I’ve been pointing for years to the fundamental rhetorical problem of wikis — that making the link target and the link label the same, as Ward’s Wiki did, moves all links to nouns and noun phrases with disastrous impact on the link structure. Wikipedia no longer uses WikiLinks, unfortunately, but almost every link remains anchored to a noun and almost no editors use links intelligently or creatively.
I don't see a problem with most links being nouns and noun phrases. Wikipedia is a reference, not a literary masterwork, and so consistency and the principle of least surprise should be followed. There's also no guideline that I know of that forbids the use of links creatively or intelligently for non-nouns, as long as they wouldn't negatively impact readability.
I also don't believe Wikipedia/Mediawiki has required link labels to match link targets in something like ten years. It's often desirable or necessary to do that purely for semantic reasons.
Using nouns rather than verbs in links does not specifically relate to wikis and WikiWords. See this more general advice from the W3C in 2001: http://www.w3.org/QA/Tips/noClickHere.html.en
> we do not recommend putting verb phrases in link text
More generally, links should identify what they link to, and more often than not that will mean a noun phrase, not a verb phrase.
Categories suck in general because they're basically tags. That is, they're not structured. I find the category system for MediaWiki in general pretty useless specifically because it is impossible to create a taxonomy and apply it in a structured or hierarchical way across your document set.
Wikipedia's core admins are pretty notorious for lording over the content they consider their domain in pretty unforgiving ways. Luckily, Wikipedia is an implementation of a specific type of platform, of which many free hosting options are readily available for your own wiki.
I think it's pretty naive to think that editorialization hasn't happened in some flavor throughout the history of the encyclopedia, as far back as Diderot. Those biases just happened to be engrained in different ways: platonism, the academy system, ethnic or national bigotry, etc.
As human knowledge expands, so comes the need for all willing people to contribute to our collective intelligence. There will always be stories of people gaming the system for their own end. MediaWiki as a platform has a revision system that allows us to review specific changes made by authors and revert them if they are in some way problematic or unacceptable.
So without getting around to the inscrutability of facts in general, I think it's preposterous to claim that Wikipedia faces an insurmountable problem due to its issues with taxonomy (a non-primary concern) or sexism (a concern relating to specific members of the community, and not of the knowledge being produced). So that being said, what's the point of this post?
> I think it's pretty naive to think that editorialization hasn't happened in some flavor throughout the history of the encyclopedia, as far back as Diderot.
Everyone has a bias. That much is taken for granted, right?
Now, the questions:
Is everyone equally biased?
Can anyone overcome bias?
Are all biases equally severe?
Are all biases bad?
Is it futile to attempt to overcome bias?
How you answer these questions will determine whether you think encyclopedias are even a valid enterprise, let alone your ideas on how to construct one. The point remains, however, that none of these questions are new, and bringing all of them up at once, implicitly or explicitly, every time the topic is discussed does nothing to move anything forwards.
It seems fundamental and exactly the kind of thing we need to get hammered out before anything else can begin, but it's really a pointless waste of time, like arguing about whether the physical world really exists every time you want to decide what to eat.
I think it is pretty clear that not everyone is equally biased or that every bias is equally bad or good. It does not seem like a useful question unless you plan on engaging in a lengthy exercise it justifying Wikipedia from first-principles or something.
I would ask this question instead: Is the editorial environment of Wikipedia specifically a place where the worst biases can flourish? Subtle edits over a long period of time from a malicious actor are one thing, but the same from someone who is simply oblivious to their bias on a set of topics is quite another. If the answer is yes, how in the heck can it be fixed?
The answer is yes, Wikipedia is specifically a place where the worst biases can flourish. It only takes a couple of cranks to keep the biases in, and there are more than enough cranks around.
I don't know that it can be fixed.
The good news, though, is that there's enough transparency to in most cases make it clear to the careful observer what's going on. On any politically-sensitive topic, the experienced wikipedia user eventually learns that you need to check the talk page and see what points of view are being deliberately and actively excluded from the main page. Checking the page history helps too - bad pages tend to have a lot of reverts with angry check-in comments.
> Everyone has a bias. That much is taken for granted, right?
Do I smell a challenge? :)
Nobody has perfect knowledge, and many people are motivated in bad ways. But "everyone has a bias" is, I think, a subtle attack on the idea that anyone can actually know anything. Most philosophy of science courses would endorse some version of that view, but I don't.
I know that 2+2 is 4, and that's not biased. I know a lot of other things, too, that aren't biased.
There are people who will claim that, say, 2+2=4 is just a way for Western imperialists to get their way over others, or even that math is a figment of the imagination without correspondence to the real world (many math professors would defend this), and those people are wrong.
Hell, some people would claim that assuming there is a "real world" is biased, but they are wrong, too. I'm not going to prove it here in the comments, though.
Wikipedia has a failure which unearths some deeply embedded problems and then proceeds to fix them. This has happened how many times now?
Until I can get the same level of information (quantity wise) from another source, Wikipedia has immense value that fills a very large market gap -- even if they can't get their taxonomy of American authors quite right.
To say "Wikipedia’s about over" ignores the actual value of the site and greatly over-reacts to another mishap in the history of the user-run site.
We don’t edit Wikipedia anymore. We don’t consult it for things that matter. It’s merely a good resource for finding odd facts no one cares much about.
Snort. Explain this then: Wikipedia's getting more views than ever, more edits than ever, and more editors than ever.
There are plenty of things broken in Wikipedia alright, but none of them fatal, least of all some misguided categorization (oh, the humanity!) by some two-bit hack with an axe to grind. Everybody on Wikipedia has their own point of view, and while some people manage to game the system for a while, Wikipedia's core strength is precisely that it's self-correcting in the long run.
• 'active editors' are stagnant to slightly declining globally for the last 3 years – and definitely down in EN and DE (largest) Wikipedias
• 'pageviews' growth has been slowing for a while, and since the start of 2013, total pageviews are down for 'all', EN, and DE
• 'edits per month' (on the secondary page) seems stagnant-to-very-slightly-improving globally, but definitely on the downtrend in EN
Sure, the the original article is an opinionated, possibly premature diagnosis of crisis... but your confident assertion of 'more activity than ever' isn't supported by those graphs!
You'll note that the primary graph on the first page is new editors per month, which is fairly stable at ~20k/mo total and ~7k/mo English. As long as that number doesn't start plummeting, there's fresh blood coming in and Wikipedia's future is assured.
Also, I think edits per month on EN and DE are slowly down simply because they're already so large and mature: these days most articles on most major topics that are, from a functional point of view, complete. This wasn't the case a few years ago, when even a layman could find things to fix everywhere and Wikipedia's article and edit counts were rocketing.
As for page views, my theory is that Google's Knowledge Graph has reduced the need for actually clicking through to WP, since most Google searches for basic Wikipedia-type info now come up with capsule summaries (often extracted from WP) that may well answer your question.
I agree that Google's Knowledge Graph is likely the largest reason for fewer visits.
New editors don't assure Wikipedia's future unless they stay around; as of the last study I can find, 2010, retention had fallen to quite low numbers...
...and each year's cohort was sticking around less than the year before. I recognize much of this may be that, in the more-established Wikipedias, the core audience of potential editors has been found, and the fun/easy 'low-hanging fruit' content has been finished (or made less fun/approachable by policy).
But that means Wikipedia is in new territory, with fewer active and experienced editors over time, doing less-exciting maintenance and defense work, while search engines retain more 'reference' views and other competitive sites (especially in the Q&A space) improve their coverage and depth. Wikipedia needs positive attention to keep the forces of decay at bay; if the attentional indicators have shifted in the wrong direction, even very slightly for now, it risks nearing a tipping point into collapse.
And counting how many edits are being made is akin to saying that you corrected 1000 loc because you found a spelling mistake in each. There is not an obvious connection to "contributing content" "meaningful content" or even "long form content".
The site really allows you to draw hardly any conclusions regarding this article's statements.
Explain this then: Wikipedia's getting more views than ever, more edits than ever, and more editors than ever.
That's like having the fastest-growing website on the internet - I got three views yesterday and 10 today, so I'm experiencing 330% growth!!! Statistics are meaningless without being scaled to something, in this case internet populations. I'm not about to write off wwikipedia, but your own graphs show the #s of editors and edits to the English language version stagnating over the last 2 years.
Wikipedia was never intended to be a formalization of semantic web (wikidata.org is more along those lines, if that's what you're looking for). So what's the emergency?
The decline in number/quality of editors is alarming, but what does it have to do with taxonomy? And how do the links fit into the picture?
TL;DR A random set of barely coherent assertions followed by OMG, Wikipedia is over
Wikipedia, for better or worse, serves an indispensable role for millions of people. So even if you disagree with its leadership and its editors with biased agendas and would like to see its demise, a more realistic solution seems, to me, to be to put on your writing cap and become an editor yourself. In this case, it is actually very possible to live according to the saying, "Become the change you wish to see in the world." A bunch of new contributors focused on fixing things might help.
Although I do remember reading stories at one point about new contributors' edits getting frequently rolled back by Wikipedians with more permissions. That could be a barrier to entry, but it's unclear to me how much of a problem this really is.
The second you try and do anything non-trivial without first "paying your dues" (and kissing the right asses) you find all your work trashed. If you complain or try to keep it in place you'll see a dozen different bureaucratic maneuvers involving obscure jargon and weird, diificult to use, administrative pages and processes to thwart you.
It's amazing the wikipedia is as useful as it is given how utterly hostile it is to new contributors. I hope it continues to be useful and thrives, but I'm not so masochistic to actually fight to volunteer.
The biggest problem I've seen over the years is that Wikipedia is not edited by experts, and in general is edited by losers.
These are the sorts of losers that watch the RSS feeds of the little kingdom of bullshit they've carved out in their own corner of Wikipedia and relentlessly fight anyone trying to edit THEIR articles. Until Wikipedia gives some sort of weight to experts (paid perhaps?) instead of unemployed shut ins, the quality will continue to deteriorate.
Taxonomy is hard, especially if you don't want it to be a collection of poorly-aging cultural biases. Even professional librarians have problems creating and maintaining future-proof ones.
Why does the Dewey Decimal System still have 10x as much space for Unitarianism as it does for all of Buddhism? [1] Well, think about what it would take to fix Dewey's initial biases post-hoc... either you have to get it exactly right the first time, or have a system that is designed to evolve.
It's unsurprising the wikipedia hive mind hasn't gotten it right. It's orthogonal to everything Wikipedia has generally done well. Seems like the best solution would be to have a separate taxonomy project that is overlaid on top of core wikipedia data by a set of flexible rules and not hardcoded in articles themselves.
So here we have a article complaining about an category problem that do no longer exist, and a user that got banned because he failed to follow the rules of the site.
Its a community site where decision is made through consensus, and consensus was to fix the problems. So where is the emergency, and why is this article trying to declare the site as doomed?
The Wikipedia ideals of having open, unbiased, and free access to information and being able to freely collaborate and edit articles directly conflict. If editing is "free", so too is slander and misinformation.
Unfortunately, I have no good solution to this problem. But then again, neither does Wikipedia.
The idea of an unbiased article is in my opinion just an illusion. In reality truth is a negotiation between different stakeholders and some form of settlement at the end where each side can somehow get away with.
I personally find the next posting of the author much more revealing [1]. It is a discussion on how much meaning one can find in a simple sentence such as "____ is a Jewish-American politician" all the way to the suggestion that some by "tagging all the Jews in Wikipedia dream of a time when those lists might be of service to the police."
Just goes to illustrate how much subjectivity can be infused into a few words that are in itself unquestionable facts.
You want axes? You want them big? You want to swing your axe into the guts of lawyers? Do the lawyers have HUGE GUTS you want to RIP AND TEAR? Then this axe is for you.
Deletionism is the only reason Wikipedia remains using. Without deletions, it would have sunk into the world of spam-trap and personal blog platform long ago.
> I still get nuclear physics in "Nuclear Physics" article, amn't I?
Not at the same level of quality, because they also won't be deleting the nonsense crank pseudo-science version of "Nuclear Physics" which would confuse you terribly. Deleting crap is one of the best ways to make the project as a whole useful.
In regards to the Wikipedia list of American Novelists, could it not be that sexism for the last many hundreds of years have rendered a collection of what many consider to be the "most prominent authors" being primarily male. I am not saying that it's right, but it would seem that given our propensity as a culture through the years to over value the opinion of a man, a list of mainly male authors is all but inevitable.
I think the point was that all the women were being moved to the "American Women Novelist" sub-list while all the men remained in the "American Novelist" top list. If they had moved all the men into one just for men also and then made the top level list just link to the two sub lists that would have made more sense. This would also open up a format for additional sub lists linked from the top list.
It's not a list of the "most prominent authors"; it's a list of all American Novelists. In addition, it was not being turned into a list of "mainly male authors"; it was being turned into a list of only male authors.
You are arguing in hyperbole. If you were to include every american author to have ever lived, the list would be untenable. There has to be a logical cut off, and often in the world of the arts, that cut off is directly correlated to prominence as an artist.
"History did it" is never a good reason to support marginalization. Especially when there are other, more neutral ways to split up "American Novelists".
I don't think sdoowpilihp said anything about supporting marginalization. It was just simply pointed out that the chosen hierarchy of the Wikipedia page could be seen as a cultural artifact of centuries of sexism.
in any event it's not a very sensible way of dividing that group. afaict there are three genera that you could use as the intension of the individuals in the sublists. you have (american)(people)who wrote(novels). taking an attribute of the (people) as the differentia makes as much sense as dividing the list by american state. dividing them by the kinds of novels they wrote would make more sense, because the writing novels thing is certainly more what the list is about than the writer's genitals.
Wikipedia doesn't have an emergency. Jimbo Wales had large ideological axes from the beginning, including topics such as who started Wikipedia. As Wales once said that conservative "[Friedrich von] Hayek's work on price theory is central to my own thinking about how to manage the Wikipedia project". Wales went from running an Ayn Rand mailing list to running Wikipedia...but he's totally neutral of course.
As an example of this bias, Wales appointed the rabid editor JayJG to the highest authority, the ArbCom in mid-2005. He was voted out in early 2006 - but Wales used his power to reappoint him any how. Most of Wales's political position pushing has been behind the scenes, but here it was blatantly obvious.
I got tired of this sort of thing and pulled back from Wikipedia around that time. I'm very skeptical of the idea that everything has changed since Wales has less power now though. It's kind of like when the military takes over in a coup, kills off or exiles the opposition, and then holds an election. Once you have sculpted the infrastructure enough to your advantage, you don't need to be as heavy-handed any more. I guess I would be one of the exiles (and some excellent editors threw in the towel over this sort of thing - very educated people who could write well).
I don't see a problem with most links being nouns and noun phrases. Wikipedia is a reference, not a literary masterwork, and so consistency and the principle of least surprise should be followed. There's also no guideline that I know of that forbids the use of links creatively or intelligently for non-nouns, as long as they wouldn't negatively impact readability.
I also don't believe Wikipedia/Mediawiki has required link labels to match link targets in something like ten years. It's often desirable or necessary to do that purely for semantic reasons.