Monday, May 31, 2004

Laurel Canyon

Watched Laurel Canyon> on DVD, and loved it. However I am amazed at the mixed response it got. Some people loved it, while many including prominent critics hated it.

To me it had all the elements: interest, a strong and successful hippy mother contrasted with her tightly sprung type-A son; amusement, the laid back left coast seduces the uptight east coast; and of course those very necessary elements in these modern times of sex & drugs & rock 'n roll. What more could you ask for?

Even many of the people who liked it hated the ending. The ending is abrupt and it is not properly signaled, but I laughed at it. Moreover, you cannot expect closure and a neat wrapping up on anything as messy as this set of lives. The relationships will just continue to morph with the changing guest list at the house party.

Thursday, May 27, 2004

Finding R

Yesterday an acquaintance sent an email that mentioned R. It had come up in another context last week, so thinking that I should find out more about it, I turned to Google. But I thought, you cannot just search for "R", that will bring up the whole world. So I searched for "r data visualization". Nothing useful.

As I knew that R is an open source project, I went to Sourceforge, but the search key there has to be at least 3 characters, so that did not work. Another thought is that my acquaintance had just been to the user conference, cutely called useR. Google that!

As it happens, if I had just searched for R, I would have found what I was looking for. What do you know!

Wednesday, May 12, 2004

Software Architecture and Modelling SIG

The SDForum has created a new Special Interest Group on Software Architecture and Modelling. Stephen McHenry from Netflix gave the initial talk followed by a Panel Session. Stephen talked about being a system architect, drawing from his many experiences, even touching on his current work at Netflix. The talk did not contain anything earth shattering or new, however he did a good job of covering the existing teritory. It was just the right thing for a new group.

There was one point made that I started to quibble with. Stephen said that you should try to delay making as many decisions as possible. This drew nods of appreciation from many audience members. I interrupted to say that there are some decisions that you have to make up front. Then the light bulb lit. A systems architecture is defined by the decisions that you make up front. All the decisions that you can delay are the implementation.