Hacker News Reader

Comment by aj7 | Hacker News Reader

What’s really remarkable to me is that senior Livesay, the same age as my mother, took a Fortran course in 1968, two years AFTER I did, as a teenager, at Stuyvesant High School in Manhattan. So the same phenomenon of coding teenagers [possibly] outdoing experienced workers was going on 50 years ago.

View on HN

I think more reasonably the same phenomenon of experienced workers needing to update their skills in order to stay relevant was going on 50 years ago.

I was 16 in 1986 when I started programming C. Replacing all the old fogeys with their Fortrans and Cobols and .. shudder .. A/PL .. This is not a new thing, this teenage thing.

Tomorrow I guide some teenage punk through the wasteland of mobile development. Undoubtedly he's gonna throw some new shit at me. Hah!

I'm baby-sitting Java devs as "senior" atm. They don't know much, but what little they know they're defending with fervor, like "REST", Spring crap, and rubbish pseudo-modularization using "microservices". Worst, taking random architecture astronaut blog posts as dogma, incapable of reflecting whether something makes sense for a given task ("SQL is too old-school and low-level (!)"). Makes you really think about what kind of graduates our universities churn out.

No phrase terrifies me so much as 'Entity Framework'.

Then better don't take a look at "Spring data" + crud apps build on top of it - as pointless a framework as it gets. Allows you to churn out a CRUD app within the time bounds of an "agile" sub task of a day or less, to then degenerate into a typical inefficient Spring magic trial-and-error project to fix for months to come.

Young programmers are a pretty good market to market to. They have lots of money, they're smart, they want new shiny things to show off to the other programmers at all costs, and they're extremely vulnerable to perceived peer pressure and social proof.

I'm sorry, do you have a specific issue with REST or microservices? Like all things, they have their place and purpose. Simply throwing quotes around them and using the fact that a junior dev recommended them as evidence against them is pretty weak, especially here on HN.

It's more the unscientific knowledge acquisition I'm having trouble with, and which I hoped academic education had eliminated. Like "best practice dictates every record must be exposed as REST resource" and the stupid "hermeneutical" discussion about REST semantics as if it were a revealed religion.

I was a very early proponent of REST in the Danish Government, for about two years before Gartner announced SOAP was dead and made a co-worker nearly cry to me at lunch and I had to make him feel better instead of doing my well-deserved victory dance.

That said when young people blithely get their smart faces on and start talking about REST I cannot avoid the old fogey suspicion that they don't know a damn thing they're talking about.

I think the root issue is that there is just simply too much to know and too little time to know it. Everything is completely reinvented every 3 years now it seems, so many of the successful young engineers I know tend to develop a herd mentality, just do what everyone else is doing. You don't need to understand it, just pick your favorite thought leaders, emulate them, and things will usually work out well enough.

I think it's OK not to know the best tools. There's a lot of things out there, a lot of the best ones take some time to learn, and bad ones can ride a marketting wave to stay popular for a good few years.

It's also understandable that once people find a tool which they like enough to sink some time into learning properly, that they then cling to it a bit.

But I'm inclined to think that anyone who chose Spring for that position either has pretty bad taste, or simply hasn't bothered to look at a second tool.

Do you mean APL? I know there was PL/I. I was 14 in 1978 when I bought a used Commodore PET with tape drive built in. I had purchased a book on DEC assembly and read it before purchasing the PET, and of course it wasn't applicable, however, books back then were written to be able to learn by reading rather than copy/paste. BYTE magazine would list the programs for you to type in. Ah, nostalgia in the morning!

I did indeed mean APL ..

http://www.computerhistory.org/atchm/the-apl-programming-lan...

Course work tends to focus on the latest fads in languages. While the language chosen for a project definitely matters, in terms of engineers, I've found the languages known doesn't matter nearly so much as the ability to learn and evaluate languages. I'm far less interested in whether someone knows Scala than I am in if they know why Scala can be a good starting point for microservices.

Reply on news.ycombinator.com