Paul's Down-Home Page

Joel Spolsky has a great book review this month that tears into the meaty topic of why Windows and UNIX programmers are so different. Contrary to popular belief, it’s not because Windows sucks, nor (as suggested by the comments here) is it because Windows programmers are all brainwashed. Joel claims it’s a simple matter of cultural differences, and that’s true to a very large extent. In my own career, I’ve written code for VMS, various flavors of UNIX, Windows, and the classic Mac OS, and I can certainly finger the cultural differences that underlie each of their APIs and programming models. It’s just that one cultural assumption most UNIX programmers bring is that if it’s not done their way it’s wrong… how very American of them.
And, since Jeremy wondered why Joel has a fan club… he writes consistently interesting essays. Whether or not you agree with them, they’re always thought-provoking, which is more than can be said of lots of other writers.

Jeremy had an interesting post about the difference between writing and explaining, which he sees as two sometimes separate topics. I’ve had the same experience he describes: it’s usually easier to explain something in a conversational way (even if it’s via email or IM) than it is to sit down and write an explanation from scratch. When you’re writing, you don’t have the same information about your audience’s assumptions and knowledge that you do when you’re standing next to someone at the whiteboard.
My solution has been twofold. First, I tend to write like I talk; one consistent comment I’ve gotten from readers of my books and columns is that they like my conversational style (although I do have to write less conversationally for more formal audiences). The other is to treat my writing like a conversation; at least for books, I actively try to find volunteer reviewers with a wide range of experience levels, then I push them to give me good feedback. This helps a great deal, especially if you can find people to bounce ideas off as you go. It helps if you’re really familiar with your audience and the things they want to know, which is why I spend so much time talking to Exchange and Windows administrators in the larger community.
Of course, neither of these approaches actually helps you get the stupid writing done in the first place. For that, I recommend Heinlein’s Five Rules, as amended by Robert Sawyer.

In this week’s column, I wrote:

Many administrators I know like spicy food. (I’m not sure why. Having grown up in southern Louisiana, I have a large extended family that glories in eating stuff hot enough to peel paint; maybe I just gravitate to people who remind me of my cousins.) I recently learned of Blair’s Death Rain habanero potato chips, supposedly the hottest chip you can buy. I have some on order; if you’re interested, drop me an email message and I’ll let you know whether they’re a worthy gift for the snack-happy admin on your list.

So, I go look in the mail folder where column comments go, and there are seven or eight messages from people wanting to know more. This is unusual, insofar that a typical column will generate one or two responses, and this one’s only two days old. I don’t even have the chips yet, but it’s cool to see my linkage between Exchange admins and spicy food being experimentally confirmed.
For my readers: I ordered the chips from IronQ.com. They were $30 for a mixed case of 2oz bags. We’re eating them at our holiday party, so I’ll post a full report once I’ve tested them.

If you do a Feedster search for Robichaux, you’ll quickly find Julian Robichaux‘s site. Turns out he’s an expert on Lotus Notes, the biggest competitor of the main product I specialize in. Since there aren’t all that many people named “Robichaux” in the world in the first place, I wonder exactly how we’re related. Hey, Julian, if you read this and decide to come over to the light side we’d love to have you!

As if the my son’s inexplicable passion for all things Wolverine wasn’t bad enough (at least Matthew can gleefully say “Michigan wucky!”), it turns out that Michigan takes, and files, DNA samples from all babies born in the state. ‘Scuse me, but I don’t see a compelling state interest in gathering DNA fingerprints for every live birth. Well, wait a minute– I do see one, but I don’t like it. I understand, accept, and support the idea behind doing DNA-based screening for health problems, but that’s not the same as maintaining the samples “indefinitely“.
This came across Politech today:

When our son Dylan was born at the University of Michigan Hospital, the blood sample was taken without parental consent and over my explicit objections. Two techs who worked for the state drew the blood sample, placed five drops on the top of a carefully marked card, and left without speaking to anyone else. Several doctors came in ahead of the techs, began giving us discharge instructions at the time the sample was taken, and they stood in my way as I tried to approach Dylan. They couldn’t understand how I would mind that a blood sample was being gathered by the state, merely saying that “it’s required by law” and that “it doesn’t hurt (too much)”. They later gave me several pamphlets explaining the importance of newborn screening. (Indeed, it is, but I can ask my doctor to do it too)
I spoke with Harry Hawkins of the Michigan Department of Community Health, who assured me that the samples were kept at an undisclosed warehouse which
locked behind a chain link fence, and that they would be destroyed after 21 1/2 years. However, he agreed to destroy the sample if both parents requested it.
His mother and I sent Harry Hawkins a letter stating, “Please destroy all samples of our son’s blood. If any identifying data such as a DNA fingerprint has been obtained from his sample, please destroy that data also.” (with all of the required details below)
I later received a notarized form indicating that the blood sample had been destroyed and witnessed. There was no mention of any DNA information, but I have no reason to believe that they gathered any, so I don’t intend to pursue the matter further.