'm not sure we can expect many innovative tools for the skilled player, at least not before they have a fairly large amount of devs. :?
As a newbie - my only experience with GNU/Linux is openSUSE and PCLinuxOS - I am completely lost.
I had the intention to learn GNU/Linux instead of being a desktop-user and for me Gentoo was a decent platform to take the plunge.
Now, with all topics and comments I have read here the last days I really wonder if I have to go the Gentoo way to learn GNU/Linux.
I have downloaded Sabayon 3.4, tried that Live DVD and it is no different than a decent DW top-5 Desktop distro to me. Where are the docs like the ones I find on the Gentoo site if I want to do some hacking with Sabayon ? Och yes they exist as a link to the Gentoo site... ironic isn't it.
Up to now the Gentoo site is the only site where I find complete docs which are written in a newbie-understandably manner for starting to install a distro from scratch
I am not at all interested in the political stuff at a higher level, nor the legal status of the project. But I acknowledge the point that such issues must be solved, and quickly, if the momentum of the project is to be maintained. But don't play it out in the open.
So, do whatever you feel best for you Daniel but do not forget that Gentoo has been the starting point of many a GNU/Linux user and that newbies will still be trying to learn GNU/Linux through Gentoo because there is no decent alternative for us.
But saying that the critical mass must revolt against the leaders -by singing the 'Marseillaise' ??? - is a thing not done. Solution come out of dialogs, not out of confrontations.
I am sure that you tried the soft way, the diplomatic way, the informal way but this article where an internal revolt is suggested leaves a foul taste in my mouth.
I hope it all cools to a manageable level and that newbies will not be left out in the dark looking for another way to learn GNU/Linux.
A.J. Baudrez
Wow, how refreshing it is to see someone with such a progressive attitude about Gentoo's direction. I especially like the bit about Gentoo's bugzilla. Unfortunately, there are too many unkind trolls hanging around there now so I refuse to contribute there now.
Daniel: You're absolutely right that Sabayon is driving innovation in the Gentoo community. One of the reason's I moved my wife over to Sabayon from Vanilla Gentoo is that hardware support is so much better. The drivers are much more up-to-date and lxnay has done a lot of innovative things with hardware detection (integrating Anaconda into Gentoo, for instance).
What I was trying to point out is that Sabayon isn't a safety net for Gentoo -- that if Gentoo ceases to be actively maintained, users can't just "switch to Sabayon and forget everything". Which is too bad. It would be nice to know that no matter what happens in the Gentoo community, I have an alternative that is very similar to what I've been using for the last three years.
Wamukota: if you're looking for a distro that's a good place to get your feet wet and has good documentation, you might check out "Arch Linux". It's a lot like Gentoo/Sabayon except that it's not a pure from-source distribution. (I guess if Gentoo does blow up, I will use Arch as my backup!)
Aniruddha: You seem to discredit the Slashdot article just because it's an outside source. But the whole point of my linking to it was to show that the reputation that Gentoo has in the rest of the open source community is very poor. Whether or not there is a serious problem, there is a widespread perception on the net that Gentoo is in trouble.
But I think it's very evident that there is a problem. I read Gentoo Planet daily and the whining, fighting, and complaining on the blogs (of Gentoo developers) makes that very clear.
The forum article you linked to actually makes this even more evident. How many developers have we lost in just the last six months?? And how many haven't officially retired, but aren't really active any more? And while Gentoo is also adding new developers to some degree, they are not as experienced, committed, or productive as, say, Diego and Seemant.
But I think the best evidence of a problem is that it seems to take longer and longer for a new package to get into the tree.
Does this mean Gentoo is doomed?
No. It's still the best stinkin' Linux distribution of them all and it's still alive and growing. But it's growing much more slowly than it was when I first switched from Slackware three years ago. I think it's clear that the internecine battles between developers are largely to blame.
it would be healthy to start working on something new.
Daniel,
I read your blog with interest but there is one point I want clearance about, please:
You are always writing/speaking about that Gentoo should "move forward". In at least my understanding you seem to have some ideas what this "move forward" means.
But you never made it clear and concrete what developments Gentoo need.
So, can you please make it point for point clear what the devs of Gentoo in your thinking have to do so that you would say "now Gentoo is moving forward"?
I am very curious about your answer.
regards
an interested Debian Stable and OpenSolaris user
OK, Aniruddha. You'll notice that at the end of my blog post, I say "If you have any other positive ideas, feel free to post them here."
I don't really see much constructive ideas coming from your critique of my ideas. I am looking for constructive comments. If you have any better ideas, please share them and the readers of this blog will seriously consider them.
The one caveat is that I would like these ideas to be things that we can do as a community, and that you can others can participate in.
Anonymous,
Sure, this is something I can talk about more in main blog posts rather than the comments. This is probably my first actual post that has gone into so much detail about Gentoo internals. As time goes on, I can start addressing these issues more.
Daniel,
I'd like to thank you for your contributions to Linux. I've been a user for a couple years now but am still very wet behind the ears.
Gentoo has been very interesting and Sabayon does everything I need an OS to do. I still have an opportunity to learn and enjoy Gentoo without spending to much time scratching my head with a puzzled look on my face.
I understand how disappointing it must have been to see some of the articles about the state of a distro you developed let alone the personal / character attacks. Sad really that jealous and insecure people are better at running down others than making themselves look good.
Peace
Hi Daniel,
I guess you don't know this but Gentoo already has user representatives. Look at http://www.gentoo.org/proj/en/userrel/ -- but I haven't seen much that I can attribute as having come directly from that program.
Hey Donnie,
Interestingly, Fabio Ecruliani (the creator of Sabayon) is listed as an active user rep. I think this might be an indication that the user rep program is not very effective. Fabio is not happy with the direction of Gentoo and has numerous serious unresolved issues with the Gentoo development team, and the Gentoo dev team has several serious unresolved issues with him (many Gentoo devs are convinced that Fabio is telling people to pretend they are using Gentoo in order to get support on Gentoo forums and bugzilla, and Fabio vehemently denies this. There have been flames back and forth about this.)
I think user reps is a good idea, but the question is - is it working, and who on the Gentoo project is responsible to make sure it works? Are the Council or Trustees making user representation a priority? Based on what I see, the answer is no.
-Daniel
Hi Daniel,
I had similar fears about gentoo's future in the past, guess it's like watching your kids come to age. :)
I left in 2001 and felt that after that the development process was alot less innovative and there was too much talking involved.
I thought about rejoining the dev-team and event talked with dan about new ideas for kde.
Soon I decided, that it's simply impossible because of time.
Watching the distribution go forward I never had to regret that decision. More specialiced developers could do the job alot better.
I always liked the way you influenced the community in a positive way but i was sometimes displeased by statements like "It is like that because it is the official gentoo-way".
Sounds like the development team leaders now use such statements habitualy. I do not read -dev so my impression might be wrong.
Let the users decide critical questions if the dev's can't make a decision whould be one approch, but online voting is not practicable nowadays i think.
But with Sabayone in mind I see a new role for gentoo as a solid base for fork's, like debian is for x-ubuntu or fedora for whatever-fork. I can imagine different fork's focused on server, desktop or embedded devices, with gentoo as the common denominator.
The gentoo developers could then put more effort into portage development and platform stabilization.
The development team leaders whould be in responsibility for archieving the forker's needs, which whould reduce their mastership role and so the power missuse.
I personaly moved to debian but I give gentoo a go every half a year or so to stay in touch. From a users perspective it still evolves in a positive way allowedly a little slow in terms of portage.
That's why I like the efforts done by paludis and pkgcore (I actualy landed here starting at http://lwn.net/Articles/240399/). It whould be nice if they develope a portage- and/or ebuild-core-api in C++ and Python which could be used by the two new and the original portage package management systems.
[kidding]Uhm, should i get my hands dirty again? I could implement the rpm and dep package backends i thought about in 2k[/kidding]
The SFC approach Grant makes seems to be the right way, wish good success with the paperwork.
Best wishes from munich
achim~
Hey Achim! For anyone reading this, Achim is one of the true legends of Gentoo who made many key contributions and even kept Gentoo alive when I took a vacation in the land of FreeBSD! :) Nice to hear from you again and feel free to send me an email or two every now and then, btw :)
I agree with your comments - I think Gentoo would be a good base for "forks," but it does not look like Gentoo wants to be in that role. Look at the bad relations between Gentoo and Sabayon as an example. Gentoo seems to want to resist change, but I am not sure why.
I have this idea of a producer/consumer model - where smaller teams produce a specific component for another "consumer" - and that consumer may be another project. So Gentoo could evolve into a distributed effort of small truly independent open source projects making components that are used to build up specialized distros by other teams. It is like your idea but with the spin that there doesn't need to be a centralized "Gentoo" that controls and influences everything. There's more freedom that way, I think. I think that this is the way that things will need to move.
Anyway, thanks for posting to my blog and it is truly a pleasure to hear from you :) All my contact info is now on funtoo.org so you have no excuse to not send me an email or IM me every now and then :)
-Daniel
Producer/consumer model, sounds better than fork-base, it makes clear that the developers are in responsibility for the consumers.
But splitting Gentoo in different subprojects does not implie that there whould be more manpower available to the gentoo-core team.
So the problem grant describes in his blog about developers spend all their time fixing bug's will not be solved that way.
Other main distro's developers are much more involved in kernel glibc or kde development than gentoo's developers are. That way they can be much more innovative and also they have alot less trouble with bug's.
The bad side of this blending is in my eyes, that documentation is not as important as it should be because the distros already have the specialists in house. And the thread exists that modifications become too distro specific, causing problems for uninvolved projects.
An enterprise gentoo (-forck) could solve that issue, it whould be possible to hire developers and let em work on the innovative and important stuff (including gentoo-core).
I don't know what the core does actualy mean in gentoo term's. I tought about portage and a minmal build system as their main competence, maybe together with a stripped down reference implementation of the os. I can understand the problems betweens the foundation based monolitic Gentoo project with Sabayon, which is more or less a one man project. Splitting gentoo development in different subprojects whould solve the liability konflict the Sabayone fork causes atm.
Enough old day talkin/thinking for now, i'll keep you up to date about personal stuff by mail.
achim~
I just wanted to mention something about Gentoo & Sabayon.
I initially tried Sabayon as a sort of way to play with Portage without having to go through the dreaded (and believe me, it is dreaded outside of the Gentoo community) Gentoo install.
I was initially very impressed by the hardware support, good looks, and bleeding edge features. I used it and that was that, but the bloat was getting to me. I decided that it was time to make the jump and do a proper Gentoo install.
The first one I did was with the GUI installer, which failed because the GTK+ installer is just not ready. The second attempt was a proper Gentoo install (stage 3). I thought that it was very easy (my familiarity with Portage that I gained through using Sabayon made getting the full system up after the install easy, too).
What I noticed from using a real Gentoo install is that Sabayon is broken. Sabayon's method of dealing with implementing its bleeding edge features is just plain bad. The ~x86 keyword is declared globally, and the world file is filled with way more packages than it should have, and an "emerge -u world" has a very, very high chance of breaking the system, hard. Naturally, the system is downgraded and loses its bleeding edge figures if you remove the ~x86 keyword.
My point is basically that while I really appreciate Sabayon and it does a lot of things right (great hardware support, all multimedia codecs pre-installed, supports KDE and Gnome, solid Gentoo base, and good looks) it just doesn't fit together smoothly enough for me to prefer it to a regular Gentoo install. For a lot of Gentoo users that are used to being able to do simple things with their system like update all out of date packages, Sabayon will feel awkward and messy.
As a side note, I think a lot of the "bad blood" between Fabio and the Gentoo devs is on Fabio. He's very passionate, but his passion is untempered, and he seems to me to have a sense of arrogance that adds to it. His just-barely-awkward English might worsen things slightly.
To give some context, so this doesn't seem like a random attack on Fabio:
http://planet.sabayonlinux.org/?p=43
http://ciaranm.org/show_post/151
I think that while the things Fabio say may not be inaccurate or untruthful, he lacks any sort of tact.
I agree that the ~x86 flag was troublesome at times, but not unworkable. With 3.4 business edition you get x86 flag but you lose some of the geewiz cool stuff. I guess there are tradeoffs in almost everything.
I run Gentoo on my server and Sabayon on all my desktops and laptops. I'm thinking of trying the core install feature vs a base Gentoo install on my next server project. The best part is that there were comments about this issue and Fabio and crew made it happen. Made it happen in short order BTW.
I guess my point is this. As a user I see Sabayon moving ahead, being responsive to their users and Gentoo just seems to be stagnating. I'm not a dev, I'm not programmer, I'm just a user. I'm just the guy that wants things to work. I'm just a guy who doesn't want to feel like someone who is developing a distro thinks I'm a pain in his backside. I get enough of that in other places.
I would agree that Fabio is passionate. I thank him for his passion. His passion helped me move away from OS choices that I wasn't happy with. I take it you have no Italians in your family? He's tame compared to my grandfather. However, I believe your comment about his English was not germane. How's your Italian?
Daniel, thanks for your work and passion with Gentoo over the years. Your work gave me my start with Linux. Ever thought about contributing to Sabayon? Best of everything to you and your family.
Daniel, I really appreciate your concerns over gentoo and appreciate your previous work on gentoo. I have one question with regards to your post: 'what are you doing behind the scenes?' You mentioned you were doing stuff, but what exactly is that? As a power user/Arch Tester/Potential Future dev I am rather interested in whats going on.
Best regards,
gentoofan23
4Walters, I agree with pretty much everything you said. The thing about Fabio's English definitely seems like a personal attack, and completely irrelevant to the discussion, and it should be both of those things. The reason I mentioned it wasn't to criticize his English, but because I've read a few posts in various places where people read something he posts about a release and think it takes away from the otherwise polished look and feel of the distribution. But you're probably right in that it wasn't worth mentioning in this discussion. I apologize for that.
--Patrick C.
Hi Daniel,
great post again.
Yes, the gentoo project trying to stay monolithic, which is unhealthy for the project as a whole. Sabayon is not the only example: The polarizing nature of the three package manager implementaions is another. This "either you are with us , or you are against us"-mentality is _very_ unproductive, because there is nothing to gain from it.
As for the "fixing gentoo"-part, I have some ideas for dev-rel/user-rel, but I need some time to present them in a convicing way. This will probably to long for a blog comment.
Yours, Björn
P.S.: As for the Fabio/Ciaranm "incedent" (although this is pretty OT), Fabio might be whatever is said of him, but an ego collision with Ciaranm is a very weak indicator of anything.
As a new face to the Sabayon community and the Linux community as a whole, I will begin by saying that Sabayon is the distro that helped me made the switch from Windows. In January, one of my friends (a Gentoo wizard) told me about it and as a Windows power user who was disgusted with the latest from Redmond, I eventually made the switch in April.
I've never looked back, and I think that has a lot to say for Sabayon and its community. They've been very helpful and the passion that Fabio has shown for the project is admirable to say the least. I understand that he is sometimes brash and comes across the wrong way to certain people, but..
1) People who post chat logs on the internet to flame someone are, frankly, quite pathetic.
2) Fabio has done a lot for the Linux community and a lot for users that want to get into Gentoo but just didn't have the initial know-how. There's no way I could have jumped through the hoops to install Gentoo. Sabayon was my window into it. For that, the only person to thank is Fabio, because he had that vision, and by deploying it, he and his team have helped thousands of would-be Ubuntu/OpenSUSE users into these parts of the Linux community.
So in the end, Fabio's attitude doesn't really matter, does it? He has a burning passion and he gets things done while a community of 10,000 continues to grow (Sabayon has recently jumped from #7 to #5 on Distrowatch.) It's still young, but Daniel is right: it wields potential for creativity and great new ideas, which I have witnessed first-hand.
As for you Daniel, I have a great deal of respect for you and what you have accomplished. You layed the foundation for an entirely new realm of possibilities with GNU/Linux. The obvious concern and support you have shown for a distro that no longer recognizes you as its creator is heart-warming, and an inspiration to all of us. Thank you for everything you have accomplished and for great things to come.
I can understand you wanting to help Gentoo and you would like for it to progress, but like aniruddha said, you don't do anything. Thus making it look like you don't care and all you are doing is writing and not actually helping. I am very thankful for you creating Gentoo but I believe you take it to far, in your role of owning Gentoo.
Though, I do believe more people should become developers, I will hopefully killing some bugs soon. All the help the better.
I love Gentoo, but I am recommending everyone install Sabayon. The install is quicker, and while the install is a bit bloated (still some redundant apps and such) you get a full system and then some when you're done. The real innovation is happening within the Sabayon team which is disappointing.
Honestly, Gentoo has needed a binary install for some time. If they were smart, they'd make lxnay a dev, and work towards making Sabayon the official binary installer.
And frankly, Gentoo devs should support Sabayon users, because it is Gentoo. It uses baselayout, and portage. Are they objecting that they can't support something because it uses an overlay or it isn't the stock config?
I imagine very few Gentoo users really have identical systems. That is what Gentoo is all about. Either you support people with varied configs, or you support none of them.
I think the Gentoo user community is the best, but I have never received any real help from a Gentoo dev, and several times I have been unfairly brushed off by them.
All in all, I think there is a major problem with Gentoo that has a very easy solution. The Gentoo team has to fight this image that Gentoo is unstable, and it is difficult to support such a varied distro.
So make the best possible stable system, with stable flags and configuration. Then package that precompiled in an installer like Sabayon.
And for the ~arch world, push for innovation. Work with many of the users in the Unsupported Software forums, because honestly some days it seems like they do more than the official devs. Bring them on board. Bring lxnay back on board.
If you really poke around dev's files you'll often see they are working on cutting edge stuff as well, but usually for personal use. They don't advertise, or publish it because they don't want to support it.
The solution is to be EXPLICIT. If you want support, stay stable. If you experiment, you can't expect the same level of support but in turn, the Gentoo devs should embrace and push for more innovation.
It is a damn shame that Gentoo was the last major distro to embrace GCC 4 for instance, and they did so almost a year after everyone else did.
Just my 2c worth - I really dont spend any braincycles thinking about the politics that happens up the tree - none of that influences any decision I make here.
What is critical to me is the portage tree, and the quality of my next emerge.
Thats ALL I care about really, and ALL that has ever attracted me to (and kept me with) Gentoo.
Its all about the portage tree.
It could be just my subjective opinion, but these days, in order to keep on the bleeding edge of whatever fringe development I mess around with next .. it seems more and more likely that I need to throw in some 3rd party overlay to get the ebuilds that I lust after.
This never seemed to be the case - like a year ago.
Whilst everything is still great, I have this irrational perception that my next emerge might not be what Im really looking for without finding a non-mainstream ebuild or overlay.
Is it just me, or have others noticed this ?
If it is the case, then what is the best way to get Gentoo back to being 'Your guaranteed source for the cleanest build of the latest big thing' ?
If there is anything I personally can do to help Gentoo acheive that, Id be more than happy to volunteer.
Daniel,
You're a genius, the work you did on Gentoo speaks for itself.
Daniel, you're failing to understand the most critical point about our desire for you to come back.
It's you.
We don't want technical solutions to various problems, or suggestions for various approaches. We want YOU.
You're the one who formed that wonderful community around Gentoo, and you're the one who held it together.
I speak as someone who tried to get involved with Gentoo. We set up an infrastructure group, targetted at stable server environments. We did all the work, then klieber came in and took over without doing a damned thing.
Considering that the initial group were formed of admins who run military environments and huge labs worth billions of dollars, we left very quickly when a snot picking dickhead was able to take over just because he was Gentoo core.
We're not asking you to dedicate huge amounts of your life to this. We're just asking for YOU to come back, and YOU can ensure reasonable process and fair judgement. Even if you spent 1 hour a week saying "yep, nope, nope, yep", that would be so much better than what is happening now.
Hey Steve,
There's better ways for me to contribute to Gentoo at this point than spending all my time trying to run the project.
I'm sorry you had the experience you did with Gentoo's infamous infra leader.
Really, the best thing I can do now is empower the community to grow and innovate, and I will be doing that. So, in a sense, I'm already back, just focusing my efforts in more productive and effective directions.
Wait a few months and I think you'll start to see things changing.
People always seem to mistake opinion for fact! Gentoo as it is now, has many weaknesses, but so does every other distribution. I use Gentoo in a huge corporate environment, and I use it at home. It works for me in both places because it is so flexible.
I too could care less about the politics of Gentoo. In fact the vast majority of users obviously care less, as is obvious if you consider the number of different people weighing in on discussion such as this.
I like the distribution and would like to see it continue it's success, but if it doesn't, I'll simply move to another distribution. Fortunately that's another one of the strengths of Linux.
The Gentoo development community is weak, at least compared to other distributions. People can argue all they want, but I have 20+ years experience with Linux and I can tell you that RedHat, Suse, and others are doing a better job than Gentoo at getting up-to-date applications in their repositories (stable ones too!), more quickly. Over the years I have found myself just compiling a lot more on Gentoo (outside of portage), because there's nothing in portage. I'm sick and tired of having to look in the unstable tree for an application that was considered stable by the app developer, 6 months ago! This complaint is often brought up in the forums and members get the same response...contribute to development or shut up. That's one of the dumbest answers I see continually repeated in the forums. Even some of the most expert Gentoo systems admins I have seen, do not have the skills or desire to code, let alone the time. The code or shut up attitude just shows that developers don't have a clue about the users of their products, and/or don't care. This attitude appears to me to be much more prevalent with the Gentoo community than with other distributions...just my experience. I think part of it is that a lot of those involved with Gentoo are young and inexperienced programmers...stereotypical computer geeks with no social skills! No, I'm not saying that applies to all programmers, but it obviously applies to many. I think when you get those involved who are using Gentoo, those are doing great things with it from a user standpoint, you get a more mature and more focused community. That's difficult to explain, but the idea I guess is that those using the distribution to do great things, have much to contribute, even if they aren't actually building the distribution itself. They need to be involved and listened too, not dismissed because they don't code for Gentoo.
On the other hand, I've found the Gentoo community to be as well educated and at least as helpful as any other distro's community. In fact, I'd probably say that the community (considering documentation and support via forums and such) is the best there is. Those comments may seem contradictory, but I can say that because in my experience the really negative things I have seen in the community, have not been enough to negate the unbelievable strengths I have seen, most of the time.
Daniel rants as much as anyone involved in the Gentoo project and there's nothing wrong with that. I don't know enough about the politics or care enough to figure out who is really right or wrong, and who has their facts straight! All I care about is what happens to the end product I use and if I don't like what happens I'll move on to something else. So will most other users.
Hello DRobbins,
I still remember the day that I found Gentoo, and then in surprise (though in hindsight not too much surprise) that it was maintained by one of the Stampede elite.
I would like to say what I liked about the original Gentoo. I still have a keen memory of all the things I liked.
It was freedom for me. Gentoo empowered me.
1) I was more directly tied into the source as it came. The ebuilds were simple, perhaps a few lines except where there was lots of room for customization. And the rest was done upstream. In fact, sometimes I could simply chance the filename and have an ebuild for the next release.
2) I could build my own distribution for my own purposes with it. I immediately left Debian, which at the time only supported ldapv3, GSSAPI through Turbo Fredrikson's excruciatingly over-burdened install. Pages and pages that seemed to scream to me, "this is what you have to do to get around the Debian developers making decisions for you". In Gentoo, I only had to enable the right use variables and it worked straight away (until openLDAP dropped the support and I had to pick it up through SASL).
3) I hate license wars. Back then Gentoo was the only way to set up what I wanted at work and home. With great delight that I found I had Alladin's ghostscript and not FSF's. And a number of utilities that I couldn't get elsewhere. And I suppose I still can.
But today it is only on my home computer and I have to admit that is because I do not want to re-install. PDF and PS support has improved a lot, most people are using Gnome, and I suppose I don't need LDAPv3 after all (*sigh* at least not when Samba4 is out). I am almost ashamed to say what really turned the tide was having three kids, and realizing that while I loved hitting the nuts and bolts that I had to leave that to others. And I was going to have to just live with their decisions.
If I were to work in Gentoo again, it would probably be to set up an applications engineering knowledge base of user-specific goals. LDAPv3? Sure then use these USE flags, select between these packages, etc...
Want clustering? Well then do you want just Storage or computational? Networked with Load balancing or not? Then these are your options.
What was great about ebuilds is that they enabled me to make those decisions on my own. Now such a knowledge base seems to be brought into the ebuilds and they are becoming Gentoo works rather than just conduits to the upstream. Under the weight of this Gentoo way of doing things, they are slipping behind and becoming harder remove their masks.
I hope that feedback helps. I am glad I found this site, and I will keep it in my reader.
Very insightful comments, On Lawn. I, and I hope others, will seriously consider them.
My advice to Daniel ...
Firstly, just get yourself back into an environment where you feel good about things again. Let the past be past, the present be the present, and get enthused about what is possible in the near future.
Get comfortable first.
Then just start coding again. Dont worry about what others are up next door - the world is always going to have its share of morons and petty politics, wherever you go.
Just code and get that smile back on your face. Do it for yourself .. and a community of like minded people with the same vision will gather around whatever you are doing, and support you.
Code.
Anything else is a diversion.
Hi Daniel, I've read most of these comments and I probably agree more with SteveAndLyn suggestions. Personally, I have wanted you back in the position you were for quite a long time. Gentoo has just seems so headless since then and wandering aimlessly. But that was the past and you have to remember why you left in the first place. Think of your family first and what makes you happy next. Otherwise you might end up in the same situation again. You/we don't want that again.
I've got most of my family on Gentoo and have been with it for a long long time. One friend that I got using Gentoo long while back, has tried other distro's and says he just feels powerless with them. But we have all noticed the lack of new packages in portage. I am creating a large app using wxpython/wxGTK and need 2.8 because printer functionality in gtk is broke in earlier versions. Is ver 2.8 in portage, no, but you can get it with other distro's. But, I can still use a local repository with portage and build my own. So can't complain too much. That still doesn't excuse the fact that just about everyone else but Gentoo has ver 2.8. I've got so many ~amd64's on my box and would actually rather not. There are many of those packages that run stable and have been tagged unstable for quite a long time.
I'm mainly commenting cause I have used Gentoo for so long and never actually talked to you and Thanked You for such a Great Product. I've used Mandrake, Redhat, Debian and a few other derivatives but haven't left Gentoo since I started using it. So Thank You Daniel!!
I for one though would like it if you 'took over' again. But we both know the vocal minority would be up in arms. No one likes to be told, 'Step aside, and let me fix it.' Even if they know their doing it wrong. I am sure there is something you can do to help out Gentoo, but please remember why you left initially and don't repeat history.
Sometimes you have to do what you don't want to, but when you do get to choose: Make the choice that makes you happy & fulfilled and takes care of you family.
-Decibels
I used Gentoo for a while on an old PC and had very positive experience, but found that compiling takes too long. Thus, I switched to Slackware, but is likely to be back after upgrade.
I never felt that users are neglected in any way.
I do not understand why it is necessary to listen to user more. I believe that Gentoo is (or should be?) a collaborative effort of its developers. Those who are not developers just do not matter, or, in other words, developers are Gentoo users.
If it is actually necessary to listen to users more, this job should be commended to payed developers, possibly of a new Getnoo based distro.
Hearing a lot about how Fedora is gathering statistics about their users I wonder if we could do something similar. For example make portage (and other package managers) report stuff like error messages if en ebuild fails(the amount of duplicate reports will help set priorities in fixing ebuilds) or installed packages which have different keywords than the default one (thus giving developers a clue about what packages users want to be stabilized for example). Or look at which overlays people are using and what packages from that overlays (thus giving a clue about what packages could be included in the tree). Possibilities are many.
But I don't know how complicated it would be to implement something like this.
"this is what you have to do to get around the Debian developers making decisions for you"
That sounds more and more like Gentoo now. :( My /usr/local/portage overaly just keeps growing and growing because of strange decisions made by Gentoo developers. I wish someone (or some group of people) would fork it.
Few thoughts about this gentoo war and the lack of new inventions in Gentoo.
So beyond doubt Gentoo is not for non-professional users. It is a distribution for those being familiar with computers, liking edit config files, manually installing things. Therefore the relationship between users and devs will allwayz be problematic. I'm not immersed in the battles amongst devs too much, but as i see some gentoo devs have too eltistic view of things.
If i understand Gentoo's mission statement correctly, Gentoo is meant to be a meta-distribution. The problem is that nowadays Gentoo does not fulfill this role. In order to be a meta-distrib, it has to have other (popular) distribs built on it.
I don't know Sabayon but that might later be one of them. So there ought to exist popular desktop, laptop distribs that listen to the end-users requests and channel them back to Gentoo.
So making the long strory short, if Gentoo wants to survive, its devs must have a good, cooperative relationship with those of Gentoo-based distribs. Devs must labor on accomodating the end-users needs, and the best way to achieve this goal is to have – at least -a user-friendly desktop distrib for the mass and a really meta-distrib for the strong fundaments.
And... an other problem with Gentoo.. there is no money it. Look at Ubuntu. A guy injected a lotta money into it and this time it is one of the most popular OS.
First of all thank you for creating such a fine distro.
When Gentoo was originally created it was the greatest thing since stainless beer.
Unfortunately things got a little strange since you left there and I had to resort to Debian since I got to get work done on my box at times...
Actually we are at a little disagreement how things could be fixed there.I think you still helping out or anybody becoming dev there for the abuse is the wrong way.
The whole thing has to crash and burn.Once the user base is down to a dozen the guys that are there to statisfy their vanity will drop out and then thing might look up again.
'You have to recompile gccbecause of:bug report' -> bug report is from 2005 for AMD64 arch;me runs i386 solaris (playing with it anyway) -> emerge --freebeer gcc -> C compiler cannot create executables
Oh - well at least the guys from prefix-bootstrap are trying something new
This post was a few months ago now, but I want to vehemently lend my support to the notion that if there's one person who should direct the future of Gentoo, it is its creator; the man who had the wonderful original vision, and who created the Gentoo philosophy (http://www.gentoo.org/main/en/philosophy.xml).
I think the ideal position would be as Benevolent Dictator for Life.
My own brief experience with the current Chief Architect (http://bugs.gentoo.org/show_bug.cgi?id=169969) has only lent weight to my opinion.
AHurst,
I am going to start to build and release stages soon. Will look into the Universal LiveCD option too.
Thanks for your post and voicing your opinion, and I'm sorry you didn't get a great response from Gentoo proper.
How about a Gentoo Fork?
Post a Comment