[darcs-devel] darcs-3 compatibility

Ganesh Sittampalam ganesh at earth.li
Sat Apr 3 15:52:24 UTC 2021


Hi,

On 29/03/2021 14:49, Ben Franksen wrote:

> What you describe here as desirable is, I think, quite typical of the
> user POV. I readily admit that my idea of restricting support for older
> patch formats benefits mostly us as developers, as it means we (that is,
> mostly, I) could forge ahead with repository changes, mostly
> disregarding compatibility issues.
[...]
> So far so good. Regardless of whether we aim for full compatibility or
> read-only compatibility, these are the downsides:
> 
> (1) We have to continue to maintain (and invoke) large amounts of legacy
>     code during the transition period.

It's a good point to split this up by impact on users versus developers.

One of the reservations I have about a "big bang" transition is how do
we know that the target state really is a good end state, and that we
won't discover more radical changes we need to make in future?

The other big reservation is that this is a bit akin to a "big rewrite",
albeit keeping significant bits of the old codebase. Even though a
gradual transition is expensive in some ways, managing and reviewing big
changes is also quite expensive.

> (2) Users may never upgrade their repos to darcs-3 because they don't
>     see enough practical benefit to outweigh the (perceived or real)
>     risk.

The alternative downside is that users may never upgrade to the new
version of darcs, or this may be a trigger for them to switch away
completely.

> (3) It is difficult to systematically test against regressions, since we
>     cannot currently invoke a legacy darcs version from the test suite.

Wouldn't we have the same problem with testing the conversion code?

Cheers,

Ganesh


More information about the darcs-devel mailing list