[lmi] Fwd: Re: [lmi-commits] master 5f53cca 2/9: Appropriately treat font sizes as static data

[Greg Chicares]

[Resending because yahoo's mailer daemon failed to deliver--apparently
the failure is theirs and not gnu.org's, as there is no specific
explanation given in their failure message:

| Sorry, we were unable to deliver your message to the following address.
|
| <address@hidden>:
| Unable to deliver message after multiple retries, giving up.

]

-------- Forwarded Message --------
Subject: Re: [lmi] [lmi-commits] master 5f53cca 2/9: Appropriately treat font 
sizes as static data
Date: Sat, 22 Sep 2018 19:23:33 +0000
From: Greg Chicares <address@hidden>
To: Let me illustrate... <address@hidden>
CC: Greg Chicares <address@hidden>

On 2018-09-19 15:25, Vadim Zeitlin wrote:
> On Wed, 19 Sep 2018 09:19:33 -0400 (EDT) Greg Chicares <address@hidden> wrote:
[...]
> GC> -        // Use non-default font sizes that are used to make the new
> GC> -        // illustrations more similar to the previously existing ones.
> GC> +    // These font sizes differ from wxHTML defaults.
[...]
>  I don't really understand why did the comment above turn into the one here
> in this commit. IMO the current commit is not very useful (as anybody who
> cares about this can check that the sizes are different from defaults) and
> raises more questions than it answers, notably _why_ do these sizes differ.
> Of course, maybe it's obvious that they were carefully chosen in order to
> match the existing outputs as closely as possible to anybody else reading
> this code, but it definitely won't be obvious to me when I'm going to
> reread it in a couple of months, after having forgotten everything about
> it.
These font sizes
  (1) differ from wxHTML's builtin defaults, because they
  (2) are more appropriate for life-insurance illustrations.

[Attempting to duplicate XSL-FO output was just a transitory means to
an end; the real end is typographical suitability in the business and
regulatory context. We have made some deliberate changes that make the
new PDFs more suitable than the old.]

To me, (2) didn't need to be mentioned, but (1) was not, because I didn't
understand what the "standard" sizes were until you explained that
|  "Standard" in this context really means "used by default by wxHTML",
here:

https://lists.nongnu.org/archive/html/lmi/2018-07/msg00033.html
| -        ,{7, 8, 10, 12, 14, 17, 20} // Standard HTML font sizes.
| +        ,{6, 7, 8, 9, 11, 13, 16} // Standard HTML font sizes for 8pt base.
|          );
| 
|      int pos_y = 0;
| ---------------------------------- >8 --------------------------------------
|
| GC> Searching online for "Standard HTML font sizes" produces such a chorus
| GC> of "Use CSS!" that I'd almost think HTML3 predated the web.
|
|  "Standard" in this context really means "used by default by wxHTML",
| although it does base the font sizes used on what antique browsers used to
| do, i.e. uses the factors {0.75, 0.83, 1, 1.2, 1.44, 1.73, 2} that (almost)

>From your POV, OTOH, what might be suitable for lmi is far
from intuitively obvious, while wxHTML internals may not need
to be explained in code that depends on wxHTML.

Soon I'll commit and push a documentary change that attempts
to make both points clearly, and references the mailing-list
post cited above.