Most online foundries and sellers let you set a few lines of text on their sites. That usually works for me if I am trying to decide on a headline face.

Back in the old days, you only had a printed catalog to go on, and had to mark up typewritten text to send it off to the typesetter. Then you had to wait until the proofs came back, and mark up corrections -- or, if the text was too long, have an editor mark up enough lines to shorten the text -- and send it back to the typesetter for another round...

My point is that we designers have it pretty easy these days, and we have several ways of testing a face before we hit the "Buy" button.

it always helps to do some research and look at the existing designs that utilize the typeface in question. Many foundries have examples of their type in use.

I can foresee many problems should a foundry allow people do download either 1 weight or a limited glyph set.

A question comes to my mind over and over: in many cases, actually you must buy the pint of beer and prove it by yourself to be sure if the beer is really what you need. And it is very frustrating when not.

sii-

stick with beer you already know. you waste less money that way.
and that newfangled beer won't get you any more drunk anyway...

-bowerbird

"actually you must buy the pint of beer and prove it by yourself to be sure if the beer is really what you need."

Yep, but since a good foundry font costs much more than a beer Cristobal's question doesn't sound so out of place. A couple of foundry who offers font fot trial, in different ways:

Underware. You can buy for 10 euro the Dolly specimen book, which comes with a cd with the font, fully working. Then if you decide to use the font professionally, you have to buy the full license.
http://www.underware.nl/site2/index.php?id1=dolly&id2=overview

B&P Type Foundry. You can download their fonts for free, but with limited glyphs - just to test them in layouts.
http://www.bpfoundry.com/fonts/fonts.htm

That said, there are some independent type designers who are very friendly and can help the buyer in deciding if the their font is suitable for him, even setting a test-layout, then sending the pdf to him.

Also, a way to see the fonts in use, though only on screen, is using the search function in Google Books (with the font name in the search field or a search phrase like "set in [typeface name])".

I've gotten test licences from Typotheque. It involves signing a form saying you understand you may only test the font, and will pay for it when you actually end up using it for any public work.

FWIW, as a not-yet-so-very-experienced designer, I appreciate such offers a lot.
Whereas display/headline stuff is nicely covered by the type testers of this world, with text fonts I often find it very hard (if not impossible) to tell if they'll work, and what kind of color and mood they'll bring to my layout.

> Back in the old days, you only had a printed catalog to go on,

It is certainly true that as far as the hoops we had to jump through to get type set (what Ricardo summarized), things are much more convenient now for designers.

But with regards to evaluating a typeface before investing in it, there were some resources of the old days that have slipped away, not adequately replaced by most current methods.

Where text types are concerned, an old printed catalog might have page after page of a given typeface set in a variety of sizes and leadings and even measures, in adequate paragraphs to truly judge the overall color and fit. Some large composing houses produced entire books of text showings -- cf. Kingsport, in three volumes. And, most importantly, they were *printed.*

It seems that in this digital age, there are plenty of opportunities for designers to see how typefaces look, but perhaps not quite as many opportunities to judge how they perform.

Or you can just print out a sample .pdf or request one if none is available.

When you can convince a hooker to let you try her product before you pay, then come back again and tell me why you should be get access to fonts before you pay.

Thank you so much for all your answers. I see I’m not the only guy thinking like this.

I know it’s not an easy question, and I am not asking only from the typographer side: I do type design as well (custom fonts, mostly; first commercial family by the end of the year).

Since fonts are being compared here to beers and paid-sex, I suppose I am allowed to do more “apples to oranges” comparisons:

—When you want to buy a new car, they let you drive it first. Full functionality, limited time.
—Demos in software are very common. Full functionality, but limited time. Or limited functionality, unlimited time.
—If I go to the bookstore, most of the times I like to read some paragraphs or even a few pages of the book I am interested in. I don’t see anything bad on this. Full functionality, limited time.

My point is: a good typeface family is not cheap. As a typographer, I would like to know better if I am doing a good decision by choosing this or that. I am not trying to get everything by free. On the other hand, as a type designer, I want that people who will (hopefully) buy my fonts, is happy and not angry because they bought something that don’t fit their needs. But I don’t want give my work for free, either.

That’s why this question comes to my mind over and over. I’m just trying to get the best for everyone. That’s it. And, thanks to your answers, I see there are more options out there.

The problem is that it is hard to test drive a font. Let's make a demo font without the e glyph. But what if the e is the character that you need for a logo? Makes it difficult. And to have different demo versions would put a lot of hassles on the vendors shoulders, and might make it so the wrong sort could build a cracked version.

I'm not a programmer, but I don't know of any way to time-release a font, and even if it happened, your customers would screw you up by asking for a modification in the sample you sent out five weeks ago (and now the demo has died). Unlimited time release demos are self defeating again.

How about a text composer (limited words, or characters), but with full functionality, i.e. the user gets to pick the size, leading, paper size (letter, legal, A4), etc, and is able to export it to a xxx dpi bitmap PDF (pretty much like the Collis specimen) which he or she can print.

Josef, Typotheque offers such a PDF tester:
http://www.typotheque.com/PDF_font_tester

If I'm not mistaken, some licenses do not allow use of the font in a PDF.

james said:
> When you can convince a hooker to let you
> try her product before you pay, then come back again
> and tell me why you should be get access to fonts
> before you pay.

for some reason, i'm reminded of the old saying:
> "i don't pay her to have sex with me,
> i pay her to leave when we're done..."

-bowerbird

Full functionality, limited time.

So how do we make that work without some kind of DRM system that builds expiration dates into fonts the same way the same way programmers handle demo software?

for some reason, i’m reminded of the old saying:

Yeah, but good luck even getting started without flashing a few benjamins.

A question comes to my mind over and over: in many cases, actually you must buy the pint of beer and prove it by yourself to be sure if the beer is really what you need. And it is very frustrating when not.

Actually, sii, one of the bars I frequent lets you try a small amount of beer before you buy anything. It's kind of like how in an ice cream shop they'll give you a taster spoon.

I'm not sure that this is the best model for fonts, though. Most retailers have great ways to test out headline faces, since you can use the "Test Drive" feature on MyFonts, H&FJ, vllg, Veer, FontShop, etc., take a screen shot, and place it in your document. But the tools for testing extended text types are limited, as Kent was saying. Perhaps foundries could have a tool where you specify line length, point size, leading, and a variety of other parameters, and it spits out a page of lorem ipsum in PDF format for you to print. It's not a perfect tool, but it would certainly help.

Of course, this is all on the retailers. Retailers that want to market their fonts well have been introducing increasingly complex and useful tools for sampling fonts before you buy them. I think that it's only a matter of a few years before MyFonts or FontShop or someone else releases something like this.

I don't think you ever know if a font is "good for your project" till you see the finished work in print or online.
Even then, you would probably refine your specs next time you use it.

Thanks again for all your comments. So far, I think the easiest solution is the limited-glyph font, since it doesn’t require any web programming, or expiration dates, &c.

Ah, so much talking about beers made me think it is the right time for one... ;-)

the limited-glyph font

...also known as "cripple-ware" :-)

It i• very u•eful for vi•uali•ing the font• in a de•ign.

The glyphs you'd limit would depend on the font. For a body text you could leave out a few letters and simply replace them with a no-width space, or even a similar letter. E.g. kill the w for a v and the m for an n. Makes it useless for a text font but visually testing it it won't change much. Or caps especially, since they are what? 1% of the letters we use, but won't interrupt the flow of a sentence for testing. For a display font, you could even mark up a few glyphs with a small slash cut. Yeah, someone could just edit the splines, but if they're going to go to that much work they're also just going to trace an image sample or edit splines on a PDF sample.

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

Fonts are tools.

A carpenter can't hammer a nail without buying a decent hammer either.

As a professional, one has to invest in fonts & other tools, and build up a collection over time.

Remember the ol' days when it was printers and type setters who owned type instead of designers? Designers used skill & imagination and then they specified the type they wanted the type setter to set. And it was up hill to the print shop - both ways!

Sorry if that sounds cranky. It's been a long week. :o)

-=®=-

I actually was thinking in something like the B&P Type Foundry way:

—Upper and lowercase complete (ABCDEFGHIJKLMNOPQRSTUVWXYZabcdefghijklmnopqrstuvwxyz)
—In the letters with diacritics, the letter without the accent.
—Punctuation: only period, comma, hyphen, space.
—No figures, no kerning.

For text fonts, that would be enough to give a good idea of its performance (how much pages will my text need to fit?, for example), and, on the other hand, this way the font lacks of some important glyphs needed to set real texts (figures, for example).

I know there’s no perfect solution, either: what if you need precisely to test the figures? Anyway, I think this would work most of the times for the majority of users (and that is a good deal for me). Of course, some people would dislike to have this crippled font, but I suppose some people would be upset for the DRM solution as well, or the expiration date solution, or the PDF-online solution...

The main advantage of the limited font would be that the user has at least some part of “the real thing” installed so it is easier to test it.

@Russell:

>A carpenter can’t hammer a nail without buying a decent hammer either.

I agree with you on this point: you must buy the best tools you can. And I do. I am just wondering if it is a good idea to let the carpenter to test the hammer before he buys (in this case, I assume the carpenter goes to the store and watch it, handle it, compare it with other hammers and just after that he pays it). He doesn’t want to steal the hammer, he wants to buy it.

Anyway, maybe it is not a good idea. Maybe it is better that the carpenter don’t touch the hammer before he pays. Maybe he should buy the hammer even if it doesn’t work so he can learn, and the next time he won’t buy that kind of hammers again (or he won’t buy in that store again, for that matter). Is this really the best for everybody?

>Remember the ol’ days when it was printers and type setters who owned type instead of designers?

No, I don’t remember those days. I wasn’t there: my first professional job was in 2000. And many typographers, increasing every year, are in the very same situation. And maybe because I live on this evil internet age, I really appreciate to test things before I buy them (NOT instead, at least not in my case). I assume some other people, including some typographers who didn’t know the old days, will think like this. They don’t care how difficult those days were; they care about solving their problems today.

If I could test the fonts, as I do with some other products, that would be great for me as a customer, since I can take more informed decisions, even if it is glyph-limited. (If it is a bad idea, it is OK, I was just wondering.) But I am a type designer as well, and I want to be someone who help their potential customers as much as I can, even if that means that someone will test my product and finally don’t buy. (Maybe it is a bad marketing idea, too.)

And maybe because I live on this evil internet age, I really appreciate to test things before I buy them (NOT instead, at least not in my case). I assume some other people, including some typographers who didn’t know the old days, will think like this. They don’t care how difficult those days were; they care about solving their problems today.

I don't think of this age as necessarily evil (or that the internet is evil, either). Like any other technology, it has good and bad qualities. And I can relate to what you say: I much prefer to go to a bookstore and thumb through a book before I put down some hard-earned cash to pay for it. I don't like buying things without looking at them, either. I might get stuck with something bad and regret my purchase.

But what you can do with the internet is educate yourself as much as you can about something before buying it. You can read descriptions and reviews of what is available, you can even find user forums and see if people are happy with product X -- or not. And as far as typefaces are concerned, in addition to setting some text online, you can request printed specimens from many foundries, who will be happy to put you on their mailing list.

—Upper and lowercase complete

So Anglophonic designers can use it for all their work for free, as long as they don't use any French?

—No figures, no kerning.
For text fonts, that would be enough to give a good idea of its performance

How would unkerned type give any idea about a font's performance? Please ...

i agree with nick, most times i only really know the right typeface was used after the job has come back from the printers. i then decide how i can use the typface better next time, or if another one is needed.

i purchased the stag family last year, and after using it many times in various settings i believe i am only just now getting used to it enough to set it just right, so it can take time to make the most of a typeface.

so at some point, you just have to dive in and buy the thing, then see if it works. the great thing is that even if it doesn't work first time, you then have a working knowledge of it and so can find the right place for it next time. my work purchased the neutraface family, so as part of me getting used to it, i set an entire book with it. it was a challenge, but i now understand that typeface so much more (and it actually worked quite well).

i've noticed that quite a few foundries will set a few pages for you in the desired typeface and send you the pdf to print out, which i think is a fair compromise.

i’ve noticed that quite a few foundries will set a few pages for you in the desired typeface and send you the pdf to print out, which i think is a fair compromise.

That's good to know, Simon. Thanks for the tip.

I think you would be surprised at how much a type maker will do to help you with your font decision. But you do have to ask, we can't read minds. (Well we actually can read minds, but it really cuts into the amount of time we have left to make new type.)

James

> Most online foundries and sellers let you set a few lines of text on their sites.

It used to work for me just fine until I had run into substantial rasterization differences just yesterday. Just 2c.

@Jan (paragraph):

In a past comment, you implied that sample font won’t be useful if it doesn’t contain some basic letter (“It i• very u•seful…”). If I say that you may include all the upper and lowercase, you say anglophonic designers can use it as it is and don’t pay for it “as long as they don’t use any French” (so, if I understand you right, you are telling that this limited-glyph font would be enough for anglophonic users). But, on the other hand, as it doesn’t have any kerning, you say it doesn’t work for testing the font performance (except for anglophonic users, right?).

—About kerning: I think kerning is important in a text font, but not essential in order to test it, given it has a good tracking. In addition: if you really think kerning is that important, I don’t see why don’t include in it, because it will contain only kerning pairs of the letters you put in the sample (not enough for real texts). Or it may include only the most important kerning values, or whatever.

—About upper and lowercase: upper and lowercase complete is not enough for a text font: even anglophonic users need to write sometimes names like François Mitterrand or César Chávez. And they certainly need the figures (not included).

But I suppose you are going to tell me now that anglophonic users may use it in this crippled condition, “as long as they don’t use any figures” (twenty-four instead of 24), and that a font without figures doesn’t give “any idea about a font’s performance”…

I believe Stefan Hattenbach of MacRhino is also willing to work with designers in order to get his fonts into circulation.

I think a limited character (full alphabet upper & lower w/o diacriticals or extended character sets) is a very generous way to see how the font will look in use. Honestly, you can't have everything, and you don't need the accents in order to see if you like the font (plus you CAN see them in a pdf). It's a test drive after all. A lot of softwares offer this sort of deal - beta or free trial version and then you pay to unlock the complete application. Makes perfect sense to me. I wish this were more readily available.

It would also be useful in terms of showing the client a sample so you can get approval before purchasing the font.

Another approach by Type Together:

“Buy a single weight now and get reimbursed if you buy the whole bundle later. This is a great way to explore a new typeface without full commitment.”

Not exactly about this, but related. To pay for a single weight certainly is more affordable, and you can use it with full functionality. The important part is that you don’t loose that money if you come back for the whole family later.

Good deal for me, too: if it doesn’t work, I haven’t lost a fortune. And I can use it later, anyway.

used to be, you had to marry someone
to see if you were sexually compatible.

the kicker was, if it didn't work out,
then you couldn't marry anyone else,
not before doing the divorce process.

you can see why this system didn't last.

-bowerbird

Cristobal, my problem with the limited-glyph font proposal it that there are lots of potential uses for a font, where the lack of numerals and accents would be no impediment: a logo, banner, headline, ad and other short stuff. On the other hand, to evaluate the performance of a font in a text context: magazine, newspaper, book, journal, I would still hold that kerning would be essential, for both aesthetic evaluation and, say, for the purposes of a cast-off or copy-fitting. I would much rather spend some time setting someone's text in one of my fonts to spec as a courtesy example, rather then spending the time artificially crippling the fonts.

Ok, got it, Jan. Thanks for your comments. It is good to know different opinions on this subject.

@ Bowerbird: in some countries there are no divorces at this day. Talking about fonts, at least you can have different “wifes” at the same time if you pay. ;)

(But talking about marriage, in some places you can have more than one wife, if you can afford it. No much difference, after all.)

Also, a way to see the fonts in use, though only on screen, is using the search function in Google Books (with the font name in the search field or a search phrase like “set in [typeface name])”.

Ncaleffi. I noticed you dealt with a rather tricky typographic situation at the end of the above paragraph by putting the full stop after the close double quotes. Is this correct practice? It's just that I always thought it looked better to put the full stop first."

Bloodtype: that's an American v British system, though more and more in the US you see the period outside if it's not part of the quotation. Most other languages (at least European) tend to use the more British style by default. Programmers tend to almost universally use the "logical" quotes with periods outside. But in Ncaleffi's case, I think he meant to type "). instead of )".

«El futuro es una línea tan fina que apenas nos damos cuenta de pintarla nosotros mismos». (La Luz Oscura, por Javier Guerrero)

For Bloodtype: I don't see anything weird in having the full stop after the closing guillemets, though I have seen the period inside the quotes, for example, when a character speaks in a novel:

Alice: "But I don't want to go among mad people."
The Cat: "Oh, you can't help that. We're all mad here. I'm mad. You're mad."

Now, going back to the subject of this thread, for sure the best way to judge a font is having it printed in high resolution by a professional service. But a helpful way for a typesetter to better understand how a font works (and not just how it looks), could be having some "typical" double page spread (a novel, an essay with footnotes and quotation, a poetry book), printable in A4. A thing that, in my opinion (and from my limited experience), most of the pdfs available in foundries' site don't do. For example, one of the reason I decided to buy a license for Whitman was the pdf made by Kent Lew, which somehow resembles a typical book-situation. I'm speaking of setting texts in books here.

Now about the limited glyphs matter: I don't think that anybody could use such a font for a professional layout (besides logo and titling, maybe) - I have made a pdf test with the free BP Romain and even an English text shows problems (no numerals, wrong accent). Unfortunately, I can't attach it here. Anyway, if Cristobal decides to go with it for his Espinosa and Fondo (which look very nice, by the way), I think that they should have kerning enabled; otherwise they're useless.

I agree with Nicola on the printable A4. Being able to print fonts and check them out first where they shine best, may it be books, magazines, or newspapers, is always a plus. It's just difficult to reproduce the production processes involved, like how newspapers are printed.

I am still up for the high-resolution (or variable-resolution) bitmap PDF generator. There is no risk of extracting the embedded font files, and the user can see if the font/s look good under conditions within his or her control.

For other type-tasting options, Type Together's is a pleasant deal.

> if Cristobal decides to go with it for his Espinosa and Fondo (which look very nice, by the way),

That felt good, Nicola.

Actually, I am doing more weights for Fondo at the moment, but those will remain, as the first weights, for the exclusive use of the Fondo de Cultura Económica. Yes, I know, I know, it is sad…

But I really think that the new Espinosa is going to be better than Fondo (at least that’s the way I see it). It is completely redrawn. Hopefully you can see something about Espinosa in the next few months. And no, I don’t know yet if it is going to be in the form of a limited-glyph font. :P

>I think that they should have kerning enabled; otherwise they’re useless.

I’ll think about it in the meanwhile, but it makes sense to me. Oh, so many options…

Some software will let you test it for 30 days. Usually you buy it anyway. Perhaps a system like that would work? That way a designer could try out a font, but not be able to give it to a printer or copy it in any way and it would just stop working after 30 days. I think I would pay more for a font if I knew that I liked working with it first.

Once again, fonts are compared to physical objects – that is, to the extent it's convenient. This time, the analogy is beer, and it makes as much sense as you'd expect.

Of course, if you try the pint and decide not to pay for it, the owner has lost one pint without compensation. The foundry, in contrast, can still sell the font.

Of course, the beer doesn't come with a BS contract purporting to limit to what you can do with the beer. You can do anything with it unless prohibited by law.

If you want to modify (or fortify!) your beer, you can. But a font, you can copy free for private study, because the legislation says it's ok.

On the other hand, beer also comes in multiple weights – the darker, the better; the light one is usually quite useless.

"Some software will let you test it for 30 days. Usually you buy it anyway. Perhaps a system like that would work? That way a designer could try out a font, but not be able to give it to a printer or copy it in any way and it would just stop working after 30 days."

Which means, if not integrating the system in the OS, which could be averted by compiling your own Gentoo build, among others, that this will result in an expiry date bundled with the font, in which a way is to incorporate a software that automatically destroys any expired fonts; thus, we're treading the path of font DRM.

The key I think is in building a relationship with the brewer, cutting out the publican and going direct to the source. Get educated in the brewing process, be able to discuss the intricacies of the fermentation process with the brew staff. Then as you build the relationship you’ll be provided with free samples, test brews, and when you need to return a bad batch there will be no hard feelings.

Where's the hammer come into this, si? :¬)

-=®=-