It can be from anything from reinventing Notepad to something so worse that no one can imagine.
So? What kind of skeleton do you have in your closet?
It can be from anything from reinventing Notepad to something so worse that no one can imagine.
So? What kind of skeleton do you have in your closet?
I tried to write a blog engine.
After a month... I just finished by settling on a tested blog engine. I did that after realizing the amount of features is required from a blog engine.
a 2D CAD system for the Commodore-64, written in Assembly Language. After getting 80% of the features implemented, we realized that the platform really could not handle a circuit schematic with more than ten or so elements in it - that and the macro-assembler took so long to build the program that the floppy-disk drive would overheat
i found out years later that 'real' C64 developers wrote their code on a PET system and used a cross-compiler
on the plus side, this project forced me to learn how to write efficient assembly language, and reinvent/discover object-oriented principles well before they became well-known in the general market
Not my idea, but... a soft PLC. On Windows NT, in Delphi 4.
In three weeks.
Someone thought it was a good idea :(
Porn for Computers Dot Com. yes we took pictures of computers without cases on. You can also tell from the front page that I had an unhealthy obsession with Slashdot at the time as well. I didn't link here because it is mildly NSFW depending on your view of things.
But essentially a 10 second loop of music, really bad tiled image background and pictures of computers, toasters, etc with no case on.
If you wish to see my horrible idea the Wayback Machine has it.
I made something more complicated than it needed to be. But I only had to learn that lesson once.
as most of the computer science students, an operating system ...
I started working on a striped down linux distribution that essentially just loaded the Mono CLR.
The idea was to create a "semi-managed" operation environment with a main parent process that would load applications into individual app domains, and function as a window manager etc.
Needless to say, I only worked on it for a few days before deciding it wasn't worth the effort.
A conversion program to reformat TextMate themes for gedit and vice-versa. I still kind of want to finish it, but it's infinitely easier just to make a new theme for either. Meh.
I wrote minesweeper in Java. I was pretty proud of it, until my roommate bluntly told me that he wasn't impressed because I pretty much just made a crappy copy of a product that doesn't need improvement.
Mine is rewriting the XML parser in Actionscript 3 using a lot of recursion to copy XML objects into standard Objects, without really considering the existing methods of reading XML data, which were already a big improvement from Actionscript 2. The code got so complex so fast I ended up not being able to follow my own logic and had to put the thing on hold. In general my experience with Actionscript 3 has not been a pleasant one. Simple tasks take a long time to figure out.
I was working with another student to create a program that would connect to a database for something or the other, i think it was supposed to be scheduling software, although a pretty basic version of it. It was my first foray into GUI programming, and we were using DevC++ with the wxWidgets. We had the program looking nice, with a menubar across the top and everything set up to read from the database and format it to the screen.
Only problem was we couldn't figure out how to actually connect to the database...i don't know if he ever got it working, i think we did pretty badly on it hah.
Ruby supports monkey-patching (also referred to as duck punching). If a class or library doesn't do what you want it to, you can extend or modify it, at run time, so that it does.
I wanted to add LDAP authentication to a Ruby on Rails site I was developing, so, flushed with power from just learning Ruby, I had the brilliant idea of taking a rather immature LDAP library and a rather immature authentication framework and monkey-patching the living daylights out of them until they worked together.
Everything worked flawlessly until I tried to upgrade Ruby on Rails, or the LDAP library, or the authentication framework, at which point the entire application broke into a million pieces.
Just because you can do something in any given language doesn't mean you should.
A Dragonball Z power level modification for the vBulletin forum software.
Don't laugh too hard.
At one of my first jobs, I was taskless for my first two weeks. So I wrote minesweeper in excel... oh wait, you wanted the worst software idea...
Digital Valet for the PC XT. View your wardrobe in beautiful 16-color resolution. Make choices about how to dress based on this information.
Where do I begin
It isn't one idea in particular, it's just way to often when I go. "I'm gonna take this tool and abstract it into something I can use. OMG why hasn't his been done before... I'm a genius"
Just to realize this has been done.... professionally.
An example. Abstracting the basic mysql functions in PHP.... when there is PDO.
I've gotten into a few quixotic projects, but I at least learned something and became a better programmer in each case. I've never had a situation where I got locked into a bad decision I've made. Generally if something's not working I'll give up on it quickly enough it won't come back to bite me in the ass.
The programmers who created things that I've had to maintain on the other hand, now those guys had some seriously bad ideas.