It’s no news that getting maths printed correctly isn’t exactly easy. That’s why Knuth came up with TeX, which gave the maths (and physics) communities better typography than they care about. Back then in the 1970s, when TeX was conceived, book publishing was in a horrible situation. Somehow doing proper typesetting had become too expensive too actually do it – particularly in the non-trivial areas where people might require the use of formulae – and DIY digital typesetting didn’t seem possible, if only because anything digital was still pretty new and – apart from AI people didn’t know what to do with it.
One of my favourite examples for this is the nice book Complex Manifolds without Potential Theory by Shiing-Shen Chern. (Who isn’t just famous because of Chern classes but was also a founder of the MSRI. He died last year and I read that there were more than 10000 people at his funeral in China.) I first worked with the properly typeset 1967 edition which was fine. But the next library I saw the book in only had the newer 1979 edition (which is the same as the current 1995 edition): Definitely a case where newer doesn’t mean better as apart from having been acquired by Springer, that new edition turned the previously nice book into a typewritten mess.
At that point, Knuth saw the potential of using the new technology to improve the problem of typesetting. Thankfully, he was educated, taste- and skillful enough to cook up a typesetting system that was very powerful. It may not be perfect – as he seems to be the first to admit – but in terms of the typographical quality of its output even most of the current software will have trouble to match TeX.
When discussion come to text processing, I am always in favour of TeX. It hasn’t got as much eye-candy as TextEdit or Word but it will produce better results. The only other difference to those applications is that TeX has very different learning curve.
Different is all I am prepared to say here. It’s not harder in total. It’s just harder to make the first few steps. But once you got those you won’t find many of the things that follow too troublesome.
The more graphical applications on the other hand make it ridiculously easy to get a first few steps done. Even little kids can do it. And as this is so easy, people play around a lot and get all sorts of bad habits which are very hard to break once they’ll have to do serious work which may need to be shared with others. At that stage they’d need to put a lot of work and learning into improving their text processing skills and the effort needed will be on the same level as for TeX. Of course, many people just don’t do that work they’d need to do and thus still ‘save’ time. And waste the time of any other people who may have to work with – or even edit – their documents later on.
Interesting elements in the text processing food-chain are secretaries. Entering text is what they’re good at. And, if they’re worth their money, their skills at handling different word processing tools are quite good. They’ll be able to tame anything from TeX to StarOffice if they had the time to look at them. Of course it will be mostly secretaries in maths and physics department who can handle tools like TeX.
But even if they can’t – unlike most other people – they’re mostly able to just adapt a new technique of typing and follow a few rules rather quickly. Back when I was at school, I once did the school’s yearbook. In TeX. It’s not the perfect tool for the job, but as we didn’t have any PageMaker or XPress, it seemed more capable than ClarisWorks, say. In addition I had done a very similar layout for our own school magazine before, so this was rather easy. The texts for the yearbook were typed by one of the secretaries who didn’t know any TeX. And despite those being the times where it was still common to type umlauts as
"a, the texts came in without any problems.
I saw a more tricky case recently. One of our professors had the idea to produce a nicely TeXed version of some nice lecture notes that a guest lecturer had made thirty years ago. He likes recommending those notes but people are often put off by the fact that they’re the typewritten mess that’s typical for the time. A secretary was asked to type those up in TeX and then sections of that version were distributed among the PhD students for the glamourous task of proof-reading.
While doing my part of that job, I was impressed that even in areas where there were many formulae, there weren’t too many mistakes. Getting all those latin and greek letters right without them being too meaningful to yourself must take a great deal of concentration. And when there were errors, they were mostly due to subtleties of notation, mostly for whether certain letters were to be in italics or not – an aspect of notation that was completely lost in the typewritten version.
At the end of the day, what makes typing maths hard isn’t that you’re using a tool like TeX that looks complicated. It’s that there’s quite a lot of convention in the the notation which you have to get right or at least handle consistently. This can make it quite painful for non-mathematicians to deal with such texts and makes it equally painful for mathematicians to let other people deal with their texts – which may be (apart from financial problems which mean that there’s less support staff anyway) the reason why most people are typing everything themselves these days.
Received data seems to be invalid. The wanted file does probably not exist or the guys at last.fm changed something.