This page is designed for me. No one else should find it useful. If you do, please consult a physician.
The contents of this page have moved to
I've been using Windows Vista Beta 2 (via the customer preview program) for about two weeks now.
Vista is aptly named. Vista provides a vista to the future. Unfortunately, it is merely a vista to the future, and not the future itself. Vista has potential, but is currently unstable, buggy, and idiosynchratic.
I like Vista. It has potential. It simply isn't quite there yet. Hopefully, RC1 will fix many problems.
Without further ado, here are the bugs I've found in Vista Beta 2.
1. On a Vista machine, use mstsc.exe to connect to another machine (I've used a WinXP machine, not another Vista machine). Maximize the remote desktop window.
2. On the WinXP machine (machine 2), log back on without disconnecting from the Vista session.
3. Return to the Vista machine and click through the "you've been disconnected" error. Reconnect to the machine. The "menu" at the top of the remote desktop session does not appear, nor is there any way to make it appear. In fact, the only way to exit from the new remote desktop session is to ctrl-alt-del and load task manager.
While using a Vista machine from mstsc.exe on a WinXP machine, attempt to add a TCP/IP printer (add a printer, local printer, create a new port, blah-printer.microsoft.com, ...)
After entering the printer address or IP address, clicking next will yield the error "The TCP/IP wizard page was not able to load" or something like that. The upshot: you cannot add a TCP/IP printer over a remote desktop connection. Ouch!
So far, the count is just at two, if you remove the 10's of crashes I've had that were "due to drivers."
Oops, we have a slight diversion.
Without further explanation, I present the
Drink for all ...
1. homosexual references,
1b. Exception: finish your drink during the volleyball game.
2. theme songs (i.e. Danger Zone, You Never Close Your Eyes, etc.),
3. lock tones,
4. utterances of Maverick, Goose, or Ghostrider,
5. references to MIG, bogey,
6. landings and takeoffs,
7. Goose deaths,
8. facebook references (i.e. "Too close for missiles, switching to guns").
Retrospectively, 1, 4, and 5 are more than sufficient. Type 3 drinks are to push you over the edge. The rest are to maintain a thorough buzz throughly the movie/game.
Sparse matrix file formats bug me. Literally, they cause me pain. Everyone uses a slightly different sparse matrix file format and they are all very close, but have slight peculiarities.
I have two ways of dealing with this problem: i) give up; or ii) figure something out. I chose ii. My solution is two-fold. First, I design Matlab codes to read and write the different formats. Second, I design templated C++ codes to read and write the different formats. The Matlab solution allows me to get up and running quickly and is sufficient for infrequent datasets. The C++ solution is more robust and allows my C++ programs to read and write these formats as well.
More coming soon...
I used to have a wine page on my Stanford web-page. I'm going to remove that one soon and move the page here.
To get the process started, check out the new wine page. The page basically has summaries of the best wine column I read. It also serves as a place to collect notes on what to buy/not buy in stores these days.
Maybe eventually, I'll add my own reviews, who knows.
I decided tonight I need to start charging consulting rates.
$0.50/Matlab "How do I" query
The conference today was very interesting.
Talks I attended:
Much on preconditioning at the end there.
Still no ideas why the diameter esitmation isn't working.
Things I need to work on this weekend: reviewing papers; testing the copying model.
Damn. There must be a bug in my diameter estimation code. (Or I don't really have a strongly connected component of my graph.) Time will tell.
I'll update this page like a blog. Let's see how it works. Never expect anything and you won't be disappointed.
Copyright David Gleich (2006).