So we just went through a terrorism alert during which the government prompted millions of Americans to buy massive amounts of duct tape, and the nation's economy is a mess. What to do?
According to the Wall Street Journal, of all places, the answer is clearly that we should all start selling prom dresses. American ingenuity is a wonderful thing....
Am I the only one who finds the Counting Crows cover of Joni Mitchell's "Big Yellow Taxi" to be far worse than the version Amy Grant did in 1994? I'd actually never heard the song before Amy Grant's version (and didn't even realize it was a Joni Mitchell song until just now), and now that the Counting Crows remake is getting radio time I wish they'd drop it and replace it with Grant's.
Mr. Rogers passed away today, at the age of 74. Of the various comments I've read, I like Judi Sohn's the most. Like her, some of my earliest memories are of watching Mr. Rogers and enjoying the time I spent watching the show. It'd be tough for a similar show to land on TV today, and for kids growing up now, that's a shame.
The Phillies played their first spring training game of the year today. They beat the Pirates 6-3.
Yesterday I was annoyed because Major League Baseball's Gameday Audio page was still talking about the World Series and didn't even mention the 2003 season. Surely there must be at least a few other people in the country who want to listen to spring training games. And what do you know -- today they updated the site for this spring, with a note that all spring training games will be available for free. I unfortunately didn't manage to see that before the Phillies game had ended, but now I know what I'll be listening to in the afternoons at work until October.
We just saw a flash of light outside and then a really loud boom. When the sound stopped a second or two later, a car alarm was blaring in the parking lot.
An explosion? Nope. Just lightning and then some thunder. Any thunder that can set off a car alarm is pretty strong in my book. Wow.
The O'Reilly Network just posted a vi tutorial by Wei-Meng Lee. As much as I can't stand vi, I've needed an article like this for a long time. There are certain situations in which neither BBEdit nor emacs are available -- a newly-installed Solaris system, for example -- or when I don't want to deal with emacs' annoying backup files, and I have to resort to vi at those times. I've learned enough to get by with vi, but now that I've read the article I'll be a bit more comfortable in it.
My great aunt Ruth passed away last night after a long illness. She was my mother's father's sister, and for years, when we all lived in the same area, I saw her and my great uncle frequently. I haven't seen either of them since they moved to Florida a few years ago (and I to California), but I certainly missed them. I'm saddened to learn that I won't have a chance to see Aunt Ruth again. All of my memories of her are good ones, and I'll treasure them.
Ruth Stolber, may you rest in peace. We'll miss the interest you took in all of us and the joy you brought to our lives.
Word came out today that Microsoft is buying most of Connectix. I've known about this for a while -- I overheard rumors of it before moving from Washington -- but I'm still not sure what to think about it.
Virtual PC is almost an afterthought in the purchase. I can't imagine that Microsoft cares about it all that much, and it wasn't a reason for the transaction. Instead, Microsoft saw the demand for virtual server systems like VMWare's and wanted to integrate that technology into Windows. I suppose VMWare was probably too expensive, so Microsoft settled for Connectix' solution. Virtual PC, meanwhile, came along for the ride.
On MacSlash, Microsoft's Dan Crevier mentions that the Virtual PC team will now be rolled into the MacBU and will report to him. That's good news -- Dan and his team are terrific.
I wonder if they'll get management's support in the long run, though. I don't really see what the incentive is for Microsoft to continue to produce a high-quality PC emulation product for the Mac. True, each sale of Virtual PC with Windows includes the full price of a Windows license, but the purchaser then runs it on a system on which Microsoft is unlikely to earn additional revenue from other pieces of software. I also wonder how many people buy Virtual PC with DOS and then install a creatively-acquired copy of Windows.
Personally, I just hope that any future versions of Virtual PC include the support I've come to depend on for installing non-Windows operating systems. At various times I've installed OpenBSD, NetBSD, and Linux on Virtual PC. It's far more convenient to install them via VPC than to sacrifice a computer for the same purpose.
It's also worth mentioning that if Microsoft decides to discontinue VPC for any reason, chances are that someone else will step into the market with a replacement product. The Mac user base has supported at least one PC emulator for the past six or seven years -- Insignia first, then Connectix when Virtual PC proved to be better than SoftWindows -- and I can't see that changing any time soon. Unless Apple increases its market share by a factor of 10, there are always going to be some number of Mac users who are required to run one or two Windows-only pieces of software and who are willing to pay a reasonable surcharge for doing so. If Microsoft provides the solution, that's terrific, since they know Windows better than anyone else. If they don't, though, I'm sure someone else will.
Nick has found a reason for me to keep my Washington voter's registration for a little while longer: Washington ballot initiative 831. It makes a statement about the problems of the initiative system and the idiocy of Tim Eyman at the same time. I'm very impressed.
Isioma Daniel, whose column three months ago set off religious riots in Nigeria, tells her story in the Guardian. "Nothing justified a religious group killing people simply because they considered a remark offensive. Who did they think they were? God? Surely God can fight his own battles." (from Neil Gaiman)
On Saturday night, Dan Gillmor broke the story that Google is buying Pyra Labs. Pyra is the company behind Blogger, which is the most widely used weblog software.
I have to say I'm a bit confused by the purchase. It's great news for Evan Williams and the rest of Pyra, who were struggling to make any money despite having over a million registered users. They suffered from the same problems as many other Internet software companies -- it's easy to get people to register, but it's darn near impossible to get them to pay. Especially when the same product can be had for free elsewhere.
So my congratulations go to Pyra, but I'm a little bewildered by what Google gets from this. Many people are saying that they now can update their index immediately when a Blogger blog is updated, but they didn't need to buy Pyra to do that. All Pyra needed for that was an API into Google, and that's trivial for Google to do at a far lower cost than the low amount they likely paid for the company. Heck, Pyra probably would have paid Google for that kind of connection if Pyra had any money.
Instead, Google gets ownership of a growing community of content creators. That's great, but Google doesn't do content creation; it does search. What do you do with a community of creators of esoteric content? In Yahoo's experience with Geocities, not much. In Homestead's experience with its products, not much. Ditto for Tripod, and for every other free web page company. Yes, weblogs get more repeat traffic today than personal web pages did in 1998, but traffic to weblogs isn't going to improve Google's search capabilities.
The only thing which comes to mind is that Google will now sell the advertising on Blogger sites. Google's AdWords program has shown that they can make money off of ads better than just about any other company, and now they have about 200,000 extra sites (the active Blogger pages) on which to practice that trade. Of course, it's not clear how well webloggers will appreciate having AdWords-style ads on their sites, since the ads will likely be keyed to each page's content. I don't know that I'd want any company's ad running across the top of this page, and I certainly wouldn't want ads from Apple, Sun, or Microsoft here.
I'm not sure what else Google might have in mind, unless it wants to move into content creation. If that's the intention, I don't think it's likely to be a successful strategy, since you can't mobilize a community whose only commonality is the software they use. Either Google has thought of something that nobody commenting on this transaction has in mind, or in a couple of years they're going to wonder why they saddled themselves with a million customers when they'd never dealt with normal consumers as real customers before.
A few days ago, Robert Cringely wrote a piece called "Sunset" about the troubles facing Sun Microsystems. Sun's in trouble, and it's interesting to think of why it's in trouble and how the company can get out of it. Cringely does a good job of both.
Sun and Apple are very similar companies. Both have survived for years on high-margin niche markets, differentiating themselves by their avoidance of Intel processors and one other major factor -- usability for Apple and reliability for Sun. In the past few years, they've faced similar challenges -- Intel's massive R&D has made the computers from both companies seem slow, and at the same time both have been attacked from the bottom. Apple has had to fight for its share of a consumer market that increasingly tilts towards low-cost (and therefore low-margin) machines, while Sun has battled against the lower-priced alternatives of Windows NT and Linux that can finally compete with Solaris in reliability and performance on the low end of the market.
With all of this, though, Apple is doing well and Sun isn't. Why is that? Apple has managed to transform itself to match the times. Usability used to be the primary reason why people bought Macs, though the hardware didn't stand out in a sea of beige boxes. Now it's design -- the iMac and Titanium PowerBooks catch your eye and make you wonder what's inside. At the same time, Apple has shifted its market, losing market share in education but picking up a number of Unix users who wouldn't have even looked at a Mac a few years ago.
Sun, on the other hand, hasn't really changed at all. The company doesn't have a response to Linux, and it can't present a compelling hardware offering for the low-end market. As Dell and others have driven prices for low-end servers down, Sun has been pushed into a high-end market, where it can wait its inevitable swallowing from below. Java won't help, either; a cross-platform runtime has never made much sense for Sun. Java runs better on Windows and Linux than it does on Solaris, so why would you buy Sun hardware for it?
I wish I could think of something Sun could do to extricate itself from this hole. It seems like every direction is blocked by some combination of Microsoft, IBM, Dell, and Red Hat. Sun needs to find a market in which they can be the best, in which business customers will be willing to pay more for superb hardware and software. I'm not sure where that is, but I wish them luck in finding it.
Today's Philadelphia Inquirer includes a cover story on the budget crisis in Oregon, which has the worst shortfall of any state but Alaska. The problems in Oregon and the way in which the state is dealing with them are simply horrifying.
I have a tough time understanding how states ended up in situations like this. We recently finished an era of prosperity, in which tax revenue increased dramatically as the economy boomed. Anyone with any sense of economics could tell you that the economy can falter as well as it can grow. It's sheer idiocy to take the money raised in good times and spend it with nary a thought of whether you'll need it later. And now we're paying the price nationwide for that shortsightedness.
You'll note, though, that the budget crises are worse on the West Coast. Why is that? It's largely because California, Oregon, and Washington, among others, have a political system that makes it easy for random initiatives to appear on the ballot. The theory is that the government's actions should be dictated by all of the citizens, so all voters can determine how certain issues are handled, or even whether to raise or lower taxes. It's absurd, and the current situation is a fine example of why it makes no sense.
Most people don't vote for tax increases unless the increase will directly benefit them. That's just the way it works. And voters simply can't reason out all of the issues on any given topic before voting for it. Convincing an entire state, then, to vote for a tax increase is nearly impossible. It's far easier to convince a majority of a legislature.
Of course, that's the argument for ballot initiatives in the first place -- that politicians spend money far too freely to be trusted to have control over it. At the same time, though, politicians can also be trusted to generally pass reasonable laws. They're not going to triple the income tax, and they're not going to eliminate all taxes. As a group, they recognize that the government has to run at least somewhat effectively. They might not do a great job of it, but they certainly have more time to think about how to make it work than most voters.
When things are thrown to the voters, though, you get ridiculous measures like the initiative in Washington to lower all vehicle registration fees to $10, regardless of the size of the vehicle. An 18-wheeler would be charged as much as a Honda Civic. I believe that one passed, and now it's tied up in the courts as the legislature tries to find a way around it. You get Floridians forcing all schools in the state to enforce class sizes that the state can't possibly pay for. And you get Oregonians turning down an income tax increase that would allow their government to at least function minimally.
There's a reason why we elect representatives. They're supposed to represent us. We're supposed to choose those who can solve problems, who can make good decisions, and whose decisions we generally agree with. When government is forced to cater to the whimsy of millions, it becomes unmanageable. That's today's Oregon, and I'm afraid it won't change any time soon.
Brent Simmons' excellent NetNewsWire hit version 1.0 a day or two ago. It's so nice to have software on the Mac that's far and away better than anything that exists for Windows. That doesn't happen very often, but NetNewsWire is a fine example.
I should also note that despite all of the criticism of the Dock in Mac OS X, one of NetNewsWire's best features wouldn't exist without the Dock. Whenever there are unread posts in my subscribed feeds, NNW's Dock icon updates with a little red badge showing the number of unread posts. It's incredibly convenient to be able to tell when there are new articles just by idling glancing at the side of my screen. There's simply no equivalent to Dock icon badging on Windows or Mac OS 9. Whatever else you might think of the Dock, that's one very nice feature.
I was the first person from the Rotor team to leave Microsoft; today, Dan Gillmor reports that David Stutz is the second.
Of course, David's departure is for reasons very different from mine. He had been at Microsoft for a long time and done quite well for himself. Over the past year, he spent an increasing amount of time traveling, growing wine, and singing (he's quite a good singer), and now that Rotor has reached 1.0, I'm not surprised to see him devote all of his time to non-MS things.
One of the strengths that David brought to Microsoft was the willingness to advance radical -- almost heretical -- viewpoints with upper management. Without him and Geoff Shilling pushing the lawyers and management, I doubt Rotor would have happened. It certainly wouldn't have been as open as it is. But David knows, as many of us who understood the open source community do, that Rotor and the other existing shared source products aren't nearly enough for Microsoft to regain the attention of the customers who are now looking seriously at open source solutions. And so there's one more memo, a sanitized version of which appears on his web site, describing where Microsoft should go. For anyone interested in how Microsoft can be successful in a world of increasingly commoditized software, David's thoughts are well worth reading.
I'm sorry for Microsoft to hear that he's left, but I'm happy for him, since I'm sure he'll love every minute of life as a farmer and singer. Hopefully I'll have a chance to head up to Washington and try some of his wine someday.
Today's Philadelphia Inquirer has an interesting story on the slow but steady decline in check usage in the U.S. More and more people are switching to other means of payment, it seems.
In my case, I've never made it through a box of checks before changing my address. And I spent five years at Stanford with the same address...but I didn't use enough checks to get through the entire box. These days, I use even less, since everything that isn't on my credit card I pay via Wells Fargo's online bill pay. The Inquirer article mentions that people still pay their credit card bills with checks, but I don't even do that -- the bill just gets automatically deducted from my checking account (and the best aspect of that is that they don't take it out till the last day it's due, so I have my money for a while longer than you might expect).
I still use one check a month, which pays the rent. The penalty for not paying that on time is so high that I won't trust Wells Fargo to get the payment there in time. I'm sure that'll change some day, though, and then I'll be just about completely done with checks.
I'm inclined to agree with Brian Palmer that there are worse reasons to choose the BSDs over Linux than this one.
Aaron Swartz has discovered the joy of writings that explain a variety of concepts to smart readers. The best authors for the works he's talking about don't talk down to their readers by adding in complicated jargon or concentrating on irrelevant details. Instead, they converse with their readers, leading them gently but intelligently on a path to understanding.
I haven't read much of Isaac Asimov's non-fiction, but I'm inclined to think that he might be among the best "teaching" authors. Asimov wrote thousands of columns and probably over a hundred non-fiction books, all targeting intelligent readers. He emphasized the important concepts and bypassed trivialities, taking pride in his ability to explain things that he had to learn before he could write about them. I intend to read a lot more of his non-fiction when I get a chance, and I think Aaron would probably enjoy Asimov's work as well.
I just finished reviewing a book proposal for a publisher. It's always flattering when someone asks me to do something like that. It's their way of saying that they think I know something about the topic, while I just find it hard to believe that at my age anyone considers me an expert on anything.
Book proposal reviews are an interesting challenge. If the proposal is good, it's easy -- just tell the publisher that everything looks great, wait six months, and read the book when it comes out. But when the proposal misses the mark, it's not easy. With technical books, especially on the kinds of obscure topics that I enjoy, the publisher doesn't necessarily have the depth of knowledge to determine if a proposal makes any sense. So part of my job as a reviewer is to help the publisher decide if the book is worth publishing, and if it'll be a good book if it's published in the proposed form.
That means that if the proposal isn't very good, I should say exactly that so the publisher doesn't waste time and money on the book. At the same time, though, my comments go to the author. It's never much fun to receive sharply critical comments from someone you don't know very well. It's also not fun to write those comments and realize that I don't know how the author will take them. I try to be very careful with how I phrase my criticism, since you never know when they'll come back to haunt you.
That's why I try to concentrate on technical and organization comments (and on copy editing if that's appropriate for the review, but only because that's something I enjoy to an unusual degree). In this case, I'm sorry to say that I couldn't find many positive things to say. The author's intended audience isn't one that I think exists, and the outline didn't seem to demonstrate that he knew the subject material very well. (Side note, if you're planning to write a book: Don't get critical facts wrong about your subject matter in your proposal. Oh, and spell-check the proposal, too.)
So off my unfortunately negative comments go to the publisher, who'll then forward them on to the author, I imagine. I wonder if I'll hear from him directly; I hope not. The issues with the outline are ones that he should correct with the help of his editor, or perhaps he can convince the editor that I'm wrong and they'll go ahead and publish the book regardless. If so, I won't buy it, but that's no great loss because I won't be in his target market.
Anyway, I'll be interested in seeing how it all turns out. The last time I reviewed a proposal for this publisher, I didn't hear anything back from them for almost two years. Hopefully they'll be a bit faster this time.
Turns out the new 12" PowerBook is not even a quarter-inch thick. Wow!
Robert Scoble wrote a terrific piece yesterday on Microsoft's efforts to build communities. I agree with what Brian Jepson says: "It was hard for me to choose a quote from all the good stuff in this piece."
I'll choose two quotes, though:
He's completely right. Microsoft would probably like to have a much better community presence, but the company just doesn't do well there. And their efforts to improve that situation haven't been very effective.
Sam Gentile has much of the explanation: "NEC doesn't constantly get frivolous lawsuits every day and have anti-trust issues. Especially because of the legal issues in the anti-trust situation alone, Microsoft people have had to watch carefully what they say and why there is all this legal control."
Microsoft's legal department exerts a lot of influence over the company. That's the way it has to be, unfortunately, since anything the company does can trigger a massive public response, and even when that's misguided it's tough to stem the tide. Legal departments don't like things like open communities; they can't control them, and lawyers don't like things that they can't control.
That said, even at Microsoft the legal department can be forced to adjust to reality. That's why things like Rotor and the other shared source projects exist -- Microsoft acknowledged that its customers were drawn to open source software and had to make some moves in that direction. But successful open source projects are characterized by their strong communities, and Microsoft's shared source projects aren't. Why is that?
The simplest answer is that the strongest communities are ones in which everyone can contribute. People aren't drawn to a community for the sake of being talked to; rather, they're drawn because they have something to offer, and they appreciate it when their contributions are recognized. That's why Linux, Apache, and Perl are so successful, but the rule doesn't only hold for open source projects. BBEdit, NetNewsWire, and Apple's Carbon and Cocoa frameworks also have strong communities. For each of those, comments, bug reports, and feature requests are all answered by the people who work on the product, and those people give honest, detailed answers...and get the bugs fixed or the necessary features implemented as soon as reasonably possible.
Microsoft's communities don't work that way, at least not in my experience. While Microsoft products are certainly dramatically influenced by the feedback the company gets from its customers, I think it's pretty unusual for a Microsoft product to change based on feedback from its community. And even the shared source products, for which the users have the source code, don't take code changes back.
If Microsoft's management wants to build communities, it has to start by realizing that you can't build a community by writing a memo saying that your vision is to build a community. Communities don't work that way. Instead, start by taking a small group of engineers who use the product that they work on (so they understand their users' perspective) and get them to talk to their users. Encourage them to encourage their users to contribute. Take feature requests, take bug reports, and act on them quickly; don't force users to wait a year or two to see the impact of their comments. Make the product extensible, as BBEdit and NetNewsWire are, so your customers can do things with it that you never imagined. Say thank you when people contribute; make them feel appreciated. Maybe even start a weblog written by one of the engineers, as Dave Hyatt has for Safari. Don't make it anonymous, don't add marketing fluff to it, and allow comments.
Don't be afraid to be upfront about the things you won't talk about. When I was effectively the point person for Apple's Java team on their mailing list a few years ago, I had to explain over and over again that I couldn't talk about release schedules and I couldn't talk about plans for Java3D. Most people understood that, and those who didn't, left. In Microsoft's current communities, I think the preference is to ignore or talk around questions that aren't comfortable. Consider the question of porting some Microsoft software to Linux, for example. Rather than saying, "We're listening to community feedback for where we should port next," go ahead and say, "We hope you'll understand, but we're unlikely to port to Linux. We'll look at other systems like FreeBSD instead." Everyone knows the first version is ridiculous if the community would want a Linux version. But everyone can understand the second.
Even if Microsoft does all of this, can the company succeed in building strong communities? I'm not sure, and I think that's part of why the company hesitates to change the way it interacts with its customers. I have a tough time picturing a strong community building up around Windows, for example, or IIS. But it'd be fascinating to see what would happen if Microsoft were to, say, open up the source to Services for Unix or something like that, and try to build a community there. Or start a package management system on top of SFU, and make it easy for people to port Unix apps to it, then make it really easy to install those ported applications from Windows. Let people contribute and help building something, and the community will have a good chance of growing.
Mary Jo Foley is putting together a list of weblogs authored by current and former Microsoft employees. I'd wonder why she cares, except that her page is entitled "Microsoft Watch". I wish she wouldn't have that list. Its implication is that by virtue of their connection to Microsoft, however tenuous, what these people say is worth watching so she can figure out what the company is doing next. Some of us would just like to write in peace, thanks.
Fortunately, she hasn't found all of us yet. Hopefully she won't.
I should mention that I wouldn't mind if someone had a list of Microsoft engineers (and ex-Microsoft engineers) with weblogs out of pure interest, or just because people with a common employer often have other things in common. But Foley's site centers on rumors and speculation, and I can't imagine that many Microsoft-associated bloggers are interested in getting pulled into that.
There's been too much stuff going on here recently for me to post much. Just when I started to figure out what my job entailed, I realized that it was actually going to take a lot of effort (and time) for the forseeable future. And I keep getting new things to work on, too.
I'm not complaining about that; rather, it's wonderful to be working on something important to the company for the first time in my life. I'm also learning a lot -- I've probably learned more useful things in my first three weeks at Apple than I did in my first three months at Microsoft. So I'm enjoying myself, but work is taking a lot of time. I think I'm going to have to cut back on my email, newsgroup, and web site reading just to ensure that I get enough sleep.
Why on earth did it take Slashdot for me to find out that Dave Barry has a weblog?
The Space Shuttle Columbia broke up this morning on re-entry. I wasn't watching at the time, but Ruby woke me up when she saw the news. It's awful -- for the families, who watched their loved ones die; for the space program, which can't afford to have anything go wrong; and for the nation, which has seen enough disasters in recent years.
I remember the Challenger explosion, and the meaning it had to a boy enthralled by the idea of exploring the universe. Strangely enough, the disaster only increased my interest in space, as I and many others paid more attention to subsequent Shuttle missions than we had to Challenger. I hope that the same outpouring of interest will occur here -- that a new generation of children will cheer on our space program with more energy than they did yesterday. If so, that's one positive that might come from today's events.
It's been painful over the past few years to watch as Congress cuts NASA's budget more and more, essentially rendering the agency incapable of doing all but the most necessary missions, which largely support a space station that has never had a real purpose. Congress and others will demand investigations, of course. I sincerely hope the result of those efforts is to give NASA the support it needs to ensure that its engineering is top-notch and that its missions increase our knowledge of the world around us. Though space missions don't always have direct benefits to our lives on Earth, we can hardly blind ourselves to the rest of the universe. The sense of wonder and amazement for all of humanity as we learn more is immense. NASA can -- and should -- lead that search for knowledge. Today's events should only inspire us to try harder.