Sunday, September 24, 2006

Calvin&Hobbs I

One of my favorite quotes from Calvin&Hobbs. Now all I have to do is find the strip:

Mrs. Wormwood: "Calvin, can you tell us what Lewis and Clark did?"
Calvin: "No, but I can recite the secret superhero origin of each member of Captain Napalm's Thermonuclear League of Liberty."
Mrs. Wormwood: "See me after class, Calvin."
Calvin: (speaking in retrospect) "I'm not dumb. I just have a command of thoroughly useless information."

Update: Found the strip

Calvin and Hobbes - I am not dumb

Friday, September 15, 2006

CEO Phooey

I've been reading Sun CEO Jonathan's blog. It was interesting at first but the more I read, the more I got the feeling its all fluff and no substance. He gushing about how Sun will change the future with things like UltraSPARC machines which has a lower TCO than any other machines, and about Solaris. He even had his friend Marc Anderson (creator of Netscape) to talk about how Marc runs his latest startup on Solaris on Sun machines. Then he talks a new service called Sun Grid at that is offering Grid computing at $1/CPU Hr.

Here's why I think its all fluff. I tried to download Solaris. It's free but the site asked me to fill in a form so they can harvest my personal information in the name providing me with a license code. I decided against it. In fact, the download is more than a couple GB in size. No way I'm gonna waste my 10GB limited bandwidth on that. If they had made it available openly like Linux, then mirrors around the world could've host it making it more attractive. Plus as more people uses it, the more software will be available for it. Right now, it takes a year for a new software to make its way to Solaris platform. For example RubyOnRails.

Don't get me started on TCO. That topic is too biased no matter how impartial everyone tries to be.

Next, Grid Computing. Sun runs a Grid compting network on UltraSPARC machines. People pays $1/CPU hr. Who in the world will use that? Amazon started a similar service six months ago and it went nowhere. The reason. Well, the only people who legitimately needs Grid Computing (and heavily will use to make the service profitable) are research lab, utility, financial and government agency. But none of them worth their salt will use the service due to confidentiality/security/privacy issues. I think that leaves lame ass smaller companies at the long tail end that needs Grid computing.

And after reading through the site, I was right. The first customer listed on their press release to use it is a biotech company who says
Sun Grid Compute Utility Helps Genetic R&D Company Get Life-Saving Products to Market in Record Time.
How many times have you heard that from a pharmacetical company in the last 10 years. How many genetic based cure do you know that saved some ones life?

Anyway, I think Jonathan as a CEO is no good. He is at heart a techie. He believes in using technology to solve all of world problem Unfortunately, thats only a wish and not a vision. Without vision, you don't stand a chance. He would be better suited as a CTO.

Why do you need vision to win. Well, that post will be coming soon.

Friday, September 08, 2006

My 43Things

I just signed up for 43things. I've been resisting the whole social networking movement for long. It simply takes too much time signing up and filling those forms. But it was only a matter of time. With 43Things, you select things/places/books/movie you did, doing or want to do. The site then tracks you against time and other people. The fun part's you get to read other people's experience who's in the same boat as you. You can read their blogs and 'cheer' them on while they too cheer you. You can see my 43Things on the right sidebar here.

The site is created using Ruby on Rails by a bunch of folks who worked at before. Recently they turned two and got funding from, so looks like they made it. Here are a few things a learnt along the way:

  • Bring your own computer.

  • Sit around one big table. Make communication easy.

  • No discussions over mailing lists.

  • Meetings take place at the pub, or as a standing meeting.

  • Use paper and notecards instead of documents.

  • Share the stereo.

  • Walk or bike to work if you can.

  • Eat lunch together. Play credit card roulette.

  • Make it more fun than useful.

  • Embrace constraints.

  • Build something you’d use. Use what you build.

  • Don’t overmonetize.

  • Rely on the software in your head rather than try to replicate culture in features.

  • Design as you build. Keep the cost of change low.

  • Be in it to win it.

Thursday, September 07, 2006

Overhead in Data Mining Class

Professor: A lot of professors will tell you that there is no such thing as a bad question.

Professor: (pause)

Professor: Yeah, they're all full of shit.

HP Goes On Dunn Hunt

Long story short:
1) One of HP's board of directors was leaking info to CNET.
2) HP's chairwoman, Dunn, authorized independent electronic-security experts to spy on the January 2006 communications of the other 10 directors.
3) These security experts used "pretexting" in order to coerce phone companies to send phone records to their own email addresses so they could determine who dunn it.
4) The slime was found, confronted, and admitted his leaking. He's decided to remain on the board until shareholders vote him off.
5) In protest of her tactics, another director, Perkins, resigned calling Dunn's tactics illegal, unethical and a misplaced corporate priority.
6) Dunn, and HP, filed paperwork with the SEC citing Perkins' resignation, but failed to identify the reason why he resigned in order to hide Dunn's investigation... a point which seems to be legally required by SEC regulations.
7) Shit ---> Fan

Wednesday, September 06, 2006


I never met Steve in person, yet I feel like I did know him since I watched him so much on TV. I have to say, I was never a huge fan - although I was a fan and I certainly admired his passion. Its rare to see someone who is so dedicated to what they do and loves it so much. I just feel at such a loss that the world has lost a wonderful bloke.