User talk:Dr pda/archive3

Your most excellent presence is requested
... urgently, at Wikipedia talk:Featured article candidates. We need some idea on articles with word counts below certain cutoffs :-) Can you give us a rough idea to move forward on this?  WT:FAC is in turmoil, and I belive this is the best solution.  Sandy Georgia  (Talk) 15:42, 30 September 2008 (UTC)
 * Provided figures there. Also updated User:Dr pda/Featured article statistics. Obviously I'm in a different timezone to everyone posting at WT:FAC, judging by the amount of stuff written since yesterday! Dr pda (talk) 03:51, 1 October 2008 (UTC)
 * Yes, the place goes wild at a certain time of day :-) Doc, what is the date on User:Dr pda/Featured article statistics?  Can you put it at the top so I'll remember in the future?  Thank you SO much for doing that work; if I had known you were going to have to do it manually, I would have done it myself.  All the best, Sandy Georgia  (Talk) 04:02, 1 October 2008 (UTC)
 * Whoops, I lost the header in the copy and paste, it's back now with today's date. I could have modified the script which makes the list to do the word counts as well, but that would have taken some time/testing, so I figured it was faster just to do it manually. (Plus the script which makes the list uses a slightly different method for calculating things compared to the prosesize script, for technical reasons, and this could have led to some confusion). Dr pda (talk) 04:10, 1 October 2008 (UTC)

graph software
What did you use to make Image:FA and GA word count.svg, gnuplot? or something else? Thanks. --Rajah (talk) 15:01, 2 October 2008 (UTC)
 * I used ROOT. Dr pda (talk) 20:07, 2 October 2008 (UTC)
 * Cool, thanks. --Rajah (talk) 22:14, 2 October 2008 (UTC)

templatecheck.js
I can't get it to work on WikiProject Persondata/List of biographies/1. It used to though. Does it work for you? Thanks. --Rajah (talk) 22:14, 2 October 2008 (UTC)
 * nevermind, got it. (Although it didn't work on section 1, but did on section 2.) --Rajah (talk) 22:15, 2 October 2008 (UTC)
 * Yes, the number of entries per section is hard-coded to be 100, and someone's deleted one entry in the first section on WikiProject Persondata/List of biographies/1. Given how few people use the script I haven't bothered trying to fix it. One other thing you might be interested in—an easy way to count the number of struck-out items on a page. Just create a bookmark with the following for the url . This pops up a message box with the number of  tags. If I ever get the time I mean to write some sort of script to go through and strike out all the names on those lists which already have persondata. Dr pda (talk) 00:12, 3 October 2008 (UTC)
 * Nice, thanks for the explanation. It may have even been me that removed that one entry. I'll do that javascript thing now. --Rajah (talk) 06:09, 3 October 2008 (UTC)

Foiled script
Doc, there's something about the Law article that is foiling the page size script on IE7; the link doesn't even pop up in my toolbox, although I'm fine with any other article. Sandy Georgia (Talk) 23:47, 4 October 2008 (UTC)
 * Very odd. I don't see why it should stop working for that article. Unfortunately I don't have IE7 so I can't check this directly. The prose size script works fine for me in Firefox and IE6. For what it's worth the current size is

Dr pda (talk) 01:04, 5 October 2008 (UTC)
 * File size: 320 kB
 * Prose size (including all HTML code): 84 kB
 * References (including all HTML code): 44 kB
 * Wiki text: 119 kB
 * Prose size (text only): 49 kB (7945 words) "readable prose size"
 * References (text only): 13 kB
 * Images: 350 kB


 * Thanks .. weird, huh? Sandy Georgia  (Talk) 01:07, 5 October 2008 (UTC)


 * hmmm, went to my son's Mac, Safari, and Page size is there on Law, so it's an IE thing. Sandy Georgia  (Talk) 01:12, 5 October 2008 (UTC)

More follow up at Talk:Law. Sandy Georgia (Talk) 08:30, 10 October 2008 (UTC)
 * I tried looking at the page on one of the other computers at work, which does have IE7, and also experienced the same problem. I found and fixed what was causing it, details at Talk:Law. Dr pda (talk) 04:55, 11 October 2008 (UTC)

Articlehistory script
Could I possibly prevail upon you to look into an issue between your articlehistory script and Firefox 3.0+? In FF 3.0 the script used to just error out completely (cue wails and wringing of hands); in 3.0.3, I am thrilled to see that I can get AH to output the list of events, but I can't get oldid on click. If I were a smart person, I might use this as an excuse to take a break from adding Articlehistory to old failed FACs, but we only have about 180 left to go and I'm *really* stubborn. Can you help save me from myself? Maralia (talk) 22:53, 13 October 2008 (UTC)
 * Hi Maralia, the problem may not actually be with FF3. I've just tried it on FF2 and I got the same result (i.e. the list of events comes up, and if I click for the manual oldid I get the dialog box to enter the date, but then nothing). I've tracked this back to a bug I introduced a month or so ago when converting the script from the old query.php to the current api.php system. I think I've fixed this now, so do a forced refresh of your monobook.js, and see if it works for you now. Dr pda (talk) 09:43, 14 October 2008 (UTC)
 * That did it - thank you! The AH script saves me so much time. If I could be so rude as to follow up one request with another: editrefs is also not fully working for me since FF 3. The list comes up right, but clicking the 'Apply changes and preview' button takes me back to a normal preview screen without applying the ref changes. Any ideas? Maralia (talk) 14:04, 14 October 2008 (UTC)
 * editrefs works for me under FF2, so this probably is a browser issue, however I don't currently have FF3 installed on any of my machines at home or work. I'll see if I can find time to upgrade one of my installations, but this might have to sit on the back burner for a while. Does the javascript error console show any error messages? Dr pda (talk) 09:49, 15 October 2008 (UTC)
 * I get a fair amount of errors in the console—serves me right for snitching someone's interface and never debugging—but none of them appear related. I wiped my .css and .js (except for editrefs), then toggled everything else on one at a time, and WikiEd appears to be the culprit. As I can toggle WikiEd off within an edit screen, there is at least a workaround for now. I couldn't tell if you're running WikiEd, since it's an option in Gadgets now; could you check if editrefs works with WikiEd enabled in FF2? Maralia (talk) 16:39, 15 October 2008 (UTC)
 * If I enable WikEd as a gadget (I don't currently use it) then I get the same behaviour with editrefs. I suspect that wikEd uses its own copy of the text in the edit box; a quick glance at the source code confirms this. My 'Apply changes and preview' button is therefore not modifying the 'active copy', as it were. I suppose I could try and modify editrefs to get the wikEd version of the textbox in the case that it is being used, but that would require me to understand the wikEd code. For the moment I think you'll just have to live with the workaround, i.e. toggle to classic text view, edit the refs and toggle back to wikEd view. Dr pda (talk) 00:24, 16 October 2008 (UTC)
 * I can certainly handle clicking a button before I use editrefs; it's just a relief to know I can use it—and articlehistory—at all now. Thanks again for your help! Maralia (talk) 01:58, 16 October 2008 (UTC)

Order of the Bath
I've added a citation to the Court Circular for 17 May 2006. This gives all the officers except the Dean, who is the Dean of Westminster, ex officio. I've added citations for the two officers I know to have changed since then (Registrar and Secretary and Scarlet Rod). Thanks for keeping an eye on this FA. Choess (talk) 15:26, 14 October 2008 (UTC)

Prosesize.js bug
The byte counts are erratic sometimes, and now I have this example. I selected a version because the page is likely to be edited. Prosesize.js says "Prose size (text only): 1061 B (271 words) "readable prose size"", highlighting the first two paragraphs before the Table of Contents (and also highlighting several blank lines between display boxes). Microsoft Word says those two paragraphs are 1531 bytes. AllWorldPhone.com agrees (almost; 1534 bytes). Art LaPella (talk) 01:29, 16 October 2008 (UTC)
 * The prose size script was never really designed for list-type articles, as it counts the stuff between   tags, and lists often have bulleted lists or tables, which are different HTML entities. However the example you give illustrates another point—I had assumed that all the references would be within the readable prose, which is not the case here or any other article where there is a reference in a list or table, say. I subtract the size of reference markers, i.e. [1] from the readable prose size, both text and HTML versions; doing so in this sort of situation gives incorrect values. (Notice that the HTML prose size for the Mortal Kombat list is actually negative!) I have updated the script to fix this problem. You will need to bypass your cache to pick up the corrected version. It now gives the size of the prose (i.e. first two paragraphs) as 1522 bytes, i.e. 1531 - 9 bytes from [1][2][3]. Dr pda (talk) 06:08, 16 October 2008 (UTC)
 * It works! Thank you. Art LaPella (talk) 20:12, 16 October 2008 (UTC)

Alternate version of prosesize.js
Hi,

First let me thank you for your excellent script, which I used in my research presented at Wikimania 2008. However, I found that I didn't quite agree with how your script worked, on a few technical points. Not counting the bytes in lists was pretty significant for the accuracy of my conclusions, so I tried my hand at writing my own script here. I've only just learned to code in JavaScript, and the result is somewhat kludgy, so my apologies in advance; it's much less elegant than yours. As you'll see, the purpose of the script is a little different; I'm trying to analyze the internal structure of articles, such as the distribution of prose within an article. As a byproduct, the script also counts images and detects errors in heading levels, which is useful for accessibility.

I made a test article, User:Proteins/Prosesize_script_acid_test, that incorporates many of the pathologies for prose-size scripts that I've encountered in Wikipedia's menagerie of articles. It might be helpful to you, and I would be grateful if you could suggest improvements to it and to my own script. As my conventions, I assumed that a prosesize script should


 * Not count blank lines as characters or words
 * Count words and characters in lists, quotes, blockquotes, and cquotes
 * Count indented text, &lt;poem&gt;s, and &lt;PRE&gt; sections
 * Not count material in tables, family trees, tennis match trees, and other graphical representations
 * Count special characters such as &amp; as a single character; also &gt; and &lt;
 * Not count lines in the concluding sections of the article, such as references that aren't part of a list, or explanatory text in the "See also", "References" and "External links" sections
 * Count words separated by an unspaced em-dash, or by non-breaking spaces.
 * Count simple superscripts such as x2 (your script does this, I think) but not tags such as this one.

I hope you won't take it amiss that I wrote my own script; I needed it for my research and it was a good impetus to learn JavaScript. As you'll see, I still have to figure out how to make a common string-editing function. I look forward to learning much more about scripting from you; thank you! Proteins (talk) 22:29, 17 October 2008 (UTC)


 * Hi Proteins, thanks for your comments. No I don't mind that you used my script; posting it on Wikipedia means it's licensed under the GFDL anyway. There were two main reasons for not including lists—first, it required extra effort, since they weren't within   tags and lists are used for many non-prose purposes eg navboxes, references, citations etc, and second, the definition of prose size at WP:SIZE excluded them. Some of your other points may indeed be worth implementing in my own script, although they will probably only be higher-order effects. I'll add this to my list of things to do. You may be interested to know that prosesize.js was also my first real foray into javascript! Dr pda (talk) 10:39, 20 October 2008 (UTC)

Thanks for your friendly letter! I'm enjoying my foray in JavaScript; it's remarkable what you can do, isn't it? I'm still having trouble getting this particular script to work in Internet Explorer, I think because of an anomalous Document Model, but I'm working on it. You've probably bumped up against that sort of browser-inconsistency problem before. Luckily, the script seems to work in the other browsers.

I took a slightly different approach in my script. Since I wanted to count the words in lists, I had to identify the beginning of the closing sections such as See also, References, Notes, etc. and then stop counting just before reaching them. That seems to agree with the definition of "readable prose" in WP:SIZE.

May I impose on you for some help? I'm perplexed by some significant discrepancies in our script results. It's not surprising that our scripts disagree on articles with lists, such as Order of the Garter (34% disagreement in word count), acute myeloid leukemia (14%) or Banksia integrifolia (10%). Quotes seem relatively rare and thus cause relatively little disagreement, such as William Shakespeare (1%). On other articles, however, where our scripts agree more-or-less on the word count, your script typically counts roughly 17% more bytes than mine does? I tried it out on ten Featured Articles, and it's roughly consistent. For example, you might try it out on Uranium (37213 vs. 31880 bytes), History of Indiana (57231 vs. 48525 bytes) or Roman Catholic Church (72361 vs. 61532 bytes). Do you have any idea why the scripts disagree so much? I probably have a bug of some sort, but I haven't found it as yet, despite some detailed testing. (There's a flag you can set in my script to see all the words, along with their byte-counts.) I'd appreciate your thoughts, since it's important to me to get this script working correctly. Thank you, Proteins (talk) 23:00, 20 October 2008 (UTC)


 * I thought about doing that with lists, but decided that trying to account for all the different names of the closing sections was not worth the effort.
 * I've had a quick glance at your script; it appears that you are only counting the number of bytes within the words, whereas I count the spaces between them as well. From other studies I've done the average number of characters per word is about 6; adding one space per word means that my byte count should be 7/6 times yours for the same word count, which is an increase of about 17%. Dr pda (talk) 23:56, 20 October 2008 (UTC)

My script does count the spaces, one per word; that was the sense of "prose_size_bytes += (num_words - 1)"? Some words might be separated by more than one space, but I thought the extra spaces shouldn't be counted. Thanks for pointing out the obscurity in the code; I made the code a little clearer there for others.

Our character counts seem to agree pretty well in the lead section; I also got 1522 bytes in the Mortal Kombat example above. I've made two test pages where we can track down the discrepancy, if you're game for that: User:Proteins/Sandbox and User:Proteins/Sandbox2. They're both taken from William Shakespeare, where our scripts differ significantly: my script gives 31003 bytes in 5855 words, whereas yours gives 35948 bytes in 5799 words. I think our total word counts differ primarily because of the "upstart Crow" and "Sir, in my heart" quotes, but the extra bytes are mysterious.

For User:Proteins/Sandbox, our counts are pretty close. My script gives 2539 bytes in 401 words, whereas yours gives 2548 bytes in 417 words. I turned on the word-by-word counting for my script, so you can perhaps see where the slight disagreement comes from. Note that the word count difference is 16 words = 417-401.

However, the addition of the "Plays" section in User:Proteins/Sandbox2 throws off the agreement. My script gives 10608 bytes in 1947 words, whereas yours gives 12151 bytes in 1963 words. Our word counts are again off by 16 (1963-1947), which is heartening. But your byte count is almost 15% higher than mine? Perplexed, Proteins (talk) 04:14, 21 October 2008 (UTC)


 * I've had a bit of a play with those Sandbox articles offline. The sixteen words difference comes from empty  tags, plus the paragraph containing the interwiki links, which contains 9 spans separated by spaces, so that's something I should fix in my script (though it wouldn't be noticeable in the main version, which displays sizes in kB). However the big disagreement does in fact come from not adding the spaces in your script. The line

prose_size_bytes += num_words - 1
 * correctly adds the number of spaces, but the problem is that it comes after you've done

if ((!show_section_diagnostics) && ((!show_lead_diagnostics) || (H2_anchor_index != 0))) { continue; }
 * If the condition is satified, then the script will, i.e. skip the rest of the loop, which includes the statement to add  . This does not manifest in the lead (or equivalently User:Proteins/Sandbox, where the only prose is in the lead), because show_lead_diagnostics is currently true. If you set it to false, articleStructure counts 401 words, 2142 bytes. The solution is just to move the line   to before the   clause (and similarly in the case of the lists)Dr pda (talk) 08:44, 21 October 2008 (UTC)

Thank you, Dr pda! That seems to have been the problem, and I fixed it as you suggested. It's ironic that the bug was caused by debugging code, but was only visible when debugging was shut off. A kind of Schroedinger bug, one that disappears when you look at it.

I ran both scripts on a test suite of 22 randomly selected Featured Articles, and for roughly half of them, the agreement in byte-count was excellent, sometimes exactly the same. The chief remaining discrepancies came with articles with long lists or quotes; examples include Søren Kierkegaard, Casablanca (film), Nintendo Entertainment System, Emmy Noether, Learned Hand and Aramaic language. Both of our scripts fail with the anomalous article Chess, because an editor inserted invisible DIV's throughout the History section, for unknown reasons. Thanks again for your help, and please let me know if I can return the favor, Proteins (talk) 13:29, 21 October 2008 (UTC)

Why subst?
First let me say thanks for your nifty scripts. But I wonder, why do you advise people to do a subst: ( User:Dr pda/prosesizebytes.js ) rather than to do an import script: ( importScript('User:Dr pda/prosesizebytes.js'); ) ... I do the latter and things seem to work, and I will thus pick up any changes or improvements you make. Am I missing something vital? The script seems to work fine for me imported instead of copied in. Thanks for any insight into this. ++Lar: t/c 15:09, 19 October 2008 (UTC)
 * That subst's, which creates code which effects an import. It may be the same code behind importScript. Those using the code get updates, too. Gimmetrow 15:27, 19 October 2008 (UTC)
 * Ah, excellent, I missed that. :) ++Lar: t/c 18:35, 19 October 2008 (UTC)
 * As of February 2007 js in fact just produces importScript('User:Dr pda/prosesizebytes.js'); //User:Dr pda/prosesizebytes.js My first script was written prior to this when the expression the template produced was more complex. js also produces a handy link to the script, so I can see how many people have installed by clicking "What links here". Dr pda (talk) 10:39, 20 October 2008 (UTC)

FAC and FAR templates
Hey, doc. We will need to resurrect the old and, and  and , so that the talk pages can be gradually botified into articlehistory after closings. I'm trying to understand the parameters in these templates; it seems they're not all the same. Some of them seem to allow an oldid parameter, others don't; some seem to allow for name changes, others don't. But I'm not sure I'm following, and I usually asked Gimme questions like this. Can you explain what parameters I should pass to talk pages when these close, or should we pass no parameters, and just let your articlehistory script pick them up? It would be truly ideal if each of these templates would reflect date and oldid in the template, so we didn't have to go to your script for any additional data. Sandy Georgia (Talk) 21:26, 22 October 2008 (UTC)
 * Of those four templates, only facfailed didn't seem to take an oldid= parameter; I added it there. They could all use an overall review for consistent parameters and syntax, though, like you said. If Dr pda doesn't beat me to it, I'll take a look later tonight when I have less distractions. Maralia (talk) 21:48, 22 October 2008 (UTC)
 * Thank you, Maralia. Perhaps someone can put the order of the parameters on User:SandyGeorgia/FA work, so we'll all know how to use them.  Sandy Georgia  (Talk) 21:54, 22 October 2008 (UTC)
 * Since Gimmetrow is back I guess this has become less urgent, however it's worthwhile to have the usage documented, so I'll add instructions to the template pages tonight. Dr pda (talk) 01:56, 23 October 2008 (UTC)
 * OK I've added documentation to each of the four template pages, so it should hopefully be clearer now how they are used. Dr pda (talk) 09:58, 23 October 2008 (UTC)
 * Thank you so much Dr pda; very nice work, just what I needed. (Well, just what I hope we won't need now that Gimme's back, but I'm glad it's done :-)  Sandy Georgia  (Talk) 19:54, 23 October 2008 (UTC)

where did he get those wonderful toys?
Hi,

Dunno if you know me; I've been around a while. Among many other things, I was a programmer for 4+ years, in a previous career incarnation. I'd like to learn to write scripts that will help Wikipedia... so .. um.. where would I start? Tks Ling.Nut (talk&mdash;WP:3IAR) 14:10, 25 October 2008 (UTC)

Hi Ling.Nut, I've certainly seen you around, but I can't remember if we have interacted. You may find some useful information at WikiProject User scripts. For myself my prose size script was the first one I wrote, and that was because I thought there must be a better way than copying and pasting into Word. I had no experience of javascript prior to that (although I did have other programming experience), so once I worked out what I wanted to do, I just googled for information on javascript to work out how to implement it. The w3schools website seemed to come up quite a lot in my searches, so I often use this for syntax reference. I also looked at other scripts on Wikipedia to see how they worked. And then having written one script, when I came across some other tedious task, it was easier to see how to write a script to do it (and of course I could reuse code!). Hope this helps. Dr pda (talk) 05:28, 26 October 2008 (UTC)

Script
Your article size script used to work for me but not now. Can you check out why? Tks.  — Rlevse • Talk  • 03:19, 2 November 2008 (UTC)
 * The prose size script is still working for me. What do you mean by not working? Do you get any javascript error messages (in Firefox see Tools->Error console). Have you recently added any other scripts to your monobook.js? Have you recently enabled any gadgets under My preferences? Have you recently changed your browser or operating system? Have you tried running the script on more than one article in case there's a problem with the article? Have you tried bypassing your cache in case the cached copy of the script has become corrupted somehow? Dr pda (talk) 03:31, 2 November 2008 (UTC)
 * Argh, it's working now. Go figure. Thanks anyway.  — Rlevse • Talk  • 12:28, 6 November 2008 (UTC)

Portal:Heraldry
When you have a chance, do you have any suggestions for P:HV? Gimmetrow 03:36, 2 November 2008 (UTC)
 * I gather this is with a view to making it a featured portal. Thanks for continuing to maintain (and automate) it by the way, it's been several months since I've really done anything with heraldry-related articles. The only real comment I have is to wonder whether the intro section is sufficiently obvious as an intro in its current location. The lack of a red header sets it apart from the other sections, but the lack of a bold header means it might get overlooked. I see you've got enough material to rotate through.Dr pda (talk) 10:09, 6 November 2008 (UTC)

Quick note
Just added a new section at Horses_in_warfare Basically expanded the section by a couple paragraphs. Because it's all new, would you be so kind as to take a quick peek and comment on anything that will mess it up at FAC? (diffs here if easier) I've sandboxed it for a bit on the talk page, tried to keep it short. Any thoughts appreciated. You are an excellent, and straight to the point, reviewer, and I want to thank you very much for that. Montanabw (talk) 07:06, 10 November 2008 (UTC)
 * Commented at the peer review :) Dr pda (talk) 09:10, 11 November 2008 (UTC)

Sort query
Doc, if you have time to look, is there a way to get the sort down to the second-level heading at User:SandyGeorgia/ArbStats? Sandy Georgia (Talk) 03:18, 21 November 2008 (UTC)
 * I don't think there is. According to meta:Help:Sorting Javascript sorting may not work properly on tables with cells extending over multiple rows and/or columns. I've used one other trick from there to make at least the other columns sort properly diff. I've only applied it to the ArbCom table, not the Sample one, so you can see the difference it makes. Feel free to revert if it doesn't help. Dr pda (talk) 04:08, 21 November 2008 (UTC)
 * If it helps, there's a very tedious, manual kluge that you could choose "one" of the second-level headings to sort by. Forex, "Mainspace" currently sorts alpha by article name. You could force the sort order to fall in line by, say "edits" using aa , ab  etc. appropriately. But it would only let you choose one field to sort by. Ling.Nut (talk&mdash;WP:3IAR) 05:29, 21 November 2008 (UTC)
 * Or actually move that top row to second level, leaving those important headers sortable. Or make that top row a standalone table, wrapped along with the actual sortable table inside another table (theoretically, anyway - don't ask how long I've just spent trying to actually make that work). Maralia (talk) 05:56, 21 November 2008 (UTC)
 * Well, I'm a klutz. I can't make even the simple columns (like total edits) sort correctly.  Tomorrow's another day.  Thanks, all !  Sandy Georgia  (Talk) 07:03, 21 November 2008 (UTC)

Shortest good articles?
Would you be able to whip up a list of some of the shortest good articles by prose? Gary King ( talk ) 16:17, 23 November 2008 (UTC)
 * I could produce such a list, but none of my current scripts will do this without some modification (unless you want data which are 5 months out of date!) The script which I use to do the equivalent thing for featured articles works because the articles themselves are tagged, rather than just the talk pages. I will try to produce the list for GAs, but as usual I have many more things on my wiki to-do list than I have time to do them. I may be able to get to this within the next week or so. Dr pda (talk) 00:29, 24 November 2008 (UTC)
 * Okay thanks, I look forward to that! Gary King  ( talk ) 15:59, 28 November 2008 (UTC)
 * The list is now available at User:Dr pda/Good article statistics. Enjoy! Dr pda (talk) 00:41, 2 December 2008 (UTC)
 * Okay thanks Gary King  ( talk ) 01:28, 2 December 2008 (UTC)

User talk:Dr pda/prosesize.js
Where and when will the output display when I implement this script? - Mgm|(talk) 12:13, 28 November 2008 (UTC)
 * Hi, if you install the script it adds a link (Page size) to the toolbox in the left-hand column of the page (under What links here). If you then go to your favourite article and click this link, the script highlights the readable prose in yellow, and uses javascript to insert the statistics at the top of the page. Hope this answers your question. Dr pda (talk) 00:41, 2 December 2008 (UTC)

Your attention
Your attention here. Many people on DYK recently, including Politizer, Gatoclass, etc, tried to claim that WP:SIZE is not the consensus based definer of what prose is, and they claimed that quotes and the rest don't count as prose. As I pointed out, your own instructions rely on Size and even refer to the fact that it wont count things like cquote, even though they count as prose. People are trying to redefine prose to exclude other pages. My one page, for example Samson Agonistes, measured as around 16k characters based on other word counters, but one individual claimed it was only at 11k. Why? Because of quotes and the rest. Every encyclopedic page that I work on includes quotes, either from famous scholars, from the text, or from the author that gives an important insight into the text. However, they are suddenly trying to claim what "prose" is to make that impossible. Ottava Rima (talk) 16:25, 14 December 2008 (UTC)
 * Thank you for the notification. Dr pda (talk) 12:34, 15 December 2008 (UTC)

not sure if you'd be interested but..
I noticed you worked on the Order of St. Patrick and brought it up to featured status; although there wasn't as much work involved I currently have the List of Knights of St Patrick up for Featured List status and your comments would be greatly appreciated. Thanks in advance. Ironholds (talk) 04:30, 15 December 2008 (UTC)
 * Actually I just prevented it from losing featured status during a Featured article review :) I've left some comments on the list at its FLC page. Dr pda (talk) 12:34, 15 December 2008 (UTC)

Need eyes again
Doc, can you peek in at Talk:Horses_in_warfare and see if you have any comments/suggestions to break us loose? I think we are just going in circles over there and third opinions on the basic issues would help. Thanks Montanabw (talk) 04:34, 16 December 2008 (UTC)


 * Hi Montanabw. I don't think I have enough knowledge of the subject to comment on what needs to be added to/cut from the article. Dr pda (talk) 09:26, 18 December 2008 (UTC)

Prosesize.js
Is there any reason I should get this output when I run through?: The bolded parts are what throws me off... Sceptre (talk) 18:46, 17 December 2008 (UTC)
 * File size: 247 kB
 * Prose size (including all HTML code): 0 B
 * References (including all HTML code): 72 kB
 * Wiki text: 73 kB
 * Prose size (text only): 0 B (0 words) "readable prose size"
 * References (text only): 11 kB


 * There was something funny with that article—when I looked at it all the ref tags had become &lt;gef&gt;! It seems to be back to normal now though.
 * File size: 246 kB
 * Prose size (including all HTML code): 69 kB
 * References (including all HTML code): 82 kB
 * Wiki text: 73 kB
 * Prose size (text only): 37 kB (6052 words) "readable prose size"
 * References (text only): 12 kB

Dr pda (talk) 09:26, 18 December 2008 (UTC)
 * I've found the problem; it doesn't work with modern.css. Yet. (And thanks :) ) Sceptre (talk) 13:26, 19 December 2008 (UTC)
 * Yes, the &lt;div&gt; which contains the text is called bodyContent under the monobook skin, and mw_contentholder under modern. I've modified the script to reflect this, so hopefully it should now work for you. (You will need to bypass your cache to get the effect of the change). Dr pda (talk) 09:08, 20 December 2008 (UTC)
 * It doesn't; I'm still getting 0KB (which makes me have to switch over to monobook; not arduous, but a bit annoying). Sceptre (talk) 20:29, 4 January 2009 (UTC)
 * Yes, there was another line where I needed to change to mw_contentholder. I've modified the script again so it should handle all skins. I've tested it under modern and it works. Bypassing your cache on modern.js should pick up the new version. (Incidentally, I'm surprised you got any output at all; when I tried running the script under modern before making changes I got an error immediately.) Dr pda (talk) 21:54, 4 January 2009 (UTC)
 * It works now, thank you. You should do the same for prossizebytes.js too :) Sceptre (talk) 22:14, 4 January 2009 (UTC)
 * Thanks, I forgot about that one. Glad it's working for you now. Dr pda (talk) 01:58, 5 January 2009 (UTC)

Can I have some more help? It's not about prosesize this time; it's about generatestats. Namely, is there any way to analyse several templates at once? I'd like to get some stats on articles that transclude this set of templates, but I think I'd have to do them seperately. Sceptre (talk) 01:32, 6 January 2009 (UTC)
 * I've modified generatestats so it now accommodates this case; just add  to the end of the URL, then at the prompt enter the name of the template category. The script will then produce stats for all articles which include any of the templates. Dr pda (talk) 08:28, 7 January 2009 (UTC)
 * Now using a template (such as featured article) doesn't work (and sorry if I'm annoying you : Sceptre (talk) 11:01, 9 January 2009 (UTC)
 * Should be fixed now. Thought the other modifications were independent of this case, but the last change I made to get them working meant they weren't. Dr pda (talk) 11:47, 9 January 2009 (UTC)

Two-page document
I'm uncomfortable with a small part of your recent edit to Order of the Bath. Italian Admiral Ernesto Burzagli's Order of the Bath diploma has two pages; and I was uncertain about the best way to handle it. By no means am I convinced my side-by-side solution was best, but I dislike yours as well. In my view, the second page signed by the Grand Master is insufficient standing alone without the context provided by the first page. Similarly, in my view, the first page alone seems a bit awkward. In other words, as I see it, in the case of a two-page document, both pages need to be posted together or not at all. If you can help me re-evaluate my point-of-view, I'd appreciate it. In the meantime, I've removed that second page of the diploma which you posted alongside the list of Grand Masters.

A corollary question: What if the two pages were combined in a single image? The result would be unreadable, but it would appear that this treatment would be preferable for you and for the many other editors who are likely to see this problem in the same light you do? --Tenmei (talk) 14:39, 22 December 2008 (UTC)


 * My concern was two-fold: first, having two images cluttered that section, squeezing the text between them; second, I didn't really see a connection between the images the surrounding text. I tried to address both of these by removing one image, and attempting to draw a connection with the list of Great Masters by mentioning the signature of the Great Master of the Order and placing the remaining image in this section . However the document (I'm not sure diploma is the correct term here; normally these sorts of things are warrants or letters patent, but I'm not sure which, if either, is correct here) really deals with an honorary appointment to the Order, and as such would fit best beside the paragraph in the Membership section which deals with this. I still think it would be best to have only one image at this point. Therefore I propose the following:


 * the two pages of the document be combined into a single image (it's difficult to read even the single page as a thumbnail, so I don't think this affects readability)
 * the image of Prince Albert be moved to the Great Masters section, with an adjustment of the caption to make it clear the subject is Prince Albert, who was a Great Master of the Order.
 * the image of Sir Richard Johns be moved to the 20th century section
 * the two-page image of the document be inserted in the Membership section, with a caption dealing with the fact of the honorary appointment (rather than the signatures)
 * Dr pda (talk) 01:39, 23 December 2008 (UTC)
 * I understand and agree with your impeccable proposals. Please allow me to take on the task of figuring out how to merge the single-page images into one two-page image; and when completed, I would expect to post it in the "Members" section as you've suggested.  In the new year, perhaps we might re-visit the questionable value of this "diploma" (or whatever it's called).  Maybe the composite image is isn't really a helpful addition to Order of the Bath?


 * In part, my reason for posting this relatively insignificant image stems from a sense of surprise I felt a couple of years ago when I first learned that the Nobel Prize medals are accompanied by individually-designed "diplomas," -- see, e.g.,
 * Metathesis as "dance" -- illustrated in Chauvin diploma, 2005 Nobel Prize in Chemistry
 * Metathesis as "dance" -- illustrated in Grubbs diploma, 2005 Nobel Prize in Chemistry
 * Metathesis as "dance" -- illustrated in Schrock diploma, 2005 Nobel Prize in Chemistry
 * Googling "Nobel diploma" brings up, for example: the Neils Bohr Institute website, the National Library of Medicine website, the Woodrow Wilson House website, etc. See also Paul Krugman's Nobel diploma.  In this context, please consider your own reaction to digital images of only Nobel Prize offered for sale?


 * One or more of the descendants of Italian Admiral Ernesto Burzagli have posted a number of photos and documents from family archives; and amongst these were images of diplomas or certificates which accompanied appointments to the British Order of the Bath, the Dutch Order of Orange-Nassau, the Moroccan Order of Ouissam Alaouite, etc. These were the impetus for posting Burzagli's unique diploma images in the context of the Order of the Bath.  I would have thought others might be similarly unaware or perhaps similarly interested.


 * I wonder whether this oblique rationale seems valid? If not, your reaction may put my other similar edits into question? --Tenmei (talk) 20:19, 23 December 2008 (UTC)
 * May I suggest a multiple image template, such as double image? Maralia (talk) 01:08, 24 December 2008 (UTC)


 * Thanks Maralia, that's a nicer solution than having to physically join the two images. Tenmei, I think the image is of sufficient interest to be included in the article (I have in fact done so), now that the space issues have been resolved, and it has been more closely related to the text. Dr pda (talk) 02:29, 24 December 2008 (UTC)


 * Maralia and Dr pda -- I appreciate your help in parsing this minor quandry. In thanks, may I share two other paired images?  As you may already know, Vincent Van Gogh copied two of the One Hundred Famous Views of Edo by Andō Hiroshige (安藤広重), which were among his collection of Japanese ukiyoe prints:
 * Hiroshige's The Plum Garden in Kameido and Van Gogh's Flowering Plum Tree
 * Hiroshige's Great Bridge, Sudden Shower at Atake and Van Gogh's The Bridge in the Rain
 * I think I've learned from working through this small problem with you. I'm pleased to include double image among the tools available in future. --Tenmei (talk) 16:18, 24 December 2008 (UTC)

Referencing stats
Nn123645 has kindly offered to create some ref stats for me, but we're having trouble. Could you look at User talk:nn123645/2009/March and offer your opinion/advice. It may not be doable, I don't know. - Peregrine Fisher (talk) (contribs) 04:03, 24 December 2008 (UTC)

Prosesizebytes.js query
Hi, do you have any idea why "Page size" on Turnpike trusts in Greater Manchester returns a zero "readable prose size": BTW I use your tool extensively on the WP:DYK project and it saves me a lot of time. Thanks --Bruce1eetalk 14:37, 6 January 2009 (UTC)
 * File size: 121746 B
 * Prose size (including all HTML code): 105 B
 * References (including all HTML code): 22850 B
 * Wiki text: 28922 B
 * Prose size (text only): 0 B (3 words) "readable prose size"
 * References (text only): 2522 B
 * Images: 194607 B
 * The reason is that the whole article has been put inside a table, and my prose size script doesn't count text inside tables. I'm not sure why the article is like this, possibly it's something to do with the image layout; on my laptop it causes all the images and some of the text to lie off the right-hand side of the screen. Dr pda (talk) 08:28, 7 January 2009 (UTC)
 * You're right, it is inside a table! And for me the images are also off the screen to the right. I'm sorry, I should have checked for something like that first before bothering you. But thanks anyway. --Bruce1eetalk 09:24, 7 January 2009 (UTC)

Maine Coon peer review
I have been meaning to leave you a message thanking you for your wonderful peer review of the article. I appreciate the hard work you put into it and am currently waiting for the GA review to be completed before I begin on your suggestions (all of which are excellent and ones I would have never seen otherwise). I just wanted to stop by and thank you for your time! – Ms. Sarita  Confer  22:31, 8 January 2009 (UTC)

Using prosesize.js in edit window?
Why have you chosen to disable the script when in an edit window? Does it not work as intended? Because, I'd love to be able to use it in an edit window since it can be useful to check the prose size for a specific section, or even prose that's only Previewed. Isn't there a &lt;div&gt; that encloses the previewed content? Gary King ( talk ) 05:51, 11 January 2009 (UTC)
 * The reason would be this request :) However the issue which generated an error then has since been resolved. It turns out that only minor changes are needed to make the script work on previewed text, so I have implemented these. The script is still disabled for the case when there's just an edit window (and no previewed text) since it works by counting the text within &lt;p&gt;&lt;/p&gt; tags, not by parsing the wikitext. Dr pda (talk) 22:07, 11 January 2009 (UTC)
 * Thanks. I've got an idea. Make the "Page size" text appear even before preview, but don't link it; and, add a title= to it. So it would be something like  Page size  Gary King  ( talk ) 22:08, 12 January 2009 (UTC)
 * Thanks for the suggestion. Since the function I use to put the 'Page size' link in the toolbox always generates a link, I've instead made the script pop up a dialog box to say that previewing the text is necessary. Dr pda (talk) 03:10, 13 January 2009 (UTC)
 * Okay, although it's a bit more annoying. Sometimes I leave an edit window open and then briefly move to another task before revisiting it, and forget that I hadn't previewed yet, so I will get the dialog box. With it showing but unlinked, it's a very quick visual cue that it is disabled. I think UI-wise, it makes more sense. Gary King  ( talk ) 03:17, 13 January 2009 (UTC)
 * Not having the 'Page size' link at all in edit-without-preview mode would provide just as good a visual cue :) However then editors might not realise the script works in edit mode at all (which in fact it won't for existing users unless they see this discussion and bypass their cache to get the updated version), which was presumably the motivation for your suggestion. Thus I have now modified the script to change the colour of the link in edit mode to give a visual cue, and clicking the link in this case will give the explanation. Dr pda (talk) 11:29, 13 January 2009 (UTC)
 * Okay it's better now. If it was removed before previewing, then I might miss that, since my "toolbox" already has a ton of items in there. Gary King  ( talk ) 16:28, 13 January 2009 (UTC)

Help
Doc, I don't know anything about the automated tools, but a very troubing situation developed last night. Do you have any ideas about what might have happened here: User talk:Slp1? Slp1, truly a stellar editor, has left the building over this; it would be nice to at least let her know how it might have happened. Sandy Georgia (Talk) 00:53, 14 January 2009 (UTC)
 * Sorry Sandy, I don't use Twinkle or Huggle or AWB or any of those tools (perhaps that's why I only have 2500 edits after 3 years!), so I don't know what might have led to that edit. Dr pda (talk) 07:11, 14 January 2009 (UTC)
 * Thanks anyway :-) When you get a chance, can you look here:  Wikipedia talk:Unreviewed featured articles?  I recall something about that article, so I don't think it's a mistake.  Not sure how you generated that list, but I suspect something is missing in the history there.  (A name change?  A move?) Sandy Georgia  (Talk) 16:17, 14 January 2009 (UTC)
 * Never mind: found, solved. Sandy Georgia  (Talk) 16:21, 14 January 2009 (UTC)

PR backlog
I have committed myself to do a minimum of two, and if possible three, peer reviews each week. If there is a list maintenance task that needs doing, I'll help there, too. I'd better warn you, however, that I am a bit hamfisted in the use of tools, so I'll need some careful instruction as to what to do. But I will help in any way I can. Ruhrfisch deserves a long break. Brianboulton (talk) 09:46, 16 January 2009 (UTC)
 * OK, let me just double-check. By changing one date, from January 12 to January 13, the PR articles dated 12 January automatically came into the backlog, right? So all I have to do, at the end of today, is alter January 13 to January 14, and the 13 January items will be backlogged? A manual check is then made to delete any items that have received reviews. If this is the case, I have no problem with the process, but am I missing something? Brianboulton (talk) 11:24, 16 January 2009 (UTC)
 * All understood now. I will take this task on as from today, and see how it goes. I will of course continue to peer-review. Brianboulton (talk) 14:32, 16 January 2009 (UTC)
 * I, too, would be willing to help. I see Brian's note above, and I've read the "update daily" explanation embedded in the PR backlog. I more-or-less understand. Perhaps Brian and I could split the task between us on some sort of rotating schedule. Or if Brian would prefer not to divide the task, that's fine too. Finetooth (talk) 18:58, 16 January 2009 (UTC)
 * I've now gone to Brian's page and printed your note to him about the mechanics of the update, so no need to explain it all again. When I'm done writing this note, I'll post a note on Brian's page so you aren't having to pass notes back and forth as the middleman. Finetooth (talk) 19:26, 16 January 2009 (UTC)
 * Not a case of too many cooks I hope! I said, above, I'd do it "as of today", but then, re-reading, I see that you are doing it for a week or so. So perhaps you should continue until next Friday, 23 January, and I'll take over then. After a week, I'll pass it on to Finetooth, and then he and I can alternate for a bit. Does that sound reasonable? I'm sorry to have got somewhat muddled, but I'm clear now. Brianboulton (talk) 19:56, 16 January 2009 (UTC)

Thanks both for your offers of help. I said I'd do it for the next week so that Ruhrfisch wouldn't have to do it while he's taking a break from PR. I'm happy to do this, but if you want to take over straightaway that's fine with me too. Dr pda (talk) 03:43, 17 January 2009 (UTC)
 * "After you, my dear Alphonse", I think the phrase goes, although I don't remember who said it. Here's a proposal: Dr pda to continue through Friday, 23 January, then Brian doing a stint from Saturday, 24 January through Friday, 30 January, then I take the baton and run from Saturday through Friday, and hand off to Brian, etc. Finetooth (talk) 18:38, 17 January 2009 (UTC)

Thank you!
[Barnstar moved to user page]


 * Thank you! Glad you found the reviews helpful :) Dr pda (talk) 09:47, 18 January 2009 (UTC)

You...
are my hero. Thanks for fixing the ref. problem I was having. You have no idea how much frustration that was causing me...NO IDEA. --Yohmom (talk) 05:08, 26 January 2009 (UTC)


 * You're welcome! (The references to tearing eyes out and chopping legs off did kinda clue me in to the frustration :) Dr pda (talk) 06:20, 26 January 2009 (UTC)

On a roll
... you're just all over the place! Thank you so much for all you're doing ... on days like today (where it seems like my work never ends, but there are few to help), I really notice! Sandy Georgia (Talk) 01:47, 27 January 2009 (UTC)
 * You're welcome, I've had a bit more spare time in the last day or two. Dr pda (talk) 02:38, 27 January 2009 (UTC)

Okay...
I used to have lovely little disambig tools that would tell me when a link led there. Now... I don't. I thought I got one added to my monobook.. but... it's not working. HELP! Ealdgyth - Talk 02:03, 27 January 2009 (UTC)
 * Hi Ealdgyth, From a quick look at your monobook.js it seems you were/are using different disambig tools from me (I use User:Splarka/dabfinder.js). I can't immediately see why yours would have stopped working, however one reason the changes you made today might not yet be working is if you haven't bypassed your cache to force your browser to use your updated monobook.js. There is one other possibility why User:Anomie/linkclassifier.js might not be working—I notice at the top of that page it says you need to install some CSS classes, which you don't appear to have done. These are what tell the script which colours to highlight the links. The instructions at User:Anomie/linkclassifier.css say to add  to your monobook.js to install these classes. I hope one of these suggestions will help fix your problem. Dr pda (talk) 02:38, 27 January 2009 (UTC)
 * Well, that didn't work (gaze up). For the Splarka, should I add all that code to my monobook, or do I have a handy import thingie? Ealdgyth - Talk 03:03, 27 January 2009 (UTC)
 * In almost every case, to add a script to your monobook.js you use importScript (Have a look at my monobook.js for example). To add the Splarka dab script, add   to your monobook.js, save, bypass your cache then look for the Find disambiguations and Find redirects links in the toolbox in the left-hand sidebar. Dr pda (talk) 03:43, 27 January 2009 (UTC)
 * thank you! That worked. I need those silly scripts, but adding them is always something of a "will it work" experience. Ealdgyth - Talk 04:23, 27 January 2009 (UTC)

Query number of FAs
Query here.  Sandy Georgia  (Talk) 19:35, 31 January 2009 (UTC)
 * Seems to have been resolved there. Dr pda (talk) 23:17, 1 February 2009 (UTC)
 * Yes, thanks anyway! Sandy Georgia  (Talk) 23:34, 1 February 2009 (UTC)

FYI: Wikipedia_talk:WikiProject_Heraldry_and_vexillology/Assessment. Gimmetrow 05:06, 2 February 2009 (UTC)

Swedish heraldry
I would like your thoughts on the Municipal heraldry section. The sampling of municipal arms still bugs me. On the one hand, we don't want to make it too listy, and on the other hand, there are so many old city arms that are quite notable to Swedish heraldry, so it's hard for me to decide exactly what should be included. I've added a proposal for comment on the talk page, and I would value your input. Thanks for all your help with this article! Wilhelm_meis (talk) 04:36, 4 February 2009 (UTC)

quick question
where's that script that compares the edits of 2 editors to look for common interests etc.? Tks Ling.Nut (talk&mdash;WP:3IAR) 13:27, 4 February 2009 (UTC)

Tools Coder Barnstar
[moved to userpage]


 * Thanks! (I like the text in the background of the image :) ) Dr pda (talk) 21:47, 9 February 2009 (UTC)

Feb 15 Dispatch
Hey, Doc, I put FCDW/2008Activity up for Feb 15. I don't expect to have any time to review it between now and then, but I'm confident that you've left it in great shape. Best, Sandy Georgia (Talk) 14:07, 10 February 2009 (UTC)

Thanks
Thanks so much for the PR! Everything you said was spot on. :) I'm starting to attack the list now. --Yohmom (talk) 02:51, 13 February 2009 (UTC)
 * You're welcome! Dr pda (talk) 03:40, 13 February 2009 (UTC)

Proposed Task Force
Hi, I have suggested the creation of a task force, within WP:HV to cover Heraldry by country articles. You may be interested in commenting on this proposal and/or joining. Thanks, - - Jarry1250 (t, c) 22:01, 18 February 2009 (UTC)


 * Replied there. Dr pda (talk) 21:37, 23 February 2009 (UTC)

persondata and microformats
You suggested last year in the closing statement of this post "Perhaps the best way of combining persondata with hCard (if you want to go there at all) would be, as you originally suggested, adding extra class tags in the persondata template itself."

In that discussion, you displayed detailed understanding of how Persondata is used, and the tools that work on this data. Do the pearl scripts that operate on these look at the resultant page html, or the wikitext? Thanks -J JMesserly (talk) 16:41, 22 February 2009 (UTC)


 * Are you referring to the perl scripts described in the section Persondata? These work on the wikitext as contained in the periodic dumps of the wikipedia database. However I don't know if anyone is actually using them at the moment. Dr pda (talk) 21:37, 23 February 2009 (UTC)
 * My apologies for the delay. I will track down who is making use of this, thanks for the pointer to the scripts.  -J JMesserly (talk) 03:53, 27 February 2009 (UTC)

User:Dr pda/prosesize.js
Technically, {{bcode| else if(wgAction == 'view' || wgAction == 'submit'){ }} should also have since that also shows the page's contents. Gary King ( talk ) 19:21, 26 February 2009 (UTC)


 * Technically, you're correct! I seldom need to purge a page, except occasionally to update a transcluded template, so the use case didn't occur to me. I've updated both versions of the script. Dr pda (talk) 00:00, 27 February 2009 (UTC)

DYK --> FA
Doc, per Wikipedia talk:Featured content dispatch_workshop, do you have an easy way to run a script to see what/how many DYKs have become FAs? Sandy Georgia (Talk) 01:32, 1 March 2009 (UTC)
 * Hi Sandy, the suggestion for a dispatch is referring to the fact that last year ArticleHistory was modified so that it would populate the categories
 * Category:Wikipedia Did you know articles that are featured articles
 * Category:Wikipedia Did you know articles that are featured lists
 * Category:Wikipedia Did you know articles that are good articles
 * if AH had a non-empty  parameter and the appropriate   (FA/FL/GA). (I happened to be aware of this at the time.) So the necessary information is contained in these categories, though some analysis would be needed to write a dispatch. On a related note Category:Featured articles that have appeared on the main page was set up to be filled in a similar way through AH just last week, though I don't think the category has finished being populated yet. Dr pda (talk) 04:03, 1 March 2009 (UTC)

Thank you!
Your input on Oiceoptoma noveboracense was very helpful. I will begin making the corrections you suggested to make the page the best it can be! Shealamartin (talk) 22:39, 26 March 2009 (UTC)

Thanks!
Thanks very much for figuring out and fixing the problem at WP:PR, Ruhrfisch &gt;&lt;&gt; &deg; &deg; 01:22, 6 April 2009 (UTC)

Math HTML
Hey, doc, maybe you know where to steer us on this? Sandy Georgia  (Talk) 16:03, 7 April 2009 (UTC)


 * Hi Sandy! Let me make sure I understand your question. The formulae in that section to me display as an italicised serif font, i.e. something like
 * Bodyfat% = (1.2 * BMI) + (0.23 * age) − 5.4 − (10.8 * gender)
 * as opposed to the TeX rendering (let me see if I can force it)
 * $$Body fat \% = (1.2*BMI)+(0.23*age)-5.4-(10.8*gender) \,$$
 * I think Tony's first comment (about the ugliness) related to the second form, whereas your comment (about the difficulty of reading) seems to concern the first form.
 * The MediaWiki software has two ways of displaying stuff within  tags—as HTML (if the equation is simple enough), or as TeX converted to a .png image on-the-fly. The version which is used depends on your user preferences (just like dates ... :). The options (on the Math tab under my preferences) are
 * Always render PNG
 * HTML if very simple or else PNG
 * HTML if possible or else PNG
 * Leave it as TeX (for text browsers)
 * Recommended for modern browsers
 * MathML if possible (experimental)
 * I agree that the HTML version is a little bit small; this is probably because serif fonts are in general smaller than sans-serif fonts (e.g. 12pt Times New Roman is about the same size as 10pt Arial). There are a couple possibilities for making it easier to read—for an individual editor. If you don't mind the ugliness, you could set your preferences to "always render PNG". Or you could edit your monobook.css to override the formatting of the HTML version. This formatting is set in http://en.wikipedia.org/skins-1.5/common/shared.css, and consists of, which simply changes the font for equations to the default serif font. If you were to add   to your monobook.css it would increase the font-size by 5%, producing something like
 * Bodyfat% = (1.2 * BMI) + (0.23 * age) − 5.4 − (10.8 * gender)
 * There may be a reason why the default setting doesn't change the size (maybe to make superscripts/subscripts fit in better with surrounding text?). I'm not sure where one would go to try to get this changed on a Wikipedia-wide basis, maybe raise it at WP:VPT? You can find more information about maths syntax and rendering at WP:MATH. Dr pda (talk) 00:35, 8 April 2009 (UTC)

Ugh. OK, thanks for the detailed answer, as always. Tony hates the larger one, but I equally dislike the smaller one. I suspect it's not worth worrying about, because MoS tangles with the Math people are never fun. Thanks for letting me know what it's about; I'll give Tony a ping. Sandy Georgia (Talk) 14:56, 8 April 2009 (UTC)

persondata.js
I have just started using your script. I am planning to add persondata to all the tennis player articles, as most of them do not have it. I noticed that the script does not work with Infobox Tennis player. Could you check why this is happening and if possible remedy it? Thanks. Leave  Sleaves  16:29, 7 April 2009 (UTC)
 * Hi LeaveSleaves, There are lots of different biographical infoboxes, and unfortunately they all seem to use different parameter names for birth date, place etc (e.g. date of birth, date_of_birth, birthdate, birth_date, born ...) When I wrote my script I made sure it knew the parameter names for the 20 most frequently used infoboxes; Infobox Tennis player uses yet another variation on the parameter names, which is why it didn't work. I've updated the script so it should now work for you, though you will need to bypass your cache to pick up the modified version. Dr pda (talk) 00:35, 8 April 2009 (UTC)
 * It is working perfectly now. Thanks for your prompt response. Leave  Sleaves  01:28, 8 April 2009 (UTC)
 * Great. Be aware that the script will put the contents of the  field from Infobox Tennis player into the ALTERNATIVE NAMES field of persondata, which is not necessary if this is identical to the NAME field. This behaviour arises because   is used differently in some other sporting Infobox. Dr pda (talk) 02:37, 8 April 2009 (UTC)

Referencing stats
On my talkpage, could you provide a link or (better yet) summarize you findings from June by class? I can't remember where we discussed it previously. Thanks so much. - Peregrine Fisher (talk) (contribs) 05:07, 10 April 2009 (UTC)

error with prosesize.js on secure
Hi, I experienced an error with User:Dr pda/prosesize.js when logged into the secure website https://secure.wikimedia.org/wikipedia/en/wiki/ which I use. It isnt an urgent problem for me, as I dont use the script very often; just thought you should know. John Vandenberg (chat) 03:39, 24 April 2009 (UTC)
 * It should be fixed now (You'll need to bypass your cache for your monobook.js (or equivalent) to pick up the changes). The URLs on the secure server are different, so when the script tried to query the API for the size of the wiki text it was using the wrong URL. Dr pda (talk) 04:24, 24 April 2009 (UTC)


 * I'm afraid it's not working for me anymore, even after I've cleared the cache. This is rather inconvenient since, in case you didn't know, prosesize.js is the standard tool we use at DYK for counting the length of contributions.
 * I'll try again tomorrow, if it still isn't working I'll let you know. Gatoclass (talk) 21:16, 24 April 2009 (UTC)
 * I tested the script after making the changes and it worked fine for me. I've just cleared the cache on this computer as well and it still works. It would help if you indicated exactly how it is not working for you (are you using the secure server as well?). In Firefox, Tools->Error console may give an error message. Dr pda (talk) 21:33, 24 April 2009 (UTC)
 * I don't know what the secure server is, I am just talking about using the program in en.wikipedia mainspace.
 * The problem is that when I click on the "Page size" link in the toolbox, instead of giving the prosesize output, I just get a blank space with four bullet points. I've always had this problem with prosesize BTW, but a simple click on the page tab would fix it. But now, reloading the page, even after a cache clear, doesn't fix the problem. Gatoclass (talk) 20:12, 25 April 2009 (UTC)
 * It sounds like the problem is at your end, rather than in the script itself. I'm not sure why my recent change should have affected the behaviour you have been seeing. I tried replacing my monobook.js with yours but prosesize still worked fine for me. Maybe it's something to do with your Wikipedia settings (a gadget turned on in preferences?) or your browser settings (javascript?). Do you have the same problem with Shubinator's DYKcheck script? That incorporates a prose size measurement. Dr pda (talk) 05:31, 26 April 2009 (UTC)
 * No, Shubinator's program works fine. Gatoclass (talk) 05:56, 27 April 2009 (UTC)
 * Prosesize now works for me on secure server. I'll tweak DYKcheck with a similar fix. Thanks! Shubinator (talk) 21:46, 24 April 2009 (UTC)

By the way, I figured out that prosesize can be used by anons. Enter  into the URL, hit enter, and the Prose size link shows up in the toolbox. You'd have to do it on each page, but the snippet above can be bookmarked. Shubinator (talk) 22:25, 24 April 2009 (UTC)
 * Thanks, I think I saw you mention this elsewhere. Dr pda (talk) 05:31, 26 April 2009 (UTC)

Thanks Dr pda! And .. that is a nifty trick Shubinator - it should be documented somewhere. Maybe we should pull "Bookmarklets" out of retirement? John Vandenberg (chat) 10:57, 25 April 2009 (UTC)
 * The trick's pretty handy because you can do it for just about any script. For example:
 * For DYKcheck,
 * For Splarka's DABfinder,
 * For ais523's edit counter,
 * The function after the importScript is only needed so you don't go to a white page (on Firefox; haven't seen on IE). It's useful if the function is the root function, so if you hit "enter" twice the script runs. However, you can use all of the above with a dummy function. You can even use Twinkle with a dummy function! (Haven't tested this one out as thoroughly, but think of the possibilities)
 * For Twinkle,
 * I mainly use the trick to test out scripts or if I rarely use a script, but I'm sure this could help out a lot of people. Shubinator (talk) 00:41, 29 April 2009 (UTC)

Re: Forbes
Yes, I saw it. Thanks a lot, it looks like one of the most through review's I've seen. Hopefully I can get to fixing everything within the next few days. After that the article should be well on its way to its FAC. Thanks again!  black ngold29  05:47, 26 April 2009 (UTC)

Fish
I think I fixed it all - thanks for the heads up, Ruhrfisch <sub style="color:green;">&gt;&lt;&gt; &deg; &deg; 02:56, 27 April 2009 (UTC)
 * The same user opened two more PRs, which were also before and after a page move (one was a duplicate - long nosed loach). I think I have also cleaned that up too. Sigh. Ruhrfisch <sub style="color:green;">&gt;&lt;&gt; &deg; &deg; 04:11, 27 April 2009 (UTC)
 * Thanks. I didn't notice that the other reviews were "before and after" ones as well. Long-nosed loach is currently showing as a redlink at WP:PR, but I guess this will go away next time VeblenBot runs. Unfortunately it looks like the moved version of the article (Acantopsis octoactinotos) is simply a copy and paste job from the external link at the bottom of the article. Dr pda (talk) 04:23, 27 April 2009 (UTC)
 * Thanks - I pared the artilce back to one sentence and the taxobox, plus the cleaned up reference. I also left copyvio notices on the article talk page and the peer review (which I archived as there was nothing left to review), and left a note on the talk page of the "author" ("cut and paster"?). Ruhrfisch <sub style="color:green;">&gt;&lt;&gt; &deg; &deg; 13:19, 27 April 2009 (UTC)
 * How not to "fix" copyvio issues ;-) Ruhrfisch <sub style="color:green;">&gt;&lt;&gt; &deg; &deg; 23:39, 27 April 2009 (UTC)

Prosesizebytes bug
First count Ta Som; the current count is 2246. Then without saving the change, go to edit mode, remove the line with the geographic template "coord", and click "edit preview"; then try prosesizebytes. The count will drop 82 bytes (currently to 2164), after removing something that isn't colored yellow. The problem isn't limited to preview mode; see WP:TDYK if it isn't deleted yet. Art LaPella (talk) 05:13, 5 May 2009 (UTC)


 * The reason the coord template with  affects the prose size count is because when it is converted into HTML the text gets put inside   tags, which are what the script counts. The reason it does not have a yellow background is that the text is inside a   which already has the background colour set, and thus does not inherit the yellow colour from the parent paragraph. I've altered both versions of the script to exclude the span that contains the coordinates from the prose size count. Dr pda (talk) 11:34, 6 May 2009 (UTC)


 * It works! Thank you. Art LaPella (talk) 21:37, 6 May 2009 (UTC)

prosesize.js - TonyTheTiger
I apologize for corrupting an archive. I probably commented on the page that had our original conversation by accident. Manual use of the code will be sufficient. Thank you very much for giving me the proper command.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTM) 06:36, 12 May 2009 (UTC)
 * Firefox keeps changing to " http://, I get the response "The webpage cannot be found."

Could you please explain in simple terms the computer-illiterate I am can understand how to download this program? Thank you for your help. LiteraryMaven (talk • contrib) 18:31, 20 June 2009 (UTC)


 * It worked . . . many thanks! LiteraryMaven (talk • contrib) 13:19, 22 June 2009 (UTC)


 * P.S. Well, it worked as far as adding "Page size" to my toolbox. Unfortunately, when I click on the link, nothing happens! Oh, well . . . LiteraryMaven (talk • contrib) 15:09, 22 June 2009 (UTC)


 * Yes, I bypassed my cache on my monobook.js page. When I tried your suggestion and copied and pasted  javascript:importScript('User:Dr pda/prosesizebytes.js'); getDocumentSize;   into the address bar of my browser I got the response "Internet Explorer cannot download. Unspecified error." Unfortunately I'm not computer savvy enough to know what that means! LiteraryMaven (talk • contrib) 17:27, 22 June 2009 (UTC)

Does this help you? When I click on "Page size," at the top of the article it says "Document statistics: (See here for details.)" LiteraryMaven (talk • contrib) 19:21, 22 June 2009 (UTC)

Thank you
... so, so much ! You're still the best! Sandy Georgia (Talk) 12:24, 1 July 2009 (UTC)
 * Thank you, thank you, thank you for once again doing the tedious bookkeeping work for me! Sandy Georgia  (Talk) 18:10, 1 August 2009 (UTC)
 * Not a problem, I was online and I noticed those pages hadn't popped up on my watchlist for the first of the month :) Dr pda (talk) 10:08, 2 August 2009 (UTC)

Doc, I can't figure out what's killing the TOC at Featured article candidates/Archived nominations/July 2009; I suspect it's because someone added a no-TOC to an individual FAC, but how can I determine that without checking every single FAC in that archive? Sandy Georgia (Talk) 18:14, 1 August 2009 (UTC)
 * Never mind, someone found it, a notoc in the Ferraro FAC. Sandy Georgia  (Talk) 19:10, 1 August 2009 (UTC)

Foreign heraldry blazons
Oyez, oyez! Greetings and blessings upon all to whom these presents shall concern: the honor of your presence is humbly requested by WikiProject Heraldry and Vexillology, that your thoughts may enrich and enlighten our discussion of non-English blazonry at Wikipedia talk:WikiProject Heraldry and vexillology/Style guide for foreign blazons. The purpose of this discussion is to craft a guide to presenting blazons translated from foreign languages in a style that is clear and consistent. The fruits of this discussion will amend Blazon as the portion of this guide dealing specifically with foreign-language blazons. Active discussion is expected to resume by the end of this week. Thank you and good day! Wilhelm_meis (talk) 12:18, 7 July 2009 (UTC)

Epikleros
Well, I finally got off my behind... and added some information. It's still pretty rough, but I did incorporate infromation from Just's Women in Athenian Law and Life, Patterson's Family in Greek History, Pomeroy's Families in Classical and Hellenistic Greece, and Schaps' Economic Rights of Women in Ancient Greece (which was the main expansion). Probably needs a bit of reorganization now to make sure information is in logical places. Are we missing any major sources? Ealdgyth - Talk 18:57, 10 July 2009 (UTC)


 * Hi Ealdgyth,
 * Yes I saw you had been busy yesterday, and figured after almost a year you'd got tired of waiting for me :) It's certainly an improvement on what was there before, though there are still some statements which seem incorrect to me, some aspects which seem over/underemphasised, and others which need expansion. If you're not in a hurry to get this to FAC, another couple of months (he says optimistically) should allow me to finish checking sources, assimilate the notes I've taken and make the necessary additions to the article.


 * Regarding missing sources, the top 14 most frequently cited works out of the 160 or so which I've consulted are the following. (The first number is the number of times cited, bold indicates the works already used in the article)


 * 56 Harrison 1968 The Law of Athens
 * 42 Schaps 1979 Economic Rights of Women in Ancient Greece
 * 29 Lacey 1968 The Family in Classical Greece
 * 19 Gernet 1921 (French) "Sur L'Epiclerat", Revue des Etudes Grecques XXXIV
 * 14 MacDowell 1978 The Law in Classical Athens
 * 13 MacDowell 1982 "Love versus the Law: An Essay on Menander's 'Aspis'", Greece & Rome 29
 * 11 Lipsius 1915 (German)
 * 10 Wyse 1904 The Speeches of Isaeus
 * 10 Scafuro 1997 The Forensic Stage
 * 10 Just 1989 Women in Athenian Law and Life
 * 10 Beauchet 1897 (French)
 * 9 Willets 1967 The Law Code of Gortyn
 * 9 Patterson 1998 The Family in Greek History
 * 8 Brown 1983 "Menander's Dramatic Technique and the Law of Athens" Classical Quarterly 33
 * So there's a few which have not yet been used, though some of the older works may not be as relevant. There's also a couple of recent books in French by Karabelias, one on the epiklerate at Athens and one on equivalent Greek institutions outside Athens, which I need to try and get hold of via interloan, and see if I can extract anything useful. But as I say, I'm happy to keep plugging away at this for the next few months, if you have a bishop or horsey article you'd rather play with :) Dr pda (talk) 07:02, 11 July 2009 (UTC)
 * I have the MacDowell on Menander article, and the Brown on Menander article, and there isn't much in there that is relevant to just epikleroi, instead of the specific play. I'm not adverse to a short discussion of the play, but I figure the indepth breakdown of it should go on the play article, not in the epikleros article. Otherwise, I'm not in any great hurry to get this to FAC, but I do think it'd do pretty well with some more work. I was kinda hoping I'd spur you on a bit. I've also had Malleus look it over for prose. Take your time, I've got a number of other articles in the works (as usual) but the poor women were moving up on the list... Ealdgyth - Talk 12:52, 11 July 2009 (UTC)
 * Consider me spurred :) I have actually been working on it off and on, though I generally only have a couple of nights per week at most that I can work on it, and the fact I prefer to do my article research/building offline would obviously make this progress invisible to you. Rest assured that I am still working on it; in fact a couple of weeks ago when I was on holiday in Europe I popped in to Bodleian library to check a couple of sources there :) I agree that it will make a nice FAC when it's done. Right, off to the library to get some more sources :) Dr pda (talk) 01:35, 12 July 2009 (UTC)

If you have the time and interest
I am trying to figure out which number is best to use for the PR stat in WP:FAS. I have posted about this here. I have also asked USer:Geometry guy (who helped set up the current bot system for opening and archiving PRs with User:CBM to weigh in). Thanks for all you do, Ruhrfisch <sub style="color:green;">&gt;&lt;&gt; &deg; &deg; 11:42, 3 August 2009 (UTC)

Order of the Garter
Good work tidying up the Order of the Garter article. I understand your reasons for removing my photograph of the mantle and hat, as the article does have a few other photographs. But I wonder if there might be a way to fit it back into the article somewhere, as it does give quite a clear view of the details and it seems a shame not to use it. On the other hand, I'll understand if you think it might make the article a bit too cluttered. -- Nicholas Jackson (talk) 09:10, 11 August 2009 (UTC)
 * I don't have any objection in principle to your photograph, it was just that there were lots of photos in that particular section, and that was the easiest to remove since the mantle and hat appeared in others. However having another look, the picture of Queen Victoria didn't really add anything in the history section, so I replaced that with the detail of the badge. The picture of the Japanese emperor I moved to beside the text on Stranger Knights, and dropped the photo of the supernumerary knights. This then made some extra room in the Vestments section which is the logical place for your photo, which I then re-added. See what you think of the current layout. Dr pda (talk) 11:27, 11 August 2009 (UTC)

Superscripts counted as prose
Prosesize counts superscripts (like [citation needed]) as prose. I've changed DYKcheck so they aren't counted, and a similar fix should work for prosesize. You could just change the line

to

Shubinator (talk) 18:43, 29 August 2009 (UTC)


 * Thanks for pointing this out. I thought I had excluded them, but I see I have only done that in the version of the script I use on the database dumps, and for the wordcount in this version. I'll fix it shortly. Dr pda (talk) 23:49, 1 September 2009 (UTC)

Clickable images
Hey, doc, do you have any suggestions on User talk:Fvasconcellos? Thank you SO much, again, for keeping up with month-end FA archiving issues ! Best, Sandy Georgia (Talk) 18:45, 1 September 2009 (UTC)
 * Answered there. Dabomb beat me to half of the stuff this time round :) Dr pda (talk) 23:49, 1 September 2009 (UTC)
 * Thanks, doc (ever amazing)! Sandy Georgia  (Talk) 02:45, 2 September 2009 (UTC)

Problem with prosesize
Could you help me sort out the problem I've described at Wikipedia talk:Did you know/Additional rules? It doesn't appear to be an error in prosesize as it's working correctly for other people, but I don't see what's causing the problem for me. Thanks, cmadler (talk) 20:29, 11 September 2009 (UTC)
 * Replied there. Dr pda (talk) 04:53, 12 September 2009 (UTC)