On 10 févr. 05, at 16:49, Bert Freudenberg wrote:

>
> Am 10.02.2005 um 15:03 schrieb stéphane ducasse:
>
>> hi all
>>
>> we (the guys from berne and roel) spent a great morning discussing 
>> about the problem of self-evaluating representation. This is a bit 
>> too long to send the result of the discussion sicne this is related 
>> to the difference between meta and base programming.
>>
>> But we came up with a satisfactory fix for Array:
>> Display the most that we can when we have the information.
>> Please have a look and let us know, because we think that this is a 
>> good fix, cheap and
>> (if you remember I was against {} but with this fix in fact I changed 
>> my mind after the discussions and I
>> prefer them compared to #(). But {} printing should be fixed with 
>> these changes. )
>
> Neat :)
>
> Two things I spotted:
>
> * I'd rather call "super printOn:" in Array if it's neither literal 
> nor self-evaluating
> * printAsSelfEvaluatingFormOn: would look nicer using do:separatedBy:

Indeed shame on us :)
But I like that you like it because I think that this is best 
compromise we can get to offer
self-evaluation when we can.

Stef

Alex can you fix that and resubmit the changeset

>
> And, could any native speaker confirm "self-evaluating" is even the 
> correct term to use?
>
> - Bert -
>
>
>