The Perfect Argument

(Moved from ThePerfectPage)

I've been looking for the "perfect study" as a presentation pattern reference work. People often argue that language/tool/technique X is better than Y, and I'd like to see "prove it!" done well, wiki or elsewhere.

Let's say we found a paper that was brilliantly written that proved clearly to everybody that blocks are objectively better than goto's. We could make that paper our reference pattern, and then re-work other topics with the same style and evidence presentation techniques.

Related: DenyingObjectiveEvidence where starts the long debate about whether blocks can be proven "objectively better" than GOTO's. --top

Almost all useful studies or debates on software organization will be somehow be tied to WetWare. (This excludes machine performance issues.) I doubt there is such a non-wetware study out there beyond some trivial aspect. Everybody wants software engineering to be an objective science, but it's not, and this is why such debates or studies are hard to come by. But many stubborn people refuse to realize "the wetware limit" and give crappy advice because of it, making up goofy "principles" in the process. (It may be objective from the perspective of given assumptions about WetWare models, but we are hardly even to that point yet.) --top

That's actually a fairly profound observation, but I find it ignores a significant point: the term "WetWare", while suggesting something personal to the individual, does not include the fact that such things exist in a broad context of a more-or-less shared history which tie or anchor it to something more than simply personal. This is why you don't see long strings of random characters or words, for example as an actual wiki entry: all language is tied to shared history and English is perhaps the most of all of them. As such, we can approach something of an "objective reality" because it is ultimately a "shared reality". -- MarkJanssen