You all suck at programming

So, I've actually read the whole thing. Yeah, I think I'm trying to do the best hack ever, but I do know programming, and I have experience.

I've learned to program by myself 5 years ago (I was 12 then). I started with Visual Basic for Applications, which is embedded in Microsoft Office programs since 97. I used to record macros to then look at the generated code and see how it worked so I could do it myself. Then I discovered UserForms (basically windows), for which there was a visual editor. I learned most of the notions of programming through the manuals that came with Office (variables, loops, events, functions and subroutines). After a while, I got stole Visual Studio 6 on Kazaa (a virus in the uninstaller??) to start making Windows executables. Then I switched to Visual Basic 2005, with the free Express edition. Then I went to C++ with GCC and MinGW. Then I learned 68k ASM :P. I'm not a script kiddie, I know how to program, but learning how to program for the Mega Drive/Genesis takes time, even more when you work in a game you know nothing of.

I believe I'm in that situation too :D.

That depends on the language. High-level languages (VB, C#, C++) have common constructs and you can switch from one to another after learning the language, but lower-level languages (ASM) don't have them and are used in a completely different way (however switching between assembly languages is pretty much the same as switching between VB.NET and C#). Sure, knowing common programming concepts help a lot, but ASM is a lot more complex, even more when the code isn't commented .

However, I understand how you feel. Programming is fun when it works, but sure is irritating when it doesn't work; it's just even more rewarding when you finally get it to work :D.

 

Completely agreed. I know Basic since I was 4, and when I had to attend a C++ class at the University I only had to learn some different syntax stuff. There's no point in learning a language, many people open their VB or VC++ environment and copy lines of codes from templates and the online help, and they're even likely to get working programs in this way, but that's not programming. You won't have the basis. This is something that's always understimate by a lot of people, you need to learn how to program, not how to use [insert language here]'s online help.
I agree with every single line of your post, programming nowadays is taking a very bad direction. I even posted about that some months ago. The direction of my rant was slightly different, but I believe it's complementary.

jman, I looked several times at what you've done (like your custom DAC thing), and let me take this chance to say I have a very high respect and consideration of you, as you've proven more than once to be a good programmer. And, what's important, you've proven to be a real programmer. People like you, Cinossu, JoseTB, Hayate, Nemesis, Puto, Stealth really deserve that green label near your names.

I'd have more to say on this topic but I'm at work right now and I don't have enough time.

 

I know I suck at programming. :P

It's one of the many reasons I've stopped working on my hack. I simply don't have the time or patience (or brain capacity, probably) to learn how to program anything.

 

I understand where your frustration comes from, but for some people, this is how they learn the way to do things. Now sadly, it is true that people will just say "this isn't documented? Ah well, not gonna bother doing it myself." but there are plenty of people who take this in another direction.

The way I learn is this - I find a program or a skill I want to learn. I find some tutorials about this skill or program. Now after a few basic tutorials, I start to get curious and I want to explore my own ideas.

BUT, there is one thing that any smart programmer should know - you shouldn't go through the hassle of doing certain pieces of work if it's already done and available. It's pointless and unwise to waste your time on something that's already been done. Just give credit where credit is due.

So do you see where I'm coming from? Yes, it's very true that some people have a total lack of real skill, and just use others work to create their own. That sucks. But it's not something you should get too wrapped up in. If you do, and it catches on, we'll end up like the Zelda hacking community. And right now, they're in a bad place.

Sites are attacking one another and nothing is really getting done on either side. There's been a divide created between the "1337" hackers and the lesser-skilled people who just want to see the cool things that come up. Sometimes they like to use the available resources to try their hand at "hacking" with tools that hackers have made available, and there's nothing wrong with that either. When someone uses another hacker's program to create something, we already know who to give credit to even if they don't mention it.

anyway, put simply:

Some people are n00bs, but some people are just trying to gain legitimate skills. Don't let it eat at you. The bad seeds will die off if that's really what they are.

 

Fixed.

 

I'll be the first to admit that I'm absolutely shit at programming, but a lot of what I've done hasn't been lifted from any guides, so I can't be doing too bad. That said, I really prefer to work on level design, but my first priority when getting into this was to add all the features to Sonic 2 that I ever wanted, and I feel that I've accomplished that. I doubt I'll even bother to do music ports and the like because that stuff is really over my head.

 

Haha.. I know I can't program for shit—I've always been more of a reverse-engineer by nature. I see how things work, and I try and replicate or modify it to suit my needs in whatever way necessary. When it came to coding things from scratch, I could always work out the process, but when it came time to write something I'd always freeze. That's probably why I'm never going to write any Megadrive programs from scratch—I don't have the testicular fortitude for it. =P

To be honest, the only reason I've even touched a programming language (and my first was still 68k) was because fuzzbuzz bugged me to do so. I didn't really "get it" until I had a summer without internet access. Within that time, I worked on a Sonic 3 hack where I had little online reference, so I started programming in machine code, starting to do some pretty neat things with it. As I got more comfortable with things, the concept of 68k ASM (and programming in general) became extremely simple to me. No longer was I stuck wondering "How do I do this?" because I understood the logical flow behind programming. I couldn't tell you the kind of feeling that was—yeah, programming is tedious and sometimes boring, but once you first understand things, you're a lot more eager to try and push your skills further than you ever thought they could go before.

My programming abilities are usually limited to analyzation, understanding, and modification. That's why I work with the Sonic engine—it's well documented enough to be able to do what I want to do, especially with the modular nature of the SST and other related aspects of the engine. Generally speaking, it doesn't take a huge amount of programming experience to do some neat things, especially when it comes to Sonic hacking, but the more the merrier, I say.

My concern is, rather, how you consider those who excel in manipulating data, rather than code. For example, since my main specialty is and always has been editing music in the SMPS format, do I "suck" too because my programming skill isn't superb? I think there's many factors to consider in a situation like this when it comes to judging a person's competency.

 

It's ironic, because I program professionally in C/C++/C#/VB/Java, but I can't program in ASM worth shit because I never bothered to learn. I'm like the opposite of half this forum.

 

I contradict the title of this topic.

Yeah, so I use guides when they exist and ask people when I can, but otherwise I damn well do it myself. I believe there's no point re-doing something if someone else has already done it in a way you can use.

 

This sums it up pretty well. In my hack, I used a guide and Puto's help to add the Spin Dash, since there was no point into redoing that from scratch; maybe I would have been able to, maybe I wouldn't, but that's not relevant. For the music, instead, as there was no public guide about using a different song in each act, I had to think about it, and so I did.
I don't like to rely on other people's code if I can avoid that, but it can still be good under certain circumstances. And this is why in xm3smps there are credits for Puto, Solaris, Tweaker, and Wiz.

You're not alone. C, C++, QB, VB, Pascal here. I'm not that bad at x86 asm btw, so I can understand some of 68k asm.

 

I think the problem isn't that there are guides, but that the guides assume you know too little and end up at the "copy this code to this point in the disassembly" level, which is really no better than the old graphics hacking techniques (decompress data from address, edit, compress, paste back into ROM). The best guides are the ones which say "here's how it works in broad detail giving pointers into the code to illustrate and allow people to read" and leaving it as an exercise to the reader to come up with different ways of doing it.

Given two guides which explain the drawing code for S2 and S3K in this fashion I'm fairly certain I'd be able to make some headway towards recreating the S3K drawing code in S2 (which is something I really want to do - the problem is I can't because I can't read the code so well due to it being so tightly optimised and using a lot of memory addresses nobody has identified).

 

Well, in any case there's a reason that my title isn't in green.

 

You know, actually, I'm seeing a tend here: Nobody wants to take the initiative. It's all about waiting for a "bigwig" to do a majority of the work, and then making rather uneducated edits based on that work. Why can't people, like... do things themselves?

Case in point, Sonic for MegaCD. There's a huge fight over why "reinventing the wheel" is unnecessary and that therefore Stealth must release his work. I'm sorry, but... what? What the hell is stopping you from going out and doing the same damn thing? The documents are there, the utilities are there, and there is nothing that Stealth used that anyone else couldn't find and use themselves!

Is it due to intimidation? Probably. But don't ever "expect" things to get done by other people. There is no wheel to reinvent—the method is, and always was, there. But if you want to make a version of Sonic for MegaCD that anyone can edit as they please, continuing this trend further, then you can be their savior, not people like Stealth and co who actually give enough of a damn to do their own research, testing, and come out with a quality product.

 

Kind of like what Quickman said, I find lists of pointers to be VERY useful. Knowing what does what seems to be the most important thing in our case when trying to understand the assembly. This is why I'm REALLY thankful for those who comment the dissasemblies. Maybe I wouldn't be TOTALLY lost without them, but I'd certainly have a much tougher time doing anything. Knowing what object number corresponds to which badnik or level object really makes all the difference. Once you know the objects, just observing how they interact in the code is really enough to try and start understand it all.

 

I believe the reason behind this is that people feel that if they don't use Assembly in their hacks, they won't be treated with much respect, even if they don't have a peticular interest in learning ASM. So they take the easy option in order to furfill this demand.

However, those things alone without any ASM often tend to recieve critisism, though there is a definate lack of good level edits. People have almost come to expect some level of ASM usage now though.

I don't see what is peticularly wrong with this for one simple reason: it's how a lot of people learn. I mean in my guides for example I explain how EVERYTHING happens, I don't just do all the legwork. There are other guides that are basically copypasta guides, and I agree those can be percieved as a problem.

I do think it's rather patronising to assume that when people do this they consider themselves programmers. I think most people know they don't really understand the ASM code they've just pasted or whatever.

This is an interesting point though. What is programming? I mean if you look at Basic, to print something to the screen, all you have to use is "PRINT". Could that be considered cheating because the person has not written their own PRINT command?

It wouldn't be unreasonable to suggest that copy and pasting specific features into Sonic was just very high level (not high as in skill) programming. Certainly not ASM programming though.

I disagree, in some cases ASM can be used as a stepping stone to learning how to program.

How depressing is that! I personally find whipping together a program quite fun, with it's problem solving. Sure you sometimes hit brick walls but just because YOU feel those feelings about it doesn't mean everyone else does.

Hardly a secret.

You do have a valid point with this. Programming is a lot more than just the codes and commands. It is a methology.

SONIC SEZ: Kids there's nothing more cool than putting ASM in a Sonic hack, but when someone tries to learn ASM without first learning the concepts of programming that's no guuud!

Seems completely justified to me. Interesting post, nice to see you care to be honest!

 

I'm glad to have brought up this topic on IRC yesterday.

I agree with most of what you say, except this:

You don't get born with programming skills, yes. To be a good programmer can take several years of experience (whether it be learning algorithms, discovering how computing/programming works, etc.)

However, not everyone can be a programmer. To be a programmer, one must have a certain set of abilities that you can only be born with. So yes, it's also a gift. Not many people can wake up one day, say 'Hey, I want to be a programmer' and become one.

Programmers need to have the curiosity, the patience, the wits, the logical way of thinking. Most people can tell OR from XOR, in regular English sentences. Most people have sucky mathematical skills, and as a result, suck at programming.

How to tell a good programmer? Easy. Good programmers tend to enjoy programming, despite all the hard work it requires. You'll always see a good programmer tinker with code, hardware, new things, functions, etc. in his free time. You'll never see a good programmer count seconds until the end of work, waiting for the magic hour to happen (unless he's finished already, or has a crappy job, but that's another thing). Good programmers generally (not always) have started their journeys with programming at an early age - 10, 15, 20 (it's all relatively early). Good programmers are akin to good mathematicians/logicians -- if you ask them an interesting question, one that will spark their curiosity, you will see them wonder for hours until they find a solution.

What about the rest?

Well, that's where the very important yet often overlooked distinction between the words 'programmer' and 'coder' kick in. It doesn't take a good programmer to code. Even a fucking chimpanzee with a keyboard could code. That's what your average code specialist or junior programmer does -- punch the keyboard like mad to produce that 100000 lines a day. Is there any thinking involved in that? Fuck no.

It's the programmers who do the actual thinking. It's a very important distinction to remember.

Of course, the code monkeys will get very defensive when you call them code monkeys, but that's the way it is, and don't let anyone tell you otherwise

This is an interesting topic indeed.

 

While I do agree with most of what has been said in this topic, I don't see what's the point on getting mad about other people's effort. Sorry but as much as some may hate it this is not about recognition. Everyone can come up with an assembly hack fully done by copy pasting from tutorials and still have nearly no idea about programming - so what? If anything, this could motivate the author to actually learn to do something real. But getting mad about it? Why?

Regarding the wheel being reinvented all the time - I'd love to have 48 hours each day to work and finish whatever I want, but the truth is that real life + real job, etc is consuming most of my time, in my case. Several unfinished/frozen projects know this quite well...

 

A bit on topic, a bit not -- a bit of personal pet peeve is when people can't tell O(n log n) algorithms from O(2^n)... Or don't know SHIT about programming (or algorithms) and write commercial software. I'm tired of people writing really crappy software only because the processing power is huge on personal computers, resulting in lots of bloat. This wouldn't slide 20 years ago.

Hopefully this will change soon, as I think we are slowly approaching a limit in CPU power (as evidenced by CPU manufacturers going for more cores rather than bigger clocks speeds / smaller chips / better code).