lol. I write a bit of code for data analysis. It's not a deliverable, aside from scientific peer reviews of the datasets and resulting papers. Isn't that kind of wild? I'm claiming to have patterns indicative of physical processes in my data, and the way about which I have put this data into its final form is a black box. So the check of my code's outputs is whether or not they can be explained mathematically. But thank god no one else has to use it. "If you'd just comment out line 71, and then un-comment line 298, like we talked about, Dave..."Issue Resolved: 317 clicks have corrected the error. :D
I write scripts to do stuff like this. I'm not a good programmer, but I'm a lazy engineer, and lazy trumps not good. Occasionally someone will say "you should share that script with the group." Terrible idea. Forgot to manually delete the top two rows of headers from the text file? Horrible errors as it tries to parse and do math on text. Have the data in the right format but columns in a different order? No errors and output that's wrong but not obviously wrong. Someone at work is going to run a study nobody but me has done. I've told them repeatedly that they can't just mimic what I did, they have to understand it. More recently I told them they can use the stuff I used but should consider it utterly unsupported. They gave me this look of shock I can only describe as a spoiled teenager told they aren't getting a new car on their 16th birthday. They invited me to their meetings. With my boss's approval, I declined the invite and took some pleasure in doing so.
I haven't, but I'm going to jot that down. I never liked that interview question, and I love the way you spun it.
Reminded me of the phenomenon web &/or print designers face, when clients/coworkers act all, "my son has photoshop so it can't be that hard" or "all you need is a computer and the software," treating you like a replaceable monkey sitting at a computer. No. You need the computer, the software AND someone with years of experience knowing how to use them to achieve what they want.
QA Consultant here. I can't stop laughing, and I'm dearly sorry for what I do - but the only thing that makes me weep is because I'm under like, 3 layers of leadership between me and the client, I can't even tell you directly, clearly what's wrong - I have to use software and methods that are often imprecise to tell you, and I can't take the liberty of adding a note on how to fix it because "it belittles you". But seriously though - most common errors made is because you look too hard to get the answer and stop caring how you get there. (And those who aren't is because you're dead tired after many restless nights because of me :D) PS: if I may give a tip, when you plan on having several very similar elements, don't instantiate. Build yourself the tools so that the original serves as a template and the rest is modifications on top of the template. Because otherwise you get inconsistencies and can even break everything else.
I refuse to repost this as it's pretty much everywhere...but check it out if you haven't yet http://www.gamasutra.com/view/news/249475/More_dirty_coding_tricks_from_game_developers.php
Oh man, I think I know what bug they're talking about with the PS3 UE3 code. I worked on a project that had a similar issue, and I suspect they used a similar fix. At any rate - good stuff.