[Haskell-cafe] I'm afraid of OverloadedLabels.

Richard A. O'Keefe ok at cs.otago.ac.nz
Mon Jul 10 03:45:23 UTC 2017



On 10/07/17 3:19 PM, Anthony Clayden wrote:
>> As a native speaker of an SVO language (English)
>> in a country where one of the official languages is VSO
> (Māori),
>> I sometimes find the OVS structure of
>> "mainstream OO languages" a little uncomfortable...
>> "Backward ran sentences until reeled the mind."
>
> lol. That's a cute comparison.
>
> I don't think "mainstream OO" is OVS. In:
>
>> ... button.colour <- red

This is normally written
    button.setColour(red)
    object verb      something

Maybe it's OV without a subject,
    object vp[+imperative]

>
> Gloss: 'button's colour goes red'.
>
> `button's colour` is the grammatical Subject. (Yes it
> represents an
> OOP 'Object.method', but that's because everything in OOP
> is an 'Object'; and the only way to access is by methods.)

Yes, but there is no law in OOP that the object has to be
first.  In Ada, I could perfectly well write
    Set_Colour(Red, Some_Object);
and have it dispatch on the 2nd argument.
In Common Lisp, I could perfectly well write
    (set-colour 'red some-object)
and have it dispatch on the 2nd argument (or indeed, have
it dispatch on more than one argument).

The object.selector(arguments) syntax is a generalisation
of record.field syntax, and there's no law requiring that
either.  Algol 68 used  field OF object.

OOP really isn't about the *syntax*, any more than FP is.
That's only a human usability issue.  (Only.  Heh.)

> This dot-suffixing is already in Haskell:
> Data.ByteString.Lazy.Builder.Extras
> 'Data's ByteString's Lazy's Builder's Extras'.

That is "path" syntax, which again older languages
would have written
   Extras in Builder in Lazy in ByteString in Data
and none of the components resembles a verb, in the way
that method selectors do.
>
> But the outcry against TDNR's proposal for dot
> really was vociferous.

Because the dot already has too many uses.

There are at least three forms of overloading:
- ad hoc, as found in PL/I, Fortran, Algol 68, Ada, and Java.
- parametric, as found in ML, Clean, F#, Haskell, and others.
- type-based, as found in Eiffel and Haskell, say.
Fitting all three together is tricky.



More information about the Haskell-Cafe mailing list