Entries this day: Funny_site Pretty_okay amazon_ec2 backups

Funny site

2:32pm JST Friday 19 October 2007 (day 13722)

I've been reading slashdot.org and came across a hilarious EULA that says "you cannot look at the source code of our page," blah blah like a bunch of babies. They also say I can't use their name nor link to their site.

To be clear, "they" are Dozier Internet Law, P.C., and their EULA is linked under their name.

I laughingly note that in the source code of all the pages *I* could find, they have inappropriate keywords listed:

<meta name="KEYWORDS" content="keywords go in here">
 <META name="y_key" content="1dfad02220b8c67b" /> <!-- For Yahoo authentication -->
<meta http-equiv="Content-Type" content="text/html; charset=iso-8859-1">
<meta name="generator" content="Adobe GoLive">

Hahaha

I guess that's what they get for using a site generator to make their pages instead of hiring a professional.

Pretty okay

4:20pm JST Friday 19 October 2007 (day 13722)

Pretty decently productive at work today. I got everything finished that I can likely do without input from my handlers (interesting word choice) and I've got a good start on doing the *next* part of that which they will likely want me to do.

7:30pm JST Friday 19 October 2007 (day 13722)

Met with 井上さん, and he suggested that instead of adding a click_count next to the view_count, that we just add a (SELECT ...) to the list of things to select. I was like, "cool," and he proceeded to write out the code freestyle, which was impressive enough for me. I didn't mention that my SQL coding isn't on the same par as being able to write stuff like that.

We ran it, and, nothing. Looks like he's not good enough to write stuff like that either.

We pored over it for a bit, but he finally gave up (I had given up already) and asked me to email him the code so he can play with it later. If he can write it, that's great; I'll put it into the project. I may try to work on it a bit this weekend, but probably I won't.

amazon ec2

4:35pm JST Friday 19 October 2007 (day 13722)

But that's not why I'm writing. I have just been poking around with Amazon's EC2, which seems interesting, though I'm not really sure how I can use it. Basically 10 cents per hour to use their CPUs. I was hoping it would be 10 cents per effective CPU hour (a bajillion times cheaper), and then I would look into ways of putting my website hosted from their virtual machine. Not sure that I can do that, though.

7:52pm JST Friday 19 October 2007 (day 13722)

Hmm. Well, I just looked up my "usage" on the service

1. They don't offer an HTML format for usage reports; only XML or CSV

2. They charge for each instantiation start as a full hour, and don't actually count the total usage and charge for each portion of an hour.

So, I can keep costs down by not starting and stopping services willy-nillfully.

I can keep costs down even more by not starting services at all.

The thing is, I can't really see how I would *use* the service. I successfully created my own AMI (Amazon Machine Image), and even included little executable files to zip it up and copy it to my S3 space.

But, .. well, I guess I could put a webserver on an AMI. I guess I could even put my site on it. But the URL would be all wonky: something like ec2-72-44-35-143.z-2.compute-1.amazonaws.com, which was my most recent one. notlong.com could take care of that, but... then what?

Hmmmm.

Are we allowed to make AMIs and then sell them?

backups

8:11pm JST Friday 19 October 2007 (day 13722)

While playing with Amazon EC2 and therefore Amazon S3, I decided I was overdue to archive my accumulation of stuff in my user space here locally. I look forward to Leopard with its "time machine," but even then it's only on local disk. Perhaps I will set up an ongoing backup with JungleDisk.