Tuesday, December 25, 2007

Virgin Island Pictures, Part the First

Virgin Islands Part I - St Croix

Book Review: While I was Gone

This is a story about two women: Jo Becker and Dana Jablonski. They meet in their twenties during the 1960s. Jo is running away from a marriage, while Dana is living the life of an artist. Jo lies to hide her past from her housemates, while Dana is ruthlessly honest, defending herself from gossip by letting everyone know intimate details of her life.

An expected narrative would be a conflict between the two women, their ways of living, and perhaps an examination of the consequences. Instead, the narrative is one of Jo Becker telling the story of her past to the reader in first person (the story begins when her three children are all adults and out of the house), all the while continuing to keep her secrets from her husband and even her daughters, to the point where they call her elusive.

Becker's otherwise boring life becomes suddenly more interesting when one of her housemates from the past 25 years moves into town and they renew a connection. Secrets become unveiled, and Becker's otherwise stalid life becomes in jeopardy.

Other editors and reviewers comment that the book is about how one must deal with secrets and be careful which you should choose to keep, so I need speak no more about it. The copy of the novel I read had an interview with Sue Miller in which she noted that Jo Becker, as a person who chooses to act rather than to reflect, had certain limitations that made her uninteresting to write or contemplate. That explains why the first few chapters of the book were so difficult to read --- they rang false on almost every note, as someone who's pre-inclined to action over contemplation is hardly likely to make such a narrative. I also note that Sue Miller chooses to use a scientist as a villain in this piece, and does so by buying into every stereotype of a lab-rat scientist. Perhaps being an artiste herself, the only person she could have fill the role would have to be someone thoroughly alien to her. This is the part of the novel that I noticed and did not really appreciate.

I read this novel as an airplane novel, and perhaps, that was the only way I could have read it --- when I was a captive audience. While it gave me quite a bit to think about, I can't say that I agree with either the premise of the story, or the means by which Sue Miller chooses to make it. It's not a waste of time, but I'm not sure it otherwise has much to recommend it.

Monday, December 24, 2007

Rainbow over St. Croix

After today's SCUBA lessons, we had a late lunch and returned to our hotel in time to see some rain clouds form and start small rains from our hotel room. Lisa shot this picture from our balcony after a particularly pretty session.
Posted by Picasa

Sunday, December 23, 2007

View from Cessna 402

We took Cape Air from St. Thomas to St. Croix, on a Cessna 402. The highlight was when the pilot invited me to take the co-pilot seat. Of course I took it! It was only a 20 minute flight, but what a view. And I didn't have to do any work. This picture was shot by Lisa from just behind me.
Posted by Picasa

Sunset at St. Croix


At Sunset, a View of the Hotel on the Cay from the Boardwalk
Posted by Picasa

Friday, December 21, 2007

Why I love Youtube

Anyone who's on my IM list probably has gotten strings of youtube links, often to unrelated subject matter. There's just so much about youtube that its probably the most TV i've seen in the last 10 years or so.

I mostly use Youtube as a means of seeing old music videos, stuff like Baltimora's Tarzan Boy, or A-ha's Take on Me. Yes, I'm a child of the 80s and over the last year or so have managed to relive much of my childhood MTV videos. Much time spent reminiscing and thinking about days gone by =).

So Youtube is pretty great for english songs, but how about foreign languages? Its spectacular there too. Case in point, and the real reason why I'm blogging about it, I managed to find a song I have been searching for years just yesterday, by randomly clicking through stuff.

It started when I was looking for a rendition of the Macross ending theme, and found this compilation of various other anime ending theme songs...and in the middle of one of those songs, found the song I've been looking for!

The song, for those of you interested is the original Japanese version of this Alan Tam song(譚詠麟- 愛的替身)...



You can find the full version here, although the version I ended up liking the most is this one.

After I had the name of the song, it was a simple matter of contacting my Japanese friends, and I will soon have the CDs in my hand =) Years of searching and I finally have the song I want. =) Granted, I wasn't looking very hard either, but its such an unexpected bonus that I simply had to blog about it.

So once again, I repeat myself. This, and many reasons more, is why I love Youtube and so willingly spent hours upon hours every week clicking rapidly through so much stuff. Becuase when gems like these pop up unexpectedly, it is very much a Christmas Present come early!

P.S. Lyrics & translations for those who like to know these things

Thursday, December 20, 2007

The Witcher Review

Over the last two weeks, my leisure time has been pretty much sucked up by this completely engrossing game: The Witcher.

The Witcher is a RPG set in a fairly typical Tolkein'esque fantasy setting. You have your dwarves, your elves, your dryads, your monsters (too many to list, but suffice to say you won't be whining that its simply a world full of short humans and point eared humans running around).

The game has you playing as the Titular hero, "The Witcher". Which in this world, refers to a special group of Monster Slayers. Far be it that they are normal, Witchers are a special breed in that they not only have mastered swordplay, but have also mastered sorcery, and also have special potions and other stuff that has mutated them far beyond that of a normal human. There are also normal human Monster Slayers in the game, but the game does a great job of explaining the difference between a normal human who happens to slay monsters and a witcher whose primary job is to slay monsters.

So far pretty rote right? Nothing special so far. Some might even be turned off that you don't get your choice in how you wish your character to be created...what if you don't want to be a sword wielding monster slayer, how about ranged weaponary? why swords? Well, the game doesn't give you any choice at all. You even have a name assigned to your Character....so those hoping for a Oblivon'esque experience have to be prepared to take this compromise or be prepared to skip one of the best RPG experiences in the last 5 years.

As mentioned previously, the game doesn't give you a choice in who to play...you have to play Geralt, who apparently is a famous Witcher in his own right. The game gets around the "this guy is so buff he doesn't need to level up" problem with playing a legend by giving him amnesia, so even though he knows how to use a sword, he's not great with it (yet). Although a bit of a cliched mechanic (how many more amnesiacs do we need to play??), it is very well handled through the great and mature writing.

Which brings us to what makes The Witcher a great game. The writing. The world that the Witcher takes place in is a very gritty world. Think Glen Cook's Black Company for a similar comparison. In this game, folks dying is just the beginning of the exploration of the mature themes. There is rape, torture, incest, and, well, Love. And that's just the tip of the iceberg! The developers however, cannot be credited with such a strong world, as most of the world buildling was done by the author of the Witcher series. Never heard of the Witcher series? Well, you're not the only one as it was previously a fantasy series sold only in Poland. An English translation of one of his novels is coming over the pond however, and it should be an interesting read.

What the developers can take credit for is how well they enhance the RPG experience. In most RPG experiences, you have your choice of being good, or being bad, with very little grey in between. When you're good, you're good as the purest snow, and when you're bad, you're like the lovechild of Hitler and Stalin.

The Witcher doesn't let you off so easily. There are moral ambiguities in almost every decision you make, and there is no right or wrong answer. Each answer can be justified one way or the other, and surprisingly enough, they even give you a choice of remaining neutral! With a character as powerful as Geralt, the game portrays what happens to neutral characters very realistically, and it is rather satisfying to see such thought put into the game story.

Beyond the whole moral "good", "bad", "netural" landscape of decisions to make, there are also other in-game decisions you make that affects how the game plays out. A decision you make in Act 1 (there are 5 acts, and 1 prolouge and 1 epilogue), can come back and affect you in the later acts. This is incredibly masterful in that it prevents the usual "save & load" syndrome that many RPGs have. By making decisions you make have an immediate payoff, and later payload unknown to you, it makes almost every decision you make seem weightier than ever! The best part is it really does make the world seem like a living world. A character you save in the beginning can come back to help you, or come back to haunt you (literally)....a decision to help a certain faction can deny you some side quests later...these decisions are never game breaking (they don't break the main quest), but it adds so much more flavor and details to the game that its hard to ignore.

A special mention needs to be made here for the story, the story is actually fairly simple, but the amount of details put into it is what makes it incredible. Add to it the choices you can make that customizes the game towards you, and it makes the story a rather rich multi-layered affair. Its fairly similar to something like Lord of the Rings (get rid of the ring) in that it is a fairly simple story, but the amount of details and layers added to it turns it into a classic. The Witcher is much the same way. Don't expect all threads to be revealed either, there's at least two pieces of ambiguity that never gets resolved at the end of the game, and its not really so much for "lets have a sequel" effect, as it is a "real life rarely reveals all either" effect.

There's also an option for the players to engage gratuitously in sex, and while some might object to it, its something easily skippable. Its also not explicit and non-interactive, for those really prudish about what they like to see in a game. I see it as an almost necessary feature in a RPG which claims to be gritty however.

The combat system is also fairly...revolutionary for an RPG. The game engine is a heavily modified NWN engine, and the combat is much the same: real time clickfest. Instead of just rapidly clicking until your enemy is dead, there's a timing aspect of it. You have to click at the appropriate moment within the animation to make the character combo into his next move. While it sounds like yet another simon says game, the timing is actually fairly complex and on the higher difficult levels, the window for the click is reduced, and there is no visual aid (on easy and normal, there's a flaming sword icon that replaces your mouse telling you "this is the time to click!").

The game graphically is gorgeous. Easily one of the prettiest games I've played, and I'll say the artwork rates higher than that of even previous heavyweights like Oblivion. Combat is a rather beautiful affair as long as you only look at what your character is doing...Geralt is very much a work of art when it involves swinging his sword, and even though it makes no sense why he has two styles of fighting with two very similar weapons, its mostly to give the players more eye-candy and more character customization options (max out silver sword? or steel sword? max out magic?).

The cutscenes in the game are mostly rendered using the game engine and I have to say I have never seen in game engine cut scenes rendered as beautifully in an RPG before. You'll swear the cut scenes are basically movies....until you get to the movies, and there its even better. They really did capture the way Geralt moves and fight. The scenery within the game are also spectaular, ranging from the complex city, a dreary swamp, to a very cheery village. Locales are varied and dungeons are kept to a minimum and only in sensible locations.

So...what are the shortcomings of a game like this? The first is that load times can be rather long, though the newest patch promises to solve this issue. The other is that you can't ever really be evil. Even when you side with the wrong faction, the game makes sure you realize the mistake down the road and corrects it for you. Though the correction makes sense, it can still grate on those who truly revel in what evils their in-game character can create. Given that this is a game about the titular character, it should be expected, but I can still see it as a problem. Oh yes, there are also numerous Crash-to-desktop bugs as well, but once again, all these problems appear to have been patched away (I didn't patch for fear that I'll have to restart the game over!).

The game is also long, 50 hours or so. Some might see it as a shortcoming, some might think its not long enough. =)

So..in summary, one of the best games I've played this year, ranking up there with Stalker, and Portal. For an RPG to engross me this readily is no longer an easy task (I stopped playing NWN2 and Oblivion within 10 hours of each), and I can heartily recommend this game with no second thoughts.

Wednesday, December 19, 2007

Steve Yegge on Code Size Bloat

Steve Yegge has a long, interesting rant about code size bloat, Java, and IDEs. At one point, he complaints:

Heck, I've never managed to get Eclipse to pull in and index even my 500,000-line code base, and I've spent weeks trying. It just falls over, paralyzed. It literally hangs forever (I can leave it overnight and it makes no progress.) Twenty million lines? Forget about it.

It's a pity Steve doesn't work at Google, then he would be aware of gtags, which indexes and provides search results for even Goolge-sized codebases without complaining.

Oh wait... Steve Yegge does work at Google... There are good points about code bloat in that entry, but then again, there's a real problem with picking a less popular language than C++/Java, which is that there are way more tools for those languages than there are for say, Rhino. It's sort of like CMOS versus every other kind of silicon technology. There's always a new technology just around the corner, but there's so much invested into CMOS that the newer technologies never do catch up.

Krugman unearths more Obama history...

It is beginning to look very much that getting Obama as president would be as unfortunate as getting Mitt Romney as president from the universal health care point of view --- we are just as unlikely to get universal health care.

Tuesday, December 18, 2007

Sunday, December 16, 2007

Review: The Accidental Time Machine

The Joe Haldeman I remember is the one who wrote All My Sins Remembered and The Forever War, iconic pieces of fiction that in at least one case, I find emotionally difficult to read a second time. In recent years, however, with Red Thunder and even the Nebula-winning Camouflage, it seems like he's gone for an entirely different genre all together --- the short airplane novel that's straight forward and easy to read, incorporates a few science fiction ideas, and has next to no character development, and leaves you feeling a bit like you've just swallowed a whole bunch of empty calories.

Take The Accidental Time Machine, for instance. Haldeman teaches creative writing at MIT, so he is extremely familiar with the setting, and can't help including a bit of the history of the infinite corridor, for instance. The basic concept, that of time travel, has been worked over quite a bit by science fiction writers, and Haldeman doesn't bring anything new to it. The story revolves around a graduate student, Matthew, who gets a machine to travel forward to time while building something else. He experiments with it a bit before subjecting himself to its effects, and then finds himself in trouble with the law. But he's always being bailed out by either providence or a future instance of himself. He interacts with future history (including future versions of MIT, as well as a past version), but the character is entirely plot-driven --- he doesn't create new devices, doesn't solve problems with his ingenuity, and is just dragged along by circumstances and his constant desire to return to MIT, wherever and whenever he is.

So as a Tom Swift type novel this book fails. As an examination of the science fiction concept of time travel, it isn't as innovative as Vernor Vinge's Marooned in Realtime. Perhaps you could consider the character developed, since he goes from a Ritalin-using wired up grad student to becoming a physics professor who enjoys the love of a good woman, but even so, it does not appear that that's the point of the book.

So I guess the point of the book is a romp through time checking out past and future versions of MIT. Which is all very good if you're an MIT-lifer, but perhaps a waste of your time if you're not.

Saturday, December 15, 2007

Review: Agent to the Stars

Agent to the Stars is John Scalzi's first novel. Not his first published novel, but his first written novel. In the foreword, Scalzi notes that he wrote it as an exercise to see whether he could indeed write a novel, and never planned to even attempt to publish it. But he had a lot of fun while writing it, and when he was done, sent it around to see whether it could indeed get published. What happened was that it got roundly rejected, and he posted it online for everyone to read. As it happened enough people liked it that he got about $4000 in donations from various folks who enjoyed it.

Naturally, once Old Man's War was published to both commercial success and critical acclaim, a publisher stepped forward to offer to publish this book. Now, if you've read Scott Adam's column in the Wall Street Journal about giving stuff away on the internet, you will see that Adams, who had considerably more clout than Scalzi did, apparently did not negotiate with his publisher to keep his old free blog entries available on the internet, but agreed to take them down. Scalzi, who's probably not in the same category of wealth that Adams is in, apparently thought enough of the internet to keep the product free while betting that the paper product would reach more people and still bring in money. I guess science fiction writers really are a forward thinking bunch of folks who aren't in it all for the money.

How does the novel read? It reads amazingly well. If I ever tried to write a novel, and it turned out of this quality, I would be very pleased with myself, even if it never got sold. The plot of the novel revolves around a Hollywood agent for actors and actresses who has just had the first big break of his career. His boss calls him in and asks him to represent the toughest sell in anyone's career, a bunch of space aliens who have visited the earth, but look like the blob and smell like fart, and want a way to introduce themselves to the Earth's population without immediate assumption of hostilities just based on appearance alone. And of course, who would know how to make a good impression but the best of Hollywood?

The plot unfolds in a straightforward manner, but not without showing off Scalzi's versatility. He can write dialogs, he can write from different perspectives (including the Protagonist's boss), he can write press releases (a whole chapter is written in press release form, which is entertaining), and he can even write funny scenes in a dry, straightforward fashion without giving the impression that he thinks constantly about how smart he is.

A very impressive first novel, and excellent airplane reading. Recommended.

Friday, December 14, 2007

Review: The Android's Dream

I seem to be on a John Scalzi kick this year, not just because his books are entertaining, but also because they seem to be exceedingly easy to get at the local library.

This book takes place on a future Earth that has had plenty of contact with alien species, but one that is not in constant war with them. The intention appears to be that of a humorous romp through a science fiction universe. Harry Creek, a genius-level programmer and investigator, has been happily under-utilized as a "Xenosapient Facilitator" for the Earth's government. But when a diplomatic incident threatens interstellar war, his old friend calls him in to help find a creature of the utmost importance, a sheep known as The Android's Dream.

Unfortunately, Creek's opponents are already well ahead of him, and soon, he discovers that there is only one specimen of that sheep DNA left in a living creature, who turns out to be a pet shop owner who had never been told of her non-human origins. To say more about what happens afterwards would give away the plot, though perhaps in the case of this book that's not all that important. Rest assured that aliens, criminals, replicant AIs and a church determined to make its kooky founder's fake religion's prophecies come true all play a major part, and the romp is in earnest, if not a little bit too kinetic.

Scalzi's sense of humor is prevalent, though nothing like the form he achieved in Old Man's War. All in all, while it's an entertaining evening's read, it is not as good as his best work.

Paul Krugman Visits Google

Almost everyone knows about the best perk at Google: the excellent food in the corporate cafeterias. But the next best perk is much lesser known, which is the authors@google program, where intellectuals from Neil Gaiman, Malcolm Gladwell, Bill Easterley, and Michael Lewis come by, give a talk about their book, get questions tossed at them by Googlers, and generally provide an intellectually stimulating environment that rivals or exceeds those of a major research university.

For the last 3 years or so, I kept bugging Hal Varian to grab Paul Krugman for a talk. Now Hal is fond of telling me that whoever I ask him to bring would be a bad speaker (though invariably he's wrong --- apparently the mere act of having to give a talk at Google brings out the best in people), but in the case of Krugman, his most response was, "I've invited him several times but he's always turned us down."

So when Paul Krugman came by, I made it a point to get to his talk, get his latest book signed, and ask a couple of questions (off the air, so don't expect to see me on the video). Hal and Meng convinced Paul to give a technical, economics talk, mostly on the housing crisis and what happened there. If you've followed his columns and blog, or read the excellent blog written by Calculated Risk, then none of it will be new. Nevertheless, several snippets Krugman said caught my ear:
  • I'm known as an international trade expert. But nowadays, when I visit conferences in other countries, people ask me "Why haven't you been writing about Brazil, or Latin America, or currency crisis." My response has been, "I'm trying to save my own damn Republic here."
  • We now have fair trade with China. We give them dollars, and they give us poisoned toys and fish.
  • I have varying degrees of indifference about the falling dollar. First of all, it can't cause a lot of inflation not just because in the grand scheme of things even the huge amount of trade can create at most one or two percent of additional inflation. There are a lot of studies that show that when the dollar drops, foreign countries reduce their margins rather than trying to reduce their market share, and that compensation effect can be as much as 40%. Basically, the prices are set by micro-economics, not macro-economics. Finally, because US corporations own so much foreign assets, when the dollar falls, their balances actually improve. Latin American countries like Argentina get in trouble when their currencies fall because their debt is denominated in dollars. Our debt is also denominated in dollars, but it's our currency, so what happens is we get debt deflation when the dollar drops.
  • The Republican nominees all have to say that we're on the downward sloping side of the laffer curve, meaning that if we cut taxes, tax revenues will go up. This is unprecedented. If you look at the 2000 election, even the Bush campaign did not say that for fear of looking irresponsible.
  • On the chances of a recession next year? I can definitively tell you that I don't know. I've been wrong on things before. For instance, the strength of the 2003-2005 recovery surprised me. And currently, the strength of consumer spending also surprises me. For some reason, nothing can stop the American consumer.
I asked my questions of him in a private session afterwards.
You've convinced me of Obama's incompetence on important issues. Are there any candidates you particularly like?
You will note that I didn't attack Obama until his campaign started bashing me. I did point out the flaws in his healthcare proposal from the very beginning, which is the free-rider problem when you do not have a mandate, which is other candidate's methods of making sure that you can't just not have health insurance,and then when you need it sign up for it because there's no penalty for not having insurance. That's what Obama's plan does. In any case, I can't provide endorsements, but my personal preferences are Edwards, Clinton, and then Obama, in that order. In any case, there's actually not that much difference between the three candidates, but what a lot of people don't get is that in this case, Obama is the establishment candidate --- compared to both Clinton and Edwards, he's has the least progressive agenda, because he's trying to be a uniter. Both Clinton and Edwards come out with good policy ideas, but Edwards comes out with them first, so I like him more.
What are the chances of getting universal healthcare after this election cycle?
If a Democrat wins the White House, and we have a Democratic congress, a little better than even odds. That's because all the candidates now at least have plans so they can hit the ground running. Clinton squandered a lot of time by waiting until he was in office before doing healthcare. Johnson signed his Medicaid bill 6 months after he entered office, because it was the first thing on his agenda.
In 2004 I asked Brad De Long about the election. He said well, if the Democrats win we'll have a competent government. If they lose, we'll all get rich by buying international stocks and emerging market stocks. As the last 4 years have shown, that turned out to be incredibly lucrative advice. Can you top that?
If Giuliani wins, you better stock up on canned food, guns and ammo. His campaign advisers are made up of people who were too hawkish for the Bush administration. Of all the Republican candidates, the most reasonable one is Mitt Romney, but I have a bias, because if he gets the Republican nomination, the Democrats have a better chance of winning.
Why do you think China is refusing to revalue their currency? It seems to be doing them a lot of damage, because of the amount of inflation it's generating in their economy.
Well, having worked in government I can tell you why. Things are going well right now. They've had that peg for years, and it works well, so there's no reason for them to change it unless things get intolerable. The last time something like this happened was when Germany allowed the Mark to float back in the 1970s. And for them an intolerable amount of inflation was 2 or 3 percent. But that was Germany, with its memory of the inflation of the 1920s. The Chinese have no such history and have much more tolerance for inflation than the mere 10% a year they are seeing.
There are many who think the CPI under states inflation. What do you think?
The CPI includes a basket of goods that include things like electronics which fall in price, whereas the prices that people remember are things like food prices. The cost of a Thanksgiving turkey is up 11% compared to last year, and that's what people remember.

I'll edit this post if I remember more, but nevertheless, I am very glad to have met the one guy who was willing to keep needling the Bush administration all throughout the last eight years, even at times when it must have been very hard to do so. He is as smart as his essays and columns make him look, and yes Hal, he was a great speaker.

Thursday, December 13, 2007

Gtags 2.0 released

Thanks to the hard work of Nigel D'Souza, gtags 2.0 has gone open source! We've actually been using this version internally since April, but it had so many dependencies on internal infrastructure that it took a full work term to untangle it. There's yet more cool stuff coming. The gtags mixer (introduced in this release) has the ability to index the files on your disk, so it is possible (though I don't know if it is easy) to use this without having a gtags server at all, which should make some small project coders happy.

I use gtags every day myself, and Google's code base is big enough that it is essential for me. I hope others find it useful as well.

Tuesday, December 11, 2007

Review: Topeak Road Morph v2


The Topeak Road Morph series of pumps have been a mainstay of cycling gear for the last 4 years, replacing the Mt. Zefal HPX series that was my trusty pump before that. When my friend Lea showed up on her fancy new bike, though, I took a look at the pump and said, "Is that a Topeak Pump?" It turned out that she had by sheer timing, managed to buy one of the first copies of the new revision of the Topeak Road Morph pump.

Well, what are the changes, you might ask, that made me enthusiastically run out and buy one as soon as it showed up at my local store (without a discount, even?). The obvious one is that the new barrel is narrower. This means that each stroke pumps less volume, and requires less arm strength, a good thing for many cyclists, even though my brother's been putting me through a weight lifting regiment in the gym over the last year. The second obvious one is that it's about 30 grams lighter. In a world where road cyclists frequently pay a dollar per gram for weight reduction, the upgrade makes sense from a weight standpoint (though truly, the first improvement is in itself dramatic). But the last bit of improvement was non obvious --- there's extra space on the pump handle so you don't mash your fingers when you enthusiastically pump away. This bit of ergonomic improvement was a surprise and much welcome.

Altogether, a worthwhile upgrade. Recommended!

Monday, December 10, 2007

The Asus EEE PC


I'm not normally a gadget freak, and it takes quite a bit for me to get excited about a gadget, and even more, to get out my wallet and plonk down money to get one. But when I heard about the Asus EEE PC, I knew it was exactly what I wanted in a personal laptop. No hard drive (4GB flash drive), fast bootup time, Wifi capable, and cheap ($409 shipped).

I got mine today, and I am absolutely astounded by how tiny it is. You think the pictures do it justice, but they don't. The first time you see it, you think: this could fit in a purse! This is my attempt to show it in scale, next to a stuffed animal and a throw pillow. To my surprise, it handles WPA wireless encryption just fine!

The keyboard is the obvious question, but I'm currently typing on the EEE PC as I write, and I find the keyboard much less annoying than I thought. I wouldn't want to write for four or five hours at a time on this keyboard, but I'm very pleased with it. In fact, I find the touchpad more of a problem, and I am definitely way faster on the EEE PC than I am on my blackberry.

I'm picking up a Windows XP Pro license and I will install that. After that, with Picasa I will easily be able to photo-blog from the EEE PC.

Highly recommended!Posted by Picasa

Thursday, December 06, 2007

Au Revoir, Mac OS X

18 months ago, I bought a MacMini. My travails were numerous, and occasionally, quite the bother. But I liked the Mac Desktop experience, and I liked the idea of running a UNIX-based OS underneath that glitzy UI.

But many things were wrong:
  • My cheap $100 Panasonic KXP-7100 wouldn't work on it. The duplex feature of the printer was something Lisa loved, and though we swapped in a Brother printer, she screamed and screamed until I brought it back. This meant that I had to run VMWare to print anything.
  • I discovered that Quicken on the Mac was unacceptably bad, so ran VMWare for that too.
  • iPhoto was too slow for my tastes, so I ran VMWare for Picasa.
  • Firefox kept crashing on Mac OS X, whereas on my work laptop running Windows XP, Firefox stayed up for days.
  • I had to run VMWare to load my Garmin GPS up with maps.
  • Similarly, my Logitech webcam wouldn't work with Mac OSX.
  • Ditto, my blackberry.
So I was basically keeping Mac OS X around for the nice desktop experience and Firefox and iTunes. But the last straw came when I bought an Infrant RAID array. The Mac talked to it just fine, but for large files, it would randomly disconnect. Furthermore, while the Windows box had an option to automatically reconnect to my Infrant box on login or startup, Mac OS X had not such option. Yes, I could edit /etc/fstab, but that's ridiculous for an OS whose reputation was for ease of use.

So there I was in this parallel universe from the one Mac users seem to love and talk about. My Mac was having all sorts of random problems, but Lisa's Windows box (and my VMWare Windows XP install) was happily churning away, taking all sorts of abuse. My Mac felt slow, but Lisa's laptop was happy. She'd complain about slowness every time she had to use my Mac, but we had pretty much identical hardware. Ok, my Mac had 2GB of RAM.

So today, I wiped my hard drive and installed Windows XP Pro on my Mac Mini permanently. What a relief. Picasa is so fast! Firefox is so fast! So au revoir, Mac OS X. Hello again, Windows XP.

What this experience has taught me about myself:
  • Others care a lot about coolness. I just want my computers to work with the hardware I buy. And price matters to me a lot. I'm not willing to buy a $600 printer just to get duplex when my $100 printer is working just fine.
  • I'm just a contrarian by nature. When everyone else wanted SUVs, I wanted a Mini-Van. Now everyone wants a Mac, I like my Windows XP. My switch back to Windows probably means it's time to buy Apple stock.
  • I can't imagine paying the premium for a Mac again. Next time, I'll build the machine myself or buy a beige box. I'm just not cool enough to own a cool computer like a Mac.

John Gruber on the Obama Health Plan

(Tip of the hat to Paul Krugman for pointing at this)

Relevant quotes:
She first points to the figure from the Insurance Research Council that states that 15% of drivers are uninsured. As detailed in research by J. Daniel Khazzom (paper available at here), this figure clearly overstates the rate of uninsured drivers by computing this rate as the share of accidents in which the driver did not have insurance. But since uninsured drivers are typically from groups that are more accident-prone, the share of accidents involving the uninsured will clearly overstate the share of drivers that are uninsured. Moreover, state reforms to improve compliance with auto insurance requirements have been very successful, with the rate of uninsured drivers (measured appropriately) in Georgia recently falling to 2%.

Mandates are clearly enforceable, and are necessary for universal healthcare to work. The Obama plan will not work. Obama is still better than a Republican, but at this point, he is the least good of the Democratic candidates, in my opinion, and is not living in the real world.

Monday, December 03, 2007

Changing my position on Obama

I was impressed by Obama on his visit to Google, but now that I've read Krugman's latest explanation of why Obama is weak, I have to say I am persuaded. Obama does not have the cojones or the intellectual horse power to figure out what's important in the economy.

My personal preference would be John Edwards, but Hilary Clinton will do as well. After all, the last time we had a Clinton in the white house it didn't suck too badly either.

Sunday, December 02, 2007

Book Review: The Conscience of a Liberal

I am a big fan of Paul Krugman, and consider him the most honest and accurate thinkers of our day. Many have called him shrill and partisan, but his record is impeccable: he predicted the bankruptcy of Bush's ideas even during the 2000 elections, and correctly called Iraq a fiasco even when it was unpopular to do so. And those who think he is partisan forget his earlier books, such as Peddling Prosperity, where he attacked the Clinton administration for what we now see as laughable minor inconsistencies on trade policy.

The Conscience of a Liberal tells a story as compelling as any fiction. Once upon a time, there was an incredible inequality upon the land. The rich were getting richer, and the poor were truly miserable. Yet the nation could not move from that ideological driven fixed point --- policies and political machines disenfranchised poor voters, and the wealthy used their wealth to buy the media to call any kind of social insurance plan communist and persuade Americans that any change to the laissez -faire economy would create doom for everyone.

Then, a massive economic crisis happened, discrediting the wealthy plutocrats and some courageous politicians created what would become known later as the Great Compression. Social insurance plans allowed everyone to be assured that they would not die poor. Medicare ensured that the elderly would have health care. Politicians, aware that corruption could undermine this plan, made government so squeaky clean that not a single blot could be found in the implementation of both plans. To everyone's surprise, the resultant, more equal society ushered in an economic boom that would last more than 30 years.

But the forces of darkness were not vanquished --- they only retreated for a while, and by the 1970s, the civil rights movement created a gap that they learned to exploit to appropriate power for themselves. The result is growing inequality, a reduction of the middle class, and many of the ills you can find out for yourself if you bother reading the news. But such ill-gotten power can only be maintained by distracting the public with wars and with the coalition of racists and religious extremists that is the modern Republican party.

Ok, that's my summary of the economic history that Krugman supplies, only he tells it in a much less fanciful way, and the facts and figures that you would expect from an economist of his stature. Krugman's prediction is that the American public is finally waking up to the bankruptcy of Republican and libertarian ideas --- even if the internet and the youth seem particularly susceptible to libertarian theology (if Salon.com's interview with Ron Paul doesn't scare you, then you've already drunk the kool aid and are no longer a member of the reality-based community).

Krugman thinks that once the Democrats/progressive movement regains power, the most important item on the agenda is to fix the health-care system and provide universal health-care. Properly administered, this will increase the public's confidence in government (after the fiascos of the Bush administration), and much as social security and Medicare secured the terms of the new deal for a whole generation, this ought to enable public discussion on what kind of society we ought to have. His explanation of what the problems in providing universal health care is as good as anything I've read, and a good exposition for those who have not seen these arguments and economic analysis before.

Krugman at the end of book discusses what a progressive government can do about inequality. He points out that family status is now so important in America that a rich dumb kid is more likely to finish college than students who finish in the top 25% of their classes but whose parents are in the bottom 25% of the economic ladder --- economic mobility is now higher in Scandinavian countries than it is in the U.S. If the Democratic party had an institutional system of think-tanks and policy apparatus, this piece of data would be all over the news and be pounded into you 24/7, but that will wait for a happier time.

Krugman ends with a declaration of what being a liberal means and why he's proud of it. I am definitely very proud to be on the political side of people as smart and diverse as Paul Krugman or Brad Delong.

Is the future that Krugman envisions possible? The piece that gives me the most hope is on page 159 of this book: By 2004, however, 76 percent of Americans saw significant differenes between the parties, up from 46 percent in 1972. Obviously, we cannot take a Democratic victory for granted --- the right-wing and libertarians still have an unrelenting grip on the media --- Fox News, the New York Times, and even CNN still pander to the Republicans. But as more Americans realize that they've been hood-winked by the Right, I have hope that even Fox News cannot fool all of the people all of the time.

This is an incredibly good book, well worth the time reading. I borrowed the library's copy but I think I should buy a copy for myself, to support Professor Krugman if nothing else. Highly recommended, and well worth buying at full price.

Saturday, December 01, 2007

Book Review: One Economics, Many Recipes

To be honest, I do not feel qualified to review One Economics, Many Recipes. It's written by Harvard Economist and blogger Dani Rodrik as a technical manuscript on globalization and economic growth --- apparently it was a collection of papers that have already been published, but properly organized into a theme with a logical progression.

Rodrik addresses many questions in this book:
  • Is there anything in common between the successful developing countries of the past 30 years?
  • Is the Washington Consensus legitimately useful for developing countries to follow?
  • Why was China's drive into a market/capitalistic economy successful, while Russia's was a disaster?
  • Can Economists provide meaningful prescriptions for developing countries?
These are very important questions, as the last decades have found the Asian economies growing by such large amounts that millions were lifted out of poverty, while African and Latin American countries seem to fall further and further behind.

The book begins with a bang, first demolishing the free-market ideology so often propounded by the proponents of the Washington consensus:
When Taiwan and South Korea decided to reform their trade regimes to reduce antiexport bias, they did so not via import liberalization (which would have been a Western economist's advice) but through selective subsidization of exports. When Singapore decided to make itself more attractive to foreign investment, it did so not by reducing state intervention but by greatly expanding public investment in the economy and through generous tax incentives.

This is important, because many market ideologues (The Economist being but one) frequently posit that the Asian economies grew not because of their industrial policies but despite them. Rodrik does a great job countering all those arguments, the best of which is a chart showing that the countries that did everything that the Washington Consensus prescribed did much worse at growing their economies than countries that attacked growth from a systematic fashion.

Rodrik then begins to provide tools for economists to analyze developing countries. The most important prescription would seem to be obvious: it is most important to remove immediate constraints to growth. For instance, Brazil and El Salvador took wholesale reforms of their economies, yet grew more slowly than the US because Brazil was savings-starved while El Salvador's problem was low returns to capital. Yet the Washington consensus prescribed the same treatment for both economies, neither of which responded.

By contrast, China's major constraint in 1978 was the lack of market incentives in agriculture. The naive Washington consensus view would have been for China to switch wholesale to the capitalistic system with land ownership. But Russia showed that that route is fraught with danger --- government insiders or powerful outside interests could easily outbid ordinary farmers and citizens, thus resulting in an unequal society that would demoralize and discredit the market based system. Instead, what China did was far more clever: they grafted a market system on top of the existing state-ordered system, allowing households who produced surplus goods to sell into the free market. This granted market-type incentives without disrupting the existing system, while slowly easing the entire country into markets.

The book is full of such examples that are both convincing and properly cited and researched. There are quite a number of equations and other mathematical constructs as well, so those who wish to dig deep in can do so. Rodrik does not pretend that the fast growing Asian economies are free of problems --- he acknowledges, for instance, that many such countries will eventually run into the constraints of their economic infrastructure and will hit a brick wall for growth if they do not keep their institutions and economic structures improving at the same pace as their economies. However, for most developing countries the first order problem of getting that growth in the first place is by far the harder problem, and Rodrik provides the tools for analyzing them, even if in many cases, it appears that the creativity involved in designing alternative institutions to the obvious free market approach has to be done by someone really much smarter and with more stake in the outcome than the average Western economist.

The second part of the book analyzes various industrial policies and institutions required for growth. One of the most important points Rodrik raises is that in many developing countries, entrepreneurship has positive externalities that are not captured by first movers. This justifies strategic government intervention in encouraging such entrepreneurship, and he provides examples as to how to do so. He does admit that government corruption and the natural incentives of a government can lead to an inability to sunset those incentives and benefits to failing businesses, but he points out that the Asian countries have successfully done industrial policies for the last 30 years, and that the ability to acquire such expertise is not restricted to Asia.

The last part of the book extends these concepts to the entire global economy, not just developing nations. What kinds of trade regimes are acceptable? Rodrik feels that the existing trade structures do not allow for national institutions or standards to stick, thereby eroding the global trade system's legitimacy. In particular, Rodrik advocates that nondemocratic countries should not be able to count on the same trade privileges as democratic ones:
Think of labor and environmental standards, for example. Poor countries argue that they cannot afford to have the same stringent standards in these areas as the advanced countries... Democratic countries such as India and Brazil can legitimately argue that their practices are consistent with the wishes of their own citizens, and that therefore it is inappropriate for labor groups or NGOs in advacned countries to tell them what standard they should have... But non-democratic countries such as China, do not pass the same prima facie test. The assertion that labor rights and the environment are trampled for the benefit of commercial advantage cannot be as easily dismissed in those countries. Consequently, exports of nondemocratic countries deserve greater scrutiny when they entail costly dislocations or adverse distributional consequences in importing questions.

In other words, the way we've conducted trade has not been fair to either the poorest of our citizens nor to the citizens of undemocratic countries that may have gotten screwed because of the almighty dollar.

I hope this review has given you an idea of what an impressively good book this is. I could probably read this book 3 times over and get even more ideas and impressively researched examples, but to do so would be to rob you of the pleasure of reading this book for yourself to get an understanding of what a really good Economist thinks about, and how the libertarians and free market-ideologues really don't have any real answers for the developing world. I paid Amazon.com price for this book because I couldn't find it in the local library, and it is worth every penny.

Highly recommended.

Book Review: iWoz

iWoz is Steve Wozniak's memoir, written with Gina Smith. One of the two co-founders at Apple, it is hard to understand what a pure engineer this man was and is without reading this book.

The book itself is well-written, with a light-weight reading style that's clearly designed to make non-technical readers comfortable. (Ironically, I wish he had chosen to keep the technical parts more technical because it would have made those sections of the book easier for me to read)

The early part of the book is easy reading, about Wozniak's early days in school, and how he met Steve Jobs and Captain Crunch and got into making Blue Boxes. His light-heartedness and spirited playfulness comes through in his pranks, and the man has not a single bit of malice at all in his body, not even when Steve Jobs screwed him over:
The whole thing used forty-five chips, and Steve paid me half the seven hundred bucks he said they paid him for it. (They were paying us based on how few chips I could do it in.) Later I found out he got paid a bit more for it --- like a few thousand dollars --- than he said at the time...

The important part of the book, about the Apple and the Apple II are very much worth reading. He really sets the record straight (e.g., he was the sole designer of the Apple and the Apple II, rather than it being a collaboration between him and Jobs). He also designed Apple's floppy drive in 2 weeks, and that included writing all the firmware for it. Make no mistake, this man was an engineer's engineer: he could design, prototype, solder, and then write the BASIC interpreter for one of the first personal computers, and still have time left over for a prank that no one could pin on him at a trade show.

Yet this man was more unworldly than a monk:
...this idea popped into my mind about two guys who die on the same day. One guy is really successful, and he's spending all his time running companies, managing them, making sure they are profitable, and making sales goals all the time. And the other guy, all he does is lounge around, doesn't have much money, really likes to tell jokes and follow gadgets and technology and other things he finds interesting in the world, and he just spends his life laughing...
Wozniak leaves you no doubt in your mind as to which person he is.

Yet a hundred pages down the road, you find the limitations of this mind-set:
During the time the Apple III was being developed, he thought we'd grown a bit too large. There were good engineers, sure, but there were a lot of lousy engineers floating around. That happens in any big company.
I can't imagine a single startup founder today taking that attitude. Not after Larry Page and Sergey Brin have shown that even at an incredible size, employee hiring is still something that you must spend time doing and worrying about, and that it is possible to keep standards incredibly high despite becoming very big. Today, Apple has a reputation for being an incredibly frustrating place for engineers to work at (even my friend the iPhone touch screen custom chip designer will admit that). But it does not have to be that way, and to a large extent, I think the fact that the engineering co-founder did not take an active management role has a lot to do with it. (An interesting note: Wayne Rosing, Google's first Senior VP of Engineering has a cameo role in the book as Wozniak's bosses boss. Rosing was instrumental in setting up Google's engineering culture, and is easily one of the smartest engineering managers I've ever had the pleasure of working with/under.)

The book ends with a chapter of Wozniak's advice to engineers. It is excellent advice. For instance:
If you're that rare engineer who's an inventor and also an artist, I'm going to give you some advice that might be hard to take. That advice is:Work alone.

There it is from the horse's mouth. Do not believe all those journalists who tell you that in this modern age, you need to work with lots of people to do anything significant. All good software is designed by one person working by himself. That is as true now as it was when Wozniak wrote Apple Integer BASIC.

When Wozniak visited Google, I got to ask him if he had any regrets about Apple. He gave me the ultimate engineer's answer: I wish we'd sprung for a better keyboard on the Apple II. A great guy, and this book's a great read. Highly recommended.