[SCL] scl 1.1

[Christopher Menzel]

On Nov 15, 2003, at 4:28 PM, Bill Andersen wrote:

> Hi all.
>
> See below.
>
> On Nov 14, 2003, at 1922, Christopher Menzel wrote:
>
>>>> "Let me emphasize Yet Again that this is an internal document 
>>>> (though
>>>> of course anyone can look at it) and is intended only to get the
>>>> concepts right."
>>>
>>> I fully understand this and hence discuss the issues which I think
>>> may have principal shortcomings in this internal document.
>>
>> Great, you get the point!
>>
>>> General comments:
>>>
>>> I find this text far too complex to be usable for people who want to
>>> build any real systems based on SCL.
>>
>> Oops, I guess you don't.
>
> One thing to think about.  Tanel is right - the spec as it stands is 
> very complex and if some of us are having a hard time reading it then 
> certainly users will too.

Folks, again: this document is for US, as a focus for getting the 
language and semantics right.  That's all it's for.  Others are welcome 
to read it, but it's not for them, at least not directly.  This, of 
course, is exactly why we need (as I've emphasized repeatedly from the 
beginning) public documents written in large friendly letters.

> I would suggest they are people building reasoners, not axiom writers. 
>  Look at the current situation with DLs.  I highly doubt that 99.9% of 
> the "users" of DLs fully understand their semantics.  They just use 
> it.  How?  They grab someone's editor and someone's reasoner and off 
> they go.  The community who *builds* those tools is small and highly 
> motivated to understand the spec.  For that reason I'm not worried 
> about this issue and I don't think any change of direction is 
> warranted - it's been three years now and it's time to wrap this up.

Well, technically, it's been a lot less than that since SCL got off the 
ground.

> That said, other documents can and should be written, as Pat suggests,

and as Chris suggests. :-)

> to popularize (S)CL.

-chris