[Standards] LAST CALL: XEP-0393 (Message Styling)

goffi goffi at goffi.org
Mon May 25 06:16:45 UTC 2020


Hello,


Le 2020-05-12 21:35, Jonas Schäfer a écrit :
> This message constitutes notice of a Last Call for comments on
> XEP-0393.
> […]
> 1. Is this specification needed to fill gaps in the XMPP protocol
> stack or to clarify an existing protocol?

no and yes. It was not necessary with XHTML-IM, but now that XEP-0071 
has been deprecated, we need something for rich text formatting.

> 
> 2. Does the specification solve the problem stated in the introduction
> and requirements?

no, the requirements says:
"Messages formatted using this specification MUST NOT hinder readability 
by users with color vision deficiency or impaired vision."

while the section" Accessibility Considerations" says itself:
"Formatted text may also be rendered poorly by screen readers. When 
applying formatting it may be desirable to include directives to exclude 
formatting characters from being read."

But there is no way to know if the text is formatted or not, there is no 
marker, discovery, or anything to indicate that styling syntax is used.

> 
> 3. Do you plan to implement this specification in your code? If not,
> why not?

No, for the following reasons:

- I don't want to implement something without proper discovery and 
flagging

- I think this is a huge step back with accessibility

- I don't like to mix wire format and meaningful text, but I could 
accommodate with that if there was something to properly mark the body 
as styled

> 
> 4. Do you have any security concerns related to this specification?

yes, it is not clear enough that formatting characters must be kept, and 
some clients may choose to hide them for aesthetic reasons.
This, added to the fact that we can't know if a text is styled or not 
may lead to meaningful characters behind removed: as a result, several 
clients can display different things.
This can be notably disastrous is a user copy/paste content such as 
code.

§7 states "When applying formatting it may be desirable to include 
directives to exclude formatting characters from being read.", but how 
can we achieve this when we can't know if the body is styled or not?

> 
> 5. Is the specification accurate and clearly written?

2 main concerns:

- The §7 quoted above doesn't explain how to apply directives to exclude 
formatting characters

- is it not clear enough that formatting characters MUST be kept


> 
> Your feedback is appreciated!

Thanks for the work done, but I'm really concerned with this specific 
extensions, and I think that a proper discovery/flagging mechanism would 
be the bare minimum to make it usable (but even with that, I'm more in 
favor of a proper separation of formatting and meaningful text).

Regards
Jérôme Poisson (Goffi)


More information about the Standards mailing list