Martin's summary does seem to cover things - ekr is wording along these
lines sufficient to address your concerns?

Alexis

On Thu, Jun 12, 2025 at 8:25 AM Martin Thomson <[email protected]> wrote:

> I think that we generally have good alignment here.  The resizing and dark
> mode are technically part of the same "reactive" part of the SVG
> featureset, but on the specifics, I think we have good alignment.
>
> I would add that there are some adaptive features that could be relevant
> for accessibility across display technology: adjustments to font size and
> stroke weight.  Adjustment of positioning.
>
> All of this is stuff that I would leave out of this document.  What we
> want is to say that:
>
> SVGs must not include animation or interactive features.
> SVGs should include only limited reactive design elements (scaling,
> dark/light mode, and perhaps minor adjustments to allow for variations in
> display technology). The intent being to ensure that how something is
> displayed does not substantially alter what is seen.
>
> That's a lot more words, but I think that it matches expectations.
>
> Technically, if we were doing what we did in the past, this would be nigh
> impossible to specify in terms of SVG and CSS syntax that is
> permitted/denied.  Now, we have an RPC that can exercise good judgment, so
> I'm not at all concerned about this being feasible.
>
> On Wed, Jun 11, 2025, at 21:50, Jean Mahoney wrote:
> > Hi all,
> >
> > On 6/11/25 12:13 PM, Eric Rescorla wrote:
> >>
> >>
> >> On Wed, Jun 11, 2025 at 9:49 AM Alexis Rossi <[email protected]
> >> <mailto:[email protected]>> wrote:
> >>
> >>
> >>
> >>     On Wed, Jun 11, 2025 at 2:34 AM Eric Rescorla <[email protected]
> >>     <mailto:[email protected]>> wrote:
> >>
> >>         TBH, all of this wordsmithing seems premature, because it's
> >>         still not clear to me what the text is trying to say. There are
> >>         a set of concrete features one might imagine implementing
> >>         here, and I don't see that we have come to agreement about
> >>         which ones we are trying to rule in or out:
> >>
> >>         - Scaling: Yes (probably?)
> >>
> >>
> >>     Yes
> >
> > [JM] yes, scaling helps make diagrams accessible.
> >
> >>
> >>         - Rearrangement : Maybe? ISTR some said no, Alexis said yes?
> >>
> >>
> >>     I don't care, happy to leave it up to RPC
> >
> > [JM] No, I think this would make creating PDFs hard, and would be hard
> > for authors to create and the RPC to validate.
> >
> >>
> >>         - Hiding and showing elements based on size: Probably not?
> >>
> >>
> >>     No
> >
> > [JM] no
> >
> >>
> >>         - Dark mode: Maybe? MT was arguing yes, but I just heard Jean
> >>         say maybe not?
> >>
> >>
> >>     I don't care, happy to leave it up to RPC
> >
> > [JM] Yes, I want to support dark mode, and the new rfc-editor.org
> > website will be able to support dark mode. If I sounded uncertain, it's
> > because I don't know the best way to construct an SVG to support it
> > (e.g., use inline SVG and currentColor?). Our current SVG profile works
> > just fine with dark mode so I'm guessing nothing too fancy is needed.
> >
> >>
> >>         - Animation: No
> >>
> >>
> >>     No
> >
> > [JM] no
> >
> >>
> >>         - Direct interaction with the image: No
> >>
> >>
> >>     No
> >
> > [JM] no
> >
> > Best regards,
> > Jean
> >
> >>
> >>
> >>         We should resolve these questions -- or at least the question of
> >>         the principle
> >>         which is trying to guide them prior to trying to write text.
> >>
> >>         If the intent is just to defer all of these questions to the
> >>         RPC, then the text
> >>         should actually say so (I would probably object to this, but
> >>         perhaps I'm
> >>         in the minority on that). However, I don't think it's helpful to
> >>         just have
> >>         vague text that doesn't actually provide the answer without an
> >>         explicit
> >>         delegation to the RPC.
> >>
> >>
> >>     This document specifically hands technical implementation decisions
> >>     to the RPC. The intro says, "The RFC Publication Center (RPC) is
> >>     responsible for making SVG tooling and implementation decisions."
> >>     Why would we need to specify it again in this bullet point as well?
> >>
> >>
> >> This is neither a tooling nor an implementation decision. It is a
> policy
> >> decision
> >> about what types of SVG are allowed (hence its appearance in the section
> >> entitled "Policy Requirements").
> >>
> >> -Ekr
> >>
> >>
> >>         -Ekr
> >>
> >>
> >>         On Tue, Jun 10, 2025 at 8:58 PM Brian E Carpenter
> >>         <[email protected]
> >>         <mailto:[email protected]>> wrote:
> >>
> >>             On 11-Jun-25 14:17, Alexis Rossi wrote:
> >>              > Okay, in an earlier email in this thread [1] Paul Hoffman
> >>             (I think) suggested wording like
> >>              >
> >>              > * SVG diagrams may not be interactive or have multimedia
> >>             or other similar elements.
> >>              >
> >>              > What about that for new wording?
> >>
> >>             That WFM too.
> >>
> >>                  Brian
> >>
> >>              >
> >>              > Alexis
> >>              >
> >>              > [1]
> https://mailarchive.ietf.org/arch/msg/rswg/iqz2v05Wt-
> >>             ncTJUsEvITw_50RGY/ <https://mailarchive.ietf.org/arch/msg/
> >>             rswg/iqz2v05Wt-ncTJUsEvITw_50RGY/> <https://
> >>             mailarchive.ietf.org/arch/msg/rswg/iqz2v05Wt-
> >>             ncTJUsEvITw_50RGY/ <https://mailarchive.ietf.org/arch/msg/
> >>             rswg/iqz2v05Wt-ncTJUsEvITw_50RGY/>>
> >>              >
> >>              > On Tue, Jun 10, 2025 at 5:37 PM Jean Mahoney
> >>             <[email protected] <mailto:[email protected]
> >>             editor.org> <mailto:[email protected]
> >>             <mailto:[email protected]>>> wrote:
> >>              >
> >>              >
> >>              >
> >>              >     On 6/10/25 4:55 PM, Alexis Rossi wrote:
> >>              >      >
> >>              >      >
> >>              >      > On Tue, Jun 10, 2025 at 2:24 PM Brian Carpenter
> >>              >      > <[email protected]
> >>             <mailto:[email protected]>
> >>             <mailto:[email protected]
> >>             <mailto:[email protected]>>
> >>             <mailto:[email protected]
> >>             <mailto:[email protected]>
> >>             <mailto:[email protected]
> >>             <mailto:[email protected]>>>> wrote:
> >>              >      >
> >>              >      >     Make the policy "generally not allowed" and
> >>             that would leave the RPC
> >>              >      >     free to apply common sense.
> >>              >      >
> >>              >      >     (via tiny screen & keyboard)
> >>              >      >     Regards,
> >>              >      >              Brian Carpenter
> >>              >      >
> >>              >      >
> >>              >      > Okay so like this perhaps?
> >>              >      >
> >>              >      > OLD
> >>              >      > "SVGs must render in a single static configuration
> >>             without dynamic
> >>              >      > elements or responsive design features."
> >>              >      >
> >>              >      > NEW
> >>              >      > "The content of SVGs should be static. Dynamic
> >>             elements or responsive
> >>              >      > design features are generally not allowed."
> >>              >
> >>              >     [JM] We should emphasize which responsive design
> >>             features we don't want
> >>              >     to support, like animation, because we do want to
> >>             allow scaling, which
> >>              >     also falls into responsive design. Other responsive
> >>             design concepts,
> >>              >     such as including CSS to support dark mode, don't
> >>             appear to be necessary
> >>              >     (SVGs in current RFCs look okay in dark mode). I
> >>             would need to more info
> >>              >     to determine if SVG CSS should be disallowed
> >>             generally. It seems like it
> >>              >     would be helpful for improving accessibility, but
> >>             maybe that can all be
> >>              >     inherited from the HTML CSS?
> >>              >
> >>              >     NEW
> >>              >          The content of SVGs should be static. Dynamic
> >>             elements or responsive
> >>              >          design features that support animation are
> >>             generally not allowed.
> >>              >
> >>              >     Best regards,
> >>              >     Jean
> >>              >
> >>              >
> >>              >
> >>              >      >
> >>              >      > Alexis
> >>              >      >
> >>              >      >
> >>              >      >     On Wed, 11 Jun 2025, 09:17 Eliot Lear,
> >>             <[email protected] <mailto:[email protected]> <mailto:[email protected]
> >>             <mailto:[email protected]>>
> >>              >      >     <mailto:[email protected] <mailto:[email protected]>
> >>             <mailto:[email protected] <mailto:[email protected]>>>> wrote:
> >>              >      >
> >>              >      >         __
> >>              >      >
> >>              >      >         I would suggest that the proscriptions of
> >>             video, audio, and
> >>              >      >         animations are not controversial.  The
> >>             only aspect of responsive
> >>              >      >         design I really think we're talking about
> >>             is dark mode, in that
> >>              >      >         if it is supported it must be done in a
> >>             way that is legible.
> >>              >      >         Maybe that is a bit TOO prescriptive, tho.
> >>              >      >
> >>              >      >
> >>              >      >         --
> >>              >      >         rswg mailing list -- [email protected]
> >>             <mailto:[email protected]> <mailto:[email protected]
> >>             <mailto:[email protected]>> <mailto:rswg@rfc-
> >>             <mailto:rswg@rfc-> <mailto:rswg@rfc- <mailto:rswg@rfc->>
> >>              >      > editor.org <http://editor.org> <http://editor.org
> >>             <http://editor.org>>>
> >>              >      >         To unsubscribe send an email to rswg-
> >>             [email protected] <mailto:[email protected]>
> >>             <mailto:[email protected] <mailto:rswg-leave@rfc-
> >>             editor.org>>
> >>              >      >         <mailto:[email protected]
> >>             <mailto:[email protected]> <mailto:rswg-leave@rfc-
> >>             editor.org <mailto:[email protected]>>>
> >>              >      >
> >>              >      >
> >>              >
> >>
> >
> > --
> > rswg mailing list -- [email protected]
> > To unsubscribe send an email to [email protected]
>
> --
> rswg mailing list -- [email protected]
> To unsubscribe send an email to [email protected]
>
-- 
rswg mailing list -- [email protected]
To unsubscribe send an email to [email protected]

Reply via email to