compatibility and the OSD

Rick Moen rick at
Thu Sep 23 18:29:49 UTC 2004

Quoting Alex Rousskov (rousskov at

> The amount of RAM required to add a "would have produced" 
> functionality may be prohibitively large to reuse pieces of AAL code 
> in embedded software (think nanonetworks, for example).

Then, some other way to show a good-faith attempt to be as close to
compliance with the literal licence terms as the AAL language allows.
Thus my point:  The law doesn't seize up and throw an exception just 
because some literal-minded computerist spots a possible difficulty.

> Is there a process for de-certifying an OSI-certified license. Can an 
> OSI-certified license lose its certification?

The Board could so vote.

But I see no reason in this case.  _Lots_ of the licences on the
OSI-approved list are really rather stupidly conceived and drafted,
such that nobody with half a brain would use them.  OSI doesn't state
that they're meritorious in any way, just that they've been found to
minimally comply with the OSD.

> Otherwise, I am sure these long arguments will re-appear, 
> especially as new technologies make it more difficult to stay 
> technology neutral.

I see enough room in the AAL to be technology neutral, if barely -- and
consider it not worth spending time on, in any event, as nobody with a
grain of sense is going to use it.

More information about the License-discuss mailing list