Django'd
Well, it took a bit longer than expected, but I've migrated this blog from Pyblosxom to Django. More technical details after the jump.
Syndicated 2008-03-09 06:28:20 from esm.logic.net latest entries
Name: Ed Marshall
Member since: 2000-04-20 17:59:23
Last Login: 2017-02-27 09:04:49
Homepage: http://esm.logic.net/
Django'd
Well, it took a bit longer than expected, but I've migrated this blog from Pyblosxom to Django. More technical details after the jump.
Syndicated 2008-03-09 06:28:20 from esm.logic.net latest entries
Adventures in blogging
I've been using PyBlosxom for the last few years for my online presence here, and while it certainly does the job, I have a habit of getting bored with stuff that works after a while. So, I'm in the process of whipping up my own blog using Django, because hey, web frameworks are the new red. Or something. :-)
Actually, real reason is that PyBlosxom is category-centric; there's no real inherent understanding of a post having multiple categories (perhaps tagging/multi-category functionality is there in a newer version, but upgrading has always been difficult because of a few changes I've made). I found most posts I made ended up touching on multiple categories; racing and geek stuff often overlap, for example. So, lacking native tagging of articles, and not wanting to mess with some of the tag add-ons I've seen for PyBlosxom, I figured I'd just write my own.
What's interesting about this move is that the most recent framework I have experience with is Rails, so it's been interesting to see a Pythonista spin on the idea. What's funny is that, from a functionality perspective, they're really quite similar. Scaffolding, database interaction, view and administration generation are all disquietingly similar. Except, you know, that Ruby vs. Python thing. ;-)
As expected, the basic blog application was a snap. I've been spending most of my time on "fit and finish"; putting together template tags for displaying the sidebar calendar, working out a quickie scheme for importing all the old data, and adapting my current "static" content to the flatpages contrib framework. I'm also trying to make sure I don't break the old URLs too badly.
Once I have most of what PyBlosxom does for me today implemented, I'll probably just cut over at that point. Afterward, I'll concentrate more on some of the specialty stuff I'd like to pull in: timeslips, logs from the cars, dive logs, etc. Fun stuff. :-) And at some point, I'll throw the whole project into SVN so others can poke at it. (I might have to post up my calendar template tag on Django Snippets at the very least, since the only other one up there uses datetime instead of the new calendar module.)
Syndicated 2008-02-29 11:53:29 from esm.logic.net latest entries
Adventures in blogging
I've been using PyBlosxom for the last few years for my online presence here, and while it certainly does the job, I have a habit of getting bored with stuff that works after a while. So, I'm in the process of whipping up my own blog using Django, because hey, web frameworks are the new red. Or something.
Actually, real reason is that PyBlosxom is category-centric; there's no real inherent understanding of a post having multiple categories (perhaps tagging/multi-category functionality is there in a newer version, but upgrading has always been difficult because of a few changes I've made). I found most posts I made ended up touching on multiple categories; racing and geek stuff often overlap, for example. So, lacking native tagging of articles, and not wanting to mess with some of the tag add-ons I've seen for PyBlosxom, I figured I'd just write my own.
What's interesting about this move is that the most recent framework I have experience with is Rails, so it's been interesting to see a Pythonista spin on the idea. What's funny is that, from a functionality perspective, they're really quite similar. Scaffolding, database interaction, view and administration generation are all disquietingly similar. Except, you know, that Ruby vs. Python thing.
As expected, the basic blog application was a snap. I've been spending most of my time on "fit and finish"; putting together template tags for displaying the sidebar calendar, working out a quickie scheme for importing all the old data, and adapting my current "static" content to the flatpages contrib framework. I'm also trying to make sure I don't break the old URLs too badly.
Once I have most of what PyBlosxom does for me today implemented, I'll probably just cut over at that point. Afterward, I'll concentrate more on some of the specialty stuff I'd like to pull in: timeslips, logs from the cars, dive logs, etc. Fun stuff. And at some point, I'll throw the whole project into SVN so others can poke at it. (I might have to post up my calendar template tag on Django Snippets at the very least, since the only other one up there uses datetime instead of the new calendar module.)
Back to the source
Following on the heels of that great old-school DEC advertisement, I get another blast from my non-academic past at University: an interview with Richard Bartle, one of the creators of the original MUD that ran at Essex University back in the late 1970s and early 1980s. MUDs are the majority of the reason I'm doing what I do today (after a fashion), and were my first introduction to social computing and network programming. What's really interesting to me is that he's saying the same things that any of us who were involved in building these text-based online delusions in the 80s and 90s were constantly harping on about, and it all still applies even though the graphics have gotten better. (In fact, he makes the excellent point that with the improvement in visual stimulus, the ability to interact with the environment has actually gotten worse, because of the nature of ad-hoc natural language.)
It was a good read; I haven't played with these things in long time (aside from a brief flirtation with creating a Python-based runtime-mutable object environment similar to the facilities that LP-style MUDs had), so it was a nice walk down memory lane, and a reminder that there really isn't much new under the sun, just a prettier interface on it.
Quick initrd note
This is mainly a note for myself, but someone else might find it handy. Working with modern initrds requires monkeying with cpio, which I can never remember the correct command line parameters for off the top of my head. So, here's a quick workflow (borrowed from Chip Shabazian) for manipulating an initrd image.
mkdir initrd cd initrd gzip -dc path/to/unzipped/initrd.img | cpio -id # Make the changes you need find . | cpio -c -o | gzip -9 > /path/to/new/initrd.img
logic certified others as follows:
Others have certified logic as follows:
[ Certification disabled because you're not logged in. ]
New HTML Parser: The long-awaited libxml2 based HTML parser code is live. It needs further work but already handles most markup better than the original parser.
Keep up with the latest Advogato features by reading the Advogato status blog.
If you're a C programmer with some spare time, take a look at the mod_virgule project page and help us with one of the tasks on the ToDo list!