Nicolas Williams wrote:
> On Tue, Mar 03, 2009 at 08:13:00AM -0800, Garrett D'Amore wrote:
>
>> I don't know about savings of effort, but if we're not making them
>> public interfaces, then I prefer them in /usr/lib/parted, where folks
>> are less likely to find them by "accident" and infer (possibly false)
>> things about their suitability for public use.
>>
>
> I think it takes more effort to argue over whether to make them public
> or not than it takes to just make them public. If the i-team doesn't
> want to support the bloody things then they should just be made
> Volatile -- they'd still be useful, so why hide them?
>
>
>> That said, I'd hope that if we ever shipped the public ones, that this
>> project could be converted to use the public ones instead of keeping its
>> own private copies.
>>
>
> More work for the next i-team. Why is that good?
>
> If every bit of FOSS was integrated like this we'd end up with a
> mish-mash of stuff in /usr/lib that third parties would *still* find
> *and* use -- that sounds like a negative architectural externality of
> letting this case through as-is.
>
> IMO, if integrating FOSS X requires integrating FOSS Y, then integrate
> both -- don't hide FOSS Y.
>
I agree that it would have been better to integrate a complete copy of
ntfsprogs, make it public, and probably run it as a separate case that
is a dependency upon this one.
That's not what the project team decided to do.
The project team hasn't committed to supporting ntfsprogs for other
uses. While it would have been nice of them to do so, they didn't, and
I don't think its PSARC's role to require them to do so either.
-- Garrett
