Hello Florian

> On 14.10.2017 11:59, Peter Waher wrote:
> > Hello
> >
> > A year and a half ago I proposed a XEP: "Content Types in Messages" [1],
> > solving the issue of describing and annotating what type of content is
> > sent in messages. At the time, many objected, since they did not see the
> > value of this annotation.
> >
> > Now, the interest seems to have been awakened again, with a new
> > proposal: "Body Markup Hints" [2]. Both seem to be based around
> > markdown, so there's an obvious common interest and common ground.
> >
> > What I wonder is: Why don't we collaborate on this? There are already
> > implementations of [1]. It solves the issue using a paradigm that
> > matches what is elsewhere used on the internet (Content Types). Why
> > invent something new, and not renew the application of the original
> > proposal? And if there's something missing, the proposal can be updated,
> > and the author list increased.
>
> Thanks for your offer Peter.
>
> Those two ProtoXEPs are indeed very similar. I have a few minor issues
> with yours, probably nothing which can't be fixed. The following list is
> not comprehensive and in no particular order:
>
> - I wouldn't define textual content for <content/>. If there is a demand
> for having different markup flavours in the same message, then I think
> something like
>
> <additional-markup-body>
>   <body>?</body>
>   <body-markup-hint/>
> </additional-markup-body>
>
> would be better. Note that BMH does not deliver that use case, simply
> because I saw no use. But that could change.

There's definitely a use, especially if you don't know the capabilities of the 
receiver, or if there are multiple receivers with different capabilities. It is 
also how XHTML-IM works: They define a method where you send a plain text 
message in the body, and a formatted message in the corresponding tag. The 
content proposal even allows for multiple formats, not only two.

> - It is not really specified where the values for the 'type' attribute
> are defined. It should probably be IANA's Media Types registry
> (https://www.iana.org/assignments/media-types/media-types.xhtml). And
> from this registry only the 'text/*' part. But I only spot text/markdown
> in https://www.iana.org/assignments/media-types/media-types.xhtml#text.
> Since not all markup languages are defined here, I choose to go with
> custom registry in BMH instead.

Internet Content Types are assumed. If unclear, it could be explicitly stated 
For Markdown, you have RFC 7763:
https://tools.ietf.org/html/rfc7763

You can use your own content types as well, as long as you prefix your subtype 
with vnd or x.

> - There is no discovery mechanism for individual markup flavours in your
> ProtoXEP. Discovery is not really a goal for BMH, it is just a nice
> goodie that I would take along. But still, it would be nice to get same
> data from the network, which markup flavours are supported by the clients.

No, but it can easily be added.

> I'm happy to collaborate. In the end, the XSF council should decide if
> any of the ProtoXEPs is worth being accepted, and then we can continue
> to work on them. :)
>
> - Florian

Great.

How about the idea of going through a layout XML, instead of sending the actual 
markdown. And then let each client transform between markdown to XML and back 
to markdown. That would allow different clients to maintain their flavours and 
still be able to interchange formatted messages?

Best regards,
Peter
_______________________________________________
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: [email protected]
_______________________________________________

Reply via email to