Archives for the Review category

The Black Swan

Black swanI just finished reading Nassim Nicholas Taleb’s The Black Swan: The Impact of the Highly Improbable. I enjoyed Fooled by Randomness by the same author, and Black Swan has a few references to his previous work. In the previous title Taleb discusses a typical human mistake of taking the result of a random outcome for skill or deserved result. In Black Swan, Taleb discusses traditional models for forecasting anything (financial markets, political scenarios, weather), and theory of chaos, which basically says that systems with high entropy cannot provide any degree of predictability.

Political analysts of the late 80s did not predict the collapse of the Soviet Union, financial markets of the 21st century did not predict Russian default with subsequent sweep on Asian markets, and security analysts did not predict two planes flying into the towers of World Trade Center. However, in the retrospect, we tend to exhibit surprising hindsight, stating that all those events were inevitable, due to [insert a variety of geopolitical arguments here].

Taleb insists that the human mind frequenty:

  1. Overestimates the odds of success when applied to oneself. Everybody buying a lottery ticket mentally assumes a much better probability of winning than reality usually presents.
  2. Attributes random events in the past as contributing to success/failure. Survivors of the shipwreck would tell the story of how they all prayed for their lives, and thus got saved. While some religious groups would tend to pick up the story as the proof of whatever agenda they’re pitching, nobody generally listens to the stories of those who prayed, but still drowned. Business journalists (with CBS MarketWatch being the worst) frequently abuse this by pitching headlines like Stocks down after Congress increasing military spending in the morning, and Stocks up as Congress approves larger military budget in the afternoon.
  3. Looks for order and sequence where it doesn’t exist. Business book writers made a cottage industry out of this by surveying the life stories of prominent individuals, and then reselling those as recipes for success. The quirkier the trait, the better it suits the public. Peter Thiel wakes up early and runs in the morning - I see, so that’s a secret recipe for running a successful hedge fund.
  4. Listens to the experts that don’t know any better. There’s a whole bunch of occupations, where expertise is learned, and re-learning it requires significant time investments. Mechanics, doctors, and foreign language interpreters all have such skills. Other kinds of expertise involve looking back at what happened and trying to draw the line of correlation among discordant and random events. Such experts involve financial analysts and government economists.
  5. Hungs up on small things without seeing the big picture. If you have significant money in savings accounts, you’re probably busy looking for a better rate. However, the bigger picture, the Black Swan of the US banking industry, is significant asset consolidation among major banks. If one of them gets hit, all of them get hit significantly. The possibility of over-exposure to subprime loans and possible collapse of the banking industry generally escapes the model that your financial analyst presented.

The book is pretty interesting, but hardly coherent. Nevertheless, it’s full of interesting anecdotes and personal observations, that make the book somewhat witty. After reading it I went around reading some other reviews - NYT seemed to not like it, Fool.com did a pretty comprehensive review of the points presented, Slate reviewer seemed to like it. There’s also author’s interview on Colbert report.

The starfish and the spider

Starfish and the spiderFor a highly acclaimed book, according to the number of accolades on the back cover, The Starfish and The Spider is surely repetitive, but makes for a quick read. In a nutshell, the author applies the analogy of starfish and spider anatomy to the corporate world. While spider has many legs growing from the center, cut off its head - and the whole organism falls dead. With starfish cutting off one of the legs won’t achieve any fatal scenarios - you will just get two starfish organisms, as the anatomy is completely decentralized. Apply that to the corporate world, and you have two kinds of companies - one where the head directs everything, and such organization, when made headless, turns into chaos. And a company where every employee is so autonomous, that removing the person on the top, as well as layers of management organization would damage the organizational body somewhat, but won’t be fatal.

Authors take a look at Kazaa, Skype, Alcoholics Anonymous, Craigslist, Wikipedia, eBay, Apache Indian tribes, and Toyota as examples of successful decentralized organizations. Successful in different respects - The eMule Project, covered in the book, doesn’t earn any revenues per se, but manages to sustain all sorts of attempts to exterminate it. Wikipedia is not a money-maker either, but undoubtedly successful as far as the quality of content is concerned. With the Toyota example, the authors point out GM’s Fremont, CA plant as an example of highly centralized, and unsuccessful, project management approach, and compare it to Toyota’s decentralized and highly autonomous approach, where bottom-up innovation is welcomed. The case of the auto plant was apparently covered in the movie Gung Ho, but haven’t seen the movie, I can’t really comment on that.

The authors fill some pages with interviews with Craig Newmark and Jimmy Wales. Valuable content otherwise, it doesn’t really add much to the book - we learn that Jimmy Wales doesn’t do any article editing himself anymore (so they’re even more decentralized than you might have thought), and that Craig Newmark is very user-centric and responds to customer support e-mails. The authors also make a passing remark on the cases when the centralized structure is preferred (U.S. Army, commercial airline, etc.), but don’t really expand on this point, so you’re sold on the idea that decentralized management is applicable to every single company out there.

Sometimes the authors seem to confuse decentralization with just employee empowerment. They quote Jack Welch and current GE as an excellent example of decentralization, while in reality it looks like the corporate structure is still pretty rigid - employees just got motivated to make more decisions themselves, instead of relying on management to tell them what to do.

Overall, the book exposes a good idea, but seems to be hung up on the idea of making the right number of pages to publish a book. Reading this review provides a pretty good overview on what to expect in nine chapters.

The Naked Guide to Bonds review

I always wanted to learn more about the bond market. While the stocks are pretty straightforward, bonds always have this air of mystique around them. Yields, treasury rates, bid and ask spread, ability to avoid taxation - it seems like there’s a special little market of its own. The Naked Guide to Bonds by Michael V. Brandes provides a pretty good and straightforward explanation of how the bond market operates.

US bond market lacks the degree of transparency that US stock market has, and since there’s no New York Bond Exchange, there are certain peculiarities a bond investor should consider. Some of the bond offerings are hard to sell, as they’re generally targeted towards the buy-and-hold crowd that wants fixed income, not opportunities for trading. Nevertheless, the bonds are traded, with prices determined by interest rates and market demand. Since most of the institutions in the market operate with the budget of $1 million and above, there’s actually a separate submarket for anyone trading in the lots of fewer than 1,000 bonds (pretty much all of them are priced at $1,000 to make at least one factor comparable). Effectively you might have two separate markets with two different prices quotes for the bond - a market for institutional (above one mil) and small investor (below one mil) trading. Bad news is that smaller investors get penalized, as prices for buying and selling bonds favor large institutions.

One of the reasons you might be looking into the bond market is tax-free municipal bonds, which are exempt from federal income taxes, and exempt from the state income taxes, if the bonds you buy belong to the same state you’re in. Municipal bonds, typically with lower yields than corporate or agency bonds, typically include the buyer’s income tax bracket in calculating the total rate. The bad news is, as mentioned above, most of the municipal bonds are issued for the period of 20 years, and if you need to sell them before the deadline arrives, you might be stuck with an offer, on which you’re losing the money, since the market is not that saturated. It’s also important not to make a mistake of getting municipal bonds for an IRA account - you’re giving up high yields for no income taxes, but since IRA income is already not taxed, tax-free municipal bonds generally do not belong in retirement accounts.

The book is easy reading. Not light, but definitely straightforward if you’re paying attention to what has been described in the previous chapters. The author then discusses some strategies that might be applicable to the reader, and talks about the common pitfalls, such as chasing the yield, or having no sense of direction in the bond market. Among online resources worth checking out, InvestingInBonds is perhaps the most well-known official resource. Yahoo! Finance also runs a bonds center with basic market data and some tutorials on bond trading. Bloomberg bonds center offers some news feeds for the market. MunicipalBonds is also a great site for researching one particular subset of the bond market, providing both research and news on new offerings.

Tale of three 802.11 b/g USB wireless adapters

With frequent changing of apartments it’s always a crap shot on whether the next bedroom/office will have a cable outlet or not. I was briefly considering purchasing a wireless bridge for my desktop machines, but then figured out since 802.11 b/g USB drives were so cheap nowadays, I might just as well go with those.

TrendNet TEW-424UB wireless 802.11 b/g USB 2.0 wireless adapterTRENDnet TEW-424UB was the first choice, since pretty frequently on CompUSA Web site the adapter will sell for $15 with a $10 rebate. Rebates for TrendNet are pretty reliable and come in the mail, but the adapter itself seems to be a total opposite. Its coverage seems to be quite spotty, and on my wife’s HP laptop with the drivers pre-installed it kept being recognized as some RealTEK networking product. Who knows, could be a Windows XP issue.

D-Link DWL-G120 -- AirPlus Xtreme G Wireless USB Adapter 802.11g, 54MbpsD-Link DWL-G120 (AirPlus Xtreme G Wireless USB Adapter 802.11g, 54Mbps) is also quite frequently on sale from Buy.com, which in combination with Google Checkout’s $10 off for new customers would sometimes yield you a free product. The driver for this guy has not been certified for Windows, of which you’re warned on installation, but little you can do at that point. Lack of certification seems to be appropriate considering the occasional blue screens of death the driver causes. Windows Error Reporting then redirects you to a page with generic “driver failure” message, which doesn’t help much. There’s not an updated version of a driver, nor a Windows-certified one.

ZyXEL ZyAIR G-220 - USB 2.0 802.11G Wireless Adapter & Soft-APWhat seemed to work for me flawlessly (so far) is ZyXEL ZyAIR G-220 - USB 2.0 802.11G Wireless Adapter & Soft-AP, which I generally got on Buy.com. The driver installation goes through pretty well, there’s an additional utility for wireless network management you can install, if you opt to use another one outside of Windows’ default. Right now they sell it for $27 with a $17 rebate (which is trackable online, and which I got). However, on Buy.com checkout screen there’s also an option to get a free magazine subscription. Opt out of that, and you’re eligible to receive a $10 rebate, bringing the price of the wireless adapter effectively to $0 (except for sales tax in California). I have not received that rebate yet, and it does seem kinda shady without online tracking. Nevertheless, so far this adapter seems to perform the best.

Why software projects fail, and review of Dreaming in Code

Dreaming in codeI finished reading Scott Rosenberg’s Dreaming in Code the other day. A story about Chandler project by Open Source Applications Foundation and Mitch Kapor, the book, written by a non-engineer, really tries to provide perspective into software development world and tries to help the reader understand why software projects fail. Through some anecdotal evidence, we mainly find out about large publicized government software project failures. Most of the private failures remain undisclosed, as within large corporations projects never fail, it’s just their resources that get “re-allocated to higher priority projects.”

While Chandler is not a spectacular failure by itself, the book does talk about a few steps that were taken perhaps in the wrong direction. So can this knowledge be summarized into distinct no-nos for anybody participating on a software project?

1. From development project to integration project. Engineers like the concept of reusability, and so do their managers, for whom the the choice between developing their own in-house product, or using a third-party (sometimes open source) solution, is usually straightforward. Unfortunately, few pieces of software integrate smoothly, and if you decide to go with third-party or open source solution, most likely you do not have people with complete expertise in it. Not to reinvent the wheel, engineers of Chandler project were discussing Zope Object Database, but as the product requirements changed in the planning stage, so did the requirements towards ZODB. No one on the team was a ZODB expert, so most of the discussions required further “looking into” ZODB internals.

2. No agreement on final architecture, or inability to settle on one. This seemed to be the case with Chandler team, all consisting of superstars. Having a dream team for the project has one disadvantage - everybody in the team respects the opinions of the others, and if dissenting views are presented, occasionally it’s hard to move forward without hurt feelings. Sometimes people in the room are thinkiing “I disagree, but what do I know”, leaving the discussion up in the air with no decisions made.

3. Feature creep. One of the problems that Chandler had outright was its need to be revolutionary, which required a redesign every now and then. As more things were abstracted (an “item” could be an e-mail, a note, a task on the to-do list, an entry in the address book, etc.), there were more features to add into the product making the next release unattainable. In the software industry, the term feature creep refers to more requirements entering the scene, making a release a highly unlikely event.

4. Poor reusability. Most of the engineers who have been working in the field for a while probably keep a personal library of routines they reuse over and over. So code reusability on a personal level is pretty good. Code reusability within organization is sometimes pretty decent, depending on how the CTO or VP of Engineering sets up the code repository and what practices are followed. Code reusability in the open source world is pretty poor. Even though there are quite a few well-written libraries out there that integrate well with other’s projects, most of the stuff requires heavy customization, if you’re lucky enough to get it in the language that you need. However, Dreaming in Code maintains that perhaps the reason you started writing your own project was to develop some new technologies, not to combine the existing ones into a single package. Therefore integration pain is the price that has to be paid for a final product to be worthwhile.

Fooled by Randomness

Fooled by RandomnessNassim Nicholas Taleb’s Fooled by Randomness explores a quite interesting idea of how frequently we confuse causation with pure randomness exhibited in nature. We tend to rever successful war generals, coaches, and corporate CEOs, studying their strategies in search of recipes for success. Success, as Taleb infers, can frequently be attributed to a totally random sequence of events.

Suppose that you have a group of 1,000 who are ready to play a game of Russian roulette once a year, with their chance of survival at 80%. You would have 800 people left after the first year, 640 after the second year, 512 after the third year, and so on. Sooner or later the randomness of the game would bring you to the point, where you had a dozen or so people left alive. Are they really good at plying Russian roulette? Do they have some winning strategies for beating the game? Should we study their family history and consult with their parents on upbringing, so that we can have more of their type? It’s pretty obvious that their path to success in this case was completely random - someone had to be among the final dozen, given the probabilistic chances of Russian roulette, and it just happened to be them.

Change the rules of the game from Russian roulette to financial markets, and all of a sudden the remaining few are treated as financial geniuses with “proven track record” and unblemished reputation. This is frequently one of the randomness scenarios that human brain tends to attribute to some past events having a historical value on influencing the present status quo.

Not to say that success is completely random. When you look at the world of financial markets, somehow one can bet that a Harvard business graduate with superb math skills and understanding of international markets will do a better job on Wall Street than a bum from Calcutta. It’s also quite expectable that a dental student graduating at top of his class from a top school is most likely to do pretty well (financially) in life. Randomness? Or deterministic behavior? Taleb argues that one can always position himself to get a better head start at an occupation, but unless you’re an Olympic runner, randomness has probably a higher role in determining your success, than skills set.

But the book is not about “defining your future.” It’s more about common misconceptions that we have about success. Most of the actors work as waiters, and most movie script writers work at fast food. Yet the public tends to perceive the job as highly desirable because of the slim majority, who actually reached the top of their profession. Human brain tends to underestimate the chances of success, when the potential payout is so large, that it outshines the rest of the 99.9%.

SearchSIG on personal search

Spock, ZoomInfo, and Wink presented at tonight’s SearchSig. The event was hosted by Google on its Mountain View campus, and moderated by Michael Arrington of TechCrunch. Each startup presented their own vision of personal search, with Spock collecting all sorts of personal information from public Internet sites, ZoomInfo crawling various directories and corporate sites in order to create a business-oriented people directory, while Wink is also parsing all sorts of public sites in order to aggregate a single profile, which they then allow the user to own.

If you like Arrington as TechCrunch writer, you’d definitely love him as panel moderator. He’s not confrontational, but he definitely cuts through marketing bs in order to get his questions answered. All participants kinda stumbled around monetization, but then agreed that currently they maintain somewhere between $1.50 and $2.00 CPM. ZoomInfo also sells premium subscriptions to some of the business-related information, and currently is profitable.

Not too many people are sure how to monetize the people search - you generally can show some contextual ads on the celebrity profiles, since you roughly know what the visitors are looking for, but search for a relatively generic name (your former coworker, classmate, etc.), and that opportunity is hardly monetizable. The Wink demo was particularly attesting to this, as the CEO was browsing the site, the only contextual ads that would show up on the right would be dating ads served by Google AdSense.

At the same time, many agreed the opportunity is there - anywhere between 1.5 billion and 2 billion searches a month are for people. If you saw Dustin’s slide from Facebook tech tasting, you know that Facebook alone generates 600 million searches a month (with actual share of people searches not being disclosed). Spock seemed to think it’s going to be great to allow people to tag other people, and Arrington pressured them into the scenario when someone would be tagged as “pedophile” or “unethical”, at which point the CEO did a little of arm-waving, referring to the “community process”. I want to see that tested when thousands of diggers would get a chance to tag anybody employed with RIAA/MPAA, or thousands of slashdotters get to tag an employee of SCO or Microsoft. It didn’t look like anybody had any good idea on dealing with tag spam, malicious tagging, or misrepresentation by claiming someone’s profile on Wink or Spock.

Overall, looks like the industry is in fairly early stage, with more questions than answers. Pressured by big search engines from one end, and social networks from another, people search engines need to come up with some winning value proposition that makes customers either reach for their credit card, or spend more time on the engines themselves, consuming some ads meanwhile.

Online video editors reviewed

ExtremeTech reviews five online video editors: EyeSpot, Cuts, JumpCut, MotionBox, and One True Media. Their conclusion? “Jumpcut offers the strongest editing and enhancing tools of the services we tested for this roundup. Unfortunately, it’s still in beta, and we ran into some uploading difficulties.” EyeSpot wins the format wars with support for ASF, AVI, DivX, DV, FLV, MOV, MPEG, MPG, MP4, RM, WMV, 3GP, and 3G2.

5 AJAX frameworks reviewed

Dr. Dobb’s Journal reviews 5 AJAX frameworks: Dojo 0.3.1, Prototype and Scriptaculous 1.4, Direct Web Reporting 1.0, Yahoo! User Interface Library 0.11.1 and Google Web Toolkit 1.0. Each framework was tested in two basic scenarios - writing a “hub” (titled collapsible link list frequently seen on sidebars of many Web sites) and a “tab panel” (horizontal tabbed navigation bar). During the process, Dr. Dobb’s Journal reviewers noted that “Dojo provides more features and HTML widgets than YUI and Prototype” but eventually “settled on the Yahoo! User Interface Library”.

Blades of Glory

Blades of GloryGenerally I am not too impressed with Will Ferrell’s recent filmography. Talladega Nights: The Ballad of Ricky Bobby was tolerable, and one would get a few chuckles here and there, but it was from being a comedy. Stranger Than Fiction is not a movie to remember either. Blades of Glory, however, manages to be a pretty funny movie without a central theme. I was afraid the whole movie was going to be based on the fact of two guys skating as a pair, but the script is heavy with gags here and there, making it overall a quite enjoyable comedy. The movie it mostly reminds me of is Zoolander, except Zoolander wasn’t funny. The same “serious look” into the world of professional skating, with Farrell’s character maintaining the same manly character that Ben Stiller was exhibiting in Zoolander.

« Older Entries