[r6rs-discuss] R5RS is not a baseline

[Marc Feeley]

On 20-Feb-09, at 6:28 PM, John Cowan wrote:

> Marijn Schouten (hkBst) scripsit:
>
>> Only if you require a supermajority for every change. If a  
>> supermajority
>> of 80% is needed to approve a new feature, only a 20% micromajority
>> should be needed to get rid of a feature.
>
> I think hygienic macros and multiple values would rapidly disappear,
> and we'd be back at R2RS quick enough.
>
> If you want R2RS, you know where to find it.

Actually where would you find a R2RS Scheme implementation these  
days?  ;-)

I'm with Marijn on this one.  It really should be that 80% of people  
must support a feature for it to be in the core spec.  For libraries  
the threshold is 0% because libraries are not imposed on users... the  
users can choose whichever suits their needs.  Anyway, that's the  
model I have in mind.

Marc