Wikipedia and sausages

Sometime in the next couple of days, the one-millionth article will be added to the English-language version of Wikipedia. It’s an impressive achievement for a project that’s only five years old , and it’s already clear that Wikipedia has surpassed its main competitors, Encyclopedia Britannica and Microsoft’s Encarta in many important respects. Neither Britannica’s 200-year history and expert staff nor the Microsoft juggernaut have proved a match for Wikipedia’s ten thousand or so regular contributors, and thousands of occasional helpers. While many criticisms of Wikipedia have been made (as with most things, the most comprehensive source for such criticisms is Wikipedia, none has really dented either Wikipedia’s credibility or its growth.

Still, as Bismarck is supposed to have said

If you like laws and sausages, you should never watch either one being made.”

The process by which Wikipedia entries are produced is, in many cases, far from edifying: the marvel, as with democracies and markets, is that the outcomes are as good as they are.

I’ve been active on Wikipedia for several months now, and found out some interesting things.

* A visit to the recent changes page shows that there are between 50 and 100 changes each minute, or something like 100 000 changes per day. A fair number are cancelled out by subsequent reverts (either in response to vandalism or in the course of an edit war). A lot of others are very minor improvements, particularly in areas like cross-linking and categorisation. Still every little helps and the cumulative impact of all those changes in quite impressive. Although the quality of individual Wikipedia entries is highly variable, the structure as a whole is more powerful and robust than that of the competition.

* If you think the blogosphere is riddled with factions, flames and fury, wait until you get involved in editing a controversial page in Wikipedia (this includes anything to do with sex, politics or religion of course, but there have been bitter controversies about quite trivial issues like capitalisation/capitalization and spelling). At least with blogs, all your critics can do is flame you in the comments section or on their own blogs. In Wikipedia, they can wipe out your brilliant work altogether or edit it into a travesty of your original intention.

* Entries for living people (and for musical groups, companies, blogs and so on) are particularly problematic, and have become more so after the Siegenthaler affair. Given the problems that can be caused by errors or deliberate falsehoods in biographical entries, the guardians of Wikipedian orthodoxy have become increasingly pernickety both about citation of evidence for claims about the inclusion of such claims in the first place. It’s quite odd to watch a lengthy debate over whether someone is or isn’t notable enough for inclusion in Wikipedia, especially since, as with most things Wikipedian, anyone, however uninformed they may be themselves, is entitled to offer an opinion

* A lot of Wikipedians are particularly dismissive of blogs – ‘the ultimate vanity press’ is one of the kinder descriptions I’ve seen. Sean Bonner suggests that the anti-blog faction think blogs are just rumors about kittens and stuff. I blame Belle for this. More seriously, I suspect that there’s a clash between the collectivist ethos of Wikipedia, at least as regards the final product, and the more traditional authorial role of bloggers.

* Despite all the above Wikipedia is good and getting better. Not long after the Siegenthaler controversy, Nature did a comparison of Britannice and Wikipedia on a set of science topics, selected without looking at either. Wikipedia had marginally more errors, but covered some topics omitted by EB (I can’t find a source for this right now, but I remember reading it).

39 thoughts on “Wikipedia and sausages

  1. JQ – do you think wikipedia credible enough for citing in research?
    If so at what level – undergrad, postgrad etc. I havent personally used wikipedia as a source but have found varying levels of antipathy to it in this regard from the lecturers I have had the pleasure of listening too.

  2. I am interested in the idea that sources of information from Wiki might systematically distort. It is widely used but not really tested. It is accessed so often because it is so easy and low cost to access. I think the wiki people themselves have a name for the systematic errors that can result from access to very inexpensive global information. They acvknowledge biases can arise.

    By the way I remain a big fan despite this concern. I use it all the time. Indeed I often use it as a dictionary. It is an amazing institution.

  3. I have been a fan of wikipedia ever since I first logged in and made my first every edit to an article (“Australian Dollar” on 3 Jan 2004). My watch list is now extensive.

    For me discovering Wikipedia represented a major confirmation of my views on civil society and the effectiveness of self organising communities devoid of any major centralised control. This was something of a surprise because previously I had been somewhat dismissive of the open source movement as it applied to computer code. And whilst I was already labling myself as a libertarian I still harboured doubts about some of the details. Wikipedia was a personal revelation.

    My first major battle in Wikipedia world was over the article on “supply-side economics”. My frustration was that certain others seemed to allow no room the article to offer a theoretical discussion of the theory of supply-side economics in its own terms. Every attempt at explaining the theory was recast into Keynesian terms. A tax cuts was not allowed to be refered to as a reduction in the domestic trade wedge or trade barrier but was recharacterised as a means to stimulating demand. The article still has many such problems.

    Wikipedia has taught me that peer reviewed actually does not mean true and correct. To be sure the process of peer review is very important at filtering out obvious errors and tiding up unclear presentation, however it does not guarantee that the result is a competely true and correct reflection of reality.

    The Wikipedia system is also a great way to gather information that does not exist in composite form anywhere else. When John Howards IR reforms were being debated I could not find a good comprehensive source of informations. So I started a Wikipedia article and tuned in to watch it grow. It quickly gathered all the information I was looking for into a single location.

    The first link below is the original Industrial Relations article I created on 12 July 2005. The second link is the latest version:-

    A similar case of rapidly converging information followed the Cronulla riots as well as hurricane Katrina in the USA. The Wikipedia articles provided much more comprehensive and timely coverage of events and details than any media outlet. Next time their is a major event find the relevant Wikipedia article and tune in for the latest breaking news.

    In about 1993 I vividly remember laying on the lawn near the library at the University of NSW. Gazing up at the library building I remember speculating that one day the content of all those books would be on a computer and available for searching and cross referencing. I remember thinking that I might see such a thing in my life time. In hind sight I think I was very myopic. What I envisaged was no where near as grand as what is currently unfolding. And it was not as far away as I then imagined.

    Blogs still amaze me. Google is great. Google news is terrific. Wikipedia is brilliant. And the Internet revolution, and the social implications it will bring, are only just getting started.

  4. Mostly for me Wikipedia is a starting point. Quick & Dirty. Then google, then academic google, then subscribed journals. (whilst not an academic I do have some considerable access through work / civic duties) What is surprising is how often Wikipedia can be an end point. I’m a fan.

    I keep reminding myself I should contribute more.

    I was tickled and had a good chuckle on the Bettina Arndt entry. Naughty but not too far from truth.

  5. lurch, citing any encyclopedia is generally discouraged. You use the encyclopedia to get an overview, or to get pointers to the real sources.

    And it’s great that wikipedia has both John Quiggin and Terje Petersen as contributors. And me, of course, who else should have started the article on Reginald Dixon?

  6. There was a study published in “Nature” recently that showed that Wikipedia has only a slightly higher factual error content than Encyclopedia Brittanica. I thought that was pretty impressive. Nonetheless, for serious academic work Wikipedia should only ever be a starting point. All Wikipedia “facts” should be verified by referring to another reputable source.

  7. Bugger, I now see that PrQ made my main above point in his article. Sorry, I should have read it properly before commenting. ):

  8. Great comment Terje – thanks.

    I’m sold on the miracle of the internet and the miracle of markets. I guess there are ways in which that does, as it should condition my view of governments, but it doesn’t make me a ‘liberarian’ if that’s someone whose preoccupation is winding back the state, who thinks that things will necessarily organise themselves well without collective vision and effort.

    Collective vision and effort will always be difficult if we can’t build it up from individual self seeking efforts – but life is full of such things. So we just press on, greatful for the miracles that come our way, and modestly trying to improve our world.

  9. Wiki’s (wikipedia’s that are not part of The Wikipedia) are also becoming a pretty useful business tool. Working in a large IT organisation we are faced with the problem of a huge amount of information that needs to be documented accurately but which is also changing all the time. Normally you rely on business analysts to collect all that information and ensure that it is up to date, but inevitably they fall behind. This is a big problem, as the most recent changes are usually the ones that people need to know about the most.

    Wiki’s are a god-send for dealing with this.

    You can actually see the wiki allowing the company to operate at a faster speed. Especially when some moron in upper management decides to turn your wiki off because “no-one’s in charge of it” and you have to go back to the old way of operating.

  10. We run a wiki in our company. It uses the same open source code as wikipedia.

  11. Nicholas Carr has a good piece pointing to weaknesses in the Nature article and in the way it was reported.

    First, it wasn’t one of the peer-reviewed, expert-written research articles for which Nature is renowned, but just a semi-formal survey by staff reporters.

    Second, Carr examined supplementary information not included in the Nature article and concluded that the reporters had filtered out some of the criticisms of Wikipedia offered by the experts, and that the inaccuracies in Wikipedia tended to be more substantial than those in Britannica.

    Third, he points out that the Nature survey concentrated on scientific and technical articles, which are a field where Wikipedia does well. That’s because such articles are generally written by specialists and generally not subject to controversy.

    “Read together, the article and the supplementary information indicate that the survey probably exaggerated Wikipedia’s overall quality considerably,” he concludes.

    Read it at Roughtype

  12. We also run an internal wiki at my company. Use it to document everything: from network setup procedures to the current state of the sales pipeline. I’ve used plenty of other dedicated tools for these tasks but nothing beats the global, web-accessible and web-editable free-for-all that wiki gives you.

    As well as making information sharing very efficient, it also helps solve the problem of people leaving the company and taking critical knowledge with them.

  13. What’s the etiquette regarding editing a wikipidea page about one’s self:-)

  14. What a great concept – and brilliantly executed. The power of democracy at work! To paraphrase Churchill – “Wikipedia is the worst type of encyclopedia in the world….. except for all the others”. Of course it’s got warts, of course there are biases, of course there will be fierce arguments on controversial topics – but the power of mass input has always got to be better than one single ‘knowitall’ source.
    The political analogy is Western Liberal Democracy – it’s always going to deliver a better outcome than any form of authoritarian rule. It’s why the US ultimately won the cold war, and why ultimately any form of religion based government (and I’m not singling out Islam here) is doomed to failure.

  15. I too like Wikipedia and refer to it often but always take anything political with a grain of salt. The John Howard entry is a case in point where ‘Further reading’ lists David Barnett and Pru Goward, Tony Kevin, Margo Kingston, Marion Maddox, David Marr & Marian Wilkinson and Andrew Wilkie.
    The tone of the entry is negative and in discussion one contributer claims that; It has never occurred to anyone to have a “Praise of Howard” section. This is of course because almost all Wikpedians (including me) are anti-Howard.

    Not encyclopaedic at all.

  16. “What’s the etiquette regarding editing a wikipidea page about one’s self:-)”

    I’ve fixed some broken links on mine, but I think substantive edits are a breach of etiquette.

    Kev, as regards “Further reading”, isn’t this a reflection of what is there to be read. AFAIK, the Barnett-Goward hagiography is the only major pro-Howard book in the field, though a new one is about to be released.

    Apart from this, I can’t see that the main body of the article is anti-Howard. For example, Labor’s 1998 GST campaign is referred to as a “scare campaign”, accurate enough but scarcely consistent with an anti-Howard POV.

  17. Tony, I read the Carr piece and it struck me as special pleading, especially the claim that you’d expect Wikipedia to excel on scientific and technical stuff.

    A standard criticism of Wikipedia used to be that its focus was too much on popular culture and Internet-related topics (these are technical in a sense, but AFAIK not covered in the Nature study.)

    Carr is beginning to sound a bit like the PFJ in Life of Brian. “Apart from popular culture, the Internet, science, technology and current events, what has Wikipedia ever been good for ? Nothing!”

  18. “The tone of the [John Howard] entry is negative and in discussion one contributer claims that; It has never occurred to anyone to have a “Praise of Howardâ€? section.”

    Maybe this problem could be overcome by having one Wiki entry for each of Howard’s faces.

  19. The Dutch mafia prevent any references to Holland as the English name for their country, always substituting “the Netherlands” on the spurious ground that (to them) Holland is only a province. But this prevents any historical perspective that includes what is now Belgium under that term, as well as encouraging Dutch irredentism (I’ve heard a Dutchman say “there is no such place as Belgium”, echoing a similar Indian view about Pakistan).

    What gets up my nose is that while it is none of my business what they call themselves (apart from the irridentism), it is also none of their damned business what my language calls them. (The same applies to this recent nonsense about winter Olympics in “Torino”, when we should no more call it that than we should call Rome “Roma” – it’s Turin, damn it.)

  20. PML, when in Torino, do as the Torinons do….

    Seriously, I have more time for the Torinons’ cultural fascism – after all they’re paying for the damn thing. What really gets my goat is that every announcement has to be made in French as well as Italian and English. Someone needs to point out to the French that half the world speaks English as either a first or second language – the same is not true of French.

  21. Like I said, what they do in their own country is their own business. What I object to is the stuff that turns up here, e.g. in google’s recent logo – that’s the cultural proselytising part.

  22. Lurch, Francis and Harald have it about right.

    One place where Wikipedia *can* be usefully cited is as a readable, short summary on a peripheral topic in an area the intended readers are not experts in.

  23. From the headline I thought we were having a barbeque in here. From the reference to “toronions” I was hopeful things were heading that way… alas….

  24. The Bettina article is interesting. The early version was arch, fun and opinionated, but it has now settled down to being dull and dishonest, because it does not point to her ideological stance.

    Generally, for me, one of the values of Wikipedia is that it can at least point to the contentious issues.

  25. I think that Wikipedia ranks with Google as the possibly the most effective resources we have yet developed for sharing knowledge and information (not quite the same thing…yet).

    Very few of us imagined either 10 years ago – I wonder what we’ll be using in 2016?

  26. John, I don’t see how you can interpret Carr’s piece as special pleading. His point is that the widely touted Nature article is not as authoritative as spruikers claim. Carr’s point about science and technical coverage is made in explaining the weaknesses of the Nature findings.

    Introducing popular claims by others is not relevant, and I doubt a skilled writer such as Carr would ever have praised Wikipedia cultural and internet content. In the sense you refer to internet articles, which is in the context of cultural discussion, they are not scientific and technical articles at all. Carr is referring to articles in scientific and technical disciplines by specialists in the relevant fields. Very few so-called internet articles fit into that category, with the exception of computer science and networking articles, which comprise only a small part of that category.

  27. Tony, I think you’ve misread my response. My observation is that it’s widely recognised that WIkipedia is strong on Internet-related topics and popular culture (though, to be fair, Carr criticises the quality of some entries in the latter field). It’s also an easily checkable fact that Wikipedia covers more topics than do its competitors.

    If you start, as Carr does, by conceding that Wikipedia is also likely to be relatively strong on scientific and technical topics, it becomes very hard to sustain the claim that the shift from EB to Wikipedia is for the worse.

    No doubt it’s useful to deflate claims that Wikipedia is comprehensively superior to the alternatives, but Carr seems to me to want to claim more than this.

  28. PML sensitivity to what we call Holland/Netherlands in English.

    At an international conference I once referred to (and supported) a submission from “our colleagues from Holland”.

    “Why do you not include me?” said the leader of the “Dutch” delegation, in English more aligned than mine with that spoken in England. “I am from Freisia!”

    So yes you can reserve the right to use a term of your choice for a group, but do not be amazed if that leads to miscommunication or worse.

  29. DC, the miscommunication – when speaking English – comes from the damned Dutch failing to learn and teach English usage descriptively but instead assert their own views prescriptively. After all, “Holland” is anyway a generic term in English, albeit archaic, used in places like Lincolnshire.

    If they want to create confusion, I am perfectly happy for them to fall on their faces. But I am not happy for them to create confusion in our own understanding of our own historical background knowledge. Do the Swiss object to being generally known for just one of their early cantons? I am no more willing to confound the issue of Holland/Netherlands than Switzerland/Helvetia.

  30. We in “The Netherlands” use the same pars pro toto for the UK: in coloquial Dutch we use the term “Engeland”. That doesn’t mean it is correct for an encyclopedia entry.

    Frustrating isnt it, when your wiki entries are censored?

  31. Another strange use of English language I encountered when living in Britain was that british people announced they were “going on holiday in Europe” when they actually were going to visit the continent.

    The english wiki is not british.

  32. Dogz,

    I seem to recall that the reason for French being an official language at the Olympic games is that a Frenchman revived these games around 1900.

  33. HE, the term “Holland” is correct for an encyclopaedia entry for the whole country – when the work in question is in the English language. “Holland” is still in common use (see the recent article in the Times on baby euthanasia in Holland [sic – it’s in the headline]).

    That means that this kind of censorship is imposing a prescriptive rather than a descriptive approach to the English language. It’s ringbarking, and quite the wrong approach for English (unlike, say, French).

    Note that none of this is a British-centric comment. It’s a comment centred on anchoring the present, past and future – retaining the value of the language as a transmitter without haveing a risk of it being reworked for an agenda no matter how well intentioned that may be in any individual case. See also Churchill’s comments on Siam/Thailand.

  34. English language wiki is international and not british. For your local use of the word Holland the disambiguation page was invented.

    Holland is properly a region within the Netherlands, now divided into two provinces: North Holland and South Holland
    In English, Holland is often colloquially used as synonym for the whole of The Netherlands
    The Kingdom of Holland was a puppet state covering the area of The Netherlands during the Napoleonic area 1806-1810

  35. Wikipedia, with a 97% share of the online encyclopedia market, has forced Microsoft to shut down Encarta. How long will it be before Wikipedia claims the prize scalp of Encyclopaedia Britannica?

    Encyclopaedia Britannica did not think that an open source product like Wikipedia would significantly challenge the credibility of its brand. They were dead wrong and Encyclopaedia Britannica’s staff seriously misread the global market. They are now very concerned about the widespread use of a free Wikipedia vs their paid subscription model. From a corporate and financial perspective, Encyclopaedia Britannica is in significant trouble.

    It will be interesting to see if Encyclopaedia Britannica survives, but recent indications do not look good. It is the combination of a) the success of Wikipedia and b) improved search engines that has put financial pressure on Encyclopedia Britannica over recent years. Many libraries, schools & individuals are questioning the need to pay for sets of expensive books, or to subscribe to Encyclopaedia Britannica Online, when the content is free on the internet, and much more comprehensive.

    Over the next year or so we will see the continued demise of Britannica as it becomes ever less relevant in a Wikipedia-dominated landscape.

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s