Thursday, January 17, 2008

Why is the Foundation funding someone else's startup?

I just heard about Kaltura, which apparently has been the topic of much discussion on the various mailing lists. I took a brief look, and I'm appalled.

This is a mashup editor, yee, fun. I see no use for this product in developing content for any Wikimedia project; all it does is allow you to mash together hunks of existing content in a superficial way that doesn't even come close to enabling content that would be of any real merit to Wikipedia or any other Wikimedia project. The examples on Kaltura are embarrassingly bad (although I suspect they're on a par with the crap mashups that Larry Lessig and his Creative Commons friends are inordinately fond of). In addition, at least one of them (the one on Naruto) is so clearly falsely licensed (it contains obvious content from the original works, which cannot possibly be relicensable under CC-BY-NC-SA) that it's evident that the Kaltura people have no concept of free content.

Nor, apparently, free software. Their product, which is released as open source, requires an activation key for use. What the hell? It's open source. You can just hack out the code that checks for the key. Furthermore, you can then take the version that has had the key check removed and republish it yourself. That's what open source means. Not to mention the fundamental contradiction of writing open source software in an inherently nonfree medium like Flash, but anyway....

It's pretty blindingly obvious that Kaltura is someone's attempt at a startup, appealing to Larry's mashup crowd, hoping to get bought by someone or another. Hey, that's fine, best of luck with that. What bothers me, though, is that the Wikimedia Foundation is providing seed capital, or at least its name and reputation, to someone else's startup.

So, Sue, what's the quid pro quo for the Foundation, here? If the Foundation is going to be supporting this startup, how much stock in Kaltura does WMF get in exchange? Cash? What's the deal here?


  1. I expect we'll see a lot more of these in the coming months: Wikimedia lends their name to some (nearly-)vaporware startup. There is a bunch of sound and fury, and ultimately nothing comes out of it.

    Kaltura's business model appears to be of the form:
    1) Collect Wikimedia Partnership
    2) Use the above to sucker VCs
    3) Profit!

    On Wikimedia's part they probably see it as a way of putting their valuable marks to use for *something*.

    It is pretty interesting to contrast the Wikimedia press release with the material Kaltura is circulating. The WMF press release is pretty careful to say nothing at all, while the Kaltura copy would have you believe that the stuff was running on Wikipedia today.

    I spoke to the Wikimedia business deals person some months ago and was clear that they are utterly desperate to find a way to make a buck off their good name, but clearly clueless as to how to do it.

    This model will probably work for a little while, at least for the startups if not Wikimedia, but the VCs will eventually realize that "Wikipedia = Touch of death", and it will dry up. Jimmy Wales' Wikia probably has them most of the way to thinking that already.

  2. I took a look at another video that Kaltura had on the devsite, which dealt with the Darfur crisis. Not only I noticed that it seemed to be a slideshow of sorts that I see on YouTube constantly, I also noticed the music was taken from another video. Credit isn't given in any of the pictures or audio that was used for the Darfur video. Fair use is a constant problem on Wikipedia, and with the way the Kaltura is just doing things, we will have a lot more on our plate than what we can deal with. Thanks, but no thanks.

  3. Since it is an Israeli company, I expect that the name Kaltura comes from the Hebrew khaltura (kh like the ch in loch). In other words, "shoddy workmanship." See

  4. Russian and Belarusian meanings for khaltura is also the same thing; poor job, cheap and/or low-cultured.

  5. I looked up the etymology on the Hebrew Wiktionary ( and am happy to translate it here: "Theatrical jargon from the Bolshevik period, when artists would perform outside the theater to earn extra money. It has since become a populr term. It generally refers to unreported side-income."

  6. "It generally refers to unreported side-income."

    So is this a start-up headed by Jayjg then, to supplement his Hasbara Fellowship income?

  7. Wikipedia allows collaborative text editing. Kaltera allows collaborative video editing. These are a natural match, in theory, so there's nothing arbitrary about some sort of collaboration.

    In practice, there's obviously a long way to go.

    The examples on Kaltura are embarrassingly bad (although I suspect they're on a par with the crap mashups that Larry Lessig and his Creative Commons friends are inordinately fond of

    Does anyone remember that cheap portable microphone setup marketed a couple years ago as "compose hip-hop while walking down the street"? The company didn't understand hiphop music, just hiphop culture, so they didn't realize that composing good hip-hop is actually really, really hard. It was like marketing a plastic portable keytar as "write rock songs while walking down the street".

    Kaltura's focus on mashups seems sort of similar. Mashups aren't intrinsically crap, they're just really, really difficult to make well (like a lot of things) -- there's no way in hell you can make a decent amv (or whatever) using tiny javascript controls.

    (Um. I'll also take this opportunity to link to a mashup I made.)

  8. Ben,

    My complaint about the Kaltera partnership isn't with what Kaltera is offering. Collaborative video editing is indeed an obvious thing for the Foundation to be interested in. And there are already several real open source projects to do this out there, some of them more advanced than Kaltera. But Wikimedia didn't collaborate with them. They collaborated, instead, with Kaltera, which is transparently a faux open source project that is clearly merely trying to play the startup game. They clearly don't understand open content or open source.

    I don't mind the Foundation supporting collaborative video editing; I just wish they'd do it with a group that is doing more than just trying to use the Foundation's name to attract VC interest.

  9. Google + YouTube.
    Wikia Search + Kaltura.

    Is Jimbo shitting himself?

  10. Kelly, this isn't a zero-sum game. There's no exclusive deal, there's no missed opportunities.

    While it's neat that Kaltura is making some attempt to take their app open-source, and we're excited if we're able to encourage them on that road, keep in mind that we can't and won't touch it unless and until it's actually suitable. (Given that it's based on proprietary Flash which supports *only* non-free codecs, that's going to be no mean feat.)

    We're not funding them, and we're under no obligation to support or use any software that doesn't meet our needs.

    In the meantime, there's lots of other projects at least as exciting, including the true open-source MetaVid which we're now hosting in our source control.

  11. A partnership generally goes two ways. In other words, it is mutually beneficial to each party. I can see the benefit of Kaltura claiming a partnership with the WMF. What, though, is the benefit to the WMF? The possibility that their product will be open source? As you say, all they are doing is "making some attempt," and the showcase of this--videos of Naruto, World of Warcraft, the Spice Girls, and Xbox 360--is hardly a shining example of free content. If anything, it just shows more creative ways to incorporate Fair Use into Wikipedia. And then there is the TechCrunch article which says: "Crowdsourcing video startup Kaltura is partnering with the Wikimedia Foundation to put its video-mashup technology on Wikipedia."

    As for helping other wikis, this may be a huge advantage to WetPaint or Peanutbutter Wiki or even to Wikia, but last I checked, assisting them is not in the mission statement of the Wikimedia Foundation.

    Brion, from your statement it seems that this is little more than vaporware. Then why create a partnership to promote vaporware? The benefit of the partnership announcement to Kaltura is obvious. What Kelly is asking is what possible benefit it has to the WMF?

  12. Brion,

    The anon immediately following your post asks the right question. Why is the WMF lending their name to THIS project, now? How does this benefit the WMF? What's the quid pro quo for WMF for lending their name to these other people's vaporware-based startup?

    I don't expect you to answer that; it's not in your bailiwick as CTO. That's a marketing and brand management issue, and I would expect an answer from Sue, or perhaps from Erik.

  13. "Collaborative video editing is indeed an obvious thing for the Foundation to be interested in."

    I'm on Ben's side here. Interest in? Sure. But look at the content on Wikipedia. The users who stick around are *serious* about getting professional results.

    For example, the photographs put up by the frequent contributors could easily stand up against any stock archive and, in some cases, even the material you find in National Geographic. (Incidentally, NG's own staff was utterly impressed (or at least feigned it well) by our volunteer contributors).

    Ditto for text, and even audio recordings on Wikipedia are increasingly held (by our community) to a *professional* standard.

    I think this is a big part of why video is not more popular on Wikipedia: $500-$1000 in still camera gear allows you to produce works that would look fine in any professional collection. Video? No. Not yet.

    So when we talk about web-based collaborative video editing, I come back to this. Can we expect any web-based video tools to produce professional results any time soon? Probably not. Even if the tools could do it, people that serious about their work would almost certainly rather use something local.

    And why should collaborative video come first? It's probably the furthest from being realistic for professional results. I think Wikimedia would get a lot more mileage out of something to serialize Inkscape's realtime XMPP based collaboration into MediaWiki's revision control. 0_o :)

    Brion, "We're not funding them, and we're under no obligation to support or use any software that doesn't meet our needs." Wikimedia staff member Jay Walsh sent out a request to the community imploring them to assist. So while the foundation may not be funding them with cash, it is trying to fund them with volunteer labor which has real value and which the foundation has failed to direct at worthwhile projects out of our own community.

  14. I don't know, I'm all for wikimedia/mediawiki engineers mixing it up with corporate talent -- this is one of the benefits from the san francisco move. (Just as long as history animations, usability overhauls, desktop mediawiki clients, and stable versions are on the way too :P )

  15. All I know is that Kaltura user BooBooGail is pretty hot.

    Great job, Wikimedia. Lord knows, you could use a dose of some hot chicks in your ranks.

  16. If its not free then forget it. If its not something thats fabulous, and instead meshes current content, then get rid of it.

    Whose idea was this.

    And good question...what would WMF get out of it.

  17. Why did we do it for *them*?

    1) They asked nicely, and were amenable to working towards the sorts of things our free software ecosystem requires.

    2) The agreement places no undue obligations on us. (We're not required to implement anything we don't like, and we're not dedicating our own limited resources to anything. Even the "beta" or "demo" is something someone *else* is doing.)

    3) There are potential benefits to us and the free software ecosystem.

    What do *we* get out of it?

    a) The satisfaction of encouraging a commercial company to work towards open source and patent-free codecs in the web video space.

    b) Another player pushing Ogg Theora and free-software Flash replacements (Gnash), two things we like.

    c) Potentially, someday it might be directly useful to us (like many other things in development in many places might).

  18. What reference can be given for Wikimedia Foundation funding Kultura, as opposed to just "partnering" with them in some way? If there is funding, I would expect that it was done with money given to WF specifically for that purpose.