Almost Wikipedias and innovation in free collaboration projects

I recently watched a presentation by Benjamin Mako Hill on Almost Wikipedia: What Eight Collaborative Encyclopedia Projects Reveal About Mechanisms of Collective Action (audio and video at the link) — or some conjectures about why Wikipedia took off, while 7 older English language internet encyclopedia projects did not.

The presentation had some interesting bits of net history, at least to someone as poorly read as myself — I had only heard about Interpedia (1993) and The Distributed Encyclopedia Project (1997) recently via a timeline of distributed Wikipedia proposals.

Through study of materials concerning the older projects and interviews with project founders, Hill arrived at 4 propositions…

  1. Wikipedia attracted contributors because it was built around a familiar product.
  2. Wikipedia attracted contributors because it was focused on substantive content development instead of technology.
  3. Wikipedia attracted contributors because it offered low transaction costs to participation.
  4. Wikipedia attracted contributors because it deemphasized attribution and “social ownership” of content.

…mapped on a grid reminiscent of many diagrams of “innovation quadrants” (example):

Or, as Dan O’Sullivan said with a broader stroke, ‘Everything is radical about Wikipedia except for the actual articles’.

At first blush this indicates that I should temper my enthusiasm for claiming that Wikipedia exploded the category of encyclopedias and that more free collaboration projects should aim to explode additional product categories.

Though early Wikipedians set out to create an , and I’m persuaded that presenting contributors with a familiar product to build helped it succeed, I think it is clear that Wikipedia, or more properly 845 language Wikipedias and other Wikimedia projects, have created a “product” that is much more useful and different from previous encyclopedias in ways that justify saying it has “exploded” the category. Yes, individual articles read more or less like previous encyclopedia articles, but then emails read more or less like letters. One approach to thinking about how big of an impact Wikipedia may have made so far would be to compare Wikipedia to surviving proprietary online encyclopedias (or perhaps hypothetical ones, had Wikipedia or similar never happened). I suspect the comparison would be akin to AOL and near peers vs. the internet.

One remain hopeful about free collaboration exploding further categories is that Wikipedia, and of course free and open source software projects, have innovated on process. There’s a huge amount of knowledge diffusion to be done, and further development of free collaboration processes, but a now new free collaboration project doesn’t automatically start out in a dead zone of innovating in too many dimensions if it attempts to include product innovation, as might have been the case in the past, as free collaboration equaled process innovation.

One small thing that we’ve mostly figured out that we mostly hadn’t figured out 10+ years ago (or 20+ years ago for software) is appropriate copyright licensing. I dimly recall from the past reading about copyright issues with keeping h2g2 going, and limitations for everything2 (no license is required, so entries are mostly solo-contributor), but I quickly looked at a few of the other projects Hill mentioned and found some curiosities. Interpedia’s FAQ on copyright doesn’t say what the project’s copyright policy is, but does express fear of an infamous patent. The Distributed Encyclopedia Project used what would now be recognized as an onerous and impractical license that at a glance I’m not sure agrees with the brief statement found at the bottom of an example article. The very first capture of states “All of the content is released under the Anti-Copyright License”, which sounds hopeful, though subsequent captures don’t have that statement, and the text of said license is not captured.

I’m really looking forward to Hill’s publication, as well as the further development of his research program concerning mass collaboration and social movements. Also, check out his reading list on AcaWiki.

11 Responses

  1. I had heard of two precursors as well — NuPedia and H2G2. For a fan of Douglas Adams, it’s sad that the latter didn’t succeed. But what were the chances humankind would get it right the first time around? Wikipedia was the culmination of many developments in both technology and culture.

  2. Ale Abdo says:


    This “processXproduct” thesis doesn’t sound quite solid to me. Or at least does not seem like the best way to frame the issue.

    Encyclopedias were structured the way they were for a reason, which we can call “usefulness”.

    It may well be that the useful model for content did not change between media, as it can be more related to how our mind interfaces with information than with peculiarities of each medium.

    So it could be that not familiarity, but usefulness, is what makes a difference. Familiarity standing out merely because the criteria for usefulness did not change.

    But of course, I haven’t seen the presentation or read the paper, and don’t know these other projects well. So I’m likely ignoring a lot of other points and evidence that must be in there. ;)

    But in any case it seemed like an interesting thought to share, specially related to the issue you raised of exploding categories.

    Thanks for the post Mike!

  3. Felix: Agreed, and there will be further such culminations.

    Ale: Of course a 2d grid throws out lots of dimensions. Still, I suspect familiarity and usefulness often are well correlated — one has to know how to use something for it to be useful, and as one gains this knowledge, the thing becomes more generally “familiar”.

  4. Ale Abdo says:


    Mike: I think usefulness has an intrinsic aspect that makes its relationship to familiarity not so simple; thus, identifying the role of each becomes important.

    Fun-ormally, I’m saying actual use is a consequence of both intrinsic usefulness and familiarity usefulness, which leads to different conclusions about, for one thing, exploding other product categories.

    I have a feeling encyclopedias were already well thought of in terms of their intrinsic usefulness (except for… the hyperlink), so it all looks as if familiarity was to blame, when in fact it could be that any attempt to modify the product made it first and foremost intrinsically less useful, with familiarity playing a minor role.

    So, if technology enabled free collaboration opens up room to make something intrinsically more useful, it might not need familiarity at all.

    Perhaps file sharing networks are a good example of such, even if freedom in this case is taken, not given. Indexed media archives were either small and private or domain specific and restricted to privileged groups.

    Even microblogging, as in practice there is an implied freedom to repeat and recombine posts, can be thought of as the explosion of a product based on free collaboration, and there was nothing quite familiar to it. Certainly not blogging.


    PS: why don’t I get an e-mail when someone replies this post?

  5. There’s no doubt that usefulness and familiarity are separate, useful (not to mention familiar) concepts and with respect to adoption (of “product”, “process”, or whatever) ought be analyzed as such.

    In this case (establishment of an internet encyclopedia) they seem to be highly correlated, but this may be illusory; I’ll have to consider that more.

    Another off the top of my head thought: usefulness might be easier to analyze. Usefulness is hard to demonstrate except through use. Identification of “intrinsically useful” but not used things seems difficult.

    p.s. Subscription to comments via email isn’t in WordPress core AFAICT. I installed and subscribed you to comments on this post, so you might get notification of this comment via email. :)

    p.p.s. “Fun-ormally” is quite a stretch visually and audibly, but I love it anyway!

  6. Rob Myers says:

    I think I’ve discussed on cc-community why failed. Basically it *was* a direct result of it being a proprietary sharecropping scheme, and it provides a usefully direct comparison with Wikipedia.

  7. Ale Abdo says:

    Mike: Hooray! I really like getting follow up e-mails, I hope this helps to make discussions on the blog more lively. Superthanks :)

    It seems I also like stretching things, that would explain me being all yoga, aikido, fun-ormal and the likes… haha

    Rob: from your cc-bizcom post, “The model proposed simply combines the disadvantages of both.” :D It amazes me how often people do that, in all fields of endeavour. Here in Brazil it’s as if there was a whole school of thought devoted to that ; )


  8. […] culturally relevant, if not dominant (the only unambiguous example of such as yet is Wikipedia exploding the category known as “encyclopedia”). One of Kẏra’s points is “The Web […]

  9. […] lot of what Wikipedia and Public Library of Science have done very successfully could also be considered “second […]

  10. […] a very popular site, even if it had been proprietary, rather than Wikipedia or other somewhat open contenders. Consider that Encarta, not Wikipedia, mostly killed Britannica, and that people are very willing […]

Leave a Reply