James Puckett

You should start by testing everything with an expansive set of proofs. If you don’t have proofs you can download mine: https://github.com/DunwichType/DTF_Proofs/tree/master/Latin

Make sure you review the diacriticals thoroughly. If you find no errors you weren’t paying attention and need to look again. If you need to compare against another font use GlyphProofer from Ben Kiel and House Industries: https://github.com/houseind/robothon
Don’t compare against a massive font like Noto or Brill. Typotheque, Typofonderie, H&FJ, Darden Studio, Commercial Type, and Underware are just a few of the great foundries to compare to. (There's nothing wrong with the Brill fonts, but using them with Glyphproofer with generate hundreds of unnecessary pages.)

Build your own proofs to test OpenType features.

Open your compiled fonts and review every single glyph in every single font for things like incorrect sidebearings or interpolation errors. If you find no errors you weren’t paying attention and need to look again. 

Test with the other popular Adobe apps: Photoshop and Illustrator. Look for problems with font names and how they appear in menus. Then play around a little to make sure that stuff like OpenType features work.

Install the fonts on Windows. Copy/paste the test strings from for Indesign files into Textedit and save as rich text. Make sure they at least work in Wordpad. If you have Office for Windows test in Word and Powerpoint.

Now go back to your Mac and check the fonts with Font Book. If you get no errors then install them. Test the fonts in Textedit, and if you have Office for Mac, test there too.

You should ask the Robofont forum if there are any common problems or mistakes people make. I don’t use Robofont so I don’t know about that (but I’ve never heard Robofont users complain. I rarely have issues with Latin fonts generated with the stable release of Glyphs. I only compile final fonts with the betas if Georg tells me to.

I think​ op-ed hacks are running out of people to point fingers at if they're pointing at fonts.

scannerlicker said:

And what was the inaugural moment for screen serifs? Georgia? I mean, it was almost 30 years ago (1993), so is there probably some earlier successful attempts?

In 1991 Microsoft added TrueType to Windows 3.1 and Apple added it to System 7. Both included nice TrueType versions of Times. But IMHO the real breakthrough for serifs on screen happened in the early 2000s when LCDs replaced CRTs and users moved to operating systems that used subpixel text rendering as the default. Georgia was a very functional typeface, as were earlier customized serifs like Courier and Times. But there’s a big difference between fonts needing to be painstakingly crafted for legibility on a screen and a font just displaying well with basic, or even no, hinting.

Hrant H. Papazian said:

As an aside, it's "funny" that a party claiming non-Latin is now big still declines to get into it... (Cyrillic? Sort of a technicality.)

Just because the market is growing doesn’t make it a great place to try making money. Designing, programming, and testing for complex scripts is far more challenging than for Latin. Marketing products and services can be an even bigger challenge. It’s not easy to do if you’re a one-person operation. 

1. I break estimates down to big line items: Design, kerning, hinting. More detail than that just bores and confuses busy art directors. If they need more detail they’ll ask. 

I haven’t produced a detailed quote in years because the real quote is done by the firm hiring me and shown to their client. I save detail for the contract. Spending a day or two writing a great contract after the client agrees to the prices is important. Wasting a day or two on a detailed quote just to get shot down by somebody who expected it to cost ten times less is silly.

If someone wants something that will eat up lots of time, like distressed outlines with hints, I explain up front what the work entails. So they don’t think I’m just taking them for a ride. 

2. After I work out an estimate I tack on at least 20% for unplanned stuff that inevitably comes up. This saves time getting approval to spend more when the AD has already run the project four weeks past deadline. And it makes everyone look good when the project doesn’t run up extra expenses.

In a contract I break down every stage of the project. I cap rounds of revisions at each stage. If the client starts pushing me past the 20% pad, or requests extra revisions, I start charging by the hour.

I generally ask for a quarter, third, or half up front. From there the contract is written with break points where I keep getting paid or I stop working. Clients don’t actually send out checks or electronic transfers overnight to keep the project going. And I don’t actually expect them to. But I want people to know up front that I’m not a piker who spends three months on a project and then begs for money after sending final files. If I wrap up stage 2 and the check for initial prototyping hasn’t cleared I just stop working until it does.

Until a check clears I don’t send a client any files other than PDFs with a huge DRAFT watermark and copyright notices on every page. My clients are ADs so they do the same thing to their clients. They know what it means and it reminds them to stay on top of accounting.

Don’t set up financing for clients. If they need to borrow money they can go to a bank. If they can’t get the money from a bank it’s because they probably won’t make their payments. Don’t feel bad for not helping people by working on the crazy terms they say they need. If they really need crazy terms from the designer they shouldn’t even be in business. And you shouldn’t help them dig their own hole.