Lonely Lion

If you haven’t seen this, you should…stat: Idiomatic Python. It explains lots of good Python style, as well as a bunch of ways to make your code more “Pythonic.” It’s a quick read, and well worth your time.

From last night’s Marketplace,

It’s hard for me not to wonder if corporations like Google feel any tug of civic responsibility. After all, their breathtaking success as news aggregators has sprung indirectly from the fruit of professional journalism.

From Andrew Carnegie on down, rich philanthropists who rose from rags have sought ways to give back to America by investing in civic endeavors vital to a healthy democracy.

Few things to me are more vital than journalism practiced according to high standards. Simply put, Google could do much more to protect this public trust: Offer support to journalism education and professional groups dedicated to truth seeking and time-honored ethical values. And assist newspapers directly, just as I think it’s time for newspapers to band together to sue to protect content.

We are experiencing a tragedy that transcends dollars, cents and jobs lost. The truth is this: If newspapers keep bleeding and dying, one day soon we may find on Google News no news at all.

I read a couple of newspapers every morning, none of which I actually pay for, other than looking at advertisements. According to Frontline, the majority of on the ground reporting is done by newspapers, most news reports just regurgitate what was first reported by newspapers. If the newspapers go away, all news goes away. It’s in Google’s own best interest to preserve the sources of one of their big draws, right?

UPDATED: Over on my facebook page, Fuzzyman pointed out that on Google News they don’t provide the full content of the article, just an excerpt. They then link back to the original article. That seems like a good way to actually help out the original content providers…so, really, the above comment, and my reaction to it…not really grounded in reality.

Last Thursday’s ChiPy meeting for July went pretty well. Brian Ray hosted. Ian Bicking demoed lxml, his current, “favorite xml library for Python.” I demoed the concurrent goodness of ipython1, and Karl led a open ended discussion on Django apps that led to a discussion of web frameworks, and a bunch of other stuff. All in all, a good meeting.

Django is two!. The framework sure has changed an awful lot in the past two years. Thanks to all involved for making a great tool. Two years ago, I would have been shocked to know that I’d be working full time on a Django based project. Very exciting stuff. For even more nostalgia, here’s the post I wrote after Adrian gave ChiPy a preview of Django in June of 2005, and here’s the follow-up in July where he got into some more detail.

Adrian Holovaty presented Django last night, in a sort of sneak preview for it’s release this week. Django clearly has a lot of features that will be welcome to Python web development. There were actual oohs and ahhs from the group during the presentation.

Neat stuff. Thanks again for your hard work Django-neers.

Last week(ish) The Washington Post published and article on micro-targeting political advertisements, essentially using data mining techniques to send super-targeted ads to households. It was a good article, and really surprising to me, essentially because I’d always assumed that political campaigns would use the same sorts of techniques that marketers were using. I didn’t know they’d be behind the curve to the point of seeing data mining practices as some sort of new thing. So, that’s naive-example number 1.

Then, the next day, I saw the author of the article had posted a blog entry about the article. I thought, “oh, that’s nice, I shall post a comment.” Fine. Done. I went back to check and see how the comments went, and it turns out the comments on WaPo blogs are rife with in-fighting bullshit. Well, duh. I should have known. Again, it turns out I’m totally naive. The open internet sure does attract some jerks.

Despite being repeatedly mistaken for a board game meetup that happened to be happening in the exact same Goose Island Location, RHS had a successful meeting last night, organized by Mr. John Melesky. As usual, no formal presentations occurred. Mostly, I just bored people with my current EVE obsession. John Melesky suggested we enter the ICFP contest next week, which seemed to be at least sort of interesting to a few people. We also spent a fair amount of time chatting up the Netflix Prize. All in all, a fun meeting.

Adrian Holovaty just released a tool for extracting templates from a sample set of HTML. What a clever idea! Rather than building a series of parsing rules for screen scraping, you can just give the tool the pages and look for differences. The differences become “holes” in the template. You can apply the template to other samples to extract the data. Super-clever.

I may give this a shot to automatically populate the Victim of Time show calendar from Myspace listings. Writing a Myspace concert screen scraper has been on my to do list for a while now. Templatemaker might make that a truly trivial task. Thanks Adrian.

Let’s be clear, I’m an Obama backer. Yet…this blows my mind.

This guy agrees:

American politicians expend untold billions of dollars on campaign commercials. Not only do these ads all appear to have been produced by the same vicious idiot, it appears that this vicious idiot has been lobotomized and then repeatedly, thoroughly concussed.

That is why as Americans we all owe a debt of aesthetic gratitude to the genius of former senator and current Democratic presidential candidate Mike Gravel, who has taken the pabulum and kitsch that is our political art and transcended it — swept it up, summarized it and broken through it into a new range of possibilities. Mike Gravel is to political advertising what Ralph Waldo Emerson is to the essay, Walt Whitman to poetry, Jackson Pollock to painting, 50 Cent to bullet wounds. He is the avant garde of the new artpolitical era.