|
|
HomeGraffitiAboutSitemapVisualDevWork
Email BrianGPG Key |
GRAFFITI -- July 07 thru July 13, 2003>> Latest: Sunday Morning <<Last Week << Mon Tues Wed Thu Fri Sat Sun >> Next Week Welcome to Orb Graffiti, a place for me to write daily about life and computers. Contrary to popular belief, the two are not interchangeable. About eMail - I publish email sometimes. If you send me an email and you want privacy or anonymity, please say so clearly at the beginning of your message.. |
WebCam
|
MONDAY
Tues
Wed
Thu
Fri
Sat
Sun
July 07, 2003 - Updates at 0700 EST
Good morning. Three day weekends make it hard to get up when the alarm only goes off 7 or 8 times on Monday morning.... Yeah, I'm running late, and I've no excuse. I was awake enough to get up at 0545, but that's a bit too early, before the alarm even. So I laid my head back down. Poof. Now it's an hour later and I'm running late. Leap out of bed and into the shower in one disjointed, fractured way (I'm too old to do "one fluid motion" anymore), shave, teeth, etc. Then pop in here to say, "Hi."
Now I must fly. There's a largish bit of stuff up starting on Friday, though, so I hope you won't be too disappointed. I did get a great bit of yardwork done yesterday, and put in several hours on the mail server howto, which is coming together. I really just need to understand now how (or even why) Debian's Cyrus packages are so much different than when I build from source. Anyway, time for me to go. See ya!
Mon
TUESDAY
Wed
Thu
Fri
Sat
Sun
July 08, 2003 - Updates at 0720
Today, I'm not in nearly the rush that I was yesterday. Good morning. That isn't to say that I've accomplished much in the intervening time. We figured out a problem with swap partitions yesterday on a Sun E450, installed and tested the major application on the first of those 4-head workstations, and had a reasonably successful day overall. Today, I'm swinging through the office to drop off an RMA package for Dell, and then off to a client site in Gaithersburg, where I'll spend the first half-day going through their Windows XP machines, applying updates, making sure the AV is working, running AdAware and so on. In the afternoon, I'll be working with their web guy to design a few Mailman lists to replace their commercial outsourced subscription email service.
I made no progress on the Email Server HOWTO last night. I'm still fuddled by the Cyrus packages for Debian, and also I'm slightly distressed that they're so far downrev, even in testing, from the 2.1 series that's current. I'll figure it out one way or another. Perhaps tonight... See you later, I'm going to get ready to go.
Mon
Tues
WEDNESDAY
Thu
Fri
Sat
Sun
July 09, 2003 - Updates at 0711
Good morning. It would appear that I managed to forget to roll the dayname at the top of the page yesterday morning. There was a short post yesterday, up at the appropriate time, and I'll bookend that one today. Yesterday was uneventful - the visit to the customer site went fine. I didn't have a chance to meet with the guy who'll be helping design the requirements for their customer mailing list, so the day was a little shorter than planned. That's alright, I suppose.
I did laundry when I got home, and a spot of yardwork, including watering all the bits that were looking droopy. It never did rain yesterday, although it was predicted. There's chances of thunderstorms and occasionally heavy rains each day through the rest of this week, however. In the evening I mucked about some more with Cyrus and the Debian packages thereof. I think I've decided for the howto, to work with source for the Cyrus IMAP and POP3 server daemon. It's much more recent, from source, with important features that are missing from the 1.5x series that's available in Debian Testing.
Today, I'm in Rockville for take three on the fibre light-up. We got some single-mode modules where we had needed multimode, so everything went on hold while those were ordered and shipped. Hopefully they're there, today. I'd best be going. See you around.
Mon Tues Wed
THURSDAY Fri Sat Sun
July 10, 2003 - Updates at 0650
The rains came back yesterday, just in time to destroy my home-bound commute. Not as badly as last week's storm, which turned my drive into a 2.25 hour piece of hell, though. Only about an hour and twenty yesterday. Yesterday's rain total isn't up yet on the weather watchers site - the beltway was being pelted while I was on it, though. I don't know how hard the rain was here, it had faded to a light drizzle by the time I got home.
Whatever Thompson's dog had managed to make its way up here. Sally was sick overnight. The first time was just before bed, and we cleaned it up. I sat up with her for a while, in case she needed to go out in a rush. But when we woke this morning, there was more to clean, on both sides of our bedroom. A quick pass or two with Resolve solved that, and she'll be going to straight dog food, light portions for a couple of days while we watch her closely. She did get to lick plates and bowls yesterday, but that's common. What was unusual in her diet was all the treats she had while at the vet's office. She was there for a 3 month post-stroke checkup. Everybody's happy ... well, everybody except Sally's tummy. She's doing really well, thanks.
Using prelink to speedup my Gentoo desktop last night turned out to be a mistake. It broke Evolution something fierce. Prelink is a tool that goes out and does all of the resolution that the runtime linker normally does, and writes all those offsets into every binary it can find (and is allowed to work on). When it's working great, it makes most programs load three or four times faster - it's quite perceptible, even on this fast dual processor box. But running prelink broke evolution. At first I thought I could get away with the easy fix, and just un-prelink the evolution binary (prelink -u /usr/bin/evolution). But nope. Then I tried recompiling and reinstalling just Evo. No joy - it's calling some other binary, which confuses Evo. So I just unprelinked the whole system. It's not like I need it to be faster, it's just that speed is well ... good. One day with more time on my hands I'll give this another try, and figure out which actual file(s) is(are) causing the problem, and just add it(them) to the prelink exclusion list.
However, that ate my evening. So I've nothing else of interest to report. You have a great day, okay? Later...
Mon
Tues
Wed
Thu
FRIDAY
Sat
Sun
July 11, 2003 - Updates at 0659
Good morning. There went ten minutes I didn't have... A productive day at the customer site yesterday. Router configured and happy, Samba next. I went to the Linux meeting last night, and that was good. Then, just after we dropped off to sleep, the Gods held their khyber-toss, NASCAR qualifiers, Metallica show. At least I think they did, I couldn't hear or see anything else underneath the huge electrical storm with crashing thunder that kept us awake for two or more hours... sheesh. I'm late and have to go. See ya!
Mon
Tues
Wed
Thu
Fri
SATURDAY
Sun
July 12, 2003 - Updates at ~1000
Good morning. The humidity is down in the sixties, the sun is shining, and I slept through the night and about half the morning (or more, it appears). That's not a bad thing - yesterday passed as in a dream. So I'd best complete my timecard for yesterday before the memory fades entirely...
Thanks for your patience... Actually, yesterday was a fairly productive day. We managed to successfully test (in one form or another) all of the links that will be used by the new network setup, including the NBase modules from MRV, the self-healing redundant fiber modules that talk gigabit Ethernet. Those remain slightly problematic, because we'd expect to see uniform error-free performance from both paths. Unfortunately, with any combination of working cables, the BoxA:Port1 to BoxB:Port1 path was unswerving in it's ability to deliver between 23 and 32 percent packet errors. Unplug those cables and put them in the Port2 path, runs like a champ. It's also true betwee each pair of the three modules we have on hand. So we did a bunch of permutations yesterday with these products, characterizing the issues so that our local fiber guy, who's back from vacation on Monday, can jump in with both feet running (hopefully).
To terminate the fiber I brought up a lab-spare Dell box, popped in a dual-fiber Intel server adapter card, installed Debian plus a 2.4.21 kernel and the latest e1000 driver from the Intel site.I put another of the dual Intel fiber adapters back in the router I've been working on. So our testing path was from router, via multimode cable to one of the NBase modules, out via dual single-fiber redundant path single-mode fiber to the other Nbase module, and then via multimode cable to the new test box I built. It was a great day because after lots of putzing about over wavelengths, cable types and diameters, terminating equipment and the like, we finally were passing packets end to end over the NBase modules. It's really too bad that we haven't got full functionality in the lab - Monday should finish telling the tale, however.
As Marcia prepares for the upcoming business travel week, we've got lots to do. There's a spot of cleaning here and there that needs doing, the lawn needs a firm hand applied, other chores are going to be attended to, as well. I'll be kept company during the week by Sally and Ebony. Eb's over to visit because Lee is going off to California with Marcia and Judy, and Jim is busy with something else, so we let the dogs keep each other company while their moms are out of Dodge.
I probably mentioned that I've decided to modify the Debian mailserver HOWTO I'm writing to use Cyrus-IMAP from source rather than Debian package. I just don't care for nor fathom how the maintainer has segmented things. So I'll roll my own for that part, and redo the front end portion of the document that's abuilding. Progress (and waiting) is. Meantime, have a great day!
Mon Tues Wed Thu Fri Sat
SUNDAY
July 13, 2003 - Updates in the morning...
We'll get to the Sunday Guest Head in a minute, but first, let's experience a little departure from reality together, shall we? I was reading an article about whisper numbers from Reuters, via Yahoo (link here) and I admit to being terribly confused. Here's the key section of the article:
Internet media heavyweight Yahoo Inc. (Nasdaq:YHOO - news) after Wednesday's close posted earnings that matched Wall Street's consensus estimates and hiked its forecast for the year -- only to see its share price skid almost 8 percent on Thursday.
Analysts raised concerns about the stock's value after the company's quarterly profit, which more than doubled from a year ago, met but did not beat lofty expectations. The stock has climbed about 99 percent in 2003 and rocketed almost 262 percent since bottoming out last September.
Correct me if I'm wrong, but let's look at this from a common sense point of view, rather than from whatever planet market analysts come from. The company has doubled it's profit for the most recent quarter over the same quarter a year ago. It did so through the bottom of a down economy. It made the numbers that it said it would. It made the numbers that the analysts agreed that it said it would. And then analysts smack the company's stock upside the head for not making liars of themselves and the company's estimate. Um, Earth to stock analysts: Is there any relevance between what you do and our planet?
If I remember correctly, it used to be the job of the CFO in a publicly held company to make an estimate of the company's profit or loss for the coming quarter, get the analysts to agree that was a good number, then hit it steady on, from a distance of 3 months. Don't come in over or under, because if you do, then it implied that you didn't know what you were doing, which is bad for the company.
Now, if I read my tea leaves right, there's absolutely nothing that a company can do to please analysts who can send the stock higher or lower without any reason whatsoever except... um, well, for no detectable reason at all. It used to be that a stock had value based upon company fundamentals. A "stock analyst" was someone who knew something about the larger market that a given company participated in, and could make buy and sell recommendations based upon "knowing" more about how the company's profits were going to be than the company's own people did.
Yes, in every real sense, that's fortune telling. But now, it doesn't matter what a company does, someone can be unhappy with the results and consequently send a stock reeling. And who's responsible for that? How about the stock analyst? Do you hear me out there, lawyers? I can hear the rustling in the brush around this fire each night. Forget stockholder suits against the company that did the best it could to maximize short-term shareholder value, at the expense of gutting the company's long term prospects. Sue the stinking analysts.
Now that's out of my system, today's Guest Head is Antony van Leeuwenhoek. He is popularly responsible for the "invention" of the microscope, but that's not true. The first compound microscopes predated van Leeuwenhoek's single-lens devices by several decades. What makes van Leeuwenhoek important is that the magnification power of his devices were an order of magnitude better than contemporary multi-lens designs. His second mark of distinction was that he observed carefully and documented well. With little more than a grammar school education, van Leeuwenhoek was elected to the Royal Society of London in 1680, on the basis of the letters he sent to the Society detailing his equipment and work. Antony van Leeuwenhoek was among the first to observe living single bacteria, as well as a stunningly large variety of other single and multi-cellular creatures. Nothing escaped his interest. You can start along several interesting paths of inquiry with this Google search on van Leeuwenhoek. Have fun!
The garden is sorta doing fine. The only real stars at the moment are the tomatoes, which are going great gangbusters, as you can see, left and right. Yep, that includes the first tomato almost ready for eating. The other plants (assorted squash, beans, scallions) really have yet to take off. And it's getting late - we'll see what happens.
Yesterday was filled with yard work, a short nap, dinner out at our favorite local Tex-Mex restaurant (On The Border), and some work up here with Greg on the new version of software for LinuxMuse. There is precious little to report on the day's events beyond that, so it will have to do. Today is likely more of the same - with some ETS website design work and software logo work based on Greg's specifications thrown in for fun. Shopping for groceries, of course, but then you knew that, didn't you? Have a lovely day. See ya!
Last Week << Mon Tues Wed Thu Fri Sat Sun >> Next Week
Visit the rest of the DAYNOTES GANG, a collection of bright minds and sharp wits. Really, I don't know why they tolerate me <grin>. My personal inspiration for these pages is Dr. Jerry Pournelle. I am also indebted to Bob Thompson and Tom Syroid for their patience, guidance and feedback. Of course, I am sustained by and beholden to my lovely wife, Marcia. You can find her online too, at http://www.dutchgirl.net/. Thanks for dropping by.
All Content Copyright © 1999-2003 Brian P. Bilbrey.
