Link to tables and figures by name

[ProgramMax]

ReSpec added support for linking to tables and figures by name rather than id. This matches the way ReSpec links to sections.

Go through the spec and see if there are links to figures/tables which make more sense to link by name rather than id.

[svgeesus]

So I honestly don't see anything terrible with

      <p>The relationships between the four kinds of image are illustrated in <a href="#image-relationship"></a>.</p>

      <figure id="image-relationship">
        <!-- Maintain a fragment named "figure41" to preserve incoming links to it -->
         <object id="figure41" data="figures/image-relationships.svg" type="image/svg+xml">
        </object>
        <figcaption>
          Relationships between source, reference, PNG, and display images
        </figcaption>
      </figure>

instead of, I guess,

      <p>The relationships between the four kinds of image are illustrated in <a>Relationships between source, reference, PNG, and display images</a>.</p>

      <figure id="image-relationship">
        <!-- Maintain a fragment named "figure41" to preserve incoming links to it -->
         <object id="figure41" data="figures/image-relationships.svg" type="image/svg+xml">
        </object>
        <figcaption>
          Relationships between source, reference, PNG, and display images
        </figcaption>
      </figure>

Either way, this was clearly not a blocker to going to Candidate Recommendation.

Close?

[ProgramMax]

Agreed @ definitely not a blocker.
IIRC, ReSpec produces different results using the different linking methods. If there aren't differences, the argument seems to just be "do it in a more ReSpec kind of way" vs. "progressive enhancement". I favor progressive enhancement.

I'll look into this.

[ProgramMax]

I just tried to add. Earlier, I mentioned that ReSpec added support. But now I'm not so sure. I think a similar feature was added but not the one I thought. I'll update once I know more.

[svgeesus]

Removing the tag until we know what exactly we want to do here