RE: test result code was Re: [Simulavr-devel] avrtest and simulavrvssimulavrxx

[Weddington, Eric]

> -----Original Message-----
> From: 
> address@hidden 
> [mailto:address@hidden
> u.org] On Behalf Of Joerg Wunsch
> Sent: Wednesday, March 11, 2009 8:28 AM
> To: address@hidden
> Subject: Re: test result code was Re: [Simulavr-devel] 
> avrtest and simulavrvssimulavrxx
> 
> As Weddington, Eric wrote:
> 
> (Timeframe for the transition from old simulavr to simulavrxx.)
> 
> > I would bet you a good German beer that it would take less time than
> > that for the "legions" of old simulavr users to switch over. ;-)
> 
> I wonder where you'd get the good German beer from over in 
> Colorado. :)

Well, I would just have to come and visit you in Dresden then. ;-)

 
> The big question mark is simply that you'll never know for sure in an
> opensource project about how many active users might actually be out
> there.  I've sometimes be surprised by people approaching me in
> private mail, when it turned out they've been using a particular
> feature for years you'd never have guessed about, just because the
> feature has once been there.  Just remember how many years we've been
> telling people that there was really no realistic reason for crying
> after the death of the old sbi and cbi macros...

However avr-libc is used everywhere. I still say that there are not very many 
old simulavr users. Avarice users, yes.
 
> Stories like these make me a little cautious about dropping certain
> backwards compatibility when it were actually easy enough to maintain.

If it's easy enough for backwards compatibility, fine. But don't let it get in 
the way of doing good things.