undefined | Better HN

0 pointsmatte_black8y ago0 comments

In general plumbing is considered the “blue collar work”. No one would put a boot camp front-end JavaScript grunt in the same league as an MS or PHD guy doing artificial intelligence research, compiler assembly, or reverse engineering malware.

And it seems ridiculous at times to even throw around the title of “Software Engineer” when the field has no standards of certification or regulation like other engineers. The only distinction from the programmer and engineer is the engineer makes architectural decisions, and the larger the scale the more accurate the title. “Plumbers” are cheap and no one cares if you fire them.

0 comments

pjc508y ago

Interesting that you put those three together; while cutting edge AI research might require that level of academic background, I wouldn't say that compiler work does. It's more within the reach of an undergraduate degree course. And malware (for and against) is a completely different field altogether, full of people with unconventional backgrounds and a large chunk of poachers-turned-gamekeepers.

I think it will be a long time before the field slows down enough that standardising it will be viable.

blt8y ago

I don't agree. The typical undergrad CS program doesn't prepare people for AI research, but that's because it dedicates lots of time to discrete/combinatorial stuff and low level systems knowledge. Drop/condense architecture, OS, compiler, exotic data structures and algorithms, networking, programming languages. Add more calculus, linear algebra, probability, optimization, statistics, symbolic AI. That undergrad would be equally good at AI work as a traditional undergrad would be at compiler work. AI field is rarefied now but will not be so forever.

matte_blackOP8y ago

The point is there was once a time when sprinkling some HTML and putting a site on the web automatically made you a wizard, and if you were young enough you might even be hailed as “the next Bill Gates”.

Those days are over. When society was illiterate, people who could simply read and write might have been held in the same prestige as those who write novels or manuscripts. As literacy grew however, so did society’s ability to distinguish between skill levels. The same will happen with code.

I remember when mobile development was at its hottest peak, declaring I was an iOS developer practically made people bow down in awe and throw offers my way for help with developing their mobile app idea (usually in exchange for equity or “revenue share”). Nowadays the field is so commoditized I don’t even mention my 8 years experience with iOS except in passing conversation.

A computer science degree is enough to call yourself a software engineer because most people can’t tell the difference these days. But for people who know the industry, a front end dev whose job is to basically push pixels on to a page is hardly an engineer, and I’d say is our modern version of a mid ‘00s website designer.

Having a timeless standard for what makes someone a Software Engineer that we can all agree on and can be verified by third parties would be helpful. Naturally, this will be met with resistance because there are many people who will not qualify, and who do not want an engineering license that would require them to be liable for their work.

cr0sh8y ago

> Having a timeless standard for what makes someone a Software Engineer that we can all agree on and can be verified by third parties would be helpful.

I would think that if this were possible, it would have happened by now.

Programming as a profession - while not as old as other "engineering professions" - is much older than what you are insinuating here; for instance, COBOL dates from 1959, FORTRAN is a couple years older, and there are a few older languages before that.

https://en.wikipedia.org/wiki/History_of_programming_languag...

But let's use COBOL as a "standard" - since there is a ton of COBOL out there still running and being maintained (and probably more than a bit being created "new"). That's almost 60 years worth of commercial software development (using various languages starting with COBOL).

If a standard could be considered and developed, it would have likely been done by now.

There are more than a few arguments as to why it hasn't, but one of the best ones I have read about has to do with the fact that our tools, techniques, etc - seem to change and are updated so quickly, that standards for testing would become outdated at an insane pace. An engineer certified on a set of standards might be obsolete before he left the building!

Ok - somewhat hyperbolic, but you get the idea. For the "classic" engineers, their tools and techniques don't change much if at all over long periods of time - so certification is more straightforward. For some engineering professions, you can pretty much take someone who was certified in 1970, and be pretty certain that he or she would be able to do the same kind of work today. That would definitely not be the case for a proverbial "software engineer" certified to the standards of that time...

j / k navigate · click thread line to collapse