[Klug-general] Linux Sound Programming

[George Prowse]

On 21/03/2010 11:02, Graham Todd wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> On Sun, 21 Mar 2010 09:30:48 +0000
> George Prowse<george.prowse at gmail.com>  uttered these words:
>
>>> PULSE and JACK don't work well together because they're designed for
>>> different things.
>>
>> All of them, with perhaps the exception of ALSA seems like it was
>> dreampt up in someone's potting shed as an answer to a problem that
>> never should have existed.
>
> Maybe not, but they do exist, mainly because other OS insist on using
> proprietary codecs and we members of the Linux community say those
> codecs should have been free to use and incorporate into our software.
> And hardware manufacturers should make it easier for us to use their
> hardware with our operating system of choice.

I agree, and a lot of the time it's not linux's fault but the laziness 
of people like hardware manufacturers.
>
>> All these fractions of applications are killing linux, you also have
>> people deluding themselves that more is better when in reality having
>> 150 different ways of doing things just complicates matters.
>
> Its called choice, but apart from that, most distros have a standard
> piece of software for sound reproduction, so in practice most people do
> *not* encounter "150 different ways of doing things".  It only happens
> when they are trying to fix a specific problem, and then the choice
> works in their favour.

Choice is great but there needs to be certain standards (which i'll add 
upon in the next reply).
>
>> Look at all the successful standards, everything from network naming
>> to HTML, and it is all governed by one standard which it worked out
>> by those most qualified to do it. If people want to add extra stuff
>> to their work that isn't in the standard then they are free to do
>> that but there is always the base.
>
> The problem is that standards - those adopted by the organisations of
> experts - are not always accepted in modern use as "standards".  The
> HTML which Microsoft is always trying to foist on us is not the same in
> every case as the HTML standard set down by the W3C, yet people regard
> it as standard; the document format adopted by the International
> Standards Organisation (the OpenDocument format) is virtually unknown
> in the world of Windows, where the .doc format is supposed to be the
> standard (even though it seems to change with every edition of Word).
> These are *not* standards merely because they are used by the majority
> of software users.

I whole heartedly agree. I feel the HTML was a good example though 
because you had the standards that were set down and then people like 
Microsoft adding their own bits around the side - the browser makers 
could then choose to support these proprietary standards or not. But if 
you wanted to write a browser that would work on any site then you can 
just support the basic functions.
>
> If it was as simple as having a standard on which everyone agreed and
> would be able to be supported by virtually all software, there'd be no
> problem.  However, we in the Linux community are coming up against a
> restriction of access to the source code and the lack of adoption of
> pure standards and these militate against the adoption of something
> which is in general use, being automatically adopted for use within the
> free software community.

I'm in complete agreement again. The linux community starts off 
immediately at a disadvantage. Even when people do come together there 
is no guarantee as well, a good example is Google and Apple disagreeing 
over HTML5
>
>> Heh, base. That could so easily lead my rant round to the LSB. And
>> abstraction. What is the obsession every coder seems to have with
>> abstraction? If you want smaller binaries DON'T USE GCC!!!
>
> Seriously, I'd like to hear your rant on the LSB (assuming by that you
> mean the Linux Standards Base and not the London and Scottish Bank), as
> its avowed aim is:
>
>> The goal of the LSB is to develop and promote a set of open standards
>> that will increase compatibility among Linux distributions and enable
>> software applications to run on any compliant system even in binary
>> form. In addition, the LSB will help coordinate efforts to recruit
>> software vendors to port and write products for Linux Operating
>> System.
>
> Now that seems pretty sensible to me, so I wonder what your objections
> could be....  The same goes for the GCC. In a world where we don't have
> accepted standards between all operating systems, the best we can hope
> for is agreed standards between open operating systems.
>
Well what I dislike about the London and Scottish bank is... :)

No, the problem with LSB is that people haven't fully implemented it. 
Even the FHS - which would have been the easiest for all to implement - 
still isn't implemented in a wide variety of distros. I remember at the 
time there was a lot of unhappiness about it because of a supposed bias 
in favour of Red Hat - and the inclusion of RPM was like a shot through 
the foot.