Re: [Tools-discuss] UTF-8 box characters with wavy lines in XML2RFC v3 draft

[Tom Pusateri <[e-mail]>]

> On Nov 5, 2019, at 11:57 PM, Adam Roach <[e-mail]> wrote:
> 
> On 11/5/19 8:39 PM, Tom Pusateri wrote:
>> We were told on this list that box characters were allowed in artwork.
> 
> That's not on my radar. I'll let Heather (copied) chime in with her perspective.
> 
> In any case, the *general* problem you're running into is that many monospaced fonts don't include much more than the basic accented Latin character set, and this results in substituting glyphs from other fonts (rather than rendering tofu or the questionmark "replacement character") that are of a different width than the fixed one used by the font in question.
> 
> This is why, had the question come up generally, I would have pushed back on the notion of using box characters in diagrams [1]: even if it works okay on your system, it's likely to fail on someone's.
> 
> The *specific* problem you're running into is that Roboto Mono (or whatever your system uses as the default monospace font if you don't have Roboto Mono installed) doesn't contain "▶", and so it's getting pulled in from some other font, where it's wider.
> 
> /a
> 
> ____
> [1] That is, absent some compelling data showing that they are generally available in all -- or at least the vast majority of -- popular fixed-width fonts. I think the current episode shows where such an assertion would be implausible.

But we completely control the HTML document and can even embed the font.  This sounds like a solvable problem.

Tom