Author Archives: DMZ

The Taking of Pelham One Two Three

I noticed today that one of my favorite movies is available on Netflix’s “Watch Instantly” (which is a DRM-ridden horror show, but anyway), and I wanted to recommend it if you’re on Netflix (and have Internet Explorer and can use Watch Instantly) or are willing to go out and rent it.

It’s a 1974 crime flick about hijacking a subway car, based on a book. It’s really sweet: it’s tightly plotted, it’s funny, it’s got tension and plot twists (which are not dumb or overly cute), and it has Walter Matthau as a worn police lieutenant and includes what I’d argue is one of the single best lines in any crime drama.

The interesting thing the movie reminds me of is something Pauline Kael used to write about: the value of workmanship. She repeatedly wrote in favor of movies being made simply, without needless directorial embellishment, and Pelham One Two Three is a great example of how it can work. Joseph Sargent’s work up to that point was undistinguished: if you look him up on IMDB, he’s got a ton of credits but there’s nothing that hints his next film will be a four-star masterpiece.

But here, it works: he lets the script and the actors do the lifting, and the story plays out simply. Sometimes, it pays to tell a story with a maximum of economy, which is what I try to remind myself of when I’m writing fiction.

Of course, the movie is being remade by Tony Scott with a cast of stars, due out in 2009. I wonder, though — if you released both versions at the same time, putting out the 1974 version as a period piece, or an “homage to 70s cop movies” or something, I wonder if the older one could compete.

Anyway, Derek says “check it out” if you haven’t already.

This is why people hate eBay

Look up something popular, say… iPods. Go the actual category for MP3 Players > iPod. Even pick a particular model.

I did this with the nano, as I was considering replacing my venerable one. Searched for items with the “buy it now” option, sorted by price+shipping, lowest. 1,349 items.

The first two pages are all miscategorized items. And of course, ebay makes it hard for you to report those. You might get to some broken ones if you’re lucky. It’s all skins, batteries, adapters — everything that shouldn’t be there. But they can’t be bothered to police even the top-tier categories they spend money advertising on Google with. It’s crazy. Lord forbid people be forced to put their ipod cases in “Consumer Electronics > MP3 Accessories > Cases > Apple iPod” where I might go to shop for those.

It’s like if I went to Target and went to the housewares section, to the toaster aisle, and instead of toasters, there were 900 people trying to sell me bread, extension cords, power adapters, non-conductive forks for sticking into the toaster…. faaaaaaaaaaaaahk, man, it’s just ridiculous.

And now I can’t shop on Amazon, because I’m mad about the whole print-on-demand issue… drives me nuts. Is it really that much easier and cheaper to run a maddening cesspool that drives people like me away that it makes it worthwhile?

The upside of sickness

After over a week of pretty much coughing continuously, I have to say my abs haven’t looked this good in years. I’m two into my six pack, and I’m betting a couple more days and I’ll be showing another two. I should write some kind of best-selling fitness book. “The Sickness Solution: Your Shortcut to Losing Weight and Looking Great”.

Sure, I might have pneumonia, but isn’t that worth it?

Programming fiction

I haven’t coded seriously in almost five years. Actually, seven, and that was writing some testing tools to do log scanning in Perl. I’ve tinkered a little since then– when I was unemployed, I toyed with using Ruby to prototype a startup idea, for instance.

I’ve been writing an interactive fiction game (think Zork) relating to baseball, in which the player is a general manager of a baseball team, and hilarity ensues.

The funny thing is I’m having the exact same feeling I used to have when I was coding, where I don’t entirely know what I’m doing, so I’ll go in and write out the desired behavior, and then in implementing it resort to increasingly ugly hacks until I can get it to work. I can’t figure out how to get a character to only take certain objects, for instance, so for now I’ve got them accepting anything you give them. While it works, it makes me wince to play through when I can give someone two redundant objects.

Or following: for a plot point I need to get someone to follow the player and then stop when they meet someone else. The only way I’ve figured out how to do that is to turn the follow rule off when they’re near the someone else, which works okay, but it’s also not a particularly good way to do it. And I’ve ended up doing a lot of making environmental behavior location-specific to the player and it just seems unsatisfying.

Part of the problem is that there’s no K&R book for Inform 7. There’s documentation and a recipe book that almost but don’t quite ever seem to tell me what I’m trying to figure out. I have a wishlist for things I want to be able to do and can’t yet.

So my process right now goes:
– pick an item off the to-do list, be it feature or bug
– re-write it
– compile (if fail, troubleshoot)
– test (if fail, troubleshoot)
– repeat

It’s slow, slow going.

The game’s 4,000 words now, which I believe would make it the longest post in USSM history if I pushed it out now, and it’s still a ways from being finished.

The really weird thing is that as much as interactive fiction’s been a niche for ages, it’s been a fairly viable one, but I cannot figure out where to get help on this stuff. I joined a list, it’s dead — though the mod approved me that day. I don’t see updates. I feel like I’m wandering around a ghost town where everyone decided to leave in June of 2007 without bothering to pack up or anything.

Even when I was programming in Fortran 77 in college, you could still find other people tasked with the same kind of work maintaining obsolete systems. But this, this is a little eerie. I can’t seriously be the only person writing a text adventure, can I?

Interestingly (to me, anyway) the only place people seem to be actively discussing this stuff at all is on Usenet, which I haven’t used seriously in many, many years. To see that it’s still living on, well… it warms my heart a little.

Now that’s customer service

Netflix doesn’t let customers sign in on the front page, which is the most absurd, ridiculous design decision they could possibly make: they have a default no-cookies-detected home page which includes fields *with which you can register* but if you’re a returning customer there’s absolutely no way to sign in there. You have to click “login” each time, get a new page to render, then submit.

I can’t think of another large-scale ecommerce site that makes this so difficult, and certainly not one that, like Netflix, you have to return to continually.

But let’s say you want to drop them a line and say “look, instead of having the login page render on a second page, even if for some reason you don’t want to immediately draw one, why not just have the “login” link create a small box where you can enter your name/password, like many other sites do?”

Nope.

“Contact Us” has no contact information.

The Help Center supposedly offers ways to contact them, but you can’t actually send them an email: there weirdly is a “Answers by Phone” box when you dead-end, but there’s never any contact information besides that. So I have to call someone, wait on hold, and then talk to someone? Why? And running a call center is expensive — it’s far easier to do something useful with email feedback.

There’s a host of stuff like this — they recently made it impossible for not discernible reason to browse some things in the same way you used to be able to

It’s amazing that Netflix somehow managed to create a help system more difficult and frustrating to use than eBay’s. Congratulations, Netflix. That must have taken some doing.

Location-based advertising

Microsoft’s thinking about targeting ads to Sync users in their car.

“We know where you are and we know where you’re headed. We could target that advertising directly to your car.”

When I worked at AT&T Wireless, I used to have an argument about this pretty frequently. They all went like this:
“We’ll be able to deliver ads to your phone based on where you are! So you’ll be next to a Starbucks and we’ll send you a coupon for $1 a latte!” (it was always Starbucks)
“Do you realize how many Starbucks locations there are in Seattle?”
“What?”
“There are dozens of Starbucks. Are you going to send me dozens of Starbucks coupons as I drive up I-5? I’d turn off my phone to avoid that. It’d be a horrible experience.”
“Um…”

You could, of course, solve this by only sending a few messages and upping the cost and the incentives for the user (“free latte at the Columbia Center Starbucks? I’m there!”) but the implementation ideas always ended up delivering horrible user experiences.

I’m not surprised that the idea hasn’t died: it’s so easy to think that there’s an impressionable customer base just out of current reach, and if only you could reach out and touch them, you could drive them to take a particular action. They just don’t ever think it through to the user’s perspective.

My presidential choice…

… will be entirely determined by whoever first promises to have the FTC shut down companies that include a “subscribe me to all your spam lists” checkbox that resets to checked on re-draw (say, if you recalculate shipping, or attempt to continue with a phone number that doesn’t quite meet their formatting expectation).

Sure, it’s petty. I don’t care.