If someone could produce a serious alternative to Wikipedia, would you join?

I don't really have the technical skills (or server space) to do it, but I've been thinking a lot about how to make a viable alternative to Wikipedia. This is what I've come up with:

Start off as a website that writes article which are licensed in such a way as to be eligible for upload to Wikipedia. Get a few good people together to write articles on topics of unquestionable notability and high quality. License these such that they can be used on Wikipedia, but only with a link from the main article page (not just some attribution in the history). As the articles move onto Wikipedia, you get 1) attention from the WP community 2) search engine juice.

Over time as the new encyclopedia builds up search engine juice and hopefully begins grabbing the better Wikipedia editors, it can become more and more independent gradually supplanting Wikipedia.

The second main idea behind my proposal involves a way to actually reach out to genuine subject matters experts. In addition to bland, "just the facts" articles, the hypothetical project would allow the posting of signed essays on the topic (e.g., An essay on WW1 might present the case that the "cult of the offensive" was the primary cause of the war). These essays would only be editable by the author, or person(s) chosen by him/her. Additionally, advertisements would be shown beside the essays, and the author would receive a set percentage (perhaps 50%) of the advertising revenue from his/her essay.

Optimistically, I would hope to receive essays from eminent scholars in the field (i.e., tenured professors) but that won't happen (at least at first). Thus, the really realistic goal is to receive contributions from graduate students. Students pursuing a PhD have several things that make them ideal contributors: 1) High quality seminar papers, which there are generally few opportunities to publish 2) A desire to see their name in print 3) A desperate need for money. By uploading a seminar paper to this hypothetical project, grad students would essentially give away nothing and could conceivably receive a decent amount of money through the advertising. Hopefully, in addition to publishing their essays, they would also help improve the overall quality of articles.

Thoughts? Anyone interested in helping to get this started?

Sounds rather like Citizendium to me?

It makes a lot more sense to publish articles on Google Knol, especially if they are articles on subjects which you actually know something about.

In some sense, I suppose what I am proposing is the combination of Citizendium and Knol. I think that there's something about combining a fact based encyclopedia with articles advancing scholarly views that is better than either of those things on its own. Perhaps I am wrong.

You will find it difficult to recruit people to participate. In general, people with the talent to write good, scholarly articles on a topic aren't going to do so on an ongoing basis without being paid for it, and it's even harder to get people to copyedit on a volunteer basis. You're going to have to generate revenue somehow, and that's not going to be an easy proposition.

I agree, that's why I'm proposing a way for these people to get paid for their work. Is your assertion that paying only for essays and not for articles won't work?

We have discussed this theme about four or five different times on Wikipedia Review. Have you done your due diligence and researched those threads?

Wikipedia has what some might call first mover advantage. This makes it very difficult for competitors to establish themselves because they must exceed the quality of the first mover from the get-go.

Think of failed Cuil search engine. It doesn't matter how evil Google is or even if Cuil could be better than Google with a bit of improvement. To most people Google is good enough and unless Cuil was better than Google from the get-go, and better by a wide margin, it wasn't worth people's time to remember the new URL, interface, etc.

Same with Wikipedia. Even if the rules of another site are a bit "better" or it is more reliable, unless it has more content than Wikipedia when it launches, most people who want to donate their time/content will give it to Wikipedia, since it reaches more people and is better known. And of course we've seen that more non-editors will use Wikipedia, as opposed to Britannica or other online knowledge sources, even though Wikipedia is not as reliable, so I am not sure that even a smaller, higher quality website would make a dent in things.

Wikipedia has blazed the path of user-generated online knowledge sources and I don't think the website that replaces it can walk the same trail.

Where did you get the idea about (1)? (2) and (3) are generally true, I think.

PhD students would make great editors since they know their topic fields and references better than most, but I'm not certain PhD seminar papers = good articles. Most of the papers I've read deal with a specific topic under study, and particularly in the sciences, would be too detailed to be useful to the general public (besides the fact they are written above the reading level of the general public).

Also, Peter, aren't websites like SSRN filling the opportunities hole?

What makes you think that Wikipedia would accept material licensed under these terms? Moreover, to post it on Wikipedia you'd have to release it under CC-by-SA, which is interpreted to mean that an article history is sufficient. So even if the material's initial licensing terms did not allow for attribution by article history, by posting it to Wikipedia you'd be licensing it under terms that, which would then allow the removal of any mainpage link you placed.

Indeed, Wikipedia will fight tooth and nail to avoid referring traffic to your site. Even putting a sourcing link back to your site in a Wikipedia edit summary or a footnote will likely result in an eventual ban for "link spamming" or "using Wikipedia for self-promotion" even if the content you're adding alongside those links is of absolutely stellar quality.

You have to remember that Wikipedia isn't really an encyclopedia; it only plays one on the Internet.

That I doubt, especially if the material is being added by somebody other than the copyright holder (in which case attribution in some form would presumably be required, depending on the license, with an edit summary link being the most reasonable way I can see of accomplishing it). I can't see an edit summary link driving much traffic, though.

I only work with people with credentials (of any level) in the real world.

So, let's start with... who are you?

Kindly,

Gregory Kohs
Founder, Wikipedia Review.com
18-year veteran of the marketing research business

The general idea is that the seminar papers would provide the essays and hopefully essay authors would also write the encyclopedia articles (I am unsure whether this is entirely reasonable). I'll admit to a fairly low degree of familiarity with what this would look like in the sciences, but I think it would work well in the Social Sciences and Arts/Humanities.


This is something I have to think about. The general idea, though, is to get someone other than the copyright holder to move the articles to Wikipedia. The hope is that someone would want to fork high quality work to WP.



Greg (or anyone else), if you are seriously interested in either helping me develop this idea or doing business together, then you can send me an email and I will give you my phone number after which we can talk through more appropriate channels.

I have no desire to write an encyclopedia and it now strikes me as an odd thing to want to do.

If beating Wikipedia was as simple as producing another Wiki with a better review model, then it would have been done by now.

There IS a way to produce an online encyclopedia which would overtake Wikipedia in a reasonable time, but first you have to think outside of the box, and get away from wikis in general as a model for how to do it.

First of all, the financial model of Wikipedia cannot be replicated (and shouldn't). Any publication that gives its editors nothing in return (not even recognition) will get the caliber of authorship and scholarship that it deserves. A reasonable financial model would encourage participation in the project and remove the limits to growth.

Secondly the Internet model of Wikipedia cannot be replicated (and shouldn't). Wikipedia's bandwidth and hosting costs are enormous and are the single greatest cause of Wikipedia's now never-ending search for money. In the current harsh economic times, it is this factor which could bring Wikipedia down. Any new online encyclopedia should have a plan to become all-pervasive while not incurring massive costs of hosting and bandwidth which would necessarily limit growth and accessibility.

Thirdly, Wikipedia's open collaborative model should not be replicated. The fact that edits from scholars can be removed, revised or scrawled over by any idiot is a key reason why no expert should even bother editing Wikipedia. There needs to be an implementation of scholarship and editorial control which recognizes and rewards excellence, both in terms of recognition but also economic self-interest. Giving the entire content away without attribution is one of the key reasons why Wikipedia is poisoning the Internet's flow of information.

Fourthly, Wikipedia's style of governance by mob acclamation should not be replicated. Instead any new encyclopedia should have a bona fide publisher and board of directors who will stand up and be answerable for the scholarship and quality of what it published.

Fifthly, Wikipedia's model of gaming Google and other search engines should not be replicated. Its too haphazard to try to do, and self-defeating if it only increases brand recognition at the expense of quality.

Wikipedia will never go away, but it can be marginalized if a better model can be implemented. The catch is that it requires a proper start-up with some financial muscle from early investors to make it happen.

I do have a plan for how to replace Wikipedia, but its not for here and not at this time. There IS a way to beat Wikipedia, rest assured.

imho, Google is the key to Wikipedia's success (if it can be s termed)
An article posted on the wiki will get the google recognition and hits
If you want your message - your pov - read (and believed) - put it up on wikipedia

unless there is an answer to this - wikipedia will dominate

However I'm curious/interested in your plan

Actually, since the switch to CC-By-SA, Wikipedia has generally been quite accomodating with regard to attribution to articles sourced to Citizendium; 73 articles are using the attribution template demanded by Citizendium, which includes a link to both the original article and the Citizendium main page. Which is actually pretty surprising considering how rubbish editors usually are at stuff like this. Of course, there's no telling the amount of articles which don't do this, but this is certainly a good step.

The general attitude seems to be that "reasonable" attribution required is being decided by the content creator rather than those importing the content.

The problem Wikipedia does pose for the suggestion is that it uses nofollow tags, so you don't actually get any "googlejuice" from the links in the articles.

Untrue. Bandwidth and hosting are only about 25% of Wikimedia's budget, and less than half of the "technology" budget, which also includes the purchase of equipment (most of which is strictly unnecessary; nobody in the CTO's entire staff appears to have any concept of what a "return on investment" is) and salaries paid to technology staff. The huge underspend they had in 2007-2008 was accomplished by deferring capital purchases that were not strictly necessary, and the $900,000 they spent that year is a good representation of the actual technical operational costs they incur. Hosting and bandwidth charges today are probably in the vicinity of $1.1 million, considerably less than their budgeted finance and administrative expense.

Wikimedia's dramatically increased budgets in recent years have been due to their desire to reach into additional directions, for reasons that are unclear to me at least, and also to their desire to offer well-compensated positions to people they like, for doing things that have no measurable connection with their mission.

I'm certain that they don't.

it's already starting to be done. now it is just a matter of all you helping to make it even better and contributing. (see ENCYC.org/wiki/ )


then cause a certain "feature":

in response to :

http://wikipediareview.com/index.php?s=&sh...ndpost&p=181989

aside from being the current logo of Wikademia.

http://ja.wikipedia.org/wiki/%E3%82%A2%E3%...5%B0?uselang=en

Which is why carefully constructed MMPORGs (like WoW) are serious alternatives to Wikipedia.


KillerChihuahua could have legitimately blocked you, had she known that for a fact.

Then again, in an MMPORG, you don't really need a valid reason for clobbering those who are not on your team.

The idea that you have to be big to be a "serious" alternative is always floating around, for some reason.

Check out Briolette of India, and its Wikipedia counterpart, and you can see that we at Encyc are just as "serious" as Wikipedians.

However, within academia-like and university-like wiki environments, they would make good content/articles!

wubidy wub

Yes. Quite.

it is just a mostly better sort of serious.

Q. Can it be done?
A. Only if wikipedia "stumbles" somehow. They have such a huge head start on content it would take at least a decade for any serious website to challenge them, even if the alternative has superior articles.

or - if google (and the others) gave better rankings to the alternatives

Sort of like this?

Does anybody ever actually make that search?

Maybe once a month or so. That's why I said "sort of" -- I realize it's not a strong empirical example.

Wikipedia has Google's algorithm pretty figured out. The mystery is why Google doesn't simply just change the algorithm to prevent Wikipedia from stealing Google-juice.

Why should search results depend on whether the proprieter of a website chooses to use "nofollow" or not to use it, or has millions of internal links like a link to "amalgamation" in an article about World War II?

Until Google gets its act together, I'm using Yahoo and Bing.

Is there any evidence that either Yahoo or Bing are any less algorithm-enamored with Wikipedia than is Google?



Here is an actual search that brought traffic to Wikipedia Review today.

They're at least as wikipedia-heavy. Bing seems to be even more so, usually returning at least three wikipedia pages in various random languages on the first page, in addition to some very... odd results. Bing is to search engines what Windows 3.0 was to operating systems. It's quite hilarious, actually.

you don't think that the big wiki has not already stumbled?

it seems to me ---> i really don't even think that that is a bad thing, even for the big wiki and all of the stakeholders involved. but silly me; what do i know?

Sure. Look up any health problem (e.g. Rabies, Polio) and Bing puts the Mayo Clinic at the top. Looks like humans made a decision.



We all agree, even Wikipedians, that Wikipedia articles are not perfect and there is lots of room for improvement.

Some of this improvement could happen more quickly and easily on other websites besides Wikipedia.

Or maybe one person's idea of improvement isn't the same as another's. Wouldn't it be great to have two articles? Or how about multiple articles, all geared towards different audiences?

Should the WMF be the only organization that does this? What are the risks with leaving it to one organization?

Searching for polio and rabies on Bing gives me en.wikipedia as the top result, and sv.wikipedia in the top 3. No sign of the Mayo Clinic. Bing seems to do some fairly heavy unasked for and quite random geolocating. Oh well, this explains why Americans I speak with don't quite understand why I think Bing is so horrible.

Google (and the others) vary by country code, as well as google.com compare with google.co.uk google.ie google.nz etc for bing its www.bing.com/?cc=uk

I am aware of this, hence why I use .com for them all. There is a bing.se and a google.se but I have no interest in either. Google.com doesn't geolocate. Bing.com does even when I have explicitly tried to tell it not to, with completely unusable results. (And I don't mean bad. I mean "Server not found", "THIS DOMAIN IS HOSTED BY SURFTOWN" and "London Escort Agency" and similar in the top 5 results, no matter what I search for. Yes. Bing is that bad). Google is now the only usable search engine, basically, and a serious alternative to wikipedia needs to game it. I don't condone google being this dominant (I hate google for what it did to DejaNews). I just acknowledge a basic fact. And I find it absolutely horrible that the three stubs I have created on wikipedia are all in the top four search results on google. It can't just be gaming the google algorithm. It has to be google adding anything *.wikipedia.org to their top results by default.

What makes you think this? I've seen pretty strong evidence that google.com geolocates.

The search results I get, compared to the search results my non-Swedish friends get (so far, they don't seem to differ. We all use .com). Google.se geolocates. If google.com geolocates, it is completely impossible to see for me. I don't get any Swedish or Swedish-related search results (that I notice). What evidence do you have for geolocating from google.com?

Same search returns different locally-relevant content when run in southern California and northeastern Illinois. Back when the company I used to work for had multiple egress points, I could get different search results simply by manipulating my egress point. (This came to my attention because someone in our Texas office asked why Google seemed to think he was in California; the answer was, of course, that our Texas office ordinarily egressed in Los Angeles.)

Hmm. Interesting. I will bug various US/Canadian friends to test this, but it may be a "geolocate within the US" thing, since there is where google gets most of its ad revenue.

I know that as an occasional buyer of Google AdWords ad placement, I have the ability as a buyer to limit or filter how my ads are shown:

Budget

Time of day of the browser

Day of week

Country of browser

State of browser

Metro area of browser

Town of browser

If you wanted to, you could run an ad about Jimmy Wales that only shows up for people geolocating to Huntsville, Alabama.

If they're doing it for AdWords, I have to imagine they're doing it (to some extent) for search. Type the word "weather" -- what happens? I can't tell, because I've asked Google to remember my home ZIP code.

It works even if I delete all google.com cookies

The easiest way to test this is access Google (or Yahoo, or Bing) using Tor, and frequently deleting your cookies. I've probably seen localized content for 50 different countries.

Well, It seems google may geolocate a bit, but in an unobvious way. With Bing it is extreme, with half the results being irrelevant sites that just happen to be .se. Annoying, google makes it appear as if their results are country neutral which is what I want. Oh well, you live and learn.