jastrachan | 2 Apr 09:04 2004

Re: Re: [groovy-user] Language Spec

On 1 Apr 2004, at 22:58, Cedric Beust wrote:
>  Please allow me to share a few thoughts, motivated by my 
> participation in JSR 175 and JSR 201, both of them heavily 
> language-focused.
>  Neil Ellis wrote:
>  I like to think we have provided food for thought for the JSR team – 
> but I’d also like to re-iterate, for myself at least, confidence in 
> those making decisions on our behalf. As has been said many times by 
> people on the list we do not want a language designed by committee – 
> we just want to feed some worthwhile ideas to the man in charge.
>  Well, I am afraid it's not going to be as simple as that.
>  At some point, a spec needs to be written and it takes some very 
> particular skills to be able to do that.  I know I can't, and I 
> realized over these past years that actually, very few people can do 
> that.
>  A JSR is a work of compromise and the epitomy of design by 
> committee.  Sorry to break it out to you, but it's going to involve 
> the participation of a lot of people with different views and 
> compromises that can, and will, sometimes conflict with James' 
> original vision.
>  What I am getting at is this:  the best way to guarantee the success 
> of this JSR is to pair up James with someone who is good at writing 
> specs and who will be able to bring some realism to the debates, aside 
> from any personal or aesthetic considerations.  Josh is absolutely 
> brilliant at that, but even he defers the final writing of the 
> language spec to Gilad (yes, writing a language spec is that hard).
>  Please do not take this as a critique of James, I am absolutely in 
> love with Groovy and James and I agree on so many points on a 
> technical level that I suspect he might be my lost brother from over 
> the Pond :-)


>  But if we want to be serious with this JSR and not become like 
> countless other JSR's that have been in limbo for years, we need to 
> get ready to face compromises.