State of Pidgin: Reverse Engineering FUD

[Gary Kramlich]

On Tue, Oct 3, 2017 at 10:22 AM, Ethan Blanton <elb at pidgin.im> wrote:
> Gary Kramlich wrote:
>> We've skirted around these concerns in the past, but I would like to
>> honestly put them behind us.  What I am proposing is theoretically simple,
>> but a kind of a logistical nightmare.  What I would like to see is
>> libpurple ship with just open standard protocols in tree.  That is IRC,
>> SILC, XMPP, Matrix.org and maybe Zephyr which is simple enough.  We would
>> then put all of the proprietary protocols into their own repositories which
>> creates the logistical nightmare.
>
> I have no problem with removing prpls that are a technical problem.
> In general, though, I'd rather see moving some of the popular third
> party prpls (particularly the ones maintained by regular contributors
> or developers) moved in-house.
>
> Splitting things out just means *more* stuff to keep track of.  We
> already have trouble keeping track of the things we have.
>
> The one *valid* reason I see for breaking out proprietary prpls is the
> ability to fix them when the provider changes something and breaks
> things.  This is historically difficult for us because it involves
> pushing an entirely new release.  However, it also doesn't happen
> particularly often.

Yeah I'm very conflicted about this one myself.  I want people to
install Pidgin, Finch, etc and have a decent base right away.  But I
also don't want to be pushing a new version of libpurple everytime
new-yahoo breaks.  Obviously this wouldn't go in tree because of those
issues, but it just convolutes everything.

> Ethan

Thanks,

--
Gary Kramlich <grim at reaperworld.com>