11/03/2005 10:34:07, Nicholas Clark <[EMAIL PROTECTED]> wrote: >On Thu, Mar 10, 2005 at 11:26:25PM +0000, Nigel Sandever wrote: >> On Thu, 10 Mar 2005 14:58:58 -0800, [EMAIL PROTECTED] (Michael G Schwern) wrote: > >> Just the acknowledgment that new features are possible is a good thing I think. > >New features are possible. I'd quite like to be able to bind in Perl 5.
Perhaps it would be possible to consider some mechanism that would allow new features without imposing them on those that didn't want them, or any incompatibilities they might cause? The low-priority keywords seems to be one, though I'll admit to not fully understanding that yet. Perhaps a command line switch that enables extended keywords. Or a pseudo-package that when c<use>d, enables them. Or maybe fixing those aspects of the syntax and semantics that prevent things like this being adaquately achieved at the perl level (fixing prototypes etc.) *is* possible? I assumed that the reason this has persisted this long was because it wasn't possible to fix, but ... [...] >It's quite likely that most users running perl interpreters don't even know >how to write Perl. > If they have a stable system, why upgrade? If they don't, and they need fixes, they will need to employ someone to find the bugs that need fixing and test that the upgrade fixes them? > >> If Perl 5 has it in use first, with extra twists over the current Perl 6 vision, >> It will be their choice whether to break with the P5 implementation. > >Since "they" are we, compatibility is usually seen as preferable over >schizophrenia. Perl 5 maintainers are still making releases under licence >from Larry, and as Larry is also in charge of Perl 6, we tend to view >keeping Perl 5 consistent with Perl 6 as a good thing. > I wasn't suggesting anything else. Simply, that if during P5 development track use, it was discovered that people preferred 'say' to only append "\n" if it was not already attached--and this was not how P6 currently viewed it--P6 might adapt to the p5 findings. Likewise if the finding was that 'say' should respect $\, rather than using "\n", that information could be fed back to the p6 team for their consideration. I've coded it all 3 ways. I think that the second is correct, but I'll see how it goes in use. As P6 has an effective clean slate, they could choose, as they already are on many other things, to break with consistancy for the higher good. Which removes some of the onus from a P5 implementation to consider the whole of the future. > >Nicholas Clark > In the end, *I* would like 'say', whether my simple implementation or something much cleverer doesn't matter to me. It's the end that I want, the route is doesn't matter. I've been trying to avoid an extended wrangle over it, as it not so important, even to me, and less so to others. I posted the patch to "show willing", and asked the question, so gauge the response. I was a little taken aback by the "no new keywords" which feels completey wrong to me... but not enough to pursue the argument beyind stating that I think it is wrong. njs.
