[Consortium] re: VST licensing progress

[Chris Cannam]

On Friday 03 Nov 2006 16:39, Leonard Ritter wrote:
> what is required here to solve the problem is to simply take the vst
> headers and write a new implementation that is binary compatible. these
> headers can be used for hosts.

The difficulty is that the VST SDK licensing agreement expressly forbids this.
You could argue for hours about the scope and enforceability of its 
no-reverse-engineering clause, but I don't think it would be proper to just 
ignore it.  The header would presumably have to be reimplemented by someone 
who had not agreed to the agreement, i.e. had never obtained the SDK.

I think it would probably be fine for one developer who had agreed to the SDK 
license to implement and publish an open source (BSD or similar) VST host, as 
source and binary but omitting the VST SDK headers, that loaded an arbitrary 
plugin and printed out information about the protocol codes it was sending 
and receiving, for the purposes of debugging it with third-party plugins. 
And then it would probably be fine for another developer who had never 
downloaded the SDK to independently take that host and work out what would 
need to be filled in to make it compile and run (i.e. resynthesise the SDK 
headers based on the "spec" that is the working host).  If those two things 
independently came to pass, the result would presumably be an unencumbered 
(if partial) API.

Not that I really know what I'm talking about, especially in legal terms.


Chris