Michael G Schwern wrote:
>
> The main thing I wish to accomplish here is to change the prevailing
> attitude from "write an RFC and maybe something will come of it" to
> "write an RFC and make sure something comes of it." Move the ball
> down the field.
Eminently reasonable.
> I wish to make the default "put up or shut up", but there will be
> outs. You just have to give a good reason why you can't implement a
> prototype.
>
> Good reasons
> Its technically infeasible to write a prototype (RFC 1)
> The prototype will take an inordinate amount of time. (RFC 1)
> This RFC is an idea/manifesto style RFC (RFC 13)
>
> Bad reasons
> I do not have time.
> I do not have the tuits.
You have not addressed my (and I suspect many others') greatest concern:
I am not a p5 hacker; nor am I a p6 hacker by dint of the fact that p6 does
not yet exist. Can I write the prototype in pseudocode? Or should I be
building on Topaz? :-/
And what of prototyping syntactic features? You want everybody to hack
yacc? You know as well as I do that every bit of yacc thus created will
be non-integrable, and thus so much excelsior. Not to mention the fact
that everyone will be expending duplicative effort cobbling together
scaffolding grammar from which to hang their good bits. Not to mention
that fact that most people can't write yacc either. Is plain EBNF
acceptable?
--
John Porter
We're building the house of the future together.
