Another post about providing font samples

by Michael S. Kaplan, published on 2007/03/27 08:11 -04:00, original URI: http://blogs.msdn.com/b/michkap/archive/2007/03/27/1962027.aspx

Mandi asks:

I pray to God you read this email. I cannot find a way to print out a sample of all of vista's font on my new HP computer that came with Vista pre installed. I know for windows xp was hard. If there is some way can you please forward it on to me or a link or anything?

Thank you

Probably no need to invoke deity when asking me a question, since I don't even belong on a pedestal, let alone in some holy place. :-)

Unfortunately, there is no good answer, assuming Mandi is looking for what I am thinking about here, though looking at In Vista, jackdaws appear to be somewhat endangered and The lazy yet foxy jackdaw I love jumped over my quick brown sphinx dog of quartz, I think the mechanism for how to answer this different but related question is clearer.

The need is to get a representative sample of each font, whether that sample goes to the Windows Font Viewer or a sample of some or all of the fonts side by side.

Now one can imagine building in all kinds of mechanisms to automate the process (and I have been, both similar to and different from the one Dean Harding suggested).

I am not swayed much at the argument (sent to me by several) that text should be added to each font, provided by the font foundry. Because I doubt this would capture each possible target language that a user might prefer to see, if they have a preferred language within a script.

So here is my working model for the ideal plan (assuming no limitations on time, resources, knowledge, or authority!) per font:

1. Determine the supported Unicode script ranges in the font to try to determine the primary script(s) for which the font was designed;
2. Determine the user interface language of the person looking for the sample;
3. When #1 and #2 intersect, use that language and pick a stock sentence in that language;
4. When #1 and #2 do not intersect, use a default language representative for the script and pick a stock sentence in that language;

I was tempted to add an item to "allow the user to specify their own string if they would like" but since they can do that now, it seems less important....

Of course item #1 is the only one that is really complicated (the task of collecting a large number of non-offensive stock sentences, one per language might be time-consuming but it is a much more tractable problem!). Determining the primary script(s) of a font is complicated by many factors:

• Many fonts will contain a basic set of characters unrelated to their target like all of the basic Latins;
• Many of the fonts in Vista (for example) contain the bulk of the Latin/Cyrillic/Greek in Unicode;
• Some fonts are meant to be more generic and contain huge subsets of Unicode as that is their goal.

And of course #4 has as whole land-mine of opportunities to offend people by the choice of "default language" any time a script supports more than one language; perhaps when #4 is being used a random choice is most appropriate? :-) 

But I don't think it is an impossible problem to solve, though I am sorry that we have nothing built-in right now to do it. Sorry for Mandi's sake too, since the question there is similar in spirit to the question trying to be answered here....

This post brought to you by ᘻ (U+163b, a.k.a. CANADIAN SYLLABICS CARRIER TLO)

go to newer or older post, or back to index or month or day