Just another WordPress site

Category: Uncategorized

  • parseMe 20070429 Update

    Here’s another update to parseMe (back story), my little GPL’ed PHP-based RSS/Atom feed reader for mobile phones and other web-capable devices.

    • Moved to object oriented, pretty much for the “fun” of it.
    • Now passing a custom user-agent in the http query to avoid problems with sources that require it (Digg, among others).

    You can find the appropriate links below:

    Keeps me from hating my phone until I can afford to get myself a nicer mobile solution.

  • Fedora Core 7 Test 4 Notes

    I installed FC 7 Test 4 on one of my home machines, which was previously running FC6, and it pretty much all went fine. Like with all new releases, there was a definite speed improvement in most operations.

    I personally like the new live CD installer. If nothing else, simply for the fact that you get a chance to see how the OS will behave on your new machine before you install it. It’s still ironic to me to see the major distros going to that format, because I remember how most Linux users were poking fun at the first developers to use a live cd install process (that I know of), back in 1999: the now defunct LinuxPPC distribution.

    For those interested, you can access a lot of FC7 screenshots and videos at the main wiki: Fedora 7 Tour.

    Besides the obvious changes, one that puzzled me for a while since I’m not a hardcore follower up-to-date with all the details, is that all my IDE hard drives were now showing up as /dev/sd* (historically SCSI) instead of /dev/hd*. One quote I could find on the wiki about this was: “In this release, all hard disk partitions follow a /dev/sd* naming convention due to a new libata driver interface in the kernel. The Anaconda installer eases the transition for release upgrades.”

    Another puzzler was the fact that despite choosing to setup my box with a manually assigned IP address in the install process, it was still acquiring one with DHCP at boot time. You can see this in the attached screenshot below (click for a larger view): note the discrepancy between the network config panel and the address reported by the ping command in the terminal. This continued, even after rebooting the machine or just the network (“service network restart” as root), until I issued a “ifup eth0” command as root, which made it all fine from there on.


    Trying to run FC7 Test 4 on my MacBook worked fine natively from the Live CD (no install), which did not under Test 3, but I haven’t been able to boot it while virtualized in Parallels Desktop for Mac. I haven’t tried in VMWare Fusion yet.

    The last note is quite a personal one: I much preferred the default theme (icons) in Test 3 (3D) than the ones delivered in Test 4 (2D)… The new ones make me feel like I’m back in the pre-BlueCurve years. 😉

    That’s it for me, for now. This box being mostly a file/web/db server, running on older hardware (P4 1.65Ghz), I can’t really play with compiz, which would be one of the major improvements over FC6, besides running new versions of everything.

    Update: here are a few notes form a friend who recently installed it on his shiny new Mac laptop: FC7T4 on MacBook (Core 2 Duo).

  • Fedora Core 7 Test 4 Released

    FC7 test 4 (6.93) has just been announced, right on schedule, and is now available as a torrent and on (most of) the mirrors.


    Get it while it’s hot and especially before it makes it to the Slashdot, Digg, and many other high profile front pages.

  • Web 2.0 Expo and Conference

    The O’Reilly Web 2.0 Expo is just finishing up, and my co-worker and I had a blast with most of it. Working in the academic sphere, it really feels great to get first hand confirmation that we are perfectly on track with the rest of the industry with our vision of what is to come. Now, if only we were given the means to do it all as fast as we can dream it, which is always a challenge in a large public institution. But we’re working on that. 🙂

    One of the highlights of the conference for me, was to find out (confirm) how close the Digg architecture is to the one that we’ve been developing for our university for many years now: LAMP (though we use PostgreSQL), memcache, Linux, clustering, etc. They actually seem to be facing some challenges that we’ve already tamed in the last year and a half. They’re hiring, by the way.

    On the other hand, it was my first time in San Francisco (about time…), and I won’t lie: it’s even better than I imagined it would be! Truly a fantastic city, populated with wonderful people. First time, but definitely not the last one, especially given that I just don’t have the time to stay a few extra days to visit Berkeley, Oakland, San Jose, Santa Cruz, Cupertino, Mountain View, etc. Next time!

    On this, I’m going to go and enjoy the city before I have to leave tomorrow morning.

  • Google@Mcgill

    Yes! We’ve done it. McGill is now using a Google Search Appliance as its main search engine backend, which is the main reason I have been so busy in the last while.

    Despite early hardware issues and a few bugs I faced in the caching engine and XML APIs (most of which have been or are being addressed by the Google Enterprise team), I have to admit that it’s been one of the most motivating and enjoyable projects I have handled at McGill.

    You can try it out for yourself on our main search page.

    We also enabled other areas, such as our advanced course search, and classified search.

    This is of course only the tip of the iceberg, since the architecture is quasi-infinitely extensible through the feeds and OneBox concepts (both of which we already use). And as usual, I already have a head full of ideas on how to further leverage the enormous amount of digital content on campus.

    Fun times ahead!

  • What Happens When…

    What happens when you’re enjoying your work a bit too much? Well, you end up working some more in the evening instead of spending some quality time with your… blog. 😉

    I’ll get more chatty again when I’m done with the projects I’m currently handling.

  • 340,282,366,920,938,463,463,374,607,431,768,211,456

    Ars Technica has a great article on IPv6 (via Slashdot):

    As of January 1, 2007, 2.4 billion of those [IPv4 addresses] were in (some kind of) use. 1.3 billion were still available and about 170 million new addresses are given out each year. So at this rate, 7.5 years from now, we’ll be clean out of IP addresses; faster if the number of addresses used per year goes up. Are you ready for IPv6?

    If you’re not in a geek reading mood, this post’s title expresses the number of addresses IPv6 will allow for. C’mon, try and pronounce it! 🙂

  • Oops, the jig's up

    From the source article, on ABC News:

    […] spring breakers, here’s a thought: Before going online to post those pictures of you and your friends dancing atop a table at Senor Frog’s, know that your debauchery will probably pop up on many more screens than you intended. Potential employers, school administrators and admissions officers, and vindictive exes can see them too, and decades from now, when college is a mere memory, those photos will still live on the Web.

    When the mainstream press and the audience it targets both start catching on to this, you just know that 5,345,961 blogs, photo/video sharing communities and other social networking sites just closed and/or lost their venture capital today alone. Did I just hear a pop? Nah…

  • March To Be Month of PHP Bugs

    From the source article, on SecurityFocus (via Slashdot):

    Stefan Esser is the founder of both the Hardened-PHP Project and the PHP Security Response Team (which he recently left). Federico Biancuzzi discussed with him how the PHP Security Response Team works, why he resigned from it, what features he plans to add to his own hardening patch, the interaction between Apache and PHP, the upcoming “Month of PHP bugs” initiative, and common mistakes in the design of well-known applications such as WordPress.

    Given the success of the Month of Apple Bugs project, I think it’s a fantastic idea. This said, our sys admin at work isn’t too thrilled by the prospect of having to patch our many PHP installs everyday in March… 😉

  • Action Plan for Killer Asteroids

    From the source article, on the Beeb:

    A draft UN treaty to determine what would have to be done if a giant asteroid was on a collision course with Earth is to be drawn up this year. […] The association has asked a group of scientists, lawyers, diplomats and insurance experts to draw up the recommendations.

    Sorry, it’s hard to type while I’m laughing so hard… Yeah, those are definitely the people to put on the job! All they’re missing are religious leaders, and we’re sure to have a speedy conclusion to that project. 😉