Live data from Hacker News

A Computer Scientist Tells Mathematicians How to Write Proofs

blogs.scientificamerican.com

41–50 of 103 posts

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#41
This is how I learned proofs:

http://i.imgur.com/omXqvos.png

Well, technically I learned proofs from Euclid first. But this book followed close behind.

Clear and comprehensible (once you learn the basic symbols). Very much structured. There's hardly any prose in the whole two volumes.

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#42
post #22

The article starts with a false equivalence between archaic verbose language and mathematical notation. For my part, I usually prefer the prose. Formulas are easier to read for people who are used to and have practice with mathematical notation , but even then they are easier to read for those people when they are sufficiently complete , and formulas seem to often come with huge leaps and unstated assumptions. Especi…

The archaic prose bit is there to point out that mathematical notation has evolved, in order to setup a presentation of a possible future evolution, namely hierarchical structure for proof parts. The article is not about prose vs. symbolic notation.

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#43

This is how I learned proofs: http://i.imgur.com/omXqvos.png Well, technically I learned proofs from Euclid first. But this book followed close behind. Clear and comprehensible (once you learn the basic symbols). Very much structured. There's hardly any prose in the whole two volumes.

Care to share the title?

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#44
Since there seems to be some confusion: the article is not about prose vs. symbolic notation in proofs, or even degree of rigor in proofs; it introduces a potential way of evolving our presentation of proofs, so that their parts are hierarchically arranged. This better facilitates reading at variable levels of detail, promotes reuse of parts, and provides useful hints guiding comprehension generally.

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#45
As noted in some comments, writing a non-trivial proof by hand with TLA+ only seems possible if there exists a good repository of non-trivial propositions that one can start from.

Does anyone know of interesting initiatives out there to build an open repository of mathematical proofs?

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#46

> He says that once you have a hierarchical proof, you can use pieces of it in new proofs very easily. For example, if you want to change one assumption slightly or focus on solutions that have different properties, you will be able to tell very easily which parts of the proof can stay the same and which parts will need to change. If only that were true of a piece of LaTeX moved from one place to another, haha!

If you move a piece of LaTeX from your prose to another proof, there's nothing at all there to help you ensure you get it right. I think the point here is this hierarchical system allows you to very easily discover what parts of the proof depend on what other parts, so if you make a change to one part, you can "invalidate" everything that depended on it easily. With prose, you have no assistance in figuring out what pieces of the proof need to be updated for any given change.

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#47
post #20

Earlier quoted context omitted.

The reason for using different scripts (Greek etc.) is to sort symbols by broad categories. Done right, it actually makes things more readable (to the sophisticated reader, which is the mathematician's target audience anyway). It's almost like an implicit type system: (within a particular context) Greek letters mean ordinals, lowercase Roman letters mean predicate logic variables, uppercase German letters mean univer…

Yes that's the pre-computer I was talking about. Sophisticated mathematicians should be concerned about if it is machine readable or not. If it's machine readable then you can use whatever font, color, or any other style you want that helps you to read it. The viewer can abbreviate descriptive names to single glyphs if you want. Programming is much less fun, and harder to follow, without syntax highlighting. But it d…

> Programming is much less fun, and harder to follow, without syntax highlighting.

That's your opinion.

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#48
post #45

As noted in some comments, writing a non-trivial proof by hand with TLA+ only seems possible if there exists a good repository of non-trivial propositions that one can start from. Does anyone know of interesting initiatives out there to build an open repository of mathematical proofs?

Something like this?

http://us.metamath.org/mpegif/mmset.html

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#49
post #48
post #45

As noted in some comments, writing a non-trivial proof by hand with TLA+ only seems possible if there exists a good repository of non-trivial propositions that one can start from. Does anyone know of interesting initiatives out there to build an open repository of mathematical proofs?

Something like this? http://us.metamath.org/mpegif/mmset.html

Great link, thanks!

Re: A Computer Scientist Tells Mathematicians How to Write Proofs

#50

This is how I learned proofs: http://i.imgur.com/omXqvos.png Well, technically I learned proofs from Euclid first. But this book followed close behind. Clear and comprehensible (once you learn the basic symbols). Very much structured. There's hardly any prose in the whole two volumes.

Care to share the title?

Principia Mathematica

http://en.wikipedia.org/wiki/Principia_Mathematica

Post reply on HN