I always thought that being an Engineer assumes you have knowledge of mathematics. The way this word is being used in Software ist kind of strange. "I built a blog in php, I'm a Software Engineer" - this feels kind of awkward to me.
The thing I have needed all the time is statistics and prob. theory, that keeps coming up literally everywhere. Calculus - not so much. If you need something, you can always re-learn it quickly. (For example, I learned quite a lot of linear algebra, but haven't used it for ~10 years, so when I had to write some 3D gfx/shader code, I had to spend like two days on quaternions, etc.)
Done
Being an engineer means using technology, science, mathematics to solve problems. Well in many cases you don't need math to solve these problems. The word itself as no root in math either, it's based on latin for devise/contrive, sure it was used for builders at first, where math was important, but context changes.
Also Software Engineer is wide and reaching, a 3D GFX Software Engineer will need some heavy mathematics to do his job and do it well, a Web Software Engineer not much, but he'll need to know a wide range of other knowledge GFX guy doesn't know (HTTP, network protocols, various languages, server technology, database technology, browser knowledge,...).
It's not about being good at mathematics, it's about knowing as much as possible about the domain of knowledge your role entails, and surprise, not every one of these requires deep math knowledge and understanding.
Your attitude is pure and simple elitism.
That's called practicing engineering. An engineer is a professional practitioner of engineer. You can practice engineering all you want, but if you're not a professional (having received an engineering degree from a certified university), it's dishonest to call yourself one. Honestly, it is elitist. But those of us who obtained our degrees worked our asses off. And I personally hate when people abuse the term to mean anything that took skill. "Candy cane engineer". "Beats engineer". "Drink mixing engineer". It's linguistic prescription to make yourself sound more important/skilled.
I applaud the effort to learn the topics of "statistics, probability, and linear algebra", but these would have been relatively fundamental courses in most software/computer/electrical engineering curricula that I've known about, and most definitely a prerequisite to calling oneself an engineer.
Don't go near any dark alleys.
But my years of engineering experience have shown me that the title "Engineer" is really more about how you approach problems and what you do to solve them and less about degree credentials. Now, the following example is in the context of electrical engineering on airplanes, not computer programming, but I think it holds true.
One of the best people I work with does not have a college degree. But over years of self-study and real world experience, he has taught himself electronics, some computer programming, and enough mathematics to get by. And when there is a technical problem to be solved on one of our airplanes, he will chase after it relentlessly, and smartly, until it is solved. His system designs are clean and well thought through. He has taught me much about designing for real world implementation. Is he not an engineer? He does more than many of my coworkers who are degree holding EEs. I am not afraid to call him an engineer, because he has earned the title in a different way.
However, I still think it's wrong for people who don't exhibit these qualities to call themselves an engineer. If you make sick beats on your macbook, that's great. But don't call yourself a beat mix engineer.
Edit: Thought of some reasons It's similar to the "doctor" title. You can be the worlds greatest doctor. Self-taught, you can do everything from intubation to surgery. However, you're still not a doctor. You practice medicine. Why? There's things that you can only learn from someone who is more skilled than you, and who is skilled at teaching. That's what a professor is (simple definition). They are an authority on their topic and are the best place to learn from. They teach things that books don't cover. They have experience. They can tell you when you're wrong, and unlike a book can teach you the most current standards and techniques.
Another point is the completeness of education. Your coworker, does he know vector calculus? Linear algebra? The forward-active voltage for a BJT? Maybe. But there's no guarantee he does. A degree from a certified university guarantees that you know the salient points of your field (not always true, but for my argument it is). If you don't have a degree, there's no guarantee. And this knowledge is important.
I agree with you, for the purposes of your argument, that degree should serve as a guarantee. It is an important signifier of mastered domain knowledge, and more importantly, a signifier of the ability to master new domains.
EDIT: That last sentence was a run-on, fixed it.
And as for experience, well, in the case of a CS student wishing to enter industry there's a good chance that the majority of your professors never even worked in industry. So if you're looking for people with experience to learn from, well, then you're in quite an unfortunate situation.
This led me to conclude that a degree offers no such guarantee that someone knows something. It offers a guarantee that someone was introduced to a number of concepts and demonstrated an understanding (or knack for cheating, cramming, what have you) good enough to pass and move forward. This is why I shudder at the thought of hiring old classmates who had to be hand-held through their 4 (or more) years of university, I know better despite what their degree might say.
Which is why that people who earned a degree...earned a degree, that's it. As far as I can tell they have no right to call themselves an engineer until they begin to practice engineering and practice it well enough to demonstrate the value of their thinking.
But the best professors I've had (and I've had a bunch) were the ones that really did combine both. (And frankly, I can forgive a lot of poor teaching in return for a "well, this is technically true, but no one really does it that way; they use this shortcut...."
As for a university education not providing immediately applicable industrial experience, well, that's kinda not the point of it. Sort of the difference between passing the FE exam and being a PE.
If he's as good, and as experienced as you say, he should be able to just do the legal/ethics stuff and get a license. Unfortunately, the professional associations are streamlined for people who take the usual path through university. At least in my jurisdiction, it is technically possible to have the experience counted, rather than the degree, but it's much harder.
The system is clearly not perfect, but when you ask the average engineer whether it's ok to do things like lie about their experience, you will get very different answers than if you asked the general public. The gatekeeper is doing a real job, even if they don't do it perfectly.
So basically, don't call yourself an engineer unless you're willing to sign off on something, and be legally bound by it. This implies a strong background in problem solving and structured design processes, to remove as much risk (both personal and to the public) as possible, which is also vital to engineering.
>So basically, don't call yourself an engineer unless you're willing to sign off on something, and be legally bound by it.
I'm sorry to say it (not really); but for a number of reasons, some good and some bad, the title has been co-opted, and there is no going back.
>This implies a strong background in problem solving and structured design processes
As far as that goes, I've met a number of pedigreed folks who can't engineer their way out of a wet paper bag.
To some extent this happens in Australia as well, although there is a movement both to require things to be signed off by a PE (or CPEng here), and to have those engineers provide documented supervision of the work they sign off on.
> As far as that goes, I've met a number of pedigreed folks who can't engineer their way out of a wet paper bag.
No argument there, certification is never proof positive of competence. I've met very good engineers who aren't Engineers with a capital E, and very bad Engineers who knew enough to fool a test board, but not much more.
The existence of these licensing schemes is far from perfect, but better than nothing IMO. Applying the concept to general purpose software is another discussion entirely!
Definitely semantics. For example, I have a BS/MS in applied math from a good engineering school university. I am a software engineer mainly working with ECEs, physics, and other math guys, who are all "software engineers".
I see where you're coming from about being a SW eng without knowing a lot of math - but there are other majors - math/physics/stats etc that will be very math heavy and not "engineering degrees".
>When you get an engineering degree from a certified university, you are an engineer.
That's not true. When you get a degree you have an engineering degree. When you get hired and employed as an engineer, you are an engineer. You have an engineering degree and I have a mathematics degree. Our employers hire people for engineer positions and call them such. If I were a professor of math, i could call myself a professor. If I were a mathematician at the NSA, a Mathematician. But my employer calls me an Engineer. The degree does not do that.
And I'm not merely referring to software. The above is true for my companies 20+ different engineering positions across all disciplines they hire for (Aerospace, mechanical, electrical, materials, software, etc).
It really annoys me that I spent 4 years studying this to find that other people who scrape together snippets of JavaScript & PHP feel entitled to call themselves an "engineer".
The other reason, of course, is that the "software engineer" term comes from a group of people who really wanted the respect that comes with "engineer" but realized that the big three don't get very far, software-wise. (And coincidentally didn't want to do all that icky math stuff. Not to mention much of the icky programming stuff.)
A filter / weedout system was required because of too many students, so its turned into something else entirely and now a engineering degree often means nothing other than having passed the weedout math classes. It really shows in some new grads that don't have any actual engineering skills but are really good at calc problems.
Lived and breathed math for 5 years in college before I could look myself in the mirror and call myself an software engineer.
Not really. Mostly just algorithms and procedures of some math concepts(Calculus and Linear Algebra) which roughly corresponds to the first year of North American math major.
In contrast, I know engineers who live and breathe PDEs and tweak compilers to solve them faster.
I doubt my CS professors would be able to solve a PDE..
Beyond the basics, not even many math professors can do that. Math is too vast and people specialize. Strong algebraic geometers are not necessarily strong analysts or algebraists or logicians.
1) No such thing as universal mathematician in this day and age.
2) Engineer's PDEs(algorithms) are not the same as mathematician's PDEs(theory). Same as comparing a student in China who learned English to communicate with English speakers to English majors from English speaking countries.
So yes, if you take a Computer Science major with a focus on Software Engineering, you will not learn enough math to minor in math "for free". If you pick a more mathematical subfield to focus on, you should probably declare a math minor for the one or two additional courses it will take you.
I made a truly stupid choice: I graduated in 7 semesters with a Comp Sci degree concentrated on PL theory without picking up the additional courses for a math minor. As a result, I'm "condemned" to learn that material independently later on. "Luckily", the Technion required me to do extra coursework for my MSc, so I've had to buck up and learn more theory.
Now get off my lawn until I'm done with my highly theoretical machine learning exam ;-)!
If you want to do good at machine learning...electrical engineering is probably a better choice; the maths learned in EE overlap fairly well with what is needed to do ML. PL theory is quite niche, even for PL researchers.
Besides, I never heard of math majors taking cryptography that early. Usually, Intro to Real Analysis, Abstract Algebra and Abstract Linear Algebra come first.