Wednesday, January 30, 2008

Give 'em Zell, Joe

Apparently Joe Lieberman is planning to attend the RNC if John McCain is the nominee (as barring a catastrophe is inevitable). I wonder if he's going to challenge any TV pundits to a duel as well.

Sunday, January 20, 2008

The $99,000 Answer

"Anything large and impressive, if it was reasonably new in appearance, was automatically claimed as having been built since the Revolution, while anything that was obviously of earlier date was ascribed to some dim period called the Middle Ages. The centuries of capitalism were held to have produced nothing of any value." - 1984, George Orwell

Watching The Honeymooners (or any 1950s TV show), there are two things that stand out. First, the episodes are much slower paced than anything on TV today; a scene that might get 30 seconds today gets 3 or 4 minutes back then.

Second, the differences in culture are striking. Have you ever heard of "Captain Video", or the DuMont network it was on? Specifically for "The $99,000 Answer", it's striking to note that none of the songs mentioned as "Popular Songs" are songs that anyone will have heard of today. "Don't Fence Me In", Cole Porter? "Shuffle off to Buffalo" from 1938? "Just Too Marvelous for Words" from "Ready, Willing and Able" from 1933? "Take Me Back to Sorrento" from 1898? "I'll Be Seeing You", Irving Kale, 1938? "It's All Over Now", Bassy Simon, 1927? "Goodnight, Irene"? "Goodnight Sweetheart", Ray Noble? Of course, I can recognize Stephen Foster, which as the punchline is the only song Ralph doesn't recognize.

I'm not sure if this is normal cultural drift, if it's related to the dawn of records as a popular medium, or if it is because Elvis, The Beatles, and "Rock and Roll" made any popular music from before that time completely obsolete. It's almost that any music obviously from before the rock era is considered "classical", and demoted to an existence solely of folk music concerts and symphony orchestras.

Tuesday, January 15, 2008

News from Africa

Malawi ends ties with Taiwan in favour of China - The war of economic bribery between Taiwan and China continues to be won by China. The most recent change in diplomatic relations: Malawi, a small, poor country in Southern Africa bordering Zambia, Mozambique, and Tanzania. "Taiwan's hopes of maintaining relations with the Southern Africa nation were dealt a blow recently when the government in Taipei said it could not match a Chinese offer to provide $6-billion in aid to Malawi, one of Africa's poorest countries." The only countries that still have formal relations with Taipei are the Holy See, some heavily Catholic Central American countries, and some poor island or African countries (full list at Wikipedia).

An optimist might have hoped that the end of the colonial era, and then the end of the cold war, would bring a close to some of these shenanigans. Yet economic battles still rage across the African continent. Deals such as this one from yesterday "Tanzania: U.S. Firm Acquires Copper Concession" are routine. Most of these deals are a lot more boring than Mark Thatcher's involvement in a coup in Equatorial Guinea over oil rights, but the sum total of all the deals is a large numbers of foreign companies and governments trying to find the best way to make money and install friendly governments in Africa.

Sunday, January 13, 2008

Theory of Computer Science

Jeff Atwood at Coding Horror wrote a recent post on how we teach computer science (although "How we teach software engineering" would be a better title). His basic premise is that teaching quicksort, A* search, and Donald Knuth isn't enough to make a software engineer in the real world. There needs to be some mention of issues with deployment, source control, dealing with bug reports.

However, I find his conclusion pessimistic beyond the point of reasonability:

"Half of software engineering is pain mitigation. If you aren't cursing your web hosting provider every week, fighting with your source control system every day, deciphering angry bug reports from your users every hour-- you aren't being taught computer science."

No, this shouldn't be what software engineering is. Software engineering should teach ways to avoid dependencies and problems with your web hosting provider. A bad source control system is just that, bad, and should be fixed, not fought with on a daily basis. And dealing with angry bug reports isn't really a software issue at all, more an inter-personal issue to figure out what is broken.

Software engineering education should teach how to design software well in the real world. It shouldn't be about dealing with minor inconveniences that are best dealt with from common sense, that in many cases aren't related to software at all.

Obama on Iraq in 2002

I think the Democratic race is going to come down to the question "Do people feel Barack Obama can execute on what he promises?" Hillary has made the case that she can do what she is going to set out to do, but it frankly isn't a very ambitious agenda. She's also suggested that Obama is promising too much, which seems like a weak criticism. It is better to promise too much and fall short, than to promise too little and not come close to doing what should be done.

Via the Carpetbagger Report, I came across this extraordinary speech from Barack Obama in 2002 on the war in Iraq.

"I don't oppose all wars. My grandfather signed up for a war the day after Pearl Harbor was bombed, fought in Patton's army. He fought in the name of a larger freedom, part of that arsenal of democracy that triumphed over evil. ...

I don't oppose all wars. What I am opposed to is a dumb war. What I am opposed to is a rash war. What I am opposed to is the cynical attempt by Richard Perle and Paul Wolfowitz and other armchair, weekend warriors in this administration to shove their own ideological agendas down our throats, irrespective of the costs in lives lost and in hardships borne.

What I am opposed to is the attempt by political hacks like Karl Rove to distract us from a rise in the uninsured, a rise in the poverty rate, a drop in the median income, to distract us from corporate scandals and a stock market that has just gone through the worst month since the Great Depression.

That's what I'm opposed to. A dumb war. A rash war. A war based not on reason but on passion, not on principle but on politics."

Tuesday, January 8, 2008

How To Find Interesting Blogs

So, you've got a bit of free time on your hands, and want to find more interesting stuff to read about. You want to read more blogs. But the internet is a big place, so you're not sure what to look for.

1) Use an RSS Reader. This will literally double your productivity immediately, and increase it significantly more once you are used to some of the tricks of how to skip over posts. There are three advantages to having an RSS reader. First, you can't forget about blogs that are updated infrequently. Second, you only have to load one page instead of dozens or hundreds of pages to see all the new posts. Third, you get a "push" update system, where you can see what updates there are without having to check each blog separately. Nearly every website you visit has an RSS feed these days.

2) Focus on quality over quantity. This goes along with #1. If you have to check each page independently, it is easier to just read a few high-volume blogs than to read 50 that have 2 posts per week. But the right 50 low-volume blogs are likely to be multiple times more interesting and useful than one high-volume blog.

3) Don't get caught up on the word "blog". There's still a lot of buzz about blogs, but that doesn't necessarily mean that you care that it's a blog. A feed from CNN or the Wall Street Journal of the day's news stories may be more useful than a blogger's take on them. Slashdot was around before the word blog was, but that doesn't mean you would ignore it.

4) Start with a few high-volume popular blogs. In general, blogs tend to link to other blogs a lot. The advantage of this is that once you develop a critical mass of blogs, you can find more blogs relatively easily. The best way to do this is to get a list of "most popular" RSS feeds in a category, or to just do a search for interesting keywords plus blog.

5) Look at lists. The "Techmeme Leaderboard" may be gamed by Jason Calacanis and Robert Scoble, but it's a pretty reasonable list of the top "Silicon Valley/Tech Industry" blogs. The "Truth Laid Bear ecosystem" may have problems dealing with the fact that liberals and conservatives don't interlink (and by ignoring DU and FReep), but it's a great list of the top political blogs.

5) Start looking old school. One easy sign of quality in a blog is how old it is; if it existed in 2000 and still exists, it must be doing something right. There's also a lot more history to see whether it's worthwhile, and possibly a healthy community of commenters. With a new blog, it's a lot harder to tell whether it will be worthwhile or not, and in fact there's a better chance it will simply disappear.

6) Add link blogs. These are a great way of finding out what other people think are interesting.

7) Seek out detailed topics. If you are looking for a "technology blog", you're not going to find that much beyond Engadget and Gizmodo just from a basic search. However, if you're interested in developments on voice recognition in GPS units, you can do a more detailed search, and find out what people were saying last year. Past performance isn't a guarantee of future results, but people who predicted the present in the past are a good bet.

Thursday, January 3, 2008

5 Missed Trends From The Iowa Caucus

The 2008 Iowa Caucuses have come and past, with Barack Obama and Mike Huckabee the big winners. There isn't good news for everyone, with both Joe Biden and Chris Dodd dropping out. Here are some angles on the caucus that aren't getting as much coverage:

1) If Mike Huckabee ends up winning the Republican nomination, this ad should be put with LBJ's Daisy ad and George HW Bush's Willie Horton ad as one of the most effective political ads ever. You can claim that FairTax proponents and evangelicals are responsible for Huckabee's support, but the media seems to completely discount the enormous popularity Chuck Norris has. As far as endorsements go, only Oprah Winfrey is even close to having the same appeal as Norris.

2) The big story is that the youth vote came out in substantial force for Barack Obama, with him getting an astonishing 57% of 17-29 year olds. A less-reported story is that Ron Paul received 21% (about double his overall support) of the Republican 17-29 vote, putting him in a statistical tie with Mitt Romney for 2nd place. Ron Paul certainly has a base of core supporters that are generally dismissed as extremists, but he also has an astonishing amount of youth support.

3) I'm glad that nobody has tried to claim that Obama is from "Iowa's backyard" this election cycle. In 2004, everyone made sure to point out that Dick Gephardt is from Missouri, and that supposedly he had a significant advantage from being local. It was a silly argument then, and it would be a silly argument now. I hope this is a sign of a better press corps than in 2004.

4) The entrance polls have cross-sections by income, and the most interesting results are on the Republican side. Romney had plurality support among voters with household incomes of above 100k per year. However, both Huckabee and Ron Paul have substantially higher support among poorer voters. This may be related to their higher support among evangelical Republicans and young voters. However, this theory is discounted by the Democratic results, where there are no significant differences by income despite Obama's higher support among young voters. It's interesting to see that class is the big differentiator on the Republican side, but not the Democratic side.

5) The popular meme in 2004 was "electability", a word which has only a vague meaning. In 2008, electability wasn't a popular term, but some people still picked it as their most important issue. Among Republicans concerned about electability, Romney is the most popular candidate. Among Democrats, Edwards wins electability voters. I'm not sure whether "electability" is supposed to mean "presidential looking establishment white male", but that looks like what it may mean to voters.

Tuesday, January 1, 2008

8 Predictions for 2018

It's the new year, the time for making predictions. Some interesting predictions for 2008 I've read include those of John Battelle and Joseph Weisenthal. Since there will be plenty of posts making predictions for this year, I decided to look a bit farther into the future, and make predictions for over the next 10 years; what the world will be like on January 1, 2018.

1) Moore's law will start to slow down; instead of computers increasing 35-fold in computational power over the next 10 years, it will be closer to 15-fold. However, computers will use substantially less power. There will not be "AI" as envisioned in the movies, but computers will solve a lot more problems than they would be expected to do today.

2) For at least one year during this time period, Microsoft software will not be on a majority of new consumer PCs. In response to this, Microsoft will retool and finally come out with an OS that is secure, not bloated with features, and has a simple licensing scheme. Desktop computers will still be in similar form to what they are today, but they will be significantly cheaper, and displays will be larger. The "smart house" will not exist as prophesied where everything is on the network, but there will be, for example, wireless-enabled washing machines that broadcast when your laundry is done (Random Hall had something like this in 1999).

3) The continually increasing number of legal issues related to content creation (like indefinite copyright extension, excessive copyright issues and patent trolls) will cause the current system to break at some point, with a significant retooling of laws allowing for more usage, not less. File sharing will cause the RIAA and MPAA to have to retool their business models significantly.

4) Passive data collection and sharing will increase substantially. There will be a "Web 3.0" bubble based around allowing you to take advantage of data in your life in new ways. You will be able to track everything you do in minute detail. This will cause concerns of "Big Brother" to explode into the national consciousness. However, problems will arise not from the government, but from stalkers and extremists. These will generally be solved by privacy restrictions and decentralization, but there will be a significant portion of the public that avoids these types of programs completely.

5) The transportation grid will look substantially the same as it does today; there will be no flying cars, computer-controlled cars, or cars that go 200 miles per hour. The car will not be abandoned for travel by rail, on bicycles, or by foot. Congestion in cities will continue to worsen, even as taxes such as London's congestion tax become more popular. However, there will be flights from New York to Sydney that take only half as long as the flight does today.

6) The price of oil will be, after adjusting for inflation, around twice what it is today. At some point, perhaps not within 10 years, but certainly within 20, there will be an actual shortage in the US with gas lines, factories shutting down, etc. (This will not be a full-blown energy crisis, just a lack of petroleum.) Political concerns will prevent the government from doing anything significant to ameliorate the situation. There will not be a push to the electric car. The hydrogen car will be promoted, but it will not be economical because of systematic issues in the auto industry.

7) There will be substantial developments in solar panel technology. A solar panel using technology similar to photosynthesis will be in development. There will also be cheaper options to produce wind power. There won't be fusion power, though it will be "15 years away" instead of the "20 years away" it has been for the last 50 years. This will be sufficient to push the upcoming energy crisis at least 20 years farther into the future.

8) The US will have at least 15,000 forces in Iraq throughout the next 15 years. There will be a declaration of "victory" in Iraq (again). However, there will not be peace in the Middle East in general. The threat of "Islamic terrorism" will stop being the primary international concern in American politics. It will be replaced by either Mexico, China, or the UN.


My name is Alex Power, and this is my (new) blog. It is primarily about technology, politics, philosophy, and anything else that seems interesting.

About Me: I attended Cedar Falls High School in Cedar Falls, IA and Iowa State University, majoring in mathematics. Now I am living in Silicon Valley and working at Google.

Disclaimer: I am employed at Google. This blog does not in any way reflect on the views of Google. I will not directly comment on any product releases of Google. I will not comment on any actions related to the stock prices of Google or any of its competitors. I will also not discuss anything related to the upcoming wireless auction. Do not taunt Happy Fun Ball. No statements should be taken to imply anything related to any confidential information or upcoming product plans Google may or may not have. Void where prohibited by law. Some restrictions may apply.