Slashdot Mirror

Slashdot needs to enter the 21st century and support wiki links. Here is the HTML version of the links I gave:

* Wikimorial
*GlobalFamilyTree
*Wikipeople

And the Sep11Wiki

*Sep11Wiki

-- mav

Slashdot needs to enter the 21st century and support wiki links. Here is the HTML version of the links I gave:

* Wikimorial
*GlobalFamilyTree
*Wikipeople

And the Sep11Wiki

*Sep11Wiki

-- mav

Check out http://download.wikimedia.org/.

The current version of en.wikipedia.org (English edition) is 301 MB, and the dump including the edit histories (ie: old versions) of the articles is 9079MB. The current version download of all languages is 686 MB.

(As I understand it, all articles are included in that 301 MB download. It is gzip compressed, however.)

As for images and uploaded files, for English, they're available as a split tar file - 1.9 and 1.7 GB for a total of a 3.6 gb download.

If I was marking some homework that relied on referencing Wikipedia, I'd have to fail them.

I can only hope that, in addition to being a coward, you are also a troll. Why? Making fun of Britannica. There are many examples which show EB being incorrect or misleading, even common items like Leap Years, the description of a coulomb, real numbers, the creator of the safety razor, and others. The list is not amazingly long but it is probably not complete, either.

Clearly you cannot blindly trust what you find in the EB. Don't blind yourself.

It's 17+GB so better wait for BluRay....

However, you can download your own copy and burn it today!

Have a look at the servers Wikipedia is using! That's some hardware Slashdot will surely like - master db is a dual Opteron with Fedora Core 2 64-bit and 1+0 RAID. The cash seems well spent.

Well, considering no less than six (6) posts before yours include the same quote verbatim, with sources ranging from woodieguthrie.com to the linked article to some commercial site that actually uses the wiklipedia entry and only cites in tiny font below the ads at the end up the page, I think we can "trust Wikipedia."

Seriously, have you any examples of Wikipedia being wrong? Wikipedia does have examples of Encyclopedia Britanica being wrong. I wonder if you'd have included such a disclaimer when citing EB? How about now, after reading this? ;)

Realise that the English Wikipedia is not Wikipedia. It is one of many Wikipedias. There are about 50 now in very different stages of development. http://meta.wikimedia.org/wiki/The_provisional_por tal_of_Wikipedia The are all encyclopedias who are growing whit there own user community and local customs. The are al using the same format and ground rules but are in practice all independent projects. The are definitely not plain translations. Many users who know other langauges look for inspiration and information to other Wikipedias (and not only to EN). And some articles are in some languages direct translations of other versions. But that is a temporary situation because of the editing process the grow appart again.
[[wikipedia:nl:gebruiker:walter]]

Actually, it's not necessarily (well, the language use bit, anyway). In English, 'he'/'his' can be used for a gender-non-specific singular pronoun for animate things (like people), unless it is known that the person is female. Inanimate things are referred to as 'she'/'her', which is why boats are usually called 'her'. It has nothing to do with male and female, it's just a feature of the language.

Note: There is mention of it on wikipedia.

The images, however, are about 3.6 gigabytes, and their copyright status is less than sound. Fair use and all that.

The bible commentary idea is part of Wikisource rather than Wikibooks. See Wikisource:Religious texts where people have started uploading the bible and other religious texts.

Angela.

The bible commentary idea is part of Wikisource rather than Wikibooks. See Wikisource:Religious texts where people have started uploading the bible and other religious texts.

Angela.

The bible commentary idea is part of Wikisource rather than Wikibooks. See Wikisource:Religious texts where people have started uploading the bible and other religious texts.

Angela.

http://mail.wikimedia.org/pipermail/wikipedia-l/20 04-June/015818.html

== See also ==
* [http://meta.wikimedia.org/wiki/Making_fun_of_Brit annica|Making fun of Britannica]

</wikify>

http://download.wikimedia.org/

Also, formatted nicely:

http://en.wikipedia.org/wiki/Wikipedia:TomeRaider_ database

A bit OT.. but please folks, the Wikimedia foundation could really use some big bucks from the government instead of relying on the few PayPal donations.

There's a possible NEH grant the Wikipedia community is working on filling out right now, that is due July 15, that could give a grant up to $500,000. If you have some time, please help fill it out here . The grant is aimed towards digital/online reference works, and Wikipedia fits the bill perfectly. It'd be a shame to pass this opportunity up.

Posted Anon for your non-whoring pleasure

It comes down to the Wiki is not paper policy. Wikipedia has the capability to have information on everything somebody might want to know, so why not?

I am almost sure a big organisation will eventually give found like UNESCO or UN

Some wikipedians are currently writing an application for a grant of $500,000 from The National Endowment for the Humanities.

It needs to be done by tuesday (tomorrow!), and they seem to be far from finished...

When you look at the RAM usage statistics of their servers, for instance for brown, you find a clear sawtooth pattern, showing a linear increase in memory usage until the server (or a service) is restarted.

At 625 MB, you could fit the text of the current database on a CD. The images will jack it up another 3.6 GB. So you could reasonably fit the current revision on one DVD. If you also want the full record of changes and revisions, it's about 15 GB just for the text.

You can download this stuff easily, and it's obvious from recent Google searches that many people do.

Your questions answered

Longer term we're working on how to scale the databases (which of the many options to use). We're using three at the moment, one primary writes, one for slow queries and one for backup, the latter two both being replicating children. For data see:

1. Squid statistics showing total and cache hits. You can see the rise when the listing here appeared at about 09:30 UTC.
2. Ganglia cluster stats showing load. The ones which are mostly blue are the web servers, the red/blue mixed are the squid caches (about 60% max is right for max load for them or connect time suffers) and Suda and Ariel are the most heavily loaded database servers. Suda is disk-limited. Ariel is memory/CPU bandwidth limited because it has faster disks, more cache and different workload.

For what we did with the previous donations from the start of the year see:

• Servers today. (The picture doesn't show them all)
• What we bought first in Feb 2004.
• Analysis in April and purchase in May.
• Work in progress plan for next purchase and subsequent budget. This will exhaust current funds on hand and we could use more for performance and reliability/ availability if it was available. The budget side includes plans beyond currently available funds. Web servers are currently the hot item we're after - those max out at peak times.

Our growth is pretty simple: when we're fast we grow to use all the capacity until we're slow again. Still no sign of us hitting the limit on demand, so it appears that we'd have no problem at all serving more people if we had another $50,000-100,000 to spend - there are ballpark growth estimates suggesting that we'd end up doing that by the end of the year if we could stay fast until then.

If anyone wants to donate, as one of the hardware people, I'd rather see monthly recurring payments of a smaller amount than a lump sum. It makes it easier for me to try to predict what we can buy based on some moderate predictability of available funds.

One common question: can we use commodity PCs as web servers? We'd like to but fitting them in the colo isn't currently practical. We're going for dual CPU 1U boxes as the next most cost-effective option for subsequent web server purchases. The Jan purchase was in part about getting enough boxes so we'd be able to switch them around to cover for failures, so those were cheaper per box 1U boxes. We've enough of those now, so it's CPU power/density time.

If anyone has any suggestions please feel free to drop comments on the talk page - we've a dozen or so people on the technical team and more input is always welcome, since we're after the most effective options we can find! Jamesday (author of much of the April planning document, one of the technical team members)

Longer term we're working on how to scale the databases (which of the many options to use). We're using three at the moment, one primary writes, one for slow queries and one for backup, the latter two both being replicating children. For data see:

1. Squid statistics showing total and cache hits. You can see the rise when the listing here appeared at about 09:30 UTC.
2. Ganglia cluster stats showing load. The ones which are mostly blue are the web servers, the red/blue mixed are the squid caches (about 60% max is right for max load for them or connect time suffers) and Suda and Ariel are the most heavily loaded database servers. Suda is disk-limited. Ariel is memory/CPU bandwidth limited because it has faster disks, more cache and different workload.

For what we did with the previous donations from the start of the year see:

• Servers today. (The picture doesn't show them all)
• What we bought first in Feb 2004.
• Analysis in April and purchase in May.
• Work in progress plan for next purchase and subsequent budget. This will exhaust current funds on hand and we could use more for performance and reliability/ availability if it was available. The budget side includes plans beyond currently available funds. Web servers are currently the hot item we're after - those max out at peak times.

Our growth is pretty simple: when we're fast we grow to use all the capacity until we're slow again. Still no sign of us hitting the limit on demand, so it appears that we'd have no problem at all serving more people if we had another $50,000-100,000 to spend - there are ballpark growth estimates suggesting that we'd end up doing that by the end of the year if we could stay fast until then.

If anyone wants to donate, as one of the hardware people, I'd rather see monthly recurring payments of a smaller amount than a lump sum. It makes it easier for me to try to predict what we can buy based on some moderate predictability of available funds.

One common question: can we use commodity PCs as web servers? We'd like to but fitting them in the colo isn't currently practical. We're going for dual CPU 1U boxes as the next most cost-effective option for subsequent web server purchases. The Jan purchase was in part about getting enough boxes so we'd be able to switch them around to cover for failures, so those were cheaper per box 1U boxes. We've enough of those now, so it's CPU power/density time.

If anyone has any suggestions please feel free to drop comments on the talk page - we've a dozen or so people on the technical team and more input is always welcome, since we're after the most effective options we can find! Jamesday (author of much of the April planning document, one of the technical team members)

Combined live stats, all wikimedia servers.

Wikipedia needs donations to stay alive.

You can download the entire Wikipedia database. It weighs in at 15GB if you include all revisions, or 600MB with just the newest copy of everything. Have fun.

A dump of the Wikipedia database is available. It's certainly big, and the content is interesting, although the structure isn't.

That's a good idea. Better might be the Wikipedia. there's probably one in whatever language they speak (well, I guess that would be English in Uganda.) and is certainly one in English.

You can download the Wikipedia database, export out static pages, and shove those on the computer. I'm pretty sure there's software to do this.

Of course, you'll end up with a lot of Star Trek and Middle-earth articles, as well as an article for every tiny town in the US, at least if you use the English database. You could sort out useful subsets using things like their new categories system.

Maybe I should ask a MediaWiki-knowledgeable person to post. I'm mostly just a user.

For some discussions have a read at the wikipedia-l mailing list.

And even better, use free (as in freedom) text books from wikibooks on the laptops.

Wikibooks has free (beer / freedom) books and textbooks that anyone can edit, by the makers of Wikipedia. A whole list of projects are found here at Wikimedia (yes they like the word Wiki alot ;-) ).

Nah, nah, got you beat! I sent them EUR150! ;-)

That's no fair! It's not my fault that I live in a country where our leaders can't do math and are consequently flushing our currency down the toilet! ;-)

the counter for funds received over at Wikimedia is now at $23,382.17

Holy moly! That was fast! Hooray for the Internet!

(For those wanting to check the latest numbers, you can find them on the letter. And there's a nice thank you note that also explains what they'll do with the extra dough.)

It's an encyclopedia -- a reference to look up background information on topics of your choice, so don't expect more than that.

Actually that link was to the Wikibooks project which is a project separate to the Wikipedia (but part of the same overall group of projects). The idea with Wikibooks is to create "modules" little sections that can be gathered up and turned into textbooks (of course with copyediting, indexing, etc).

There are also other Wikimedia projects underway.