Comment by throwaway2037 | Hacker News Reader

Comment by throwaway2037 | Hacker News Reader

Parent Full thread

I see this type of comment on HN frequently. I have a dim view of national certifications for software engineering or computer science. Are there any highly advanced economies that have "a formal assessment by the national engineering council" for software engineering or computer science? If so, are they useful? Do they actually indicate quality? For me, the field(s) are changing so fast, that it hardly makes sense. If you are working in an area that is not life threatening, I am not a believer in certifications. It all seems like a bunch of gatekeeping.

Today, I was listening to the Lex Fridman podcast with John Carmack. It reminded me that John Carmack does not have a university degree, yet, he is one our generation's globally recognized masters of software optimisation. There are few in the world who can do what he does, and he has no uni degree, nor (I assume!) any "national certifications". Michael Abrash is similar.

On a personal note: (US) Wall Street has similarly ineffective gatekeeping with the Series 7 & 63 exams: My father called it "toilet bowl knowledge" when I studied for it. He said: "Once you are finished the exam, you can flush away that knowledge. You will never use it again." He is right about more than 90% of the "knowledge" required to pass those exams.

> Are there any highly advanced economies that have "a formal assessment by the national engineering council" for software engineering or computer science?

The EUR ING certificate from Engineers Europe can be awarded to those who have attained a degree in engineering, but also for those with "no exemplifying formal qualifications, but will have engaged in professional Career Learning and peer review via the individual route".

https://www.engineerseurope.com/what-eur-ing-certificate

Since Engineers Europe is a private organization (albeit one that is widely recognized), I would imagine that individual European countries have their own rules about when someone can call themselves an engineer. I am pretty sure that Germany only requires a degree for instance. Looser still are the rules in Britain: the UK does not require any certification at all for the basic term 'Engineer', although there are more specific titles that are strictly protected (ICTTech being one of them, and yes they really do use that silly abbreviation, obligatory italics included). Theoretically, a country could refuse to recognize certifications from Engineers Europe, but that would not exactly endear them to the European Commission or to other European countries!

And so what?

Those gatekeepers might not accept a John Carmack or a Palmer Luckey (or Bill Gates or Larry Ellison or Vitalik Buterin, etc, etc), but how many of their certificate-holders have done any comparably significant engineering work?

Does it matter for them? Would any employer/investor, seeking to hire/fund carmack etc hold back because they didn't have a cert? The vanishingly few with those levels of talent stand out quite easily and don't need "proof of competency" to be evaluated.

For the _vast majority_ of others, certification allows employers to have externally-validated trust in an engineer who mightn't stand out as obviously.

Early in their careers, even exceptional people face obstacles from gate-keepers. It wouldn't surprise me at all if there have been Germans of similar temperament, credentials and talent as Carmack or Luckey but who struggled much longer to stand out to a sufficient degree that they could get great job opportunities or funding in a that much more credentialist environment.

I like to imagine that such people would have found a way to emigrate and make their impact abroad. Either way, the cost to Germany itself (and Europe generally) has been immense.

> For me, the field(s) are changing so fast, that it hardly makes sense.

This applies to CS degrees too. But while there's immense amounts of churn in the hot language/UI framework of the day, the basics like algorithms and data structures barely change.

On personal note, Series 7, 63, and 62 are not hard tests....one can pass it without much study...I did pass the series 62 in my sleep.

But keep in mind those are legal compliance tests...WTF?

In short words, they give the legal basis for why in finance we have to follow a sales script as it then covers the legal boilerplate mess...

On the other hand in the late 1980s I had a chance to be in the Turtle Trading class in Chicago....obviously that would have been more helpful as I think the success rate was over 50% for students of that pratical turtle trading class.

Reply on news.ycombinator.com