Trivia: Algol 68R and colons in conformity clauses

[Jose E. Marchesi]

> Jose and everyone,
>
> On Fri, May 8, 2026 at 3:54 PM Jose E. Marchesi <jemarch@gnu.org> wrote:
>
>>
>> I have often wondered why I kept finding Algol 68 snippets out there
>> reading like:
>>
>>   case foo in (int) bleh, (real) blah esac
>>
>> rather than
>>
>>   case foo in (int): bleh, (real): blah esac
>>
>>
>> An example is the ELLA 2000 sources I am porting to GNU Algol 68.
>>
>> Today I learned, it is because the Algol 68R dialect didn't use colons
>> after the specifications in conformity clauses.
>>
>> They fixed it in Algol 68RS, which implemented the Revised language more
>> accurately.
>>
>> Just the trivia of the day ;)
>>
>
> Somewhere in the house I have a hardcopy of the non-revised report; I need
> to find it and refresh my memory as to what crept in (besides the closed
> syntax).

Note that omitting the colons in conformity clause specs appeared in the
Algol 68R already adapted to the Revised language.

The unrevised language didn't feature conformity clauses.  Instead, you
used these clauses whose name I can't recall now, with a representation
involving ::=, in which you basically forced an united value as one of
its moods in "free form".