February 9th, 2008

Jill and Sean in the Washington Post

Jill and I appear on the front page of today’s Washington Post business section. We were interviewed about our family’s financial goals and so forth. There are some misleading quotes, but I have come to expect that from journalists, and overall it is accurate.

Granted, I had always hoped that our first appearance in the business section would be coverage of a cutting-edge new stock derivatives analysis we jointly published in the Journal of Mathematical Finance, but this is also pretty neat. ;)

February 5th, 2008

Monowheel

February 5th, 2008

Snow Lions

Found at Incredimazing.

February 3rd, 2008

I finally got around to installing a COBOL compiler on my computer here at home. I downloaded OpenCOBOL which built from source with no problems on my Mac. I went ahead and wrote the typical “Hello World” program to test that everything compiled and installed correctly:

1
2
3
4
5
000001 IDENTIFICATION DIVISION.
000002 PROGRAM-ID. hello.
000003 PROCEDURE DIVISION.
000004 DISPLAY "Hello World!".
000005 STOP RUN.

And compiled it:

cobc -x hello.cob

And ran it:

./hello
Hello World!

Brilliant. Now I need to find a good book on the COBOL language so I can get down the syntax. And also find some fairly sizable COBOL code that I can read through to get a sense of idiomatic COBOL.

Fun!

December 10th, 2007

I killed off my Facebook account mere minutes ago. Since I have this blog, another couple of blog ideas in development, and other ways of connecting with people I care about, the Facebook account seemed superfluous for me. If you look for me in Facebook, I might still appear, but I have changed most of my information to nonsense. This is because Facebook does not allow you to delete your account, only to “deactivate” it, where “reactivating” it is done by logging back in. That sounds more like I’ve just logged off than deactivated.

I am now quite convinced that web applications should always allow you to delete your account. 37 Signals does that very nicely, I might add.

UPDATE: 03 Feb 2008. I ended up recreating my Facebook account mere days later. There are still things I want to tinker around with in the Facebook applications system.

November 28th, 2007

Word to those who are having this problem:

If you are using Opaque Menubar to turn your translucent menubar into a solid color, you might experience a problem with the desktop background.

If you set a desktop background that is not *exactly* the same size as the screen, it will be immediately scaled to fit the screen. Trying to tile? Won’t work, it will rescale the single tile to fit the whole screen. Etc.

I’ve had this problem for a week before I got tired of it and dug in to fix it.

Turns out that the OpaqueMenubar does some fancy trickery with setting the desktop background in a certain way to turn the menubar a solid color. The unfortunate side effect is that is causes this background scale-to-screen problem. Killing the Opaque Menubar process reverts the desktop background to whatever scale/tiling/centering you want.

It’s a bit frustrating, because I do like the solid menubar, but I like a nice tiled background more.

Note: this bug also causes any sort of dynamic background (like the very awesome EarthDesk) to revert to a default (and static) Mac background image.

November 24th, 2007

In the end, after all the work I did to get w3m and other Emacs add-ons working, I went back to Vim. There’s something in me that just prefers modal editing and the simple structure of Vim to the multi-faceted complexity of Emacs.

Currently, I am using MacVim, which does a great job of integrating Vim in with Mac idioms.

November 12th, 2007

Short note.

I decided to start using emacs as my primary editor for my programming. Since I use a Mac at home, I went ahead and installed both Aquamacs and Carbon Emacs to see which one I like better.

For both of these, I wanted to install w3m, a textual web browser. This was an enormous pain. Installing w3m from MacPorts failed — it either locked during compilation, or it would compile and then ram to 99% cpu usage when starting w3m.

So I tried Fink, which was a bit slow and not as nice as MacPorts. But w3m ran.

I didn’t want to keep Fink, since it looks a little behind the curve, so I tried to get MacPorts to compile and run correctly, using the source and patch information from Fink to help me.

I finally got it running, hurrah. I don’t know what the actual underlying problem is, but w3m depends on boehmgc. MacPorts uses version 7, and Fink uses version 6.7 So I edited the Portfile in the MacPort and changed it to use boehmgc 6.7. Ran “sudo port install boehmgc”, and that installed fine, so the 6.7 version was ok. Then ran “sudo port install w3m” and that compiled and installed fine; and it ran fine.

So, if nothing else, replacing the boehmgc version 7 with 6.7 seems to do the trick.

Sorry for the messy posting. I am a little tired after spending a couple of hours on this, and am not writing at my best.

October 10th, 2007

Louis and Chmee donned impact armor: leathery stuff, not unpleasantly stiff, which would go rigid as steel under impact from spear, arrow, or bullet.

From The Ringworld Engineers by Larry Niven, 1985.

d3o technology is a specially engineered material with intelligent molecules that flow with you as you move but on shock lock together to absorb impact energy.

From the d3o lab, 2007

October 10th, 2007

The Decennial Response Integration System (DRIS), which will collect and integrate the census information collected from various sources, also is behind schedule, according to GAO. DRIS is on budget but is expected to deliver reduced functionality for the April 2008 dress rehearsal, in which the bureau tests all equipment and business processes needed to carry out the census, requiring more testing later in the process. A third project, called Data Access and Dissemination System II (DADS II), is also behind schedule. The Census Bureau originally planned to award a contract in 2005, but just awarded it in September.

“Delays in functionality mean that the dress rehearsal operational testing will take place without the full complement of systems and functionality that was originally planned,” according to GAO. “However, bureau officials have not finalized their plans for testing all the systems. . . . Without sufficient testing of all systems and their functionality, the bureau increases the risk that costs will increase further, that decennial systems will not perform as expected, or both.”

Been there. Done that.

From Government Executive.