layout hack
layout hackMain Pagelayout hack
layout hack
RSS Feed RSS Feed Main page
layout hack
layout hack
layout hackAbout Melayout hack
layout hack
Software I wrote
Resume
Friends of mine
Pictures
Musicianship
Stuff I have for sale
layout hack
layout hack
layout hackPersonal Newslayout hack
layout hack
2010:
March, April.
2009:
January, March, August.
2008:
Jan, Feb, Apr, May, July, August, September, October.
2007:
Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2006:
Jan, Feb, Mar, Apr, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2005:
Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2004:
Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2003:
Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2002:
Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2001:
Jan, Feb, Mar, Apr, May, Jun, Jul, Aug, Sep, Oct, Nov, Dec.
2000:
Jan, Feb, Apr, May, Jun, Jul, Aug, Oct, Nov, Dec.
1999:
Jan, Feb, Jun, Oct, Dec.
1998:
Jul, Aug, Sep, Nov.
layout hack
layout hack
layout hackGeek Stuff (computer related)layout hack
layout hack
Digital Music
Java
Why LiveWire Sucks
Why ASP Sucks (a bit)
Linux
MacOS
Unix
Oracle
Perl
Emacs
O'Reilly
layout hack
layout hack
layout hack(some of) My Interestslayout hack
layout hack
Humor
Sony Playstation
Cars
layout hack
layout hack
layout hackSearchlayout hack
layout hack

layout hack
layout hackAdslayout hack
layout hack



Valid HTML 4.01!

spacer
November 27, 2004

Not only do they make shockingly inexpensive carbon-fiber guitar and bass necks, but Moses Graphite makes really fancy necks too. This gun guitar might be more to your liking if you are in the G.I. Joe age group, or Japanese, or both.

I think I know why science education in the U.S. is falling behind other countries, and why hard science investment is waning in the U.S.: we as a nation don't like science, don't like its results, and want it to just go away. Why do I think so? I wish I had a subscription so that I could read Third of Americans Say Evidence Has Supported Darwin's Evolution Theory (Almost half of Americans believe God created humans 10,000 years ago). But I can get to an NSF science literacy poll (32K PDF), which reports that 67% of American adults don't think the Big Bang happened, 47% don't think humans evolved from earlier species, 52% think that humans and dinosaurs lived at the same time, 25% think the sun goes around the earth, 46% don't know that the earth takes a year to go around the sun, and 78% can't explain what a molecule is. I guess if you don't know any more about how the world actually works at a practical level than people did 5000 years ago, you shouldn't be expected to have any idea of where we came from, what we're doing here, or our place in the universe.

Microsoft has done a lot to make managing large numbers of PCs easier, via group policies and remote admin tools and the like. However, this sort of leverage can really backfire if the software is buggy or confusing to use, or if the operator doesn't know exactly what they're doing: DWP kills 60k+ PCs in Windows XP upgrade lash-up - oops.

Nobuo Uematsu, the composer for the Final Fantasy video game series, is forming his own studio. Awesome! Kim and I are tired of playing the Final Fantasy series (could have something to do with the fact that the games are fundamentally all the same, with different plot and characters and graphics and music, but nearly identical repetitive gameplay and skill/magic systems across titles). I don't want to miss out on his music but I'm not interested in any of the new games. I hope he does some music for a totally different game that might be interesting enough to play.

The Target Stores Wake Up Call service is the most retarded marketing bullshit I have ever seen. Ever. I sorta respected the idea of the wake up call... but then I saw the implementation. So, so lame. I'm sure the Franklin Mint / Mr. Winkle crowd will love it. Naturally, Flash is required to see the web site (see above, "retarded marketing bullshit").

The fact that Darth Vader is making wake-up calls for Target stores so that holiday shoppers can remember to get up early and fight to be the most materialistic, just goes to show yet again that George Lucas is a whore.

faisaljawdat (6:30:13 PM): lord vader, blue light special in aisle 5?
jamieflournoy (6:30:37 PM): *INHALE* *EXHALE*
jamieflournoy (6:30:57 PM): (cuts soccer mom in half with lightsaber) All too easy.
jamieflournoy (6:31:13 PM): JOIN ME and TOGETHER we could SAVE AT COSTCO
faisaljawdat (6:31:44 PM): i find your lack of savings... disturbing!
jamieflournoy (6:32:07 PM): ROTFL


November 24, 2004

When you buy a guitar or bass, chances are good that the setup will not be to your liking. When I say setup, I'm referring to the various physical tweaks that you can make to the instrument. A beginner might only think to tune the instrument using the tuning pegs and to change the tone with the various knobs on the front of the instrument, but that's only the beginning.

You may find that an open A string sounds good, but playing high up on the neck, the notes are all out of tune. You may find that the strings are very high off of the neck, requiring quite a bit of force to fret (which in turn makes it difficult to play accurately and in time, and which leads to rapid muscle fatigue in the finger flexor muscles in your forearm). You may find that some notes on the neck sound good, and others buzz annoyingly. These are all examples of an incorrectly adjusted instrument.

Every pianist probably shouldn't be expected to also be a proficient piano tuner, because tuning a piano is quite complex and time consuming. However, a guitarist (or bass guitarist) should learn how to set up his or her own instrument, because it has a major effect on how hard the instrument is to play, and whether the tone of the instrument is as good as it can be. What's the point of buying a $4000 instrument if it's going to be hard to play and sound muddy and out of tune like a $150 instrument?

Someday when you're rich and famous and can afford to take a dozen guitars on tour with you, you can hire a tech who has the whole professional tool set to tweak your axes just the way you like, to re-tune them every time the temperature or humidity changes, and to re-string them when you break a string or when the strings start to sound less than fabulous. Until then, you should know how to get your instrument pretty close to an optimum setup for your preferences. It'll make you play better and sound better.

The major physical things to worry about are neck relief, intonation, and pickup height. (Pickup gain and battery freshness matter too, but those are electrical tweaks.) I found a nice, simple, sparse, well illustrated setup guide for the Gary Willis Signature bass that explains things very well. Feel free to ignore the part about the "ramp", since that's a Gary Willis idea (unless you like that idea so much that you want to add one to your own instrument somehow).

Remember, if you change your strings, stay with the exact same manufacturer and style, or you may end up throwing your instrument out of whack. Why? Well, if you play light strings (like I used to), and switch to medium-light, they're thicker, so there's more mass, so the pitch is lower at the same tension. That means you have to tighten the tuning pegs further to get them in tune compared to the light strings, so the neck bows (unless it's made of carbon fiber, in which case it laughs at changes in string tension), and you have to adjust the truss rod. Even if you stay with the same gauge (.040"-.100", for example, is the typical "light" bass string range) but change to a manufacturer like DR (my favorite) who compresses their strings during manufacture, the mass of the string has still changed (which is what really matters) compared to an uncompressed light set, so you still have to tweak.

Pickup/electronics genius Bill Bartolini has a really good, if concise, page about mounting and adjusting pickups. It's intended for installers of replacement pickups but the summary for adjusters of existing pickups is clear: get your pickups as close to the strings as possible. That's what those screws are for - they're not just to keep the pickup from falling out; they're there for height adjustment too. Obviously if they're too close the strings will hit them and/or they'll get in the way of your fingers or pick as you play, so find a happy balance. But hey, in 30 seconds with a screwdriver you can improve your instrument's gain dramatically. After that, it's up to you to stop doing all those horrible squeaky sloppy things with your hands that you could never hear before. :) Gain is a double-edged sword.


November 22, 2004

I got a couple of new basses; pictures are here.

NBA brawl/riot video (WMV)

Tourist's Guide To Driving Around Washington D.C.

Canada Busy Returning Bush Dodgers

Mesh on Gavin and Kimberly

Making Sense of Java's Dates


November 20, 2004

I've worked at a series of tech companies in the Bay Area, ranging from tiny (<5 employees) to medium (>3000 employees) in size. Here, from the INSIDE of the hiring process, are two key things that folks looking for tech jobs need to know:

1) Hiring total strangers off the street is a royal pain in the ass for employers.

A few years ago the job market was so tilted in job seekers' favor that it was hard to find any qualified people, so there was the cost of the recruiter on top of the high employee compensation that comes from such a tight market. This was, I kid you not, sometimes 30% of the first year's salary. Yes. For an overpaid dot-commer out of college with a week of skimming Learning Perl under their belt, if that candidate gets hired at $100K/yr, the recruiter gets an additional $30K. So, lots of companies during that time offered referral bonuses of just a few thousand dollars, so that same gratuitously underqualified putz could then get his employer to hire his brother, girlfriend, roommate, etc and be rewarded handsomely for it. Shockingly, these companies were not successful in the long run.

Now the problem is just the opposite (too many job seekers for the number of available job positions), with the same result: extremely high overhead to hiring someone. There are so many job seekers that if you post a public job listing on Craigslist or Monster.com, you'll be inundated with resumes - hundreds per day.

In 2002 (when the job market was still extremely ugly for job seekers) I posted a carefully worded job posting on Craigslist with detailed and accurate skill requirements, as well as phrases like "you MUST have all of these skills or you WILL NOT be considered" and still got over 400 resumes in less than one business day. Most of them were completely underqualified and didn't even write the cover letter that the job description specifically stated was necessary in order for the application to be considered.

Many of the resumes you will get are written by job seekers who are simply lying about their skills, just hoping to B.S. their way through the interview, or maybe hoping to get a chance to be honest and plead with the employer to hire them anyway based on a claim of eagerness to work or to learn the skills required for the job. But employers (right or wrong) want to hire someone qualified for a near-market rate. They could hire you and train you, or train and promote from within, but there's little incentive for them to do that.

Example: $50K/yr employee gets $5K of training to do a $75K/yr job. (The company has invested $5K in that employee.) That employee can immediately leave for a similar job for $75K/yr. So, the company has risked $5K. Most companies I've seen would rather not try and come up with some kind of training lockup or probation phase to counteract that risk; they'd rather just hire someone from outside the company who already has that skill set. I'm not excusing this practice, just pointing it out. Personally I think the cost and risk of hiring someone new is higher than the cost of attaching some kind of golden handcuffs to employee training (i.e. if you quit within 6 months you owe the company $5K, or something similar).

Since there are so many applicants, it would seem that hiring externally is manageable if only there were some sort of filtering mechanism they could use. That's probably why employers inflate their actual requirements, and then recruiters pad the years of experience, and do keyword searches. That cuts down on the number of resumes that trivially match the job description, and is still legal.

This is all avoidable, though, if the manager who has an open position can just ask his or her employees if they know someone qualified. Instantly you reduce the pool of job seekers from 500-1000 to about 5. That's a number that you could reasonably phone screen or even interview. You get the bonus of having someone you already know who can vouch for that person, which is better than a random reference attached to the applicant's resume. If you ask your best rockstar employee if they know anyone good, you can bet that person will be good too, unlike some jerk who may well have padded their resume and fast-talked their way past the HR department's preliminary phone screen.

So, it's true what your parents told you: it isn't what you can do as much as who you know. If you're awesome but nobody knows you, you're much worse off than if you're mediocre and have lots of employed friends.

2) Being unemployed for months and months looks really, really bad to employers.

So let's say you got laid off six months ago. You have no job. None. Not even a part time crappy job, not even working for free (a.k.a. equity only) at a startup, not even delivering pizzas. Nothing. You've just been looking for a job Really Hard.

This looks very suspicious to an employer. Have you really spent 1000 hours (40h/w * 25 wks) looking for a job, and couldn't find one? No one would hire you? Or... you wouldn't stoop to take a job that isn't perfect for you. Either possibility looks pretty bad. Either you're unemployable, or you're inflexible.

It really, really helps you a lot to be employed. It means you're employable, reliable, etc. and that your layoff was a fluke rather than your creative way of saying that you got fired. It means that you can be flexible. Sure, you're not a chump, because you are looking for a better job, but at least you came up with a temporary solution to the no job no-monay problem.

The solution to both of these problems is to try and get a job at any relevant tech company that you can. You'll get a paycheck, which is better than no paycheck (unless the paycheck pays less than unemployment, but unemployment still runs out). You can still look for a better job, but you also MEET PEOPLE who will be references for you and may jump ship and then call you asking if you'd like to join them at their new company. Your skills will get freshened up; you might even get lucky and work for a company that trains its employees. Heck, you might even like it there and work on renegotiating your salary once they figure out that you're not a mutant. And finally, recruiters looking at your resume won't wonder what the hell is wrong with you that caused you to stay jobless for six months.


November 11, 2004

I'm selling my PowerBook and getting another just like it, except faster and with twice as much RAM. Software developers really do need fast computers, which I've been kind of trying to resist because I want a little bitty laptop. This is the best compromise I could come up with: a really fast little bitty laptop.

I ordered the new one on Thursday afternoon and it showed up on Friday afternoon. That's ridiculous. I was unprepared. My 1GB RAM upgrade didn't show up until yesterday. Let me be clear about this: if 640MB isn't quite enough, 256MB is really bad. Actually because the CPU is faster, the swapping wasn't as noticeable, but it was still bad. Once I got the RAM installed, it's just amazingly fast. Wow. TextEdit launches in one second. Safari launches in two seconds. The behemoth known as Eclipse that I actually do my coding in takes 21 seconds, which I will just tell you is a good thing. For reference, my wickedly-fast Linux server (Athlon64 3500+ CPU, mirrored ATA/100 drives doing alternate reads, ReiserFS, 1GB RAM) takes just over 10 seconds to do the same thing.

More importantly, when I use Eclipse and Mozilla (with a bunch of huge documents open in multiple tabs of multiple windows on multiple screens) and Acrobat Reader and Terminal.app (with 10 windows) and Fire and Stickies and Palm Desktop and LaunchBar and Emacs and... all that other stuff, it's actually fast, not beating up the hard disk swapping things in and out of memory and making me wish I had bought a desktop computer instead. I did get pretty good at using Eclipse on Linux via X11, but that was utterly horrible unless I was at home because I have DSL, and the upstream bandwidth is not sufficient for an interactive GUI session (even using TightVNC; see Chicken of the VNC below). It would be OK even if it took 5 minutes to launch all this stuff, as long as it's all really responsive once I get it all started and spend several hours switching around among all these apps. If I have to bail out of one of them to do the other, my productivity goes to about 10% of what it would be if I could just run it all at once and focus on the problem instead of the shortcomings of the tools I'm using to solve that problem.

BootCD is useful for situations like the one I had in which you may want to boot off of a disk other than your internal disk, because (in my case) you want to securely erase that disk. Well, BootCD works (provided you first get the latest version that actually is compatible with the OS you're trying to make a boot CD for), but it's painfully slow. It's even worse if you did what I did. I took the 512MB SODIMM from my old PowerBook and put it in my new PowerBook in hopes that the new one would just slow itself down and work anyway. It didn't. So the old PB had 128MB RAM in it, and as I lamented earlier, the new one had 256MB of usable memory in it. Guess what? When you boot from CD, there's no swap file. When you have no swap file and 128MB RAM, it sucks. So. Damn. Slow. It improved a lot when I put the 512MB module back in the old PB and rebooted.

I learned something else in the process: it takes about 28 hours to securely erase a 4200 RPM 60GB hard disk. Formatting erases almost nothing. Zeroing the whole disk once doesn't do much either. Gutmann's method, now eight years old and recognized by Gutmann himself as obsolete on modern drives, performs 35 passes, which would have taken even longer than the seven that I settled for. I don't have state secrets, just financial data and passwords and stuff. Fortunately I didn't really need that laptop for anything so it was OK that it sat there for over a day erasing itself. Anyway, if you need to secure erase a drive, be ready to wait, and wait, and wait. It's not actually very taxing on the host computer to do this; the erasing programs I've seen all bypass write caching as much as possible, so in terms of CPU and I/O bandwidth they're not working very hard. I suspect that if there were a reliable API at the operating system level, it would be possible to do this much faster, by using some kind of "write one pass with caching" scheme followed by a sync to disk. I think what they do instead is to bypass caching entirely, and very slowly write a small chunk of data, sync it, and repeat. I've read that drive manufacturers are working on a much lower-level standard for asking a drive to securely erase itself, which would be great; that could be incredibly fast if implemented correctly, compared to the current way that software seemingly has to work in order to guarantee that each different erased pattern gets all the way to the disk.

Jason inspired me to do something a bit clever: I went back to using the old laptop but with the new laptop as a FireWire Target Mode external disk, so I didn't have to copy all the data back to the old laptop, and later sync it back to the new one's drive, or do some kind of similar hard disk switcheroo. (The 12" PB G4 hard disk replacement process is not easy.) I really should have tested the RAM in the new laptop before switching over my data, but I didn't, so this was the best thing to do to hold me over while I waited for the RAM to show up.

Funny links: Sorry Everybody, FireWire Dino, Llama Llama Duck, Fuck the South.

Funny in name but serious in purpose is Chicken of the VNC. Perhaps the best part of this VNC program for the Mac is the fact that it works well and does TightVNC encoding and all that, but that's not very funny. The icon, however, is funny, and the documentation is even funnier: screen shot (87K JPEG image). "brawwwK!" But wait, there's more. Google brawwwk. 64 pages? I had no idea that onomatopoeia had reached such epi[demi]c proportions.

Next week I'll be performing on stage for the first time in about six years. The ensemble I'm in at school is pretty good, and I've had fun so far, but the original drummer flaked out and hasn't shown up to several of the ensemble's weekly rehearsals. The instructor who's in charge of the ensemble had to substitute another student basically at the last minute. He was literally walking past the room we were practicing in, and she asked him to come in and sit in with us. The poor guy has 6 days to prepare, and has no time to prepare, really, so he's basically going to stop by after class and immediately perform in front of an audience. Balls! Damn. He did a good job sitting in with us in rehearsal... I would be scared out of my mind if I had to do something like that. He doesn't even really have to do it, he just agreed when asked. Wow. Next time I attend one of these rehearsals I'm going to bear in mind that something silly like this very well might have taken place behind the scenes.