[plt-scheme] why should black holes be exceptions, rather than divergence?

Perhaps the right conclusion here is that you should not talk to
theorists too much about what behavior is desirable when actually
programming.

Robby

On Mon, May 31, 2010 at 8:59 AM, Keiko Nakata
<keiko at kurims.kyoto-u.ac.jp> wrote:
>> On Mon, May 31, 2010 at 8:47 AM, Keiko Nakata
>> <keiko at kurims.kyoto-u.ac.jp> wrote:
>> > From: Robby Findler <robby at eecs.northwestern.edu>
>> >
>> >> Errors seem to always be preferable to divergence to me.
>> >
>> > Personally I agree.
>> > But I think this is a strong commitment,
>> > since I can catch errors but I cannot catch divergence!
>>
>> I think you're suggesting that the inability to catch divergence can,
>> in some contexts, be a good thing? Is that right?
>
> Just that I do not have a definite clue to argue for my preference.
>
> When I talk about Scheme's behavior to theorists,
> they tend to be a bit surprised.
>
> It is not that they dislike it, but they often said to me that they have never
> thought of it this way.
>
> Keiko
>
>