Hacker News Reader

Comment by seanos | Hacker News Reader

I think the 10,000 hour rule is more applicable to skills that are somewhat stationary. Technology is a moving target, as new technologies are introduced some of the skills in which you were an expert become obsolete. On the other hand, something like playing Piano is a stationary target - if an expert in playing Piano travelled 30 years into the future, they would still be an expert in playing the Piano. So I think you have to be constantly learning and the most important skills of all are learning how to learn efficiently and having the ability to teach yourself.

View on HN

Well, you cited music performance, but so did the guy who did the research everyone who talks about "10,000 hours" is referencing:

"Even in music there is evidence for improved skill. When Tchaikovsky asked two of the greatest violinists of his day to play his violin concerto, they refused, deeming the score unplayable (Platt, 1966). Today, elite violinists consider this concerto part of the standard repertory. The improvement in music training is so great that according to Roth (1982) the violin virtuoso Paganini "would indeed cut a sorry figure is placed upon the modern concert stage" (p. 23)." [1] There are no stationary targets.

He uses this as evidence that the notion that "a sufficient amount of experience or practice leads to maximal performance appears to be incorrect" [2]

[1] http://projects.ict.usc.edu/itw/gel/EricssonDeliberatePracti... p.366

[2] Ibid.

But the actual task and skills haven't changed, just the elite performance level.

What kind of tech related skills that makes one obsolete pretty quickly?

Take Rails and Merb as examples. Both are using the well known enterprise patterns (despite the crowds hatred against the word enterprise, the ActiveRecord and DataMapper patterns came from Fowler's Enterprise Pattern book). These enterprise patterns dated back since early 2000. Unfortunately there are many developers who do not want to take their time to read books and decided to write their own frameworks.

Java is a 15 years old platform and these days it is still relevant. Hadoop, HBase, Cassandra, Hudson, and other infrastructure stuffs are being written in Java. Scala, Clojure and Groovy are written on top of JVM (and the idea of VM, which is roughly 15 years old or more, is still around).

LISP has been around since late 1950 yet these days some "elite" still considered the best language out there (or at least the paradigm of functional programming is considered the best one out there).

Object Oriented Programming has been around since early 1980 and I don't see it to be obsolete anytime soon.

The practice of data structure and algorithms (list, binary search, sorting, tree, graphs and so on) are still relevant until today.

The concept of distributed VCS has been around since early 2000 (probably even earlier than that) except that it's not F/OSS. (Remember BitKeeper?). The concept of VCS itself has been around since 1990 or maybe even earlier than that.

C and VHDL are still the strong players in embedded/system programming. Yes, there's "Go", but it's a newcomer/niche right now. People who program in C are still programming in C if they are in the embedded world.

Linear Algebra and Statistics which have been around since probably forever are still relevant today and I don't view them as a moving target.

Though I do understand your point of view of certain technologies that are considered as moving targets. Here are a few examples: .NET related frameworks (WinForms to WPF, WCF, WF, Silverlight), C# from OOP to mixed with functional, Flash to Flex/Air, Windows specific development, JS/HTML/CSS.

Though one could argue that since 1996-1998, there aren't too many changes in the HTML/CSS specs except between HTML4 to HTML5.

Another point of view as to why Technology is a moving target is that if a programmer/developer is moving from one ecosystems to another: Embedded to Web development, Desktop App to Embedded, etc. But one could argue that it's like a Pianist decided one day that Guitar is his/her next playground.

Yes you put it much better than me. I agree that many of the underlying abstract aspects are largely unchanging, it's just the technologies that shift.

If I were the best piano player in the world today, I'd probably still be one of the best in the world if I time travelled to 2030. On the other hand, if I were to have brain surgery in 2030 I probably wouldn't want the world's best surgeon from 2010 to time travel and do it.

> On the other hand, if I were to have brain surgery in 2030 I probably wouldn't want the world's best surgeon from 2010 to time travel and do it.

Or even, would you rather by a book on Java written 15 years ago, or written yesterday ?

An interesting question might be: I decide I want to be an expert racing driver and I know in order to achieve that I need 10,000 hours of deliberate practice. So I look at the controls and I see that I have a clutch, an accelerator, a brake pedal and a steering wheel - I decide to spend 2500 hours of deliberate practice on each. After 10,000 hours I am an expert in accelerating as hard as possible without spinning the wheels (accelerator), heel and toe gear changes (clutch), braking as hard as possible without locking the wheels (brake) and following the optimum racing line (steering). The next day they introduce semi-automatic gearboxes, anti-lock brakes and traction control. Some of my 10,000 hours were spent on obsolete skills. Am I still an expert?

Touche; If you are racing manuals (or your codebase is jdk 1.0 er.. what it was 15 years ago), then no.

On the other hand, if I were to have brain surgery in 2030 I probably wouldn't want the world's best surgeon from 2010 to time travel and do it.

I agree with your broader point, but your specific example is unfortunate. Surgery, and brain surgery in particular, has a lot more in common with the 19th century than the 21st. A disturbingly large amount of brain surgery is still accomplished via drill and very tiny ice cream scoops. I'd be surprised if that changes very much even in 20 years.

Reply on news.ycombinator.com