Or, Broken Hearts Club: A Romantic Comedy, The (2000). The beginning was interesting enough for me to decide to stay up past midnight (I feel so old!) to watch it.

It's been a while since I've been sick enough to contemplate not going to work. I've got two projects I'd rather not lose time on. I seem to slow down incredibly when taking pseudoephidrine, it seems - I'm having more trouble than usual trying to remember the words for various concepts; like the word for when something just seems so different, almost unreal- I suppose some people would say it's like being on acid. But I know there's a word for it, damnit.

My project at work (replacing an expensive and slow archiving system with a fast alternative using Open Source software) is nearing the end of its first phase. I initially exported samples from the archiving system using its MacOS client, and had only one slight surprise - the Mac could handle files with '/' in them. But, unzip on the FreeBSD side replaced the character, and I was happy with one minor hack to fix that up. But Windows...

Foreign readers should note that in South Africa Sentech is currently the only available alternative to the poor service and heady pricing of the still-monopoly telco, Telkom. But they seem to be as ept as their competitor on providing information.

I really like it when Python makes it easy for me to do something - like change my catalog functionality to take a file-like object rather than a string and it'll take care of it (iteratively). Now I can hand the catalog a large local file without worrying about memory (all that much), and I can also hand it a urllib.urlopen'd file. I've indexed about 25 megabytes of content via the web, and that's currently taking 25 megabytes to store (in a ZODB FileStorage). That's not particularly good, even if I'm currently indexing stop/common words. The search does take less than a second though, and that's the important part for now. It's designed for web log entries, after all. But I'll look into it. Also need to involve mxTidy (if available) to take care of files that seem to hang my StrippingParser...

This time, Jon Udell's using Twisted along with mxTidy and XPath to rewrite web pages in a Personal Web Proxy, in his Proxy Power article. I'd actually just been thinking of using my nope catalog in a personal proxy to remember what I've been reading. Add some time-related criteria to the search, and I can look for all content pertaining to Subversion I've looked at in the past week, for example. If I ran the proxy on a network-central, I'd have a total view of my web usage. It'd also be a quick way to find out how well my catalog scales in ZODB...

The past long weekends and public holidays has allowed me to start writing some code again, and nope (Neil's Object Publishing Experiment) is the codename for what I've been playing with. It uses the Twisted/ZODB integration from The Shuttleworth Foundation's SchoolTool (headed by Zope developer Steve Alexander) to provide a really simple object publishing environment, stealing many ideas/concepts from Zope 3.

They grow up so quickly. Or something. I've mistakenly thought Barendjie was over 21 for years, so I can't say I feel that. Anyway, Jaco threw a massive 21st supper at Banana Jam in Cape Town, with about 40 people.

Alex Martelli has nice things to say about Twisted in this Open Enterprise Trends interview.

It only took Barry a few minutes to find some bugs in the previous release of sisynala, after I wrote some quick start instructions. The 0.9.18 release has this quick start guide, and fixed Barry's bugs. I've decided I might as well place it on PyPI, more users should help shake out a few bugs.