Re: More functions accepting strings and symbols

[Jim Blandy]

thi <address@hidden> writes:

>    From: Dirk Herrmann <address@hidden>
>    Date: Wed, 11 Oct 2000 19:29:29 +0200 (MEST)
> 
>    Up to now this has not been a big deal, since strings and symbols were
>    represented similarly.  However, as I am currently trying to make accesses
>    to different types more explicit in the code (by for example splitting
>    SCM_LENGTH into SCM_STRING_LENGTH, SCM_SYMBOL_LENGTH and so on), it
>    becomes obvious that there are some functions which are not very explicit
>    about their inputs.  I suggest to restrict the accepted input parameter
>    types to strings in most of the cases.
> 
> nooooooo!!!!  [sounds of hair ripping out]
> 
> i can't describe how unhappy i would be if string/symbol
> interoperability were dropped.  :-(

Well, now, why don't you tell us about what you're using it for?  :)

Guile's behavior in this area isn't R4RS, and it strikes a number of
us as being odd design.  Of course, that's not the best reason to
break people's code and piss them off.  But if you want us to continue
to support this, you need to explain why it's the Right Thing.