Hello everyone -- my name is Dave and I love to ride!

Finally a day that was warm enough to get on my newly-inflated wheels.

I rode north to 9th and Lafayette. Went west to 6th where I jogged south on Christopher and over to my favorite Hudson River Bikeway. Then the big question -- north or south? Considering this was the first ride of the season, there were good arguments to go south. Shorter ride. Well-defined terminus. Possibly other routes to come home on. But I went north.

Familiar route. Turned around at the Intrepid, and crossed the highway at Horatio. Rode south on Washington, turned east on 10th and rode it to Cooper Sq where I turned south. Considered taking Bleecker St. Maybe next time.

It was a bit chilly and my legs burned a bit and I got tired sooner than I would have if I had been doing it every day, but IT FELT FUCKING GREAT TO BE BACK ON MY WHEELS.

Here's the route and stats thanks to CycleMeter.

I've also hooked up CycleMeter and DailyMile, so my trips will automatically (knock wood) show up over there too.

 4/9/2011; 6:26:56 PM.   .

There was something troubling about Larry Page's edict to his employees that tied their year-end bonuses to Google's success in social networking. I couldn't put my finger on it, but Mike Elgan at ComputerWorld helped me see it.

Page's edict tells employees: "Stop working on Google's mission and start working on Facebook's."

As we used to say here on Scripting News -- Bing!

When have we seen this before?

Heh.

When haven't we?

This is the cycle of the tech industry. A company starts off scrappy, and against all odds, not only does it vanquish every startup of its generation, but it also takes the lead from the former incumbent. Then it grows big, hires lots of employees, many of them from the companies they defeated, and then are surprised when they become the companies they defeated.

Meanwhile a new crop sprouts, one of them rises to the top, and takes aim at the leader...

The founder, whose brilliance and tenacity and drive was what made the company surge past everyone else thinks he can do it again, only this time instead of leading a small team of banditos, he's running a fleet with aircraft carriers, battleships, supply lines. And btw, the company they took the lead from is still kicking around, and capable of making trouble.

We know how this story ends. The upstart takes the lead away from the incumbent.

Microsoft didn't take the graceful exit, which was to become a services company and banker, and fade into the background. But destiny forced them there anyway. (They thought their enemy was Netscape, but it turned out to be Google.) It was more of a struggle than it had to be. Same with IBM, before them. And now Google seems to be going that way.

There is of course the Steve Jobs story. But to go down that path, Google would have to return to its roots -- search -- and forget about dreams of being Facebook.

 4/9/2011; 12:38:51 PM.   .

Today I'm going to try to make something work that may be hard

I have a domain, let's call it domain-x.org, and I want it to return a feed.

The content-type has to be text/xml and the request must return XML.

No redirecting.

Since it's not a sub-domain, a CNAME will not be possible.

Which means we will not be able to use Amazon S3, which is a real shame because in every other way it's the perfect way to do it, from a scaling and cost standpoint.

Pretty sure I'm going to have to use Apache. And since I don't have a Linux server lying around, it's going to have to be Apache/Windows.

This isn't just an idle exercise, we need to solve this problem in a way that can be easily replicated for lots of users, without them having to configure anything as low-level as Apache or S3. This is a feature for true end-users. So in the next step I will need that REST-programmable DNS that I've been talking about.

Update: It works.

And it was very straightforward.

Here's what I did.

1. I created a folder on my Apache server called davelinks.com.

2. Went to my registrar and pointed it at the Apache server.

3. Added a .htaccess file in the folder with a single line:

DirectoryIndex rss.xml

4. Added a callback when my feed is built that in addition to writing it to the usual deeply-nested place on S3, also writes it to the top level of the folder I created in step 1. The file is of course called rss.xml.

5. http://davelinks.com/ works.

Took about 1/2 hour.

 4/9/2011; 10:46:05 AM.   .