Wikisource:Scriptorium/Archives/2010-11

Questions

Category needed for articles with no scanned djvu to verify contents

Most articles on Wikisource have no means of verifying the content. Some give only a vague source, such as a telecast with no transcript. These articles lower the quality of Wikisource as all articles should have scanned versions in order to verify. Further, articles are much harder to read when there is no scanned version, and the extra wide formatting of the final versions are extremely hard to read. A large number of these article are unfinished and poorly copy edited.

Therefore, I think there should be a category to identify these below par articles, so the reader can easily skip them if so desired. Many of these articles are inadequately copy edited also, and since there is no scanned version, it is not possible to improve them. These articles decrease the credibility of Wikisource.

Is there any means of assessing the page views of Wikisource articles to assess how often they are read? Just wondering. It seems to me that Wikisource serves the interests of a small group of editors that cater to their own interests and don’t serve a wider public. I think some attempt should be made to improve the quality of Wikisource so as to draw a wider interest group. Also, some "truth in labelling" is needed. Too often I embark on reading a work only to find it poorly copy edited and no where near finished. Since there is no scanned djvu, it is not even possible to try to improve it.

Just my two cents! Another editor (talk) 08:13, 21 September 2010 (UTC)

Does anyone else have an opinion? Another editor (talk) 18:41, 21 September 2010 (UTC)

As mentioned below, the pace of response, discussion, consensus and adaptation is not so much "slow" here on WS as it is methodical and purposeful. Please be patient and keep this in mind in the future. George Orwell III (talk) 22:02, 21 September 2010 (UTC)

It's usually possible to find a source in Gutenberg or archive.org. Without a source they are just less credible. You have to remember that the djvu extension was added just recently. feydey (talk) 18:48, 21 September 2010 (UTC)

Keep in mind that discussion here moves at a slow pace. :) I think we had a variation on this discussion a while ago and the general mood was that it isn't yet time to transition near-exclusively to proofread page. It had a lower rate of adoption then, though, so maybe more people will bite this time. I'm for a slower reform--continuance of the general recommendation for use with newer texts, and a trial effort to transfer some of the larger non-using texts e.g. Anti-Nicene, Nicene and Post-Nicene Fathers.

Also, regarding displeasure with lack of notices for texts which have problems, I vaguely remember a stray reference that the Swedish (?) ws conducted an audit of their texts. Might be worth considering. Prosody (talk) 18:55, 21 September 2010 (UTC)

Moving toward the increased usage of proofreading via the Index: and Page: namespaces is, by far, the absolute most desired form of contribution to WS one can make & was only somewhat recently agreed upon as such (in general terms) as we move forward on WS. It is by no means an absolute must - especially in relation to those works added pre-2009-ish (well about or around that timeframe; I know of no concrete date personally). As these works were added before the realization and subsequent implementation for the current desired, but not mandatory, guidelines for contributions, they are to be re(viewed) somewhat differently than what we now do as community moving forward.

These works may or may not hold the same fidelity as if they were created with the additional benefits we have currently in place; we won't know that until there is a overall proposal on how to bring them up to a specified uniform standard or designate them as "pre-standardized", etc. At the time they were added, the standards were different and they complied with that standard so they were acceptable.

If the standard moves one way or the other and the time comes that these works need to be addressed to improve/maintain the credibility of WS as reliable source. Your proposal may indeed be needed at such a theoretical point to help resolve this issue, but currently it isn't all that pressing and can be more easily ascertained by the presence or omission of a "source" tab normally appearing along with the other tabs at the top of each page in the namespace. It also does not explain to the unknowing reader adequately enough the difference between the pre- and post-accepted guidelines for contributed works and the relation of those works to actual fidelity of the work. I'll end there for now. George Orwell III (talk) 22:02, 21 September 2010 (UTC)

The other thing that we have noted is that many of the Gutenberg texts are for editions that are not readily available in the scanned format, and this makes the proofreading a little more complex. In other words, different (US vs UK, 2nd vs 3rd) or not specified editions. It has slowed us down, or stopped, a conversion on more than one occasion. To also note that for transcluded versions, they no longer utilise {{PageQuality}} so that becomes an indicator for many texts of works without scans. — billinghurst sDrewth 12:20, 24 September 2010 (UTC)

If the work is labeled as coming from Distributed Proofreaders, the scans are probably available on request. Enough requests might get them to actually go through with making that publically available archive of scans that has been on the proposal table for a long time.--Prosfilaes (talk) 03:38, 25 September 2010 (UTC)

A preference for dejavu is an implicit preference for stuff that was scanned by one of the big scanning projects. This is potentially limiting. Unless there is a good way for a contributor to create dejavu from a hard copy at home, then we are implicitly depending on large, expensive equipment that is under the control of large organizations that may or may not have their own agendas. We risk restricting ourselves to the set of works that have already been scanned. To avoid this bias, I think we should be happy to accept any input that has an attributed provenance. I do agree that we should ask for the strongest reasonable proof of provenance, such as pictures of the cover, title pages, and other front matter. A Worldcat reference will usually permit other folks to find a copy of th esame edition in a library. (Personally, I'm working on the DNB, using dejavu images from the big scanners.) -Arch dude (talk) 21:48, 26 September 2010 (UTC)

It's no problem making a DJVU file at home, though I suspect many might find it easier to make a PDF file, which we could also load. Doing a one-man type-in job is not usually a productive idea; it's very hard to produce a good typo-free edition that way.--Prosfilaes (talk) 22:48, 26 September 2010 (UTC)

I notice that files are still frequently being added that do not have a DJVU backup. Since Wikisource is still encouraging them, why not identify them by a category? The addition of a category does not disallow them, but acknowledges them by a different category of verifiability. Just a suggestion. Mattisse (talk) 16:14, 27 September 2010 (UTC)

It's worth remembering that DJVUs are not a panacea. Even when they're used, the final product cannot be better than the original. If we base our text on an edition riddled with misprints, it will not be as good as one corrected against a more reliable edition. However, if the latter is post-1923, we can't post DJVUs.--Longfellow (talk) 19:55, 27 September 2010 (UTC)

"if the latter is post-1923" then we sometimes can post DJVUs. For US publications, they would have needed a new explicit copyright (prior to 1978 and in most cases 1989) for changes from the older edition to be copyrighted, If we can post the text, in the US typography is not copyrightable, so we could post the new images sans illustrations or new material. If the book is not clearly out of copyright, then whether or not we could use the text is less bright-line...though that doesn't really change depending on whether we're using scans of the images or not.

Also, my big love of DJVU as an admin is that we get fewer miscorrections, and it takes less time to fix them when we do.--Prosfilaes (talk) 20:13, 27 September 2010 (UTC)

WIP on Horses and roads

Horses and roads, as I told you, is a very interesting book about horse welfare, but it has been an "opportunity to test ideas". Presently, I'm adding experimental {{HAR}} to its index pages, beginning from Page:Horses and roads.djvu/241. The template adds anchor-links to analytical index entries, "proofread-proof", so that the same link runs both in nsPage and in its transclusion. It uses internally another experimental template, {{Anchor2}}, that highlights its target; as you see in its doc, it needs an additive row of css code to show highlighting of the background.

All the stuff is a little complex to understand fully, but IMHO this complexity is hidden, since the template call is very simple: as an example, see the first link to book page 100, that is simply {{HAR| 100|Abelorna, experience of}}: param 1 is the number of book page, param 2 the name of the anchor. The template calculates the djvu page corresponding to book page 100, calculates the name of Ns0 page where page 100 is transcluded, and gives a different link when it is called from nsPage or from ns0. The idea has been appreciated into it:source, and I'll implement it there as soon as possible.

Please don't be confused if some templates are lacking/incomplete into pages following Page:Horses and roads.djvu/241, ando consider that anchors will be seeded at end into the text; now only some anchors have been introduced into the pages (those pointed by the first row of links, under "Abelorna, experience of". And remember that you will not see highlighting if you don't add that row to your personal css. If you like the idea, consider the possibility of adding the code into Common.css. --Alex brollo (talk) 09:43, 27 September 2010 (UTC)

I began to introduce anchors pointed by analytical index entries. A terrible work, they are hundreds! The sorted list of anchors to seed is here: User:Alex brollo/Sandbox. I'm going to post them into right pages by User:Alex_brolloBot as lists into html comments, just to have a small help, then I'll go through pages moving them into the right place into the text. My bot is unflagged, I'll use a slow throttle speed (min 60s, perhaps slower). --Alex brollo (talk) 08:00, 4 October 2010 (UTC)

help please

Could someone please help me format Wakulla/Chapter 6 for me? There is a section of the text that I need help on. Thanks. - Tannertsf (talk) 21:17, 29 September 2010 (UTC)

Did you need help with the song near the end? You can either use {{gap}} and <br /> to format each line, or you can enclose the song in <poem> and </poem>. For more information, see Help:Editing poetry. Cheers, stephen (talk) 21:34, 29 September 2010 (UTC)

Yes. Thanks for changing into poem format. I'll be sure to read about editing poetry. - Tannertsf (talk) 21:37, 29 September 2010 (UTC)

Image caption font-size and line height.

In PSM, there are hundreds of multi-line image captions and small font text paragraphs, in which the combination of 85% font-size with a line height of 1.15em (13px) is optimal. I familiarized myself with the numerous templates dealing with font sizes and there is a great variety. The {{fine block}}, {{font-size}}, {{font-size-x}} etc. and I really wouldn’t want to make another if possible. Examples of all three, with their calculated line heights ARE FOUND HERE, and an example of the image caption where it’s needed IS HERE, using the {{font-size-x}}, but the line height is still too large.1.2em (17px).

The most desirable option would be the {{font-size-x}} template modified with an optional line-height size, with an if statement for the 85% font-size to be 1.15em (13px) because the template is already referred to in numerous text through the{{fsx}} redirect. Unfortunately, this kind of template modification is beyond my technical knowledge at this point. Comments and ideas are appreciated. - Ineuw (talk) 01:30, 30 September 2010 (UTC)

We have {{smaller block}}, which uses <div style="font-size: smaller;">, and more recently, {{Fine block}}, which uses <div style="font-size:85%; line-height:1.2em"> for "fine print", but not as fine as "smaller". The line height in the first is what you get with style="smaller", the 'fine block' template tweaks font-size and line height to create a style that is 'smaller than normal, but larger than smaller. The purpose of a smaller block of text in print is to distinguish it from the 'regular' text of the page, a quote, footnote, caption, or long lists. Smaller blocks are intended to have a semantic value, or compress 'fine' details for convenience, or to be unobtrusive and ignorable. The comment has been made that {{smaller}} is 'too small' and I agree that is the case when my settings show regular in a small font, this is especially true of smaller smallcaps. There are conflicting approaches on how to handle this, this page introduces some of the pitfalls and recommendations; one of these approaches states that anything smaller than normal should be reserved for fine print. Perhaps the solution to this site wide problem should be derived from how this site, and its sisters, defines the default text size; en.ws should be taking its cues from en.wp on this matter, unless there is a very good reason and consensus to deviate. cygnis insignis 08:33, 30 September 2010 (UTC)

style and scope

Prompted by an increasing number of unproductive distractions, I will try to identify what I see as becoming a source of disruption. I use the template and tag for refs at wikipedia, how small that makes the text is not something I or another user have to consider, question, debate, or negotiate, to include the interesting bit. I also use an infobox there (only one), the finer points design of line weight and colour change with the wind, unless it has impact on the reader I don't care which preference triumphed and who got blocked. What I know about style and display has been irrelevant to contributing content at that place, it is not the point of the site. If I'm tired and have to squint to read something, I hit a key, make the text bigger, then carry on. I don't decide that wikimedia sites makes text too small, or has lousy kerning, and set about making some display issue a cause celebre. Others do, if they are right then good luck to them, but I'm certain that nearly every reader of the site doesn't notice.

We should be very cautious of replaying discussion of finer, or trivial, matters of style from the larger communities at en.wp and Commons. I feel this discourages users who are interested in the content of their contributions, and shifts focus from the more delightful aspects of our scope. This is a cultural problem, not a criticism of individual's opinion; someone who wants to share a text only needs a handful of unobjectionable templates, but they are confronted by, and drawn into, an ever-increasing number of tedious and divergent stylistic considerations. New user adds a text, 2nd user adds their theory of optimum heights and weights, sizing and kerning, and the crucial decision on serifs; it is understandable is new user thinks that is WIW! This is especially true if they come from a sister, where they could have created featured content with no understanding of these issues. It is also the case that fugitive and mutually exclusive ideas on these issues are implemented here, the site has been used as a means of fostering what is deemed noisy and unproductive elsewhere.

When copy/pasting from Gutenberg was the norm this didn't have much impact, however, completion now requires coöperation; it would be sad if the basis for collaboration was agreement on the font, rather than love of the work itself. There is an notion that discussion would be required be every time a new user, or new login, insists they see and add a space before punctuation or some such matter. The circumstances lead to thousands of tiny tyrannies, users advancing or denying a notion that could not gain a consensus, regulars either override or indulge others on a whim. There are things that a) should be transcribed, there is b) things could be transcribed, and there is c) things that open the floodgates to tedious wonkery and slavish photo-facsimiles of inconsequential elements of a scan we have a click away. Of these a) is easy, b) be should be unobjectionable, nice, and c) should be avoided like the plague. Not assigning a preference, honouring site and user preferences should be the default position, not something that has to be justified to those with different views. cygnis insignis 08:33, 30 September 2010 (UTC)

Against the proposed policy of limiting editors' choices in formatting and the squashing of the "thousands of tiny tyrannies," a counterexample: if you look at the scans of Sanskrit Grammar you will notice that many different sizes and styles of type are used. To reproduce them here would doubtlessly seem to you to fall under "tedious wonkery and slavish photo-facsimiles of inconsequential elements of a scan we have a click away." Yet, at least one reviewer praised this particularly for its salutary effects on the text's organization. Perry's Sanskrit Primer, in a footnote, reads: "The student is therefore advised at this point to read carefully the chief rules of euphonic change in Whitney's Grammar §§ 139-232 (the two larger sizes of print)" [emph mine]. I hope it's clear that what another editor who has nothing to do with the text at hand might conclude regarding formatting has nothing to do with whether that formatting has any utility. It is these tyrants, and these tyrants alone, who, by love of a particular work rather than works in general, have the requisite information to make decisions regarding what is beneficial in creating a particular edition. Prosody (talk) 16:50, 30 September 2010 (UTC)

Nice example, that is "a) should be transcribed" by that reckoning. Limiting any one user from changing everyone else preferences is my position, it is not our purpose and no one should be 'editing' anything here. "To reproduce them here would doubtlessly seem to you to fall under" [should quote] "The purpose of a smaller block of text in print is to distinguish it from the 'regular' text of the page, a quote, footnote, caption, or long lists." [me, from section above]. I make use of multiple font-size, smaller and greater than normal, when it is given and benficial. The exceptional example given would be unobjectionable, the tyrant we should obey is the quoted author. My love of individual works, and works in general, has produced what I hope are faithful transcriptions, occasionally with challenging formats, if there is something to be improved then I hope someone lets me know. Try assuming there is something in what I say, and avoid personalising the discussion where possible [everyone]. We should add complexity with sound reasons, not because it is possible, or deprecated at a sister. cygnis insignis 20:54, 30 September 2010 (UTC)

I was overly pointed in my response but by no means was I attempting any sort of personal attack. Sorry for not being civil enough to make that obvious. Of course, we're trying to make a sort of platonic ideal of a particular edition, hewing to what sprung out of the author's mind and disregarding extraneous information from the hands of the text editor or typesetter. The point I meant to make is that the people best able to make the distinction between those two categories of information here are the people who spend many hours preparing the individual work, not the community at large, nor the random 'drive-by' editors (who can only be emboldened by any sort of advisory guidelines). Prosody (talk) 15:46, 1 October 2010 (UTC)

I didn't see a personal attack, what I observed was an example that I would have transcribed anyway, a decision that would have been reinforced by an outside authority, being attached to the a quote of the most strident part of my statement, which was applicable things we all probably agree are (c). I think it possible to reach agreement amongst the first group, acceptable by the community at large, on definitions of (a) and (c), and come to accord on the conventions and standard practice for (b) "could be transcribed". The merits having a style guide are obvious, efficiency and reducing noise are amongst the best, the reasons for deviating need to be calculated by authoritative and content related concerns.

We have a house style on some matters, we generally never think about them because they work so well. The advantages of using the ref system for footnotes allow us to focus on the interesting bit, yet it is possible to conceive of a problem with disposing of the asterisks, double daggers, or superscript numbers on the page. They are, after all, 'in the scan' or recommended by a style guide. I don't have to justify this deviation every time a new account comes by, it is unobjectionable SOP. I ignored this convention when I thought was required by a work, about two exceptions that I asked for a sanity check on. There is a lot contrary guidance on trivial matters, and examples of these that users assume is de rigueur, I am repeatedly distracted from contributing content to justify the omission of (c) on the basis that is arguably (b).

I don't want to be placed in the position of owning, deciding, or negotiating on a work by work basis, or discouraged by the additional labour on hundreds of its pages. There is a laundry list of less-important considerations deprecated in the wikipedia style guide, problematic in digital documents, or found in a fraction of the works here, or drawn into a discussion on something that might be okay if everyone has already had a good hard think about it. There is a strong correlation between enthusiasm for any and all of these, then boredom and desertion around about half-way through the second chapter, the culture that allows this did in fact reduce noise, but I doubt it increased the content as much contributions that attempted to match a simple and regular style. Someone sees a space before semi colons, or around a curly quote, and they are unaware of the consequences of transcribing it, or 'decide' it is an integral aspect of transcription and the second user is personally wrong for objecting, because the is no style guidance. Leaving out space before punctuation becomes 'discovered' to be a preference, though there are good reasons to avoid it, it is being imposed by most users most of the time. If someone suggests otherwise, we add it to the style guide, if we need to change it we discuss it, this depersonalises the whole business and stops the fragile and fragmented approaches, incompatible style preferences, and opportunities for misuse of the community's time and patience.

The virtue is the simplicity is provides those who want to produce content. Guidance is not intended to stifle evolution and improvement of our content, it would primarily facilitate expansion and collaboration. I recognise that the unspoken or implied guidance on allowing users to apply what they want, and that was workable before collaboration became prevalent; I believe it is no longer viable to have our incompatible preferences of form where they are trivial. I'm not denying anyone has the right to use the site to convey something meaningful, just to have everything they like. cygnis insignis 02:37, 2 October 2010 (UTC)

if you can establish an norm that makes things unambiguous, so that two different users proofreading the same scan would end up with the very same wikitext, even though they do not know what the other is doing, then I fully support your proposal. ThomasV (talk) 06:57, 2 October 2010 (UTC)

That is the one of the objectives, supporting one of the proposals below will achieve that obvious advantage. cygnis insignis 07:33, 2 October 2010 (UTC)

quotation marks

• Proposal: The Wikisource style guide should recommend the use of straight quotes, " ", rather than “ ”

quotation marks and apostophes

• Proposal: The Wikisource style guide should recommend the use of straight quotes and apostophes, ' " (on the keyboard), rather than combinations of ‘’“”`'" (various and discrete input methods, platform dependent). Special requirements for exceptions are documented.
• Discussion: see w:Wikipedia:Manual of Style#Punctuation, or provide a reason to ignore it.

The same as above, but someone may see a distinction. cygnis insignis 03:12, 2 October 2010 (UTC)

• Precendent: "Curved single quotes may be replaced by straight single quotes (apostrophes: <'>.) Many OCR conversions of the original text perform this change ." [1]

I disagree with these two proposals. Wikipedia may choose to use straight quotes to make them more accessible to the various editors, but I prefer using the usual curved marks for quotations in Wikisource. The texts here are not edited as frequently as they are on Wikipedia, and I think it just looks better that way. It should of course be up to the individual, and those wishing to use the straight marks can. And if anyone wants to spend time changing straight marks to curved, that is fine. I do use straight apostrophes only because I prefer to make them distinct from the single quote marks. —Mike 04:23, 2 October 2010 (UTC)

I like it too, but I prefer a good reason to vary from simplicity. What is the reason to do it, it would be almost indistinguishable to many readers. Can you formulate a counter proposal on how this is to be decided, who, when why? Can you justify the extra work in applying and checking these open and close correctly? Are you certain there is no affect on access, that the reasons it is recommended at en.wp, technical, practical, predictability, do not apply here? Have you considered that this requires changing the ocr of hundreds of pages, maybe dozens of corrections to something that needed several to no keystrokes? cygnis insignis 04:39, 2 October 2010 (UTC)

The reason to do it is because they're the right characters for the job. Only history has made the straight quotes and apostrophes more available. Typesetters have always used curled quotes. There's no effect on access; it's been well over a decade since every system supported basic Unicode, and no system can handle any fragment of the modern Internet that can't handle it. I don't see a problem in making this a mixed practice, in letting users so interested convert them over the proper quotation marks.--Prosfilaes (talk) 06:53, 2 October 2010 (UTC)

If that is correct that make that the style, and let en.wp know they have it wrong. To appreciate one of the problems I identified, contributing to the Page namespace, requires a reasonable amount of experience using it. The second point needs to clarified. Wikipedia does not, that is sufficient reason to wonder whether assertions of 'no effect' are that simple, it is certain it is more work. Someone changes the first 10 pages of an index I'm doing, what do I do in that circumstance? cygnis insignis 07:28, 2 October 2010 (UTC)

My rule of thumb on style is: are we conveying to the reader what the author intended? If he/she was making a point by using one type of quote rather than another, in particular if there is one type in one place and the other type in another place, we must stick to that. If it seems that the author didn't care, what we do is of no more import than deciding to use a font different from the original.--Longfellow (talk) 21:45, 3 October 2010 (UTC)

Good points. If the printer uses two styles for quotation, we should assume there is meaning in the author/editors intent, and consider whether this is conveyed in our transcript. I'm not aware of an example of this, only things like relative size (that have no effect on searches, etc.), but will point out that different printings of the same text also vary with no effect - typesets rarely affect meaning/author intent. We don't decide to use a different font, as a rule, we have a default or the user (not User:) decides on another with their preferences. cygnis insignis 08:08, 5 October 2010 (UTC)

FWIW... Most U.S. legal/legislative texts use the different quotes to differentiate between the usage as understood as given above (quoting someone; the spoken word etc.) and to indicate (even illustrate or instruct) new or amended text from the old as needed for proper codification. Frequently, you will find that such text is always given what has been called "curly quotes" (‘‘ ’’ or ‘‘ ’’ among other ways [“ ”] that all render the same on WS apparently) an open-quote for each new line, or line item, but only a single close-quote on the very last line of the new or amended text as a whole. This is different than when these types of legislation or legal texts cite existing case names or previous Acts where the normal (or the straight) quotes are used as they normally would be (as laid out above basically). The older the publication, the less likely one is able to ascertain this without some subjective input of course, but it's all based been on that differentiation to smooth the subsequent codification for the past 200 years (well, in the U.S. at least). I also vaguely remember something similar being done for certain types of scientific/mathematical notations/papers and their publication but I can't swear to that. George Orwell III (talk) 09:29, 5 October 2010 (UTC)

Intriguing, and a possible basis for IAR. The trend among some publishers of other types of works was to begin each line wrap of quoted material with quote marks; this was often with the same size type, so quotes were still denoted across page breaks. This 1750 work, though I have seen later than that, doesn't have that page break problem after transclusion as a whole section. The erudite discussion of the legal example points out these variants begin new lines; if adding a second type for quotes is needed, then it is worth heeding how this is transcribed a site that specialises in the material. Keeping in step with an authoritative or 'official' website, government or university for example, or accepted Standards for digital pages on the material would be pretty compelling. cygnis insignis 19:38, 5 October 2010 (UTC)

Well, there is Index:United States Public Law 110-2.djvu already on WS and that page links to all three (text, PDF & XML) types of U.S. government provided docs of the same law. Not the best example of all the possible incarnations but a copy & paste or simple inspection can show the underlying non-use of straight quotes for differentiating ones (i.e. the "curly" quotes, the "slash" quotes, double &lsquo ; &rsquo ; -- however one may call them). George Orwell III (talk) 19:58, 5 October 2010 (UTC)

I believe it's the same as capitalization; it's usually moot, but the disambiguation between opening and closing quotes is occasionally needed in disambiguation. The intent of most authors was to use quotes that show the difference between opening and closing.--Prosfilaes (talk) 18:53, 5 October 2010 (UTC)

Can you show an example of needed for disambig. The intent of the publisher was to show what was quoted, within the constraint of a printed page: this is not a problem with digital pages, except where it is retained as an affectation. cygnis insignis 19:38, 5 October 2010 (UTC)

I don't know what your point about the printed versus digital page; curved versus straight quotes work the same way on both. Try

Susan said "And ' and ' and ' and ' and " and " and ' and ' and ' and ' and".

Is that as easy to follow as

Susan said “And ‘ and ’ and ‘ and ’ and ” and “ and ‘ and ’ and ‘ and ’ and”.

? Or stretching the rules more does

" I have found that ' to be " should be " I have found that to be ' "

mean

“ I have found that ‘ to be “ should be ” I have found that to be ’ ”

or does it mean

“ I have found that ‘ to be ” should be “ I have found that to be ’ ”

? In any case, I don't why we should use the same rules as Wikipedia. Once in a final form, we don't need easily editable text; the ongoing cost of switching from straight quotes to curved quotes after most of the work is minimal. I believe a simple script can get it 99% right, except for stuff like "'Tain't ri't", and other dialect junk.--Prosfilaes (talk) 21:10, 7 October 2010 (UTC)

This makes no difference to what you do here, but what is your point? Wikipedia has one rule for everything, the MOS and guidelines are derived from that. Once it is in it's final form, maybe, barring all other considerations, but there is no guideline to getting it to that point. Pages of works I 'finished' are changed in passing to some other style of quote, and there is no guideline to how contributors can agree on getting to the point. Do I change the rest to honour the notion that users denied something at wikipedia can be indulged in their preferences here? The overwhelming circumstantial evidence, and impracticality, and the very real problem of interfering with searches, and the inconsequential impact, if any, of using the given straight quotes means we need a very good treason to deviate. If they are wrong, or inadequate, then wikipedia is wrong. We need agreement, and guidelines, for collaborations by users who contribute content (more than a couple of hundred pages). We don't need to champion the right for one lot of users to do something they could not do at wikipedia. Let the libraries that produce the thousands of pages of ocr know about the script, when it exists. Speculation is unhelpful, as is seeking weakness in my presentation of something that doesn't belong to me - a wide-spread tendency to avoid them. This is a slippery slope to replicating every on the printed page. cygnis insignis 22:15, 7 October 2010 (UTC)

I don't understand that comment well. Wikisource is not Wikipedia. This is not about users denied something at Wikipedia; this is about representing what is on the page in a way that not only preserves the content (including semantic content) of the original, but also in a way that maximizes readability.--Prosfilaes (talk) 22:36, 7 October 2010 (UTC)

Why does 1750 work (link given above) go to an error page? I’m sorry but am I missing a point here? If the page doesn’t exist, why does it have a link? Mattisse (talk) 20:17, 5 October 2010 (UTC)

I think I fixed the link(s) to point to where Cy intended. George Orwell III (talk) 20:21, 5 October 2010 (UTC)

Thanks. I'll see if I can this link right: I think the loc page is the closest thing to what we aim for, not splitting to two pages and accessible and clear formatting - as you would expect from the Library of Congress (Shame about the table code display). They use one style of quotes, which I would do too ;) cygnis insignis 20:34, 5 October 2010 (UTC)

Not sure if you are rendering is what is really there - a common "problem" when trying to illustrate this. A copy and paste into WS from all three types (PDF, XML, ASCII) show the differentiating quotes...

PDF
	This Act may be cited as the ‘‘House Page Board Revision Act of 2007’’.

XML
	This Act may be cited as the “House Page Board Revision Act of 2007”.

ASCII
	This Act may be cited as the ``House Page Board Revision Act of 2007''.

... not straight quotes, being used. Just because they appear as the same under the various browsers and set-ups does not mean they are not being used - the plain text version is the easiest baseline to view since it also shows where the side-notes should be anchored to as well (also hidden but exist at the same time when applied digitally). George Orwell III (talk) 22:30, 5 October 2010 (UTC)

I'm not sure we are on the same page. The example I indicated was viewed on three browsers, using various settings and fonts, it showed a full page with one style of quotes, curly as it happens. Straight quotes are semantically equivalent, and applied with a good deal more ease than if the process was reversed. cygnis insignis 22:15, 7 October 2010 (UTC)

They aren't semantically equivalent; " is both an open and a closed quote, which is semantically different from “ begin solely an open quote.--Prosfilaes (talk) 22:36, 7 October 2010 (UTC)

From the the pov of the reader, _" is open, "_ is closed, the same information is conveyed. And only those who unaffected either way would actively seek some hair splitting, or spacing, as a basis for deviation. We could does not mean we should. Curly quotes worked well when spacing varied in justified texts (see below), which comes with all sorts of other subtleties that we don't need with 'modern'. cygnis insignis 05:53, 8 October 2010 (UTC)

Right - I just copied and pasted the first line using from all three to illustrate the Government Printing Office's application of quotes. When you drill down further in the Public Law you'll see the simple examples of single opening quote per amended/added line in use. I did not want to clutter up the illustration with a multiple page amendment and the resulting Christmas tree of instructive text and quote-marked amended text.

No we are not on the same page. I will follow the reasoning and examples laid out above as it pertains to a narrow type of work; one that should not infringe on the majority of the other types of works here. I can see the benefits of straight quotes where the main concern is with literature (the spoken word or quoting someone) but straight quotes can never be a mandate when it comes to certain other types of works - legalease and legislative in this case. George Orwell III (talk) 22:48, 7 October 2010 (UTC)

Any variation on straight should be widely agreeable and documented, to provide the same advantage as the guideline. I would also test the search function and otehr emergent properties. I apologise if there is a ambiguous situation where there is no page break, though I see it where there is, I'll read through again later. I'm happy to help with documentation for key areas, like the development of legal texts, and you would surely appreciate being relieved of the task of repeatedly explaining, or justifying, the scheme and sop to every new contributor. cygnis insignis 05:53, 8 October 2010 (UTC)

I can appreciate your perspective and of course welcome any such guidelines with any possible caveats clearly carved out within it but I'm also well aware that the impact of such a guideline will not sway the contributor, new or old, one way or the other, very much. When confronted with the "evidence" (a reliable source, scans of the work) containing X type of quote, Y type of dash or Z amount of spacing running contrary to that guideline -- all convincing arguments about collaboration or uniformity will fall to the wayside and this will ultimately remain a matter of user choice & preference.

Though it is well beyond my particular skill set, I think time is better spent developing a script (or bot?) to convert each of the bulleted proposals to one style or the other as need be. As someone pointed out above, it is not as if these WikiSource hosted works are common WP articles. They aren't typically subject to further & constant editing and refinements as found in the main WP namespace but ones that, at some point, are considered finished and not open for debate or amendment(s). Playing quote cop is waste of time (imho). George Orwell III (talk) 06:31, 8 October 2010 (UTC)

Liberation from something that is a tangent to our scope is what is needed, not giving host to every possible permutation of fugitive style choice. It would be impolite to point out the series of one and a half chapters of an unfinished work, users demand a right and become bored when no one pays attention; they are affronted when their preference is removed, but using that right is tedious when they are ultimately disinterested in the content. Users take a while to see what needs transcribing, they produce complexity from looking at it, not reading it. WS needs it more because it is more vulnerable to tendentious overflow from the big brash sister.

I'm not disagreeing with you - formatting on WS can be a distraction and so can quibbling over which quote to use as well. My online expierence, albeit not primarily wiki-based, has taught me that some "battles" aren't going to go away ever, never mind easily, and this is one that can be nuetralized with the application of such technical means. That hypothetical contributor is now sheepish on some project because he/she wanted curly and now everybody has been adding straights -- run the bot; make them all curly; and get back to proofreading the remaining 200+ pages already is all that I was trying to get across. In other words; remove the road-block before the traveler has a chance to get to it - good or bad. The same goes for the points that follow - a guide is good thing to have but, imo, it's not going really solve anything for you nor motivate contributions. Finally, I find little comfort or sympathy when one aspires to the WP standards in general. In fact, I'd be more likely to take a knife to them than some fork (but that's just me venting). George Orwell III (talk) 08:40, 8 October 2010 (UTC)

Doing this through something like the new 'layout' options would be the way, reader choice if they want it. They are still subject to editing, and are, that is a key point. I'm working through an old FT that has every possible combination of these two (or three) issues. The basis for a clear guideline at en.wp is even more important here, in building, consistency, predictability, maintenance, and preservation. The formulations and apparatus for managing a community of users who must agree when their preferences differ, the MOS still greatly eases my burden when I'm creating content there. It also forestalls those are disgruntled with wp's mos coming over to a smaller, quieter community and advancing a [trivial] variant in an unproductive way - making us a style fork on wikimedia. Simple example: I had to work out why my copy paste to en.wp didnt convert titles to italic, '‘Blinky Bill and the Rogue Apostrophe’', and I was engaged in this discussion! This trivial example amounts to a lot of confusion when previous attempt worked fine. Creating content and avoiding discontent is the primary aim. No guidance is a poor option, it creates the situation where adopting the simplest and acceptable means is regarded as arbitrary; the more trivial the issue, the greater interest by those uninterested in content. [User_talk:Cygnis_insignis|cygnis insignis]] 08:01, 8 October 2010 (UTC)

Speaking of disambiguating between open and closed quotes, perhaps you meant something like what follows. It is a clear example of the way I use the different quote marks, as previously mentioned. —Mike 17:00, 6 October 2010 (UTC)

He shrugged his shoulders. “Well, perhaps, after all, it is of some little use,” he remarked. “‘L'homme c'est rien—l'œuvre c'est tout,’ as Gustave Flaubert wrote to Georges Sand.”

It looks partly right, replacing 2 types with 4, but why is there a 5th, the straight apostrope?. As with all others, I had to zoom to check that, this is a couple of dots or pixels either way - there would be almost no peceptible difference to most readers, no loss of meaning, just a great burden and potential source of noise in the community. It less work for me to pursue this, than engage in a fragmented conversation, and various rationales, month after month, when I am trying to focus something that is much more interesting. No one thinks there should be two approaches in one text, this is the best way to avoid it. cygnis insignis 06:11, 8 October 2010 (UTC)

spaces

• Proposal: The Wikisource style guide should recommend the main text use one space, or none at all.

The aim of this proposal is not intended to relate to special formatting, tables, or unobjectionable enhancements to presentation. It is intended to define the style in the greater body of the text proper, and the regular use of headings. Spacing (kerning) around punctuation, incremental or full, is absent, one full space follows : ; , . ? ! (though two spaces may follow a full stop/period, it doesn't display;) before and after the quote mark and parentheses. As with wikipedia, the dashes of parenthetical em dash and hyphen has no space surrounding it. And anything else any one else can think of. cygnis insignis 03:09, 2 October 2010 (UTC)

help

How do I match and split? i need it for 1911 encyclopedia....and i have no clue - Tannertsf (talk) 19:25, 30 September 2010 (UTC)

Sorry, I meant to answer this. Help:Match and split has the details, here is the skinny: switch on the gadget in your preferences, find the page in the scan, add that to the code at the help page. If that is too fiddly, though it may save you time if you are doing lots of them, you can copy and paste the text into the Page: namespace, next to the scan of the article. cygnis insignis 00:36, 1 October 2010 (UTC)

im getting error and text does not match messages...so now what? - Tannertsf (talk) 01:30, 1 October 2010 (UTC)

Adopt the second option is what I would do. Please link examples or name the article, then someone can show you what to do.. cygnis insignis 03:20, 2 October 2010 (UTC)

Index:EB1911 - Volume 01.djvu is an example. any of the red page numbers. - Tannertsf (talk) 10:36, 2 October 2010 (UTC)

you are doing it wrong. m&s should be used in the main namespace. you tried to use it in the page namespace. see the help page. ThomasV (talk) 10:53, 2 October 2010 (UTC)

I've read the help page for match and split. still don't know how to do it...if anyone could show me how to...i would appreciate it. - Tannertsf (talk) 12:35, 2 October 2010 (UTC)

EB1911 is not a good example for m&s, because articles are in general smaller than one page. it will not help ThomasV (talk) 12:42, 2 October 2010 (UTC)

I tried it and got the same result, the section tags appear in the text layer and M&S wants to create those. Copy what I did at 1911 Encyclopædia Britannica/Amygdaloid, you just need to change the page number from= and to= and prenamed section fromsection=s2. The header need prev and next changed too. Comment: I'm becoming more convinced that creating the smaller pages are helping no one, especially the person creating the pages. cygnis insignis 13:02, 2 October 2010 (UTC)

Nearly a thousand pages, with an estimated average half-a-dozen articles per page, and that's just volume 1! There has to be a better way to browse this clicking through one-line articles. cygnis insignis 13:08, 2 October 2010 (UTC)

I agree that creating micropages is not the right way to do it. check this page. it uses a script that spares you page creation. For the moment it needs javascript, but I plan to develop an extension tag that will do this server side. ThomasV (talk) 13:08, 2 October 2010 (UTC)

I was planning on using that for a dictionary of Painters, but I opted to divide it into initial letter when someone rushed it to mainspace. The incoming links to the notable entries use the page number, near enough if I plan to do more than one dictionary, or anything else, in my life-time. Your method is probably better, I still intend to try it out, another brilliant innovation. cygnis insignis 13:17, 2 October 2010 (UTC)

I see what you mean, but i was talking about the index: pages for the 1911 encyclopedia. - Tannertsf (talk) 13:39, 2 October 2010 (UTC)

Pardon, it was an interesting diversion. I answered at 13:02, 2 October 2010 above, go to this link and adapt what I did. cygnis insignis 13:47, 2 October 2010 (UTC)

And i do the same, even on encyclopedia pages that have nothing on them? - Tannertsf (talk) 14:07, 2 October 2010 (UTC)

Yes, but first do this. Create the redlinks (Page:) from the index (Index), proofread the section you want, then create the 'empty article' . Read Help:namespaces to get an overview of what is going, and Help:Side should help with the rest if you need it. cygnis insignis 15:01, 2 October 2010 (UTC)

Sister project link(s)

I've been noticing some development work on the link to Wikipedia articles etc.: the small message about other Wikimedia content with the little logo, as it is currently. If there is a Wikipedia article identified in a header field (I see the DNB00 header, but I presume this is site-wide and a component of other headers) it says at the moment "Wikipedia article" in a light blue link. Well and good, but is there an actual objection to displaying the WP article title? In the case of biographies with name inversion, which you might think self-inflicted but is how the DNB does it (like surname+forename) this may be the only place on the page where the name in correct order actually appears. Which is quite serious for search. I used to like the non-hidden way of doing it for that reason alone. Charles Matthews (talk) 09:01, 1 October 2010 (UTC)

If you mean {{Plain sister}}, it wouldn't be a big problem to implement that practically. The easiest way would be to remove the piped text in the link or duplicate the parameter (eg. [[w:John Smith]] or [[w:John Smith|John Smith]]). However, if there are links to several sister projects the result could be confusing (eg. "sister projects: John Smith, John Smith, John Smith, John Smith"). Another way would be to include override parameters (eg. "Wikipedia_override = John Smith's biography" or similar) to customise the display. However, this would require changes to both the sister template and all of the headers; which means an admin will have to do it. There is also a potential problem in that non-standard wording might cause confusion or undermine the point of having a standardised template in the first place. - AdamBMorgan (talk) 12:10, 1 October 2010 (UTC)

I expected more input on my tweaking, hoped for it actually. I presume there will be a close match between the titles, or a redirect at en.wp. As I used it I had the same thought about displaying the target name, sections of works presents ambiguity that WP titles would help clarify. This requires more labelling to name the sister, though they are likely to be just a few. There is some musing on that at Template talk:Plain sister, and at Template talk:Header. Thanks should go to AdamBMorgan for the development of this, I'm responsible for the instability and ongoing uncertainty. cygnis insignis 08:13, 2 October 2010 (UTC)

While we're on the subject, I would also like to see the "biography, media", etc links in the author header made into explicit sister project links. The "biography" one especially irks me. We're a library of public domain texts. When I'm on Samuel Johnson's author page, and I click on "biography", I expect to be taken to Boswell's Life of Johnson, not whisked off to some completely different website. ... Okay, maybe I'm painting this a wee bit too thick, but I do think the link text is way too vague. Hesperian 23:05, 1 October 2010 (UTC)

Incorporating the first part of Plain sister should cover that, as it starts with the words "sister projects." Also, cygnis insignis' alternate format mentioned at Template talk:Plain sister would better handle Charles' suggestion. For example, the standard "Biography at Wikipedia" could become "{{{Wikipedia}}} at Wikipedia," which would automatically give the name of the link (eg. a link to the article "John Smith" at Wikipedia would link to w:John Smith and display as "John Smith at Wikipedia") without causing the confusion it would in the current format (nor would it require any changes to header templates). - AdamBMorgan (talk) 11:31, 4 October 2010 (UTC)

Broken file - needs deleting

See Index:Devonshire Characters and Strange Events. I'm mentioning it here as it doesn't seem to appear in the category structure at all (seemingly there's a bug in mediawiki/the proofread page extension that prevents this). Could an admin delete this? Thanks. Mike Peel (talk) 21:23, 1 October 2010 (UTC)

Done

And another: Index:Encyclopedia Britannica 1911. Mike Peel (talk) 21:27, 1 October 2010 (UTC)

Not sure what the story is with this one - something about a TIFF demo & was created by User:John_Vandenberg. I'll leave a note on his talk page pointing here so he can clarify if its still needed or not. George Orwell III (talk) 21:36, 1 October 2010 (UTC)

Sorry, I didn't realise that it wasn't a broken file. Thanks for deleting the one that was, and following up on this one with the creator. Thanks. Mike Peel (talk) 22:22, 1 October 2010 (UTC)

I've deleted it. John Vandenberg ^(chat) 00:36, 3 October 2010 (UTC)

The main page looks empty

Because it's missing the October featured text? feydey (talk) 11:06, 2 October 2010 (UTC)

Fixed this the other day. I was away and no one else did it. We do try to get it done earlier, and we are looking for nominations, otherwise we are going to need to look to what we should be doing if a suitable candidate does not exist. — billinghurst sDrewth 14:21, 6 October 2010 (UTC)

Easy LST

Labeled Section Transclusions are complicated, especially for new contributors. They are probably the most difficult aspect of ProofreadPage.

Part of the problem comes from the syntax : sections must be placed within tags, and the syntax of these tags is difficult (except with the lsth syntax, but this option is not active here and it would constrain us to display section titles).

Here is the LST syntax:

<section begin=chapter1/>
blah
blah
<section end=chapter1/>

However, if you are familiar with html, you would expect something like :

<section name="chapter1">
bla
bla
</section>

The reason for that difference is that the creator of LST wanted to make it possible to transclude overlapping sections. This possibility, which is completely useless for ProofreadPage, is the reason for that complicated syntax.

In order to simplify this, I wrote a script that replaces tags with a pseudo-title :

## chapter1 ##
bla
bla

These pseudo-titles are converted to sections when the page is submitted.

I added this tool as a gadget (called easy LST) ; please test it and let me know if you experience problems. I would like to make this the default behaviour of proofreadpage, because it would mainly help newcomers.

Please note that this tool considers that any text on a page with sections belongs to a section; if some text is found between two sections, a new unnamed section is created for it. [2]

ThomasV (talk) 11:12, 2 October 2010 (UTC)

If it's decided that this will be the default behaviour of proofreadpage, will there be an easy way of opting out? (Easy=something in "my preferences"; not easy=fiddling with scripts) The Appendix to A Dictionary of Music and Musicians contains a mixture of new articles and corriegenda for articles in the main body of the Dictionary. I'm deliberately only putting section tags around the new articles because the corriegenda don't need to be transcluded. Beeswaxcandle (talk) 07:56, 7 October 2010 (UTC)

Thomas is talking about an opt-in gadget. It won't become the default behaviour of proofreadpage, because of the very issue you raise: sometimes there is spurious crap that you don't want to transclude, in between two sections that you do; e.g. a rule dividing two sections. Hesperian 09:00, 7 October 2010 (UTC)

Would it be easier to type ## Blah ## above the material not being transcluded? Instead of two section begins and section ends (i.e., four lines of text), you have three single lines. Then when transcluding just ignore the middle section. —Spangineer^wp (háblame) 12:21, 7 October 2010 (UTC)

Lol, speaking as someone who obsessively removes the references tag from the footer when a page has no footnotes, I certainly wouldn't want to inject into the page content a labelled section that I have no intention of ever using. Hesperian 06:24, 8 October 2010 (UTC)

I slightly modified the script : type #### above material not to be transcluded (no need for "blah") ; this will not add unnecessary sections.

Since this script is supposed to ease beginners work, it should be the default behaviour ; it was an opt-in gadget so that it could be tested.

It is now default. You can still opt out, by adding this to your configuration "self.proofreadpage_raw_lst=true;"

ThomasV (talk) 20:12, 19 October 2010 (UTC)

I opted out when I couldn't get this page to display properly. cygnis insignis 20:47, 19 October 2010 (UTC)

you should reload your javascript ; it was not up to date when you edited the page. ThomasV (talk) 20:54, 19 October 2010 (UTC)

The situation is complicated by there being two sections to be ignored, your undo worked, but this was done because I assume an unwanted break is introduced. cygnis insignis 21:14, 19 October 2010 (UTC)

no, the problem was an extra space at the end of the line containing the title. it is fixed now ThomasV (talk) 21:17, 19 October 2010 (UTC)

Thanks, it seems to work. cygnis insignis 21:20, 19 October 2010 (UTC)

Is it possible to still leave it as a gadget, though default to ON? I would prefer that we set it up so that it is easy to turn it off. Doesn't seem right to have the instructions not readily available, ie. you have to have read this thread on the wiki. — billinghurst sDrewth 07:10, 20 October 2010 (UTC)

sure, we can have make it a gadget. ThomasV (talk) 11:12, 20 October 2010 (UTC)

footers

"Return to the top of page" is a bit useless for very short pages. I think we should display a footer only if the page is long enough that header and footer are not both simultaneously visible ThomasV (talk) 12:56, 2 October 2010 (UTC)

I've thought about whether it is desirable, there are some other considerations. The previous and next functionality becomes redundant if the same links are visible in the browser, which depends on the device, but is otherwise useful - especially as you expect to find it. A more conceptual, perhaps vague, concern is how the header and footer serve to delineate the source from the site's navigational, notation, and other meta-content; everything between the header and footer is a transcript of the whole work or a subpaged section.

BTW, the Category part of our pages, empty or not[!], is still inserting itself above the footer where there is no transclusion. I think you (Thomas) made the fix to pages with scans already, at least the result of the change stopped it happening on my platform. cygnis insignis 07:46, 5 October 2010 (UTC)

Looking at the pages without the footer, I preferred it with the footer. It finished the pages nicely and separated from the category information and the category box, especially as they are different widths. Can we please put the footer back, at least until we have completed the discussion, there was no urgency to have it removed. — billinghurst sDrewth 10:40, 7 October 2010 (UTC)

It seems to disappear if there is no previous or next in the header, or has it always done that? Does it happen elsewhere? cygnis insignis 12:43, 7 October 2010 (UTC)

Page numbers alignment

Hi!

Does anybody knows what should be made in order to get the page numbers aligned with the beginning of each page in pt:Elementos_de_Arithmetica/Capítulo_0#4? The link takes us to the right place (the line Parecer sobre a obra..., from this page), but the page number does not. Besides, if we change browser width, the number stay in a fixed position while the text wraps to fit the new width.

Am I doing something wrong? Helder (talk) 18:52, 7 October 2010 (UTC)

I notice indented first lines lines seem to be the default, as well as the regular empty line. I suspect the line height may be factor too. These may be redundant, non-compliant, and not the way wikimedia does things. You might also look at this template, which produces

Lorem ipsum dolor sit amet, consectetur adipisicing elit, sed do eiusmod tempor incididunt ut labore et dolore magna aliqua. Ut enim ad minim veniam, quis nostrud exercitation ullamco laboris nisi ut aliquip ex ea commodo consequat. Duis aute irure dolor in reprehenderit in voluptate velit esse cillum dolore eu fugiat nulla pariatur. Excepteur sint occaecat cupidatat non proident, sunt in culpa qui officia deserunt mollit anim id est laborum.

… a hanging indent. cygnis insignis 19:49, 7 October 2010 (UTC)

What do you call a "regular empty line"?

There is this CSS #text p {text-indent: 2em;} in pt:MediaWiki:Common.css. Is it incompatible with the script wich highlights the pages and align its numbers in the left of a chapter? If so, what is the proper way to indent paragraphs?

What "line height" my be a factor? Is it the vertical spacing I've added with <div style="height:10em;" /> at pt:Página:Elementos de Arithmetica.djvu/7? Helder (talk) 20:23, 7 October 2010 (UTC)

An empty line, two returns, as you did after "Hi!".

If you look at the scan, you will see there is no empty line between them - the indent means new paragraph in print. So does an empty line. The easy way is not to indent, and changing the commons.js for all Pages is not the way to do it. It is extra work, harder to read and make nice, and not necessary.

Yes. What you will get is something that looks good on your own computer. Again, the easy thing is to avoid doing it, the reader probably doesn't care. cygnis insignis 20:38, 7 October 2010 (UTC)

This looks like a Javascript bug to me. Ask ThomasV. Hesperian 23:30, 7 October 2010 (UTC)

I do not see the problem. has it been fixed in the meantime ? ThomasV (talk) 09:29, 13 October 2010 (UTC)

Yeah, WFM now. Hesperian 11:02, 13 October 2010 (UTC)

Wep, for me too. Thanks anyway =) Helder (talk) 16:19, 22 October 2010 (UTC)

Template citation help.

The current {{DNB link}} template properly cites the original 19th century release [see template]. This link also allows for a second term to be entered that indicates the year of publication, other than the original. When this 2nd term (e.g. "01") is used, the link accesses the proper DNB01 article, yet the citation remains unchanged indicating an article published in 1901 was released between 1885-1900. The link is in use for a couple of 1901 biographies. The doc template has been annotated with "comments" that provide more accurate citations for the 1901, 1903, 1904 and 1912 releases, all made by the original publisher. Help...and thanks...JamAKiska (talk) 14:13, 10 October 2010 (UTC)

I made this in my userspace to avoid stepping on anything I don't know about. Please look it over to confirm they come out right, and if so you should be able to copy and paste it. Prosody (talk) 18:05, 10 October 2010 (UTC)

Nicely done...it is already installed on 3 pages. Thanks ! JamAKiska (talk) 19:47, 10 October 2010 (UTC)

After reviewing a large sample of the current uses of {{DNB link}} and squashing a bug related to the use of year= instead of simply passing the parameter anonymously, I've gone ahead and rolled out the changes to {{DNB link}}. I don't think there will be any problems, but please keep an eye out. Prosody (talk) 19:16, 11 October 2010 (UTC)

Thanks, I sure will. Am getting more comfortable with templates, but may need some help later this week on DNB01. Much obliged... JamAKiska (talk) 19:55, 11 October 2010 (UTC)

Manual for a U.S. ICBM system

• T.O. 21M-LGM25C-1 (AFD discussion page) (Wikipedia's Guide to AFD discussions)

Wikipedia is considering dumping a manual for the Titan II Intercontinental Ballistic Missile system onto you. Please comment in Wikipedia's AFD discussion as soon as possible if this is not what you want. Uncle G (talk) 18:49, 7 October 2010 (UTC)

• I don't think WS collectively should decide what to do with a WP article. Let someone upload it here. If it is a validly published book and not in copyright we should keep it; if not, it will be deleted.--Longfellow (talk) 09:16, 8 October 2010 (UTC)

• Looks like they deleted it from the mainspace already. Remnants can be found at w:User:Dwarner30uk/T.O. 21M-LGM25C-1. At first glance (& if its real) it would most likely fall under Public Domain / work of the US Government. George Orwell III (talk) 09:56, 8 October 2010 (UTC)

Looks like a valid page, but the contributor is still learning his way around and posted to the wrong project. So the full history was moved to a user page. I've copied it to T.O. 21M-LGM25C-1, since it does seem to fit here.-Steve Sanbeg (talk) 01:34, 9 October 2010 (UTC)

Internet Archive

I uploaded this scanned book to the Internet Archive, but several days later no OCR text or alternative file formats have been generated. How come? Does this only work when I upload PDF and not when I upload Djvu? --LA2 (talk) 21:18, 10 October 2010 (UTC)

I have no specific knowledge about the issue. Can you put it through the derivation process again? — billinghurst sDrewth 11:27, 15 October 2010 (UTC)

How do I do that? Upload it again under a different name? --LA2 (talk) 21:49, 15 October 2010 (UTC)

Nope. When logged in … Patron info > tasks that are done > Find the task, and click Mgr > Derive. 05:50, 16 October 2010 (UTC)

Did you try uploading the pdf? might be worth a try. Or get the software and do it yourself. You ought to be able to get an answer there. cygnis insignis 12:14, 16 October 2010 (UTC)

I'm learning more here, thank you! I'm having success with PDF uploads, not with TIFF+ZIP or Djvu uploads. --LA2 (talk) 21:24, 22 October 2010 (UTC)

Need a bit of Hebrew and Greek help

Please could someone knowledgeable check the bits of Hebrew and Greek text under the heading "John" on Page:EB1911 - Volume 15.djvu/459, to see if I've transcribed them correctly? Thanks in advance. - Htonl (talk) 17:17, 11 October 2010 (UTC)

Greek is OK. Charles Matthews (talk) 20:17, 11 October 2010 (UTC)

I only noticed one word there in Hebrew but it looks fine. Dovi (talk) 04:08, 13 October 2010 (UTC)

Yes, it was just the one word in each language. Thanks! - Htonl (talk) 14:25, 13 October 2010 (UTC)

A wiki table question/help

I am designing a wiki table HERE and would like to pad individual rows of cells, but the style="padding-bottom:4px;" doesn’t work. I studied the help pages of various sister wikis, but there is no reference anywhere to padding a cell/row. I am aware of the cellpadding= parameter, but that seems to apply to the whole table. Can anyone please help? - Ineuw (talk) 21:23, 13 October 2010 (UTC)

Not sure you can pad an entire row in one shot using style=. I think you need to pad all the individual column cells within that row to get that to display as desired. George Orwell III (talk) 21:34, 13 October 2010 (UTC)

George Orwell III, Thanks for the reply and you are right. Padding only works for individual cells, or for the whole table. - Ineuw (talk) 00:42, 14 October 2010 (UTC)

Wiki table help #2 requested

I would like to pad the left and right margins of all cells in THIS TABLE so that they don’t hug the border lines, but my previous assumption is wrong - the style="padding: 0px 3px 0px 3px;" in the table definition doesn’t work. Can someone set me straight as to what I am doing wrong? Thanks. - Ineuw (talk)

Nothing as far as I can tell. I tried it a couple of ways myself and nothing padded (must have something to do with the <sub>'s ??). Anyway adding the old-school line... cellpadding="3" before the table header's style= parameter(s) seems to work here. George Orwell III (talk) 23:54, 23 October 2010 (UTC)

Thanks for taking a look. The <subs> could be the reason. - Ineuw (talk) 02:09, 24 October 2010 (UTC)

• CSS doesn't allow table-wide assignment of cell padding from the table element. Instead it requires it to be done on the individual TD elements. This is logical when you assume that tables will usually conform to certain standards on a particular site, or within a particular Web page, with other named variations defined in the CSS headers. Because the access to those headers are restricted on the wiki, custom tables would need to be created by adding individual padding requirements to each cell in the table. —Mike 15:55, 25 October 2010 (UTC)

Bilingual works and pagelist

Hello, I have a doubt with bilingual works (for instance one page in Catalan/the opposite one in Spanish). Would it be possible to use a <pagelist> do display only the pages relevant to each Wikisource? For instance pages 1,3,5.. would be displayed in the Spanish Wikisource and 2,4,6... in the Catalan Wikisource. Otherwise we have to divide the book in two to be able to use the "Bilingual Extension". --Micru (talk) 11:51, 14 October 2010 (UTC)

I would suggest that you use <pagelist> and for the pages where the language is not native to the wiki of interest, label them with a hyphen/ndash, and mark them as notext for that language. So langauge one would use and language two would use When you then transclude them using <pages> the NO TEXT pages will be ignored, well that is how it works in enWS, so if it doesn't then there may need to be tweak made for you locally. If you run into difficulties, give us a call. — billinghurst sDrewth 11:33, 15 October 2010 (UTC)

I've seen the solution but I don't really understand it. See Page:Fundamento de Esperanto.djvu/13. Hesperian 12:13, 15 October 2010 (UTC)

That then would mean oldwikisource:MediaWiki:InterWikiTransclusion.js and oldwikisource:Template:Iwpage. I will look to import the documentation. — billinghurst sDrewth 13:56, 15 October 2010 (UTC)

Done and inhaled the documentation and combined to the one doc page. — billinghurst sDrewth

To note that I imported Template:Iwpages and Template:IwpageSection — billinghurst sDrewth

In, for example, this edit, Hesperian stated that we shouldn't be marking other-language pages as notext. Can one use {{iwpage}} even when the page hasn't yet been transcribed in the appropriate language? And what language parameter does one use to get the text from the multilingual WS? - Htonl (talk) 17:41, 15 October 2010 (UTC)

Generally we don't have that many experts in the interwiki/inter-language space. :-( I would have thought that we would be hosting the various components in their respective language spaces, and as that would mean we would have Index pages in respective language wikis that we would mark one without text in the other language. Truth be known, I think that we are yet to work out best guidance yet until people play and work it out through some trial and error. Maybe see what has been used at sister language wikis to see what they may be doing.

ThomasV would be the expert on how we quote oldWS and its language bits as it his application. — billinghurst sDrewth 05:58, 16 October 2010 (UTC)

A line of asterisks

Can someone show me how to properly format a line of asterisks, as I am trying to do over here? Thanks, Alcmaeonid (talk) 21:02, 14 October 2010 (UTC)

Here's three options:

******

******

******

The last is probably the easiest: {{***|6}}. Hesperian 23:29, 14 October 2010 (UTC)

Thanks, Alcmaeonid (talk) 02:35, 15 October 2010 (UTC)

Text surrounding images

Can the text be inserted around SUCH IMAGESas in the original? I uploaded the clean image HERE. My other approach would be to break up the image and separate the horizontal from the vertical. Would this be acceptable? - Ineuw (talk) 00:10, 15 October 2010 (UTC)

It can't be done without stratifying the image, and even then it is a pain. See http://meyerweb.com/eric/css/edge/raggedfloat/demo.html for more info. Hesperian 01:27, 15 October 2010 (UTC)

I agree with Hesperian, and this is one of the occasions where acknowledge that we are in a web format, and not in a book and need to present differently. The readers won't mind. — billinghurst sDrewth 11:24, 15 October 2010 (UTC)

Thank you both for the input and the link. I will use the image as is for now, but admit that it is challenging. - Ineuw (talk) 19:49, 15 October 2010 (UTC)

Text surrounding images [continued]

I segmented the image in question HERE but still can’t get the text to wrap around the image. Is this possible to fix, or I should just give up? - Ineuw (talk) 15:56, 20 October 2010 (UTC)

decline and fall of roman empire

can we make this an index page? it needs proofreading, for which index pages would be better for. - Tannertsf (talk) 08:53, 16 October 2010 (UTC)

Do you have a file(s) uploaded to Commons? When I look at http://www.archive.org/search.php?query=decline%20and%20fall%20of%20roman%20empire I see that there is a plethora from which to choose and multiple volumes. We would want to have a complete set of volumes of the same edition. Which is more authoritative? Which is better scan? Etc. then becomes the subsequent components of the choice. — billinghurst sDrewth 10:27, 16 October 2010 (UTC)

no file uploaded yet. i am still looking for the best scan and best editions...i need help though. - Tannertsf (talk) 11:00, 16 October 2010 (UTC)

I am not understanding what is the help that you desire. — billinghurst sDrewth 11:20, 16 October 2010 (UTC)

I am desiring that someone help me to find the best scans to upload and the best editions. I have the paperback book, but need to find good scans. just asking to see if anyone could help me find them. - Tannertsf (talk) 11:35, 16 October 2010 (UTC)

I like a book with a happy ending. There is a bewildering choice, but I noticed a revised edition from 1909, eg vol. 4 of 7, that is expanded with later commentary. If you go to the link on the title at the top of that page, you will find shorter and earlier editions. Avoid stuff sourced from google, the brownish-looking scans from public libraries and universities are usually complete and good. cygnis insignis 12:03, 16 October 2010 (UTC)

Thanks. What format does it need to be uploaded to Commons? - Tannertsf (talk) 12:26, 16 October 2010 (UTC)

DjVu. There are some instructions at Help:Adding_images#Text_images, and more at Help:Side by side image view for proofreading. cygnis insignis 13:11, 16 October 2010 (UTC)

Do you mind doing it for me? - Tannertsf (talk) 13:22, 16 October 2010 (UTC)

Done: Index:1909historyofdec04gibbuoft.djvu (It only occurred to me to change the filename after I clicked the upload button. Nevertheless, it's here now and it won't matter in Wikisource's mainspace.) - AdamBMorgan (talk) 01:17, 17 October 2010 (UTC)

Oops, that's just the fourth volume (I just clicked through Cygnis' link; apparently this is a poor day for my reading comprehension). Attempting to solve this now... - AdamBMorgan (talk) 01:46, 17 October 2010 (UTC)

Start with this: Index:Decline and Fall of the Roman Empire vol 1 (1897).djvu. I couldn't find any more of the 1909 edition, so this is the start of the 1897 edition instead (which might actually be lacking volume 2; at least, I couldn't find it). I'm not having a lot of luck finding this stuff on the Archive right now some I'm taking a break. More files can be added or found at Commons:Category:The History of the Decline and Fall of the Roman Empire. - AdamBMorgan (talk) 02:33, 17 October 2010 (UTC)

Someone pleae block User:Dune Harrow

This guy is currently actively vandalizing user pages. -Arch dude (talk) 00:06, 19 October 2010 (UTC)

Already done. For future reference, messages like this are best posted on the administrators' noticeboard. Hesperian 00:08, 19 October 2010 (UTC)

Another section tag convention?

Between Volumes 1 and 25 of the PSM project, I used the <section begin><section end> tags to mark the relevant article title pages. Now, I am proofreading these pages, and starting with volume 17, I noticed that the tags were converted to ## ## LIKE HERE. I searched Mediawiki and the installed plugins of Wikisource, but found no reference to this convention.

For my own knowledge, I was wondering who, and what process did this change, because the page history shows not edit activity since I made the initial edits months ago. Is this something new? Can I use this convention to identify article sections of future volumes? Finally, wish I knew about this because it would have saved lots of time (50%) and thus would have progressed further. - Ineuw (talk) 23:55, 19 October 2010 (UTC)

See Easy LST a little bit above on this page George Orwell III (talk) 01:22, 20 October 2010 (UTC)

Thanks GO III. Very interesting. It will certainly help to reduce work - past Volume 25. For now, I have to use the old way, because I inserted them without proofreading the text, and the spacing convention is a bit different. Tip of the hat to Thomas V. - Ineuw (talk) 02:54, 20 October 2010 (UTC)

Reverting to the old standard

My supposed knowledge is not showing :-). Which configuration file do I put "self.proofreadpage_raw_lst=true;" to revert to the old section tag? Vector CSS? Vector JS? Do I have to enclose it in some javascript wrapper??? Help is certainly appreciated. - Ineuw (talk) 03:10, 20 October 2010 (UTC)

if you want to revert to the old way, you need to add that option to your "vector.js", not css. before you do that, it would be helpful if you could explain which page poses you a problem ; maybe I need to adapt the script ThomasV (talk) 04:57, 20 October 2010 (UTC)

Hi Thomas V. First, I was confused with the lack of <section end> tags, and then their re-appearance after clicking Preview. The old way is begin and end tag pairs, but the LST omits the first end tag, and each click of the Preview inserts an additional end 2 tag as follows:

Top of page



<section 1 begin>

text . . . . 

<- Missing <section 1 end> ->
<section 2 begin>

text . . . . 

<section 2 end><section 2 end>

Bottom of page

You can choose any article title page where the section tags were used from THIS LIST. But now that you corrected my Vector.js, (ty) all is as before. I just want to let you know about the multiple end tags when using the LST. - Ineuw (talk) 06:28, 20 October 2010 (UTC)

I just tried these pages, but I do not see the problem. are you saying that you end up saving a page with unbalanced tags when the script is active ? ThomasV (talk) 06:36, 20 October 2010 (UTC)

It’s too late for me now, I will set it up as it was, and if it happens again, I will make a screen shot and give you a step by step of the circumstances. Thanks for the help. - Ineuw (talk) 07:21, 20 October 2010 (UTC)

---

For Thomas V: I retraced my steps and on closer attention, I noticed what is different between the two methods. In the context of my use, after saving the LST style, section 2 becomes nested within section 1 and that’s how I got two section end codes at the bottom of the page:

Top of page



<section 1 begin>

text . . . . 

<section 2 begin>

text . . . . 

<section 2 end><section 1 end>



Bottom of page

This page was sectioned in April 2010 with two consecutive paired section codes as in begin/end begin/end. After the implementation of LST, when proofread and saved, it automatically converted back to nested old section codes.

Since Volumes 1 to 25, the old section codes were inserted months ago and only now are being proofread, I will continue with the old style until these are completed because I am used to the line spacing needed to maintain consistency. - Ineuw (talk) 18:24, 20 October 2010 (UTC)

that is strange. I cannot reproduce that bug. Try to edit this version of the page, and see if it happens again. Also, what is your browser ? ThomasV (talk) 05:50, 21 October 2010 (UTC)

We have some problems with the change at the Spanish-speaking Wikisource. We do use overlapped sections, so it is becoming a mess. I don't know exactly how can I turn this feature off at es:wikisource, but I think this should be only an optional gadget (a really good one, in fact) and not a default one. --B1mbo (talk) 18:31, 21 October 2010 (UTC)

Hi Thomas, I spent a lot of time today trying but couldn’t reproduce the error either. My browser is FF 3.6.11. I retraced my steps, thinking that it may be the text editor which is set to UTF-8, and the browser which is (Windows) Western ISO-8859-1 but this made no difference. Now everything is OK, so please don’t waste your time, this editor is content. - Ineuw (talk) 22:04, 21 October 2010 (UTC)

Page:1965 FBI monograph on Nation of Islam.djvu/61

I'm a new user. Can someone take a look at Page:1965 FBI monograph on Nation of Islam.djvu/61 and see if I formatted it correctly? Thank you, Protector of Wiki (talk) 07:11, 20 October 2010 (UTC)

I validated your work and it’s fine but place the page number at the bottom in the footer section. - Ineuw (talk) 07:23, 20 October 2010 (UTC)

Okay, thanks! But why is the pilcrow (¶) on the first line not directly beside the first line of text? Protector of Wiki (talk) 07:31, 20 October 2010 (UTC)

A good question, I assume that it’s a newline below the header. It’s like this on all (proofread) pages, just look at other projects. - Ineuw (talk) 08:17, 20 October 2010 (UTC)

It is an artefact of the javascript that displays pages in Page: namespace and it is in the <noinclude> section, hence nothing about which to be concerned. — billinghurst sDrewth

New Gadget - UserMessages

• UserMessages: Adds a script for welcoming registered and IP users, with {{Welcome}} and {{Welcomeip}}. (See bottom left, below toolbox.)

Enjoy! :) -- Cirt (talk) 07:17, 20 October 2010 (UTC)

Symbol help

Symbol help : I am trying to find a symbol for male and female species markers. one is a circle with a cross subtended, the other is a circle with an arrow extended. Can anyone point me to the right place please?

On a keyboard with a separate number pad at the right, you can create them by holding down the alt key and entering 11 and 12 respectively on the number pad: ♂ ♀ --Longfellow (talk) 09:57, 25 October 2010 (UTC)

Sorry, I don't have that luxury !! Any other suggestions welcomed. David ... Actually I'll try copy and paste on the symbols which you have shown.

Yes, copy and paste should work.--Longfellow (talk) 20:59, 25 October 2010 (UTC)