Re: [Jack-Devel] JACK 1.9.10 to test (for 64/32 bits compatibility)

PrevNext  Index
DateMon, 07 Jan 2013 14:57:27 +0000
From John Emmas <[hidden] at tiscali dot co dot uk>
ToJACK <[hidden] at lists dot jackaudio dot org>
In-Reply-ToPaul Davis Re: [Jack-Devel] JACK 1.9.10 to test (for 64/32 bits compatibility)
Follow-UpPaul Davis Re: [Jack-Devel] JACK 1.9.10 to test (for 64/32 bits compatibility)
On 7 Jan 2013, at 14:35, Paul Davis wrote:

> 
> server+compiler "foo" interacting with libjack+compiler "bar" is not something we should even be thinking about.
> 
> we've spent years telling users that (a) you cannot have two versions of JACK installed (b) JACK is a system comprised of a server and a library.
> 
> I can see no reason to put any effort whatsoever into supporting a mixed-compiler build of the server and library. it is just insane. 

As a Mixbus developer, marketing a program that we cannot debug is even more insane.

To make Mixbus work on Windows I've needed to build many dozens of 3rd party libraries.  With the exception of pthread and zlib (which we don't yet build from source) we can debug every one of them ourselves.  In fact I can say confidently that if we hadn't been able to debug those libraries Mixbus would never have seen the light of day on Windows.

Stephane has been enormously helpful but even so, there have been some occasions when his time was very limited (or maybe he wasn't able to reproduce the problem) and it only got solved because I had an alternative means of debugging it.

And the same applies to practically every other library we've used -  sigc++ / lbsndfile / gtk / glib / pango et al.

The ability for us to do our own debugging can't be valued highly enough.  I'd certainly be loathe to give that up now, after it's served us so well.

John
PrevNext  Index

1357570657.25119_0.ltw:2,a <EAFBC4D1-B64C-49D0-8131-5BF7843CC913 at tiscali dot co dot uk>