I don't need architecture because we just need to get it done quickly.
Update: A friend once told me he bought a system with a 100MB hard drive because that was bigger than all his neighbors' drives. Admittedly, at the time I was still swapping 3.5" floppies, often putting the contents of the first floppy into a ram disk on a system with 1MB RAM so that I could avoid all the swapping to run my PASCAL compiler. The poll option regarding 640KB RAM reminded me of this.
A friend once told me he bought a system with a 100MB hard drive because that was bigger than all his neighbors' drives.
The 640k RAM one similarly reminded me of when someone donated an old TRS-80 Model 12 to my high school, and they turned around and gave it to me. ("Well, we can either store it somewhere for a few years until we finally sell it at an auction for 50 cents... or we can ask Dave if he wants it.") In addition to being my first *nix system (Xenix, to be exact), it also included my first-ever hard drive, holding a whopping 12M. At the time, I was absolutely convinced that it would be all the disk space I would ever need for the rest of my life...
It was working yesterday or was not working yesterday but is working today may be feeble but I've come across it several times, especially on systems with lots of caches, so you have no idea what state the system really is in...!
I'm not sure about worst excuse ever, but one I've heard a lot recently that's pretty awful is some variation of, "we just did what the data indicated."
I actually first learned Perl to essentially do some low level Data Science type of stuff, but I quickly realized the individuals who were asking me to do it (middle managers and low level executives) weren't so much interested in finding out what the data said, but more so in figuring out which data to look at and which to ignore to reinforce their already formulated conclusions when they presented them up the chain of command (higher level executives) to indicate their department is doing a super great job, or whatever.
I got pretty tired of being asked to tweak little things, change how this is measured going forward, or update the calculation to include/exclude things to bolster the numbers in favor of some individual's argument.
I'm not in the same position under the same folks anymore (thankfully), but I still have this Dilbert tacked to my wall as a reminder.
Just another Perl hooker - My clients appreciate that I keep my code clean but my comments dirty.
Ugh. Reminds me of doing some testing of a 3201 (poll-select) communications system with (as it turned out) my ex-employer. There would occasionally be system failures, and one of them was explained away (during the Winter) as 'Snow on the lines'. I was gobsmacked. This was the last 90's. I had nothing.
Fortunately, there was a tech at ex-employer that I'd played hockey with, and he was totally on the ball. When I had him on the phone with a problem, I got it solved right away. So at least there were *some* competent people there. That was a weird period of employment -- leading a major communications project with a company that had .. laid me off as being 'below standard'. Huh.
One debugger I used actually had an option to display the phase of the moon :). And actually that debugger (which was otherwise quite good) had a bug that made the execution misbehave when stepping over some instructions under specific conditions. As you might guess, I was using the debugger trying to find a bug, so I ended up with what I thought was a single bug that was behaving differently depending on how close I was trying to observe it. Lunar Phase Dependency alright.