Monday, January 28, 2008

As if You Care: Catch-Up Edition!

Get it? Ketch-up? Photo by user Automania on flickr.
Being out of town for 5 days means coming home to about 400 items sitting in Google Reader. Here are the highlights:

1) O'Reilly Radar: Books that Make You Dumb (or smart?) What's interesting here? You mean besides three of my favorite things (data mining, books, and information visualization)? How about the Book of Mormon ranking 30th on the list of smarter books? Just sayin'. Although the obvious counterpoint is to say that if we wanted to be REALLY smart we'd be reading Lolita instead.

2) While we're on the subject of information visualization (and religion), let's hit Strange Maps: Regionalism and Religiosity (not a Jane Austen novel, actually). So here's a map of what the largest congregations are in all the counties of the U.S. It's just the kind of thing Rob and I might spend an hour discussing at work, if only I still worked with Rob. So, instead, we all get to discuss it here.

3) Who's job is it to make me a Homebrew Snow Machine? Somebody get on that. Better yet, build it at my sister's house in Massachusetts, where there's a good hill and colder temperatures (and 3 boys to help with the building).

4) I don't have HBO, and don't watch The Wire, but I know its about criminals in Baltimore. That doesn't stop me from being inexplicably interested in the guest post series "What Do Real Thugs Think of The Wire?" on the Freakonomics blog: Part One Part Two Part Three. WOW. For the faint of heart, I should say there is censored strong language, but if you went to junior high you'll know what the curse words are. I'm probably going to have to get this guy's book. I don't know if I'll ever watch The Wire, but it's just interesting to see this particular demographic react to their own portrayal on TV.

Sunday, January 13, 2008

The 10-minute album cover meme

I first saw it here, but decided to do it after seeing it here. The rules:


The first article title on the page is the name of your band.


The last four words of the very last quote is the title of your album.


The third picture, no matter what it is, will be your album cover.

Let's see 'em.

Sunday, January 06, 2008

As if you care about... Daveipedia

I have this idea; it's not that great but it's persistent and I'm probably just going to have to do it to make it go away.

Daveipedia! my own collected knowledge, on a wiki, because for whatever reason I can't seem to get enough of them. So, if you're reading this, I probably know you and you probably deserve an entry on Daveipedia.

But! Lots of stuff is already on Wikipedia, and I shouldn't have to duplicate that stuff by hand. I could just make a local copy of everything on wikipedia - excluding the history pages, it wouldn't even break 10GB, and then I could just modify and add to it. That's easy, and totally unappealing. That's just another wikipedia clone with a few additions and modifications. Daveipedia needs to represent what I've come in contact with, not all the things everyone has come in contact with. It's only going to exist locally, on my home machine, for now, to keep from prying eyes and because I wouldn't be able to get to it from work anyway. But it's a place for me to play, and store things I might otherwise forget.

So here's how I want it to work:

- It's a normal mediawiki install, plus my favorite extensions like semantic mediawiki and so forth. Starts out empty; no preloading info from Wikipedia - that's cheating.

- Normally, on mediawiki installations (including Wikipedia), if you make a link to something that doesn't have a page on the wiki, it shows up as a red link; clicking on a red link leads you to a new edit window for that page, so you can make and save it. I want a script or a little mediawiki bot that will find red links on Daveipedia and look to see if a Wikipedia page exists by that name. If it does, great! Go get the page, and load it into Daveipedia. I referred to it; therefore I have come in contact with it. Therefore, it gets a page.

-Obviously, I should only let my red link bot do this to pages I've viewed; otherwise it'll cascade when it gets a page from Wikipedia with new links - it'd go get those, and then their links, and so on forever until I'd have a Wikipedia clone again. No, it should wait until I've visited the local copy of whatever page, then pre-fetch only those pages directly linked to on that page. It will grow organically as I use it, to encompass what I'm putting on there, plus what I'm viewing on there, plus one more link layer out, silently waiting for me to read those pages, and when I do, automatically growing one more layer out from them. I think I know enough to do this, and make it work the way I want.

You knew I was crazy, but you had no idea it went this far, did you? DID YOU?!?

Also, here's 2 quick reviews of movies I've seen in the last 24 hours:
1) I saw Waitress, and I thought it blew (sorry, JPB!) Stupid people making stupid choices and being stupid to each other for an hour and a half, then 15 minutes of it all suddenly getting fixed in the most unbelievable way possible. Unless you're a huge Keri Russell fan, don't waste your time. It did make me hungry for pie, though. 1 star out of 5.

2) I saw No Country for Old Men. This movie is NOT for the squeamish - lots of people get killed, some of them very graphically. But it was excellent - top notch Coen Brothers of old, like Blood Simple or Miller's Crossing. Super-creepy villain. 4 stars out of 5, but not recommended for most of you.

If all goes well tomorrow, I will watch A Midnight Clear, which I saw as a teenager, and remember as being really good. We'll see. High hopes, though.