[ANSI-Smalltalk] Next STEPs

[Martin Kobetic]

Yup, I like this approach. I'd add that we should be careful and not get carried away piling things up too high. There are many good ideas out there that are no doubt handy in many circumstances, but not all of them should go into the standard. It's important that in the end the standard as a whole is powerful, but coherent and that it sticks to the principle of a minimal number of simple but powerful building elements. So let's consider as many suggestions as we can, but let's be very selective in what we actually put in, considering each one in the context of the whole set.

If we don't manage to keep the standard "lean and mean", few will be willing to plow through hundreds of pages of it, let alone bother complying with it. Especially if we intend to stick to the existing compliance criteria, where compliant implementation MUST implement everything defined in the standard.

Martin

Bruce Badger wrote:
> Andres,
> 
> All STEPs are enhancements, it's in the name: Smalltalk Enhancement
> Process, so the collection of all STEPs will be the collection of
> proposed enhancements past present and future.  At least this was my
> understanding.
> 
> The example you give of: SequenceableCollection>>beginsWith:
> aSubSequence I would expect to find in a STEP about collections (or
> something like that).
> 
> Taking collections as an example, I would expect there to be a STEP
> that reflects the current ANSI specification and then another step
> that (when it completes the STEP process) makes the first one
> obsolete.  This is how RFCs work and it seems to have been a good
> model for them.  I particularly like the fact that I can see the
> histroy of HTTP (for example) by following the chain of non obsolete
> RFCs.
> 
> How would this fit with what you have in mind?
> 
> All the best,
>     Bruce
> 
> On 06/03/2008, Andres Valloud <andres.valloud at gmail.com> wrote:
>> Bruce,
>>
>> I think that another STEP for enhancements to make to the ANSI standard is
>> in order.  My motivation for this is that, over time, all the dialects have
>> accumulated convenience methods which more often than not one misses when a
>> different dialect is used.  An example of this would be
>> SequenceableCollection>>beginsWith: aSubSequence.  I am
>> sure we can think of others.
>>
>> I looked at the Smalltalk syntax STEP and I have a feeling others may want
>> to take care of it.  I wouldn't mind taking over the enhancements one.
>>
>> Andres.
>>
>>
>> On Wed, Mar 5, 2008 at 9:57 PM, Bruce Badger <bwbadger at gmail.com> wrote:
>>> Paolo's STEP proposal had been out for some time not, so I think that
>>> we should perhaps start to use it for real.
>>>
>>> I's like to suggest that we start two STEPs.
>>>
>>> o The errata of the current ANSI standard.
>>>
>>> o Smalltalk syntax as defined in the current ANSI standard.
>>>
>>> John O'Keefe is currently wearing the errata crown.  Would you be up
>>> for expressing the errata as a STEP John?
>>>
>>> Does anyone want to step up and be the author of the Smalltalk syntax
>>> STEP?  I'll have a look at it if nobody else will, but I think the are
>>> some people who are much better qualified than I who hang out on this
>>> list.  Did I see a hand go up at the back?
>>>
>>> By getting these going I think we will be testing STEP 1, the STEP
>>> process itself, which will be a Good Thing.
>>>
>>> All the best,
>>>    Bruce
>>>
>>> --
>>> Make the most of your skills - with OpenSkills
>>> http://www.openskills.org/
>>>
>>> _______________________________________________
>>> ANSI-Smalltalk mailing list
>>> ANSI-Smalltalk at lists.openskills.org
>>>
>> http://lists.openskills.org/cgi-bin/mailman/listinfo/ansi-smalltalk
>>
>> _______________________________________________
>>  ANSI-Smalltalk mailing list
>>  ANSI-Smalltalk at lists.openskills.org
>> http://lists.openskills.org/cgi-bin/mailman/listinfo/ansi-smalltalk
>>
>>
> 
>