[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: Various things



> From: Felix Winkelmann <felix@xxxxxxxxxxxxx>
 
> The reference implementation allows escape sequences in lower- or uppercase, 
> but it's not  specified whether uppercase escape sequences are 
> case-insensitive.

That is an oversight on my part.  The escapes are case-INsensitive.  I will 
update the text/html.


> What's the point of ~_ ?

Compatability with T.  As with ~K for ~?.  

I am not particularly wedded to this one.  Hmmm.. in looking at the T manual, 
the specification is to either output a space pr a newline.

I am not particularly wedded to this one.  Perhaps we could drop it.

Anyone care to argue to keep "~_" ?

 
> Why not go the full way and use Dirk Lutzebaecks full CL format
>  (as can be found in SLIB)? It's way more powerful and just as portable.

Historically, it has been difficult to gain complete concensus within the 
scheme community (or at least the scheme authors' group 8^) for anything more 
than minimalist solutions.  One argument which can be raised against a more 
complete solution (i.e. SLIB) is that an implementation running on a 
microprocessor or an experimental, academic, implementation should be able to 
be relatively small but still be called "Scheme".  

The SRFI process is one response to this problem.  It mandates nothing but 
engenders common interfaces and functionality.

I personally consider SRFI-28 to be too minimalist.  I have encouraged someone 
(you?) to propose an "advanced format" which could be the interface from 
SLIB.  The version here is meant to be "a higher baseline" and something 
which is small enough to be used in a microprocessor/microcontroller 
implementation (e.g. BIT or Scheme48) but useful enough to be universal.
 
 
Cheers,
-KenD