Hacker News Reader

Comment by Rmilb | Hacker News Reader

Does anyone else see a future where programmers need to get a licence to practice their craft?

For engineers who have to sign off on matters related to regulatory bodies, that's been the case more or less forever. I assume that includes software engineers who approve software that has to be approved by regulators. It's certainly true of construction plans, etc.

In my first engineering job (offshore drilling rig design and construction), pretty much all the senior engineers had PEs.

> I assume that includes software engineers who approve software that has to be approved by regulators.

Not that I know of.

According to this article in IEEE [1] "Thirty states currently require licensure for software engineers working on systems that affect the health, safety, and welfare of the public as well as those offering their services directly to the public." That seems a reasonable source but I don't know how common it is in practice.

[1] http://theinstitute.ieee.org/ieee-roundup/opinions/ieee-roun...

Yeah, and thats when you know "tech" has actually changed the world instead of just imagining it does.

A civil engineer pointed out to me that the licenses come when it really starts to affect people and infrastructure. Programming does this by proxy, but not directly enough yet.

New York State's bitlicense was the closest to burdening software engineers with licensing requirements, until the one person at the head of the financial agency clarified it wasn't intended to be interpreted that way, and modified the wording of the law.

So one can't change the world unless lives are endangered. That doesn't seem very logical.

Making people's phones overheat and changing file extensions on their PCs isn't enough of a problem

Thats all it comes down to.

When programmers are doing things that actually endanger people's lives directly, actually alter the physical living environment directly, then you will see governing authorities pushing for licenses.

Engineers design passenger aircraft, rockets, cars, and medical devices without needing licenses - the products themselves are strongly regulated. Requiring the engineers developing the product to carry licenses will just raise the overhead cost of developing such products.

I think the key difference is certifying the product when there are multiple instances of the product being made is cheaper. For civil engineering, most of the projects are unique (and even if the design is something of a copy - then the sites present unique challenges). So maybe it makes more sense there to license the engineers.

>When programmers are doing things that actually endanger people's lives directly...

Toyota famously had some issues with their software and fatalities. It'd be hard to license software development though because it's such a quickly changing field.

https://news.ycombinator.com/item?id=9643204

You don't have to (yet), but there is a PE exam for software engineering: https://cdn.ncees.org/wp-content/uploads/2012/11/SWE-Apr-201...

Companies already need licenses to produce software for various safety or life critical applications, such as medical devices.

Since many of us fully expect AI to accomplish all programming in the future, I don't think so. Unless a licensing regime is humankind's last stand against AI.

I think a lower hanging fruit would be to push for the adoption of Rust, Ada, or similar languages that close or mitigate all too common errors.

Then perhaps we can put to bed the stupid "Engineers or developers" argument

It's not that difficult to put that to bed today.

Did you receive some form of a Computer Engineering degree from an ABET institution (just using the U.S. for this example)? Congratulations you are an engineer.

Did you receive a Computer Science degree, Computer Information Systems degree, boot camp graduate, self taught or otherwise? Congratulations you are a developer.

But what about an ABET-accredited Computer Science degree? Does that make me a Develogineer?

ABET has separate accreditation for Computing and Engineering.

http://www.abet.org/accreditation/

So you're saying you have to have a four year engineering degree to be an engineer, and you don't count a computer science degree?

Did you know that MIT, Stanford, Berkeley and Harvard do not offer Computer Engineering degrees (only Computer Science or Electrical Engineering or EECS).

Also you've completely discounted self taught practitioners, some of whom are often far more competent than people with actual Computer Engineering degrees.

Hence, it really isn't "put to bed" yet.

Course 6-3, the "computer science" major at MIT, is officially titled "Computer Science and Engineering". Maybe MIT 6-3 grads can call themselves both developers and engineers?

Computer Science and Engineering:

https://www.eecs.mit.edu/academics-admissions/undergraduate-...

That's a single course, not a degree.

This was really annoying to go dig up in my attic, but I found it:

http://i.imgur.com/eiF1INY.jpg

Did I say this was really annoying? Because it was. At least it wasn't as tacky as posting my diploma. Please don't tell people from MIT what MIT offers.

You could have just linked to the website. :) I guess things have changed in 20 years.

But really you only proved my point. OP said that a computer engineering degree makes you an engineer and a computer science degree makes you a developer. So what does a computer science and engineering major make you?

My main point was you can't decide if someone is an engineer based on the name of the degree they were granted.

Except that engineers do not require a license to practice their craft either, except in certain situations.

Reply on news.ycombinator.com