[Ardour-Users] subscription support down, your ideas sought

[Ross Johnson]

John Rigg wrote:
> On Thu, May 15, 2008 at 06:01:47PM +0200, Ralf Mardorf wrote:
>   
>> that sounds fine but I don't think anything will change for you without
>> becoming proprietary. I'm working for free, too and I need state benefit
>> for survival. I'm 41 and my experience tells me the only solution seems
>> to become proprietary. Sometimes we have to make hard decisions.
>>     
>
> Unfortunately working just for fun doesn't pay the rent or food bills.
> I do sound engineering because it's fun. I charge money for it because
> I need to eat. I see no problem at all with free software developers
> accepting money for their work - how can they devote enough time to
> develop a serious project if they have to do something else to pay the
> bills? 
>   
That's the point of the open source model - to spread the work across as 
many hands and minds as possible so that each can contribute a little 
without having to earn a living from it. The problem then is to attract 
enough hands and minds.
> Taking your argument to its logical conclusion, much of the
> Linux kernel would be proprietary because there are developers being
> paid to work on it.
>   
Those companies paying developers to work on the Linux kernel are 
earning income from related products and services and recognising that 
cooperation in the development of the kernel itself is better than 
competition. The difficulty for Ardour is that it isn't commodity 
software like Linux is - it's an end-product rather than a platform or 
catalyst for things that people are willing to pay for.

AFAICS the problem for the Ardour project is how to attract more hands 
and minds to a project that has a fairly narrow user-base (and even 
smaller potential developer base).

Based on my view of the problem I have a couple of [possibly quite 
fanciful] ideas to add, that aren't light work but might be useful to 
think about longer term:-

1) Would it be feasible to create an online journal dedicated to Ardour 
developers, as distinct from Ardour users? It could provide a focus to 
clearly explain the internals of Ardour, Jack, plugins, file formats, 
standards, specifications, exposure of future features, etc. in 
technical language. If it was done well it may help to encourage 
corporate sponsors to support Ardour, especially vendors of audio devices.

For anyone first getting into Ardour development, writing a technical 
article is often a good way to reinforce the discovery process and, 
provided the articles are reviewed, corrected and edited by more 
seasoned developers, would be publishable while giving great feedback 
and encouragement.

In my own experience with large projects, anything that makes it easier 
to start contributing code and ideas to the project would help. I can 
program in several different languages but I find that it requires a 
very large investment in time and effort up front before I start to make 
even small inroads into understanding the framework and APIs, etc. in 
any large project. That is most likely because I'm getting a little old 
but it is a barrier - I have an interest in using Ardour, Linux, and can 
program, but I haven't been able to contribute even though I'd very much 
like to.

I should also explain that my motivation for looking at the code is in 
solving problems that I experience directly, which I would then be more 
than happy to feed back for other users to benefit from. This is quite 
different to wanting to code for fun.

Should I just subscribe? First of all, that doesn't solve the problem 
that I want to fix now (to get my real job done), and secondly, I'd 
prefer to see the FOSS model working the way it's supposed to, rather 
than pay money and see the contributor pool shrink to just a few 
developers paid by subscriptions or remunerated in other ways, such as 
social benefits etc. Since the subscription model was introduced I just 
haven't been convinced that it's sustainable, and I'm very relieved to 
hear that Paul doesn't need the money on a personal level.

2) Paul has told us that there has been corporate interest in the past, 
but I assume they all wanted some kind of exclusivity. Perhaps Ardour 
could be split into a general purpose DAW platform framework plus user 
level layers.

Jack already provides a very low level general framework and Ardour 
provides a user-level application on top of that technology, but I think 
that splitting the stack into additional clearly-defined layers may 
provide more opportunities and encourage corporate involvement to value 
add, integrate, and possibly even provide a vehicle for new start-ups. 
Deconstructing the Linux success, the trick may be to provide a common 
technology base that they all have an interest in supporting while still 
enabling differentiation or competition at the value add or service level.

Other projects that this model appears to work for are Apache, Eclipse 
and OpenOffice, etc.

Ross Johnson