Comments on: Gentoo's decline: A case of missing leadership (or rising Ubuntu?)

Is Gentoo's decline a question of leadership or a question of Ubuntu?

[The_Decider]

No Ubuntu is not good for Linux. In nearly every way it is like MS. It produces a bloated me-too OS and thrives only on marketing.

Linux desptop would be much further along if Ubuntu didn't exist.

[ferretboy88]

Most people that use windows computers have to work for a living and don't have months to waste setting up a Gentoo system. Ubuntu is a great way for people who are new to linux to get a taste of something new and better. Gentoo has not had much of a support base for a few years now. The creator of linux said he doesn't like Gentoo. He rather use Fedora. If I still lived in my parents basement and spent all day long configuring Gentoo I would not have a life or a girlfriend.

[deepwave]

As an ex-Gentoo user, I'd say that Gentoo never had a real chance as a serious Linux distribution. From a user perspective, compiling source code to install the latest version of a software is a pain. Most developers are not interesting in hacking on all parts of the software stack either. The majority of developers are interested in a single aspect or project. Finally dealing with all the configuration of a system is real PITA. And the fun of having a single update wreak havoc on your nicely setup system, eventually drove me away from Gentoo to Ubuntu.

Ubuntu would of won anyways, cause Ubuntu is Debian done right.

[dberkholz]

There's always a "hot new distro." For a while it was Gentoo, now the flame's moved on to Ubuntu. Who knows what will be next? The "hotness factor" affects many developers as well as users. After some years of getting screwed over by a few of those poisonous people you hear so much about, I think Gentoo's renormalizing to its new status as a smaller distro with a unique niche.

[trentreviso]

It is silly to claim Ubuntu has been bad for Linux. Ubuntu comes as close to a Linux desktop for the masses as any distribution of Linux ever has. Of course, there are all kinds of things about Ubuntu that could be improved. Of course, there are all kinds of things about Linux that could be improved. Both are works in progress. Neither will ever be all things to all people.

Ubuntu is not responsible for any decline in Gentoo. Ubuntu's success has not come because millions of Gentoo users have abandoned Gentoo for Ubuntu. Ubuntu has drawn its user base primarily from disaffected Windows users. To the extent that Ubuntu has drawn attention to Linux generally, increased enthusiasm for Linux, and expanded the Linux user base, it has certainly been helpful to Gentoo's cause, not detracted from it.

Many Linux distros are too small to survive much beyond the devotion, enthusiasm, and industry which the founding members of the distro impart to the first few years. Gentoo may or may not prove to be one of these. But Gentoo's problems are in spite of Ubuntu, not because of it.

[robbat2]

I left this comment for James Bellenger as well, but I'm posting it here for wider distribution.
There are some serious errors in his numbers, esp. with regards to number of developers.

========
I think you have some numeric errors. I'd like to speak to those, plus the general perception here. Of the Linux distributions I've seen, they tend to have explosive growth in the early years if they are going to be successful, followed by much longer periods of slower but sustained growth.

Firstly, in measuring the size of the tree simply by snapshot tarball, you neglect that we removed all of the $PKG/files/digest-* files at the end of January this year, as the final removal of the old Manifest1 system. This saved 3MiB in the snapshots overnight when it was done. We also used to permit much larger patches in the tree, and they are now on the mirrors instead. It used to be any patches larger than 400k had to be on the mirrors, now the limit is 20k, and we'd like to reduce it to 10k. Right now, there is 1.8MiB in patches larger the limit, which would be a reasonable saving, but not anywhere near as much as it used to be years ago.

I'd like to see somebody dig out old data and chart the number of packages that Gentoo has had over time, with markers for when things like X and KDE went to split versions. Plot it both as absolute graph, and as a relative graph

Secondly, there is something badly wrong in your counting of the developer tally.

Your 'developers retired' column adds up to 633.
However, if you count the number of active and retired developers, there are only 654 developers in history of Gentoo. 407 developers are presently retired, and 247 are listed as active. By the 654 total, less your 633 loss, we should only have 22 non-retired developers, which is VERY far from the truth.

Your join counts are also too low. Here's the join data pulled out of LDAP:
9 gentooJoin: 2001
78 gentooJoin: 2002
87 gentooJoin: 2003
143 gentooJoin: 2004
103 gentooJoin: 2005
70 gentooJoin: 2006
45 gentooJoin: 2007
27 gentooJoin: 2008
It's off by 4-5 developers for exact year, but the total is accurate because if they left and came back, the join date is their most recent return. It also adds up to 562, as there are about 90 old developers that we don't have join dates for.

I don't have LDAP data recorded for when old developers retired, I've got one data source starting at about mid-sept 2007 for recording the exact dates, but nothing from early on. Please also note that it's only in the last 2 years that we started heavily retiring for inactivity.

How were you counting data from bugzilla? Did you just go by the summary line, or did you read each bug? There is no consistency in bugzilla between reuse of a bug for recruitment and retirement. Sometimes they are seperate bugs, sometimes not. The bugs might also get reused for disciplinary issues. Not all the bugs are public either, because sometimes the developer has their private info in there (like birthday or real-world contact info).
========

[pablonhess]

I've used Gentoo since 2003 and have followed most of what has happened to it. I like it less today than I used to in 2003-2005.

I really don't think Ubuntu has eroded Gentoo's user base.

As I see it, Gentoo's largest failure was not being able to give birth to a user-friendly distro. I think Gentoo is quite similar to Debian in lots of ways. However, Debian has always been the basis for dozens of new distros for every little niche every year. Gentoo hasn't. Its only major spin-off is Sabayon, but there could be room for many, many more.

Gentoo is a VERY GOOD meta-distribution. It might even be that its only real value is its "meta-distribution'ness", but it remains unexplored. But why? To me, THAT is the real question.

[ferretboy88]

Linux will never win the OS war if it has all of this infighting. Cut it down to one or two distros and then have armies of people work from there. having 1000 distros and all these attacks on Ubuntu from the dudes living in the parents attic is doing nothing but hurt.

[thiagords]

I am a Gentoo user. I used Ubuntu for some time and came back to Gentoo, for just one reason: Gentoo has the best package manager ever, and it is called Portage.
I never saw in any other distribution a package manager which is so robust as portage. It has a extremely good dependencies tracking and offers a easy install and uninstall mechanism (it takes a bit longer than other package managers though, because all packages must be compiled).

I think that the combination of easy-to-use linux distro, like Ubuntu, which provides a powerful package manager, like portage, has the chance to go really far in the OS war.

[Squichie]

Ubuntu I think is good for the community, that is, if you think attracting users is good. I tried Gentoo, and while it was cool, I stuck with Ubuntu because it has a bit more vision, and it attractive for other users. Its interesting, most of my good friends who are tech savvy, I have got them using Ubuntu now. Lets also consider that they are all entering their first years of computer science, its great to know that our up and coming developers will have Linux in the back of their mind. Eventually, I hope to move the company I work for to ubuntu. I am surprised at how many people have actually used it from where I am from. Out of a staff of 80 or so, I can count 9 people who have chosen it as their main OS, and many others who have at least tried it.

[Quag7]

The interesting thing about Gentoo, and I've been using it exclusively on my desktop since 2002, is, with the possible exception of the KDE 4.x delay, I would never have known there were issues with the distribution, had I not read articles such as this. From my perspective as a desktop users, things pretty much chug along as they always have.

The forums are still quite active, the #gentoo channel on Freenode presently has 951 people logged in, and my system still works great for my needs.

What saddens me about articles like this is they tend to dissuade people from giving it a try. If what ails Gentoo is lack of "heat" (and I'm not sure long time Gentoo users see this as much of an issue), these stories kind of perpetuate more of the same.

But as for a desktop distribution, it works fine for me, and while my world would not come crashing down if Gentoo suddenly got blasted into space by an asteroid strike, I've seen nothing in six years that has enticed me to switch. I still try other distributions in virtualization, and I like some of them, but they tend not to give me, on the desktop, anything more than Gentoo does, even in Gentoo's apparently (so say articles like this) embattled state.

Gentoo is unique enough that I don't see it going away anytime in the forseeable future. For many Gentoo users, nothing else quite plugs into the needs and wants of its user community.

Gentoo maybe needs a PR person or something, I don't know. But functionally it works just as well now as at the height of its "flavor of the month" popularity from a few years back.