Post SXSW

SXSW: Blogging Where Speech Isn’t Free

Sunday, March 11th, 2007

On Blogging Where Speech Isn’t Free, moderated by Jon Lebkowsky…

Robert Faris of the showed a worldwide filtering map and a Venn diagram grouping jurisdictions according to whether they filter for political, security, or social content. Most that filter do so for all three. Filtering is very hard, so excepting a few jurisdictions that disallow net connectivity period, most attempt to induce a climate of self-censorship.

Ethan Zuckerman showed the map of press freedom and pointed out that blogging takes off in moderately repressive jurisdictions that restrict the formal press, sending journalists to the net.

Shahed Amanullah said there are many Muslims in the US who want to debate radicals on their websites but are afraid to because merely visiting those sites will catch the eye of US security. He also said that among other things we can do is to highlight the persecution of bloggers in the Muslim world.

Shava Nerad took on a number of FAQs about .

Jasmina Tesanovic mentioned the popularity of , which has a very impressive Alexa rank (1,376) considering its small and relatively poor potential audience (Serbia). The site is hosted in the Netherlands.

A questioner gave examples of the importance of expatriate media about repressive jurisdictions, which Zuckerman reiterated, using the term “” to describe expatriates and the stateless.

I completely forgot to ask a question about the overlap between filtering for political and economic protectionist (i.e., copyright) purposes.

Update 20070313: Read Zuckerman’s in-depth panel writeup.

SXSW: Semantic Web 2.0 and Scientific Publishing

Saturday, March 10th, 2007

Web 2.0 and Semantic Web: The Impact on Scientific Publishing, probably the densest panel I attended today (and again expertly moderated by Science Commons’ John Wilbanks), covered , new business models for scientific publishers, and how web technologies can help with these and data problems, but kept coming back to how officious Semantic Web technologies and controlled ontologies (which are not the same at all, but are often lumped together) and microformats and tagging (also distinct) complement each other (all four of ’em!), even within a single application. I agree.

Nearly on point, this comment elsewhere by Denny Vrandecic of the Semantic MediaWiki project:

You are supposed to change the vocabulary in a wiki-way, just as well as the data itself. Need a new relation? Invent it. Figured out it’s wrong? Rename. Want a new category of things? Make it.

Via Danny Ayers, oringal posted to O’Reilly Radar, which doesn’t offer permalinks for comments. This just needs a catchy name. Web 2.0 ontology engineering? Fonktology?

SXSW: Web hacks copyright

Saturday, March 10th, 2007

Sergio Villarreal and Kent Brewster gave an interesting, if mostly offtopic talk on Web Hacks: Good or Evil (or: Welcome to Web 2.666). Many web technologies started off as “hacks”, notoriously <img> and .

The rest of the presentation followed loosely from the premise that “content” is, will be (via services like Dapper) and needs to be “everywhere”, largely via feeds and now . From this came three observations:

  • JSON everywhere as an alternative to feeds
  • “IP” is a questionable concept
  • Suddenly, everything is hackable (e.g., via a service like

And three recommendations:

  • Don’t wait for pipes to drain your feed (publish JSON)
  • Don’t stop writing!
  • The web hates authors and loves writers (continue to create, as opposed to selling previous creations)

The last seems like an observation, or a repeat of the previous recommendation, but is a really nice soundbite.

The presenters struck me as being far too optimistic (or pessimistic if you want) about the impact of their technologies (Brewster is a technology evangelist for Yahoo!) — closing slide “Copyright is dead” and imagining a copyright-ignoring YouTube appearing in Kazakhstan, and having an impact.

Factoid: Brewster said Yahoo! has about two dozen full time people reviewing content flagged as porn, mostly moms, with higher than standard cubicle walls.

SXSW: Commercialization of Wikis

Saturday, March 10th, 2007

Evan Prodromou gave an excellent presentation on Commercialization of Wikis: Open Community That Pays the Bills. Check out his slides.

A few points:

  • Other stuff will be recognized as having wiki nature, e.g., .
  • Four categories of wiki businesses: service provider (Wikispaces, Wetpaint, PBWiki), content hosting (wikiHow, Wikitravel, Wikia), consulting (SocialText), content development (WikiBiz). My comment: at first blush Wikia would seem to be a service provider, but they are also deeply involved in content creation and community management.
  • Down with and the notion that wiki contributors are suckers or sharecroppers. Better to think of wikis (and wiki businesses) as platforms for knowledge. Contributors use your wiki to help each other, not to give you free content. My comment: I’m not so down on crowdsourcing. Yes, it is MBA language, but the usually involve compensating contributors. Crowdsourcing shouldn’t be conflated with sharecropping, nor confused with community purpose.
  • For wikis purpose more important than friends or ego for blogs (cf. blogs and social networking).

Seven rules for commercial wikis:

  1. Have a noble purpose — e.g., shared knowledge (use a free license), help a community.
  2. Demonstrate value — most interesting example is “carry the torch”; wiki communities can be transient, an entity that keeps focus helps.
  3. Be Transparent.
  4. Extract value where you provide value — most obviously, advertising for hosting.
  5. Set boundaries.
  6. Be personally involved.
  7. Run with the right crowd — e.g., open source and open content, or you will be suspect of being a crowdsourcer.

It appears that Prodromou’s Wikitravel lives by these rules and has succeeded.

Update 20070317: Prodromou has a roundup of blog responses to his presentation. It was great indeed catching up with him.

SXSW: Web business anecdotes

Saturday, March 10th, 2007

Web App Autopsy description:

There’s a lot you can learn from just looking at your own code line by line. Join us as we dissect a live web application that uses modern web technologies and see how the code can show us what it took to create a web app from idea to launch.

Instead the panel consisted of random anecdotes from four people who run web businesses, each of which has tweaked something or other. Oh, and look, conversion ratios!

A panel such as this desperately needs focus, and probably a moderator who has deep and varied experience in whatever the focus is, probably a consultant or academic.

Update 20070313: On the other hand, Sean Ammirati has a very postive writeup of this panel.

SXSW: Art, like, inspires me to design

Saturday, March 10th, 2007

The nearest The Influence of Art in Design came to its topic was showing a couple web sites that use color schemes to match fine art visuals incorporated in the site. Otherwise, it was all about “inspiration.” You should really try listening to some different music and see how that changes your creative process. Or maybe go to a museum and think about what you like about the pieces you see. Yeah.

The best quote I’ve heard attempting to relate art and design is “Art is the experimental end of design” from Caleb Chung, designer of the Furby. Sounds nice to me, but I don’t know how much water it holds. Is there a website or book that explores this using concrete examples?

SXSW: Online Publishers & Ad Networks

Saturday, March 10th, 2007

Skipped to the second half of Online Publishers & Ad Networks. I must have missed the part considering the question of “selling your own ad space, being part of a small ad network or teaming up with a search engine” but I imagine the first two have lots of explaining to do.

Nobody knows what form advertising accompanying short form videos on the web or small form devices will work best. I suggest that it should be as unobtrusive as text ads on a web page, or I will skip your video.

SXSW: Why XSLT is Hello World

Saturday, March 10th, 2007

Arrived about half an hour into Why XSLT is sexy to see in on the projector. What the heck were they talking about for the previous half hour? Left.

I have long wondered about using XSLT as an (untrusted) code distribution mechanism (e.g., acquire and run XSLT as an alternative to invoking a web service), but I suppose performance and functionality constraints make it a really niche case.

SXSWi wrap

Saturday, March 18th, 2006

There were a surprising number of panels more or less concerning entrepreneurship. I only attended one of these, Sink or Swim: The Five Most Important Startup Decisions. It was very mildly amusing but as far as I could tell the only important decision discussed was whether to look for outside funding or not, a well-trod topic if there ever was one. There was even one panel on Selling (Big Ideas to Big Clients).

I understand that was mentioned in passing on many panels. Attendees coming to our booth were much better informed than in years past, part of a greater trend.

The Digital Preservation and Blogs panel I was on was interesting for the self-selection of the audience — I imagine every librarian and historian attending were present. A writeup, photo, and my narrow take.

Both accepted panels I helped conceive went very well, especially Open Science. Though an outlier for SXSW the audience Q&A as high quality. Moderator John Wilbanks did a great job of keeping a diverse panel (open access journal editor, synthetic biologist, IT standards person, and VC) on point.

Commons-Based Business Models included Ian Clarke of Revver, which encourages sharing of short videos with an unobtrusive advertisement at the end under a CC license that does not permit derivative works. This licensing choice was made so that stripping out the advertisement is not permitted. Jimmy Wales challenged Clarke to think about opening up some content on an experimental basis. Sounds like a good idea to me. I suggested from the audience that attribution can require a link back to Revver, so even modified videos are valuable. Clarke responded that advertising at a link away is far less valuable. True, but the question is whether derivative works that could not otherwise exist become popular enough to outweigh those that merely remove advertising. I suspect many derivatives would be uploaded directly to Revver, allowing the company and original creators to take full advantage of additional revenue and to become the leading site for explicit remixing of video, a la ccMixter for audio. Seems worth an experiment — Revver is in no danger of becoming the leading video site at the current rate.

I also asked Clarke about interest in his patronage system. He said Revver is aimed at the same problem (funding creators) but was easier to implement. In the same vein I met John Pratt of Fundable, which is based in Austin. I got the impression he didn’t think the service could be viral (I disagree). I’ve written about FairShare, Fundable and related ideas several times in the past, mostly linked to in my Public Goods Group Shopping post and its comments. The field is ripe for a really good service.

The EFF/CC party was very well attended, even not considering its obscure location (an Elks club). In the middle of the facility was a room of Elks members, playing cards and other games, oblivious to the SXSW crowd that outnumbered Elks even in that room. I gave a very brief thank-you speech for CC, which I closed with a prayer (because we were in Texas) to J.R. “Bob” Dobbs (because we were in Austin).

At the end of the trade show Rob Kaye alerted me to the giveaway of every book at a well-respected computer publisher’s booth to “cool geeks” or similar. 5-10 years ago this would’ve really excited me, but this time I was mostly concerned about bulk and weight. I took a few. I suspect they’ll be among the last computer books I obtain, free or otherwise.

James Surowiecki gave a presentation which I did not attend but I hear focused on prediction markets. I should’ve made the time to attend simply to see the crowd reaction. Several of the latest sites cropping up in that field certainly look like they were designed by potential SXSW attendees — circa 2004/5 generically attractive web applications. I should have some posts on that topic soon, starting with Chris F. Masse’s 2005 Awards.

Bitzi as Tagging 1.0 Metacrap

Sunday, March 12th, 2006

On the Tagging 2.0 panel just cited as (more or less) a non-successful predecessor to Tagging 2.0 applications, saying something like “things like Bitzi (mumble) Cory Doctorow called .”

Vander Wal recently explained in a comment at Joho the Blog:

The big thing that was different, from say Bitzi, was people tagging information in their own vocabulary for their own reuse. Tagging information for others as a priority seems to make it far less accurate as a person may not understand the terms they are using (well understand them as other may).

He’s right. There’s too little private benefit to “tagging” at Bitzi, largely because what interfaces to what you have individually contributed are lame to the extent they exist. The Bitzi use case is rather different from and but it can learn a lot from them.