asked on

VB Programmers: Debate/Discussion topic offered

https://www.experts-exchange.com/questions/20543705/Good-vs-Streamline-when-not-same-DEBATE-wanted.html

I've posted a request for some debate/discussion regarding good coding practises and streamline coding practises, and what happens when they may be in conflict.

Evaluation of a scenario I postulated is requested, as is a "I would do this because.." or "I would do that IF.." input. I'm not looking for a right/wrong answer to a specific problem, I just want to get a feel for the general community opinion, and help create a guideline/ruleset of conditions for that "would do IF".

ASKER CERTIFIED SOLUTION

aikimark

membership

This solution is only available to members.

To access this solution, you must be a member of Experts Exchange.

Start Free Trial

GPrentice00

ASKER

tweaking still, but my intent was to say "if we extrapolate from this point to a point where perception is slowed, then what would be done at that point to speed it up, and is not valid to design that at the smaller scale?"

Some things work on a pure linear scale to infinity, others have a switchpoint. A sort routine for example, some sort methods work excellent on <n items, but function horribly for n+1. The method that just flies through n+1 items spends a rediculous amount of wasted operations on a set of n items. Sort method A may use Na operations for some number of items, method B may use Nb operations. For small values, Na may be quite small and Nb quite large, then suddenly Nb becomes smaller than Na.

--
Just because if I expect to be dealing with a bunch of values most likely a bit larger than the switchpoint, where the number of operations is similar for each, should I use the simpler approach method A with a few more operations, or should I do the more complex method B which will have less, but is inperceptable in the end.
Tough call, I guess, but if the possibility that my set will become larger is higher than the possibility of my set becoming smaller, its not a hard call.

But for a more linear process, that doesn't really have a switchpoint, extrapolation and interpolation work both ways equally, and the question then becomes at what point do we toss out the window the best solution for an extrapolated extreme case, since that same soltuion can be interpolated to the simplist case.

And then of course, there were the answers that you did supply that filled in part of that gap.

--
Unfortunately, it wasn't as much a problem/answer as it was a theoretical discussion intention.