pouët.net

iq: while i partly agree with the sentiment, i think your statement is self-contradictory. also given that higher education / academia is largely a social requirement / peer pressure nowadays, don't you think it's more important to be original and to encourage people to break the norm if they feel it's the right thing, rather than to destine them to something that might not even be healthy for them?

for me, being part of an academic world is not much different than demoscene, to be honest. preparing papers (releases), going to conferences (demoparties) incl. after-conference boozing, doing lots of archival research and doing really interesting stuff. the only significant differences are that i get paid for it, and also receive more recognition than in the demoscene.

Just to clarify: I'm not advocating that people should skip getting an education, I'm just saying that in my case getting an education has sadly not been an option (for various reasons). It's worked out anyway, and one of the best programmers I work with is entirely self-taught as well, so it doesn't have to be the end of the world.

I would've loved to have gotten a firmer grip on for example maths, and what I know I probably would've learned much faster attending a university. Still I learned it though, and I keep learning more because, really, once you stop learning stuff you might as well be dead. Plus it's damn annoying not understanding something fully.

In my experience the ones who tend to stick to their domains are the less talented programmers, regardless of level of education.

I honestly didn't thought this about self-teaching before. One would always think that if you are self-taught then you will also be succesful in your studies. But if you don't have patience to get into new territory and your studies require that you learn things you can't relate with your current knowledge then shit happens. So, maybe the university is also a test whether you can adapt into new kind of knowledge you can't relate with what you already knew.

I am with Gargaj with this though. Studies require you to learn new things in a very short time and when you are especially that kind of person, it doesn't work well. Sometimes I even doubt if some of the succesfull students I met have learned more than people who have failed. There surelly will be some exceptions but I know people who tell me they didn't managed to understand the subject yet they seem to be much better than me in passing the exams. It's like the skill is about doing everything in your hand to pass the exams and get good grades (which never interested me) rather than learning something which you can learn at your own pace as a self-taught.

If you allow me to give you my 2 cents:

At least here in Vienna, studying at university revolves very much about teaching yourself. You have to learn from books. So studying in Vienna is the same as autodidactic learning, with the only difference that a university student gets evaluated in exams, and this way receives feedback and ultimately a degree.

Students are expected to be proactive, to be actively interested in the subjects of the university curricula and to obtain information from various sources themselves. Success at university depends very much on your personality and interests.

For a demoscener I believe it would not be any problem to study in Vienna, as most demosceners are autodidacts and therefore used to acquiring knowledge from textbooks or papers. Or do I have a wrong impression of demosceners?

Yes, you do.

Adok and his shit "Captain: I'm in Mensa."...

At 14 one of my early programs was a text-mode RPG with a builtin interpreter. However I had no clue how to make a compiler and ended up with something laughably bad. From this day I considered compilers to be something magical.

Going through Uni years later was enlightening and made me realize it was only one of the many holes in my self-taught programming.
The problem with self-teaching is, if you are like me you will learn the same thing over and over again until you know all x86 instructions by heart, instead of focusing on areas where your knowledge is weak. I think education and paid work get you out of this scheme.

the place I go to, aarhus university, information studies, is centered around a "reflective practicum" meaning that we essentially teach the students theory, skillsets and ways of thinking in the classroom using the books - and then the students have to go out in the world and do different projects that put these skills into play.

It's based on the idea that you wont really know the theoretical stuff until you try it out in practice (and I believe that to be correct at least for interaction design which is my field of study). We're a bit of an odd study place though, since most places just feed the students theories and let them figure out how to use it for themselves :)

ok, this is offtopic, but:
@iq: i really admire your demoscene stuff. your articles on your webpage taught me a lot!
@optimus: your cpc-wolfenstein engine is one of the cooles things i have ever seen!

you can learn alot by hanging out with the right people. the people who are smart. i found out in my early days that if i hang out with the smart guys i got more smart myself, if i hang with "stupid" guys i got more stupid. lol. unfortunately when those people move away, get girlfriends, married etc.. you wont spend much more time with them.
- if you are smart i think you learn by people who are much smarter than you. and you can't learn from someone who are more "stupid" than you. you can learn to them instead if they are interested in learning from you, but most of the time they dont want to. by using the word stupid i dont use it as a negative term.
doing homework is great as long as you understand it, but if you're stuck and dont have anyone to ask, then you're lost. or you're just have to take a break, and try again later. homework, exams, etc.. put alot of pressure in the person, its important that he or she doesnt let this pressure take over. its not the end of the world if you fail your exams. Einstein failed his.

@rac: Thanks! I thought though the coolest thing would be Batman Forever or something like that :)

Hmm...

If somebody was interested in computer graphics programming and already did it as a hobby (such as in the demoscene), then I can't see going to university to study computer graphics to be all that useful. Wouldn't it be better to take some physics or mathematics course that is not directly related to computer graphics, but still covers useful concepts that could potentially be applied to one's programming interests? After all, isn't it about expanding oneself rather than confining oneself?

I think the main problem is the conditions in which one goes to the university.

For example there's a difference between going to a super high rated/expensive university where teachers are themselfs researchers that are up to date on the latest trends and are themself pushing the envelope, versus being in some third grade rated public university in some god forsaken location where obviously you tend to have troubles attracting good teachers...

Then there's the difference between having the studies paid so you can actually work, versus having to spend hours doing mundane jobs just to be able to survive.

This directly impacts the type of people you will meet, how much you will benefit from it, etc...

If I take my particular case, I had the misfortune to be accepted at the University of Besançon (France), somewhere on the eastern border, far from everything basically.

The main teacher (university dean something?) was teaching us architecture, which would have been fine if he had been updated on what has happened in the 20 last years. He also taught us assembly language, and gave me a 0 on the exam because I wrote the required program in 68000 instead of x86. Not my fault if the exam was just asking to write in "assembly language - no dialect specify". (Had to go in front of the university board of teachers to plead my case. I was allowed 20 minutes to do it again, and in the 20 minutes I gave them back a fully commented 8086 + 68000 + 6502 versions just because I could and because I was fucking pissed off.)

The teacher in charge of computer graphics had no clue either, and I had to gave him some explanations about why my fixed point line drawing routine was 4 times faster than his naïve Bresenham routine... etc... etc...

So in my case, waste of money, frustration. I did learn few things, mostly vocabulary and methodology. The most important thing I learned is that one should never trust the diploma indicated on a resume: You can have a Bachelor Degree in Computer Science and be clueless in computing.

For me, the university was like 98% of time lost. But, the remainder 2% was really useful. I agree so much with what Iq wrote.

It is not that you can´t not learn absolutely everything by yourself by just reading some books or studying in internet, but it is that by yourself, you will probably miss reading about really important things that doesn't seem like so.

I remember statistics, databases, the O notation, logic, numerical methods, and a bit etcetera were a pain to study. In fact, it was like "why I need to study this thing!?". Then, at some point I've needed absolutely all of that, both for hobby programming and professional, even the most unexpected things were at some point useful.

About bad teachers... unfortunately they are everywhere. Don´t focus yourself in the teacher, but in the things you are studying. If the teacher is not good, buy a book instead. In the end, university is not about the people who teach you, but about you acquiring a good knowledge in a lot of different subjects, that are going to be really helpful in your future.

rudi: So while you are getting smarter, your friend is getting stupider lol You're something like a knowledge leecher. :<

So far this is some of what I've learnt:
- Some maths (mostly linear algebra)
- Boolean logic (discrete maths)
- Formal systems in general
- Grammars
- State machines
- Semaphores
- Proper parallel programming and pitfalls
- A good bunch of useful algorithms
- O notation
- OpenGL, shader programming
- Some more physics
- Basic functional programming
- OOP
- Some cognitive psychology!

I've also done a good deal of other things (among them: embedded systems, raytracing) but do not feel that it taught me a whole lot of new stuff.

Stuff I should have learnt but didn't find useful in the moment:
- Being academic
- UML diagrams

Stuff that I feel was a drag and ultimately a waste of time:
- chemistry (obligatory for all civil engineering lines)


Besides that, some social/networking benefits:
- Got to know a good deal of international culture
- Eventually ended up bartending, both in the bar and at huge parties
- Got into a venture project
- Got plenty of contact to companies with open job positions

And social cons:
- Not enough women! New girls in the everyday are scarce, unless you start cold approaching like a pickup artist in a wildfire.

rudi the Brainsucker?

I think we can draw a conclusion of sorts: you could get an equivalent to a degree education with self-study, but only if you study a load of subjects you have no interest in or use for, plus the subjects that you already 'know'. Both of which are well worth doing really, although it feels like a waste of time quite often :)

@Lord Graga:
i think i also learned a lot of interesting things! somtimes the "sense-making" comes some months (or even years) after the course...

social pros:
on the other hand, if you meet girls who are actually into maths/technology/theoretical cs/physics whatever, a relationship can be more rewarding in the long run! :)

lord gravko: uml diagrams and off springs will pay back ten fold when you get a job. documenting decisions, reading past designs is a key skill in the real world.

the rest of your items will most likely remain in the box of "cool stuff you did at university" :)

@rasmus: Yeah, I know, I also regret not diving further into it when I had the chance.

The most useful skill I learned in academia is the ability to reflectively frame problem in systems design and interaction design - that is useful outside of academia too. Furthermore I learned a set of concrete methods such as prototyping and so on and tried them out in practice, and a lot of strange theories that were nice as well as probably served to shape my way of thinking.

I'm very happy with my time at university, but as opposed to some of the other posters I seemed to possess the exact mindset that fits well academia here. Which is, among other reasons, why im applying for a phd :p