PDA

View Full Version : Universal font classification scheme


ktinkel
02-05-2005, 10:42 AM
Take a look at the “Type-Expertise Universal Font Classification System (http://www.type-expertise.com/whitepaper.htm)” white paper if you interested in this topic.

Type-Expertise has come up with a database scheme (patent-pending) that is meant to accommodate the roughly 50,000 fonts in existence today (up from 10,000 two decades ago, according to the white paper), as well as any number that may be released in the future. It attempts to meet the needs of users, type designers, font vendors, and application and OS developers.

The French organizers say this about their proposal:It is a global, inter-connected network of information that constantly enriches itself from the changing knowledge and font practices of every participating font user.
Interesting. They do not mention the PANOSE System developed in the 1980s by Benjamin Bauermeister, but I wonder how that compares to this one. It too is fairly extensible.

Cristen Gillespie
02-07-2005, 08:46 PM
Interesting. I was just thinking how difficult it is to sort out fonts. I just d'loaded the latest Suitcase for the Mac, and saw it allows for adding "styles" and "keywords" to one's own fonts, although this is a bit tedious, not to mention irrelevant if your terms quickly change. But more useful than not, even still.

I was looking for a font that was similar in kind to, but not identical with, a common font. Finding something with the same "feel" is difficult for those of us without enough experience to recognize all the names of all the fonts we wind up with.

With a lot of fonts, it's obvious that there are several made that are almost identical to, or closely derivative to, the well-known fonts. It would be useful to know what they are, too, and also useful to avoid buying fonts that are almost identical to what you already have. This service could be useful, if I understand rightly, for all of that.

Of course, it could also be too expensive for many people to use it.

ktinkel
02-08-2005, 05:06 AM
Interesting. I was just thinking how difficult it is to sort out fonts. I just d'loaded the latest Suitcase for the Mac, and saw it allows for adding "styles" and "keywords" to one's own fonts, although this is a bit tedious, not to mention irrelevant if your terms quickly change. But more useful than not, even still.

I was looking for a font that was similar in kind to, but not identical with, a common font. Finding something with the same "feel" is difficult for those of us without enough experience to recognize all the names of all the fonts we wind up with.

With a lot of fonts, it's obvious that there are several made that are almost identical to, or closely derivative to, the well-known fonts. It would be useful to know what they are, too, and also useful to avoid buying fonts that are almost identical to what you already have. This service could be useful, if I understand rightly, for all of that.

Of course, it could also be too expensive for many people to use it.I am afraid the real solution is to buy methodically, one or two font families at a time, and use them until you know everything useful about them. I have known many designers who rely on the same small collection of fonts for years and years, and that was during the time when it didn’t cost anything to change typefaces at will (because the typesetter bought them, not the designer).

Most of the fonts you happen to collect — bundled with applications, say — end up not being all that useful anyway unless you complete the family.

To me, this classication scheme is an academic exercise as much as anything.

Michael Rowley
02-08-2005, 10:57 AM
KT:

'classication scheme is an academic exercise as much as anything'

With font families so large (some of the more popular have 50 members), a scheme that selected the top 500 families would be more than enough for anyone's needs. I'm not in generally favour of birth control, but fonts need it, if necessary by exposing their more sickly offspring on the nearest mountainside.

Michael

ktinkel
02-08-2005, 11:56 AM
… but fonts need it, if necessary by exposing their more sickly offspring on the nearest mountainside.It is odd, isn’t it, that people who understand perfectly well that there are only a few really great movies or books or kitchen knives or shoes seem to think that all fonts are created equally wonderful when of course they are not.

Fonts, like most things, tend to fall along the classical Bell curve.

Cristen Gillespie
02-12-2005, 10:01 AM
Most of the fonts you happen to collect — bundled with applications, say — end up not being all that useful anyway unless you complete the family.

To me, this classication scheme is an academic exercise as much as anything.

We know that most of the fonts are nearly useless for the lack of styles, or just because they're badly constructed and impossible to set well. But they come with apps, they come with clipart packages, they come with home-user card and album apps, and until we know what they are, we're unlikely to dump them. Not as novices who haven't the experience to tell if they're the start of something wonderful, or just plain icky.

Many times fonts are useful not for reading, but for background, for just being art objects in a many-layered file. Plus, clients know what they want--they want unique, silly, useless, but they do want a display font that isn't what you have if you limit yourself to the 2 or 3 families you are getting to know.

The display font may look plain hideous to someone such as yourself with good training and years upon years of study, but it's a rose to the average person. Something that addresses our need to get a handle on these rampaging fonts isn't merely academic to those of us who are trying to get the experience, doing it all, even what we know we're not properly qualified to do. We didn't get to start with just a few fonts. We started with an overwhelming number of fonts, and getting through to the ones we can use and trust and still satisfy the client is not easy. I'm not sure the average person much notices the text font so long as it is legible, but the display font, that they want BIG to catch attention, they do pay attention to.

I didn't see this classification system as something experienced designers particularly want or need, but something the rest of us might benefit -- and learn -- from. Of course, there is plenty of room for that not to be the case.

Michael Rowley
02-12-2005, 11:23 AM
Cristen:

'The display font may look plain hideous'

Possibly the vast majority of fonts are display fonts; but in the days of lead type the most innocent beginner did at least know which were the fonts intended for serious use for books, pamhlets, leaflets, etc. and which were the fonts intended for posters, newspaper headlines, and similar things where large letters and letters intended to catch the eye were needed: he knew, because they were only available in 7–14 point sizes or 30 point (smaller than that?) and greater. Of the text sizes, the fonts differed mainly in body size, width, and colour, with the exception of the sans serif faces, but they were used mainly in continental Europe: their variety was not so great as to prevent the learner from fairly easily memorising the different faces. For instance, the 2nd edition of Simon's Introduction to Typography (1963) contained illustrative material on 23 text faces (all seriffed), most of which are still well known and most of them still used more often than not for book production.

Now we have an enormous range of fonts given free by Microsoft et al. that we can use at any size, but should not. It is like giving users the free run of a pharmacy without giving them any idea of the dosage or application of each drug.

ktinkel
02-12-2005, 11:45 AM
I didn't see this classification system as something experienced designers particularly want or need, but something the rest of us might benefit -- and learn -- from. Of course, there is plenty of room for that not to be the case.Standardized schemes are based on type features, like the Styles listed in the Suitcase list.

It sounds to me as if you would like to assign personal terms to particular fonts (create Suitcase Keywords, in other words). That you can do with Suitcase. As you say, tedious.

If nothing else, it will be interesting to see if the face that made you think of romance in May will still seem romantic in the fall. Not sure there is any agreed-upon reaction to typefaces (any more than colors) — even within a country or culture. Or one person, from one time to another.

I went rummaging, and somewhat miraculously came up with the Directory of Evocative Typography by Renee Lewinter (1980), which attempts to attach a mishmash of terms to type.

Here are the fonts she considers to be “festive”: Mardi Gras, Parisian, Balloon, Sapphire, Vivaldi, Campanile, Kismet, Tango, Mistral (leaving out faces or names unknown today).

For “science fiction” she lists Microgramma, Dominante, Avant Garde, Univers, Amelia, Orbit-B, Modula, Bauhaus, Revue, Glyphic, Inverserif.

Is that the sort of thing you’re thinking of?

Cristen Gillespie
02-14-2005, 09:21 AM
Now we have an enormous range of fonts given free by Microsoft et al. that we can use at any size, but should not. It is like giving users the free run of a pharmacy without giving them any idea of the dosage or application of each drug.

That's where I think some sort of system of classification, beyond saying "transitional" or "modern" could help novices understand what the face is good for. Just telling us "not for use below 24pt is a help. I think I'm capable of picking a text font--I stick with Adobe's (don't have a huge need for more) and while not perhaps inspiring, I know they set better than I can adjust any font, and they're legible.

I'd like to own a few of those families that come with 25 fonts, and I'm confident I could design almost anything with just a few of those families, but design doesn't appear to be about making the most of what you have. It's become a lot more in your face and "different for difference sake." Maybe that will turn around, seeing as how we're overwhelmed by an unholy amount of display faces these days, most of which are more work than they're worth to try to make work.

Cristen Gillespie
02-14-2005, 09:52 AM
I'm feeling a bit frustrated with Suitcase's long, long list of styles. Oddly enough, as long as it is, it doesn't have every single style a font gets named. I want to be able to remove about half of their styles, and can't see how to. I can only add and subtract my own. No Slab Serif, but Extra everything<G>

On the other hand, my personal keywords (classifications) are much more literal than your Ms Lewinter's. Probably my art history, but I do think in terms of the era I associate with a font, or I use descriptive words (splatter, heavy ink), to help me find something fast.

However, some descriptive keywords certainly can help. Very tedious to enter. Even more tedious is realizing that the same "apparent" font is under several different names, and do I keep them all? Or do I attempt to figure out which one has been created the best, without a clue how to decide on that? There is no way I can see in Suitcase to say "another Bodoni" or "another Bauhaus." Not without making my keywords list longer than their Styles list.

IOW, I have far more fonts than I need or want, but have little idea how to eliminate them while leaving the best behind.

I agree that it could be very difficult to say "Sci-Fi" or "Romantic" and have it mean the same thing to everyone, or even the same thing to the same person next year<G> But if you have a database where you can start to eliminate fonts, starting with suitable size, bold or thin, working down to sci-fi or delicate script, we'd certainly find our fonts a lot faster. Pairing fonts (like food--goes with beef, or best with shellfish) would also help a lot of us out here.

I guess what I'm saying is that the few classifications that fonts traditionally have aren't really enough to help those of us who don't know by the name what the font is. Add to the classification some descriptive words and at least it cuts down on the noise. How exactly we decide what those descriptive words are and which fonts they belong to is the very dickens, but some attempt ought to be better than none at all.

Or, of course, we could simply ban most of the fonts floating around. I'd be willing to let you be one of the team that decides which ones to ban<G>

ktinkel
02-14-2005, 10:26 AM
Or, of course, we could simply ban most of the fonts floating around. I'd be willing to let you be one of the team that decides which ones to ban<G>No doubt! But if you think you’re frustrated now …

tphinney
02-15-2005, 08:06 PM
The Type-Expertise stuff is interesting. Despite its organizers attempts to market it more excitingly, it is at its heart a keywording system. If one uses the keywords of the masses, then it is an oddly democratic keywording system. I'm not sure if that's a good thing or not! They do also support the Vox classification system within theirs (and maybe Panose too, I forget).

With the continued decay of the RIT database, I had suggested last fall to ATypI that we do our own font database. Unlike the Type-Expertise system, it would be oriented towards including non-digital fonts as well, more of a historic bent. The ATypI executive (myself included) has been busy with some other things recently, but I hope to revive this idea soon.

Cheers,

T

P.S. Nice to be back on the DTP Forums. Not sure how often I'll be visiting....

ktinkel
02-16-2005, 05:49 AM
The Type-Expertise stuff is interesting. Despite its organizers attempts to market it more excitingly, it is at its heart a keywording system. If one uses the keywords of the masses, then it is an oddly democratic keywording system. I'm not sure if that's a good thing or not! They do also support the Vox classification system within theirs (and maybe Panose too, I forget).Wonder who would pay for it, though. It is a commercial venture.

With the continued decay of the RIT database …That is very sad. Are they looking for a new home?

I had suggested last fall to ATypI that we do our own font database. Unlike the Type-Expertise system, it would be oriented towards including non-digital fonts as well, more of a historic bent.That sounds like a good idea, although it might be a bit labor-intensive for ATypI to tackle. I have the notion of establishing a Wiki for that sort of thing, but it is not at the top of our to-do list.

Now that you know the way, you can always drop by now and again. Always glad to see you.

tphinney
02-16-2005, 06:46 PM
Wonder who would pay for it, though. It is a commercial venture.

Yes, that's one thing. Personally, I would want to have any such database more openly available, among other things.

That is very sad. Are they [RIT font database] looking for a new home?

No. Archie Provan isn't interested in giving it up. But besides that, the overall consensus seems to be that the DB is not in good enough shape to make it worthwhile. Among other things, it was indexed to a collection of physical type specimens, most of which were tossed out after Archie retired. Very sad.

That sounds like a good idea, although it might be a bit labor-intensive for ATypI to tackle. I have the notion of establishing a Wiki for that sort of thing, but it is not at the top of our to-do list.

Well, I think ATypI would rely on a combination of volunteers and input from member type foundries as well. Leastways, that's my plan.

I love Wikis, but would it be easy enough to impose sufficient structure in that format, do you think? Or do you think structure might be overrated for this sort of project?

Cheers,

T[homas]

Cristen Gillespie
02-16-2005, 07:27 PM
No doubt! But if you think you’re frustrated now …

<BG> I'm sure you'd pare me down to the essentials. Of course, then someone would come alone wanting something with "pizzazz" whatever that is.

I was cruising Adobe and dropped into their type section. It looks from the font page that you can buy each font in a family separately, or by the family. I thought they were only selling by the family not long ago. More expensive in the long run to buy one or two at a time, but a way to build up a useful family, even if very slowly.