Do you want to play a game?
Everyone always tries to make it about themselves…
ToDo List Apps are the Mouse Trap of programming. Everyone wants to build a better one, most of them only work soso, and really the simplest is what gets used.
Do you want to play a game?
Everyone always tries to make it about themselves…
ToDo List Apps are the Mouse Trap of programming. Everyone wants to build a better one, most of them only work soso, and really the simplest is what gets used.
I wonder how long it will take for an equivalent of “found object” art to arise in programming?
Yes, just slavishly cut’n’pasting is n00b behavior; but if one challenged oneself to strictly cut-and-paste only, no editing, putting together an actually functional result could actually be a nontrivial challenge. As perverse as it is challenging, of course; but a nontrivial challenge.
Maybe it was in college. eh?
I was reading the paper and I see a description of the kinds code snippets their algorithm explores:
So, to be a full buzzkill, the algorithm doesn’t copy code off of stackoverflow or anything like that. It just has a small list of functional programming primitives plus basic math and conditionals in a domain specific language (DSL). Those functional programming primitives are the “snippets”.
The programs they list in the appendix look simple enough (longest 7 lines) that they should be both human understandable and provably correct.
“It could allow non-coders to simply describe an idea for a program and let the system build it”
Getting the “non-coders” to actually describe the program they needs is hard. So hard that most people don’t do it and skip directly to build a prototype/MVP and iterate.
"The function should return a scalar… that is, unless the input distribution has both independent and dependent components, in which case the output is a vector between M and P units in length. If this is the case, the correlation matrix and a vector of Booleans should also be returned… no, wait… "
I wish I was joking.
It’s Friday. Woo.
The most important question I have ever learned to ask in response to a feature request is “Wait, what problem are we actually trying to solve here”
Because half of all software requests are the equivalent of asking to upgrade a revolving door to turn constantly at 60 RPM - and when you finally get to the question of why someone requested that, it turns out that what they really wanted was improved airflow, so you do the software equivalent of opening a window
^^^YES THIS^^^
This topic was automatically closed after 5 days. New replies are no longer allowed.