I'm a freshman in college and don't know what to major in. EE and CS seem the most interesting to me though. But i can't decide which to do. Can anyone give me some advice on what to do?
If you cannot use proper capitalization and punctuation, drop out, as both will require you to write papers.
Name:
Anonymous2010-01-18 17:49
Open the case of your running computer. Touch an exposed wire.
You'll either find a respect for electricity and a fear of computer science or a respect for computer science and a fear of electricity. Either way, win-win.
I'm doing a CS course right now. I chose it because I wanted a piece of paper that said I'm ub3r.
The expected level of engineering is low.
The expected level of paperwork is high.
The main hate I'll bring out of this course is a dislike towards UML. Why you have to write down and then draw a whole range of pictures to describe what could be described in class diagram tables is beyond me.
Also, you'll never understand how important modeling your program is until you get a several thousand lines into a project and realize you could have done things in a much cleaner (possibly more efficient) fashion. There are few feelings worse than ripping apart your code structure because you forgot to consider one dinky little feature.
Doesn't apply to /prague/ too much, but in an industry setting it's hugely important.
>>92>>93
As I stated at the fore; this is just for the piece of paper.
The bits I dislike are use case diagrams and component diagrams. At the level I've been taught, the systems I've designed and built can justify the inclusion of design, but not drawing a stickman for every input and expected output.
Component diagrams can be worked together by using the class diagram tables, adding some flow chart style links and choosing a method of identifying child parent relationships and interaction.
Using clean room development style on your pseudo code and actual code after these designs have been done is all I feel you require.
Name:
Anonymous2010-01-19 3:15
>>94 Using clean room development style on your pseudo code and actual code after these designs have been done is all I feel you require.
This may be fine for dinky little software or if you are the only coder that will look at the code. This is woefully inadequate if other programmers are supposed to comprehend anything more than a trivial section of the software.
Why all this work? It's all about documentation, it should be complete, relevant and up to date. It's all about making your codebase easy to follow and maintain. The abstract logic allows people that are unfamiliar with the system to focus on different parts of the system design (input/output, system entites, data storage, system processes) without having to assault their minds with irrelevant implementation details.
Name:
Anonymous2010-01-19 3:22
>>95
You're arguments are correct and I agree. I just don't like doing them because my experience is limited to ditzy little programs where there have been 4 people in the whole team at most. In these projects it felt like a waste of time and thus I've grown a dislike to them. In large projects I can see the attraction of having everything split up and organised.
UML is ass shit, and I've actively worked on forgetting it and using the brain space for something more useful like how to knit condoms for cats. Planning is good, done in a smart way that's suitable to the particular task at hand, but that rarely requires enterprise stickmen and knowing 50 different UML5.0XP2000 standard ways to draw a class in MS Visio.
But if you apply for a job in programming today, there's a 80%+ chance the person interviewing you has never heard of a pointer, thinks 'Algorithm' is the capital city of Greece and XOR is the name of that new alien movie. But they WILL babble Design Patterns BS, and rather than solve the fizzbuzz problem, they'll have you draw some UML2K enterprise pictures to show you're a good programmer.
>>98 But if you apply for a job in programming today, there's a 80%+ chance the person interviewing you has never heard of a pointer, thinks 'Algorithm' is the capital city of Greece and XOR is the name of that new alien movie. But they WILL babble Design Patterns BS, and rather than solve the fizzbuzz problem, they'll have you draw some UML2K enterprise pictures to show you're a good programmer.
I have met people like that and still can not fathom how they can function in a software project.
Name:
Anonymous2010-01-19 13:38
>>102
I have met people like that, and they are the ones who do non-technical interviews, not function in a software project.
Name:
Anonymous2010-01-20 9:08
>>103
But if they're the ones judging your development skills, surely they must be part of at least one of the software projects in their company.
Name:
Anonymous2010-01-20 11:55
>>104
No, they're usually not. In my experience, the few times I've interviewed with someone fucktarded like that there's a technical interview as well.
Name:
Anonymous2010-01-20 13:40
>>105
If there's a later technical interview, then why does the fucktard have you draw UML diagrams in the first place? Shouldn't this be part of said technical interview?
Name:
Anonymous2010-01-20 14:34
>>106
Well the technical interviews I've had are really just some mid-level guy who will give you a `fizzbuzz' type problem or ask you to psuedocode an algorithm, 15 minutes at the most. He just wants to see that you know what you're doing and you know how to think. I know people who've had them over the phone.
The other person usually is aiming to get to know you and find out what you're like as a person. Granted, I've never had to actually sit there and draw out a UML, so I can't say for certain there aren't companies out there like he was describing, but for the most part they just like to talk to you about programming-related stuff to make conversation (...to a certain extent). Also, if you respond with `UML? DAT SHITS GEY', they probably won't hire you.
Name:
Anonymous2010-01-20 15:59
>>107 Also, if you respond with `UML? DAT SHITS GEY', they probably won't hire you.
What if I say highly structured project management (PMBoK style) is bureaucracy?
Name:
Anonymous2010-01-20 16:01
>>107
Will they hire me if I say UML is unscientific and ultimately destructive?
Name:
Anonymous2010-01-20 16:12
i studied Information Systems Engineering at Imperial which is a mix of comp sci and ee.
it's actually more like three quarters comp and three quarters ee cuz they teach you so much in order for it to be useful.
basically me + all my mates are earning > £50k a year so it's prolly best to do a split degree.
basically it shows you have a dynamic approach to learning.
comp sci guys are aspie fucks... usually.
basically i work now as an architect for exchanges using both skills.
Name:
Anonymous2010-01-20 16:28
people who do EE won't learn alot of useful things like design UML and code syntax... but then again they do learn about systems.
in your spare time you can read all the USEFUL comp sci stuff from one book "learning java" or "learning c#".
meanwhile in EE classes have fun making shit that goes whoop whoop bang.
useful things like design UML and code syntax useful UML
Name:
Anonymous2010-01-20 18:27
>>111
Hell yeah, I remember throwing aluminum foil planes at a couple rods that were plugged into a huge capacitor bank. BOOM And trying to make a railgun, though we could never get that to work very well.
Name:
Anonymous2010-01-20 20:53
>>115
its all about directed electromagnetic pulse devices