Re: Repetition syntax

I think options 1 and 3 are preferable to the other options.

I have a question about 1; how tolerable might it be considered to have alternative characters for the `<` and `>`?

So that it would be syntactically acceptable to use less-than and greater-than signs, but ALSO acceptable to use some other non-US-keyboard characters as brackets?

The base proposal would be proper syntax:

| 1 | a, b<3>, c  | a, b<3,5>, c    | a, b <<3,5>> ",", c  |

Some symmetric substitution (for those willing to type the characters and not necessarily using guillemets, those are just me picking an example) would ALSO be proper syntax, that is, both `<` and `«` would work:

| 1a | a, b«3», c  | a, b«3,5», c    | a, b ««3,5»» ",", c  |

but mixing `<` and `«` would NOT be proper syntax:

| 1b | a, b<3», c  | a, b<3,5», c    | a, b <<3,5»» ",", c  |

I'm seeing the keyboard as most of the problem; the glyphs exist, they're just not all equivalently readily typed.

-- Graydon

On Wed, Dec 3, 2025, at 11:08, Norm Tovey-Walsh wrote:
> Hello again,
> 
> I thought I’d see if I could summarize the proposals that I’ve seen so far on this thread. Here’s my list (I’ve used 3 and 5 as stand-ins for any arbitrary number):
> 
> |---+-------------+-----------------+----------------------|
> | 0 | a, b*, c    | a, b*, c,       | a, b ** ",", c       |
> |---+-------------+-----------------+----------------------|
> | 1 | a, b<3>, c  | a, b<3,5>, c    | a, b <<3,5>> ",", c  |
> | 2 | a, b#3,c    | a, b#3,5, c     | a, b ##3,5 ",", c    |
> | 3 | a, b&(3), c | a, b&(3,5), c   | a, b &&(3,5) ",", c  |
> | 4 | a, b 3 *, c | a, b 3 * 5, c   | a, b 3 ** 5 ",", c   |
> | 5 | a, b x 3, c | a, b x (3,5), c | a, b xx (3,5) ",", c |
> |---+-------------+-----------------+----------------------|
> 
> Row 0 is just the ordinary “*” operator for comparsion.
> 
> I think the proposals in rows 2 and 3 could work with most non-name punctuation characters. (Consequently, I haven’t repeated them for “@”, for example, or “%”.)
> 
> I think the proposals in rows 2 and 3 could also work with either “:” or “..” or any of a variety of other separators between the numbers in place of “,”. Again, I haven’t expanded the table. I think we can make the choices serially.
> 
> I think the proposals in rows 4 and 5 are impractical. I think they’re too confusing for users and they’re also too confusing for the grammar. Using a delimiter that might be a name character would be much harder to make work unambiguously in the iXML grammar for iXML.
> 
>                                         Be seeing you,
>                                           norm
> 
> --
> Norm Tovey-Walsh
> Saxonica
> 
>