Re: planning 2.0?

Hi,

"Andy Tai" <[email protected]> writes:

> I think it is well agreed that the current tla's strong ties to "certain
> aspects" (but not all) of the POSIX environment is a weakness and to work well
> on Microsoft Windows is a desired feature for Arch 2.0 and Tom's prototype
> clearly shows that.

GNU Arch is/should be targeting the GNU system.  Therefore, portability
to other systems can arguably be an advantage, but I don't think it
should be a strong requirement.

So, if Windows-portability has an negative impact on the design, then
I'd reject it.  If it doesn't, then we can make it part of the plan.
That's roughly the "official" GNU policy [0].

For instance, it seems that `revc' could potentially work fine on
Windows because it doesn't rely on any Unix/GNU idiosyncrasy.  OTOH,
`tla' builds on the ability to use long paths, hard links, and so on,
which Windows doesn't provide.  Thus, maybe we should not worry too much
about Windows for `tla' because porting it there may require
unacceptably constraining the design.

Thanks,
Ludovic.

[0] http://www.gnu.org/prep/standards/standards.html#System-Portability