November 2003 Archives

I'm finally getting a chance to catch up with posting my notes from the 2003 ECAR Symposium that happened last week. This was a really good forum - smallish (140 attendees), short (day and a half), with great content.

ECAR is the Educuase Center for Applied Research. It is largely the brainchild of Richard Katz, who is not only Vice President of Educause (the largest professional association in higher ed computing), who is trying mightily to get us to focus on finding out what is actually happening in our field so we can make decisions based not only on intuition and anecdote but at least partly informed by real data. Not an undertaking for the faint of heart - something nobody would accuse Richard of.

Alan Kay gave the keynote presentation. His talk was titled "Back To The Future - Real Math, Real science, Real Children, Real Computing" and he spoke of how it is possible to teach very young children math, science, and computing in ways far more meaningful than is commonly done today. He talked about how children learn from actually doing things and observing results, and can then generalize from those results, rather than by memorizing wrote answers. He showed a fifth grade class where they recreated Galileo's gravity experiment, dropping balls of differing weights from a height, and recorded the results on video. Using the Squeak programming environment, they were able to analyze results side-by-side for the different drops, and time the ball drops.

Kay's wry comment: ""In every class of children there's usually one Galileo - interestingly enough it's usually not the teacher"

My complete notes from his talk are posted here.

The Sakai project, a new project to combine several open source projects in higher ed including OKI, uPortal, CHEF, and OSPI, has put its prospectus on the web here. This is a project that has the propsect of being very important, and is well worth keeping an eye on.

I've been meaning to write about KEXP for a while now, and looking over their new web site redesign reminds me to just do it.

We in Seattle are fortunate to have this great listener supported radio station in our midst, but everyone else can listen too, in multiple streaming formats and bit rates - including a 1.4 mb/second uncompressed full CD-quality stream, if you've got the pipe to handle it.

It's not that I necessarily like everything I hear on KEXP, but rather the fact that the DJs are all wonderful fans of the music they play and they each have the latitude to explore the music they love. Today I heard a story on NPR that noted that there are something like 30,000 albums released each year - I increasingly am relying on the experts to help me sift through that, and that means paying attention to folks like Amanda Wilde, Michele Myers, and specialty folks like Jon Kertzer (African music), Don Slack (Country twang), and my all time favorite DJ anywhere, John Gilbreath (new jazz and associated explorations).

The station keeps a two week running archive of all of its shows, plus archived playlists so you can see who played what when, as well as streaming archives of live broadcasts it does.

I'm more likely to discover new exciting music here than anywhere. KEXP's slogan is "Where the Music Matters"...and it's true.

These are just too good! Thanks to Adam Woog.

We got our first dusting of snow here in Seattle this morning, and (just coincidentally) we're off to San Diego for the Educause Center for Applied Research Symposium.

I'm looking forward both to the event, which features a keynote presentation by Alan Kay, who was at least partiallly responsible for inventing computing as we know it today (the GUI, object-oriented programming, etc), and to being at the way cool Hotel Del Coronado.

I'll be reporting from there as possible...stay tuned.

Recently my wireless at home has been exhibiting strange behavior - the laptop makes the connection and works fine for about a minute, then network applications just stop working, though XP reports that it has an excellent wireless connection. The Airport management software (I use Freebase on my Windows XP laptop) doesn't see the Airport on its scanning, but if I manually tell it what IP address the Airport is on I can connect fine.

So I spend several days futzing around with settings on the Airport, making sure that I can connect fine with a wired connection on the same port on my hub, etc. Then I bring the Airport in to work and let my colleague Mark McNair take a look at it - works fine for him - and he even updates my firmware for me (there's apparently no way to update an Airport from a Windows machine).

So I bring it back home, plug it in - same deal. Finally I have the bright idea to disable the Internet Connection Firewall on the wireless connection - voila - everything works fine.

I have no idea why it worked ok for months with the ICF turned on and all of a sudden stopped working. But now that I have it working again, I'm not messing with it.

I think this is the box I've been waiting for to get my mp3s to my home stereo - though I'll have to make sure it can cope with AAC files.

This has been widely reported, but it's too mind-boggling to pass up - David Sifry of Technorati (a site that tracks which weblogs are linking to which other weblogs) reports:

"One year ago, when I started Technorati on a single server in my basement, we were adding between 2,000-3,000 new weblogs each day, not counting the people who were updating sites we were already tracking. In March of this year, when we switched over to a 5 server cluster, we were keeping up with about 4,000-5,000 new weblogs each day. Right now, we're adding 8,000-9,000 new weblogs every day, not counting the 1.2 Million weblogs we already are tracking. That means that on average, a brand new weblog is created every 11 seconds. We're also seeing about 100,000 weblogs update every day as well, which means that on average, a weblog is updated every 0.86 seconds."

Wow!

My colleague Josh Larios wonders how many of those weblogs actually get updated regularly. An excellent question, and probably someone's upcoming PhD thesis topic.

John Babich has posted a list of his Top Ten Holiday Blues Records, for those looking forward to a blue, blue Christmas.

Back in the early '80s when I was playing in Eddie and the Atlantics we had a Christmas R&B show that we toured around the Northwest, so many of these songs bring back fond (at least in retrospect) memories. I particularly remember one hungover morning schlepping all our gear up several flights of stairs to perform a set on Gary Crystal's show on Coop Radio in Vancouver. We tried to give away a date with Ray Downey, our sax player and back-door Santa, to no avail.

Wired is reporting here that Virginia Tech's Mac-based supercomputer, made up of 1,100 dual processor G5 boxes, has been rated the third fastest supercomputer in the world. Wonder if my cgi processes would run faster on it...

Betsy Devine (who I just discovered, thanks to Joho) writes here about the liberating shock of coming from Java and C++ to scripting languages, in this case PHP:

"What a whole new world of freedom--it's like being able to say, "Mom, I'm bringing a friend for dinner..." Then you show up with 7 friends, a puppy, plus (Scott adds) a parakeet and a goldfish--and Mom says, "Hey, no problem, who ever runs out of spaghetti?"

John Gruber reports that BBEdit 7.1 supports sftp for secure file transfers - that's a good update by the Bare Bones folks!

Maybe it's just the end of a long week, but this made me laugh - thanks to Bob Jamieson.

Today I used a Creative Commons license for the first time for my own work. I received a request for the presentation slides from my Educause 2003 presentation and the message requesting them suggested that I include a copyright statement on the materials. Instead I decided to use a CC license called Attribution-ShareAlike, which allows anyone to make use of my work, as long as they attribute it to me and share any derivative work under the same terms.

The process of choosing which license to use was really easy, using the process set up on the CC web site, and they provided a complete code snippet to use on my web page to notify people of the license and link to the terms.

This is great stuff, and the folks at Creative Commons are doing really terrific work that points the way into usable forms of protecting intellectual property while allowing for sane reuse of the IP.

Here's Dylan Evans, a psychologist, arguing that in the future it will be necessary for average folks to actually understand something about how computers work and a bit of coding, and that operating systems that make it visible and transparent are to be preferred to those that try to hide the guts behind a facade of user-friendliness. Interesting.

Thanks to Tim Bray for pointing this one out.

Back home again, trying desperately to catch up.

Our session on the Chandler project on Friday morning went very well - standing room only (that's what happens when you have luminaries like Mitch Kapor on your panel) and great feedback from folks in the room. I remain really excited about the potential for both the Chandler software and the collaboration between higher ed computing types and the developers at OSAF.

My slides from the talk are available on the web at http://staff.washington.edu/oren/presentations/educause2003/ . There's an html version of the outline as well as powerpoint slides.

I heard the closing session with Richard Clark was good, but I missed it as I was still talking to folks about Chandler.

I realized that I forgot to mention the interesting sesion I went to on Wednesday about how MIT and Stanford have been jointly benchmarking their help desk operations. I'll post a link to their site when I retrieve it from my notes - I think I'll want to use many of the measurement techniques here at the UW.

I closed out the week with a shrimp po' boy and an Abita amber at Ralph Brennan's in Downtown Disney along with my colleague Tom Lewis - the ersatz New Orleans atmosphere made me long for a real trip to the Big Easy...ahh well, maybe next year.

And kudos to the hard working Educause staff are definitely deserved - the largest Educause conference yet (6400 attendees!) went off very well, the program was terrific and rich, and the events, especially the night at California Adventure, were great!

Well, I've guess I've arrived if someone is posting spam as comments to my weblog...sigh.

Interestingly, all the entries are to a single (old) posting in the blog, which leads me to believe that url just got added to some automated process somewhere. I can't believe anybody actually finds this an effective marketing strategy.

At the file sharing lovefest yesterday, Graham Spannier, President of Penn State University, announced that Penn State has reached an agreement with Roxio to offer unlimited access to the new Napster service to students on campus. The story is here. It will be interesting to see how students like it, and how other campuses either do or don't follow suit. It will certainly be a good workout of Napster's (and Microsoft's) DRM technology.

Educause threw a terrific party last night at Disney's California Adventure theme park. That California Screamin' is one hell of a roller coaster ride, and the Soaring Over California is a terrific demonstration of just how good immersive virtual reality can be - there is none of that kind of cognitive dissonance that makes me get queasy on the older motion platform rides like Star Tours at Disneyland.

Fun was had by all, as near as I could see.

Yesterday's general session on P2P file sharing issues in higher ed, which included luminaries Jack Valenti from the MPAA, Cary Sherman from the RIAA, Graham Spanier from Penn State, and Charles Phelps from University of Rocherster, produced none of the anticipated fireworks.

The panelists, along with moderator Mara Liasson from NPR, did a good job of congratulating each other on the work they've accomplished in educating students about the illegality of sharing copyrighted works and how they are open to moving to services which legally allow online distribution of music and film.

There was basically no discussion of any of the difficult issues such as price points, availability, digital rights management, fair use, or first sale doctrine.

While it's good to know that at least the entertainment industry won't hold higher ed institutions to blame for being uncooperative, it would've been nice to see more frank discussion of the places where we differ. Ah well...another time, perhaps.

I imagine that the video from the event will be made public at this site, though it's not there as I write this:

http://www.educause.edu/conference/annual/2003/resources.asp

It was great yesterday to see Ken Klingenstein, leader of the Internet 2 Middleware Initiative, get the Educause award for Leadership in Information Technologies. Ken has been a driving force in the profession for many years, and for the last several has been at the center of important and hopefully far-reaching advances in distributed middleware architectures.

Ken's award-acceptance speech was titled "Noteworthy Failures, Ulysses, and the Heart of Rock and Roll" and was full of characteristically humorous, warm, and penetrating insights as he looked back on his career.

Way to go, Ken!

There was a good keynote speech yesterday from Shirley Jackson, President of Rensselaer Polytechnic. Dr. Jackson talked about the challenges facing higher education in the new century, and how information technology both is a forcing function of change and offers enabling tools for coping with changes. She then went on to talk about how progress in coping with the issues facing us are being undermined by the decreasing interest of students in the US in science and technology, and how we have to work hard to correct that.

While there was not a lot of new information here, it was really great to see a higher education leader with some real vision of the issues and the place of IT in helping our institutions - it made me feel the lack of leadership at my own institution, which has been without a President for over a year now.

Went to a reception hosted by Apple last night, where they tried hard to make the point that the iTunes Music Store is a good way for campuses to fight illegal file sharing, and how they're looking for ideas from campuses to pilot joint programs with them. I dunno... as much as I love the iTunes store (and I am a very regular customer of it), I don't see any reason a higher education institution should help Apple (or any single commercial entity) do their marketing for them - legal music downloads will probably supercede illegal downloads when they have as wide a selection and better quality at the right price point. I look forward to that day.

Yesterday the University of Michigan, MIT, Universit of Indiana, Stanford, and the uPortal consortium announced a major new effort called Sakai.

The participating institutions intend to integrate and synchronize the software and specifications that have grown out of the Open Knowledge Initiative, the CHEF project at Michigan, and the uPortal framework. The intent is to coalesce this effort around the newly ratified JSR168 spec for Java portlets.

This could turn out to be a major advance for open source course tools, and one which we'll be following closely. Our major problem at the moment is how to take advantage of this effort in our current, Perl-based environment for the UW's Catalyst tools - and generalizing from this, how to open up this effort to developers working in non-J2EE environments.

There's a lot of buzz here about a company called Xythos, which makes technology for web-based file management. Their solution has been implemented at U Texas in Austin, and lots of other schools are working on implementing it, including the University of Wisconsin at Madison.

The Xythos technology allows people to manage files on a remote filestore via either a web-based interface or through the WebDAV protocol. It supports WebDAV over SSL, allowing for secure connections from desktops. They store the files in a proprietary manner, with all of the metadata about the file stored in a database. That allows them to add features not found in traditional filesystems, such as versioning, access control lists, and recognizing when the same file is being stored multiple times (by using an MD5 hash of files that gets stored as a key in the database).

While I understand the attraction of those additional features, it means that you can't interact with these files except through the Xythos interface. I do wonder if a simple web-based and WebDAV interface to traditional Unix file systems isn't more along the lines of what we're looking for at the UW.

The sun has come back out in Anaheim and the fires are out, so things are looking up !

Yesterday morning I attended a not very inspiring session on IT governance in higher education. It's too bad, as it's an important topic and one on which we've been spending a lot of time at our institution, setting up new committees and other vehicles for communicating with the folks who use the services we provide.

In the afternoon I heard a great talk by Tracy Mitrano from Cornell on the state of federal laws affecting IT security and privacy at higher ed institutions. The talk was based on a paper she wrote titled "Civil Privacy and National Security Legislation: A Three-Dimensional View". The paper will appear in the November/December 2003 issue of Educause Review. I recommend that everyone dealing with computing policy issues in the US read this paper - it's scary as hell.

In her talk, Tracy noted that over 13,000 warrants have been issued under the Forign Intelligence Surveillance Act of 1978, 1,000 of them in the year 2000 alone. The only public records of these warrants are the annual reports that that Vice President issues to Congress - no other details are available. Yikes.

Another great obsesrvation by Tim Bray on the falsity of claims that the web browser can't provide a rich user interface experience.

"Not once has an end-user type person written in saying they wished they could have a richer interface like the kind they used to have in compiled desktop apps. "

I'm down here in Anaheim for the Educause 2003 conference. I'll be appearing on a panel on Friday with Mitch Kapor, Jack McCredie, Ira Fuchs, and Chao Lam - we'll be talking about higher education's participation in the Chandler project.

I'll try and blog events and conversations from the conference as they take place, though it may be spotty because Educause has apparently made a conscious decision to not have wireless connectivity in the meeting rooms in the Anaheim convention center - some of the speakers think it's too distracting. A pox on them, I say, if they can't manage to be more interesting than the net :)