Reflections on college.

[The_Tao_s_Revenge]

@Derleth

C++ is a huge mess, but getting the concept of pointers is important: It’s a big part of understanding hardware organization and getting a general handle on how OSes work the way they do.

If I understand it correctly a C++ pointer is a variable that points to the memory address of another variable. Was not sure why thats useful in C++ though. It sounded like it’d cause weird bugs, especially if a pointer pointed to the wrong variable, or the original variable memory was released and another program started using that memory.

But I just googled them and now I get some of why they’re useful and I get linked lists now too! You couldn’t make those in C++ without pointers. Good stuff.

It’s a powerful concept. The idea of an extensible program with its own programming language is one of the best ways to manage complexity, because it allows the application to implement the structure efficiently (usually in an inconvenient-but-fast language like C++) and allows the extensions to implement the complex specific behaviors (usually in a convenient-but-relatively-slow language like Javascript or Lua).

Indeed. It’s like Legos. You use lower level languages to fabricate the various types of bricks, and base building board (application) and higher level languages to build things with the bricks. In some cases like Firefox even the GUI is made of lego bricks.

How do you know arrays are quicker? Have you profiled? That’s the only way to find out. As to bad habits, every language has its own notion of what’s good style and what’s bad style. Don’t worry about not being able to treat Python like Javascript or even like Perl (which, incidentally, has hashes as well).

Hmm good point. Finding out could make a fun educational project.

Here’s a short list, chosen because they’re all very different from anything you know and from each other:
[ul]
[li]Common Lisp, because it is a programmable programming language: You approach problems by extending the language, both using functions and macros (nothing like C++ macros), turning your application into a domain-specific language. Common Lisp removes the artificial barrier between the language designer and the language user. Read this book alongside the one I just linked to. It is less practical but it will teach you the Lisp Philosophy a lot more effectively. Some elements of that philosophy only apply (easily) to Lisp, but others (the bottom-up methodology, for example) apply to any language.[/li][li]Haskell is beautiful in other ways. It is probably the best language to learn about type systems from, and when you figure out how to stop fighting the type system you’ll forever after appreciate what type checking is capable of doing for you. (The downside is that the type systems in languages like C++ and Java are a half-assed parody of what Haskell provides.)[/li][li]GNU bison is the modern version of the historic parser generator yacc. If you don’t know what a parser generator is, you really need to learn one: Formal grammars are your friend and so are declarative languages. Bison itself isn’t as important as finding some tool that understands EBNF (Extended Backus-Naur Form) and will generate the low-level parser code for you.[/li][li]Forth is a great example of how you can do a lot with a very primitive language. Like Common Lisp, the main strategy in Forth is extending the language. Unlike in Common Lisp, most Forth systems give you as much access to the underlying hardware as assembly language. It’s worthwhile to build your own Forth interpreter (in fact, building new Forth systems seems to be the main hobby of serious Forth programmers ;)). Understanding the deep similarities between Forth and Lisp will go a long way to helping you ignore outward appearances and recognize common patterns in all designs.[/li][li]C is the most important part of C++ (pointers) without all of the ugly cruft (just about everything else). In addition, you might have noticed that POSIX speaks C natively. Understanding C will teach you a lot about why most modern OSes (Linux included) are the way they are. (Yes, I want you to buy that book specifically. It not only teaches you the language, it teaches you a good, terse style, and has examples complex enough to get that style across effectively. A tasteful style is important to reading anyone’s code, even your own after six months.)[/li][li]At least one assembly language. Don’t limit yourself to x86 because that’s what you have sitting on your desk: Emulators for older, more interesting architectures are readily available. I personally like the PDP-10 and the VAX, but there are many others out there. SIMH is a whole collection of emulators.[/li][/ul]

It’s also a good idea to pick up this book somewhere. It isn’t language specific, having examples in multiple languages, and it won’t teach you any language. What it will teach you are techniques that will improve your ability in any language.

Thanks! That looks pretty interesting. I’ll have to give it a go and see what I can get out of it.

I agree. I’d love to see programming language design taught, in fact.
“The interface is misdesigned: It’s hiding risk instead of making it apparent. Someone is bound to believe it and do something stupid.”

Poetry and philosophy are central to programming. You need to create metaphors that are beautiful, because beauty is simplicity and simplicity is understandable. Having a coherent philosophy (which C++ lacks, Common Lisp has deep down beneath all of the cruft, and Python tries very hard to express at all times) enables you to see beauty more easily.

Keep the sense of wonder. Programming is a creative art and a sense of aesthetics goes a long way.

Thanks man. Much truth in that.

[TVeblen]

Doors, perhaps I’ve misread much of what you’ve posted over the years but the overwhelming impression I’ve gotten from you is free-form angst. Everything is complicated, nothing turns out the way you hoped, you’re ultimately disappointed or disillusioned to some degree.

If you couldn’t find worth in four years of college then the work world will be a really serious bummer for you. I wish you luck and all but it doesn’t sound like you even noticed the staggering chances and good fortune life already handed you.

[Indistinguishable]

@The_Tao_s_Revenge

If I understand it correctly a C++ pointer is a variable that points to the memory address of another variable. Was not sure why thats useful in C++ though. It sounded like it’d cause weird bugs, especially if a pointer pointed to the wrong variable, or the original variable memory was released and another program started using that memory.

Those are bugs you can have. (Also, very commonly, there is the bug of trying to dereference a pointer which was never initialized to any meaningful value to begin with, or which was purposely set to a default value meant to indicate “I don’t point to anything”). But C/C++ are, by design, close-to-the-metal languages, so such risks come with the territory (perhaps they needn’t, but they do). Incidentally, in case you don’t realize it, but you do seem to, a pointer need not address the memory of a variable in your program, as such; it can address an arbitrary location in memory, regardless of whether there is a particular named variable in your code stored there. This is key for the next bit:

But I just googled them and now I get some of why they’re useful and I get linked lists now too! You couldn’t make those in C++ without pointers. Good stuff.

Indeed; pointers are the key to manipulating dynamically allocated structures in C and other low-level languages (where you have to manually manage memory, rather than relying on some out-of-sight built-in garbage collection system to do it for you). If you couldn’t manipulate memory indirectly through pointers, you could only ever be using as much memory as would correspond to the named variables in your code, each of which has a fixed size; this would, of course, be rather stifling.

[wonky]

Like many others in this thread, I wasted a lot of opportunities at college. I was at the wrongest of wrong schools for me and that made everything… wrong.

But I still learned things. I read, I wrote, I discussed, I learned to dream in Greek.

I gained confidence, too, after going to a terrible high school.

If I had it to do over, oh the changes I’d make.

[Aankh]

@Fear_Itself

I got my degree in 1979, and it would be ten years and three companies before I worked for someone who also graduated from college.

Welcome to 2008, sir! It’s a whole new world.

As to the OP, while it’s possible that you have learned nothing, it’s not very probable. More likely than not, you’ve absored all sorts of intangibles, developed your thinking processes even further, and gained exposure to the type of people who go on to manage corporate America. Don’t discount that last one considering that your goal is a better paypacket!

Since your degree is in pol-sci, I’m sure you put in the effort to, at the very least, do the prescribed readings. If you can tell me with a straight face that none of that helped you at all, gained you any perspective, or added to your knowledge-base over four years, well, then I’ll be forced to chime in with the “It’s not college, it’s you” crowd. All in all, I think you’re an intelligent but pessimistic person, and it’s your pessimistic side that wrote the OP. How does your wife feel about this issue, by the way? (You’re the one whose wife is a regular here too, right?)

As for the crowd jumping on Airman for not paying for his degree, frankly, you guys are twits. If he earned his education by performing national service through the USAF, more power to him.

[Ogre]

Jesus, I must have gone to a great school (I didn’t), because I was constantly learning shit I had no idea about beforehand. Protein chemistry, quantum mechanics, short stories and poems I had never read, organic chemistry, linear algebra, you name it. How does anyone take 4 years of classes and never learn anything?

[ultrafilter]

@Ogre

Jesus, I must have gone to a great school (I didn’t), because I was constantly learning shit I had no idea about beforehand. Protein chemistry, quantum mechanics, short stories and poems I had never read, organic chemistry, linear algebra, you name it. How does anyone take 4 years of classes and never learn anything?

By studying political science, apparently. (I feel justified in joking a bit as my sister is a political science major, and I’ll be the first to admit that I couldn’t cut it in her program. All that reading, and no equations…)

[F.U.Shakespeare]

I went to college after working for a few years. My undergrad degree was in electrical engineering, and I learned a helluva lot of valuable stuff earning it. But I learned almost nothing about history or other liberal arts subjects that I didn’t already know. I graduated in 1985 from a large public university. Based on anecdotal data, I think this is/was fairly common for informed adults majoring in technical subjects.

It was especially galling to have my writing ‘corrected’ by people who couldn’t write. Although there are exceptions, engineering profs and grad assistants are not an especially literate bunch. I once ‘lost’ an argument with a grader who insisted that ‘firstly’ wasn’t a word.

[groo]

I got a lot out of college, but perhaps that’s because I started a little younger. From AD’s posts, it’s obvious that he was more of a fully formed adult when he started college. I learned many, many things in college, and physics, math and writing were the least of my lessons. I learned:
[ul][li]That I couldn’t coast through life cramming for tests the night before.[/li][li]That no matter how smart I thought I was, there were many, many people much smarter than me, and I needed to adopt a good work ethic.[/li][li]That if I’m facing a seemingly impossible problem, that I can and should think of ten thousand different ways of looking at it, and at some point, I may end up solving it. Or I may just end up knowing ten thousand things that don’t solve the problem.[/li][li]That no matter how noble I may feel about my own story, there are lots of people who’ve overcome even greater adversity.[/li][li]That I can, when the chips are down, work and work and work and work until the job is done.[/li][li]That nobody owes me dick, and I have to earn their respect, and that with some people, I may end up never getting their respect, and that I’ll have to deal with that.[/li][li]How to play foosball and pool, how to fix pinball machines, when to stop drinking, how to avoid arguments, how to start arguments, how to budget, how to give CPR and how important it is to cultivate good relationships with quality people.[/li][/ul]
Through my various part-time jobs in college, I also learned:
[ul]
[li]That geologists spend a lot of time walking around in the desert and sleeping in the back of their trucks.[/li][li]That a totally cool awesome job of analyzing Voyager II data could actually just mean many, many nights of tedium in front of a computer, and if I wanted to continue with that, I’d better love it. (It turned out I didn’t love it that much).[/li][li]That people won’t necessarily pay me to be a smart guy, but they will pay me if I can apply my smarts to help them get useful work done. And, of course, they’ll certainly pay me to stack books, prepare surfaces for welding, operate a switchboard (when there were such things) and flip burgers.[/li][li]That other people I see flipping burgers or sweeping floors may actually have a lot of interesting things going on just beneath the surface.[/li][/ul]
Of course, much of this means, “I grew up a little in college,” and it’s likely that these are just lessons correlated with my age at the time. But I think that in the context of academia, I had to grow up while also trying to learn difficult subjects in a disciplined way, which added pressure and brought some clarity. And I was fortunate to have several jobs that were related to fields I was interested in, to help me figure out what I wanted to do. I imagine it’s a different story if you start college long after you know for sure that you’re an adult.

Also, none of you had any choice in the matter, but Pell Grants funded by your (parents’) tax dollars helped me pay for college. Not a lot of help, but it was better than nothing. So, thanks! I hope some of my tax dollars will help repay the favor.

[pseudotriton_ruber_ruber]

@F.U.Shakespeare

‘firstly’ wasn’t a word.

It’s not a very elegant word. Firstly, “First” is much better. Secondly, it’s redundant. Thirdly, it’s longer than “First.” Fourthly, it’s slightly pretentious. Fifthly, it’s unnecessary–if you’re listing things, it’s usually pretty obvious what the first item on your list is. Sixthly, such lists as these sound pretty stupid once you get much beyond “secondly,” don’t they? Seventhly, does anyone really care how many items you’re discussing, or are the items themselves what you want your reader to be focusing on? Eighthly, doesn’t this seem like a stupid and padded way to contruct an argument? It does to me. Ninthly, I’m giving up now, but I could go on for quite a while in this idiotic vein.

Tenthly,

Yours,

PRR

[Mellivora_capensis]

Good post, groo.

[Capt.Ridley_s_Shooting_Party]

@Derleth

[ul]
[li]Common Lisp, because it is a programmable programming language: You approach problems by extending the language, both using functions and macros (nothing like C++ macros), turning your application into a domain-specific language. Common Lisp removes the artificial barrier between the language designer and the language user. Read this book alongside the one I just linked to. It is less practical but it will teach you the Lisp Philosophy a lot more effectively. Some elements of that philosophy only apply (easily) to Lisp, but others (the bottom-up methodology, for example) apply to any language.[/li][li]Haskell is beautiful in other ways. It is probably the best language to learn about type systems from, and when you figure out how to stop fighting the type system you’ll forever after appreciate what type checking is capable of doing for you. (The downside is that the type systems in languages like C++ and Java are a half-assed parody of what Haskell provides.)[/li][/ul]

Good suggestions. I’d remove Bison, though, and suggest Haskell’s Parsec: a good demonstration of why people are excited about Monadic programming. There’s also SML, which is a strict language (as opposed to Haskell, which is lazy) with a well designed module system (something Haskell doesn’t really have!). If you really wanted to blow your mind with type systems, you’d be looking into Epigram, though

[Cubsfan]

@Captain_Haddock

I see your point. Isn’t this why some people take degrees in science, maths and engineering?

Quoted for truth.

People who waste their time getting degrees in meaningless areas of study like history, and any liberal arts have never made any sense to me. Those types of things can be learned just as well by reading books and watching documentaries. There is no advantage to learning the material through a college. There are disadvantages like paying 30K a year to learn what you could learn at the library for free or on a 1K trip to Europe and the fact that your degree means nothing to 99% of the hiring workforce.

Most (not all) people who liberal arts degrees and similiar types of studies do so because it’s easier and less intellectually demanding than the sciences and they just want to check the COLLEGE box on their job application.

If you wanted to learn something useful that truly is a discriminator for hiring (and rightfully so) you would have done science, math or engineering.

[MsRobyn]

Not to put too fine a point on things, but when you’re an adult student, as Airman and I both were, a lot of courses really are easy to the point of being a joke. It may not be the professor’s fault; he’s got enough to worry about when a student is failing, let alone finding stuff for a student with a 95 to do to keep busy and interested. A lot of it comes down to life experience; you’ve seen a lot of what the professor is talking about already, so all you need to do is fill in the details. Even Human Biology, which claimed the GPA of many traditional students, was ridiculously easy for me since I’d been working as a coder and billing specialist and had a good knowledge of anatomy and physiology.

Where we live, there are few, if any, employers who will accept a bachelor’s degree in a hard science; most want at least a master’s degree. The big exceptions are the federal government and state governments. Unfortunately, the state is currently on a hiring freeze.

Robin

[Derleth]

@Capt.Ridley_s_Shooting_Party

Good suggestions. I’d remove Bison, though, and suggest Haskell’s Parsec: a good demonstration of why people are excited about Monadic programming.

The Tao’s Revenge: Monads are absolutely fascinating. I’m telling you this because at first glance they look utterly insane. They’re like macros in Lisp, or first-class continuations in Scheme: Incomprehensible until you realize just how powerful they are.

There’s also SML, which is a strict language (as opposed to Haskell, which is lazy)

Tao: Laziness and strictness are a fine example of how tastes can differ in language design: Laziness makes defining new functions out of existing ones very easy, but strictness can make software easier to read by restricting the syntax further. There is no ‘right’ answer.

with a well designed module system (something Haskell doesn’t really have!).

Haskell does have a module system, but on a brief look at SML’s it isn’t as elaborate as what you’re talking about.

If you really wanted to blow your mind with type systems, you’d be looking into Epigram, though

I learn something new every day: I’ll have to investigate Epigram, because I’ve never heard of it, and I’ve never heard of its type system.

I completely blanked on another language for you to investigate, Tao: Erlang is probably the single best way to explore concurrency. Learning how to think in terms of multiple communicating threads of execution is a truly mind-expanding experience, and Erlang has various tricks (mainly the idea of single-assignment and no state mutations) to keep you from tripping over your own feet too often.

[Capt.Ridley_s_Shooting_Party]

Haskell does have a module system, but on a brief look at SML’s it isn’t as elaborate as what you’re talking about.

Haskell modules aren’t first class, although SPJ has a paper on an extension to Haskell that includes these, and you cannot write functions between modules (“functors”). I also don’t like how type classes and modules interact in Haskell, but cannot seem to put my finger on the problem I have with them!

[capybara]

@Cubsfan

People who waste their time getting degrees in meaningless areas of study like history, and any liberal arts have never made any sense to me. Those types of things can be learned just as well by reading books and watching documentaries. There is no advantage to learning the material through a college. There are disadvantages like paying 30K a year to learn what you could learn at the library for free or on a 1K trip to Europe…

They don’t write books on engineering or maths? Does something special happen in the hard-numbers classrooms that doesn’t happen in the lib arts classrooms? Would you say that you learn to “be” an engineer or mathematician, and that it’s not just rote memorization of procedures and equations? If I read all the textbooks once, that doesn’t make me an engineer, right?

Might you accept that something analogous happens in the lib arts, maybe that one, ideally, learns to write, and think, and be a historian or at least think historically or historiographically, and become a producer of new knowledge rather than just memorizing dates? That one develops critical thinking skills and a breadth of knowledge that would probably not happen at the city library reading “I Claudius” and “Girl with a Pearl Earring”?

Can I assume that you’re an engineering or maths graduate?

[Spice_Weasel]

@even_sven

So they didn’t expect much from your writing? Who cares what they expect? You could have pushed yourself anyway. Whenever I had to write a paper, I never chose the topic that would be easy to write- I chose something that I knew was going to involve a lot of research and thinking. I didn’t have to. I could have done fine with the easy and obvious topics. But I actually wanted to learn something. So I went out and learned it. So what if I could have gotten the same “A” by choosing something easy. The point wasn’t to get the “A”, the point was to learn.

even sven, between this and your postings in the Opal thread, I am liking you more and more with every passing moment.

College changed the way I think forever, and I’d like to think that I was more than just lucky in that regard. I made myself learn. I challenged myself at every opportunity. And it couldn’t have just been the caliber of the University, either–I took a psychology class at the local community college once, and I got a fuckload out of that. If you’re intellectually lazy, fine–maybe school isn’t for you. It isn’t about memorizing and regurgitating a compendium of useless facts, it’s about learning how to use the resources at your disposal in order to develop a cogent analysis of ideas, and finally to communicate that analysis effectively. I would expect anyone who posts here to understand the true value of an education. Don’t go around generalizing saying college is worthless when YOU are the one responsible for the quality of your schooling.

[villa]

@Cubsfan

People who waste their time getting degrees in meaningless areas of study like history, and any liberal arts have never made any sense to me. Those types of things can be learned just as well by reading books and watching documentaries. There is no advantage to learning the material through a college. There are disadvantages like paying 30K a year to learn what you could learn at the library for free or on a 1K trip to Europe and the fact that your degree means nothing to 99% of the hiring workforce.

You really don’t understand the purpose of education do you. That such a narrow, depressing, close-minded viewpoint. The fact that you describe the process as “learning the material” shows you just don’t have a clue.

[Dangerosa]

@MsRobyn

Not to put too fine a point on things, but when you’re an adult student, as Airman and I both were, a lot of courses really are easy to the point of being a joke. It may not be the professor’s fault; he’s got enough to worry about when a student is failing, let alone finding stuff for a student with a 95 to do to keep busy and interested. A lot of it comes down to life experience; you’ve seen a lot of what the professor is talking about already, so all you need to do is fill in the details. Even Human Biology, which claimed the GPA of many traditional students, was ridiculously easy for me since I’d been working as a coder and billing specialist and had a good knowledge of anatomy and physiology.

Sure, meeting the requirements was easy - after all, as an adult learner in an easy school you really only need to know 90% of the easy stuff to get an A. I coasted through and had three A- - the rest As - one A- in an impossibly hard course, one from a instructor who was nuts after I ended up missing quizzes I couldn’t make up for work - and one from a group project with an incompetent group…but I did LEARN. I possibly learned less than students who worked their butts off to get Cs because I’d been exposed to some of the subject matter before.

But Airman learned nothing…got 100% on every test he ever took. Was exposed to zero new ideas. Refreshed his memory on nothing…other than having the degree, completely wasted his time…

Now that is an accomplishment.