There’s much ado about Knol, but I haven’t noticed many Wikipedian voices. Well, I’ll speak up, and I’m sort of a Wikipedian. Actually, less so recently, which is relevant.

The Target Audience · I know exactly who Knol will appeal to. People like me. Frustrated editors. People who’ve enjoyed participating but find that they’re doing it less.

When first got interested in Wikipedia, I used to spend the occasional evening just following the "random article" link and cleaning things up; and also doing spot quality improvement when I looked something up and saw flaws or breakage. I still do a bit of that.

Also, I took responsibility for maintenance of a few articles where I had expertise: Vancouver, T.E. Lawrence, Audiophile. And, before I found out that you’re not supposed to edit your employer, Sun Microsystems.

I don’t really do that any more, for two reasons. The first is that, as with being a serious gamer, being a Wikipedian is increasingly time-consuming. There is a forest of acronyms, and there are all these macros you can slap on pages to indicate problems, and they have subtle cultural implications that I don’t really understand. For example, as I write this, T.E. Lawrence has a badge at the top saying there’s a neutrality dispute. So I went and looked at the (immense, huge, disorganized) Talk page and couldn’t figure out who was disputing what. But I’m no longer confident enough of Wikipedia culture to just delete the stupid thing.

Second, there are irritants. People who know less than I about one of “my” subjects would come along and splash in stupidly-erroneous material, and people who write less well than I would uglify formerly-clear sentences. If I were willing to engage them on the Discussion page and do the dance, I could eventually end up with a good result, but it was kind of discouraging and very time-consuming.

If I were twenty years younger, with more time and general enthusiasm, it probably wouldn’t be a problem.

Also, I consider myself thicker-skinned than average, so when someone I consider an intellectual lightweight says that I misunderstand the Versailles Conference or phono cartridges, in a rude tone, I don’t let it get me down. But there are people in the world, in particular people who’ve put in the time to grow expertise, who are going to stomp off in a temper.

I bet a lot of them already have.

The Solution? · Those kind of people are going to really like the proposition: “I’ll write a carefully-crafted Knol on T.E. Lawrence, those wankers won’t be able to pee in my pool, it’ll get the link love and I’ll be the acknowledged authority. Yee-hah!” In fact, that very thought crossed my mind when I read about Knol.

The Problem · At the end of the day, I don’t buy it. Whether or not you like being edited (and lots of people don’t), on balance the majority of edits are improvements. And on balance, most Wikipedia articles are pretty decent.

Yeah, it’s a problem that being a Wikipedia editor is a less-lightweight activity than it used to be. But it’s not fatal. Being a gamer is less-lightweight too, and Warcraft draws millions. The fact that I can’t contribute as easily as I used to irritates me, but at the end of the day Wikipedia doesn’t really need me, it seems.

Udi’s one of the Really Smart Guys and Knol is a genuinely new approach. But I don’t think that there’s a really big problem that will drive people toward it.


Comment feed for ongoing:Comments feed

From: Carolyn (Dec 14 2007, at 17:24)

Well, Tim.

If you’ve seen (one of the) “Back to the Future” movie(s), I’d say you’re tempted by Knol the same way what’s his name was tempted by being called a chicken.

Is that the same as being considered a serious gamer?

I highly regard you, Tim. Be aware, I’m sure you are, of environmental distractions that serve you no useful purpose.

(Which is not necessarily to say that Knol is not… )

You know how you got my attention… Shalom.


From: pwb (Dec 14 2007, at 18:05)

I wonder if one simple change would make Wikipedia better: drastically limit article length. You would think that longer is better but I think longer tends to be superfluous, more prone to inaccuracy and controversy, disorganized, etc. When people go to Wikipedia are they really looking for depth?

I also think the increasing requirement for citations is causing article degradation. Again, it would seem that citations are a "good thing". But I think the result ends up being an article composed of a list of semi-related citations rather than a well-crafted, cohesive whole.


From: Tim G (Dec 14 2007, at 18:43)

I was really excited about the idea of Knol. It's a great way to "micro-publish". Rather than write a book on something, write some articles now and then. You'll get paid if they are good. Who knows, you might get more than you would from a publisher.

Then I realized what is going to happen. Someone (or some hundred people) will simply bulk-load every wikipedia article into Knol and attempt to collect ad royalties on it all. I think license-wise, they might even be allowed to do so. Then what?

I think the signal to noise ratio is going to be really high. Maybe google is right, maybe as the "search company" they will help you find the good articles over the cruft, but I'm a little skeptical.



From: ju! (Dec 14 2007, at 18:46)

you probably know them, but just in case:


From: Alistair McMillan (Dec 14 2007, at 19:51)

That POV tag was added back in July, by an anon editor with only one other edit.

He/She didn't leave any comments on the Talk page explaining what he/she was disputing, so I've pulled the tag.


From: Mark (Dec 14 2007, at 21:44)

(Google employee, not associated with the Knol project; I don't speak for Google, etc.)

I'll say now what I said when I saw an early internal talk on Knol a while back: I'm a bit concerned about waking up in 2012 and searching Google (or The Next Google, or whatever) for 100 different things throughout the day, and seeing the same 10 sites show up on the first page of search results. Centralization of content is dangerous for the same reason that centralization of services is dangerous -- single point of failure, juicy target for hackers, mass gaming/spamming problems, the ever-looming spectre of censorship, not to mention the indirect effects of shoehorning different topics into the same page structure. (I honestly don't know what sort of structural templates Knol will offer, but the general point stands. But I suspect that most knols will "feel" similar, in the same way that Wikipedia articles do now.)

I mean, the stated goal of Knol is to be the first page you visit while researching anything. Even assuming that Knol and Wikipedia and other sites don't overlap too much, what happens to the web if the top 10 search results for any research-y query are the same 10 mega-content sites?

I still worry about that, and I never heard a good answer. Maybe there isn't one.


From: Aristotle Pagaltzis (Dec 14 2007, at 23:40)

Says Alistair McMillan: “That POV tag was added back in July, by an anon editor with only one other edit.”

You know, the one thing I often miss when I do editing on random Wikipedia pages where I’m not subscribed to the page history feed (ie most of them) is that it takes so much work to do some proper revision sleuthing. It would be very useful if I could select a particular stretch of text in the page to ask “when was this bit changed?” and be shown only those revisions which touched it. That would make it a lot easier to see what was there previously and who changed it to the current form, both extremely useful for judging the merit of any change.


From: Janne (Dec 15 2007, at 02:49)

Mark, you have a point about concentrating a lot of relevant content into only a few sites making things brittle.

But you're working for the largest single point of failure today, and that situation looks to be worsening. You may or may not see the same ten sites pop up in your search in 2012, but you'll be hard pressed to find more than two or three sites from which to do the query. I'd be rather more worried about that.


From: Henry Story (Dec 15 2007, at 08:16)

I am just reading Benkler's "The Wealth of Networks" [1]. I am only a quarter through the book, but I can see that his point against knol would be that the size of the individual task is probably too large to make for a big commons based production success. Benkler mentions a few projects like this, including the wikipedia books project that have failed.

I would add that I don't quite see the point of knol, since one can get the same effect by just publishing an article on one's own web site.

Ie. it seems to be too close to the web peer production system, which works very well (we all just keep our own web site up to date), and link to other things that interest us.



From: Charlie Anzman (Dec 15 2007, at 12:58)

Tim -I too felt the same way and blogged it. Then it seemed things were improving but that was short-lived. Many of us have donated time and energy to make Wikipedia relevent....only to see someone come in and re-write the piece 'with an agenda' or spam links all over the place.

I have mixed emotions about Knol. On th other hand, Wikipedia is a GREAT concept. They need to clean up their act. Maybe make the 'rules' a little clearer and force editors to sign their work (with an on-file e-mail or something). There does need to be more moderation.

Is it my imagination or do they have a spam-bot going now??


From: TB (Dec 16 2007, at 08:59)

I just read about the Knol concept today, and to me it seems like the most interesting aspect of Knol is the fact that you can upload a quality article and get paid for it, while still offering the content for free to the reader.

I've checked up on a few of the sites that allow authors to upload and act as curators/editors for their own articles, and the articles there are consistently too sketchy to be useful from a research POV. Personally, if I want to find a research-quality article I use citeseer or google scholar. Wikipedia is useless for finding expert knowledge.

So as a publishing site for real research, I belive Knol could be successful (given the possibility of being paid for your work), provided efficient mechanisms to stop plagiarism are put in place. And since money is involved, plagiarism WILL be a problem.


From: Wes Felter (Dec 16 2007, at 17:37)

TB, can't you also get paid for your content using AdSense?

The coverage of Knol has focused on getting credit and getting paid, but you can do those already. The new part appears to be the content management tools, and maybe some free PageRank.


From: Stavros Macrakis (Dec 16 2007, at 20:53)

Hi, Tim,

I don't think "Knol is a genuinely new approach" -- has been doing pretty much the same thing for 14 months without notable success. But perhaps with the magic of Google, it will succeed.

As for Wikipedia, I agree with you that it is just too much work to maintain articles there, and the dynamic tends to drive out the competent and encourage the stubborn. I'd think that WP procedures could be improved to avoid this, but its ideology seems to keep it in this faux-egalitarian rut. Oh, well.



From: Paul W. Homer (Dec 17 2007, at 12:35)

Implicitly, information has a quality rating. When lots of it is dumped into wikipedia and fought over, it increases the quality. Blogs on the other hand cover the full spectrum of quality, and many blogs contain intermixed high and low quality information. Knol seems like a mini set of slightly structured sub-blogs, ranked by popularity. Can you actually trust the overall quality of that? If not, then what value does it really provide you?

I know lots of people stress that personalization is the next great thing, but with all the reams and reams of questionable information out there, for me the next big thing is summarization. What good does access to massive amounts of info. do for you if you can't trust it?



From: Wes Wilson (Dec 17 2007, at 12:56)

Google is under tremendous pressure to aggregate adwords revenue as soon as possible before an economic (read: advertising spending) downturn.

Imagine the gross revenue loss from all that wikipedia traffic. Knol is a prototype for google to figure out how much they have to spend to deal with wikipedia traffic control, either through an acquisition or through some type of negotiated search results cross-linking.

Th bigger picture is that wikipedia itself is blocking Internet users from getting to other sites that are relevant to the topics discussed. Wikipedia is slowly evolving into the substantive search start page, so they have to figure out how to promote useful pass-through. Otherwise, someone else will; this is the only real opportunity Knol has.

What we would like to see is a wikipedia-like first-fold with google-like search results following and (maybe) some paid-links in the middle. Obviously this sounds really awful to both parties. But Amazon is proving that it is the objective.

Highlighting authors is no different than highlighting columnists or algorithms, just another feature. Users have to feel comfortable that somehow the data is fairly accurate, and scoped as far as it is economically feasible to do so accurately.

Knol may turn into a database of columnists, which will provide an alternative reading, but it cannot replace mundane helpful fact presentation as a resource just because users prefer a pass-through experience.

Knowledge is an artifact of skills, not of information processing. Google has stated repeatedly that their main goal is to organize society's information for search, not to aggregate knowledge units.

In fact, it is very possible that eventually we will see even more wikipedia listings in the search results because that will stabilize google's role in providing general knowledge access. It will always be true that you can search wikipedia more effectively using google than using their own search engine.

One can only hope.


From: Anothr Mark (Dec 27 2007, at 12:06)

The fact is that more and more Google searches turn up Wikipedia sources as one of the top hits. So Google is sending people over to the competition. And people will start to ask: "why not just start every search at Wikipedia?"

Google doesn't like that.

It's all about money.


author · Dad · software · colophon · rights
picture of the day
December 14, 2007
· Technology (77 fragments)
· · Publishing (155 fragments)
· · · Reference (13 more)
· The World (107 fragments)
· · Life Online (263 more)

By .

I am an employee
of, but
the opinions expressed here
are my own, and no other party
necessarily agrees with them.

A full disclosure of my
professional interests is
on the author page.