Why companies don't support Debian (LinuxWatch)

Hmm... The article seems to be saying that Debian is doing a bad job at making money for it's
community. But Debian's community just won't become wage slaves for Debian; they like the
freedom volunteering allows. Hasn't this lazy tactic of describing a volunteer group in ways
that would normally be used to describe a failing for-profit business been used many times
before?

Debian is always in a "crisis". Gentoo is always in a "crisis". Volunteer organizations are
both strong and fragile. Strong because they focus the common interests of a group of people
to meet common goals. Fragile because if the original goals are lost, volunteers disappear,
much faster than paid employees in similar project scenarios. So, Debian's community, by being
vocal and fighting tooth-and-nail to keep true to their vision are actually not putting Debian
in crisis, but are solidifying the mortar that keeps Debian's volunteers together to create
and maintain a productive project.

The article reasons solely from the ill-informed premisse:
-Debian is mostly made of anti-corporate volunteers

The truth is that Debian is made of debian developers (and a large network of non-dd besides
that), who all have their own reasons for participating (including: being employed by
corporations to do so).
Very few of these reasons actually include an "anti" agenda, rendering the article useless.

The main point seems to be that Debian is not exactly what the author thinks it should be, so
he writes an article that contorts facts and history to create an opportunity to bash Debian.
That is at least how it struck me.

So, I'm guessing that this has been suggested before, but why not take contributions centrally
and distribute them through community-elected fellowships?

Of course, it may also be the case that there is no problem to solve here...

What the article describes as a disadvantage is actually Debian's greatest 
strength: Having no significant money involved leads to full independence. 
Debian can't go bankrupt (think Progeny) or be bought out by a competitor 
(think Novell ;-) )

The same goes for Ubuntu: It's essentially a gadget of a dotcom 
millionaire, who also afforded himself a space flight. If he finds himself 
a new hobby, Ubuntu is doomed.

No other distribution than Debian (except maybe Slackware) has a track 
record of delivering quality releases for fourteen years.

This is a horrible inaccurate article with extreme bias.

I don't complain about lwn posting crappy articles often at all. But this is one that
shouldn't have been posted.

I will do my best to be civil.

This is probably the most mis-informed article I have read in some time.  And it comes from
someone who I thought understood his topic matter.  I guess I was mistaken.

As a long time user of Debian and not a developer or maintainer, I can say with confidence
that from where I sit Debian is not unfriendly toward corporations or other organizations, but
is unfriendly toward anyone who intends to restrict the freedom of us, the users.  The author
uses as one of his skewed examples the Firefox issue.  The fact is that the Mozilla Foundation
licensed their artwork in such a way that it is in conflict with the Debian Free Software
Guidelines.  An amicable agreement could not be reached regarding the licensing impasse with
the result that Iceweasel was created to protect the freedom of us, the users.  

It really wasn't the first choice of the Debian developers to do it this way.  The Mozilla
Foundation left the project no other choice if the browser code was to remain in the Main
Debian release and the project would continue to adhere to its core principles.  (In fact, the
Mozilla Foundation has certain restrictions on the use of the name FireFox, as I understand
it, which also conflict with the DFSG.)

This is the point the author consistently missed while he tried to build a case against the
Debian developers' alleged hostilities toward corporate entities (never mind the fact that the
majority of Debian Developers that I'm aware of have some means of gainful employment), that
first and foremost on the minds and in the actions of the Debian developers is the protection
of the software freedoms outlined by the Debian Free Software Guidelines for both themselves
and us, the Debian users.

With the list of Linux distributions known by Distrowatch, Steven, don't you think there is
room for just one that values software freedom above corporate expediency?  Evidently not.

Leave Debian alone.  It's just fine as it is.

P.S. Perhaps the author should look up Software in the Public Interest at
http://www.spi-inc.org/ if he would like to make a hardware or monetary donation to Debian.

This article is really poorly informed, and I'm surprised at that as I 
don't recall Mr. Vaughn-Nichols being quite so hasty to publish before 
checking facts when I was Debian Project Leader.

Companies can and do support Debian, not only through donations of 
hardware, but conspicuously by supporting DebConf.  For years it has been 
the practice that a significant amount of DebConf's budget has gone 
sponsoring travel expenses to our annual conference, in direct 
contradiction to Mr. Vaughn-Nichols's assertion.

The DebConf organizing team has for the last several years published a 
detailed report, which includes budgetary matters.

We also make no secret of our sponsorships; journalists who attend DebConf 
would be hard pressed not to notice the ubiquitous T-shirts which list the 
year's numerous sponsors on the back.  (While many developers wear their 
hair long enough to obscure such a listing, this should not present an 
insuperable obstacle to the assiduous fact-gatherer.  At DebConf 5 in 
2005, for example, the sponsors were also listed in the official program 
of the conference.)

Not only is it not the case that "there's no one to write a check to", but 
big firms like Sun, Google, and HP have long known that, in the United 
States, Software in the Public Interest, Inc. (SPI), is the 501(c)3 
corporation of first resort if you want to Debian out.

Checks from Google for Debian, PostgreSQL, and other Summer of Code 
projects come directly to SPI, and the funds held in trust for the member 
projects.  Michael Schultheiss, a friend and fellow Progeny alum, is the 
current SPI Treasurer (as was I, at one time), and could no doubt clarify 
many of these matters.  As could Sam Hocevar, Anthony Towns, or other 
folks named in the article.

That the piece calls itself out as an "opinion" should not relieve the 
author of all responsibility to check his or her facts.  That big name 
op-ed writers like David Brooks and Jonah Goldberg (just to name two 
examples) fail to hold themselves to such a standard does not make them 
worthy of emulation--it makes them figures to be transcended.

I too would like to ask LWN to filter out these kind of "news" and keep them from being spread
further. This guy is so far off it almost hurts. 

I worked on collecting funding for both Debian Conferences and other things within the Debian
Community for three to four years and found that when you approached the companies with
specific requests and a sound plan for spending the requested funds that people were very
forthcoming and supportiv. Debian is regarded highly within the industry and its work is very
appreciated. It is a stong brand with positive associations (stability, reliability,
enthusiasm, vision, altruism, quality) attached to it. The author of that article is known for
writing in a dismissiv and negativ way about Debian. So much so that one would suspect a
hidden agenda by now. I doubt that facts could throw him off. 

The Linux-Watch author may be wrong in some points, he completely missed some others, IMHO:

  * From a business perspective, a Debian installation is hard to support. 

This has two main reasons: first, it's extremely hard to define 'a Debian install'. Because of
the complete freedom to the user, you'd have to do a lot of investigation to ascertain what
configuration actually is installed. Second, Debian systems tend to be more customized (less
standardized) than other installations of other distro's.

  * No release cycle.

Corporations like predictability. Debian releases seem to happen randomly.

  * However, one thing is certain: slow release cycle.

You may not know when Debian is going to release, but you can be certain of one thing: the
release cycle is slooooowwww. This either forces you to run a development release (testing or
unstable) or at the very least compile some custom software yourself (the kernel comes to
mind).
Please note almost all distributions make a development version of their distro available.
Think Rawhide, think Opensuse factory. Most users of thos e distributions don't need to use
the development versions, while lots of Debian users seem to run testing or even unstable.
While you may argue testing is 'good enough' for most purposes, you really only want to run a
stable distribution in a business environment.

  * Conservative

Debian is created from a very conservative standpoint. The base system is guaranteed to run on
very modest hardware. While this is a good thing, businesses won't care. Their hardware will
almost never be older than 4 years.
This wouldn't be too bad if this wouldn't affect Debian on modern hardware. But it does. When
a developer assumes his software is going to run on moderately modern hardware, he can offload
the user and take his workload on the computer, for instance by creating a more efficient (to
the user!) interface to the software.
Even without user interaction, you still have to deal with a fundamentally different
environment in contemporary hardware.

  * Freedom without compromises

From a political point of view, Debian is completely right: Be Free!

However, when this harms business on the short term, principles tend to be thrown out of the
window very quickly. So they want their driver right now, and bug free. Not alpha 0.6 of some
Free reverse engineered driver, which may reach stability in another 6 months. Or so.



Debian is a very good distribution, for its purpose. But its purpose isn't to be used in a
corporate environment. It's developed to cater it's developer's needs, and none other. This
makes Debian an excellent distribution for developers (eg the audience of this site), and an
excellent basis for more business-like distributions like Ubuntu.

I think this kind of specialization is a good thing. It makes software ideally suited for its
purpose. A very broad, generic Linux distribution essentially would be good at nothing, it
would just not suck in almost all areas.
 

He's at it again...

Steven J. Vaughan-Nichols is someone who does not like Debian. For those of you who didn't
know so: just search on linux-watch.com for Debian, and see for yourself.

Of course there's nothing wrong with having an opinion; I'd just suggest that nobody take his
writings as "objective"...