To me the term didn't come from HR it came from trying to get "programmers" to think in terms of constraints, value, risks etc. Seems a pretty clear and simple definition.
The reason I think its not taken off in the same ways as say mechanical engineering is because in general customers cannot actually see the product, only certain elements of its outputs.
If you couldn't see bridges either in use or after a failure. You just saw cars disappear at one end and appear at the other, not even able to measure the gap or time taken to cross in any meaningful way then structural engineers would bullshit and self grandise as much as software engineers.
Maybe one day the average person will know enough to hold us to account, then we will begin engineering properly.
The trend of giving high sounding titles started decades ago as more and more educated people entered the workforce and involves all jobs.
Manager? Vice president.
Secretary? Executive assistant.
Developer? Engineer.
Reports and collects some data for his bosses? Data analyst.
Warehouse worker? Logistics and distribution specialist.
It doesn't have to always be self-aggrandizing for it to be true that lots of people are now called VP who used to just be middle-managers.
As opposed to learn about a type of algorithms, some math, some language or some pattern.
10 years latter, I’m glad for the exposure.
It was more useful to understand and talk to folks on the sides of the dev department : security, ops, qa, support. Heck, even product.
I felt that with those, the dev could monkey patch whatever together and still be part of some engineering process.
Now I don’t know. I sure do push code around.
If not, then, at least in my mind, it seems difficult to pin down exactly how software engineering can occupy a space in the domain of modern engineering.
Some sort of specific liability in the legal code could make sense, though.
In physical engineering, you really get one chance. If you make a mistake or miscalculation, there is a cost associated with undoing what's been done, as well as trying again.
In software, there is no such additional cost. You can trivially start from scratch, or any point between scratch and here, at any given time. There is no disposal cost.
Security breaches, locked in data (poor dB structure or no direct access) reverse engineering undocumented functionality, even just dismantling supporting infrastructure and migration costs are huge. Another huge factor is when business logic is codified and the knowledge is lost from the business and only temprarily held by transient consultants or developers.
An sap project for a single region for a $1bn per year company can easy cost over $10m and no software is being developed, only config, process change and data migration. Never mind increased staff turnover reduced productivity.
But very few people have the experience and expertise to understand what they are really signing up for.
Just think about a 3d model can do for a construction project and what that would look like for code and you can see how poorly we do this really.
(Make software actually be engineered)
That's without even considering all the kinds of rules-lawyering that would be applied afterwards.
Yeah, we call them civil engineers.
In the US, in order to be an engineer one either had to have a license to call themselves an engineer, or a company could confer that title to employees. These days, it's less about the specific education, and instead the process of solving problems.
Engineering is a discipline which may be applied to many domains, including software. I define an engineer as someone who uses fundamental principles and processes of science and engineering to solve real world problems.
Our customers need to ask a skilled expert a question and receive an answer.
If I merely implemented code to achieve this goal, I would not consider that engineering, but rather coding.
Instead, if I first defined the problem statement, gathered requirements, documented assumptions, broke down the problem into discrete deliverables, defined metrics and success criteria, and created a plan of record, that would be engineering.
I have a formal education in engineering, and aside from the mathematical and science fundamentals required for engineering, a large portion of my education was dedicated to the process of solving problems (also proudly, a good portion was dedicated to learning about ethics as well).
Engineering involves risk tradeoffs, and if your developers are not making your risk decisions then they are not engineers -- and also you're going to spend a lot more money producing software (the authority to do risk assessment should always be as close to the "in the trenches" work as possible, every degree of removal invites the risk decisions to be made on lower-quality more-aggregated more-gutfeel information rather than clear metrics).
Most of computer programming is instead tinkering -- build a proof of concept, put it under stress, see what is being stressed, buttress those parts. The fundamental idea of architecture, "let's add some structure in advance so that we can build a self-sustaining system and take away the scaffolding later," is only really seen today maybe in test suites or so? Those are the scaffolds of our day I think. We're still in very early years for computing as a proper engineering discipline. But at least we do make some simple risk calculations.
Most of mechanical engineering is still tinkering. Not all problems are solved. Many of the people here pining for "real engineering in software" need more interdisciplinary experience, ffs.
I can only speak from my undergrad experience, but almost every engineering discipline took the same background courses, as well as mechanical, thermal, fluid, and electrical engineering classes and from there, went on to take courses related directly to their field.
From my experience, tinkering is another way to say prototyping or building a proof of concept.
As we've gained experience, the need to prototype the same solutions decreases, but when learning new tools and methodologies, experimenting and prototyping is very much part of the engineering process.
We do not. Civil engineer is a specific thing, with training, education, and licensing. The person who designed the system being installed or serviced was likely a civil engineer, the plumber is a plumber. And that's ok, everyone doesn't have to be an engineer. Being an engineer isn't inherently better than being a non-engineer.
> an engineer as someone who uses fundamental principles and processes of science and engineering
Do you see how saying an engineer is someone who does engineering isn't useful to other people? I'm sure you know what you mean, but basically everyone is and is not an engineer according to your definition.
Or a plumber can just be a guy following a plan and cutting and gluing pipes together.
Yes. That is the joke (among non-civil engineers with civil engineer friends).
> Do you see how saying an engineer is someone who does engineering isn't useful to other people?
No. Because that is not what I wrote.
I also provided a concrete problem, along with an engineering and non-engineering approach to solving it.
This is the first I've seen this point in the discussion. Engineering is considered a profession, distinct from a vocation. The term derives from professing an oath to serve the public in an ethical manner. If developers had to take a similar oath, I wonder if it would open them up to civil/criminal liability when they acted unethically (e.g., creating code in a social media application with the goal of manipulating behavior for profit)
With software, true engineering only happens if the development model can follow such a process. Nobody wants to put in the work to do this for all code though.
Being a good engineer is adopting the right combination of methodologies to fit the problem space.
Being a good software engineer means knowing what you should plan out in advance and what you need to test and design as you go.
I'm not convinced there is much utility in conferring the attribute of engineer to someone who is not practicing engineering when it's just as possible to say "She studied engineering" or "She majored in engineering" to denote the status of having an engineering degree but not engaging in the doing of engineering.
I don't think this is actually useful without strict definitions of what constitutes engineering. If I go to a homeopath who prescribes an herbal tea to cure my cancer are they "practicing medicine"? The law says they are not, and cannot, because of clear definitions of the term.
More seriously, the first programmers were electric/electronic engineers, so the engineer title was probably inherited. And game engines actually require lots of math and physics.
>Economists don't call themselves engineers and engineers don't call themselves economists.
There are quite a few universities that offer courses/degrees in "financial engineering" and they often are in the mathematics/economics colleges and don't have a typical engineering curriculum as you laid out.
I think shat this article fails to note is this engineer light/technician class which tends to be the more hands on portion of "engineering" in the case of sw, they would be more light engineering more programmer less overall design. AKA your systems architect/principal enginner is probably doing "engineering" the software engineers are likely just programmers.
Some would define it differently as someone who has an engineering license. There are States who have actually presented lawsuits aimed at preventing people from describing themselves "engineer" if they don't have that license. And, while an engineering degree is the more common first step in the licensing procedure, it's possible for one to get a license without an engineering degree. In the legal sense, they are more of an engineer than someone with a degree and no license.
Regarding the plumber, I think the mechanical engineer who designs the plumbing system would be the engineer, while the plumber is the technician. Just like the electrical engineer designs the system but an electrician installs it. Both roles are equally important but shouldn't be conflated.
(FWIW, I don't subscribe to that notion of what "engineering" is)
You can have the exact same curriculum and be called engineer or not be called engineer, depending on the university. You can even learn nothing related to engineering and have an engineer title, or the other way round.
I'm not saying you're wrong (especially the first part), but there's a lot more to it, especially in countries where 'engineer' is a protected name/profession.
Personal opinion: I don't care. My degree is "Computer Science", not "Computer or Software Engineering" and I see myself as a Software Developer (although I will allow programmer)
Every degree in my school's college of engineering was BS/MS. And I think something like 25 of 27 were "degrees in engineering".
If you want to go full circle, they are "degrees in engineering" because they're accredited by ABET, and satisfy the first requirement towards licensure and a PE.
There is no simple definition, which is the whole point of accreditation and licensing boards.
ETA: But if the programs in your hypothetical have an identical curriculum (and both are accredited), then yes, they would equally be "degrees in engineering".
If you design pacemaker software, you're an engineer. If you build RESTful websites, you're not. I say that as someone who builds RESTful websites.
The reason engineers need difficult certification exams is, so they don't destroy lives or expensive property.
There was a programming PE (USA) at some point, but it was closed.
That just describes problem solving.
On another tangent, I wish we’d redefine what it means to solve engineering problems. I have noticed that there’s this property of engineering that as problems are solved, new ones are created. So it’s this never ending cycle of new problems blossoming.
In terms of creating new problems, making life better is an under-defined, open-ended problem. There’s always a lowest-hanging fruit, that doesn’t mean we should stop picking them.
What if their solution is non-technical, still an engineer?
By that definition, almost any profession - from surgeons and taxi drivers to construction workers and cooks - are all doing "engineering"?
By the definition given - "a process of optimization within a solution space bounded by constraints" - a taxi cab driver is certainly an engineer, with no fuzziness. Nor would it be an extreme case testing that fuzzy borders of that definition.
And that's why herval disagrees with that definition. As do I.
Definitions can also be overly broad, which is why we no longer use the term "star" to refer to planets, nor refer to the sun and moon as "planets". https://en.wikipedia.org/wiki/Definition_of_planet .
Otherwise engineer is just a title too, so I guess it is mostly meaningless?
I think this is exactly what the author points out as "software engineers don't know real engineering". What makes you say that no other engineering practice don't optimize on cost / engineering time?
There are degrees of quality in engineering, and one of the factors is definitely "time spent on engineering". If you are making cheap frying pans, I doubt that you're going to hire a rocket scientist to check whether the screw holding the handle is strong enough to resist 5000 heating cycles...
This would put industrial engineering deeply in "less engineer-y" territory, IMO.
I think someone is very engineer-y when they pick a simple, easy bridge design which requires less skilled labor to produce, even if this takes them away from working with more juicy physics problems.
AND...
If need be reduce it down to the underlying science and mathematical principles from which it came from, and because of that can analyze why things are "good" or "bad" at a much more fundamental level.
Practitioners/Crafters are repeating "recipes" with far less comprehensive knowledge of the principles that led to those recipes. They may know historical evolution and practical reasons why one recipe is better than another, but their depth of knowledge is far less.
So getting back to software, you can see two types: people that know algorithms and can mathematically design and analyze problems. Or you have people that are just following the api docs and basic examples of loops + if-then but don't know how CPUs work inside, how OSs work, and why.
And do you need full engineers for a website layout or ruby on rails site? No. But if you are scaling or have complex data models or will have actual load, then ... you need engineering.
This is a fantastic definition.
It shows very clearly why most software devs are not engineers. I know I'm not.
As a field, we don't "optimize within a solution space bounded by constraints".
We throw crap at the wall to see what sticks, hacking up something like what the users said they wanted, then changing it all to do what they ask for once they start using it.
It's normally done as a craft where we're using fuzzy human instincts to get to a good-enough solution and move on.
And for systems where lives aren't on the line, nor millions of dollars every five minutes (financial trading, massive server farms at Google, etc), that is actually the correct tradeoff. Engineering is overkill for most software.
NASA needs software engineers. Microsoft doesn't.
Even Google and financial services are really just services that scaled and acquired a massive userbase.
“Software Engineering” is really just an application of civil engineering project management to programming projects. The job title of software engineer is used too liberally however.
Microsoft needs a _few_ software engineers.
Most of their developers won't be and shouldn't be, though.
Conversely, NASA will have plenty of software systems that have nothing to do with spaceflight, like their website.
They shouldn't be paying software engineers to engineer those systems, but rather software developers to build and maintain them.
To me, the difference is that when we're writing software and just trying semi-random UX ideas and implementation approaches out, we don't have any real constraints that we're genuinely bounded to.
So, yes - throwing crap afu the wall is an optimization process.
It's the other half of the above definition that it misses, IMO.
I'm not great at typing on my smartphone.
An engineer is a guy who measures stuff upfront to create a plan of action (like measuring how much concrete we need if need a bridge to carry 500 cars). He never gets his hands dirty nor does he build the actual bridge. He has a ton of responsibility and must call on vast knowledge to know what patterns will be needed and if they will cope with the loads.
On the other hand, programmers are more akin to gardeners (my mom is super into gardening) and carpenters - we have to get our hands dirty and build the thing. We have to use and stay true to the spec (from the engineer above) as close as possible, measuring a piece of wood 3 times before we cut it (not the fail fast nonsense) and carefully plant new plants that won't kill the whole garden. We also need to pull out weeds and prune the bushes and so forth. With a bit of careful implementation you can have a gorgeous garden and bird/insects/squirrels visitors and so on.
Software Developers I would say, is then a combination of the above, depending on your experience you might lean more toward one side or the other. The best developers have a good balance of Knowing stuff and Doing stuff.
That's just regular old optimization. There's no point in optimizing anything that doesn't have constraints, because there aren't any bounds.
As the author correctly notes, there aren't necessarily hard boundaries on what qualifies as engineering, but I think there clearly are axes on which something is "more engineer-y" vs "less engineer-y", and as the constraints of the solution space go from physical (tension, voltage, pressure) to non-physical (interpersonal relations, public perception, aesthetic judgment), the activity goes from more engineer-y to less engineer-y.
In software, we're always optimizing on cost / developer time, but that's closer to the non-physical side of the constraint physicality axis. As you optimize for things like memory usage, execution time, binary size, then you get closer to the physical end of the constraint-type axis, and are thus more engineer-y.