Re: On Not Merging new and old behavior ( was Re: Test-Stream + MongoDB scalar refcount issue )
[email protected] (Kent Fredric)
| Newsgroups | perl.cpan.workers |
|---|---|
| Message-ID | <CAATnKFAGszUhXbX9ysGodx7+zXnDB0jKT=7YDQ8UVDhq+SpDFA@mail.gmail.com> |
On 3 May 2015 at 10:34, Aristotle Pagaltzis <[email protected]> wrote: > Essentially the only sane thing is to declare dependencies on both of > the ecosystems. Then you can also run your tests against each of them. > There's a few things here that also need to be thought about, and that's the future of merge/split packaging. It is suggested that at one future point $SYSTEM2 and $SYSTEM1 may ship together. There are benefits to that, and there are downsides. I imagine the prime rationale for merging is not because there's a benefit in merging, but because that will get the modules in perl itself classification. Merging just to ship with perl I'd say is a bad idea. Especially if we hope to *ever* ( talking 20 years maybe ) decide $SYSTEM2 to be better than $SYSTEM1 and eventually get it out of perl. ( I'm not saying that will ever happen, its just something to think about ) Coupling what may one day be branded "legacy" with "the new system" seems like a logical regression of some kind to me today. -- Kent *KENTNL* - https://metacpan.org/author/KENTNL