Front page | perl.perl5.porters |
Postings from December 2013
Christian Walde wrote: >Zefram in #p5p failed to show me any other possible reason, since he >himself tentatively agreed that the arguments boil down to >"tradition". That's a misstatement of the position I was expressing. I argued that the difficulty in understanding XS-generated code comes from the concepts, not from the cryptic variable names. I also argued that, due to the stereotyped use of the XS-generated variables, it is appropriate for them to have short names. These issues do not boil down to tradition, and I did not agree to any statement that they did. Where tradition comes in is that, as these concepts appear in every xsub ever, there is a network effect in favour of using the same names for them everywhere. There is value in using the traditional names, as Steffen does in his article. This effect of tradition is distinct from my argument in favour of short names per se. Tradition does not underly the psychological factors that make short names appropriate. More the reverse: the psychological factors are part of what caused this tradition to arise in the first place. -zeframThread Previous | Thread Next