4 ms·
sometimes big companies hire me to tell them what to do with their beans, then they are disappointed that i did not tell them to stick them up their noses. in
by bauchidgw 15y ago
sometimes big companies hire me to tell them what to do with their beans, then they are disappointed that i did not tell them to stick them up their noses.
in the next step they claim i missunderstood the question, the question was not really about "what to do with the beans", it was more about "how should we stick them up our noses in the best way"
- JoshTriplett 15y agoThere's a reason that Open Source communities often react to strange questions with "can you explain what you want to do?" or a similar request for the bigger picture. Too many people start down a chain of reasoning based on how they want to solve a problem, get to step 7 of 10 of a bad plan, and ask for help proceeding to step 8. Explaining how to get to step 8 solves the problem they think they have, but asking for the big picture can lead to a much better solution to the problem they actually care about.
- JangoSteve 15y agoThe problem with this sort of response is that, somewhere down the road, someone else using that open-source project will be working on an entirely different problem and actually have a legitimate need to make that same jump from point a to point b. Naturally they'll hit up the googles and get excited when they see that someone else had that exact problem before. But then, instead of a solution, they are presented with 15 answers that all say "you're doing it wrong". This is why my favorite answers are those that say something along the lines of, "technically, the answer to your question is... but the fact that you're having this problem may be an indicator that there's a better way."
- JoshTriplett 15y agoOn the other hand, I love when I search for the answer to some specific question I think I have, and find an answer that points to the thing I ought to use instead. People often follow the same wrong paths to a solution.
- JangoSteve 15y agoWhich is why my favorite answer includes both (e.g. "if you really want to do it this way, it's like this... but you should consider this way..."), rather than simply assuming that I don't know what I'm doing.
- pnathan 15y agoI think for small scale problems, it's easier just to have a 'wrong' solution that works right, than it is to go the 'right' solution which inevitably involves more complexities and analysis.
- JeremyBanks 15y agoI've found this to be a common issue on Stack Overflow. Someone will ask a question that might indicate they're doing something wrong, and all of the answers will be pointing that out, usually ignoring the possibility that the user has a legitimate reason for wanting to do whatever it is. A current example is a question I asked recently: How do you safely use variable table/column names with SQLite in Python? [1] I want to write a slightly high-level wrapper over Python's SQLite API, so I need to be able to escape table/column names. The first response is one of Stack Exchange's top users telling me the premise of my question is "evil". It's very frustrating. [1] http://stackoverflow.com/questions/6514274/how-do-you-safely-use-variable-table-column-names-with-sqlite-in-python http://stackoverflow.com/questions/6514274/how-do-you-safely...