Ian Schaefer

Recently we styled the main navigation menu for a Develisys client site. The nav was challenging from a production point of view for it combined overlapping transparent tabs or unequal widths; normal, hover, and active states; was styled on a clean, accessible unordered list in the html; and finally, we wanted it to be as fast as possible, deciding ultimately that the nav should request only a single image from the server. After some struggle, we arrived at a workable method that looked and worked like a charm in the current Firefox, Opera, and Netscape, IE 6 and 7, and degraded gracefully in the downlevel browsers. Not bad.

But it soon emerged that the near-final version of the nav was crippled and unusable on Safari 1.3 and 2.0. In Safari, anywhere you clicked on nav would reload the current page. To the designers, it became immediately apparent the anchor on the currently active tab was a large as the entire button background image, about 650px x 110px, or so, preventing the use click from targeting any of the inactive tabs. Dead navigation.

I have yet to determine if this is a bug in the Safari browser or an undefined behavior. The behavior in question occured under the following circumstances:

• absolutely positioned (+) ul
• absolutely positioned (+) li, overflow hidden
• absolutely positioned (-) anchor set to display block, with a background png image of all navigation tabs
• negative text-indent to make the original list text disappear (The Pixy Technique, if I’m not mistaken.)

I’ll cut to the chase, it turned out to be the text-indent that threw everything out of whack. Not just negative, but for any value. We decided to hide the anchor text with {display:none} instead of the negative text indent, as this works equally well across all browsers, and did not necessitate re-thinking the entire navigation styling.

Because this was an interesting case, I decided to try to generalize the trouble with Safari. I’ll be back with that soon…

Have you ever wanted to typeset and print a passage of text in a particular face, but knew you had only a limited quantity of the type? Or, have you wondered if a single 12A 25a font would be enough type to finish composing a small job? Just curious how much type it would take to print The Gettysburg Address? For the seasoned printer, the answers to these questions may be a matter of ‘horse sense’, but for students of lettepress printing like me, a little help is in order. I created Typecount to address this simple need. It is a rudimentary application written in php.

What Typecount does

Typecount counts the characters in a passage of text; but, of course, anyone can do that with a run-of-the-mill word processing application. What is unique about Typecount is that it can count characters as if they were individual pieces of metal printing type. At your discretion, typecount will consider ligatures, dipthongs, quaint characters, or any other characters that you designate as a single piece of type.

If you are printing an old style type, say Caslon with quaints, check the boxes for ct and st and Typecount will count them as psingle pieces of type. Only after it has counted these quaint ligatures will it go on to tally up the remaining c’s and s’s and t’s.

Typecount will also consider any string of characters that you designate in the “Additional Combinations” textbox. For example, I have an ample supple of Qu ligatures in my case of 14-point Monotype Garamond 156, so I would include it in my query.

Try Typecount, the character and ligature counter for letterpress

Support for the long-s is coming soon!

12/23/07 I recently learned that some visitors are receiving errors when they attempt to use Typecount. I haven’t updated it a good bit, so I will try to work on it soon. – Ian Schaefer.

Now that my website has been re-invigorated—if not beautified—I want to diversify and focus a bit in hopes of better leveraging my interests and skills. ‘Diversify and focus’ sounds somewhat contradictory, but the idea is to develop a few websites that address specific aspects of letterpress printing. By creating targeted content, each site should rank high for specific, related search terms, thereby generating highly qualified, but probably fairly light traffic. Each could become a platform for a community, for services, and for targeted advertising. Creating interest, building traffic and participation will, in turn, help to further develop the content. And so on.

The letterpress printing community already has a small number options on the Web, like Briar Press, LetPress, PPLetterpress, and the recently launched Typecasting Mailing List.

My new letterpress adventure begins with this site about letterpress type. Not much there yet, but it will grow.

It’s been quiet around here lately. I just ducked in from my other work — web design at Develisys and for the .918 Letterpress Club in Lancaster — to spruce things up. This site is now built on Textpattern. The design is not complete; I decided it was more important to redeploy it with the new back-end, rather than languish over all the design details, so that I could get back to publishing some content. Since I actually dismantled my Movable Type installation a couple of months ago, I had no publishing platform other than good old html, making the switch more or less imperative.
Both the design and the entire site framework will continue to evolve, but still, I hope that even the current state of affairs is an improvement over the previous incarnation.
One interesting aspect of the new site, for me, at least, is the use of three sections to divide and manage three, somewhat distinct, areas of interest for me: namely, web design and development , letterpress printing and the manufacture of letterpress type , and the remaining arts, sciences, and miscellaneous diversions that engage me. The flexibility of Textpattern allows me to publish articles to each section, maintaining distinctly different pages for each section, while aggregating content from each section on the homepage, and in secondary elements, like a sidebar, throughout the site. I am sure that this is all quite possible with other CMSs and blogs, including my former Movable Type, but the Textpattern approach makes sense to me.