Linkstew

It's somehow reassuring that after a year it looks like Amtrak's made even less progress on their scheduling algorithms than I have in blogging (where last year's post about this is still on the front page):

That's right: the default selection it's now offering is a worse schedule that costs more than the train I actually want (714).

I'll make you a deal, Amtrak: if you fix this in 2009, I won't blog at all. ;-p

Really, Amtrak? How did you decide to pick the trip that requires a transfer, takes significantly longer, ultimately transfers to the train I wanted anyway, and is the same price as the default option? And don't even get me started on your session handling.

hackdiary: Using Wikipedia and the Yahoo API to give structure to flat lists - Not only is this a neat hack, but it also pointed me at a couple of things I didn't know about Yahoo!'s API.

Sure, there are lots of ways you can donate to the hurricane relief effort, but here's a reminder that if you're a Dreamhost customer, they'll match your donation if you make it through their site. Read your latest dreamhost newsletter for details.

Heck, it might even be worth it just to sign up for some quality web hosting just so you can double the effectiveness of your donation. ;-)

Mapping Google

A pretty good examination of the way Google Maps' interface works.

Quirksmode.org

This site has been indispensable in the heavy JavaScript work I've sadly found myself doing in the last week. I can't recommend it highly enough.

Microsoft won't go soft on Mike Rowe

Man, this is almost unbelievably absurd. Funny, but still absurd.

Today was a long hot day of trying to work around HTML's shortcomings using JavaScript, only to end up hating JavaScript even more and fixating on how useless HTML's form tag is.

"The form tag!?" you might ask.

Granted, there's an awful lot wrong with HTML, but as a web developer, the form tag (and the input tag) are the only things I really care about at the end of the day -- and unfortunately, the form tag sucks.

So, if I could change one thing about HTML, I'd move the "action" attribute from being an attribute of the form tag to being an attribute of a submit input tag. Then, it wouldn't matter that HTML can't deal with nested form tags, because I could just submit to different actions depending on which button was pressed.

When Safari came out back in January, one of the features they touted was its "dynamic bookmark interface with a familiar single-window iTunes-like interface." Actually, here's exactly what the Safari site says (emphasis mine):

Many people don't even bother organizing their bookmarks because of other browsers' confusing, complicated interfaces. In Safari's Bookmarks Library, you'll find the familiar, single-window interface like iTunes, which lets you edit bookmark names and addresses in place as though you were renaming an icon on your desktop. You can create any number of folders in your library, and keep them in the bookmarks bar or menu, like the preinstalled news folder. ...

The problem is, Safari's bookmark features don't deliver on the promised "iTunes-like interface." The only resemblance they have to iTunes is the folder column and the contents column. Let's take the "iTunes-like interface" idea and run with it, and see what kind of bookmark management system we'd end up with:

The first thing we'd have to introduce would be the "Bookmark Library," which would contain the master copy of every item in your bookmarks. Each bookmark folder would then contain a subset of your Bookmark Library, much like a playlist in iTunes is a subset of your Music Library.

Your browser history would be automatically added to your Bookmark Library, so that an item in your history is in exactly the same place as any other bookmark.

In addition to that, we'd want to add a lot more meta-data about each and every bookmark. We'd want to add last visited, visit count, referrer information, how the site was left, keywords, comments, and even a summarized version of the text on the site (taking advantages of OS X's "summarize" service features.

Then, with all that meta-data, we could add "smart bookmark folders," like "sites I visit a lot but haven't visited this week," or "sites I visited that were linked from URL X," or "Prius sites I looked at last week."

Yeah. Now that would be putting an iTunes-like interface on the tried and true bookmarks interface, and it'd rock.

Discussion and Citation in the Blogosphere...

This article was somewhat interesting, but it didn't seem particularly profound. I also read most of the articles this article links. I kind of felt like there should be a meta-article written about the article, though. As it was, it felt like the author ignored how what he was saying was just another example of what he was describing.

Remember how I wrote about the Save and Restore Safari URLs AppleScripts awhile ago, and how I'd had ideas to do all of that using OmniWeb and OmniOutliner, only to be thwarted when I discovered those programs lacked the AppleScript support to do what I wanted?

Yeah, well, it turns out the newly released OmniWeb 4.2's "AppleScript support has been improved and now allows access to your bookmarks."

Oh, and on the OmniOutliner Exras page, there's a "Save OmniWeb URLs" script: "It's useful when you have a bunch of pages you don't want to lose, but you don't necessarily want to keep the browser running all the time or take the trouble to bookmark them all by hand." Oh, and it was last updated on December 5, 2002, exactly two months before I wrote about my ideas.

So, yep, just as I suspected: There's not an original one in my head.

First up, I read this article about the Windows build process. As I read it, I was a good combination of horrified and fascinated. I can't even begin to comprehend the complexity involved in that thing... Five thousand programmers? That's almost as many residents as live in the UC Berkeley dorms, and I can't imagine all of them being organized towards one common goal, with the possibility that something someone does will impact someone else's system in some unpredictable way...

And yes, I was horrified by both the complexity being described, and horrified that I was fascinated by what I was reading. Proof once again that I am a geek.

As I read the article, I imagined that the programmers at the lowest levels didn't even know what they were working on -- that they were given a specification for a subroutine, and were just asked to make it do something based on the input. This image was fueled by Stephenson's description of Y.T.'s mom's job with the Feds in Snow Crash, where the programmers are "interchangeable parts" turning out tiny widgets, and the actual software engineering is done by people who aren't actually programming.

The quote that most caught my eye from the article was about the significance of a broken build, which boggled my mind almost as much as the complexity involved in the first place. And I

"We've sent out calls at 3 a.m. when the build is broken, find the developer that broke it, and get him into work right then and fix it immediately. The developers are on call 24 hours a day. There's definitely an escalation process. A broken build is considered a critical, severity-1 problem."

It also got me wondering how Mac OS X and Apple compares. Maybe I'll find out some day.

Second, this interview with Marc Andreessen (one of the co-founders of Netscape) was kind of short, but he did say one thing that I found very thought provoking:

"Things like the back and forward button, we never intended that to be a permanent part of the interface. But people get locked into metaphors. You have to be careful with the metaphors you put in front of people because once they click onto one, that's it.

Consider how pervasive the back and forward button have become these days -- to the point of being "basic" features of our file system browsers -- and you'll see how significant it really is.