Graffiti, photos, urbanity, free culture. What's not to like about Jonathon McIntosh's StreetJams.

Speaking of free content, the United Nations has a cartographic division that's providing a number of good quality, free maps. And their terms of usage are quite liberal.

Hat tip to Resource Shelf

Being deeply interested in social media, I'm always on the lookout for freely available media to seed projects. Our very own United States Library of Congress is embarking on a project to digitize a huge number of US newspapers. Fab-u-lous!!

Then good old US copyright rears its ugly head. It's not so much that there the materials might be copyrighted, it's that you can't know and can't find out. Until you make some money and the vultures start to swarm.

Anyhoo, the initial seed for the project is "Stars and Stripes", an old wartime US government publication. I have to imagine that's in the public domain.

Hat tip to Waxy

Flickr's image annotation is neat, but relies on Flash for presentation. I'm not inherently opposed, but I've thrown a few students on the shoals trying to recreate it. They've mostly foundered on the rampup to learning Flash.

Gina Trapani and Stuart Langridge have come up with imagemap and DHTML combinations that have a similar effect. I'm guessing they're both a little easier on the learning curve, and much easier to generate programatically.

Okay, I was little burnt out after getting sick, attending the CSCW social networking workshop, attending parts of CSCW, and throwing in a couple of lectures to boot. More about the workshop and conference later, but time to get back in the saddle.

An observation about the Web in general, and the blogosphere (or blogospheres as our President probably believes), in particular. They both suck for retrospection.

Suppose you have a link, or a blog post, and you want to get a picture of the history of that content over time. All you've got is the present. You can't ask Google , or Yahoo, or Technorati, for query results "from" a given time window, e.g. give me all the relevant pages for an old post of mine, but limit them to a year ago. Then you could start to build a picture of the time varying context of a piece of content and better understand where it fits in to the blogoshpere and how it got there.

Not much out there enabling content forensics.

David McDonald, Shelly Farnham, and Danyel Fisher put together a small workshop on social networks at CSCW '04. This will put me in some fine company, including that of one Barry Wellman, network analyst extroidanaire. Trip report late next week.

I posted earlier about interesting things that might happen if more people blogdumped as Tom Coates did, and some nice visual examples of blogging behavior are coming out. After my post, Coates subsequently lamented that there wasn' t a deluge of interesting vizes, but these things are hard work! Heck, it probably takes a little bit of time to figure out his data format and then a good way to interanlize.

In any event, through the follow up work I ran across Tom Carden's processing sketchbook, a collection of interactive applets generated using processing.

If you'd like to program a distributed application that has producers and consumers decoupled in time and space, you can try to use any of a bunch of off the shelf message queue services, open source or closed from the big boys.

You could also try to ressurect the glory days of tuple spaces. Good luck!!

Or you can sign up for an Amazon developer token and use their remotely hosted, Web services based messaging queue. No muss, no fuss, no questions asked, just queue away. The dang thing even has a REST interface!!

How sick is it, that the world's biggest bookseller is the world's best web services provider? Google's not evil, and Yahoo!'s rich, but nobody walks the walk better than Amazon. They don't strut about telling us how many PhD's they have or cooking up obtuse billboards as job applications. They just create and support interesting tools that people can really build upon.

I have to believe this is some scheme to get real world pounding on a core piece of their infrastructure, for relatively low risk and payout. Oh and if it really catches on, they can start billing to boot.

Okay, we'll ignore that ugly little 1-click patent wart there for the moment.

Greg Linden announces that Findory redesigned and reworked its infrastructure.

A whole lotta movin' parts there. Pretty doggone brave in my book, but they seem to to have pulled it off. It would be interesting to hear the backstory of what their architecture did and now looks like.

This year it seems the in thing, at least in the blogosphere, to make it clear who you're voting for, presuming you have a US vote. We're pretty much a politics free zone here, but in the spirit of being a full participant in this nascent media ecology, here are my thoughts.

In general, I'm inclined to vote Democratic. Last election I claimed that the two candidates were just different flavors of vanilla, or castor oil depending on your point of view. Boy was I wrong.

I feel the executive branch has failed many Americans in the past 4 years. Many more than it has benefited. While some problems were due to circumstances that were uncontrollable, many were failings arising out of poor policy and/or planning. The talents, faith, and lives of many citizens have been flat out squandered.

I have no illusions that Kerry will turn out to be a barn burner, but due to lack of execution, heck not even sticking to their own rhetoric (c.f. nation building, growing the government), this administration should not be returned for another 4 years.

So I'm voting for Kerry.

One of the neat things about Laszlo Systems toolkit is the way it turns Flash into just a runtime system. Basically, you have a compiler for an XML and JavaScript like language, that targets the Flash player. Missing from the toolbox were components that allowed one to do really dynamic, highly visual, interactive elements ala processing. Granted, I didn't break a leg looking for one, but I did some extensive browsing through the documentation.

This just came across my del.icio.us transom, but Nicolas Cannasse has released MTASC, the Motion-Twin ActionScript Compiler. This is a command line application that takes ActionScript code and generates an SWF movie, again treating the Flash player as a runtime engine. Not exactly the same as Laszlo, but in the same spirit.

Two big wins. First, for those of us who are visual ide impaired, this gives us entree into hardcore Flash programming. Second, it serves as an examplar for targeting other languages at the Flash player. Think writing Flash movies in something Pythonic.

Apparently MTASC is open source and written in OCaml. Vive la programming du functional!! (Please correct my French.)

Mark Glaser recaps some recent hyper-local citizen's media efforts. This includes some guarded praise for GoSkokie.com. Yup, we never got over the sustainability hump, but hopefully with Medill's brand name on such a project other j-schools are legitimized to take a stab.

Also, GoSkokie isn't completely dead. I should note that a section of Medill students is now posting more traditional stories to GoSkokie. We'll see if that brings things back to life.

Tom Coates is providing a complete archival dump of his longrunning blog. If a concerted number of folks do this, it could provide a lot of grist for blog tool developers. Think what kinds of interesting studies would be enabled by having clean captures of a significant number of the top 100 blogs, however you decide to rank them.

One of the issues that blog researchers looking to do empirical studies have is simply getting raw data, especially of a historical sort. Writing crawlers seems simple on the surface, but let me tell you, once you decide to scale to any significant number things get hairy fast. If you want to get full archives, you'd better have IBM class skills and resources.

The real scotcher though is redistribution. It should be obvious that if a researcher collects their own archives of publicly available pages for private analysis things are fine. If the copyright isn't clear though you could potentially get into hot water giving out someone else's content. Maybe there's enough Creative Commons sites out their though to make a start.

I wonder how the TREC folks get around this issue? I know they have a Web track and sets of data.

Anyhoo, coordinating the collection, verification, and redistribution of volunteered blogdumps would be a good task for some enterprising academic.

I'm sure having Cal Henderson around to make the presentation would be really helpful, and probably much more entertaining, but the PDF of his slides for a presentation on Flickr to the Vancouver Php Users meeting still has lots of crunchy goodness.

Having seen two or three of these slide decks from other compaines, I'm struck how these systems become polyglot so fast. In Henderson's talk, a template language (Smarty), a server page language (Php), and a systems language (Java) all make appearances. Not to mention good old HTML, XML, and SQL. Oh and throw in some Perl to help coordinate all of this stuff.

Slides hosted by Niall Kennedy. Received via some twisty path probably passing through del.icio.us.

And yet another vote for Php sucks, although I'm guessing it was tounge in cheek.

Over on the Yahoo! Search Blog, Jacob Rosenberg documents some searchbox tips for focusing your news hunting.

Searchbox as command line yet again.

A few days stale, but worth reading is Greg Linden's summary of why apps shouldn't ask people about preferences. A modern app should watch what they do and then divine the rest. I take it that doesn't mean "don't help people get things done", but don't ask them to explicitly rate stuff, or maintain preference data, or respond to repeated off task queries.

I think this is the direction Aggregator 2.0 will eventually have to head. For a number of folks, maintaining blogrolls and the generated flow just won't work. People who need to read tea leaves like business intelligence analysts, political operatives, public relations, cool hunters, etc. etc. I'm still working out the profile, but people who have to observe communications in fast changing environments where trusted sources may not even exist.

The trickiness is that some of what will be needed is social information. This leads to needing critical masses of people buying into new infrastructures. The only obvious way to do that is through Web interfaces and plumbing, which unfortunately makes it somewhat more difficult to watch what people are doing.

Link parkin': Luke Wroblewski hijacks Edward Tufte's concept of sparklines and takes a serious stab at applying the concept to blog postings.

On a lark, I requested an account on Rojo.com through an e-mail address posted in Jeremy Zawodny's Web 2.0 summary on the company. In case you're a newcomer (apparently there's something of an upsurge in interested folks) I'm a bit of a webfeed aggregator nut, even claiming to do some research on supporting technologies. So joining Rojo is a natural.

I'm planning on reporting in ongoing depth on my experience using Rojo, but I have to say the out of box experience was less than compelling. First, I took the default set of feeds for a new user. It's a large number, which isn't a problem, except that you have a ton of items in each feed, so out of the box a naive user is facing information overload.

My second issue has to do with the way social networking is integrated. In short, it's Friendster style as opposed to del.icio.us style. You have to invite people in as opposed to being able to easily see what other people are doing.

The interface does seem nice though. That might sound like small potatoes, but doing a good interface in a browser is no small feat.

The option A out of the box experience seems broken to me, though. I'm overwhelmed, not much of an obvious way to catch up, and I have to work to connect to people. I must admit that I am way biased as a webfeed aggregator power user and a Bloglines user to boot, so I may be just the wrong audience. But I haven't dug in, so all is not lost.

By the by, if you'd like a Rojo invite, send me some e-mail. No guarantees, but they do allocate a fixed number to pass out.

Update I'm out of Rojo invites. A recent (Feb 4, 2005) rush of publicity seems to have quite a few folks now looking for them.

Noted earlier, but link parkin' for now: Alexa Web Information Services APIs

Jon Udell posits some relatively plausible scenarios that could drive the growth of linkable continuous media. The ability to point at chunks of audio/movies from given offsets is severely lacking with current formats and players. Even worse is integration of the presentation of these chunks with other media, e.g. accompanying text. Podcasting and VOIP may be the drivers for improvements in this area. Of note is the connection by Real Networks' Rob Lanphier to the open source, Helix toolkit, which provides a hackable version of Real's player and server software and could form the foundation for making this easier. SMIL is the technical underpinnings of it all.

It'll be an interesting time when the world easily and routinely links into more than web pages.

You know what I'd like to see on del.icio.us? A way to filter out bookmarks into del.icio.us. I never understand why people need to bookmark pages of del.icio.us folks or tags. Just subscribe to the dang RSS feed!! Having an RSS subscription is just as good as a bookmark. Even better you don't have to visit the site manually.

Actually, I have a sneaky suspicion that these irritating urls start out as life in browser bookmarks which get exported/imported into del.icio.us.

Of course this is anathema to the del.icio.us way, but power users can wish can't they?

Is that a GPLed Linux server in your pocket, or are you just happy to see me?

I wonder if it's a Usermode Linux VM that somehow gets booted on the machine or a Windows application that networks into a remote Linux VM?

Via Josh Lucas

Link parkin': How to make your own photo mosaics.

Via Engadget

Greg Linden does a great job keeping an eye on Google visiting UW and opening an office in Microsoft's backyard. He posits that the move is an attempt to capture excess MS and Amazon talent. I'll also point out that it puts them near the UW CS department, one of the top systems departments in the land. A proximity that'll make it easier for student internships, joint research projects, grad student hiring, faculty consulting, and faculty defections to get off the ground.

Just a thought, but 20 years from now will we be talking about Google as the Xerox Parc for the Web era? With better management to boot?

Many bloggers have years worth of content they've produced, but where are the personal tools that help them take better advantage of it? I've link parked the same link multiple times. There are long running, latent threads in my archives that could be packed up and re-presented to me. Why isn't there some little agent that looks at my blog posts, picks out the urls, and hits up del.icio.us, Google, Alexa, and Yahoo, for related material. It doesn't even have to post it directly to my blog, but just e-mail it to me to jog old memories or add different perspectives.

The blogosphere has a supremely short event horizon. Mining the entirety of an author's or self identified community's past publishing is an area where someone could make some real hay. And with even more sociable media mechanisms for people to splice themselves into, there's even more grist for the mill.

Peter Merholz neatly summarizes the excitement around ethnoclassification, a.k.a. folksonomies, social tagging, etc. His contribution to the conversation is a call for system designers (del.icio.us, Flickr) to figure out a way to pave over the agreed upon and highly trafficked tags.

To add my 2 cents, as far as I know you can tag urls (del.icio.us), photos (Flickr), email (GMail), webfeeds (Rojo) and, slowly, blog posts. Someone needs to fix up the music players so people can start tagging tracks and playlists. Granted, this can be reduced to tagging urls, but the tags can actually be carried around in the audio files and playlists. Individual tracks might be too fine grained and too much work to be worthwhile, but tagged playlists would be fun!!

Apropos of nothing ever posted on this blog.

The One Ring was destroyed.

The Berlin Wall fell.

The Evil Empire has collapsed.

And there was much rejoicing.

This one post on Jeff Jarvis' Buzzmachine, singularly illustrates why blogging is just horrible for tightly coupled conversation. C'mon, what percentage of people on Earth can and want to actually track a discussion that way, much less try and make sense of the spatially and temporally distributed fragments.

I grant that USENET is a bit of a mess, but it sure as hell got some things right.

Apropos of nothing, del.icio.us has zero egoboo. There's just about no way to distinguish one user from another other than by their bookmarking behavior. With the outbound link on the profile page, you can do a bit of namebranding, but that has little import for how good a user's link stream will be given the varied ways people use the system.

I wonder what an appropriate network growth model for del.icio.us subscriptions would be and whether the system actually conforms to it.

As evidenced by the wastage of the comments, Jeffrey Veen launched a flaming Molotov cocktail towards an open source beehive. I can summarize his post in four simple statements:

• Make installation easier
• Write better documentation
• Use CSS to make the design flexible
• Push system administration to the background

There, that wasn't so painful now was it?

The open source part is a red herring which simply generated a bunch of noise. These features could be demanded of any cms. The real question to be asked is why commercial vendors haven't filled the void if open source stuff is really that bad.

Frankly, instead of starting with open source cmses, he might have had better luck seeing if wiki software would have been a better fit.

A number of folks, including me, have wondered why the term "social software" has picked up traction, when there's a whole field of academic computing (computer supported collaborativ work (CSCW)) which encompasses just about every aspect attributed to the term.

Adina Levine is working on drawing out the interesting new features of social software. I'm still puzzling out if there's really something there, but elements ring true. For a half a minute, I'd bought into loose coupling but realized that many of the services cited (Technorati, Flickr) are even more centralized than USENET ever was.

Maybe what it boils down to is that these systems embrace the two aspects highlighted by Clay Shirky: first class group formation, and sociable media (my term). Anything else is just reaching.

Hat tip to Seb Paquet

Trolling around in the backwash from Web 2.0, I discovered Ludovic Dubost's XWiki, a Java based Wiki toolkit with a programmatic API. Open source and GPLed.

The research arm of the Australian Commonwealth, CSIRO, has been working on tools for making continuous media of the web. Called Annodex it may be the solution to many of Jon Udell's issues with hypermedia.

Hat tip to Wesley Felter.

You'll hear about it from a bunch of different places, but Google has released a, Win32 only, desktop application that augments your searches with local stuff. Hat tip to John Battelle, who has a lot of deep dirt on the release.

Via the Online-News mailing list, the Missouri School of Journalism is making a foray into citizen's media with MyMissourian. Looks very much like what we we're doing at Medill with GoSkokie except with more spit and polish, and as Clyde Bentley reports, more and better organized manpower.

While not busting out all over, it appears that journalism schools are at least giving their students an opportunity to encounter citizen's media. It may take a few crops of students, but eventually the viability of the concept should trickle up the management hierarchy.

Of note as well is MyMissourian's use of the open source Mambo content management system. As opposed to Geeklog, which is mainly a volunteer effort, Mambo appears to have the semblance of a professional development team backing it. As an example, the quality of the documentation is markedly different between the two. Also, since Mambo isn't specifically blogging focused it has a touch more sophistication in dealing with other media. Plus, as opposed to MovableType, Mambo is still both free as in beer and free as in freedom.

Don't quite know what's going to happen at the Online News Association 2004 Conference super panel, but it should be entertaining. Panelists include (according to the ONA website):

• Arianna Huffington, nationally syndicated columnist and former gubernatorial candidate
• Joe Trippi, Howard Dean's former campaign manager, MSNBC commentator and author of "The Revolution Will Not Be Televised"
• Mickey Kaus, Slate blogger
• Jehmu Greene, Rock the Vote President
• Dave Winer, Democratic convention credentialed blogger and organizer of ConventionBloggers.com

Media Lab projects are so tantalizing. Witness John Maeda's Physical Language Workshop, a follow on to the Aesthetics + Computation Group. The PLW is supposedly working on a project called the Treehouse Studio, underlying the OpenAtelier. The Treehouse Studio embodies a concept that's just about due: really nicely designed and integrated rich media authoring and management through the web. About a year and a half ago, I was talking about a similar project that had Alexa and John Seeley Brown involved. Bits and pieces are floating around, but I was hoping the PLW had made some major strides. Weblogs made text read/write for the Web, and photos are getting there, but there's plenty of strides to be made for other media.

What can I find related to Treehouse Studio? Not much other than year old pre-alphaish demos. No papers, no source code, no updates. The rhetoric on the PLW web page is great, but digging deeper all I get is vapor. Still I can hope that another hardy band of students, many of the current crop seem to have moved on, wash up on Maeda's shores and pull it all together.

Spotted on the SmartMobs website, with Open Semacode, now you too can start building your own "Find Nearest Printer" application.

I normally don't pump The Firm's activities too much in this venue, but there's a few exciting developments that may be of interest to the 6.5 folks reading this blog.

First, our School of Communications is trying to build up the Technology and Social Behavior specialization. Justine Cassell and Eszter Hargittai are pushing things along, and the program has a fabulous speaker series for those in the Chicagoland area, or happen to be in the area on the right dates.

Also, NU is placing a big bet on the interdisciplinary study of complexity. The university has established the Northwestern Institute for Complex Systems (NICO). Not only is NICO hiring postdocs, they're also having a great seminar at the end of this month. Unfortunately, I'm not sure it's open to the public.

And if you want to know what it's like to be an intern in Medill's Global Program, check out the postings at the informal Global Program interns blog.

Danah Boyd notes some contrasts between her observations of youth communication technology usage and the hype coming out of Web 2.0. There's probably a kernel of truth in her observation of a generation gap, but I'd counter that the Web 2.0 crowd, while important and influential, isn't the totality of computing. While those are the big money toolmakers, there's an element of youth culture out there that can make their own tools. Ultimately, I'm not sure if there's a problem here.