Hacker News Reader

Comment by nathan_long | Hacker News Reader

I wouldn't argue for ageism, but I would say programming is different from law because technology changes much faster than law. There aren't giant paradigm shifts in law every 10 years.

A new technology may in fact be much more productive to work with than an old one. Does someone with COBOL experience have an advantage or a disadvantage in learning Python vs someone who is just starting? Which is larger, the amount of useful knowledge they bring along, or the amount they have to discard?

There are sage old programmers that I look up to. But there are also a lot who just burned out and quit, tired of the new technology treadmill. I'm not sure which I'll become.

View on HN

The volume of small, incremental, changes in law is higher. Let me put it this way. The Dodd-Frank regulations that have come out so far in the last two years is probably the same number of pages, printed 50 lines per page, as the C++ source for Oracle's Hotspot JVM. And it's only 30% done. And there is no ChangeLog.

The pace of paradigm shifts is indeed much faster in technology, but I disagree there are major paradigm shifts in technology every 10 years. There are major developments within specific scientific domains like AI, image recognition, etc, but in software engineering itself there is little new under the sun. There is nothing in Javascript that wasn't done better in Smalltalk 30 years ago. CoffeeScript, C#, Lua, Ruby, etc? Derivative stuff that any Lisp-er plucked out of 1985 could pick up in a week or two.

I feel like a lot of what you're talking about is applicable to languages and technologies in general. Advantageous technological knowledge seems to be more about frameworks and domain-specific experience more than anything else; something that a Lisp-er from the 1980's will have quite a steep learning curve ahead for.

Not disagreeing with your point. Just pointing out something that has been bothering me a bit about the fragmented platform market.

Your field has a guild and you enjoy the benefits of a professional culture that is centuries in the making. Fully agree with your points regarding the comparative rates of change, and, the sad fact (isn't it?) that the lack of deep exposure to the existing body of work in field due to a lack of effective apprenticeship of the non-existant-guild gives rise to the "major paradigm shift" every 10 years.

Law has much faster changes than programming. In programming you can use 2 year old ideas and be fine, in law no so much. A single case before the supreme court can completely change what's legal. And just about every bill changes something significant for someone.

Ditto doctors, you are required to stay up to date, where programmers can come back after 10 years and learn on the job doctors need to pass a "Maintenance of Certification Exam" or break the law.

PS: Programmers do need to keep learning, but it's no where near as time consuming.

Law I think is a field with unique responsibilities for their members to stay current and keep learning, things do change fast and you have to keep up to stay effective.

Paradigm shifts every 10 years? Not really. COBOL to Python ain't no paradigm shift.

We also have the issue of "digital natives."

In programming, there is a difference in someone who learned computers in college, someone who learned computers in high school, and someone who learned computers in grade school. Then the next wave was kids who grew up with the internet. I suspect it will continue with kids who grew up with touch devices. I don't think there is a similar parallel in other industries.

I'm 37 years of age I've been programming since the age of 7 (1982), and programming for the Internet since 17 (1992); that's 30 years of software development and 20 years of Internet/network programming; mostly web and mostly for government or in commercial environments and in more languages and frameworks that I care to remember.

Today I consider my judgement, that is to say, my ability to forsee the path of a project, or anticipate and avoid complexity to be better than it ever has been in the past, and think this is one of my strongest and most marketable skills.

Yes I still learn new things (faster than ever), and yes I am still very productive.

Am I a "digital native"?

I don't really buy who "kids who grew up with touch devices" might make better programmers than the generation of people who built Windows, NeXTStep, etc, who are in their 50's and 60's right now. Hell, Knuth is 74.

The fundamentals of programming, as an engineering field, haven't really changed since the 1980's.

"Kids who grew up with touch devices" may not necessarily make better programmers, but there's such factor as being comfortable with tools, I guess.

It's like the language. You spend a lot of time observing and using the system (writing, speaking)—even without full understanding—and you get comfortable: you easily distinguish between characters, recognize speech, can quickly and legibly write. It takes much less effort to apply this knowledge to do successfully things that may require it.

Likewise, if you spend a lot of time around computers, it takes much less effort then to use them for many things, including programming, since computers (or, maybe in future, touch devices) are tools we use for programming.

Cannot emphasis that enough. I would even argue the opposite: people who are now older than 30 did have the possibility to hack much nearer to the hardware of the devices, and could understand how they work. Now, because the tech is both so well packaged and more complex than before, no one tinker with devices really anymore. Sure, there is stuff like Arduino, but this is really reserved to a couple hackers. You don't have a Arduino hacker in every classroom. So maybe the digital natives are more comfortable with SMS and social media, but that doesn't mean they do better software or hardware developers.

Would you argue knuth is a good modern software developer? Have you ever used latex? It's an outdated unmaintainable piece of crap.. Did he develop anything we would call good software in the past ten years?

Knuth is not a developer, he is a computer scientist. The standard CS career path is not yet well trodden, maybe we are all supposed to turn into managers and high-level researchers as we get older?

Knuth didn't develop LaTeX.

Reply on news.ycombinator.com