Monday November 17, 2003
Pragmatic Version Control
I emailed the Pragmatic Programmers
(Dave Thomas and Andy Hunt) to see if they would send copies of the Pragmatic Starter Kit
to our chicago.pm
group for review. Next day I got an email from Dave saying that the first two volumes are on the way. I just got Pragmatic Version Control, and Pragmatic Unit Testing last Friday, and I quickly devoured the first. In fact, the review
is finished as well.
Wednesday November 12, 2003
gives you a quick way to send your House Rep a fax or email letting them know that you want to cast your votes using an open
system, not a blackbox that could be riddled with bugs and security holes. On the EFF site there is more information about the issue. Included is a link to a paper
by some folks at Johns Hopkins and Rice who review the Diebold
system that was used in Georgia state wide elections in 2002, and is getting more and more business since the last Presidential election. Diebold's source code was obtained
and posted to the net by a hacker who exploited a vulnerability in a webserver (any guesses) and gained access to their corporate "intranet".
Our analysis shows that this voting system is far below even the most minimal security standards applicable in other contexts. We highlight several issues including unauthorized priveledge escalation, incorrect use of cryptography, vulnerabilities to network threats, and poor software development processes. For example, common voters, without any insider privileges, can cast unlimited votes without being detected by any mechanisms within the voting terminal.
I like the idea of voting electronically...but if we're going to do it, let's do it right.
Tuesday November 11, 2003
I've been working
off and on with Jane Jacobs at Queens Public Library
in New York City to help them with some data munging. QPL has lots of patrons who speak Russian, and a large collection of Russian materials. Unfortunately their catalog data has transliterated titles, authors, subjects etc. This means the original Cyrillic has been romanized so that cards (and more recently records in a database) can be sorted and displayed. Now that there's Unicode, these Cyrillic characters can be sorted and displayed pretty easily. So I used Perl with MARC::Record
to detransliterate their data, which was then imported back into their online catalog. Now patrons who read Russian can search for and read bibliographic records for Russian titles in Russian
. It's kind of neat to see your name in the news, even if it is just a tiny blip in the days events. I just wish they mentioned Perl in the press release
Wednesday November 05, 2003
Last night's chicago.pm
meeting went very well. Lots of new faces, and people with interesting backgrounds...and I learned a thing or two about command line switches that I had forgotten after reading Data Munging with Perl
. Andy's presentation should be available soon here
. I spent some time after the meeting encouraging some people to volunteer to talk at a future meeting: in particular I'd really like to hear hachi
talk about POE
some...it sounds like he's involved with some interesting work with it at pobox.com
. Also, there was a guy (I'm terrible with names) from Northwestern
who is doing some interesting genetics work with BioPerl
and some slime. Steven Lembark volunteered to talk about debugging Perl next month, which should be interesting as well. Jason, Steven and I had some beers at a nearby microbrewery. They had a nice India Pale Ale, which helped since Steven got talking about the differences between swapping/paging until my brain hurt
:) All good fun. Having the new space with the projector (and pool table) really helped alot. I think this marks the beginning of good things to come.
Friday October 24, 2003
I got an email from Patrick Hostenbach at Los Alamos National Laboratory asking for me to tweak Net::OAI::Harvester a bit. His OAI-PMH repository is particularly strict and it turns out N::O::H was ocassionally passing a name/value pair that caused problems.
I'm really happy to hear LANL is using the module since Herbert Van de Sompel at LANL was one of the chief architects of the OAI-PMH. It was devised partly as a way for people to harvest metadata from the arXiv eprint archive.
Wednesday October 22, 2003
I've been working with Martin Emmerich at the Max Planck Society
to get the Net::OAI::Harvester
docs up to snuff. We've even turned up a few not so minor bugs, and added some missing unit tests. It turns out their eDoc
server has an OAI-PMH interface. eDoc is "an easy to use electronic platform to document, share, manage, archive, disseminate, publish scientific information" of the Max Planck society. Thanks CPAN.
Tuesday September 16, 2003
I didn't manage to make it to last night's chicago.pm meeting. But I did happen to visit Michael Graham's website. Michael was visiting from toronto.pm. I learned about the Album-a-Day
project, which is kind of a neat idea. If you are looking to listen to some random/free music it's worth a visit. I didn't see it, but Spin did a writeup
on the project. I also happen to use Michael's Palm::Progect, which works pretty well for me. Thanks Michael.
Saturday September 13, 2003
On a tip from autrijus I took a look at psh again (perl shell). Last time I looked it didn't catch my fancy...but my how things have changed. The full power of a familiar shell, (cmd line completion etc) plus all the power of Perl at your fingertips. I feel like psh is going to make the amount of one off programs that are scattered all over the file system rapidly diminish.
Just as an example, I wanted to look at the comments inside some image files I had.
% use Image::Magick;
% $i = Image::Magick->new();
% forfiles *.jpg ( $i->read( $_ ); print $i->get( 'Comment' ),"\n"; }
That's all from the command line, and this is just the tip of the iceberg! psh is the most fun I've had in Perl for a while.
Friday August 29, 2003
eff & sco
Thanks EFF for making this
Thursday August 28, 2003
mjd on patterns
I just discovered mjd's piece "Design Patterns" Aren't
....I need to really read the GoF book before I can make an educated comment, but it seems to me that MJD is right on about the larger meaning of Alexander's work.