History

This page and associated pages describe some of the history of TeXShop. Users new to TeX should skip this material and get on with the task of learning TeX and LaTeX, or using the program.

TeXShop has been around for 24 years; development started even before the official release of macOS in 2001. The many dramatic changes in the development of the modern Mac had a profound influence on the program. Several stories are woven together in the history below: the role of open source software and contributors from all over the world, the significance of Object Oriented programming and Objective C, Apple's two dramatic processor transitions to Intel and then to Arm, the transition from 32 bit machines to 64 bit machines, memory management and automatic reference counting, and the solution by Apple of the "fragile base class problem" in object oriented programming. Some of these stories are well-known and some are not. In particular, the last item is crucial and yet most Macintosh fans seem unaware of it.

In October, 1988, the NeXt computer was released, originally only for education. The University of Oregon decided not to carry the computer, but our faculty could obtain it at Oregon State. By that time, I had been at Oregon for twenty years, and I thought I deserved a “reward”. My goal was to have fun giving NeXt demos to friends, but in fact, I never had a successful demo. Macintosh friends claimed that the machine did nothing that the Mac didn’t do. “It doesn’t crash,” I’d say sheepishly. Unix friends would open a Terminal window to full screen, type like the devil, and then claim that curses.h was out of date.

The NeXT was the opposite of flashy, but it couldn’t be beat for everybody work, at least in mathematics. My colleagues struggled with typesetting jobs taking two minutes to typeset ten pages, while my machine typeset 100 pages in seconds. It had LaTeX, Mathematica, and internet access. What more did a mathematician need?

Meanwhile, Apple was in serious trouble. In December of 1996, I had just finished grading my classes when I heard a rumor that Apple would adopt NeXtStep as their future operating system. But I had seen IBM, and then Sun, license the software and then do nothing with it, so I didn’t get excited. After a couple days, it transpired that Apple bought NeXT lock, stock, and barrel, and Steve Jobs was returning as an unpaid adviser. Then I got excited.

There were struggles ahead. The University of Oregon Provost announced that since Apple was failing, departments could only buy Macs with his special approval. For the only time in my career, I sent an angry response, which he did not answer. There were no repercussions for that letter, but happily in the next decade Steve Jobs created the greatest second act in American business history. My advice is to write angry letters only when you are very, very, very, very sure of your position.

As soon as betas of OS X appeared, I switched to using them for my work. Someone had compiled teTeX for the beta macOS, so I could write source in TextEdit and typeset from the Terminal. I discovered pdfTeX, then at version 0.14, but still, it worked. So a dvi display wasn’t necessary and I could preview with Apple's Preview. This gave a very primitive TeX system.

I kept pestering our local Apple representative for real TeX and for Mathematica. He assured me that Apple understood how important both were. But I didn't see any progress. Finally one day I thought “you wrote a few programs using OpenStep, and you wrote a commercial 3D program for high school teachers with object Pascal. It ought to be possible to write a front end for TeX on the new Mac with Cocoa. Cocoa has NSText and NSImage and NSTask. Why, the program almost writes itself.”

I began writing the program in the early spring of 2000. The 2000 Worldwide Developer Conference, WWDC, was in San Jose in May. At this conference, Apple was slated to give developers the release version of OS X, which would be sold to users several months later. I was only teaching one course that spring, a Discrete Mathematics course for Computer Science students. The authors of the book were also at the University of Oregon, so I asked them to teach my class for a week and went to the conference. Before leaving, I bought a 17" PPC portable, erased the hard disk, and partitioned it so I could install macOS at the conference.

In the conference keynote address, Steve Jobs said that sometimes in the computer industry, there were name changes for marketing reasons. He then announced that the release version of OS X would instead be called OS X Public Beta and would be sold for a $15 handling fee rather than costing $125. After the keynote, a friend said to me "Wasn’t that smooth? Jobs just announced that OS X is delayed a year." My friend was right.

By this time, TeXShop was definitely working on beta versions of macOS, but the Apple PDF software had a significant bug: the software could not read embedded fonts in files. Consequently, TeX documents written with Times Roman displayed beautifully, but any mathematics in the document would appear as blank areas. I hurried back to the motel at the end of the Monday session, installed the new beta, and ran TeXShop. Embedded fonts still did not display.

Soon after the conference, I began releasing preliminary versions of TeXShop on the internet. Meanwhile, Apple was working on macOS, and in March, 2001 they announced the first non-beta release. This time, developers got their copy only a week before the official release. When I got my copy, I immediately tried TeXShop. Apple had fixed the embedded font problem and TeXShop could typeset LaTeX code containing mathematical equations. The future had arrived.