User talk:Samwilson/Archive 2

Hi from Alex brollo
I took a look to your common.js and I found that you are running boldly eis.js. Ok! This means that you are testing it here into en.source :-)

I'll import your common.js into my one and I'll see what will happen.... I guess, I'll be fashinated by what I'm going to discover!

PS: I use my "test account", User:Alex brollo bis, to test anything new. --Alex brollo (talk) 15:04, 14 October 2016 (UTC)
 * Ok, I found the first (surely not the last...) issue: while importing your common.js as it is into User:Alex brollo bis/common.js, eis tool doesn't run, since a global object  (the global object I use anywhere to see what's happening inside my "iffy" and to allow objects sharing among different tools) is need into main javascript namespace. Such a need should be avoided for sure, but presently it is needed. Please add it into your common.js and test eis.js again. --Alex brollo bis (talk) 15:26, 14 October 2016 (UTC)
 * I moved eis.js here into User:Alex brollo/eis.js, please feel free to use it and edit it as you like, as yours (you've needed privileges for sure). Presently I'm using it into it.source, but I can fastly point again into it.source code if anything goes wrong. I already translated into English somethoing, Iì'll go on adding some comments into the code. --Alex brollo (talk) 07:24, 15 October 2016 (UTC)


 * Sorry for my slowness in replying. I'll let you know how I get on with eis! :-) I'll fork it to User:Samwilson/eis.js for testing. Sam Wilson 15:27, 16 October 2016 (UTC)
 * Excellent!
 * As soon as you like.... I'd like to share a couple of other exotic, but running, ideas to support a faster editing; but I feel better to deal one thing at a time. ;-) --Alex brollo (talk) 16:04, 16 October 2016 (UTC)

Bot archiving available
Hi SW, we have Wikisource-bot that is able to archive progressively, if that is of interest. It uses the standard pywikibot script from elsewhere at WMF that archives based on time passed on each section. — billinghurst  sDrewth  22:12, 16 October 2016 (UTC)
 * Hm, well maybe if I become more talkative! :-) I've only just needed to archive after 8 years... hehe. I'll keep it in mind though. Sam Wilson 01:12, 17 October 2016 (UTC)

Toolbar button for curly quotes
Hello! I just wanted to thank you for creating this toolbar button. I've been reviewing the curly/straight quotation mark debate (having not realized previously that they were even possible until I ran across them when editing an EB1911 article) and decided to use them for my project Index:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu. I went looking for a bot to do the conversion and found your toolbar button instead. Not only will it make the conversion of already-proofread pages orders of magnitude easier, but it will make it easy to edit new pages. So thank you! Laura1822 (talk) 11:10, 29 October 2016 (UTC)
 * Oh I'm glad it's of use! That's good. I use it on some novels that I'm proofreading. Curly quotes aren't super-liked by many Wikisource contributors, but they're tolerated (so long as works are consistent in their usage). Do let me know if there are any typographic situations the toolbar button can't handle. Most of the conversion steps I brought in from the Gitenberg project's clean up script. Sam Wilson 00:53, 31 October 2016 (UTC)
 * Yes, actually, I found it didn't work on a couple of pages in an article that had long quotations with some phrases in italics. There were a few opening or closing quotation marks next to italics that I had to convert manually, but usually it got them correct on the other end.  Take a look at Page:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu/98 and the following page or two.


 * Even with this imperfection, the button is still so useful I'm thrilled to have it!


 * Since you, like me, seem perhaps to prefer old-fashioned typography, I wonder what you think about emdashes. My complaint about them is that they are far too short.  I have been using bar (with |2 parameter) instead of emdashes.  I'm considering changing over to a pair of so-called longdashes, which are of course much too short for their professed purpose.  The main difference between them and bars appears to be that the longdashes are a little above the midpoint (which is nicer!), while bars are exactly at half-height.  You can review the differences at User:Laura1822/Ackermanns.  What do you think?  Is the length of dashes controlled by the font face or family in the user's browser (thus rendering the effort nil)?  If so, do you know what fonts use longer (longest) emdashes?


 * Maybe we should start a support group for people who prefer "old-fashioned" typography (much of which is still current practice in actual printing practice). I can't tell you, for example, how much I despise sans-serif fonts and would like to get my hands on the person at Apple or Microsoft in the 1990s who decided that sans serif was "easier" to read on screen while serif was easier to read on paper.  A completely bogus distinction which has since been disproved by study after study, to no avail.


 * Sorry for the rant! Keep up the good work!  Laura1822 (talk) 10:24, 31 October 2016 (UTC)


 * Yes, I'll join such a group!! :-) I'm a bit fan of getting typography right. It's one of the things that annoys me most about ereaders, but I'm discovering that lots can be done to make things better (like curly quotes). With regard to dashes, I quite agree that bars are often what's needed, and yet they're not very well supported (on ereaders, I mean; see an example here). I don't like the CSS hack of struck-through em dashes constituting a bar, but it's better than a unknown-glyph grumpy-face. I reckon we need a sort of "hint of the day" or something, in which we could have little explanations of the various typographical things, because I reckon lots of people don't know about the richness possible (and the differences in meaning). Sam Wilson 04:05, 1 November 2016 (UTC)


 * Looks like bars work (not ideally), but longdashes don't work at all. Well, that's enough to make my decision to stick with bars instead of switching to longdashes!  And the emdashes with or without hairspaces are moot, not to mention ugly.  Also, nice to see the picture of the Kobo.  I'm going to have to replace my 1st-generation Nook one of these days, and there is no way I'm purchasing another proprietary ereader.  Do you like how the Kobo handles djvu files?


 * Is that your personal wiki?


 * I like your idea for a hint of the day for our group. I too like to get typography right;  30 years ago I did some desktop publishing using Pagemaker, and spent inordinate amounts of time choosing fonts and tweaking layouts to get them just so.  One of my favorite books, of which I happen to have a first edition hardcover from 1955, and the printing is just beautiful, with good spacing and extra extra long dashes.  It is so much easier to read!  I hold it up as my ideal, wish I could format all fiction like that!


 * I see that you use the DBCustomMono font for proofreading too! When I first joined WS I asked for help to figure out how to use it.  I got the help, but contrary to my expectations no one said, "wow, what a great idea!  Everyone should be using this--let's make it the default!"  -- more like, "huh?  why would you want to use that?"


 * Some months ago I contributed an essay to Wikisource talk:WikiProject Proofreading comparing the process here to that at PGDP and suggesting improvements to implement here. The project is completely dead, however, so it garnered no response and I've been meaning to bring it up in a general discussion forum but haven't gotten around to it--don't have the energy.  Laura1822 (talk) 09:52, 1 November 2016 (UTC)


 * Yes, I think bars have to be done by combining multiple em dashes, unfortunately. At least if one day we figure out a better way, could be changed to output a proper bar. I'd love to see how other ereaders render things, the Kobo is perhaps a bit old. New ones might be different. And I do recommend them! They work with pretty much any format (although I haven't actually tried reading a djvu, instead spending most of my time reading epubs generated from Wikisource!).
 * Have you come across the Standard Ebooks project? They're trying to pay attention to great typography (and sometimes use WS books, although unfortunately when they do they usually haven't submitted their corrections back here).
 * I'm surprised that you didn't get a good reaction about DPCustomMono! I hadn't seen your write up either, and it's really good to read. :-) I absolutely agree, there is so much more we could learn from PGDP — I've done a bit of proofreading over there, and it's a pretty good experience (ultimately, I prefer being here because we keep the link to the original scans, and Wikimedia feels generally more robust and egalitarian). Do you think a gadget for enabling DPCustomMono would help (people still have to download and install it, but at least they wouldn't have to edit their CSS)? Or, actually, perhaps we should talk about making it the default as part of the ProofreadPage extension?
 * Have you heard about the Community Wishlist Survey? It starts on Monday. You must think up three proposals! :-) It sounds like you've got some good ideas, and especially ones that are not just "make proofreading better" but actually are discrete accomplishable things. (I'm going to put some proposals in, but because I'm a member of the Community Tech team I won't be voting on anything.)
 * Sam Wilson 00:29, 2 November 2016 (UTC)
 * I was unaware of the Wishlist; thanks!  I might have an idea or two for Commons as well.  I prefer WS to PGDP for much the same reasons you do:  I think the links to the original scans are important (I am always frustrated with books from PG because they do not give the edition information), and I agree that one of our strengths is that anyone can do anything, from proofreading a page to setting up a project.
 * I am somewhat of an ideas person. I am good at thinking things through, organizing them logically, and writing persuasively, but I am not as good at following through to the bitter end and my technical skills are on the low-to-moderate side (I was once a techno-geek but have had to exchange my geek card for a moderate Luddite card as tech has raced past me).  So if you are a person who makes things happen, and you like my ideas, I am really glad I met you.  :)
 * I have links on my user page to a few Jane Austen texts that appeared to have had the text imported from PG rather than freshly proofread. This is a huge problem, IMO, because the scans here on WS are first editions and the PG texts are much later;  there are significant differences in spelling and punctuation.  Having the original text reproduced exactly would be an important resource for us to have here online since academic research has in fact focused on that very issue in the recent past (some academics are arguing that the verve of her style was actually imposed by her editor, or later editors).  But even without that particular research issue, it is still important to create WS texts that exactly mirror first editions!  So I am thinking that a WS Project dedicated to finding and correcting such texts (which seems to have been a habit in the early days of WS?) would be valuable.  But only after, perhaps, a revival and revision of the Proofreading Project linked above and POTM.  Laura1822 (talk) 09:03, 2 November 2016 (UTC)

Toolbar button error report
I just noticed that on Page:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu/93, when I click the toolbar button, it mis-converts one of the italics marks on the fourth line to a curly-single-quotation mark. The phrase is "as yet." I didn't look closely at the others on the page, just noticed this one. Laura1822 (talk) 10:51, 2 November 2016 (UTC)
 * Hm, thanks for pointing it out! I can't quite see immediately why it's being stupid. :( I'll get it fixed though!
 * (Oh, you probably do it already, but it's a good idea to add any scripts you include in your common.js to your watchlist, that way you'll be alerted to any changes or discussions there.)
 * Sam Wilson 00:55, 3 November 2016 (UTC)
 * Okay, maybe is all good now? Sam Wilson 01:10, 3 November 2016 (UTC)
 * The error on that page appears to be corrected. Thanks!  You do good work!  :)  I added your script page to my watchlist;  thanks for the suggestion.  Laura1822 (talk) 11:23, 7 November 2016 (UTC)

Sorry to report another page where it seems to be confused by numbers and italics: Page:Repository of Arts, Series 1, Volume 01, 1809, January-June.djvu/12. Note that at the end of the fourth paragraph, for "15th," it works correctly, but at the end of the last paragraph, it doesn't handle "1st" correctly. (This work has a lot of this sort of mixing because apparently the printer did not have, or chose not to use, italics for numbers for the principal typeface.) Laura1822 (talk) 15:44, 7 November 2016 (UTC)

invisible button with reversed colors
Noting your discussion below, I'll go ahead and point out that I don't see the button. (I tried reloading the page (as below) and that did not fix it.) I did not report this earlier because I presumed it was related to my non-standard setup:  I use a black background with light foreground for everything (as I am extremely light sensitive), and it's pretty normal for me to find that things like buttons don't show up (for example, I can't see the previous, next, and up buttons on an editing page, nor does the switch-from-vertical-to-horizontal button appear any longer after I use the previewbut I think that in that case the button really disappears, and is not just there-but-not-visible). As long as I know what and where the button is, I don't worry too much about it. Laura1822 (talk) 11:34, 7 November 2016 (UTC)

FullScreenEditing.js
Hi. The script works fine in over/under mode, have lots of workspace. Many thanks. — Ineuw talk 05:46, 7 November 2016 (UTC)
 * Oh cool! I'm glad. :-) Let me know if you have any ideas for it. :-) Sam Wilson 05:52, 7 November 2016 (UTC)
 * The only issue is that the icon disappears, once it's activated. — Ineuw talk 06:05, 7 November 2016 (UTC)
 * That's frustrating! Do you mean, as soon as you click it and it goes fullscreen, the toolbar button disappears? :-( So you can't return to normal? Sam Wilson 06:16, 7 November 2016 (UTC)


 * Please don't be frustrated. I was told (rightly or wrongly), that Rome wasn't built in a day :-) I retested it thoroughly, and I suspect that it's a caching issue. If the user reloads the browser's current page a couple of times, all controls reappear. Which means that when activated initially, all controls disappear, meaning the user controls row of preferences etc., the page controls like read edit etc., the Charinsert bar (mine is set to display on top), and the editor toolbar (I am using the advanced version). However, a couple of page reloads and all controls return to normal. Will try to create a screenshot of what I get in full screen editing mode. — Ineuw talk 06:52, 7 November 2016 (UTC)


 * Hm, yes, I think I've seen what you describe (or some part of it). I sometimes have to reload a page to get all scripts that I've got in common.js to work properly; I think it's a fact of these scripts not being part of the normal resource-loader world, but rather addons after the fact, and with the vagaries of downloading them sometimes they get missed. :-( The upcoming Gadgets 2.0 stuff, things should get better! :-)


 * Oh, and do you like the way that fullscreen editing remembers your last-used state (it sets a cookie)? Or is that annoying? Sam Wilson 06:56, 7 November 2016 (UTC)


 * Now everything is working fine, all controls are visible in all views by opening the page for editing without refreshing. Perhaps it was a cache issue, or the cookie was not set the first time? And, I like the storing the last state feature. It's logical.  — Ineuw talk 07:03, 7 November 2016 (UTC)

Problems into la.source
As you know, la.source is a very small and poorly active project, its a pity since it could be considered the "second multilingual project" of wikisource family, since Latin is something like a international language. Its present advantage over mul.source is, that its pages can be linked fully to wikidata, while pages of mul.source can't (I don't know if this big issue has been fixed?)

I'm a (almost inactive) sysop there, and sometimes I try to import templates and tools; presently I'm trying to run FullScreenEditing.js, but I found it painful, and I'm far from satisfied by the result, since it randomly fails. Often the whole tools group (where FullScreenEditing icon should be appended) is not loaded at all. Could you take a look to general settings of that project? Is perhaps some importent setting lacking?

A question about FullScreenEditing.js: I see that it uses cookies. I abandoned them for my tools and presently I only use localStorage, with pretty good results. Is there some good reason to avoid the use of localStorage and to prefer cookies? --Alex brollo (talk) 15:33, 10 November 2016 (UTC)


 * Yes, I think the idea is to make mul.source be linkable from Wikidata, but I'm not sure of the state of progress. Can't find the phab item right now. :(
 * There are known and rather large issues about the system of using mw.loader.load to load user scripts. Basically, from what I read, there's no absolute guarantee that scripts will load, or load in order. The new gadget system will fix this, but I've no idea when that's happening. It'll mean that gadgets etc. will be minified and served along with other page scripts in the proper way. I don't really know how! :-) Do you get it to work on la.ws when you repeatedly hit reload? (not that that's a solution, but it narrows the problem down; I'm getting that behaviour on en.ws).
 * And that's a great idea re localstorage; I didn't have any particular reason to use a cookie, just that it was simple. I'll look at changing it; I'm sure it's more efficient that way, and it's probably generally good to reduce the number of cookies being set.
 * Sam Wilson 01:48, 11 November 2016 (UTC)
 * As soon as I understood what localStorage is, and how it works, we "itsourcians" used it a lot for heavy jobs too; i.e. there's a tool to add "author's references" (links to Author namespaces) that uses a localStorage object containing the whole list of Author page titles, with a pick-up box that lists existing authors for similitude of names (you know howoften authors names change from book to book). The list is more than 6000 items long, nevertheless seach is extremely fast. Obviously localStorage data are local, but they are uploaded and updated by shared js tools, so that in practice they are shared among all users that use those tools.
 * About la.source: the intriguing issue is that the right tools group disappears after refreshing the page (i.e. by Ctrl+R); this doesn't happen into it.source nor into en.source. I've been so much confused by ResourceLoader that I've been tempted to stop any js and css contribution. I'll try to convert FullScreenEditor into a gadget for la.source - just a try. --Alex brollo (talk) 11:27, 11 November 2016 (UTC)

Looking for an old script.
Hi,

I am using over and under editing, and when I had the problem of the full sized OCR display window, you gave me this script to deal with the problem. Now everything is OK, but I was wondering if the script can be modified to resize this OCR window with my specified height. I find the current window too small under certain work conditions. Thanks. — Ineuw talk 02:16, 2 January 2017 (UTC)


 * Hm, I thought the over-under setup was that it'd resize itself to be ~⅓ of the window height? Is that not happening? Maybe you've got an old version of the script; you can just use User:Samwilson/FullScreenEditing.js instead of your sandbox (there are a few differences between them now). I'll have a look at what can be improved too. Sam Wilson 05:07, 3 January 2017 (UTC)


 * In order to avoid sowing confusion, please forget about the reference to past problems, and the following is much clearer (I hope): Currently, my text area edit window is set to 11 lines, measuring 252px. The opening which displays the original page is currently 250px (using a pixel ruler) I would like to be able to open the upper window (the original page display) set by my preferred size, which I don't know but certainly more that 250px. The code pasted in the sandbox is from my archive folder. Currently, I have none of your code in my common.js and I wouldn't know where it is set. — Ineuw talk 05:41, 3 January 2017 (UTC)


 * Uploaded THIS IMAGE to further clarify. — Ineuw talk 06:15, 4 January 2017 (UTC)


 * Cool, makes sense. So you want the height of the page image to be more? Will that mean you have to scroll to get to the summary and save buttons etc.? And is this about big screens or small ones (primarily)? Do you need to be able to switch between sizes? I guess the best solution would be to have a draggable bar at the bottom of the page image, and for it to remember your previous setting. What do you think? Maybe the sort of generic improvement that should be incorporated into proofreadpage? Sam Wilson 06:22, 4 January 2017 (UTC)


 * have you first considered sizing the editable window size through special:preferences? — billinghurst  sDrewth  06:29, 4 January 2017 (UTC)


 * , I am familiar with the edit window settings, but they affect only the lower edit window. I played with the settings between 10 to 14 lines which depend on the selected font size. But, that doesn't help me see more of the original and that is my wish. Also, Thanks for the quick reply. I don't want anything fancy. Just thought that a small script in the common.js (or the .css) where I can set the window size to about ~300px so that I can read a few more lines. As for accessing the various objects and controls, I am used to scrolling up an down for access, or make extensive use of the keyboard tab key, and alt+shift+p, or alt+shift+s to get the job done. By now, both methods are 2nd nature.  — Ineuw talk 06:44, 4 January 2017 (UTC)

Author/doc
Parameters "birthyear" and "yearyear"? Is the latter a typo? --EncycloPetey (talk) 00:18, 17 January 2017 (UTC)
 * Ha! Yes, definitely. Oops. I'm still working on the docs (and still on the module too, so no doubt lots will change). I'm going to ask about it soon on scrptorium. Sam Wilson 00:22, 17 January 2017 (UTC)

I'm curious to see what happens with BCE dates, such as Author:Plato. --EncycloPetey (talk) 06:19, 30 January 2017 (UTC)

Are the unusual red-linked birth/death date categories at Author:Aristophanes the result of your work? --EncycloPetey (talk) 06:39, 30 January 2017 (UTC)


 * Yes I'm afraid so :( They're because Q43353 lists the dates as 'circa'. I take it that these shouldn't exist? We have other approximate-date categories like Category:440s BCE births but Category:C. 446 BCE births isn't quite exactly the same as that. Or we could treat it as the same, maybe? Or just leave it out all together? Sam Wilson 06:45, 30 January 2017 (UTC)


 * The dates should not be circa, unless that has been added somehow. Wikidata lists multiple birth/death dates for this person, but gives preference to one of each.


 * Also, in searching for disambiguation dates, you might wish to include parenthetical BCE dates, which will not have a number as the last character before the closing parenthesis. --EncycloPetey (talk) 06:48, 30 January 2017 (UTC)


 * Ah, that's a good point. I could fix it up to look for any parentheses that contain two or more consecutive numbers; do you think that'd catch everything and not catch the wrong things? And for the dates for Aristophanes, they are indeed circa in Wikidata (including the 'preferred' ones); if that's wrong, you can update it over there. Sam Wilson 06:55, 30 January 2017 (UTC)


 * I'd have to check the original publications from which the dates were pulled, and some of the are not in English. I'm not sure how valid the "circa" is, but there certainly is uncertainty about his dates of birth and death, as with just about anyone born that long ago.


 * Do you mean two or more consecutive numbers or two or more consecutive digits? There will be at least some dates with single digits, and some with only a birth date or only a death date. --EncycloPetey (talk) 07:08, 30 January 2017 (UTC)


 * Hm, okay, to flip it around then: are there disambiguated titles with one or more digit within the parentheses where those digits do not represent years? We could probably also do with something like Category:Authors with disambiguated titles or something. Sam Wilson 07:11, 30 January 2017 (UTC)

Writings of Thoreau scans
Hello. I had mentioned to at their Talk page that there is also an updated 20-vol set of Thoreau's Writings published later than the 11-vol set (see here). It contains more journal writings and a few more poems. EP had not time yet to see which set might be most desirable, but added, "even the 11-volume set looks to mostly duplicate works we already have in earlier editions." Just wanted to give you a heads-up and possibly save you some work. Londonjackbooks (talk) 09:24, 26 January 2017 (UTC)


 * Ah no worries! Thanks. I'll not import more for now, but will keep an eye on the page here and help out when you've figured out which is best. I wonder if The Writings of Henry D. Thoreau have published any of their research about the different collections. Sam Wilson 23:29, 26 January 2017 (UTC)

RE: Familiar Letters of Henry David Thoreau (1894), edited by Franklin Benjamin Sanborn (is this the same as vol. 11 of the 1893 collection?)

Yes, these are the same, but the copy you've linked to is a later edition (1899). The first edition of his Letters came out in 1894 (which is the one we already had). It was presumably appended later to the "1893" collection of Thoreau's Writings, but the scan you've selected is from 1899. --EncycloPetey (talk) 04:35, 8 February 2017 (UTC)


 * Hmm, okay, I think I'm probably just confusing things for you! Sorry. I'm keen to help with the Thoreau stuff, but actually I'm currently mostly looking for things to test the ia-upload tool with! Which is why I keep flying by and selecting scans to upload. Sorry, I'll stop now, it sounds like there's more complexity to the editions and dates than I thought. :-) Do you have any DjVus you want imported from IA? I can do them for you! Sam Wilson 04:40, 8 February 2017 (UTC)


 * Thanks for the offer, but no, nothing right now. Unless a file is REALLY large (like some audio files), I can manage well enough. I just did some Thoreau and Pindar uploads last week.


 * Where you might be able to help with Thoreau in that regard is to track down scans of the periodicals in which certain famous essays originally appeared. Quite a few of his essays were published in the Dial or Atlantic Monthly prior to being collected into a book. I'm not sure how many of those even exist as scans at IA, and haven't checked. But you can get particulars for several of the essays from the header notes of essays in Excursions and Cape Cod, and might be able to locate additional publication details from the preface of the Writings volumes. I'm not sure how many of those we'd want to upload and transcribe, but I'm also not sure (as I say) how many exist as scans. --EncycloPetey (talk) 04:53, 8 February 2017 (UTC)


 * The actual first edition of Familiar Letters was a "large paper" copy (15 June 1894)—150 copies printed. The 1894 copy we have hosted here I believe is the second printing as noted in Henry David Thoreau: A Descriptive Bibliography, Borst, 1982 (containing title page images/descriptions of first editions) Londonjackbooks (talk) 10:56, 8 February 2017 (UTC)

Dates are only handled once in Wikidata
In my opinion it is premature to remove the dates visibly from the templates. I would prefer that at this stage that we flag that dates of life should only be added if person is not in wikidata. Many are not, and I do not wish to miss such life data either due to people knowing and not having the ability to add, or failing to add at Wikidata for whatever reason. Until we have the means to directly edit Wikidata from enWS then we need to better guide, and supply information. — billinghurst  sDrewth  03:09, 31 January 2017 (UTC)

Strine and pottymouth ...
are mine, though I choose to not list them.—(anon)

Please have a look
An anon created Template:Birth date and age, which could have been a simple issue except that this template (which did not exist) seems to be called by several major templates, including Header and Author. Is the call to this template a result of the new Module? Are there any other similar calls to "ghost" templates that we should be protecting from vandalism / spam? --EncycloPetey (talk) 20:28, 27 February 2017 (UTC)
 * It's because of UF-hcard-person which is transcluded in Template:Header/doc and contains links to Birth date, Birth date and age, Birth-date (bday), Birth-date and age, and b-da. So it's not a problem that it's not protected. It seems it's just a copy and paste from Wikipedia. I've removed that section of UF-hcard-person. Sam Wilson 23:53, 27 February 2017 (UTC)

Circa Yo(B|D)
Can we please put it onto the agenda to categorise years of birth and death that are circa either into the designated year, or into the designated decade. I have a preference and would argue for the year, but in the end, I would just prefer to see something happen to remove potential for creation of unbound categories, or more red cats. Thanks — billinghurst  sDrewth  01:53, 24 March 2017 (UTC)
 * yes! good idea. I feel bad that I've ignored it for this long; I just wasn't sure what to do. You reckon just drop the 'c.' and use the year that's left? That's easy enough I think; will do it this arvo I hope. (Personally, I don't use those categories for anything much, so I'm not actually very sure in what manner they are used and what people expect from them.) Sam Wilson 01:57, 24 March 2017 (UTC)
 * Fixed. Sam Wilson 04:28, 24 March 2017 (UTC)
 * Would I really lay that sort of compliment on you? You just wish for that higher rating! Re change ... beaut, thx. I know that I utilise YoD for copyright stuff, YoB na-ah, though some may, and it does match xwiki. I just don't wish to have to doubled their number to cater for circa. — billinghurst  sDrewth  04:48, 24 March 2017 (UTC)

d:Property:P3919 contributed to published work
Hi. To let you know that I have shepherded through Wikidata this new property, and have now (re)populated the DNB contributor data the right way around. Obviously we have a stream of other biographical, encyclopaedic and periodicals which can be fed into this property. One of the things that I would like to look to better manage and display is how we can list contributors to compilations and periodicals within author. At the moment we utilise a string of templates like Template:DNB contributor, EB1911 ..., DMM ... and so on, especially with their string of initials. [Noting that I will be looking to feed the initials into WD when I have worked out whether I can do it with Pasleim's PLtools]. Anyway, if you look at something like Author:John Morley I would be interested in your thoughts on presentation means, and the future ability to manage it through WD. Use for categorisation is something that we may or may not wish to consider, though isn't my priority. Thx. — billinghurst  sDrewth  14:02, 1 May 2017 (UTC)


 * that's great! Good work. I've added it to d:Wikidata:WikiProject_Books. I'm not sure about author display here; we can't do cool things like tell (from an author's page) whether they're contributors-to-published-work, unless it's listed in their item. Certainly can list contributors better on works' pages though. Sam Wilson 06:40, 2 May 2017 (UTC)
 * Sorry, I understand a bit better now about the two-way relationship between P3919 and P767. So, I guess each published work will be listed on the authors' items? Sam Wilson 06:48, 2 May 2017 (UTC)
 * That is correct, recorded against author for author display. Also if we ever have portal pages for periodicals, then we can potentially pull our contributors for these works. Done that way as major periodicals would not be listing all their contributors over time, so we will need to list contributors to their works. I am unsure of exactly how and where it would be used on pages, however, that is just output and positioning. Some of my thinking was along the lines of how things like plain sister/authority control pull components and display, ie. if we populate at WD they automagically appear, or they can be manually written into a template. Numbers of the DNB authors were often noted in obituaries as being prolific in Letters to the Ed of the Times, or writing in the other prolific periodicals of the time. Knowing that, knowing initials utilised, may also allow us to accredit some of our currently initialled cum anonymous works. Have to start with data in, and build to the outcome.  Also noting that Pasleim's harvest tool cannot apply qualifiers, so he will look to see if he can bot the data into WD. — billinghurst  sDrewth  07:08, 2 May 2017 (UTC)

hiccoughing for IA-upload
log says which looks like something at IA or here has not dealt with the diacritic thingies. OR As a newly created file it was still doing its business even though it said that it was all finished. Your help would be appreciated. If these are better as phab tasks, then please say so. — billinghurst  sDrewth  22:15, 17 May 2017 (UTC)
 * https://tools.wmflabs.org/ia-upload/log/RepubliqueFrancaiseConstitution1848
 * Thanks ; I'm sure this is a daft mistake on my part for handling non-ascii characters. :-( I'll look into it today (am at the hackathon in Vienna!). And yeah, adding bugs like this to phab would be great, then they can be slotted into the fortnightly cycle of commtech work more easily. :-) Sam Wilson 07:31, 19 May 2017 (UTC)


 * Another thing, when djvu file is created by the IA Upload tool job queue, the ocr layer is frequently mis-matched with the pages. Examples: Index:Sánchi and its Remains.djvu and Index:The Gods of India.djvu. The ocr is also substandard. Lower portion of the page is frequently ocr-ed in the upper portion of the text layer, and other variants of such mixing also occur; quotes are in curly form; line-breaks not suppressed, etc. Moreover, when IA does not create jp2 folder, the djvu creation gets halted (e.g. https://tools.wmflabs.org/ia-upload/log/in.ernet.dli.2015.100758) So I got fed up with it and created the djvu offline. I think the IA upload tool ocr should use the Google OCR; not the version available with the Google OCR button here, that is sub-standard, but the version used by the OCR4wikisource software, which can handle multiple languages and scripts. Hrishikes (talk) 07:48, 19 May 2017 (UTC)

Override of living author categorisation required
At some point, can you please consider the best means to have the ability to override the categorisation of living author. With something like Author:Svayambhuva there is no pertinence to dates of birth/death or living authorship. Such pages should sit nude of such categorisation. To note that there have been multiple instances of our users forcing a vague or indistinct dates of life at Wikidata as they don't have the knowledge. I see such forced vague dates as problematic when we do wish to find appropriate authors and assign dates. I would much prefer that if dates are not known that users put in locally an approx. date or period, rather than force at WD in indiscriminate means. — billinghurst  sDrewth  05:13, 20 May 2017 (UTC)
 * Yes, I agree that making wrong data just to help with this sort of thing is bad. Sorry that the author template is so opionionated! :-) I shall fix it up. My first thought is that Q178744 should have 'no value' for birth date. This would also then stop it adding the 'era' category. Ultimately, Author:Svayambhuva should just be in Category:Authors with missing birth dates and Category:Authors with missing death dates, yeah? I've added a test to Module:Author/testcases. Sam Wilson 05:51, 20 May 2017 (UTC)
 * Nothing for which to apologise; all consequences of difference; and what we think is like for like sometimes isn't. Tweaking as we progress and develop. It all becomes on what Wikidata wants with their guidance for data, then how we implement it. One could argue that the author may not even be a human, and then we need to think what we would do if fictional human, or deity is the assignation. The other side of (big) metadata. — billinghurst  sDrewth  07:33, 20 May 2017 (UTC)
 * I've updated it to only treat humans as living. That's not strictly true, but perhaps we can live with it? Cats and fictional/mythical authors can still be given birth and death dates and they'll render as normal. (And hullo from WikiCite in Vienna, by the way!) Sam Wilson 16:23, 23 May 2017 (UTC)
 * Irony that the fake (co-)author gets a WP article, though the protagonist and author-proper does not. Moral: Never do science! Thanks for the update, and I will presume that the module will continue to evolve over time to suit requirements and quirks. — billinghurst  sDrewth  04:04, 24 May 2017 (UTC)

css fix needed
Hi. A Dictionary of the Booksellers and Printers who Were at Work in England, Scotland and Ireland from 1641 to 1667/Bellamy, or Bellamie (John) utilises sfrac nobar and class="leftoutdent", in combination the templated text doesn't move to the right and being css mostly incompetent I cannot figure out why. I tried to have the margin and text indent inherit, though that was clearly not right. Are you able to assist? Thanks if you can. — billinghurst  sDrewth  13:14, 17 June 2017 (UTC)
 * I've had a look, but I'm a bit confused! Do you mean there's something in PDBP that's going wrong just for this page? It's working everywhere else isn't it? Sam Wilson 07:02, 20 June 2017 (UTC)
 * It is broken whenever used in PDBP. The overarching class=leftoutdent and the spans don't play nicely. Something is not inheriting something else, and that is beyond me to work it out. — billinghurst  sDrewth  03:42, 8 July 2017 (UTC)

Let me try again Template:Sfrac nobar and class="leftoutdent" do not play together nicely
 * A Dictionary of the Booksellers and Printers who Were at Work in England, Scotland and Ireland from 1641 to 1667/Meighen (Richard)


 * 1) in Firefox it displays with forward thrusted fraction
 * 2) In Chrome it displays with similar forward thrust (as first example or with line split

I have tried putting an inherit into the template in various means and places, to no avail. I am clearly too clueless with such stuff. — billinghurst  sDrewth  13:06, 14 August 2017 (UTC)


 * I think it's fixed now. The problem was that the  was inheriting a -2 em text-indent from   in mainspace, so that just had to be reset to 0. Seems to work on Firefox anyway; can you confirm? Oh, and the phab files appear to be private somehow, and I can't view them (I guess they default to only be viewable by the uploader? I didn't know that.) Sam Wilson 13:49, 14 August 2017 (UTC)
 * Thanks for the fix, that has resolved the issue, and I will look to utilise nowrap to resolve the other issue. Re phab images ... really? weird! There does not seem to be any control for or against sharing their love, and just created as I would with a phab ticket. For the purpose of testing, I will add you as a subscriber to one, and learn something. — billinghurst  sDrewth  05:34, 15 August 2017 (UTC)
 * The images were restricted! Seems the name on them is their permissions, not the creator, who would have thought. You have to go in and edit them, where edit is basically permissions. Learn something every day, they are now visible fwiw. — billinghurst  sDrewth  05:43, 15 August 2017 (UTC)
 * That's odd about the files. And I've added a note to the sfrac nobar docs about nowrap. Sam Wilson 20:45, 16 August 2017 (UTC)
 * thinking a little further, can you think of an occasion that we would want it to wrap where we have parameter $3? Wondering whether we just apply it to the code, and I remove my use. — billinghurst  sDrewth  23:09, 16 August 2017 (UTC)
 * Good idea. Someone's already tried to do that, with the  class on the outer span, but that class doesn't actually seem to be defined anywhere. We could just add   to mediawiki:common.js I reckon. I'm not sure where else is assuming the existence of this. Sam Wilson 00:10, 17 August 2017 (UTC)
 * I've bin bold and reshuffled that template a bit, and added nowrap to the outer span. Sam Wilson 03:41, 17 August 2017 (UTC)

IA uploader stuck?
Hi Sam, someone tried to upload some files for me from IA to Commons and they appear to be stuck. The files are "France and the Levant peace conference 1920.djvu" and "Zionisn peace conference 1920.djvu". I was able to upload a different file, so I went back and retried the other two files, but they didn't upload. Any ideas? Avery Jensen (talk) 06:58, 2 August 2017 (UTC)


 * this appears on first sight to be an out-of-memory problem, so I'm running the conversion job directly now, and it seems to be working. I'll let you know when it's done. Sam Wilson 07:19, 2 August 2017 (UTC)


 * It completed successfully this time; the file is File:France and the Levant peace conference 1920.djvu. Sam Wilson 07:57, 2 August 2017 (UTC)


 * Ok thanks it looks good. Why does File:Zionism 9204 Peace Conference 1920.pdf look so bad? Other pdf files don't do that, for instance File:Syria and Palestine WDL11774.pdf I made an index file and it looks okay. Avery Jensen (talk) 05:56, 3 August 2017 (UTC)

barnstar
Thanks! It's great to meet you! I'm looking forward to lots of Wikisourcery this week. Sam Wilson 09:32, 9 August 2017 (UTC)

Mobile sidebar gadget
Could you please elaborate on what this does and who uses it? My technical knowledge is lacking. Thanks, Londonjackbooks (talk) 10:18, 18 August 2017 (UTC)


 * It adds a new tab next to the watchlist star, showing a little icon of a smartphone. When clicked, a right-hand sidebar opens, showing the current page in a mock-up of a smartphone, with a live preview of the page in mobile view. It's pretty cool. Takes up a bit of space, but is easily toggled on and off by clicking the tab. I guess we can just install it as a gadget and see what people think? We probably need at least a few people to say it's worth it. Oh, and you can enable it for your account on English Wikpedia to check it out. Sam Wilson 10:23, 18 August 2017 (UTC)


 * Thank you. My concern is less for myself or other Users than for mobile visitors to WS works. Londonjackbooks (talk) 10:36, 18 August 2017 (UTC)

WS Twitter account, etc.
Hello! In my tweets about poet Florence Earle Coates, I frequently reference Wikisource pointing to her works here, as well as to other authors and their works. Usually, I use the hashtag #Wikisource, or now in the future, #wikisource_en so I don't "spam" the actual accounts resulting in pesky notifications. Sometimes, however, I will use @Wikisource or @Wikisource_en if I think it may lead others to WS. But I will try not to overuse them. Don't ever feel a need to like or RT posts! Thanks for all! Londonjackbooks (talk) 01:56, 30 August 2017 (UTC)
 * No worries! It's all looking great I reckon. :) I sort of keep retweeting things cause otherwise i get spammed from twitter telling me to retweet more often hehe! :) But I'm going to try to get out one thing a day on each of the twitter accounts. Seems that twitter likes the firehose approach to things... Sam Wilson 02:00, 30 August 2017 (UTC)
 * It does :) But stick to one's purpose. Better to make a lasting impression than to gain fleeting followers :) Londonjackbooks (talk) 10:08, 30 August 2017 (UTC)

On this day in 1838, Frederick Douglass escaped from slavery. May be worth an #OTD mention at Twitter given that My Bondage and My Freedom is current PotM. Londonjackbooks (talk) 16:30, 3 September 2017 (UTC)
 * Good idea, done: https://twitter.com/wikisource_en/status/904497721490288640 Sam Wilson 00:14, 4 September 2017 (UTC)
 * Morning!... or evening! I believe the "keeper of the keys" for the WS/WSen Twitter accounts has a new role, so not sure if responsibility for accounts will be transferred at any point. The request to update enWS Twitter intro to suggested text (Wikisource—the free library—is a @Wikimedia project maintained by volunteers to create a free content library of source texts. Visit to read or help contribute!) is still pending. I sent an email request/follow-up some days ago, but I am sure these are busy times... Londonjackbooks (talk) 12:19, 14 September 2017 (UTC)
 * Morning! (It's always morning in UGT.) I've emailed Aubrie asking her to change the text. Sam Wilson 00:04, 15 September 2017 (UTC)
 * Forgot to say thanks. Thanks :) Londonjackbooks (talk) 02:22, 16 September 2017 (UTC)

ia-upload is giving error
Would appreciate if you can take a look at why the files are not getting uploaded to commons. Thank you. Shree (talk) 16:58, 12 October 2017 (UTC)

[2017-10-12 09:32:14] LOG.INFO: Unable to load XML file or no BODY element found in 'language not currently OCRable' [] [] [2017-10-12 09:32:14] LOG.INFO: Uploading to /mnt/nfs/labstore-secondary-tools-project/ia-upload/ia-upload/jobqueue/in.ernet.dli.2015.541557/in.ernet.dli.2015.541557.djvu to Commons ShriTatvaNidhi.djvu [] [] [2017-10-12 09:32:14] LOG.DEBUG: Getting fresh token {"type":"edit"} [] [2017-10-12 09:32:14] LOG.CRITICAL: Client error: `POST https://commons.wikimedia.org/w/api.php` resulted in a `413 Request Entity Too Large` response:  413 Request Entity Too Large   413 Request Entity (truncated...)  [] []


 * IA often fails to create djvu, in which case, you can add the pdf to Commons by using url2commons. In my case, IA converted to djvu this file on Oct 6 but failed to convert this file on Oct 10, so I shifted the pdf instead. Hrishikes (talk) 17:44, 12 October 2017 (UTC)


 * I tried url2commons - that is giving an error in autorization. Error retrieving token: mwoauthdatastore-request-token-not-found. IA-upload is working when djvu files are there, but does not work when choosing pdf or jp2 as source for djvu. Shree (talk) 18:42, 12 October 2017 (UTC)


 * Don't pay heed to the authorisation error. Click to authorise, and then go forward as if it is authorised. It will work, it does for me. Hrishikes (talk) 00:07, 13 October 2017 (UTC)


 * I think URL2commons requires a wikitech login, not the universal wiki login. I have now created that account and will retry. Thanks. Shree (talk) 03:53, 13 October 2017 (UTC)


 * This is a known problem I'm afraid. But luckily, it's being worked on right now! See T175680 for more info. It's because the generated DjVu is over 100 MB and is being rejected by Commons (because we're using an overly-simple upload method). Should be fixed soon. Sam Wilson 22:53, 12 October 2017 (UTC)


 * Thanks for working on the fix for this. Shree (talk) 03:53, 13 October 2017 (UTC)
 * The phab bug is about bypassing/removing the 100 mb upload barrier. But the point is, why should the size cross 100 mb? Djvu size is supposed to be lower than the pdf size when the original source file is pdf. See these examples of pending uploads: 1. 1, pdf size 45.46 MB, later uploaded by url2commons 1A; 2. 2, pdf size 48.16 MB, later uploaded by url2commons 2A; 3. 3, pdf size 26.9 MB, not yet uploaded; 4. 4, pdf size 14.25 MB, log has disappeared, later uploaded by url2commons 4A; 5. 5, pdf size 8.6 MB, successfully uploaded after djvu conversion by ia-upload 5A. So it appears that if pdf size is 10 MB or less, IA-upload has no problem with djvu conversion from the jp2 folder. For bigger pdf files, djvu size crosses 100 MB. Hrishikes (talk) 05:25, 13 October 2017 (UTC)

Books and Bytes - Issue 24
 The Wikipedia Library Books & Bytes

Issue 24, August-September 2017

 Arabic, Kiswahili and Yoruba versions of Books & Bytes are now available in meta! Read the full newsletter Sent by MediaWiki message delivery on behalf of The Wikipedia Library team --MediaWiki message delivery (talk) 04:53, 21 October 2017 (UTC)
 * User Group update
 * Global branches update
 * Star Coordinator Award - last quarter's star coordinator: User:Csisc
 * Wikimania Birds of a Feather session roundup
 * Spotlight: Wiki Loves Archives
 * Bytes in brief

Template:Authors
To avoid confusion, could I recommend instead the name Template:Author-list or Template:Authorlist? Otherwise I fear future editors will frequently become confused by the similarity to Author, which has a very different function. --EncycloPetey (talk) 21:03, 1 November 2017 (UTC)
 * Good idea! I've moved it to author-list. I've a little more work to do on it, then I'll bring it to the Scriptorium for discussion. Sam Wilson 00:34, 2 November 2017 (UTC)
 * Waving that I have and have fiddled with wikidata edition at some point as a means to recreate what I do manually at my user page. We were missing sufficient data at WD that time, though I was approaching it somewhat differently. Something that allows us to automate process and auto-generate lists rather than the crud we (badly, or rarely) do manually would be noice!— billinghurst  sDrewth  03:40, 2 November 2017 (UTC)
 * (Sorry, only just noticed this msg as well as the below.) What I'm trying to do at the moment is figure out a small first step with datadata'ing mainspace pages. Do you think the author list is a good spot to start? So far, it links where possible, and has schema.org markup, and will traverse the edition-or-translation-of property when searching for authors to list. I need some representative works to test with. Sam Wilson 03:50, 2 November 2017 (UTC)

Metadata'ing links
For our underlying templates like authority/link and article link do you think that we should be doing some internal class/id/... labelling/wrangling to these templates so there is scoopable metadata, and maybe the ability to format these more particularly if someone so chose? — billinghurst  sDrewth  03:34, 2 November 2017 (UTC)
 * yes, probably! It never hurts to have more structured data. :-) I'd say it's worthwhile, if there's a schema.org type/property for what we need. On a related note, I've been experimenting with author-link, to make links (with structured HTML) to authors; see what you think. (There are still lots of problems, e.g. we link to Portal:United States Navy (Q8373949), which isn't the same thing as the entity that we use as an author (Q11220).) Sam Wilson 03:46, 2 November 2017 (UTC)

our linking to wikipedia through header
Is it time for a more resilient, and automated means to wikilink to wikipedia out of the header than manually completing ? How able are we to have the system register to place a wikilink to a WP article based on the "WS-article Wikidata link -> WD main topic item -> target article WP-link" ? I ask as we are starting to explore the interwiki linking based on the "edition(s)/edition of" links, and this is probably a similar relationship (or maybe it isn't.) — billinghurst  sDrewth  11:20, 23 November 2017 (UTC)
 * Or said in another way when I properly WD populate something like will it start to add the pertinent WP link without manually adding it. — billinghurst  sDrewth  11:24, 23 November 2017 (UTC)
 * We definitely can get this working! I don't think it's too hard. We seem to be agreed on the work-edition relationship through P629, and we can support both that linked situation and the one in which the WS and WP sitelinks are on the same item, and the one in which there's only a WP sitelink via 'main subject (P921)' from either the edition or the work. I would probably think this could be created as a new template, and inserted into header; perhaps wikipedia-link? Or do we want to modify plain sister? That might make more sense I guess... I started that earlier this year, but it got confusing... :-) Anyway, I'm totally keen to get this working better. The other thing I'm working on at the moment is author-list. Sam Wilson 02:27, 24 November 2017 (UTC)
 * Neat. If it isn't going to be ugly to do, I think that we create as a 'separate' template that is subsidiary beneath plain sister. We could do our usual soft replacement strategy; uses active parameter if in template, though where empty that it uses WD. I think that it will be a case where we will be wanting to push the WP ref into missing "main subject" fields. Community needs to be cuddled through these changes. I would also want to see if we can have the WEF framework worked to have a means get the data entry for these (sub)pages into WD with an easy addition of main subject. — billinghurst  sDrewth  03:28, 24 November 2017 (UTC)
 * Yes, my thoughts were to add features to Module:Plain sister, doing away with all the manual listing of sitelinks, and looking everything up from Wikidata. I think the best way to do it will be to add a new exported method to that module, next to the existing  and then once we're happy it's working as we want it to, swap the invoke statement in plain sister. I'm trying to gather all test cases in Module:Plain sister/sandbox testcases, so it's easy to confirm things are working and that future changes don't break things. Sam Wilson 05:01, 24 November 2017 (UTC)
 * okay, I will try to think of the situations we may wish to link and get to it over the weekend, RL willing. — billinghurst  sDrewth  09:12, 24 November 2017 (UTC)

IA upload problem with Petty
Hi Sam,

how are you? I was quite happy to find the IA-upload tool, and I understand that you know a lot about it. I used it recently, and all went well, with Index:Petty_1647_Advice_to_Hartlib.djvu. As you can see proofreading is done already! But with another file I encounter some problems: Petty, William – Economic Writings (1899) vol 2.djvu. It is in the list of IA Upload for some days now and nothing seems to happen. The tool says it possible failed. I tried some links, but that does not help me. I don't understand a word of all the log-texts. But perhaps you can take a look at it? What has gone wrong? Is there anything I can do to prevent this kind of messages in the future? Or can I just download it, and then upload manually to Commons? Thanks in advance, --Dick Bos (talk) 08:03, 8 December 2017 (UTC)


 * Hm, sorry about this failure! Basically, some jobs are failing for various reasons; this one has come up before, and is being tracked in T159796. I'm not quite sure what's going wrong, but hopefully will figure it out soon and get things back on track. Don't worry about downloading that DjVu just yet (it will be missing its text layer); I'll have a dig into this in the morning, and see what I can find. Thanks for reporting it! Sam Wilson 10:23, 8 December 2017 (UTC)
 * I had uploaded it by alternate means as c:File:Petty, William – Economic Writings (1899) vol 2.djvu, but after Sam's explanation about text layer, I have marked it for speedy delete. Sorry for the inconvenience. Hrishikes (talk) 10:35, 8 December 2017 (UTC)


 * No I'm sorry too; it shouldn't taunt people with that DjVu download link! :-) It's sort of useful to at least have some DjVu, because for some things there is no text layer. But it's misleading most of the time. Sam Wilson 10:39, 8 December 2017 (UTC)

Books and Bytes - Issue 25
 The Wikipedia Library Books & Bytes

Issue 25, October – November 2017

 Arabic, Korean and French versions of Books & Bytes are now available in meta! Read the full newsletter Sent by MediaWiki message delivery on behalf of The Wikipedia Library team --MediaWiki message delivery (talk) 18:57, 15 December 2017 (UTC)
 * OAWiki & #1Lib1Ref
 * User Group update
 * Global branches update
 * Spotlight: Research libraries and Wikimedia
 * Bytes in brief

before I boldly ...
I have been exploring and fiddling with NOP usage for the umpteenth time as it is one of those issues that gets impacted by the minutiae changes of mediawiki. I think that for the current rendition of mediawiki parsing that nop with its empty div is problematic, especially with tables. I have now experimented with nopt (nop for tables) which is just an empty comment, and it seems to resolve our table issues, and where it is falling foul of linter extension. I would appreciate if you could confirm that. If you don't have your own pages from which to select, generally you can find some in Special:LintErrors/fostered.

As I see it, our one of suggestion is currently failing, so if we separate nop/nopt to different templates (even though that entails the difficulties of explanation), if things ever stabilise we can than just make one the redirect. I am not advocating that we stop using nop at paragraph ends as there I haven't played, and I think there we need the hard placement, and it will allow for us to modify the behaviours separately. Eseentially we have been cheating in trying to get the one template to do two similar though different placeholding.

Presuming that it is confirmed, we can then just bot replace  with   without horrendous issues. Thanks for your viewing if you have the scope. — billinghurst  sDrewth  22:28, 18 January 2018 (UTC)
 * Guys: the whole matter has been hashed and rehashed before and amounts to a slightly dodgy interaction between MediaWiki:Proofreadpage_pagenum_template and MediaWiki:PageNumbers.js in the presence of truncated table fragments (basically things go off the rails when the sequence (page number tokenising span-wrapped-span)  crops up. Tidy used to "migrate" the token upward to the top of the   tag - effectively overlaying and overlapping all page-number processing. God alone knows what happens under the current parser!) You may also consider an old experiment into reversing the problem at Template:Table-page? This isn't really a mediawiki issue but rather is an effect of the system evolving away from the assumptions made long ago unique to ProofreadPage. 114.73.1.186 06:34, 19 January 2018 (UTC)
 * Knew I'd miss a bit. Precisely the condition I described above arises on page 12 within Bauer_v._Glatzer_et_al./Second_Amended_Complaint. You will note there is no back-link to Page:Bauer v. Glatzer - Second Amended Complaint.pdf/12, although no such problem arises with pages 11 or 13. Apparently the current parser treats the token as impossible and casts it out entirely. The older system kept it but moved it to the wrong place - the situation neither gets better or worse - it is still wrong.
 * The idea of an empty variation on nop is not bad but some solution which neatly provides an innocuous DOM node to build upon (   is tempting but illegal HTML5 in      context) would be better. 114.74.130.95 aka 114.73.1.186 06:59, 19 January 2018 (UTC)
 * Yes, I know that it has cropped up before, and that it will crop again. Our issue is that we have tried to have a universal NOP that fixes everything, so we repeatedly play whack-a-mole due to code change n... n+1, and the solution may not be universal in working. I want to move to separating the solutions with separate templates, and we can worry about/fix/play/resurrect the content of the respective templates separately. I have proposed today's internals, but we can fix those internals for any future update. Sick of playing the fffffixit game. — billinghurst  sDrewth  13:30, 19 January 2018 (UTC)

Books and Bytes - Issue 26
 The Wikipedia Library Books & Bytes

Issue 26, December – January 2018  Arabic and French versions of Books & Bytes are now available in meta!
 * # 1Lib1Ref
 * User Group update
 * Global branches update
 * Spotlight: What can we glean from OCLC’s experience with library staff learning Wikipedia?
 * Bytes in brief

Read the full newsletter Sent by MediaWiki message delivery on behalf of The Wikipedia Library team --MediaWiki message delivery (talk) 17:35, 31 January 2018 (UTC)

Bulletin
It looks as though The Bulletin is the newspaper, and "The Bulletin Newspaper Co." is the Company, if you're categorizing by publishing company. --EncycloPetey (talk) 23:26, 21 February 2018 (UTC)
 * Hm, good point. I guess that was its official title when On Our Selection was published (is usually known as just The Bulletin). I'll change the cat and the portal. Thanks! Sam Wilson 23:30, 21 February 2018 (UTC)
 * You can probably leave the Portal name as-is. The magazine still had that name. --EncycloPetey (talk) 23:32, 21 February 2018 (UTC)
 * Ah, good point. And I'll create a new wD item for the company as separate from the magazine. Sam Wilson 23:35, 21 February 2018 (UTC)

Index:The Roman index of forbidden books.djvu
Mind if I work on this one? —Beleg Tâl (talk) 01:32, 25 February 2018 (UTC)
 * not at all! :-) I can't even really remember why I uploaded it. Perhaps I was testing something. All yours though, go for it. Sam Wilson 09:37, 25 February 2018 (UTC)

ia-upload build process clearly hates me
tells me that it has failed, though not an indicator of why. It appears to have taken all the pages, though not constructed at the end. Timeout? Too big? Phabricator ticket? Walk away cover my eyes? — billinghurst  sDrewth  00:49, 26 February 2018 (UTC)
 * Hm, yeah this is bad I'm afriad. There's a few of these failing like this. T159796 is tracking it, but there's no great info there yet. Alex Brollo has a better approach to these conversions, but I've not yet found time to change ia-upload to that system. :-( Sorry to not have anything more positive to say! I am, on the other hand, still hoping to hear back from the IA that they'll reinstitute DjVu generation (they have at least not definitively and finally said they will not). Sam Wilson 00:58, 26 February 2018 (UTC)

Share your experience and feedback as a Wikimedian in this global survey
Hello! The Wikimedia Foundation is asking for your feedback in a survey. We want to know how well we are supporting your work on and off wiki, and how we can change or improve things in the future. The opinions you share will directly affect the current and future work of the Wikimedia Foundation. You have been randomly selected to take this survey as we would like to hear from your Wikimedia community. The survey is available in various languages and will take between 20 and 40 minutes. Take the survey now!

You can find more information about this survey on the project page and see how your feedback helps the Wikimedia Foundation support editors like you. This survey is hosted by a third-party service and governed by this privacy statement (in English). Please visit our frequently asked questions page to find more information about this survey. If you need additional help, or if you wish to opt-out of future communications about this survey, send an email through the EmailUser feature to WMF Surveys to remove you from the list.

Thank you! WMF Surveys, 18:36, 29 March 2018 (UTC)

Reminder: Share your feedback in this Wikimedia survey
Every response for this survey can help the Wikimedia Foundation improve your experience on the Wikimedia projects. So far, we have heard from just 29% of Wikimedia contributors. The survey is available in various languages and will take between 20 and 40 minutes to be completed. Take the survey now.

If you have already taken the survey, we are sorry you've received this reminder. We have design the survey to make it impossible to identify which users have taken the survey, so we have to send reminders to everyone. If you wish to opt-out of the next reminder or any other survey, send an email through EmailUser feature to WMF Surveys. You can also send any questions you have to this user email. Learn more about this survey on the project page. This survey is hosted by a third-party service and governed by this Wikimedia Foundation privacy statement. Thanks! WMF Surveys, 01:34, 13 April 2018 (UTC)