September 2009 - gutvol-d - lists.pglaf.org

re: In search of a more-vanilla vanilla TXT
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

jim said: > Which is a "good" submission format for submitting a > transcription of Shakespeare to be read on a cellphone, > PG txt format, or HTML? > Answer: Neither file format works worth a dang for > specifying Shakespeare to be read on cellphones. > Yet both file formats contain the lists of the words. again, jim, you've come up with the wrong answer... as i've said, repeatedly, the iphone app "eucalyptus" does a great job of rendering the p.g. e-texts in a way that makes them quite beautiful, according to reviewers, and i agree, for the most part. (it ends up eucalyptus is kind of flawed as an e-book program when it comes to some capabilities that i consider vital, such as _search_. but in terms of rendering the pages, it does that nicely.) format wonks think the format needs to describe beauty... it's far better, however, for a viewer-application to elicit it. because in the end, everything really depends on the viewer. > Here's a brief excerpt from Dove: > "'Go?'" he wondered. "Go when, go where?" well, that's on page 227 of this version in google. > "And proceed to my business under your eyes?" > "Oh dear no -- we shall go." > "'Go?'" he wondered. "Go when, go where?" > " In a day or two -- straight home. Aunt Maud wishes it now." http://books.google.com/books?id=B9AOAAAAIAAJ&client=safari&pg=PA227& ci=89%2C537%2C776%2C177&source=bookclip" http://books.google.com/books?id=B9AOAAAAIAAJ&pg=PA227&img=1&zoom=3&hl=en& sig=ACfU3U0L1tXMRlU83MbayfHNmmTl27CrXw&ci=89%2C537%2C776%2C177&edge=0"/ but i don't see anything special about that, anything that would be missed or lost in the plain-text version. > And another one: > She particularly likes you. ok, that's on page 18. the "you" is italicized, so you should have put underscores around it, so the viewer knows it's emphasized. it's also embedded in dialog, so the way you have pulled it out -- as if it was a paragraph by itself -- is rather misleading here. again, if you purposely disfigure the plain-text version, then yes, it will be inferior. your solution is to stop disfiguring it... -bowerbird

2 1

Re: z.m.l. can do what you want
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

gardner said: > I know BB talks a good deal about ZML > and it sounds pretty cool wow, that almost sounds like it could be a compliment. > but I've googled till my fingers bleed and > I cannot tell for sure what exact ZML he has in mind. oh, you just should have asked me, gardner. google can't find everything, especially if something ain't all that concerned with being found by google... on the other hand, it's pretty easy to remember it: > http://z-m-l.com and oh yeah, that stands for "zen markup language". *** at the site, click "see some examples." which takes you to: > http://www.z-m-l.com/go/vl3.pl you've come to a page that has filename links on the left, and then a respective button for each one of those files... let's say you click the link for the top file -- test-suite.zml: > http://www.z-m-l.com/go/test-suite.zml make sure to click the text link -- don't click the button yet. the file you're viewing -- the test-suite for project gutenberg -- has been "marked up" with z.m.l. -- zen markup language. (z.m.l. is "zen", so you won't actually "see" any markup, at least not any anglebrackets with tags inside of them. but you'll see that it's formatted regularly; that's z.m.l.) i've appended the topmost lines from that file -- the cover page and some lines from the contents -- to this post. once again, this is the .zml file... *** now click the "back" button to go back to this page: > http://www.z-m-l.com/go/vl3.pl this time, click on the "test-suite" button on the right side. this will perform the conversion of the .zml file into .html, and take you to the resultant .html file right on the web... text colors are used to signify different structural elements. you can save this .html to your own machine to examine it. or put it side-by-side with the .zml file to see the conversion. this .html, with its c.s.s., certainly won't be your cup-of-tea, but changing the c.s.s. is a mere matter of template editing. it validates as is, to .html 4.01, which might work for kindle, or might not, i haven't done any checking on that specifically, but again, we modify the conversion by editing the template. i pointed you to the test-suite first, because reading it will give you an introduction to the overall philosophy of it all... also, the second file is "the 11 rules of z.m.l.", which might also give you a good orientation, even if the file is quite old. (but as i'm reluctant to tie anything down, it's not outdated.) i can also turn out a mean .pdf, with that conversion routine, and since you can size the .pdf page (and all other variables, like font, fontsize, leading, margins, etc.) however you want, that ends up being quite usable on a fixed-size e-ink screen. so we have .html (important in mounting a web version), and .pdf (for those situations where it's the user-chosen solution), but the best part of all is that zml-viewers are easy to program. i've coded them in basic and perl, and python should be simple. it's very fundamental coding, so it'll be portable to any language. it also facilitates open-source efforts, because it's easy to hack... zml-viewer-programs turn the .zml file into a beautiful e-book, customized to the user's preference, and offering a wide variety of high-functionality capabilities that make a .zml file powerful. this combination of beauty and power make z.m.l. hard to beat. -bowerbird p.s. the top of the test-suite file goes like this: the test-suite for project gutenberg a document containing the full range of features found in project gutenberg e-texts by bowerbird intelligentleman greetings, earthling... this is an e-text brought to you by project gutenberg, a 35-year-old volunteer effort to put literature online. please see the web-site for news and information on usage conditions for e-texts, volunteering, and more... http://www.gutenberg.org table of contents the test-suite for project gutenberg table of contents dedication chapter 1 -- welcome aboard chapter 2 -- the sections of the book chapter 3 -- text "styling"

2 1

Re: In search of a more-vanilla vanilla TXT
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

jim said: > In the case of this book, the answer was more than 1000 "losses" -- > or an average of about 3 losses per page. And this is NOT counting > about an addition 1000 losses in representation of emphasis. what was the book? i'd like to compare the versions myself. the question is, "why are you having _any_ losses in the .txt files?" the answer, i am sure, will once again be, "you're doing it wrong". it sucks, yeah, but it will be important to fix your broken workflow. seriously, if you are stripping out emphasis, you're making a mistake. (unless the "emphasis" had no meaning, and was simply ornate decor.) in the meantime, if you do things intentionally that harm the .txt file, then yes, the .txt file is going to seem awfully incapable to you, so it's not a big surprise that you keep wanting to insist that such is the case. -bowerbird

2 1

Re: In search of a more-vanilla vanilla TXT
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

jim said: > Now maybe to some of you -- you consider this result > to be a good thing, an acceptable thing, a thing that > well-represents the considerable efforts of the PG volunteers. i doubt there is anyone who thinks that. what you ended up with is pure shit. that's because you did it wrong. you rewrapped lines that weren't supposed to be rewrapped. if you would have done it correctly, it would've come out right. but you did it wrong. now, your point is probably that we should make it easier for our users to do it correctly. and nobody will disagree... we _should_ make it easier for our users to do it correctly. and there's an (awfully) easy way to make it easier, which is to mark all lines that should not be rewrapped with leading spaces. but the whitewashers won't do it. why won't the whitewashers do it? i dunno. you'll have to ask them. i've certainly asked them. i've asked them to do it, pretty please. i've asked them again to do it, pretty please. i've asked 'em why they haven't done it. i've said, repeatedly, that i think it's stupid they haven't done it. and they still don't do it. not all the time, anyway. they do it some of the time. i consider that a very slight victory. -bowerbird

2 1

Re: In search of a more-vanilla vanilla TXT
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

jim said: > Again, TEI P5 is 1350 pages, which is a lot more inaccessible > to volunteers than anything describing HTML tags that I have seen! > I'd say the DP tagging documentation is already painful enough > for most of us. I am about 100 pages into the TEI documentation, so > in maybe two weeks I can tell you more about what I think about it... wow, jim looks to be a bit masochistic. could be a perfect candidate. :+) -bowerbird

2 1

Re: In search of a more-vanilla vanilla TXT
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

carlo said: > The only route to have more TEI submissions is > to have a version of guiguts producing TEI instead of HTML. yeah, right. extend the rickety workflow right out the window. that's the ticket. that will make everything all better, for sure... and thus again we learn to appreciate the open-source approach. > And of course improve the automatic conversion from TEI to HTML and the automatic .pdf conversion. and all the other conversions. or, you know, just use all the "standard" routines already out there, in this thoroughly-explored and well-documented standards arena. or hey, maybe once we get an e-book into .tei form, we can just stand back and admire it, and pat our backs on our achievement. who needs to convert to .html, when we can just bask in our glory? -bowerbird

1 0

z.m.l. can do what you want
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

z.m.l. can do what you want. as soon as you're ready to act, and not just yak. -bowerbird

2 1

Re: In search of a more-vanilla vanilla TXT
by Bowerbird＠aol.com 16 Sep '09

16 Sep '09

jim, i don't think you're saying anything that hasn't already been said here before. many times. over the course of years. i just think you're saying it less clearly... but hey, if anyone thinks jim _is_ saying something that can use some attention, please do tell us just exactly what it is... thanks. -bowerbird

1 0

World's most heavily pirated books
by Greg M. Johnson 15 Sep '09

15 Sep '09

I was listening to some podcast (I have since forgotten exactly which one, but it was about the celebration of sci-fi culture) where they talked about "the world's most pirated books." At first one of the hosts went into a little tirade that *book *piracy was the purest form of evil-- worse than any other, I guess because the book publishers weren't as evil or something. The list did include a recent book about Photoshop, which is unfortunate. But the list also included *The Kama Sutra*, and a few other really old classics. Eventually it dawned on me, if not entirely to both hosts as well, that they were conflating "piracy" with "downloading by bit torrent." < s i g h >. -- Greg M. Johnson http://pterandon.blogspot.com

1 0

Line Art
by James Cloos 14 Sep '09

14 Sep '09

I see that the HTML versions of a number of PG works include images of art from the original books. Are higher-resolution scans of those images available anywhere? I'd like to experiment with automated vectorization of scans of line art, and providing SVG format vector files of the line art back to PG would make the effort doubly useful. (Encapsulated PS and PDF files also could be made available, if desired.) -JimC -- James Cloos <cloos(a)jhcloos.com> OpenPGP: 1024D/ED7DAEA6

2 1