extension cords are always useful

Well, I managed to justify buying a new 12″ PowerBook. I like it. It’s so nice and smooth. And the antialiasing on the fonts is just SO nice. Every web page just looks better. And it holds a lot of my music. And it’s nice and small. And I can get away with not caring about how the disk is partitioned, or how the registry is registered, or how the start menu is arranged, or how many processes are running on startup. It even accepted my NTFS-formatted external hard drive without a hitch! Wow, I’m flabbergasted. And Expose will take some getting used to, but I think it will be a very used feature on this meager 1024x768 screen.

Now I’ve gotta buy protective stuff for it. I don’t want my baby to get all scratched up.

Last night, I spent a couple of hours writing up my thoughts on the PS3 and the Nintendo Revolution. Well, of course the computer crashed, right as I was nearly done. Maybe I’ll write it again later. But that’s not the point of this entry.

I saw Star Wars Episode III: The Revenge of the Sith today. I really liked it. I thought it was a great movie. I want to see it again, and maybe then I will have more to say. But for now, I have come up with a Star Wars Geek Code:

c–z+V++D+PM+AC+RS+VG+++B+++C+FL+

http://www.wbwither.net/swgc.txt

Comments, suggestions, and improvements are welcome. It’s 6 in the morning, so I know I could probably do better.

First off, as I commented on Scoble’s blog (heh, wonder if he’ll read this) I didn’t watch the MTV thing. I think that launching the Xbox 360 on MTV was a stupid move that alienated much of the core audience (which prefers to get their news online from gaming-related websites), while building (or attempting to build) hype among a general audience way too far in advance. I doubt we’ll see Longhorn ads on MTV 7 months in advance, let alone 30-minute specials. Everyone whose opinion I value, who watched the show, said that it sucked. This includes personal friends of mine and Gabe and Tycho.

Next. It looks pretty damn cool, but that is to say that it looks like a big, elongated iPod. There’s even the circular thing that looks just like the scroll wheel. I mean, MS did a good job with the design, although I think it’s a bit too derivative of the Apple aesthetic. Don’t get me wrong, the Apple aesthetic is a good aesthetic, and the Xbox 360 looks nice (much nicer than the old Xbox — I painted mine).

Thirdly. I’m not sure if I’ll like what they did to the controller. It certainly looks pretty cool, but the exact way they moved the white and black buttons to the back gives me pause. I thought they were going to simply make more trigger buttons, below the regular trigger buttons — so you could depress them with your middle fingers. From what I’ve read, the new buttons are actually above the triggers, so if you want to have quick access you’ll have to either use the triggers with your middle finger (not ideal) or move your trigger finger up to mash the other button, necessarily keeping it from holding down the trigger while you do whatever else you’re doing. To use Halo 2 as an example, I think it’d be great to be able to jump while shooting both guns and maneuvering around. As it is with the current-generation Controller S, you’ve got your thumbs maneuvering (one on each analog stick) and your trigger fingers shooting — to jump, you’ve got to remove your thumb from the right analog stick and hit the jump button, thus removing your ability to look around for the split second it takes you to hit the jump button. It’d be great to jump by making use of a currently-unused finger — for example, the middle finger, hitting a second trigger located underneath the first trigger. Easy. But I haven’t seen pics of the back of the controller yet, so I might be wrong.

Fourth. Wireless controllers = awesome. Infrared controllers?!?! What were they thinking?! Talk about throwing back to the good ol’ NES days. I mean, come on. This limits the usefulness of the controller in SO many ways. Say I’m laying sideways on the couch, and the controller’s not pointed toward the Xbox. Say I’ve got my Xbox on the bottom shelf of my media center, and I’ve got a coffee table that keeps me from having line-of-sight. Say it’s cold, and I’ve got a blanket over me, including my arms and hands and the controller. This is a huge gaffe. Given the excellent battery life of my Wavebird, I don’t think MS went with IR for power consumption reasons. Of course the Wavebird doesn’t have force feedback (the biggest single drain on the batteries), but still. Cut the Wavebird’s battery life in half and it will still be reasonable.

Fifth. They’re including an Xbox Live headset, and the “communicator” is kaput. Thank God. They charge ridiculous amounts for those now. It’s like $30 for a crappy microphone headset and a plastic bit with a potentiometer.

Sixth. Technical specs. My jaw is still dropped. 3.2GHz 3-core PPC chip with dual threads per chip? And the real kicker — an ATI graphics core with 48 pixel shader pipelines?! Wow wow wow. Top-end graphics cards today max out at 16 pipelines — and those are the $500 cards. Even spending $300 today on a video card just gets you 8 pipelines. It was a given that the number of pipelines would be at least doubled in the next generation, but wow. 48 is simply insane. The video card will be like 75% idle at anything less than 1080p — seriously. Current-gen video cards have no problem giving playable framerates at 1600x1200 for the best games (e.g. a GeForce 6800 Ultra (16 pipelines) gives 70+ FPS in Doom 3 at 1600x1200). 1600x1200 is 1.92 million pixels, while 1080p (1080x1920) is 2.07 million pixels. Not much of a difference there (and *all* computer displays are progressive-scan). Is there a higher HDTV resolution than 1080p? If not, they’d better come up with one, because at 1080p, this GPU will just be stretching its legs. If you’ve got the Xbox 360 hooked up to a standard-definition television, it could probably shut down 46 of the 48 pixel shaders and you couldn’t tell the difference. Don’t believe me? The “old” Xbox’s GeForce3-derived GPU has 4 pixel pipelines and runs at 233Mhz.

Miscellaneous items. Glad they kept the hard drive. Glad they made it so you can stand it up on its end. Glad they made the top so you can actually stack stuff on top of it. Glad they’re only doing one model, at least initially. The “Circle of Light” makes me laugh out loud at the name, but it sounds like it might be useful — and it provides the meaning for the “360″ part of the name. Free emasculated Live is kind of cool, but I’ll still pay for the Gold level. I just hope that they don’t turn things into a nickel-and-dime-you-to-death scenario where every game has 20 50-cent downloads. Hopefully, most games will not have non-free downloads. The Halo 2 map pack model of “pay to get it now, or wait and get it free in 4 months” really appeals to me. I hope that all developers will see the downloadable content situation the way that Bungie sees it –you already paid for the game, so you ought to be able to get the full game. (Even if the ending sucks.)

I’ve been loyal for years. Never even looked at anybody else. I’ve been perfectly happy with what I’ve had. And I was pretty sure I was getting the good end of the bargain.

And just tonight, I cheated. I flirted with another. More than flirted — I experimented. I had an insatiable urge, a curiosity that had to be indulged. I dipped my toe in, to see if maybe it could be better with someone else. It wasn’t. And now, I feel terrible about it.

I got an online quote from Geico. You know, the car insurance. I’ve been with Progressive for some 3 years now, and their premiums have always been very reasonable (and recently decreased by like $30 a month! — possibly related to the fact that I turned 24), and the customer service is top-notch (fortunately, I’ve never had to make a claim). So I got a thing in the mail from Geico. Save up to 15%, blah blah blah, you know the drill, they’re blasting it from every radio and television in sight.

I had to try it. I had to see if maybe they would be cheaper. They weren’t. Never again. Progressive, I swear, this is the first and last time I’ll cheat on you. How can I make it up to you?

Last night I watched a video interview with Dave Probert, a senior Windows Kernel Architect. It’s a pretty fascinating hour and a half, particularly if you know anything about UNIX and how it works. (I don’t think I’m alone in being unaware of Windows NT and how it works; I had assumed that such knowledge wasn’t readily available.) I’m not that well-versed in operating system theory, but there wasn’t anything in the video that I didn’t understand. In fact, I was proud of myself when I knew one answer before he said it — the “kernel of the kernel” would be the scheduler, and when I knew that the HAL stood for Hardware Abstraction Layer while he mistakenly called it the Hardware Adaptation Layer (he admitted that he’d been using the acronym so long that he had forgotten what it stood for).

Anyway, he explains how really one of the biggest and most important parts of modern Windows (based on the NT kernel — i.e. NT4, 2000, XP, Server 2003) is basically an emulator for the Win32 APIs that Win9x introduced. There are native NT API calls, but almost every piece of software — even Microsoft Office — uses the old Win32 APIs. NT is actually designed to be remarkably flexible; when it was built, it had HALs to match up with Alpha, PowerPC, MIPS, and x86 architectures, and APIs to match OS/2, Posix, and Win32. But as things progressed, the OS gradually became more closely tied to Win32 on the API side and x86 on the hardware side (although there is now a version of Windows that runs on the IA64 (Itanium) architecture).

So what we have here is a situation in which the most popular OS/architecture combo in the world forms an emulator running on an emulator. WinNT emulates Win32; modern x86 is a RISC architecture emulating the CISC x86 instruction set. You’ve got to wonder whether all this emulation on top of emulation isn’t just an awful waste of computing power. I doubt it’s an order of magnitude difference — probably not even a two-fold increase. But it’s no doubt there nonetheless.

Order of magnitude, now there’s a concept we haven’t successfully binarified. I just got through with The Mythical Man-Month, and Brooks kept talking about an order of magnitude increase; he repeatedly and explicitly stated that an order of magnitude is a 10-fold increase. Now…. why do even computer scientists stick with the old base-10 definition of order of magnitude? Why is 10 (base 10) so special? Going from 1 to 10 only means a doubling in binary; it means an 8-fold increase in octal; a 16-fold increase in hex. It seems to me that for a computer scientist or a programmer, an “order of magnitude” increase ought to be a fluid concept, meaning at least a doubling but as much as, say, a 20-fold increase.

Ugh. I’ve been sick for several days now. It’s pretty much just a cough… runny nose, sore throat. The cough is the worst part, though.

The weirdest thing is how I feel mentally when I wake up. I woke up this morning and promptly had a coughing fit that lasted about 5 minutes. The first thing that came into my mind was that this was a bug, and I needed to get online and submit a bug report or do something to fix it. Not in so many words, but that was the feeling I had — that this was a technical problem, and I needed to fix it.

Several months back, when I was working on the site and I had to have an alarm to wake me up every morning, I would often have almost the same feeling (hallucination?). The alarm would go off, and somehow I would think in my head that the alarm was connected to my programming. That I had set it up to test the site or something, and it had found something and it alerted me, so I just needed to acknowledge the bug and have it keep going (i.e. hit the snooze button). I swear I could picture my alarm clock somehow having a TCP/IP stack and doing Important Things for me…. it was connected to the global Internet and was bringing me important news and results. And I would be so sure of it, too, in my screwed-up semi-awake mind. This happened on almost a daily basis, probably for a couple of weeks.

Is this a sign of something? Am I an actual programmer now? Or am I just losing my mind?

This is in response to Robert Scoble’s post “A church switches to open source and teaches us a lesson” and the comments on that post.

I’ve got to agree with John Welch, especially his first set of comments. At least in terms of what he’s describing. In terms of what he’s suggesting, I don’t know if that’s really where Microsoft wants to go today.

When MS decided to give out the $60 billion stack of cash as a dividend to its investors, that sparked quite a conversation among my friends. $60bil is quite a bit to just let evaporate from the corporate coffers. My final take on the situation was as follows. MS isn’t evil. They’re not even out for world domination. It was a sound business decision, a statement along the lines of “We’re pretty much on top of the world right now, but things might not be so great in the future, so we’re giving our investors a good return on their money right now.”

It’s a given that just as computer hardware has become commoditized, so computer software is quickly becoming commoditized. MS had (and will continue to have, for a while) a good niche in the market. There was a very great need for a company such as Microsoft to essentially be a monopoly in what I’m sure people 50 years from now will look back and see as the dawning of the computer age. MS helped democratize computing. All these Linux hackers wouldn’t have the cheap Intel hardware to hack on, had it not been for Microsoft. Now that the market is largely saturated, with most home computer users on their 3rd or 4th machine, with improvements becoming increasingly incremental both hardware and software-wise, there is less opportunity for companies like Microsoft, and less need for them as well.

IMO, the dividend payout was an admission of as much by Microsoft: The only way we’re going is down. It’s the unfortunate fact faced by those on top.

However, it looks like MS is making a play for a horizontal sliding move to the consumer electronics side of things, which personally I find pretty exciting. Xbox Live has succeeded wildly — MS is the first company to transform console video gaming into a subscription service, and I and millions of other gamers happily pay MS $5 a month to play wonderful games on MS’s wonderful system and MS’s wonderful network. Whether this system can eventually turn a profit is another question, but the amount of consumer goodwill in this area is frankly pretty astounding, given the initial resistance to Xbox.

Once again, Microsoft has first-mover advantage and it’s pretty much MS’s market to lose. MS has shown that, given a closed system, they can create software — more than software, an experience — that makes people want to buy and use a Microsoft product, which is sadly not the case when it comes to most software, where it’s more of a default choice: “I have to use XP because…” “I have to use Office because…” “I have to use IE because…”

Again, given that Xbox is a closed system, I really think that this is Microsoft’s best opportunity for future growth, while the core PC business will gradually shrink. Probably 5 or 7 years ago, I read some quote from Bill Gates about MS being in everybody’s living room 10 years down the road. At the time, it seemed like a joke. Right now, though, it seems like it’s largely true, and becoming more true by the day.

So, in summary: No need to worry about the decline in computing this, servers that, desktop this, IIS that, SQL Server the other, oh no we’re losing market share, LAMP has a bigger community, blah blah. All that is just fine. Those markets — they’re a losing place to be. Being profitable in a commodified market is VERY difficult, especially for a company like Microsoft where margins are VERY high. MS has spent a lot of money breaking into the video game market (and the home entertainment market more generally), and the reason for that is that the people in charge understand what I just said about commodified markets. There’s no need for hand-wringing. It’s all going according to plan. Sure, there will be rearguard movements, but I’ll bet you Bill Gates doesn’t see MS being a server operating system and office suite company in 20 years. He’ll have moved to where the profit is — and it’s not in web development languages and database software. Just ask Oracle or Sun.

Well, I’m done with a part of my programming at work which involves manipulating data and finding what you want within that data. We provide two ways of finding things, searching and browsing. I’ve found that I’m much more of a browsing person, and I’m not sure why that is. I would rather click 4 times to find an item I want, than type a quick search term and have it come up on the next page. (Of course, I know exactly how many clicks it will take to find it — if I didn’t, a search might be more appealing.)

I take iTunes as an example. There’s really something very special about the way it finds your music as you type in your search term. It fills me with glee to see the numbers of matching songs count down as I add each character. And yet, I really don’t like this as a way of finding the music I want to play. I’m too used to browsing through folders that I’ve arranged — of course, nobody else could find anything in them, but I know pretty well where everything is, even with my ~50GB of MP3s — and choosing the songs that I want and putting them into a playlist. A playlist itself is quite mutable, although mine tend to remain static for weeks or even months at a time — maybe 8-10 hours long that I’ll set looping endlessly. If I have a new song I want to hear, I’ll add it to the list, then maybe delete it when I’ve heard enough of it. Or maybe it’ll find a new place at the bottom of the list. When I do revamp my playlist, it’s often kind of a fun process as I go through my various folders. “Oh yeah! I haven’t heard that in a long time.” It’s necessarily less directed than a search would be, although as I said before, I’m always able to find whatever I want pretty quickly, if I have something particular in mind. “Shuffle” functions don’t work for me because much of my music is connected track-to-track. I guess I wouldn’t mind an intelligent Shuffle function that keeps groups of tracks together. I don’t particularly like the way that iTunes does playlists.

I’m the same way when it comes to e-mail. I’ve got a gMail account, and I really don’t quite see the point of searching my e-mail. Maybe I just don’t use e-mail enough — I’ve certainly had a severe drop in my e-mail usage since I left college. Anyway, I use Yahoo! Mail for webmail, and even though I think they have a search function, I never use it — I’d rather scan back through my saved e-mail folders to find what I want, because I know generally where it is. I guess that I do use the search function in Pine on a fairly regular basis, but it’s usually just to find a person’s name in the From: header. The more-or-less immediate sorting in Pine is pretty nice too… with two keystrokes, I can sort based on numerous criteria, the most useful again being the From: header. Anyway, the point of this is that I’m simply used to dealing with my e-mail in the old-fashioned folders/sorting way, and although I can appreciate the theoretical coolness and utility of being able to search my e-mail with something more algorithmically complicated than grep, I don’t find it so useful personally.

Maybe this is all part of a control thing? Like, I feel that I *must* organize my music/e-mail/whatever — because if I don’t, then I’m not really in control of it? I’d rather put my stuff in rigidly defined files and folders and playlists than have it “at my fingertips” at the other end of a search query. I wonder if other people are different and find the search method more intuitive.

To the Yahoo and Adobe thing.

Hehe, I was right. Yahoo search results pop up a new browser. On Windows XP, they pop up IE, even if something else is the default. Interestingly enough, right-clicking on a word in Acrobat Reader (just right-click with the normal Hand Tool) will bring up a context menu including the item “Lookup [whatever that word is]”. Choosing this option will search for the word on Dictionary.com…. in the default browser (Firefox in my case).

I stand by my earlier conclusion that this is a stupid, stupid product. Yahoo is the only beneficiary of this arrangement. They get their plug-in bundled with Adobe Acrobat, the most ubiquitous browser plugin around. (Well, maybe next to Flash). What does Adobe (or Adobe’s users) get out of it? Less screen real estate and crappy functionality that is actually already beaten within their own product (the Dictionary.com search is initiated by clicking on a word in the document, while the Yahoo search is lacking this functionality. So if you come across a word in a document, and you want to do a Yahoo search for it, you’ve got to open the Yahoo search thingy, type in the word, and then when you click submit, Internet Explorer shows you the results.)

Geez, you’d think the marketers actually programmed this one. I can just picture an Adobe engineer…. “Wait! We’ve already got functions to call up the default web browser and…” Marketing: “Hey, don’t worry about a thing! We’ve got it from here.”

Hate bloggers?

What I find funny is that even his original LA Times piece (PDF) was pretty dumb. He misunderstood the purpose of Google’s library-scanning thing. It’s not all about the searching. A huge part of it is just the archiving of all of those millions of volumes, copying them into a form that will be easily replicable and will one day fit in the palm of your hand — everybody’s hand. This is a project that has to be undertaken only once, and its fruits will outlive Google, the Internet, blogs, and maybe even humanity as a species.

The digitization of our libraries is perhaps the most exciting project, knowledge-wise, to hit in a long time. It’s the modern-day equivalent of Gutenberg’s press: a technology that democratizes information, takes it out of the centers of learning and puts it into the hands of people. God knows I’d love to still have regular access to the library at the U of C, but that’s impossible for me right now, and it’s impossible for millions of people for all their lives.

Besides, I dunno what kind of scholarly research he does, but when I was in college, I didn’t have time to read an entire book to get “context.” The containing chapter at most, and sometimes just the few pages surrounding it. And it doesn’t seem that many of the academics who actually wrote the articles and books that I cited, read the entire books either. Taking stuff out of context is par for the course in academia. What might have been one author’s afterthought in one chapter of a book becomes a key piece of support in another author’s article.

But those concerns aside, it is a given that making all of these volumes searchable will increase the productivity of researchers by orders of magnitude. The one reasonably in-depth research paper that I wrote in college started out relying on just a few sources, because those were the only ones that the professor and the Ancient Studies librarian (who, besides being a total babe, was extremely knowledgable and actually knew several articles off the top of her head that would be relevant to my paper) knew of. And then I went to the library, copied those articles, checked out those books, read the relevant parts, checked their citations, found those cited books/articles, checked their citations, etc. etc. in an ever-widening circle of information that took weeks before I found every source I ended up using. And it was frustrating because there might have been a very good, relevant article that I simply didn’t find in time.

Google’s search algorithm actually mimics the citations-give-authority model of academic research, and therefore it is perfectly suited to this kind of research. Academic articles have been “hyperlinking” for quite some time, and the levels of inter-linking and robustness of authority on a topic are MUCH simpler to model within this closed system than they are on the Web. It’d be just as easy to find, not only what articles this paper cites, but what papers cite this paper. Seriously, what took me weeks of crawling the library could happen in an afternoon once this system is in place. And the folks who currently don'’t have access to a knowledgable professor or librarian…. they can find the same information I did. Possibly even better information. Er, “knowledge.”

And several of his objections are simply poorly-thought-out. No, nobody wants to print out a 500-page book…. but nobody wants to spend an hour and a half standing in front of a copy machine, paying 5 cents per copy for 15 articles at 40 pages apiece, in what amounts to manual labor as you flip each page, hold it down, press the button, rotate the book, hold it down, press the button, flip the page….. since most journals aren’t check-out-able in most academic libraries. In this case, the higher-tech solution is most welcome.

Plus, I disagree with his assertion that full context is required. Most academic writing isn’t so opaque that a reasonable person can’t understand what’s on page 142 without reading pages 1 through 141. In fact, many books explicitly refer back to important ideas (”As discussed in Chapter 3, …”). We’re not reading James Joyce here. A reasonable level of knowledge about the subject at hand is assumed. If I’m writing a paper about Sparta, I don’t care if the author is using the example of Sparta to prove that political and military hegemony cannot last more than 3 generations… I just care what he says about the particulars of Sparta. And I don’t care about the other Greek city-states and their political systems either, except to the extent that they interacted with Sparta.

The proof is in the pudding. There are already plenty of invaluable search tools used by academic researchers. I know several professional social-science/psychology professors/researchers whose first resource is always Ovid’s PsycINFO search with fulltext abstract, keyword, and topic searches. In the law arena, Lexis-Nexis provides the same service, with millions of opinions, judgments, and articles fully searchable at the click of a mouse. There’s a reason why lawyers pay hundreds or even thousands of dollars a year for this online service: it works, and it’s much better than the non-tech alternative.

Anyway, I guess I can sum up by saying: Michael Gorman has completely missed the boat.

Hmm, I wish there was a way that WordPress would automatically say “This post tracks back to the following: http://drbacchus.com/wordpress/?p=900