Timeline for Naming a method that does the same thing faster but only approximates the result?
Current License: CC BY-SA 4.0
17 events
| when toggle format | what | by | license | comment | |
|---|---|---|---|---|---|
| Jul 7, 2023 at 18:03 | comment | added | user949300 | In addition to several other good suggestions, I'll toss "QuickAndDirty" into the ring. | |
| Jul 7, 2023 at 10:15 | history | protected | gnat | ||
| Jul 6, 2023 at 17:06 | answer | added | gnasher729 | timeline score: 2 | |
| Jul 6, 2023 at 14:49 | answer | added | svidgen | timeline score: 4 | |
| Jul 6, 2023 at 12:18 | answer | added | LoztInSpace | timeline score: 0 | |
| Jul 6, 2023 at 12:15 | comment | added | Martin Maat | I would add an enum argument to the original method. The enum's nembers are Accurate, Optimal, Fast. If you want to be explicit make it AccurateAndSlow, Pragmatic, FastAndCrude. You can use the same nomenclature in the names of your specialized private methods. | |
| Jul 6, 2023 at 12:12 | answer | added | user130934 | timeline score: 0 | |
| Jul 6, 2023 at 12:10 | comment | added | csstudent1418 |
@ThomasOwens batchSize is only a parameter of the faster version. In theory both would be the same with batchSize -> ∞ but that would take Terabytes of memory. So no, the implementations of the both have completely different parameters.
|
|
| Jul 6, 2023 at 11:54 | comment | added | Thomas Owens♦ |
Does your language support default arguments? If so, why can't you use default arguments to specify a default batchSize that you believe gives the best tradeoff between performance and accuracy while letting callers choose to override it for improved accuracy or improved performance at the expense of the other?
|
|
| Jul 6, 2023 at 11:29 | comment | added | csstudent1418 | @Ewan The accuracy sadly isn't a direct parameter, it is indirectly (and unpredictably) related to a certain batchSize that is used for the approximation calculation. | |
| Jul 6, 2023 at 11:28 | comment | added | csstudent1418 |
@Jasmijn I like to have multiple functions that do the same thing start the same way, which is why I avoid prefixing variations of them. That said, approximateResult(data) has the issue I mentioned which is that it doesn't indicate the drastic speedup. You may take calculatePairwiseRelation(data) as an abstract proxy for the function name if it helps.
|
|
| Jul 6, 2023 at 11:16 | answer | added | Ewan | timeline score: 1 | |
| Jul 6, 2023 at 11:12 | comment | added | Ewan | First Approximation, Estimate, CalculateResult(enum Accuracy) ? | |
| Jul 6, 2023 at 11:05 | comment | added | Jasmijn |
approximateResult(data)? Answering this question probably needs a specific function name.
|
|
| Jul 6, 2023 at 10:51 | history | edited | csstudent1418 | CC BY-SA 4.0 |
edited title
|
| Jul 6, 2023 at 10:31 | review | Close votes | |||
| Jul 12, 2023 at 3:07 | |||||
| Jul 6, 2023 at 10:14 | history | asked | csstudent1418 | CC BY-SA 4.0 |