 |
lua-l archive |
|
|
||
|
On 11/21/2010 8:17 PM, Lorenzo Donati wrote:
KHMan wrote:[snip][snip]...[snip] and one becomes attuned to quickly scanning source code patterns in that way. [snip]...Since bit32 is a new library, there is no legacy burden (beside possible clashes with Mike Pall's bitlib), why adding noise and inconvenience?
Yeah, I think there is no need for two sets of names when the library names no longer clash. One will work just fine -- both are bit manipulation libraries after all and the old names have been in use for years. If people gets sloppy and run into say the shift difference, then they need to be pointed to the documentation.
Well, I can see that Luiz has cleverly deflected all arguments by remarking on the alias thing (which I had anticipated in advance and so I did earlier say specifically I had no problem with aliasing.)
See the SHA1 example that was linked earlier. All the aliased locals in all caps like ROR etc. Was that what they were thinking? Would someone be typing that every day? If I were writing Lua embedded code (where processor header files have hundreds of constants in all caps), I wouldn't do that.
Perhaps two sides are just talking past each other. Fine, as long as nobody shouts the other side down without reasoned arguments.
In journalism there is this thing called the "smell test". When intuition says it "smells", the former is usually based on experience, and so there are valid viewpoints being discussed -- to dismiss everything out of hand is poor debating. Let views be heard; we can disagree.
-- Cheers, Kein-Hong Man (esq.) Kuala Lumpur, Malaysia