February 2010 - gutvol-d - lists.pglaf.org

Re: d.p.'s undeserved superior attitude about "quality"
by Bowerbird＠aol.com 25 Feb '10

25 Feb '10

carel said: > You don't need to talk to me about having your time wasted. > I am the somewhat infamous Carel that left DP nearly a decade ago > (after spending around 500 hours creating a front-end and > revamping the horrific code that resided on the back-end). oh, hey, glad to meet you. :+) i do believe i am the one who has grabbed the "infamous" crown at this point in time, however. i see they still let you post there. myself, i've been banished, so as not to corrupt the youth there. > It is nice that DP strives for quality in the work that they do. yes, it is. :+) and it's nice that vancouver is finally getting some snow... ;+) > I am not however surprised that > they ultimately became backlogged. well, yeah, quite a few people saw that coming... when you have 3 rounds of proofing, and you want all of your material to go through all 3 rounds, and you have thousands of proofers in p1, and hundreds in p2, and only dozens in p3, a backlog is pretty inevitable... especially if you expect your p3 people to take _more_ time on each page than your p1 proofers. it just will not add up... > DP has managed to produce a huge number of etexts and they could have produced 3 times as many if their workflow were 3 times more efficient, something it could _easily_ attain... according to brewster, internet archive scans 1,000 books a day. (google probably scans twice that many before their lunch break.) d.p. digitizes 2400 books a _year_. i predict they cannot keep up. > and for that their efforts should be appreciated. i appreciate the proofers who volunteer their time _immensely_... and the people who format, and postprocess. they're champions. but the "powers that be", who _refuse_ to change a bad workflow, and clamp down on dissent that reveals why that workflow is bad? i have no respect for them at all. not even one shred of respect... > Their etexts also have a decent level of quality. i'm with michael hart on this. (even if he refuses to see that.) i think the first mission is to get text up with acceptable quality. (which, by the way, most of the in-process stuff at d.p. is _not_) and _then_ we can use the _public_ to drive it toward perfection. (and no, i don't get the impression that that's what the current "pre-release" plans will entail; seems to me it's "post and forget", which is the attitude that p.g. takes towards _all_ of its e-texts.) > If the volunteers at DP are blissfully unaware that their time > is being wasted then they do not feel like victims of the process but they _are_ victims, whether they _feel_like_ victims or not... and because they are being used so inefficiently, they are _not_ receiving anything near the satisfaction they _could_ be getting. and, like i said, despite a constant influx of volunteers sent there by p.g., the steady d.p. base seems to be a fairly constant number, which means they are experiencing severe "churn" and "burnout", neither of which bodes well for the future. the well _will_ go dry. (especially since more and more people will come to see google as "the source" of the books that they try to find online, since google offers people several million more titles than project gutenberg.) > If the volunteers at DP are blissfully unaware that their time > is being wasted then they do not feel like victims of the process, > but rather champions of the cause. well, they _are_ champions. too bad they're not treated as such. > So, efficient or not, no harm is being done well, as i just argued, harm _is_ being done. it most certainly is. > and books are being produced. some. not nearly enough. and not nearly efficiently enough. > If the inefficiency bothers you and me (and others), > it's a moot point because it is not our time that is being wasted. but the potential of our society _is_ being spent inefficiently, and the time of _good_people_ is being wasted, for no good reason. so i think the point is _not_ moot. i believe i should _speak_up_. > And, we have no power to change what is at DP. > At least that is the way I look at it. i'm not looking for "power". i'm looking to enlighten people. -bowerbird

1 0

Re: ok, let's take a look at gardner's book, just for the exercise
by Bowerbird＠aol.com 25 Feb '10

25 Feb '10

gardner said: > Abbyy 5.0. yeah, i could tell it was an old one. (he/be confusion was the sign.) i think most people say abby v7 is where recognition hit a plateau. (versions later than that have added features, yet no better o.c.r., while all of the versions earlier than v7 are decidedly inferior.) if you can find a used copy of v7, it would help you immensely... -bowerbird

1 0

Re: ok, let's take a look at gardner's book, just for the exercise
by Bowerbird＠aol.com 25 Feb '10

25 Feb '10

let me now _contrast_ the undeserved superior attitude that some d.p. members manifest about their product with the _humble_ nature of the request gardner made, where he _invited_ me to analyze one of his books and report back to him how he could improve his workflow. my general attitude here is to try to be of _assistance_. but so much of the time, i am vilified, and the nature of the interaction turns confrontational and viscous. so it's nice when someone like gardner comes along, and reminds us all that it doesn't have to be that way, that we can talk pleasantly about how to do better... *** you might recall that i found 159 differences between the archive.org o.c.r. and gardner's proofed version... > http://z-m-l.com/go/gardn/gardn-159diffs.html i've resolved those differences, and found that they split almost exactly down the middle between cases where the o.c.r. was wrong, and gardner was wrong. this was an _intermediate_ version of gardner's file -- the last one he had before he rewrapped lines, which he generously gave to me to save me time -- and he likely made some corrections after that act, plus the whitewashers might've fixed some errors. so i'm going to now test my diff-resolved version against the file that was actually posted up at p.g. what i _can_ say now, after resolving the diffs, is that gardner's o.c.r. program didn't do very well. ("very well" sounds so stilted; i prefer "very good".) whether this is because the scans weren't good, i cannot say... gardner, what app/version is it? in particular, there was a problem with he/be diffs, and other h/b glitches, as well as with punctuation (in general). so i'd want to track down that aspect... and gardner, next time you might want to hold off on the rewrapping until later in your workflow, as that'll make it easier to compare against the scans. i should have a final report by tomorrow or friday. -bowerbird

2 1

Re: Translation project
by cmiske＠ashzfall.com 25 Feb '10

25 Feb '10

>-------- Original Message -------- >Subject: [gutvol-d] Re: Translation project >From: "Keith J. Schultz" <schultzk(a)uni-trier.de> >Date: Wed, February 24, 2010 4:28 pm >To: Project Gutenberg Volunteer Discussion <gutvol-d(a)lists.pglaf.org> >Hi Carel, >What you are forgetting that every language has its >cultural idiosyncrasies and idioms. You can not take >a pre-school text book and simply translate it. >Furthermore, there are also cultural differences in >teaching. I deal with people of a variety of cultures and languages on a daily basis. Problems and frequent episodes of humor occur as a result of these differences. I also realize the existance of idiosyncrasies and idioms or I would have thought that google translate was the answer rather than human translations. I have instructors from many different cultures and they openly discuss and often use the teaching methods of their culture of origin. So, that is something I would have a hard time forgetting as I have several decades of personal experience with it (I'm wht is known as a 'life-student' or 'professional student' --depending on how rude you feel like being). [snip] >Furthermore, a German normallydoes call their dogs >Spot. Or translate Dick and Jane. Well I could start >with "Every dick and Jane" being Hinz und Kunze, but >Hinz und Kunze are last names. Exactly. Again, the reason google translate is not adequate. As books intended to help a native speaker of one language gain some familiarity with the sentence structure of another (by reading books they are familiar with from their own language), I fail to see how most of this applies in any real depth. The individuals translating the books would have to decide how best to deal with the issues of how "See Spot" should best be phrased in another language: I believe that is the role and purpose of a translator is it not? And, I've rarely seen names changed in translation as that would remove a large portion of the cultural of origin. The purpose and nature of what I am looking for is why I doubted that PG would be interested in the end result: It would not be an 'authoritative' translation, but rather a learning aid. >It does not work that way. Sorry. I'll be sure to let the publishers of books that do exactly what I am describing know that they are doing the impossible. Since my question was actually whether or not such a translation project exists, I will take your response as the simple 'no' that all of this can be translated to mean. I'll search around a bit more for an existing project and then start one of my own. If it serves no other purpose than to help me (and some of my fellow students) to get a firmer grasp on a few languages and to meet a few interesting people who are willing to help out, then the project will be a success. >regards >Keith. Yes, thank you.... Carel

1 0

Re: d.p.'s undeserved superior attitude about "quality"
by Bowerbird＠aol.com 25 Feb '10

25 Feb '10

carel said: > The fact that they even strive for quality at DP is a mark in their favor. not really. everyone is gonna tell you they strive for quality... (trust me, nobody's gonna say, "quality doesn't matter to us.") besides... it depends on how you go about trying to attain that goal. and d.p. has a lot of problems -- a lot -- in that respect... but it's not the notion that d.p. "strives" for quality that's irritating; it is their blind stubbornness that they have already _attained_ it, coupled with an unwillingness to measure the price they pay for it. let's take the first item first -- the idea that d.p. quality is _great_. this has been a staple of the propaganda served on the d.p. forums for _years_ now. equally troubling is their accompanying attitude that solo producers create shitty e-texts. if you want, i can point to some listserve posts from years back where this led to some highly embarrassing interactions for d.p., where one solo producer showed just exactly how _crappy_ some d.p. e-texts were, and dared d.p. to uncover any errors in the work that he had done. they couldn't do it. that led to a big shakeup at d.p., as d.p. got spooked about "quality", and started piling on the rounds to make sure they got things right. it also resulted in a slight reworking of the standard propaganda, where "the early d.p. e-texts had flaws, but the later ones are great." this is even better, because it's even more insidious towards rebuff, and you see this superior attitude expressed regularly on the forums. now, i do believe d.p. quality is better these days. but now they are paying a _very_ high price for it, in terms of the time and energy used. when d.p. piled on the rounds, and made people "earn" a p3 badge, they significantly increased the time and energy used, and they also bought themselves backlogs that are threatening to strangle them... which means that in spite of a constant influx of people (thanks to that banner on the p.g. website that directs volunteers over to d.p.), production hasn't increased as a result (it's been flat for years now), and the burnout factor has become palpable. not a good situation... i don't mind if _you_, as a _volunteer_, decide that you want to spend hours and days and weeks and months getting your e-text _perfect_. but when you start wasting the time and energy of _other_people_ -- who are _volunteers_ who've donated time and energy to you -- by putting them into a workflow that is decidedly inefficient, i mind. meanwhile, the attitude about "quality" at d.p. has gotten ridiculous. i ran a poll there years back, on "how much quality would you like?" > http://www.pgdp.net/phpBB2/viewtopic.php?t=32479 the question: > for a 200-page average-difficulty book, > how many errors should we > tolerate as "acceptable" for > posting that book to project gutenberg? the poll gave people options, which received the following votes: > none -- it must be perfect -- 4% -- [2] > 1 error in 200 pages is good enough -- 16% [7] > 2-4 errors in 200 pages is good enough -- 30% [13] > 5-10 errors in 200 pages is good enough -- 23% [10] > 10-20 errors in 200 pages is good enough -- 9% [4] > 20-40 errors in 200 pages is good enough -- 2% [1] > 40-50 errors in 200 pages is good enough -- 0% [0] > 50-100 errors in 200 pages is good enough -- 0% [0] > 100-200 errors in 200 pages is good enough -- 0% [0] > you can't have your pudding until you eat your meat! -- 13% [6] > Total Votes : 43 as you can see, even though i gave a lot of options trying to stretch them out, of the d.p. people who expressed an opinion, the majority wanted no more than 4 errors in a 200-page book, and roughly 9 out of 10 of them wanted no more than 10 errors, and i have no doubt they think they are _meeting_that_standard_. which is just plain ridiculous. 4 errors in a 200-page book is a _very_ polished book -- most digitizations aren't that good. (many aren't anywhere close to being that good.) but d.p. people _think_ their work is that good. they _believe_ it. they _proclaim_ it, as if it were a proven fact. but it's not a fact. many years back, "zora" from d.p. was one of the loudest people proclaiming the superiority of d.p. quality, so i analyzed a book that she had postprocessed. i found _hundreds_ of errors in it, even _obvious_ stuff, like out-of-sequence footnote-numbers... i haven't analyzed any d.p. books lately, but given the fact that people there have an undeserved sense of their _superiority_ when it comes to the issue of quality, perhaps i should do that. it might be fun to knock them off that high horse. of course, they won't listen over there. but we will know here... -bowerbird

1 0

Re: Translation project
by cmiske＠ashzfall.com 24 Feb '10

24 Feb '10

Keith and Andrew: Thank you so much for your replies. I am in the process of learning Spanish (the 'textbook' version they tend to teach in the US school system). In my readings I have already encountered many moments of "I know what it is saying, but what does it mean?" I lack the cultural background and ability to 'play' with words that is necessary for high-level translation. What I was actually thinking of was more along the lines of the translation of early readers (such as the Dick and Jane series) into a multitude of languages. As books of this nature are intended to have a rather limited vocabulary scope and to facilitate the learning of an individual's 'native' language, I thought these books might also be helpful if translated into multiple languages. When reading these books it would also be possible for the language learners to refer to the edition in their native language for guidance. I'm not sure that this sort of translation process would actually create something that PG would even want. Is there much interest beyond my own to see "See Spot. See Spot run...." translated into a dozen or more languages? I assumed that if there were much interest, there would already be such a project out there on the internet. There are, of course, books of this nature on the market, but they are rather limited in availability and somewhat pricey. A free archive would be nice. I agree that the majority of works of literature should be left to the experts and native speakers of both languages. Andrew's project was a fine example of a collaborative effort. I hope more people get together in small groups (or individually) to translate more works. Their efforts are appreciated. Carel -------- Original Message -------- Subject: [gutvol-d] Re: Translation project From: "Keith J. Schultz" <schultzk(a)uni-trier.de> Date: Wed, February 24, 2010 2:57 am To: Project Gutenberg Volunteer Discussion <gutvol-d(a)lists.pglaf.org> Hi Carel, Translation is a fine art. You have to be proficient in both languages and should be a nativer speaker of the language you translate into. Furthermore there is more to translation than taking the text and translating the words, systanx, and semantics. You have to do research on the author, historical research to understand the mood in which a text was written. As proof of fact just take a look at Shakespeare. I have to agree with Andrew that a collabarative approach is not good. You must have strict rules in place when a group is working together. I do not mean to discourage you, but want to make sure you understand the task of translating. regards Keith. Am 23.02.2010 um 07:00 schrieb Andrew Sly: > No particular site springs to mind. Perhaps if we knew the > language you have in mind it might be easier to go searching > for possibilities. > > One barrier to that kind of collabarative translation is that > there is usually not one "proper and correct" way to translate > a given phrase. There can be different word order, differences > in shades of meaning and tone, etc. If you have different people > translate different parts of a text, you can end up with great > lack of a consistent style. > > It's not impossible though. Take this example: > http://www.gutenberg.org/etext/28971 > > It was worked on by a small group of people specifically > for inclusion in PG. A couple people translated different > chunks of the text from English, then after the whole thing > was put together, we did a few proof-reading runs, tried to > even out inconsistencies, ran a spell check, etc. > > This was done through email and using a wiki. > > --Andrew > > On Mon, 22 Feb 2010 cmiske(a)ashzfall.com wrote: > >> I've just recently started studying a language that is foreign to me. >> Although I am a long way off from being able to translate anything >> beyond an early reader, I was wondering if there is a translation >> project out there that is similar to DP? I looked about the PG site, but >> could not find a link to anything of that nature. >> >> Thanks; >> Carel >> > _______________________________________________ > gutvol-d mailing list > gutvol-d(a)lists.pglaf.org > http://lists.pglaf.org/mailman/listinfo/gutvol-d _______________________________________________ gutvol-d mailing list gutvol-d(a)lists.pglaf.org http://lists.pglaf.org/mailman/listinfo/gutvol-d

2 1

Re: so what is so important about pagination?
by Bowerbird＠aol.com 24 Feb '10

24 Feb '10

keith said: > I agree with BB, but we have different ways about doing it. well, to put it in a way that is just a bit more accurate... i have actually developed a system for doing it, and coded tools that prove both the viability and the efficiency of that system... and you've written some posts where you waved your hands around... so, yes, in a sense, we do have "different ways" of "doing it"... ;+) -bowerbird

1 0

Re: !@!!@!!@!Re: Re: so what is so important about pagination?
by cmiske＠ashzfall.com 24 Feb '10

24 Feb '10

I agree as well that the original digitalization should be as lossless as possible. The various outputs of this original can then be as lossy as anyone cares to make them. If the original is lossy, then the information is lost 'forever.' Lossy originals reduce the choices of readers and forces a convention upon them of what is and is not important. The readers of the books should decide what they want/need. Carel -------- Original Message -------- Subject: [gutvol-d] Re: !@!!@!!@!Re: Re: so what is so important about pagination? From: "Keith J. Schultz" <schultzk(a)uni-trier.de> Date: Wed, February 24, 2010 3:23 am To: Project Gutenberg Volunteer Discussion <gutvol-d(a)lists.pglaf.org> Hi All, Been following this thread loosely. But the fact remains that the problem is the encoding of the digital text!!! NOT, the output. I have said it time and time again, the the encoding has to encode the information of the original scan or p-text. Thats is a rendering of such an encoding will produce a fairly close copy of the original. (I use rendering here as a way of producing the text; either in text or display) For the output you simply use those parts of the encoding for your output that you need. The page number problem can be handled because out of convention they are placed in the header or footer of a page. So if you do not want the original pagenumbers then just skip them. If you want the original page breaks use them and output it where you want it. I agree with BB, but we have different ways about doing it.

1 0

Re: so what is so important about pagination?
by christine 24 Feb '10

24 Feb '10

I have been following this conversation, and would like it to lead to a slight change in the PG pagination rule. As post-processor at DP-Int, I often have the feeling of uploading unfinished works because I have to remove the pagination from the text files. I don't mind in novels, but for books with index, it is another matter. Some time ago, I received a mail asking me why I format my page numbers as invisible in the html files. One of my reason is that I can not see the point of showing them, since most of everything is linked (TOC, LOI, Index, etc...) When I started post-processing, I included the page numbers in the text files, and often had to remove them to have the project posted. When I asked about the reason, I received the answers "it disturbs", "there is no need for it, the reader can do a search", etc. Now, it does not seem very logical to show them in a linked file (html) and leave them out in the text file. Doing a search is not a very good solution either, in most of the case, it is like searching for a Smith in a british telephon book, more finds than you care for. As example: I am currently working on "The Life of Napoleon", one of the index entries says "N. returns to Paris". Well, searching for "Napoleon" or "Paris" in that project would brings too many finds, as for "returns" this would not bring you to the proper page either, because, there, you find "is driven back". So, would it be possible to just relax that rule and let the volunteer decides *if* the pagination in the text file of a project is needed or not. Christine (aka Lostpaces)

5 6

Re: !@!!@!!@!Re: Re: so what is so important about pagination?
by Bowerbird＠aol.com 24 Feb '10

24 Feb '10

michael said: > If you have a point to make, you missed. i think lots of people got the point. but you evidently missed it. > Take better aim and try again, or give it up. my aim is true. i've done enough. i have no need to "give up", _or_ to continue. the future is the entity that will deal with this. doesn't matter how much friction we generate. the future will deal. and this is how the future will do that dealing: 1. it will need to tell which texts are accurate and which are not, because it will find itself awash in e-books without provenance... 2. it will do this determination by comparing digital text to scans. (it will test the validity of the scans by comparing them to p-books.) 3. digital text which retained pagination/linebreaks will lend itself better to this comparison process, so it will generally tend to "win", and the "best practice" will come to require this aspect, and strictly. 4. digital text which is difficult to get to (e.g., zipped up in .epub, obscured by some opaque format, unduly burdened by d.r.m., etc.) and text which has discarded its pagination/linebreaks will lose out. i have no interest in engaging in "debate" on these 4 points, because it's perfectly clear to me that they're all _inevitable_. the question here is whether p.g. wants its e-texts in group #3 or #4. -bowerbird

3 2