Perl 6 - the future is here, just unevenly distributed

IRC log for #marpa, 2015-10-07

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary

All times shown according to UTC.

Time Nick Message
00:27 Idiosyncrat I got some advice re amsbook via private email ...
00:27 Idiosyncrat and have just converted.
00:28 Idiosyncrat I was frankly reluctant to concede that the "paper" is now a small monograph, but so be it.
00:28 Idiosyncrat Traditional math publishing has been aimed at dead-tree journal, where space was at a premium --
00:30 Idiosyncrat math can be compressed and naturally editors given the choice of accepting 12 papers or 10 will make the choice that prevents what might be a historic result from never seeing the light of day.
00:31 Idiosyncrat So, for example, in a traditional math article, if you can simply refer to an already published result, you do so, rather than take the space to spell it out.
00:32 Idiosyncrat This meant readers had to be familiar already with much of the literature in the field (many of them were), or prepared to go to other references a lot (most had access to a university library)
00:33 Idiosyncrat I'm taking the approach that if I can gain readability by using more space, I'll go for the readability -- the Internet has changed the tradeoffs completely.
01:05 Aria That makes me so happy.
01:05 Aria Because math is so completely inscrutable as usually written
01:32 Idiosyncrat Aria: I've been working on finding new conventions that make the math more readable, and I hope they work.
01:33 Aria That would make me so happy if you discover some.
01:33 Idiosyncrat already in the current Theory paper, I scrapped single letter variables in favor of names
01:34 Idiosyncrat I had a very hard time with sentential forms written as strings of single letter variables, mixed Greek & Latin
01:34 Idiosyncrat As in, like quick, which is epsilon and which is eta?
01:35 Idiosyncrat And then when the single letters aren't enough, they use sub/superscripts ...
01:35 Idiosyncrat when the main font is already 10pt type.
01:36 Idiosyncrat They say most mathematicians are unproductive by age 40, and the reason is the synapses slow down ...
01:36 Idiosyncrat but I wonder if it isn't eyesight. :-)
01:36 hardmath123 I would very strongly support semantically sensible variable names in math literature.
01:37 Idiosyncrat So already I went with 12pt type, suggestive names, and no Greek.
01:39 Aria That right there eliminates a significant part of what I struggle with reading.
03:26 Idiosyncrat joined #marpa
05:06 Idiosyncrat Good night!
05:26 ronsavage joined #marpa
08:37 ronsavage joined #marpa
14:37 Idiosyncrat joined #marpa
17:41 Idiosyncrat joined #marpa
17:49 Idiosyncrat joined #marpa
17:53 Idiosyncrat http://nlp.stanford.edu/~manning/papers/CICLing2011-manning-tagging.pdf
17:53 Idiosyncrat A paper on the current state of the art for Penn tagging.
17:55 Idiosyncrat "Penn tagging", for those who haven't heard of it, is the modern academic version of "sentence diagramming" -- identifying parts of speech in a sentence.
21:46 ronsavage joined #marpa
22:35 Idiosyncrat joined #marpa
23:24 Idiosyncrat Other changes I make in my math notation -- almost no use of italics
23:25 Idiosyncrat The AMS and TeX styles use italics for variables and for theorems
23:25 Idiosyncrat The variables usage is yet another thing that makes sentential forms hard to read in traditional parsing papers.
23:26 Idiosyncrat And most italic fonts are annoying to read long passages in.
23:26 Idiosyncrat Finally, contrary to what most typographers seem to believe italic does *not
23:27 Idiosyncrat *not*
23:27 Idiosyncrat add emphasis.  ...
23:27 Idiosyncrat It makes you more inclined to skip over the text.
23:28 Idiosyncrat I find the italic font is useful only in very special circumstances, where you want to set off very short bits of text ...
23:29 Idiosyncrat in a way that does not add emphasis.
23:30 Idiosyncrat Btw, the comments about italic font are based on *most* italic fonts --
23:31 Idiosyncrat I have seen *some* italic fonts that would work for emphasis.
23:32 Idiosyncrat But most italic fonts, the common TeX fonts included, do not work for emphasis, and do not work for text of any length.

| Channels | #marpa index | Today | | Search | Google Search | Plain-Text | summary