Quoting Daniel Kahn Gillmor (dkg@fifthhorseman.net): > >> I personally think the foundry should not be present at all unless > >> there are multiple packages with the same name (ala netcat-traditional > >> / netcat-openbsd). > > > > I think it is convenient if you have a collection of fonts from the same > > foundry and you would like to identify them all at once, e.g. bacause > > you consider all of them of general high/low quality. > > I tend to lean towards pabs's perspective here. There could be > foundries whose work varies, and some fonts are higher/lower quality in > the same foundry. The package name isn't the place to indicate that > (though of course a meta-package for all fonts from a given foundry > should include the foundry name in the package name). There's also a bit of history in this, particularly when it comes at SIL-endorsed fonts and thus, as a conservative person, I was tempted to follow the direction set by this family of fonts. As a compromise, we could leave this up to the maintainer. So, the "foundry" component thus becomes an even more optional component of the package name with a mention along the lines of "the package name may contain the foundry name following "fonts" and prefixing the font name. This should be considered only when there is a need to make it clear that fonts belong to a same "family" or origin.
Attachment:
signature.asc
Description: Digital signature