Hacker News Reader

Comment by xyzzyz | Hacker News Reader

When professional certifications are required to write car software, you won’t have top tier people writing it. Instead, you’ll get mediocrities who will jump on the certification as a way to secure a sinecure at a company requiring it.

Consider existing professional certifications, like, for example, Cisco certification. How many top tier people have it? Some do, but in my experience, not many. Instead, they are overwhelmingly held by low-to-mid skill people, who see it as a ticket to a better job. Top tier people can easily get top tier job, and have no need to waste time on some certifications.

View on HN

I think you've missed the point. A professional engineer isn't just a certification. It also grants the power and indeed the duty to say "no" to things that are unsafe. If a cost-cutting change is proposed that means the bridge might collapse, you say no.

But in software? If management says you need to cut corners to get something out by an arbitrary deadline, well, you cut the corners or they'll find someone who will.

Well said. And if you fail to say no to something that is unsafe, you can be sued for malpractice. The fact that management pressured you is of no consequence.

In theory, yes. In practice, probably not. Most stuff that "falls over" after being built is due to poor maintenance or skipped checks. I never read any news about "PEs" (people seem to be drooling over that title) being sued.

I like how a city engineer at little rock put it to me one time. “Having a license does not mean you have all the answers, but you are kind of responsible for them.“

> A professional engineer isn't just a certification. It also grants the power and indeed the duty to say "no" to things that are unsafe.

At the end of the day, you still work for them, so with enough "nos" they'll just fire you or assign you to projects which don't matter.

> you still work for them, so with enough "nos" they'll just fire you or assign you to projects which don't matter

They'd have to replace you with someone else similarly certified.

And surely, there will never be anyone who’ll just cynically use their professional credential to make a lot of money this way. Of course, there are plenty of people who do shady things like this today, but clearly obtaining professional certification will make angels out of humans.

If you lose your PE because you do what you suggest you lose your ability to work. Your career ends, formally.

Software certifications, afaik, are not like this, but instead are sort of like merit badges.

If. Many people will take a risk, as the risk will pay off in most cases that aren't totally egregious. Most shady scenarios are not "people will certainly die if I sign off on this", they're rather "eh, this is somewhat against the rules, but it's not that important, it just increases the risk from one in 1000 years to 1 in 300 years". People who will accept the risk to their personal careers will be in high demand as "people who make projects happen", and they'll be highly renumerated for their increased risk profile.

In any case, if PE was required to work in software, I wouldn't have had a job in the first place, so I'll always oppose these kind of efforts to shut out the outsiders and underprivileged people.

I work as an electrician in industrie automation. To say "no" is very important. It's not easy, but I don't have to worry they ever fire me for my "no" because they know I'm right.

As an electrician for example, we have to check almost everyrthing with various tests and sign it with my name.

For me, the software world still looks and feel like wild west.

A PE at an engineering firm is a bit like a partner at a law firm. They carry a big stick.

And get paid 1/10th the amount of money? Don't forget that.

A PE is not a certification, it's a license, legally-recognized by the state. (Like other licenses, such as for mental health counseling.) You have to take 2 tests, one to start, and another to finish, after 3 years of professional work. As with most engineering students, I took the first part (the EIT), but never took the second, as my job never needed it. I wish I had, though. It would have been a nice feather in my cap.

In any case, I can ASSURE you that it was comprehensive, and you have to "know your stuff" to pass it. I've been Microsoft certified, so I've seen this side too. If there were a similarly-reputable license in software engineering, there would NOT be "certification factories" to crank out people who can pass the cert test, and who can do little else. I don't know what the analogue for doing actual mechanical/electrical/civil engineering problems would be in the software engineering realm, but that's the KIND of thing that would have to be tested. It would require a working computer and internet access, and not a duffle bag of engineering textbooks, like I had. ;-)

I think certification is secondary. The most important thing is liability. The problem is that software problems are ephemeral and often gets disregarded. If a software is slow, people seems to just live with it and/or just buy faster computers. Just think about how weird it is if things just randomly fall of, it happens so irregularly that I don't expect them to happen. This is simply not true of software.

If software engineers in a specific field (automotive) were held liable for bugs in their software, no one would want those jobs. Would you?

then why would engineers in other industries accept theirs? IMO software should move a lot slower, but the economy does not work that way. There's way too many experimental things that go live in production.

I think it was either Bjarne Stroustrup, Scott Meyers, or Herb Sutter who said that if we "went slower" (took less risks / program more like embedded / NASA / JAXA / ESA), we would still be living in the computing stone age. The real trick is to take calculated risks given your business environment and constraints. Usually, on any given team, there is one person who thinks they are "flying to the moon" with their accounting software (or whatever). And their pace is ridiculously slow, and code vastly over-designed and tested. They are always planning for once in 10,000 years events.

> If software engineers in a specific field (automotive) were held liable for bugs in their software, no one would want those jobs. Would you?

If almost no one wants them, then they would pay very well and be in high demand, and that comes with lots of benefits and perks, and you can require the use of the best tools to ensure those disastrous outcomes don't happen. So yes, I'd take that job.

(I am not disagreeing with your claim): This assertion is a bit at odds with the notion of certifications, but only when considered without other factors. Economic theory suggests then the achievement of the certification is not costly enough to cause a separating equilibrium of talent versus not.

Reply on news.ycombinator.com