Ongoing Discussions

The comment system here is definitely turning out to be a worthwhile addition, generating responses on quite a few articles. The most interesting of these so far has been in response to my Fibbertarian Economics article, in which Derk Lupinek (of Mind Floss) and I are having a pretty spirited discussion about the principles involved in sustaining or restraining the upper-upper class. Some readers might be surprised to see which side I’m taking.

Thoughts About Work

OK, I’ve been at the new job about two full weeks now – long enough to figure out what it’s going to be like over the longer term. Here are some observations in no particular order.

  • It’s a complex and specialized product, but I’m fortunate enough to have the right background to understand the parts and I think I’ve managed to get a pretty clear picture of how they fit together already. There are still some good-sized conceptual/algorithmic problems for me to sink my teeth into, but the basic architecture seems sound.
  • The people I work with are even cooler than I expected. I knew many of them and thus had some idea (or else I wouldn’t have come here), but people change and there were many I didn’t know so it’s nice to get beyond that “getting to know you” stage and not find any interpersonal glitches.
  • I’m loving the short commute.
  • I’m not wild about being in a cube (I don’t rate a full office, of which there aren’t very many) but they’re actually pretty good cubes with high walls and good sound characteristics. It’s actually quieter than my old office at EMC was. I have put up my New Zealand flag, and many of my trademark accessories (the Flying Kiwinis, dot-com toilet paper, Valdez Penguins, desktop catapult) but the moo-bag is still at home. I can’t remember where I put my calendar, but someone else has one so it’s OK.

I’ve set up an in-house instance of the same weblog software I use to run this site, so I can “think out loud” and collect other people’s input on my ideas. It’s a kind of writing that I’m comfortable with, and the weblog/comment environment offers some advantages over the typical email atmosphere, so I figure it’s worth trying.

Serendipitous Humor

The weekly Arcata police log is written in a style that just makes me grin every time. The following excerpt from this week’s edition was particularly good:

2:01 a.m. A young man and some potted plant friends strolled westward on Sunset Avenue, inviting inquiry. But when a police car entered the picture, the lad went deciduous, dropped his leafy charges and took off running down some handy nearby train tracks, disappearing like the freight trains which once plied the rails.

“Went deciduous”? If I ever happen to be in Arcata I’d like to buy that cop a drink, or a donut, or whatever, just for that phrase alone.

Got game? Guess not.

Everyone’s favorite EMC VP continues to show that he’s a little out of touch about the current state of technology. Chris Gahagan has been quoted several places as saying:

A lot of these vendors that are proposing virtualization say, `You basically put JBOD behind the virtualization appliance,’ but that doesn’t work for a whole lot of reasons. If you look at the DataCore product, [for example], it can’t scale because it’s based on off-the-shelf PC technology… .If you wanted to put DataCore on a fully loaded Symmetrix, you’d have to put two dozen DataCore boxes in front of the Symmetrix just to manage the I/O load. That doesn’t seem like a good consolidation strategy.

So it’s bad because it’s based on off-the-shelf technology? Do you know much about the history of RAID, Chris? Aren’t some of EMC’s products – e.g. Celerra – basically off-the-shelf components behind the fancy nameplate? What ever happened to “it’s the software, not the platform”? Scalability (or lack thereof) does not reside in the components; it resides in how the components are combined and used.

The quote is making the rounds because DataCore has responded by challenging EMC to a duel with a Porsche 911 Turbo as stakes. (The story was also reported on Byte and Switch, but those folks are effectively a mouthpiece for EMC’s competitors and wannabes so I wouldn’t look to them for the truth any more than I’d look to EMC’s own marketing department.) The duel looks interesting. I admit that I’m a little skeptical of DataCore’s claims, because the Symmetrix is still a pretty big piece of muscle even if it’s not in the shape it used to be and doesn’t have the brains it should, but it’s within the realm of possibility that they really can pull it off. Even if they “lose” it’s worth it for them. They get lots of exposure, and a victory for EMC at 5x the cost would be Pyrrhic indeed. Not surprisingly, because not everyone there is stupid, EMC is aware of this and has declined the challenge. That’s a shame, but maybe some people at EMC will be encouraged not to make statements they’re not willing to back up.

Only My Scorn is Inevitable

Apparently a California court has soundly rejected the doctrine of “inevitable disclosure” for trade secrets. I first heard about this idea in Novell v. Timponogos (which the new decision cites) while I was at Dolphin, because we had relationships with some of the people involved. The doctrine says that if I worked on designing frobulators for employer X, I can’t just forget what I know about frobulator design – including potential trade secrets in that area. If I then get a job designing frobulators for employer Y, according to the doctrine, it can be assumed even without specific evidence that I’m using knowledge that is the property of X.

It actually does make sense in a certain context, if we assume that everything I learn at X is the property of X. Many would reject that axiom. Many more might object to this doctrine on the basis of assuming guilt. For example, my frobulator design at Y might be based on an entirely different set of principles than those I designed at X, and involve none of what I learned at X, and yet the doctrine of inevitable disclosure allows X to assume otherwise. The idea of a legal doctrine saying that wrongdoing is “inevitable” under certain circumstances is quite scary.

Of course, the decision only affects California, and was partly based on the reasoning that the doctrine is being used to create de facto non-compete agreements in a state where those agreements are strictly limited when explicit. That means it’s a very small victory, limited in both conceptual and geographic scope, but it’s a victory for employees nonetheless.

Barn Doors and Missing Horses

After all the fun I had with Verizon blithely allowing JTLnet’s worm to spoof email from me, it seems that Verizon finally Got It ™. Today I got mail informing me that they’re going to require SMTP authorization, which is exactly what they should have done before to prevent such spoofing. I can’t help but wonder how many incidents like mine must have occurred to bring the issue to their attention.

Somebody Else’s Problem

There’s a story on Wired about an FBI child-porn sting. It’s pretty disturbing stuff all around, from the sympathetically-portrayed protagonist who might actually be guilty to the disingenuous way the FBI agents claimed “not to know” certain information that might have limited the scope of their investigation. Most people will notice those. What’s less obvious, though, is the problem with the following passage:

If viewing child porn online acted as a gateway to molestation, an explosion of it on the Web should have triggered rising rates of child sexual abuse. And it has Â? in the impoverished countries where most of the illegal material is manufactured these days, such as Thailand, South America, and Eastern Europe. In the US, however, where the contraband images have their largest audience, rates of sex crimes against children are falling sharply.

The author is trying to address the claim that passively viewing child porn inexorably leads to active exploitation of children, and s/he concludes that it does not…in the US. What’s missed, because it doesn’t matter to the American writer, is that sexual exploitation of children worldwide is increasing. It’s good old supply and demand. The web has increased demand, and someone somewhere is responding by increasing the supply. For there to be new pictures or videos of naked children for people in the US to view, there must be actual naked children being filmed or photographed somewhere in the world. Is that not a problem? Apparently not, to the Wired writer.

I’m not going to say that the fact of exploitation abroad actually justifies prosecution of US citizens who merely clicked on a few links on the web. There’s a whole morass of constitutional, moral, and other issues involved. There are arguments to be made about targeting supply vs. transport vs. demand, as has been done with regard to the drug war. The only thing I’m trying to address here is the way that exploitation of children outside the US is treated as though it doesn’t even matter by the Wired article. It’s not that the problem was considered and balanced against other concerns and determined to be outweighed. It wasn’t even put on the scale. As too often happens in US media, all that suffering was deemed to be without weight because it happens Somewhere Else.

Good Screws

Bram talks about different kinds of screws today. If you find that interesting, I suggest One Good Turn as a good bit of light reading. It’s quite amazing how much was involved in inventing the screw and screw-making equipment, and the book also describes several different kinds of screws. I’m actually rather partial to the Robertson screw, which uses a square driver head; somehow it just seems like it’d be the hardest of all to slip or strip, although I can’t quite explain how the angles and such make that so.

Fibbertarian Economics

If you’re coming here from Derk Lupinek’s “Mind Floss” site, please be aware that he has so severely misrepresented my views that even I would argue against the version he presents. Try reading my views in my words instead of his, and I think you’ll find that I’m hardly the sort of Diana Moon Glampers that he makes me out to be.

Steven Den Beste’s Rush Limbaugh impressions are starting to annoy me. In one of yesterday’s entries he quotes the following, allegedly from a correspondent named Nathan:

One thing has bugged me for the last few months. To the best of my knowledge, no one has ever explained exactly what the problem is with a widening gap between the rich and the poor. The only thing I can see as a possible argument is perhaps the poor will be resentful. But as long as the income of the ‘poor’ continues to rise in real dollars (you can have an air conditioner, 2 TVs, a car, cable TV, and Air Jordans, and still be below the poverty line), I don’t see it as a bad thing that we have a collection of capital in the hands certain individuals. Accumulated capital is a prerequisite for many of the great accomplishments of any century. The government has the greatest accumulated capital, since they can accumulate from everyone, willingly and unwillingly, and at a greater rate. Therefore, the capital the government gathers should be used on projects for the use and good of all (it IS good for everyone to not have children starving, insane people running around, etc). Accumulated capital in private hands can fund private enterprise. Our economy and nation would collapse without this situation.

I have thought, pondered, and considered, and cannot think of any possible reason anyone would hate this (except for knee-jerk liberal reaction); after all this land was built on the prospect of opportunity. Right or wrong, there is an implied lack of limits on that same opportunity. I feel if I can understand the logic behind it, perhaps I can argue against it, or maybe even be convinced.

Now, I think “Nathan” is missing a couple of points, which I’ll get to later, but first I want to deal with Den Beste’s response, which follows:

Do you believe in freedom, or equality?

That’s it. That is the single most fundamental question everyone has to answer, and everything else they believe in will be based on this answer. Within this question, “equality” means “equality of results”.

Right off the bat, Den Beste shows his talent for equivocation. Even as he claims to be using “equality” to mean equal outcomes, he’s clearly arguing against equal opportunity. It’s a slimy start.

You can talk about “leveling the playing field” all you want, but even in a situation where absolute everything has been evened out and the playing field is undeniably level, once you let people freely compete, inequalities will emerge and become greater and greater, and the only way to prevent that is to continually meddle with the situation to undo inequality as it spontaneously arises.

First, Den Beste assumes that the tendency for momentary advantage to increase and become permanent is inevitable, even though the degree to which that happens is in fact very dependent on how “the game” is set up. Some systems are subject to such “tipping points” and some systems are not. We as a society have some say over which type of system we live in.

Den Beste then goes on to use the emotionally-loaded term “meddle” to describe what is necessary to keep the system in equilibrium. Is it “meddling” for a referee to enforce the rules of a football game? Obviously not, but somehow letting anyone but the plutocrats have a say in the economy is meddling. One wonders by what twisted logic he can call that meddling, and not call it meddling when the US takes upon itself the responsibility to enforce UN resolutions in ways that the UN itself deems inappropriate. That should be fun to see.

Those who are complaining about the “widening gap” are, ultimately, opposed to freedom. They believe in equality, and if you make that axiomatic choice, then the current situation is wrong on that basis.

And thus Steven presses a false dichotomy into service. Note also how he has by this point abandoned the clarification on what he means by “equality” so that his equivocation can be more effective.

Enough of Steven. Too much, in fact. Let’s go back to Nathan (if, in fact, there is such a person). The crux of Nathan’s error is this:

Accumulated capital is a prerequisite for many of the great accomplishments of any century.

Really? What great accomplishments have had the accumulation of private capital as a prerequisite? This argument can be stated more simply as “big stuff costs money” and could just as easily be used to justify socialism as plutocracy.

But that’s not even the most significant missed point in Nathan’s argument. What great accomplishments have always depended on, even more than capital, is individual motivation – in particular the motivation to innovate. Drastic income disparities reduce that motivation. In the case of general motivation, the problem is despair. Why bother working harder or longer if you won’t be able to get more than a smidgin ahead anyway? The motivation issue is behind the failure of communism, but can just as easily affect a plutocratic system where outcomes are guaranteed to be equal (or nearly so) for the non-privileged majority. Equal opportunity provides motivation; equal outcomes destroy it.

In the case of the specific motivation to innovate, there’s a different problem. Innovation requires time and energy, and often money. People who are one step away from poverty can’t afford to invest any of those resources in pursuing their creative inventions. Even people who are reasonably comfortable can’t afford to risk seeing several months’ work turn into nothing, and will tend to stay at their boring non-innovative jobs. Deliberate public efforts to find and nurture innovators (more meddling, I suppose) don’t seem to work all that well. The most effective method seems to be to allow the existence of a large middle (or perhaps upper-middle) class of people educated enough and comfortable enough that they feel they can afford to take a few risks bringing their ideas to fruition. That large group cannot coexist with a permanent super-rich upper class; there simply aren’t enough resources to support both. Entrenched class systems also inhibit the social mobility that allows motivated and creative people to move into that necessary middle class.

The net result of having a great gap between rich and poor is stagnation. Disaffected workers have no incentive to work at peak efficiency. Innovation, which by its nature tends to spring from unexpected sources, stops because the vast majority of people capable of serious innovation lack the resources to pursue their ideas. That’s sure as hell not the kind of economy/society I want to live in. The only reason Nathan or Steven think that’s a good world is because they have never actually lived in it. They are, undoubtedly, the ungrateful beneficiaries of a system that is most definitely not structured according to the principles they espouse.

Email Go Bye-bye

I am now officially angry at Microsoft. My new laptop has a Firewire interface, and I’ve been using a portable 30GB for a while now as a backup medium for all of my machines. This drive seems to work fine with the new laptop to copy files around in Explorer, play video files, etc. However, when I try to do a backup to it using Retrospect I get all sorts of of disk errors. Because the I/O is all done through the paging system (which I have always regarded as a mistake, on any platform) this causes the system to lock up. When it did this the first time, the only way I could unfreeze the system was to remove all power. When it came back up there were no notices of filesystem problems, but all of my email folders were corrupted. They were there, all right, just as large as they had been before, but Outlook Express (which had created them and been the last one to write them) couldn’t figure them out and thus refused to use them. I’ve learned a few tricks to do with recovering email folders, but these ones seem beyond my capabilities to repair.

Why do I blame Microsoft? The drive works fine on three other machines, so it’s not the drive. Retrospect works fine on three other machines, so it’s not Retrospect. There could be a driver or a laptop-side hardware bug, so perhaps Fujitsu deserves some share of the blame. However, Outlook Express had been shut down for quite a while before the backup. Nobody was writing those files, and any flushing that needed to occur should have done so already. There’s no excuse for OE not being able to read those files, no matter what happened during the backup and subsequent reboot. There’s no excuse for the reboot being necessary in the first place, but that’s quite likely a whole separate error on Microsoft’s part.

Don’t tell me that this wouldn’t have happened under Linux, either. I’ve had very similar things happen under Linux, specifically an XFree86 video-driver install that ultimately led to a totally trashed root filesystem and a reinstall from scratch, so don’t hand me any of that garbage. On this particular day it just happens to be someone from Microsoft who has earned my wrath and contempt. It’s really pathetic that they managed to screw something up so that doing a backup destroyed my data.