Personal Logs

Saturday, December 17, 2005

Bill Gates' "Think week" (2004)

Bill Gates has a habit of "Think Week". It is taking a week off every year and spend it reading the latest research papers that he wouldn't have time to read normally. This is not that a unique concept nowadays because a lot executives do "corporate retreat", which is a very similar thing.
The link here points to a "eWeek" interview with Gates about his think week in 2004. I know we are approaching the end of 2005 but I am still excited to read about it.
He has talked about his routine in the think week. Basically it's just Reading + daily necessities (sleep, eat, etc). He makes commentary about each piece he has read. I think this is a brilliant idea! Everybody knows the benefit of writing comments on what we read. But we don't do it often at all because there are just more stuff than we could ever read! However, I think it is wise to trade a bit of reading time into writing time because the benefits is just so huge that we wouldn't consiously give away. We need to write something to really digest the information and recall it easily when we need later.
Blogging is a perfect place to write the commentaries. Especially a blog without many audiences, I wouldn't worry about boring my audiences to death :)

OK, here is this article's summary:

Gates is paying attention to microporcessing.
Gates has started reading about wireless tech in depth. He has pointed out Ultra-wideband and WiMax (802.16)
Gates is closely watching security. He has written a memo about IPSEC (IP security protocol)
Machine Translation (a research funded by Microsoft) is making good progress that English help docs can be translated to Spanish automatically and users don't tell significant difference from the human translated stuff.
Gates considers XML to be the next single big thing in IT. Every web services on the planet can be used as a subroutine of your system. This is a really powerful statement. MSFT and IBM has organized a few interop fest to crystalise the standard. Smart move !
Security is the highest pirority in MSFT. Longhorn is pushed back for this reason.
The big products coming before Longhorn are mainly infrastructural. Normal customers wouldn't even notice when they are deployed.
Visual Studio 2005 (codename) Whitehorse will allow visual designing of web services and visual customisation of relationship between application and deployment. Visual modeling is a hot topic. IBM's Rational group is doing some interesting things on this as well and they have Eclipse. Gates said UML is not the full answer to modeling. He think Modeling should be so powerful that businesses can directly use it to customise their enterpise solutions visually.

Saturday, December 03, 2005

Working hard and Reward

I'd like to share an interesting and important point that is brought up in a conversation with my manager.

Today we talked about my recent progress. I told him how much I enjoyed my last project sincerely. But I have also expressed my concern about the next few months scheduled for the team to concentrate on defect fixings. I said I enjoy working on new projects because they are more fun and make me grow faster. I like working hard and take on extra responsibilities. But I can't really work hard in defect fixing because there are only a small number of interesting defects and a large number of boring ones. I would end up with all the crap defects if I work hard and clear up all the interesting ones faster than they come in. He is quite shocked by this and agreed this is an obvious omission in his planning. He used the analogy that if you work hard on peeling potatoes and only get rewarded with more potatoes, you would be pissed eventually. But if you work hard at eating cakes and get rewarded with more cakes, you will work even harder. I'm sure he's been thinking about this a lot after the talk.

This problem exist only in a small number of domains where the workers actually do their job because they enjoy it. Of course not all the programmers enjoy writing code. But the ones do definitely have a big plus over the ones who don't. A culture of rewarding good programmers with good works can be much more effective and healthy than monetary reward.