Greatest programming achievement

[SolarFlare]

What is the best thing you ever programmed? What is the most unusual thing you have programmed?

[Yves M]

That's a really hard question to answer ;) Some programs are really hard to write down, others really hard to design, others are great because of the circumstances etc...

Well, the programs that I wrote that are memorable to me are these :


At highscool, the last day of school in the programming class. The teacher was bored and said we could do anything we wanted. There was no internet access, so mainly people were chatting or playing solitaire *sigh*. I told my teacher that we could be playing a game of Tron (like snake but for two people). He didn't believe I could get that to work. In a matter of 10 minutes I had programmed from scratch a version of Tron which was playable. And the people in the class started competing against each other.
Still in highscool, I entered a programming competition and had to solve quite a hard problem. It would have been easy with recursion but I soon figured out it would be much too slow. Spending a whole week thinking about it, I came up with a dynamic programming solution without ever having heard anything about this technique.
College - solved a hard pure math problem on a supercomputer for my final year project. It took me pretty much six months only to understand the maths behind the problem and then programming a supercomputer was a lot of fun. even though I wasn't doing this full-time, I guess it was one of the most intense moments of programming for me.
Summerjob in College Years - implemented a highly specialized database engine for linguistic queries in the space of two weeks. I started off by reading three Phd thesises on several aspects of the problem and then implemented my own solution. It was amazing to me how I could ever accomplish this.
Present company - implementing a matching algorithm for linguistics, improving on the design and again reading Phd thesises and research papers.

I know this all sounds like boasting, but you wanted the programs I'm most proud of. I'm not gonna tell you about the grids I've been implementing and which fail on design one after another, so I'll stick to these ;)

[SolarFlare]

cool!
;) :cool:

[Yves M]

Oh I forgot the most unusual program... That's hard too, what do you think of as unusual ?

I go for this program then :

As a second year project in college I was in a team of six that had to implement a Java animator for labeled transition systems. They are basically like graphs, but each link in the graph is assigned a label and the nodes themselves don't matter. This kind of stuff is used in program verification or design verification. I did all the graph stuff and the logic behind it. When our group met up with the supervisor, the rest of the group would shut up while I was discussing internals with the supervisor. Maybe this was because I was the only to understand scottish, I don't know ;) Anyways, we completed the project, but the supervisor objected to some design issues in my part. In my report I managed to reject all of his issues and actually convince him that we had done the right thing. I guess it was my first experience at "commercializing" one of my products. I guess that worked pretty well since we got rated top, above people who had implemented a 3D editor and viewer, a radio network complete with radio devices and drivers...

[Yves M]

Originally posted by solarflare
cool!
;) :cool:

Well, tell yours :)

[SolarFlare]

Actually, although I'm an enthusiastic programmer, I have never taken a programming course longer than a week- meaning I don't really know much (even if I claim to), and everything I have done in vcpp is self-taught. I have never done anything professional; most of what I do is for my own personal benefit.

Well, due to popular request ;) here are my memoriable progs (which wane in comparison to Yves'):


A program which is like a virtual flashcard: it asks you one thing and you have to give the equivalent- you can save and load lists, etc.- basically for study purposes
A program which records the time of each visitor to my computer (whenever it is turned on), and gives them a status, so I know if someone used my computer when they thought I wouldn't know
A program which emulates IE which I wrote for my mom so she could read her email and use the internet without it being really complicated (she is computer illiterate)


Obviously, nothing at all amazing, which is why I didn't jump out and say them from the start. Just stuff I made cuz I love to program.

:D :D ;) :)

[Yves M]

# A program which emulates IE which I wrote for my mom so she could read her email and use the internet without it being really complicated (she is computer illiterate)

Well, honestly, I think this IS amazing ;) It may not earn you a PhD to implement such a thing, but if your mom can use it without getting confused then this means you're doing something right which other people aren't ;)

[Xeon]

Need we say more?
Yves is totally a genius........Leonardo Da Vinci and Albert Einstein of the programming world! Crazy!
I think your mom accidentally drank some Machine Language potion or ASM milk while giving birth to you, and your dad probably ate the wildest and most powerful mainframes for breakfast and gulps down C++ and the universe's craziest and sleekest algorithms for dinner, coupled up with everything that was seen in the advanced programming world.

This makes one ultimate super-human, undestructible : Yves M.
U really should be more than powerful to survive on programming contests alone, Yves. Go quit your job and enter into all those programming contests. If you lose even 1 of them outta 1000000000000, I make myself into a ****.
That means, u're fated to win, pal.

Aye~! If only I was something like you.........ohhhh.......

[Gabriel Fleseriu]

I wouldn't dare saying that I have programmed something amazing :) The coolest thing was an operating system for some custom boards that used a 16 bit Mitsubishi MPU. The whole thing worked distributed using a weird kind of network. I had a hardly documented C cross-compiler and a couple of books (no internet!). The amazing part is that it worked. :)

[Xeon]

That's crazy! Hmmm......I wonder how u folks get the chance to touch these stuff in the first place......cool! :) :) :)

[SolarFlare]

Xeon- Greatest programming achievement?

[Amn]

call me Bond...Xeon Bond. hi hi.

I really dont have any greatest programming achievement. This is because i am a perfectionist. On each project i complete, my head is already miles ahead. Many may think thats an advantage and a road to fast-learning, but i disagree sometimes. It s like drugs, when you only program for the sake of programming. However i had my moments too...

Once me and mah friend started a discussion on 3d modelling software, and those days only 3DS 4 was available (DOS times yeah...). So i came home and spent a week with EGA modes ( i needed high resolution and wouldnt bore myself with VESA) and inline assembler. The result was pretty tiny program which modelled wireframe 3d objects. Interface was dead simple and very lowlevel. I remember i was high of my achievement a whole year after.

Also our company had a guy who developed a chat-server which communicates with Macromedia Flash clients in browser. THe guy suddenly quit the project, taking the source code in DElphi and all other files with him. All there was left is a binary from the server running and lots of kids that chatted online. Later it turned out the server was buggy and had a habit of deadlocking every day sometimes twice a day. The boss was freaking out, and I proposed a thought, which i hardely believe i had that time, namely, that i could reverse-engineer the protocol and build another server in C++ without anything having to be replaced except the server binary. It took me 4 months, and lot of work, not to mention my selfesteem oscilating like a sine function ;) After all i was a victim of my big mouth. So i had to do it. And i did.
My server also had bugs at first, but i eliminated every last one of them mercelessly ;) That is probably one of my greatest achievements if you will.

Funny thing is most of my "greatest" works are made when i was employed, i just dont come up with software when sitting at home, only frameworks and R&D... I guess i work hard under pressure... hehe.

By the way dont you agree guys, that best works are not always those that are perfected and debugged, but those that are like sex - very short lived and very violent in production ;)

[dimm_coder]

There are programms I wrote:

- At 1-st course I wrote some cool DOS-programms, elustrated some math and phis processes.

- When I was studing VC++ by myself, it was early
than in University, I wrote Tetris clone. It was
cool.

- At 3-d course of University I wrote program SysInfo (like Task Manager on Win NT/2000), but with more extend information: loaded modules, threads, virtual address space for each process, info about kernel drivers.

- Then on 4 course I began to work ( ooohh it even was at 3-d course but only little ) in firm which developes antivirus software. At first I developed code for analyzing virtual address space of processes for checking it on viruses.

- Last time I was writting software for remote administrate of our products. It like M$ DCOM (of caurse more simple) but used on cross-platform base. On this base I wrote common console for administrate programs.
For ex: We have server antivirus monitor for M$ Exchange 2000. It's like win-service running on server machine. It's contain special object for my "DCOM" and resources of dialogs it needed for it configuration. Then I run my console on remote computer. I connect to this server. My software create object on server and put pointer to it interface for console. After that console work with it like with simple object (like DCOM). Console put pointer to callback object to server. This special callback object can work with console view and change it. Dialog resources convert to my format and put on console where will create real view of configurator.
Then user can GUI configurete remote program.
Each GUI event go to the server and then server by callback object can reorganize console view.
I wrote it not only for Win but Linux now to.

- Soon I will be write antivirus software for Linux mail servers.

fuff... :rolleyes:

[Xeon]

By the way dont you agree guys, that best works are not always those that are perfected and debugged, but those that are like sex - very short lived and very violent in production

Hmmm......violent.......that's right. That's why you need to buy lubricants, or else you're gonna see yourself in a hospital or something.
This brand called "Ulti-Lube" by F-Xeon Inc., is really cool. Cheap and powerful too. :D

And nah, guys! Unlike you gods who have created stuff that are worth boasting, my greatest achievements include using the MessageBox(...) function and some simple CDC stuff in MFC. Nothing more than that.

Of course, that's totally laughable, but please don't laugh, cos' u're gonna injured my ego seriously. :):D

[Platinum Plus]

Originally posted by Gabriel Fleseriu The coolest thing was an operating system for some custom boards that used a 16 bit Mitsubishi MPU. The whole thing worked distributed using a weird kind of network. I had a hardly documented C cross-compiler and a couple of books (no internet!). The amazing part is that it worked. :)
Gees Gabriel, that is truely amazing! If you wrote an OS that worked why don't you get going on your own OS - call it GFos. You can place it somewhere on the forum and give us tips n' tricks on how we can develop it further into a fully fledged OS to compete with any OS and take over all the machines !

[cup]

I suppose my greatest achievement when I was a youngster/fresh graddy was bypassing the dongle on a commercial Wordprocessor on a Commodore PET. This was in the early 80s when all the magazines were saying that the dongle was a fantastic anti-piracy device and that it would take months to break it. In fact, it only took 2 days the first time, 1 hour the second time. I sent an article to the magazine that made this declaration, detailing how easy it was to break the security. It was entitled "How to break the golden on Dr Fat Crow", golden and "Dr Fat Crow" being anagrams of dongle and the wordprocessor. Needless to say, it never got published as it probably showed how little these reporters knew. Anyway, wordprocessor vendor might have sued them. Since then, I've always read magazine articles from 'experts/consultants' with a pinch of salt.

If the security for the system had been written by a bad programmer, it would have been difficult to crack. As it was, all the security was based in one routine that was called from all over the place. All I had to do was figure out where that routine was. This was easy: it was the most frequently called routine. Then unscramble the code, bypass the dongle checks, rescramble the code, patch in the new checksum and that was it. In case you are wondering, this was all done in 6502 assembler/machine code.

I suppose the moral is: if you want a secure system, get a bad programmer: someone who isn't organized, doesn't think logically and doesn't read specifications properly. Also, they have to be the Not Invented Here type of person. They'll put in so much crap all over the place, it will be really difficult to break it, even if you have the source code and documentation in front of you. If you've ever had to unravel code written by this sort of person, you'll know what I mean. Only problem is, how would you put out a job advert for someone like that?

[Gabriel Fleseriu]

Originally posted by Platinum Plus

Gees Gabriel, that is truely amazing! If you wrote an OS that worked why don't you get going on your own OS - call it GFos. You can place it somewhere on the forum and give us tips n' tricks on how we can develop it further into a fully fledged OS to compete with any OS and take over all the machines !

There is a little difference in complexity between a Mitsubishi MPU board and a modern mainboard :)

For a single programmer, I think it would be an immense achievement if he/she would be able to clone DOS3.0 or so.

[cup]

For a single programmer, I think it would be an immense achievement if he/she would be able to clone DOS3.0 or so.
I'll second that.

[dimm_coder]

It's real cool cup!!!
But in some I disagree with you is that for good security U need bad programmer.
That's system U described was very bad planed.
But, there is practically no programs that can't be forced, I think.
Especcialy if it middle programs, like WinAmp, WinZip, RAR, etc.
All you need is good debugger ( for ex. SoftIce ), and then U can explore internal logic of program (every function call , etc ).
I did it some time for hobby only.
So if U want hacker was muddled some, you mast write more difficalt internal program logic, and it's possible may be for programmer with not ordinal thinking.
And also :
If you want to protect your program good U mast write such algoritms that will make energies for it breaking > than need in your program for some one who want break it.
;) :D
And there is no anything totally protected :cool:

[Gabriel Fleseriu]

Originally posted by dimm_coder
If you want to protect your program good U mast write such algoritms that will make energies for it breaking > than need in your program for some one who want break it.
;) :D
And there is no anything totally protected :cool:

I think that is the idea behind protecting software: make the protection hard enough to break so it doesn't pay off breaking it.