MathematicalOrchid

TL;DR: FP addition requires finding a common denominator. FP multiplication is just multiplication. (Not 100% accurate, but that's the gist of it.)

OK.

You want to comment your answer to help the next poor soul?

I don't grok awk. Maybe someday...

It looked simple, but it seems it isn't.

Maybe this is just too complex for a shell script... (?)

Ouch! That's not fun...

There shouldn't be any other dupes.

For every line that ends :FOO, there's always a dupe on the line before.

In my actual data file, only about 4% of the lines contain a colon at all.

@enderland True. But to me, valiantly working around a problem can make it seem to management that it's not really a big deal. Sometimes you can end up inadvertently "hiding" the problem from management. Although, as you say, it seems like this isn't the issue the OP has; sounds more like management don't actually care...

All of the suggestions above seem to be about working around the problem, rather than fixing it. If anything, surely this just makes the problem even less visible to management. (?)

I stopped using Word after it ate my assignment, 16 hours before the handing deadline. From then on I wrote all my assignments in HTML. If only I discovered TeX sooner...

Really, you're not missing much.

And the default styling is quite ugly, although the latest versions improve that slightly.

It's actually really unreliable for large and/or complex documents.

@DavidCarlisle As I say, the primary purpose of Word is editing documents. It lets you immediately see what you're doing, and it makes it (too?) easy to change stuff.

All I can say is, I'll take TeX over Word any day!

Oh, right. TeX limitations. EBCDIC. History.

Wait... how did we get to arguing about XML?

@DavidCarlisle As I said, getting a machine to generate MathML should be super-easy. It's just writing it by hand which is tedious and error-prone. I didn't mean to imply the entire thing is a waste of time.

All I said - plz don't kill me - all I said was it's very hard to write by hand.

It's searchable, text-to-speach works on it, etc.

@DavidCarlisle Reading MathML is obviously a load easier than reading some ugly GIF.

For THAT purpose, I can immediately see why it would be useful.

I gather the main purpose behind MathML is to provide a common language for math tools to talk to each other.

@DavidCarlisle OK, let me rephrase: the MathML rendering in Firefix looks awful.

Every programming language has XML parsers, etc.

@DavidCarlisle I did see one guy ask me why they don't just use TeX. The problem being, the only thing that can parse TeX is... well... TeX. By making it all XML, you can transform it with XSLT and do all sorts of other stuff

@Johannes_B Typing an equation by hand is one thing. Typing the entire parse tree by hand is quite another.

I wouldn't mind, but it looks awful too. ;-)

@DavidCarlisle That's proper hardcore, right there. I thought browsing the Internet with Telnet was hard, but writing MathML by hand was just painful.

Then again, that's not its purpose, so hardly surprising.

@DavidCarlisle Oh God, MathML... Nobody wants to write that by hand, ever.

Ah yes, XML... You just write stuff in angle brackets and it works. If only it were actually that simple... Come to think of it, given the choice between getting arbitrary XML to work and getting TeX to work... TeX is at least documented better.

@DavidCarlisle The scary thing is... you're almost certainly correct about that.

(I once had a go at reading the source code to see if I could extract the formatting engine... You can guess the answer.)

I particularly enjoy how the TeX source code checks whether your mainframe is ASCII or EBCDIC...

Also fun: DVI preview requires all your figures as JPEG, PDF output requires them all as PNG. Or is that the other way around? DAMN >_<

@DavidCarlisle As I say, back when TeX was invented, such things weren't an issue. The packages that exist do an admirable job of working around the limitations of TeX, but that's not the same as saying TeX isn't limited.

@DavidCarlisle It was a while ago now, but IIRC with DVI the colour stopped at the page break, but with pdflatex the entire document came out green somehow.

@DavidCarlisle It broke for me. In particular, it broke one way with DVI preview, and a different way for final PDF output.

I didn't think HTML even does hyphonation... (Another plus for TeX!)

To be clear: I use LaTeX A LOT. I like it. I just sometimes wish it could be a bit less awkward, that's all.

It works, just not ALL the time.

Let me put it this way: I couldn't get it to work reliably.

I repeat: TeX does not understand colour. There are packages that make TeX spit out specials that the driver on the other side understands (you hope!) to kludge around the problem, but this is not the same as "TeX understanding colour". For example: my paragraph is green, and then a page break... oh, the rest of the paragraph is black again. Oops! Because TeX doesn't know what you did to the last page; only the package knows that. Like I said, you can do it, but it's not very reliable.

@Johannes_B If I want to change the font of an HTML page, I just write 1 line of CSS to say what font I want. If I want to change the font of my TeX document, I have to take the entire LaTeX page rendering engine apart and monkey-patch the code to change its behavior...

@percusse Can you be a bit more specific about which packages you feel contradict my list?

@Johannes_B Certainly lots of people have written useful packages. But you won't find a package that changes every single style detail in the way that you want for your particular project. It's times like when I wish I could just CSS the thing...