I’ve worked my whole career in startups of one form or another. I wonder if one of the reasons I don’t relate to or agree with a bunch of the advice in this thread is because that doesn’t seem to be the typical background.

Follow up to my job search fail post, again mostly culled from this thread. But much shorter than the last one!

I don't think programming will ever be something everyone can or will do.

No. 1 it requires an iq above 100 at least. That cuts of half the world right there, maybe more. Secondly, it requires dedication to a task that many (most?) consider extremely frustrating. I.e. trying to understand how a machine "thinks". And it requires valuing logical thought over emotional, which is not true of many people ime.

Anyone (with a college level iq) can become a programmer, sure. But not many people want to badly enough to deal with what that costs.

There are kids toys that get them doing programming. The basics of programming aren’t at all complex. And it’ll be even less complex (and frustrating) as kids are raised and taught important root concepts from an early age and we build higher and higher level languages.

It’ll also become a more important skill in other fields. Those who can use excel macros (or even formulas) are more useful than those that can’t. As software takes over the world the people that can do even basic customizations/extensions to that software will be more useful than those who can’t.

The trend over time has also shown that programming in general gets simpler and more accessible.

Can anyone comment on how important whiteboarding skills are for web-dev positions in particular? I would think they shouldn't be that important as web-dev work typically doesn't require really strong data structures / algorithms knowledge. But based on what I've read in this thread and on cscareerquestions, it sounds like grueling technical interviews are the norm. Is this more true in Silicon Valley than in other locations (I'm in Canada FWIW)? I'm approaching graduation and am trying to get a feel for how much time I should be dedicating to studying/practicing CTCI relative to developing my portfolio of personal projects. Some of the recent comments in this thread have made me think that employers will overlook my github projects altogether or be entirely unimpressed with student-level projects.

Doing anything of value in programming requires a lot of dedication, frustration and constantly being humbled by how little you actually know.

You can teach people how to create basic programs, sure. But the dedication to be a software developer requires a certain type of person in my opinion. I've had this discussion with plenty of CS grads that used to be devs but made a lateral move into something that wasn't as challenging on a technical level. They just didn't want to keep dealing with this **** every day, was the gist of it.

IQs are super important ITT.

Hint: we are nothing special. To be a successful median SE takes the equivalent effort to be a successful median accountant. If you disagree with that show your work.

You should be able to bang out leetcode easies. Basically array method one liners. Wouldn't bother with On stuff.

The majority of people applying to cs programs now because it’s profitable are dropping out because the weeder classes are too hard, so there’s that.

The equivalent doesnt really happen in say, nursing, the other booming profession. Everyone finishes those programs. Hell even if you dont you’ll probably get a job anyway.

Calm it down buddy. I never said devs were required to be exceptionally smart. Just that there is a minimum bar to get coding, probably around the average college level iq which is 110 iirc. That already cuts out more than half the population.

But my main point is that to stick with it, you need to have a certain type of personality, probably maybe a little bit related to a certain spectrum, to be willing to suffer the frustrations of constantly interacting with an inanimate object. And those types of people aren't very common.

Hey champ, I don't agree but not going to waste my time arguing more. When you make a claim like this chief its on you to provide evidence of this. I think IQ gatekeeping is very dangerous and should be avoided, and again, slick, show your work if you want to make this claim.

Why am I required to have such academic rigor when nobody else itt is? It's common knowledge that IQ is normalized at 100 so by definition half the population is below that threshold. And it is also pretty common knowledge that college graduates are higher than the average, roughly 110, I guess I could look that up for you.

I use that to infer that most CS grads have that type of IQ, but that's just my speculation... you know... like the majority of posts in this thread are. I framed it with a bunch of "probably's" as well. Its not gatekeeping, since I don't wield any power here. I'm just guestimating population sizes.

And I'm not writing a thesis. Nothing I wrote that isn't framed as speculation fulfills the "Extraordinary claims require extraordinary evidence" maxim, at least not in my opinion. But I guess IQ is taboo to some, so I'll tread carefully in the future.

edit:
btw, I also happen to think a successful median accountant has an iq around 110 or higher.

You first.

I do go back and forth on this but overall side with grue.

At my last company we had raised 50-100M and did important and nice stuff for human life/ the healthcare market. But the business side was flat for many years and it was hard to recruit and pay good people.

I worked with some horrendous people. Lazy, behind the times, just dog**** engineers. I also worked with a few great ones (this is where I spent my energy and who I worked with). But I can't believe the bad ones were really high IQ or were exceptionally smart. I would absolutely not want them anywhere near accounting.

Now, that environment frustrated me endlessly so I left, but from the stories itt and from dozens of other life experiences I've found companies in basically any industry of more than 30 people require technical resources. So the demand is insane and the average competence is therefore low.

And maybe I'm wrong but I would guess that to be a successful median engineer you need to not be complete dog**** and you're there.

But it doesn’t. Web sites, excel macros, hacked together scripts, simple apps, and so on and so on provide a ton of value and don’t take much time, effort or knowledge.

Hell, we literally have a whole class of developers that only took a few months to become productive and useful programmers. That’s not even possible in a whole bunch of other disciplines.

And all of this ignores the trend that things get simpler and easier over time. So even if it was true now that it was hard to be useful that’s not at all indicative of what the future holds.

Source? This is news to me.

Most of the best devs I know think it's something 80% determination, 20% raw brain power. But of course you can fit a lot of stuff into those definitions.

I think programming is way too broad of a field to make a claim like this. If you’re building simple websites for a low hourly rate - it doesn’t take much of either. If you’re building the latest military super robot killer - it’s probably going to take a lot of brain power.

I know an ada programmer who programs satellites. It seems like C++ with endless validation and testing. Much more about diligence than smarts.

When I first started coding my brain was much more capable than it is now in terms of raw cycles and juggling many things at once. I wrote the worst spaghetti code you ever saw - but I could reason through it by brute force.

Me now is probably too dumb to do that, so me now writes more modular and cleaner code. Well I mean hopefully I'd be writing cleaner code even if I was as smart. But the point is if your code requires Ed Witten-level mental gymnastics, you might be doing it wrong. I'm reading a Kent Beck book now where he makes the point that your #1 priority in coding should be to make sure it's readable and can be picked up by future devs.

My best skill/value-add at this point is probably my experience in how systems evolve over time and noting what changes designers would make if they could do it all over again. None of that requires a lot of brainpower, just perseverance and knowing what to pay attention to.

Based on my experience I'd guess raw brainpower has a huge factor in how quickly you pick up a new skill - which can be very important - and also impresses your bosses. But in terms of mastery of that skill, or being aware of the bigger picture issues that go into incorporating that skill into good software architecture - I'd guess raw smarts is much less important.

like hard data? I doubt that's tracked in any meaningful manner. Just anecdotally from long conversations with the big boss at my job, who was a professor of mine last year. He used to sit on the board of the CS school at my alumn. I was complaining to him about what I thought was the overly broad and simplistic curriculum of CS i experienced and he said it's still too hard for students, they've had to dumb way down the weeder classes, and he said he knew it was that way at other schools as well.


Take it for what you will but from what I typically saw of the average student at an above average school - i was not really that impressed.

fun fact, though - my class, 2018, was the best CS class the school has ever seen. I finished with a 3.8 and didn't even qualify for honors awards.

Another fun thing I learned from him, I guess professors that write their own books usually get boned. He signed a book deal where the publisher got basically everything, and said if you added the hours he spent writing it, it probably was less than minimum wage for what has been returned on it.

I was of the popular view that proffs just wrote books to make money off students.

The more people who switch to CS the easier they are going to have to make it.

College books are one of the biggest scams going right now. Pure evil.

I’ll admit I don’t entirely understand parts of this because I’m probably missing a bunch of context / US academic knowledge. For example, I don’t know exactly what a weeder class is or what you mean by your fun fact.

But I don’t really get how this is talking about completion rates of CS programs. I know that’s always been measured at my school and it certainly hasn’t changed in any significant way even with increased enrolment.

In highly sought after majors like CS they have several lower division courses that are meant to catch early failures. Usually it’s the math requirement but at my school the early programming classes are meant to be pretty hard.