The future of Linux: what it means for Wikipedia (apc) [LWN.net]

apc covers an LCA talk by LWN's Jonathan Corbet. "Kernel release 2.6.24 came out on January 24, just before linux.conf.au began. Corbet estimates 2.6.25 will be finalised sometime around April. That rapid cycle represents an astonishing volume of new code. "We are adding about 2000 lines of code to the kernel every single day of the year, without exception," Corbet said. "Nobody can really keep up with this [on their own] any more. It's an amazing process, and it seems to be working." The project which those numbers immediately bring to mind is Wikipedia, which uses similar open source principles, along with an "anyone can contribute" ethos."

to post comments

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 18:40 UTC (Thu) by einstein (subscriber, #2052) [Link] (3 responses)

The comparison to wikipedia seems seriously flawed to me. Are you really taking the position
that any random bozo can shovel random crap into the linux kernel in drive-by fashion? That
would be a propaganda coup for the anti-linux crowd, for sure.

Fortunately, the linux development process doesn't work anything like that.

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 19:01 UTC (Thu) by rsidd (guest, #2582) [Link] (2 responses)

I agree.  Anyone can send their contributions to linux, but it has to go through several
filters before Linus will add it to his "official" tree. This is what is missing in Wikipedia.
It is remarkable that it has done as well as it has; a year or so ago I found Wikipedia a
superior resource to most traditional references, but now I find random crap in pretty much
any article I see there.  (It is still a superior resource, provided you are able to filter
out the crap and willing to double-check every fact you need.)

Someone could try starting a "wikilinux" -- a centralised source tree repository that anyone
can contribute to or modify, with similar minimal safeguards as Wikipedia -- but it probably
wouldn't even build, much less work for most people.  May be an interesting experiment though.

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 21:46 UTC (Thu) by smitty_one_each (subscriber, #28989) [Link]

Also, the mechanical learning curve for making Wikipedia contributions (irrespective of domain
knowledge of the topic being edited) is trivial compared to that of doing kernel work.
Of course, one could always whine for more introductory material for Linux: waaah!

The future of Linux: what it means for Wikipedia (apc)

Posted Feb 8, 2008 17:20 UTC (Fri) by kingdon (guest, #4526) [Link]

My experience with Wikipedia is somewhat similar although I guess a bit more positive. You can find problematic stuff there (and cleaning it up is one of the hobbies of the dedicated wikipedian), but for me the amazing (and surprising) thing is that Wikipedia works as well as it does. Oh, and with policies, blocks, reverts, page protection, etc, it isn't quite as anarchic as it appears (although I suppose that is one of those "glass half empty or half full" things).

As for "wikilinux", there's forkolator and a long time ago Ward Cunningham had some kind of wiki which let you edit perl code on wiki pages (although I'm not sure whether/where this was published). And of course there is fitnesse where people (not necessarily the whole world) edit acceptance tests (sort of a kind of code) via a wiki. But all of these are very much embryonic in terms of whether there is a a workable concept lurking here.

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 19:02 UTC (Thu) by sjj (guest, #2020) [Link] (7 responses)

It doesn't seem that the Wikipedia comments were made by our esteemed editor. To me it sounds
like they were tagged on by the writer of the apcmag.com article.

Can you clarify this headline and blurb, please?

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 19:37 UTC (Thu) by sfeam (subscriber, #2841) [Link] (6 responses)

I agree. A similar request came up a couple of weeks ago. If the LWN 
headline is actually a quote from some other source, rather than a capsule 
summary by the LWN contributor, could we please see it in quote marks?

In the present example, I had no idea what the parenthesized (apc) meant 
until after the fact.

Tagging of press articles

Posted Jan 31, 2008 20:48 UTC (Thu) by corbet (editor, #1) [Link] (5 responses)

It is interesting to me that this sort of confusion seems to be arising now. The "title (source)" convention for reporting on articles found elsewhere has been LWN's way since the beginning, and, for the first 9.99 years, nobody complained. If there's now a better way to headline such articles, we could certainly change, but the prospect makes me just a little nervous.

Tagging of press articles

Posted Jan 31, 2008 21:11 UTC (Thu) by hathawsh (guest, #11289) [Link]

Well, (apcmag.com) would be lot clearer.  Just (apc) makes me think of a UPS and all the other
TLAs I'm already swimming in.

Also, the lead-in "apc covers an LCA talk..." does not warn the reader of just how much
reporter opinion got mixed into that report.  Something like "apcmag.com wrote an opinion
piece regarding an LCA talk..." might serve better.

Tagging of press articles

Posted Jan 31, 2008 21:12 UTC (Thu) by ncm (guest, #165) [Link] (1 responses)

I don't think the problem is the title.  It's the segue from apc's quote from  you to his own
editorializing.  An inserted "[apc digresses to add]" after his quotation would make the
transition clearer.

formatting nested quotes

Posted Feb 1, 2008 16:40 UTC (Fri) by jabby (guest, #2648) [Link]

I think another source of confusion is the use of double-quotes within double-quotes. Even if an original text uses double-quotes, if I'm quoting it within something else, I'll usually convert them to single quotes and use double-quotes for the outermost level. If there are quotes within quotes within quotes, I just keep alternating with each nested level.

This approach increases the visual parseability, removes ambiguity, and avoids confusion should a typo result in the accidental omission of one quote mark.

For this blurb, it would have looked like this instead:

"Kernel release 2.6.24 came out on January 24, just before linux.conf.au began. Corbet estimates 2.6.25 will be finalised sometime around April. That rapid cycle represents an astonishing volume of new code. 'We are adding about 2000 lines of code to the kernel every single day of the year, without exception,' Corbet said. 'Nobody can really keep up with this [on their own] any more. It's an amazing process, and it seems to be working.' The project which those numbers immediately bring to mind is Wikipedia, which uses similar open source principles, along with an 'anyone can contribute' ethos."

Maybe there's not a significant difference for other readers. Still, I think it helps.

Tagging of press articles

Posted Jan 31, 2008 22:41 UTC (Thu) by zooko (guest, #2589) [Link]

For what it is worth I'm one of the people requesting more noticeable markup, and I've been a
reader for more or less all of those years.

Maybe it's because my eyesight is weakening with age?

Maybe it's because the mainstream press is getting more prolific and inane, or because LWN is
becoming less selective about which mainstream press articles it references?

Regards,

Zooko

Tagging of press articles

Posted Feb 2, 2008 12:36 UTC (Sat) by roblucid (guest, #48964) [Link]

The attribution method has always seemed clear enough to me, and whether 
it's "apc" as their site style themselves or apcmag.com seems a 
triviality.

The real isssue is "Is that article news?".  The Grumpy Editor's talk is 
reported creatively with lots of Wikipedia comparisons (may be the journo 
uses it a lot for research?) and there's not the context of the actual 
talk, nor comment by the GE on whether the Wiki stuff was contained in the 
GE's talk.

Perhaps the GE getting quoted on other sites and reported on is news, but 
pieces on general background around Linux conf in the media, may be would 
be better presented in a batch, so those interested in mass media buzz 
surrounding the conf can get a general impression.  Especially if there's 
a link to the text of the talk, so reporting accuracy (versus 
creativeness) can be better judged.

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 19:29 UTC (Thu) by jordanb (guest, #45668) [Link] (3 responses)

> While the majority of kernel contributors (17%) aren't paid by an 
> obvious employer, a large chunk are.

Interesting definition of "majority."

The future of Linux: what it means for Wikipedia (apc)

Posted Jan 31, 2008 21:26 UTC (Thu) by tetromino (subscriber, #33846) [Link] (2 responses)

I suspect the word our editor was looking for is "plurality".

s/majority/minority/

Posted Jan 31, 2008 21:38 UTC (Thu) by stevenj (guest, #421) [Link] (1 responses)

Actually, it's not even a plurality. By the article's own numbers, at least 33% of kernel developers are paid to work on the kernel by their employers (from Red Hat to IBM to Google).

I think they must have meant to write "minority" instead of "majority," and it was just an editing screwup. "Minority" makes more sense in the context of the article too, because it is arguing that the kernel is not "the product of volunteers" any more.

s/majority/minority/

Posted Feb 1, 2008 8:13 UTC (Fri) by alexl (guest, #19068) [Link]

I assumed he meant that its the largest group, larger than from any one particular company.