Re: SVG's future

On Sun, Feb 12, 2017 at 11:27 PM, Sebastian Zartner <
sebastianzartner@gmail.com> wrote:

> On 13 February 2017 at 06:05, Rik Cabanier <cabanier@gmail.com> wrote:
> >> No own suggestions/ideas, what to do to get SVG implementations
> complete?
> >> I think, there are meanwhile millions of authors with a lot of content
> >> around,
> >> much interested in complete implementations, without a need to worry
> about
> >> different bugs and caps in different common user-agents, surely a lot of
> >> them
> >> interested as well in a new version of SVG with new features,
> simplifying
> >> their work or even allowing new types of images
> >
> > SVG is more successful today than it's ever been. Maybe those incomplete
> > features weren't needed in the first place?
>
> That's a claim which adds fuel to the fire for those trying to defend SVG
> 2.
>
I'll rephrase: with SVG being so popular, if a feature has bugs why were
those not fixed?

The previously mentioned mesh gradients are a good example proving
> that this claim is not generally true, because there's a lot of
> interest for them.


Mesh gradients are currently underspecified and from my experience very
difficult to implement. For example, Apple Preview still displays some of
them wrong after 20 years.

>
> > I believe that the future of SVG does not consist of new graphical
> features
> > but of a deeper integration with the rest of the platform as well as
> offer
> > more consistency. (ie common matrices, CORS/CSP, CSS)
> > That can be done outside of SVG and AFAIK is still moving ahead.
>
> I agree that SVG profits from more consistency with the rest of the
> platform - and SVG 2 does a giant leap towards that, already. Though,
> as the feature support spreadsheet[1] indicates, browser vendors only
> partly share that opinion.
>

Thank you for pasting that document. Amelia is a great example of an author
who helps with the process. There would be no SVG2 document without her.
My advice is for you: pick a feature, build consensus, write test files and
help with the implementation.

It's a lot of work and slow.


> But I believe SVG's future does not only lie in consistency with other
> standards, it also needs to provide new (partly already long-demanded)
> features to stay successful, like it's done for any other standards of
> the W3C is working on.
> And I hope that the CSSWG keeps that in mind if it takes over parts of
> SVG 2 as discussed[2].
>
> Sebastian
>
> [1] https://docs.google.com/spreadsheets/d/1kkqzcxY53h7liRYppLSSFG2sjaJ8V
> 8TCP5rWLZK0AxA/edit#gid=0
> [2] https://lists.w3.org/Archives/Public/www-style/2017Feb/0046.html
>