The Soul of a New Machine

I just finished Tracy Kidder’s “The Soul of a New Machine”. I
thoroughly enjoyed it, reading all but the first chapter during my
flights from Pittsburgh to Oakland, CA. Actually, the first chapter
was my least favorite, and after reading it I put the book down for
several weeks before resuming. The book describes the creation of the
32-bit Eagle, backwards compatible with the 16-bit Eclipse, at Data
General Corporation during roughly 1978-1980. This project stands out
in the degree to which management isolated the engineering team from
the rest of the company and gave them a great deal of technical
freedom. This allowed the engineers to design and build what they
thought was a good machine. I do not say “great” machine because the
engineers worked under enormous time pressure, as has proven to be
typical of the industry. Still, as regular a participant in endeavors
for which panic-stricken is the best description of the schedule, this
book struck a chord with me. One of the main characters, Tom West,
sounds like a fascinating man. The prologue describes him as “a good
man in a storm”. I hope that people may someday say the same about
me.

2007 Dad Vail Regatta

I raced in the Dad Vail Regatta for the last time this past weekend in the Men’s Varsity Pair with Adam Suhy. Though the weather was not cooperative this season, we had some strong racing results. Here is a summarized version of our season:

Date Race Location Result
March 24, 2007 Murphy Cup Regatta Philadelphia, PA Second to Pitt
April 1, 2007 UAA Championships Pittsburgh, PA Second to Washington University
April 14, 2007 – April 15, 2007 Knecht Cup Camden, NJ Second to University of Buffalo
April 21, 2007 Car-Du-Pitt Pittsburgh, PA Cancelled (weather)
April 28, 2007 Governer’s Cup Regatta Charleston, WV Cancelled (weather)
May 6, 2007 Mid Atlantic Championships Fairfax, VA First!
May 11, 2007 – May 12, 2007 Dad Vail Regatta Philadelphia, PA Third in our semi-final

By far the most exciting race was the Dad Vail, which is our national championship. It is arranged into heats, semi-finals, and finals, with 3 advancing from each heat, and two advancing from each semi-final, to create a six-boat final. We placed second in our heat to Georgia Tech, which was good enough to advance to semi-finals. Our heat was not an all-out effort.

Our semi-final was the most intense race I’ve ever rowed in. Here are the results for all three semi-finals from the Dad Vail webpage.

Men’s Varsity Pair – Semis 2 to Final
Evt Lane 1 Lane 2 Lane 3 Lane 4 Lane 5 Lane 6 Start
116
Fordham 42
Place: 6
Time: 7:32.48
Carnegie Mellon 17
Place: 3
Time: 7:05.86
Boston College 7
Place: 5
Time: 7:31.2
Oklahoma City 78
Place: 2
Time: 7:05.72
Bowdoin 8
Place: 4
Time: 7:16.6
Iona 49
Place: 1
Time: 7:05.49
8:35
117
Florida Tech 40
Place: 6
Time: 8:06.18
Washington U.114
Place: 2
Time: 7:08.58
Temple 102
Place: 4
Time: 7:26.89
Buffalo 13
Place: 1
Time: 7:01.8
Cincinnati 25
Place: 5
Time: 7:32.62
Rollins 90
Place: 3
Time: 7:25.71
8:42
118
U. Rochester 89
Place: 6
Time: 7:34.5
Cal Long Beach 16
Place: 1
Time: 7:03.06
Toronto Miss 104
Place: 2
Time: 7:07.21
Georgia Tech 45
Place: 5
Time: 7:26.61
Grand Valley 47
Place: 4
Time: 7:19.74
Duquesne 37
Place: 3
Time: 7:07.64
8:49

You can see that we took third to Iona and Oklahoma City, with the sorted times being: 7:05.49, 7:05.72, and 7:05.86. That means we missed second place by 0.14 seconds (which would have been sufficient to advance), and first place by 0.37 seconds. Our time would have been good enough to advance in either of the other two semi finals. Also interesting is that Georgia Tech, who finished ahead of us in our heat, took 5th in their semi-final, and it was not a close race for them. Even more interesting is that Iona went on to take second place overall.

Thus, while it is disappointing that we did not advance to finals, we were a bit unlucky in the way the semi-finals stacked up, and I feel like we had one hell of a race and were definitely in there with the fastest boats.

VMware vs Parallels on OS X

The battle rages! I just installed VMware Fusion and it worked without a hitch. I downloaded a pre-installed Debian VM image and it worked great. I also just did an Ubuntu install into a Parallels VM and it went smoothly. This is one of those rare moments where I have two software options and both actually work! I should note that I paid money for Parallels, while VMware Fusion is free.

So far they’re both tied for first. Parallels is likely to emerge the winner because of its more integrated file sharing between the host and VM (I believe VMware workstation offers similar functionality) and because of Parallels’ “coherence” mode, where Windows apps appear to be OS X apps.

Still, this is cool. 🙂

Serial port confusion

So there is an IDC header (two rows of pins) on the motherboard of one of my machines for a serial port. To make use of this, I need an IDC – DB9 adapter. It turns out these come in two flavors, depending on whether they’re aligned based on pin number of physical layout. Examples of each kind are here and here. Here is the pin numbering for each of the connectors:

IDC10:
1 3 5 7 9
2 4 6 8 10

DB9:
1 2 3 4 5
6 7 8 9

Wiring diagrams for each kind of alignment:

Physical alignment (aka DTK):

IDC10 DB9
1 1
2 6
3 2
4 7
5 3
6 8
7 4
8 9
9 5

One-to-one pin number alignment (aka EVEREX/APEX10):

IDC10 DB9
1 1
2 2
3 3
4 4
5 5
6 6
7 7
8 8
9 9

Other pages that discuss this issue are here, and here.