See Also: AppleHTMLOperating SystemsPerlRadioSemantic Web
The following entries were written in the Technology category:
I’m not pleased at how .Mac handled the recent disk size increase. My locally-mirrored disk increased to 10GB, causing my laptop’s drive to fill up. This was followed very shortly by Mail, Aperture and iTunes all freaking out at a completely full disk and crashing (or automatically quitting in a not-too-clean way). Boo .Mac!
I’ve finally gotten the GPS connection working with my D200, and am loving the simplicity of it all: turn on GPS, turn on camera, wait for satellite fix, take pictures, enjoy the extra geospatial metadata.
The iPhone came out yesterday, and I’m liking what I’m hearing about it.
This morning, I gave a talk at the Scripting for the Semantic Web workshop on my paper Extensible SPARQL Functions With Embedded Javascript.
I’m in Innsbruck, Austria attending the Fourth European Semantic Web Conference. I’m putting some photos of the conference on Flickr, and will be posting some notes on papers and presentations that stand out. Up first is “SPARQLeR: Extended Sparql for Semantic Association Discovery”, an extension to SPARQL for constrained path queries.
After a tricky upgrade of glibc, you may have noticed that this site was down all of yesterday, and part of today. The problem been tenuously fixed (although the more permanent solution is still unclear), but things might be a bit spotty until the underlying problem can be determined.
Basically, my brand new Nokia N93 that was cutting edge as of yesterday seems like a weak imitation of the iPhone today.
Version 1.041 of RDF::Query hit CPAN a few days ago.
I’ve been working on a bioinformatics project involving lots of suffix trees. The project involves generating suffix trees for each genome in a set, and using them to generate small profiles of the genome signatures.
The International Semantic Web Conference is currently taking place here in Athens, Georgia, and it’s been really great so far.
In my recent work on PhotoStuff, I’ve been working in Java on the Mac. PhotoStuff is full of features, but I often find the interface frustrating and confusing. This is especially true on the Mac where many of the UI elements end up being in the wrong place and looking very un-Mac-like. It’s worth spending some time to make some (mostly simple) changes to PhotoStuff to make it act a bit more Mac-like.
I’ve been putting a lot of work into pure-perl RDF classes recently, and have been frustrated by some ongoing issues.
It’s been around for a few days now, and I’ve got mixed feelings about Flickr’s new geo-tagging support. On one hand, they’ve done more for the cause of spatial photo annotations in the last few days than anyone has ever done before. On the other hand, the data still ends up being controlled by Flickr, and the simple coordinate approach to geo-tagging means that you can’t attach any other interesting data off of locations.
It bothers me that when discussing An Inconvenient Truth, WGBH DJs refer to it as being a “Power Point presentation” instead of a “Keynote presentation”.
Why is it that I used to use my Ericsson R280 to read content online quite a bit more than I ever did with my Sony Ericsson T68i, and the T68i got used quite a bit more than my current K700? In fact, I’ve all but given up on using the internet on my K700 because of how bad the interface is. It’s all I can do to type in two text fields to get directions with Google Maps.
I went to Audrey’s fantastic “Deploying Perl 6” talk last night at the Boston.pm meeting at MIT.
It made my day to hear online yesterday that my pet project, RDF::Query, was saving someone’s ass at work.
New RDF::Query (version 1.038). Available from CPAN shortly.
ESWC went really well. The Scripting workshop was quite good, making me a bit sad that I missed last years’ at which I was supposed to present a paper. The conference proper was also good, but the hallway (beach) track was far superior to most of the presented talks.
New RDF::Query (version 1.035). Available from CPAN shortly.
New RDF::Query (version 1.034). Available from CPAN shortly.
As I promised in Personal Report, the other day I sat down and wrote some code to pull together some personal data and consolidated it on my about page. Currently, the data I’ve got updating nightly is: recently watched movies, recently listened to music, and beer I’ve had recently (which is compared to my beer data from 2004).
New RDF::Query (version 1.032). Available from CPAN shortly.
For those that haven’t heard, I was recently accepted to a Ph.D. program in Computer Science starting in the fall at the University of Maryland. So starting in August, I’ll be moving to DC for roughly five years.
I try to make a lot of the information in my life public. Photos, travels, movies, music, how I organize files on my computer, and even what beer I’m drinking. But I’ve never really felt like I’m conveying this information in as interesting a way as it deserved.
New RDF::Query (version 1.031). Available from CPAN shortly.
I’ve just made public a feature on this website I’ve been playing with for several months. After discussion privacy issues at ISWC with Libby and Bijan, I added support on the photo pages for marking individual photos as private. The idea here is that if you’re in a photo, you have the option of taking it offline.
New RDF::Query (version 1.030). Available from CPAN shortly.
In preparation for Aperture, I’m in the process of upgrading my memory and disk setup on my two computers, trying to increase efficiency and make room for more photos and applications. The upgrades involve a gig of RAM for my PowerBook and two new 400GB disks for my PowerMac.
I’ve just uploaded a new version of RDF::Query (version 1.028) to my site, and it should be available from the CPAN shortly.
I’m finally back from Ireland (and England) and ISWC. There are pictures online from the conference and my weekend in London.
Discussing consumer electronics with Tim earlier today, and then consumerism and Wal-mart, finally leading to peak oil and consumption, I was startled by his reaction to the problem. Startled because despite the fact that I think about these things a lot (likely a significant portion of my life, recently), it’s all become rather muted with repetition. Tim’s reaction brought the impact of the situation flooding back to me.
The BBC ran a piece the other day on wind turbines becoming popular in residential settings, describing the potential economic and environmental benefits to generating power with a wind turbine. However, towards the end of the article, the ability to be totally self-sufficent on wind power is called into question.
I’m drooling over the announcement of Aperture. The demo videos (and quite frankly everything I’ve read about it) make Aperture look like a near-perfect post processing application for photos. It’s clear that Apple took all the serious issues photographers deal with when working in digital, and thought long and hard about how to solve them. From the stacking of rapid-fire shots, to the backup vaults, the pervasive metadata support to the beautiful user interface, Aperture looks like a winner.
You can now “Gift” any iTMS album or song to someone. It doesn’t look like you can do this with full iMixes yet, and iMixes are still always public, but I like where things are headed. We’re getting close to being able to send people full mix tapes via iTunes.
I really liked the idea of this meme from Om Mali about listing your ten favorite “small freeware and shareware applications that have helped you get the most out of your Mac.” Here’s my list.
I went to Jamba Juice yesterday (jamba jamba jamba!), and in a parking lot of thirteen spaces, there were three Priuses, plus the one I was driving in. Of course, there were four or five SUVs, but I’m just blown away at how many Priuses there are here. Even more than the dense numbers I see everywhere else. Which is very cool.
I’m back at the airport. Again. Back to LA for a week. I’ve got lots of work to do while there, and lots of other things to take care of, but I promise, I’m going to start posting more often.
I’ve just released RDF::Query 1.027, tracking some of the changes in the latest SPARQL draft.
I’m psyched to hear about Josh McAdams’ proposed BitTorrent Peer Library For Perl.
Just got an email from Sören Auer alerting the authors from SFSW that the papers, including my MT-Redland paper, are now online.
Well, that was shocking. Apple did announce plans to move to Intel processors within two years.
I’ve uploaded RDF::Query version 1.024 to CPAN. This release has a significant number of changes, including some serious bugfixes for the RDF::Core support.
I’ve just released version 1.020 of RDF::Query including preliminary support for SPARQL OPTIONAL graph patterns. It hasn’t been tested a tremendous amount, but it seems to work on the simple stuff. Also included in this release are support for the SPARQL Query Results XML Format and miscellaneous bug fixes.
I’ve now got initial support for ASK, DESCRIBE and CONSTRUCT SPARQL queries in RDF::Query.
Installed Tiger tonight, and things went surprisingly smoothly. GPGMail freaked out, and some of my extionsions were missing after the upgrade (Synergy, SSHKeychain), but mostly things are working. Haven’t tried MySQL or Redland, yet, though.
My MT-Redland paper just got accepted to SFSW at ESWC2005. Now I need to figure out how to make some changes without going over the page limit.
Spurred on by Libby’s work with cross-querying with Joseki, I’ve put up v1.014 of RDF::Query which will run queries with proper FILTERs, including custom function filters. Here’s an example of a custom function for finding images within a certain radius of a point.
I’ve put up pictures from the WCCS Station Dinner at Paragon last night, with a few additional shots of the flowers that are springing up all over Providence.
I’ve resurrected my RDF::Query project, and added initial SPARQL support (with the new Turtle syntax!).
OSNews recently ran the piece “The Problem with Global Menu Bars” in which the author criticizes the usability of global menu bars (“ala Mac”). I would disagree with almost the entire article, and find most of the criticisms to be based on a Windows-centric mindset notably lacking an understanding (or basic knowledge) of the Mac HIG.
I’ve just emerged from two weeks of writing a paper on MT-Redland for SFSW2005.
Shelly over at Burningbird has been doing some really great things with Wordform. It’s a much grander project in scope than MT-Redland, but they share some common themes of using weblogging tools and RDF to allow the creation of new semantic metadata, and I’m very impressed with the work Shelly’s doing.
I’ve just put version 0.007 of MT-Redland online. This release mostly addresses the speed issues I talked about yesterday, which have been greatly improved. Also, I’ve got initial support for SKOS categories in CVS.
With a few code cleanups, and checks to make sure the query returns the appropriate nodes, I’ve got a working version of an RDQL search returning RSS, and using it in an aggregator.
Because Google refuses to find files that are linked to in both my RDF and in my HTML rel=”alternate” headers, here’s a page with links to all the individual RDF files describing photo albums on kasei.us.
I’ve put up version 0.004 of MT-Redland, and I think it’s coming along really well. Here are some updates including screenshots of the interface to add semantic metadata, and a new search results page based on an RDQL query.
I’ve put up a new version of MT-Redland (v0.002), with a bunch of new features including hooks into the MT CMS code, allowing interface changes to the edit entry page.
In the spirit of release early, release often, I’m posting a first draft of a Movable Type ObjectDriver (storage backend) that uses Redland and mysql.
I’ve been doing terribly at updating this month. Really quite awful. So here’s a recap of the month.
I’ve just submitted a lot of changes to the Image Description FAQ over on the FOAF Wiki.
Well, PostgreSQL didn’t scale well, either. It hit a wall above 3 million rows (data from New England, California and Alaska). I’ve moved on to developing an R-Tree-based database in hopes that it will scale to the size of the entire Tiger/Line dataset.
MySQL support for spatial geometries is broken. There is no Distance() function, despite there being documentation for one. I’m trying to use PostgreSQL, but having only limited success.
With a working computer back in my life, I’ve been cleaning up some code, and making some changes to the website.
I’ve got a new computer.
My laptop is working!
Only a day after getting a new DVD drive for my powerbook, and having the ability to read CDs for the first time in two years, my laptop appears to be broken.
I never thought I’d say this, but I’m going to miss driving in LA. I still hate the driving-in-LA part, but for the past two weeks I’ve been driving a brand new Toyota Prius, and it’s an amazing thing.
Optimizing the image search continues. It’s much faster now than a few days ago (again), but still slow for interactive use. Here’s a screenshot, though, to tide you over.
I had intended to write a short bit about why you wouldn’t be able to browse pictures on this site by location for a long time, but two days of solid hacking on an good-enough solution using RDQL has gotten me so much closer than I would have thought possible just a few days ago.
There are now 8642 photos, at least partially annotated, on this site.
I’ve been putting tons of work into the photo annotation code, and am about ready to tackle the trip photos. To test the code, I’ve been adding lots of new pictures, and going back and annotating lots of old pictures.
Just a few days short of two years after getting my T68i, and about a month after losing (stolen?) it in Ireland, I am now the proud owner of a slick K700i.
Libby wrote up our (Morten, Libby, and myself) thoughts on image annotation, including the geo-spatial stuff, and put it online at the ESW Wiki
I was thinking about how nice it would be to hook it up to the IRC channels I frequent, both on the front-end and the back-end.
Nils Erik Svangård just posted to the dashboard-hackers list regarding the possibility of adding a voice recognition cluepacket generator to dashboard.
Morten, Libby, and I sat down over pints last night to hash out the image creation event semantics and markup.
Libby has posted a first cut at the issues brought up for an Image Description FAQ at yesterday’s breakout session.
The image annotation session at FOAF Galway was really good.
I’ve arrived in Galway, and am now waiting for the FOAF Workshop to begin.
RFID tags have been all over the place recently.
I talked with some of the folks on #rdfig today about my RDF mapping, and got some good suggestions and pointers.
I’ve hacked up some geo-locator code to pull locations from RDF and map the result on a world map.
I’ve been seeing a lot of the family, and doing a lot of programming.
Of all companies, why must Apple have issue with my entering Zip+4 information in a registration form? Pity.
I can’t believe there are still programmers who don’t realize that Zip+4 is a valid zipcode. The burden shouldn’t be on me to work around stupid programming.
Does iChat AV order the chat participants in a coherent fashion such that each participant sees the other participants in the same relative place?
I’m not sure how useful it would be in practice, but I like the idea of being able to infer location of people based on the RDF in my weblog, photos, and beer feeds.
You can now search for photos on this site by place name, latitude/longitude pairs, and Wordnet terms.
Almost all the photos in the pictures section now have individual pages describing who appears in the picture, the date and location at which the picture was taken, and the camera used to take the picture.
And now I’m at MIT waiting for the start of the Boston ACM meeting and Dan Sugalski’s talk “Parrot in Detail.”
So Simon and I drank yards together last night at Sunset. On the topic of beer, as of today I’m making my beer drinking habits (“problem”?) public for all to see. I’ve started generating graphs of total beer consumption, beer types, and locations of drinking. Also, the list of beers these graphs are generated from are available as a text file, or as a hacked-up, work in progress, RDF file.
I don’t like the fact that I have to proxy my requests from web browser (using XMLHttpRequest) to Wordnet because of a MIME time.
So how long will it be before we get an AirTunes equivalent for Quicktime?
I’m bitter that I just bought an Airport Extreme base station a week before Apple announced the new Airport Express.
Unplugging your computer because a new worm is terrorizing the campus network is not a viable solution.
Encouraged by a rumor heard at last night’s Boston.pm meeting that my forests and frogs made the cut, I’ve finally slogged through the whole of Apocalypse 12. Sure enough, my RFC 254 gets a full thousand words of coverage, being solved in a most excellent way.
It appears that the RFC I wrote nearly four years ago for Perl 6 has been accepted by Apocalypse 12.
I feel compelled to point out just how short-sighted Gizmodo comes off in their commentary on the rollout of broadband via powerlines.
I can’t believe there’s a food ontology that defines the class “PastaWithSpicyRedSauceCourse.”
I’m finally making use of all that RDF I put so much time into writing. All of the picture pages are now just a bit more helpful, showing who appears in the pictures, and roughly where the pictures were taken.
This weekend, I bolted an RDQL query onto RDF::Core in roughly 225 lines of code, and 25 lines of grammar. No matter how many times I look at it or use it, I’m always blown away by the power of Parse::RecDescent to implement a parser and lexer in such a perl-ishly concise manner.
Ok. Morning Edition totally just used the word “unopportune.” wtf?
The Dashboard project website is claiming that Microsoft is “biting” off them. But dashboard, whether they know it or not, is following directly in the footsteps of Bradley Rhodes’ Remembrance Agent.
I can’t figure out why google keeps putting me on the first page of really bizzare searches.
foaf:tipjar has been proposed, and I think it could work nicely with some new ontologies and media player support.
Apple’s new bluetooth update doesn’t work with my bluetooth adapter. Foiled again!
Bluetooth support gets better with version 1.5.
Rural Cambodian schools are getting online with roving motorcycles routing the packets to a satellite dish in Ban Lung.
The server I use to connect to the freenode and rhizomatic IRC networks has gone missing, and I’m going into withdrawl.
I’ve been doing some hacking on the site, and things are looking better. I’ve now got MT 2.66, the code pages are being generated from RDF/XML, the search page is styled to match the rest of the site, and search results may now be retreived in RSS.
The cartoon theory, based on how easily referenced it is, is proving tiself with Blackwatch Plaid.
Hello? Apple? Yes, I’d like to inquire as to why my Bluetooth headset does not show up as a choice of external speaker and microphone.
I haven’t decided yet whether this is mildly clever or absolute nonsense, but I recently started using a Quantum::Superpositions object as my debug flag.
I’ve got some ideas of where I’d like to see the BitTorrent protocol, client and tracker go in the future. I’d like new ways to see and use the data in torrent files, as well as ways to integrade that data into a tracker.
I have a strong distaste for the sharp-corners of wiki use, and wonder why things aren’t different.
Safari timed out while trying to download the Safari Timeout haxie. Bah.
I didn’t realize Panther would default to IPv6 when using rendevouz.
I feel obligated to state for the record just how much this CSS fix for IE annoys me.
I seem to have fixed part of the redering problem on IE6 for Windows.
I’ve put in a new title image using the LIR CSS method. Tell me what you think, and if it breaks anything.
Dear KDX, I hate you. So stop sucking, kthanks.
I pulled the drive from my ailing beige G3/AV box (fuosing), and put it into my P3 workstation (mangala), replacing the dead IBM SCSI drive.
The paper I wrote with Mike has finally been published in Computers & Geosciences as “CompSurf: An environment for exploring surface reconstruction methods on a grid.” The paper reflects work that I did along with Nick Doolittle and Trevor Agnitti in the summer of 2001.
Tim Bray recently posted thoughts on the Spam Problem, reminding me of a talk given at DefCon.
If only Microsoft weren’t so incredibly evil, I’d probably be all about them; Their research department has the resources to do some really good work. The World-Wide Media eXchange is one such project.
What would be slick is if over bluetooth you could start a “conference call” during an ongoing call such that one of the phones is connected to both a GSM call and another cell over bluetooth.
Shutup and Code has a simply fantastic idea regarding bluetooth cellphones.
Amazon’s API just added the TextStreamSearch method. I think this would also make an interesting addition to the Remembrance Agent.
Excellent. KDS is shipping me a replacement monitor for my aging Radius S-3F.
More downtime.
In a rational way, I’ve always understood and agreed with the oft-repeated idea that as a programmer, it’s really quite difficult to step back from what I know of technical issues and approach a UI problem from the view of the (uninformed) user. Yet there’s always been that non-rational part of me that wants to believe that I could at least approximate that mind-set.
As you might have noticed, there have been some server problems (*ahem*) around here at kasei.evilfunhouse.com.
The current iChat can save chat transcripts, one chat per file, to an archive folder. I think the same feature needs to be applied to the AV features.
Two hardware failures in two hours. Bah.
I installed RedHat 9 on mangala a couple of days ago, and I’m not impressed.
Seth Godin’s What Should Google Do? has some really great ideas.
Last night the power adapter for my Revision A Titanium laptop (the yoyo) died.
I want a new hotline-ish client and server that support BitTorrent.
Keyboards that make you use the shift key to type a forward slash are stupid, stupid, stupid.
Oh dear. This may be enough to get me to start using Yahoo! again on a regular basis.