User talk:Dispenser/Archive 2

Links to Google Book Search Breaking?

Whenever I put in a Google Book Search link it returns an error:

GBS page: http://books.google.co.uk/books?id=Ix2eYaG82hYC&dq=francis+bacon+sound+houses&pg=PA407&ci=88,252,893,766&source=bookclip Image: http://books.google.co.uk/books?id=Ix2eYaG82hYC&pg=PA407&img=1&zoom=3&hl=en&sig=BqE1OnF8zttsp03KpfntSa2oSwY&ci=88,252,893,766&edge=1 —Preceding unsigned comment added by 65.57.245.11 (talk) 03:22, 17 June 2008 (UTC)

Re: Frivolous bot edits

Could something be done about some of the pointless edits that some bot do? I just can across User:VeblenBot which updates User:VeblenBot/PERtable and a good chunk of its edits are to simply update the timestamp. Isn't the "does not consume unnecessary resources" suppose to cover this? This isn't the only one that operates in this fashion either. —Dispenser (talk) 01:31, 3 January 2008 (UTC)

I'm not sure anything needs to be done about the policy itself since, as you have pointed out, it already covers this. The approval request for the process you mention can be found here, and it seems the bot was indeed given approval to edit once per hour whether it needed to or not. I assume that the bot is coded the way it is because doing so is usually simpler, and thus less error-prone. I also assume that Mets501, the approval group member who approved the request, decided that one edit per hour was sufficiently insignificant that it didn't really matter – correctly so, in my opinion, since this wiki often sees 20,000 edits in an hour.

I agree, though, with both your statement and the policy's statement that unnecessary edits should be avoided wherever possible. I suggest that you contact CBM, the operator of VeblenBot, and suggest that he modify it to avoid editing when no actual changes to the table need to be made. Such a change would not actually alter the function of the bot in any significant way so shouldn't need approval. You may wish to do the same thing in any other instances you may find – Gurch 13:03, 3 January 2008 (UTC)

Stable geohack

Hi, you're quite welcome! I have enough work pouring out of my ears as it is ;-) You'll have to get access to the stable server and the geohack tool by asking River Tarnell or Dab.. Somewhere, there was a list where you could enter your name as maintainer, but I can't find it anymore... Thanks, Magnus Manske (talk) 20:35, 7 January 2008 (UTC)

SMOG Copyright Violation

I contacted the copyright holder in regards to the SMOG article, and it appears that he is the original author and frequent maintainer of the article. He has since added a GFDL statement to the bottom of his website.

The only other issue I can see is that he apparently invented this system of measurement, being a doctor in Applied Linguistics. His works are original works and may not be suitable for Wikipedia as he is the sole authority on the matter.

I do like his article though, and would like to see it expanded upon. I would like to forward our email correspondence to you if you're interested. ~ Agvulpine (talk) 11:32, 9 January 2008 (UTC)

named versus unnamed parameters in templates

Halló Dispenser! Are you using IRC? Or Skype? My IRC nickname is gangleri the Skype name is irelgnag . Maybe we can chat about the topic. Best regards
‫·‏לערי ריינהארט‏·‏T‏·‏m‏:‏Th‏·‏T‏·‏email me‏·‏‬ 05:49, 10 January 2008 (UTC)

Weblink checker

What a wonderful thing you created! (BTW, "automaticly checked" needs to be corrected...) I thank you again - it really helps. --andreasegde (talk) 17:55, 14 January 2008 (UTC)

Ditto, I've used it on a few article I've created. Very helpful! :) Mahanga^Talk 03:35, 6 February 2008 (UTC)

AutoWikiBrowser Userbox image

I think it looks good. -Gwguffey (talk) 16:36, 15 January 2008 (UTC)

What the heck?

What the heck is a good machine parseable datasets? I reaaly don't know about how to program bots and the like. Could you please reply on my talk page? Editorofthewiki (talk) 17:07, 19 January 2008 (UTC)

I really just want to know whether you can create a bot from the Insee info and the info from the French Wikipedia. Is that possible? Editorofthewiki (talk) 18:54, 21 January 2008 (UTC)

Re: Single quote in #ifeq

I checked out m:Talk:Migration_to_the_new_preprocessor#Single_quote_in_.23ifeq, and couldn't replicate. --Splarka (rant) 04:44, 21 January 2008 (UTC)

Spelling

Could you fix the spelling of "foundation" at [1]. A bit more documentation would also help. Thanks and happy editing, Geometry guy 23:16, 25 January 2008 (UTC)

AWB Feature Page

Thanks for helping with the organisation - [2]

=)

—Reedy Boy 11:09, 28 January 2008 (UTC)

Just a quick thought. It can probably be categorised better than it is now... Maybe worth moving some more about to sub sections! —Reedy Boy 11:11, 28 January 2008 (UTC)

Webchecklink bug?

Is it a bug from webchecklink here, that makes it show all pages from www.experts-exchange.com as "Service Unavailable", while they are accessible from a browser? Or is it a problem with experts-exchange itself? huji—TALK 09:59, 1 February 2008 (UTC)

Link check problem with Bath, Somerset

Hi, Thanks for the really useful link checker, however Bath, Somerset is at Wikipedia:Featured article candidates/Bath, Somerset and I've been asked to check the links. It shows 2 x 404 & 2x 400 errors but these all work when you click on them & the URL it takes you to is the same as in the refs. The last one is to a journal that requires a subscription & not a problem - but any advice on the others would be great.— Rod ^talk 18:31, 5 February 2008 (UTC)

nice work on your link checker

Hi, I saw your "shameless self-plug" and gave it a try. I knew about those redirects, but had never bothered to update them. (Actually, I tested it first on Andrew Sullivan, which is a real mess, but decided not to make any changes for lack of knowledge or interest.) I was so impressed that I added a more detailed description here. One quirk I noticed is that the "Show preview" and "Show changes" buttons seem to work the first time only. --Jtir (talk) 22:43, 6 February 2008 (UTC)

reflinks.py

Thanks for adding the interactive version of DumZiBoT. I have already put it to good use twice.[3] [4] (I changed the edit summary from the default.) The expanded description could probably go at the top, since several people have asked for DumZiBoT to be run on particular articles. --Jtir (talk) 20:45, 12 February 2008 (UTC)

It fails with the following message on articles with foreign characters in their title; e.g., Kemal Kerinçsiz. --Adoniscik^{(t, c)} 18:06, 20 April 2008 (UTC)

UnicodeDecodeError: 'ascii' codec can't decode byte 0x.. in position ..: ordinal not in range(128)

linkchecker

Hi !

A little bug in your tool :

In Humanitarian response to the 2004 Indian Ocean earthquake, there is the source text

[[Sweden]] || [[Swedish krona|SEK]] 500M (USD 72.2M)<ref>http://www.regeringen.se/sb/d/4823/a/36245</ref> || || [[Swedish krona|SEK]] 1100M (USD 159M)<ref>[http://www.frii.se/index3.shtml Frivilligorganisationernas Insamlingsråd - Aktuellt<!-- Bot generated title -->]</ref> || || 177.2 || 0.5,

and it is converted into

[[Sweden]] || [[Swedish krona|SEK]] 500M (USD 72.2M)<ref>{{cite web |url=http://www.regeringen.se/sb/d/4823/a/36245 |title=]</ref> || || [[Swedish krona|SEK |archiveurl=http://web.archive.org/web/20050101025709/http://www.regeringen.se/sb/d/4823/a/36245 |archivedate=2005-01-01}}] 1100M (USD 159M)<ref>[http://www.frii.se/index3.shtml Frivilligorganisationernas Insamlingsråd - Aktuellt<!-- Bot generated title -->]</ref> || || 177.2 || 0.5

:)

NicDumZ ~ 19:47, 15 February 2008 (UTC)

Speedy deletion of Template:UK patent/doc

A tag has been placed on Template:UK patent/doc requesting that it be speedily deleted from Wikipedia. This has been done under section T3 of the criteria for speedy deletion, because it is a deprecated or orphaned template. After seven days, if it is still unused and the speedy deletion tag has not been removed, the template will be deleted.

If the template is intended to be substituted, please feel free to remove the speedy deletion tag and please consider putting a note on the template's page indicating that it is substituted so as to avoid any future mistakes (<noinclude>{{transclusionless}}</noinclude>).

Thanks. --MZMcBride (talk) 20:50, 19 February 2008 (UTC)

Readability tool

Sometimes the tool comes back with statistics fairly quickly (For example; Introduction to evolution, Bees and toxic chemicals and Dog), but othertimes seems to be slow, so slow that it might be broken (Evolution for example). What is going on? Are those articles just too complicated? Is something else wrong?--Filll (talk) 01:23, 20 February 2008 (UTC)

Thanks for implementing a readability analyser. I'm sure Wikipedia needs one of its own, because Wikipedia's conventions require so many elements that should not enter in to the calculations, e.g. UI elements, references list, "See also", links to editions in other languages, various templates that appear in certain article categories.
Even so, I was surprised by the difference in the scores given for Evolution of mammals by your analyser and Juicy Studio's Readability Test, which I've been using and recommending for a few years:

	Your analyser	Juicy Studio
Words	8,401	10,033
Sentences	300	1,078
Words per sentence	28.00	9.31
Polysyllabic words	1,327	2,306
Syllables per word	1.75	1.75
Words per sentence	28.00	10,033
Flesch Reading Ease	30.4	49.23
Flesch-Kincaid grade level	16.42	8.71
Gunning Fog	18.29	12.92

The differences may be due to the factors I mentioned above, but they need to be explained - I expect editors who are sufficiently interested in readability to use your analyser will know and use others.
More generally, your analyser's output needs some explanation, for example:

• What does "Readable proses:" represent? It's wikilinked below the big diagram, but should also be wikilinked at the top of the page.
• What's the big diagram all about?
• Just below the big diagram there's a paragraph that includes the sentence "No direct action should result from the information provided here." What does this mean? If it means "This article is within the acceptable range" it should appear at the top of the output with a suitable caption, e.g. "Overall evaluation". If it's a reassurance that your analyser will not change the article being analysed, it should appear on the input form.
• The same paragraph includes "..highlighed words indicate puncuation." What does this mean? I couldn't make a guess because there were no highlighted words in the output for Evolution of mammals.
• I think I see the purpose of the 2 outer text columns, headed "Wikitext" and "Proses (Marked up)"; but what's the purpose of the one headed "Text"?

You may also want to make the algorithm ignore:

• Directives like the T.O.C. one at the top of Evolution of mammals (I don't want to reproduce it here in case it confuses Wikimedia).
• ASCII art. I used ASCII art for family trees in Evolution of mammals because I thought the clade template had significant drawbacks for both readers and editors.
• That may mean your analyser should ignore all templates except those on a whitelist, e.g. for tables.

Despite all these concerns, I'm doing to put a link to your analyser on my User page - it has the potential to be extremely useful.
PS If you like I could also do comparisions of results for less scientific subjects, e.g. a couple of articles on computer games, which might contain fewer sesquipedalians. Philcha (talk) 09:57, 16 April 2008 (UTC)

Collapsible templates

Hi there. I'm an admin over on Memory Beta, the non-canon Star Trek wiki, and we are looking to include the collapsible templates into our wiki. I have tried to do this before by copying over the code, but it didn't work. I wonder if you could take the time to explain step-by-step how to set this up in advance. Thanks for your time (-: --82.37.50.162 (talk) 10:49, 13 March 2008 (UTC)

Navbox rewrite (which adds CSS)

Hi Dispenser, I recently rewrote {{Navbox}} to fix a number of known bugs and to add a whole bunch of new features, including CSS support. Please check out my new code (User:CapitalR/Navbox), the test and description page (User:CapitalR/NavboxTest) and the Navbox talk page to add comments (Template talk:Navbox#Complete Navbox rewrite). Seeing as you've done some work on this in the past and have been pushing for CSS to be added, I'd appreciate your comments, corrections, changes, and suggestions. Feel free to directly edit the code and test page in my user pages. Thanks, --CapitalR (talk) 19:17, 17 March 2008 (UTC)

external link checker tool

Hi dispenser,

http://tools.wikimedia.de/~dispenser/cgi-bin/linkchecker.py only works for Wikipedias, as far as I can see, is that correct? —Pill (talk) 13:28, 5 April 2008 (UTC)

Could you tell me...

Could you tell me why this isn't working? Thanks! archanamiya · talk 14:48, 18 April 2008 (UTC) Never mind! archanamiya · talk 18:55, 18 April 2008 (UTC)

Link Checker Tool not working for Call of Duty 4: Modern Warfare

Your Link Checker Tool is not working for Call of Duty 4: Modern Warfare. Could you look into that please? Thanks! Gary King (talk) 04:08, 21 April 2008 (UTC)

Also, if we are watching a page, is it possible if when we use your tool to save a page, it saves the watched status, too? If I am watching a page and use the link checker to change URLs, it will have the watchlist checkbox unchecked. But, I guess this may not be able to be done because the script can't check if the page is watchlisted by a user or not? Gary King (talk) 18:33, 21 April 2008 (UTC)

For your Readability tool, it does not work for Timbaland. Gary King (talk) 21:11, 25 April 2008 (UTC)

Hello, Dispenser. You have new messages at Gary King's talk page.
You can remove this notice at any time by removing the {{Talkback}} or {{Tb}} template.

Hey, this always seems to hang. Could you look into it please? Gary King (talk) 16:38, 30 April 2008 (UTC)

Nevermind, it finally worked after a few tries. Gary King (talk) 16:51, 30 April 2008 (UTC)

Sort template metadata

The template metadata that you have added is a very useful column. Could you sort the keys in it alphabetically, though, so that accessdate is always before publisher, etc.? This would make it easier to sift through a lot of links and see which ones are missing information, and it would make it easy to compare between two links. Thanks! Gary King (talk) 04:00, 5 May 2008 (UTC)

Dead Link Checker error

I got this error when I hit 'Save changes':

<class 'wikipedia.NoPage'>: No textarea found
      args = ('No textarea found',)
      message = 'No textarea found' 

Gary King (talk) 01:54, 12 May 2008 (UTC)

Hello, Dispenser. You have new messages at Gary King's talk page.
You can remove this notice at any time by removing the {{Talkback}} or {{Tb}} template.

A minor issue, but can you change the contents for <title> to something else besides "Main Page"? This is so that I can reach it more quickly using Firefox 3's new address bar completer. Thanks! Gary King (talk) 04:29, 13 May 2008 (UTC)

What's the timeout time for connections? It seems to be longer than usual for some reason. I feel like it's around 10 seconds? Gary King (talk) 18:34, 25 May 2008 (UTC)

Multi-article wrapper to Pywikipedia

Is there a multi-article wrapper for the tools on http://tools.wikimedia.de/~dispenser/view/Pywikipedia, particularly reflinks.py? I'd like to run it against 50-odd articles in a human-assisted-bot format, so that as soon as I save a page it goes on to the next one in the list. davidwr/(talk)/(contribs)/(e-mail) 17:38, 17 May 2008 (UTC)

For you

		The da Vinci Barnstar
		For the work you do on the link checker tool, which makes FAC so much easier. Thank you. Ealdgyth - Talk 14:53, 18 May 2008 (UTC)

WikiProxy

Copied from User talk:Daniel. 06:03, 19 May 2008 (UTC)

WikiProxy isn't handling : in titles correctly see Call of Duty 4: Modern Warfare. — Dispenser 16:01, 18 May 2008 (UTC)

I believe you're after de:Benutzer:Duesentrieb :) Cheers, Daniel (talk) 06:03, 19 May 2008 (UTC)

AWB's List Maker

Hey Dispenser, Just thought i'd let you know that im in the process of starting to overhaul AWB's List Maker.

We've made it more modular, so that plugins can be added for extra search criteria and such.

Its removed some of the things that made it more complex, and i've made the combo box now sorted as per your suggestions back in Wikipedia_talk:AutoWikiBrowser/Archive_18#Interface_Tweaks

If you SVN update you'll be able to see how much more simpler the code behind is... Was wondering what you were thinking regarding overhauling the designer (as now would be a good time to do so..)

Thanks

—Reedy 10:52, 28 May 2008 (UTC)

Outhouse article

Thank you for reforming and putting in the corrected links and references on this article. Really helps! 7&6=thirteen (talk) 12:16, 1 June 2008 (UTC) Stan

Link checker

Can you help me with a question about your reference tool that was left on my user page? I'm not at all familiar with your tool so I can't answer the question. Thanks. -- Gogo Dodo (talk) 04:59, 2 June 2008 (UTC)

Re your message: Can you leave a message for Haha169 about the issue? Thanks. You are better able to explain what happened and how he can ignore the problem then I since I'm still not quite sure what your tool does.

On a side note, when you left the comment on my talk page, your edit altered all of the timestamps on other comments to UTC-4 notation. Not sure what happened there. -- Gogo Dodo (talk) 05:41, 2 June 2008 (UTC)

Captive Market

I was just about to use autowiki browser to link Captive market - but you seem to have beat me to it :) Chendy (talk) 11:58, 4 June 2008 (UTC)

Possibilities of normal distribution numbers

It looks like you tagged all the normal-distribution redirects from 68-99.7% as having possibilities. It happens that the article 68-95-99.7 rule covers (I think) pretty much everything those numbers have to say. Is it all right if I remove {{R with possibilities}} from them? $\sim$ Lenoxus " * " 03:46, 10 June 2008 (UTC)

Worlds End State Park thanks

	<font=3> Thanks for your comments on Worlds End State Park which made Featured Article today! Dincher (talk) and Ruhrfisch ><>°° 23:57, 13 June 2008 (UTC)

Request

File:TOC template.JPG

TOC preview

I want someone to make a TOC(Tables of content) template for wikipedia like this[5]. If you know how to make, then please put on your effort as it will be good for wikipedia and its users. *The image is an edited one*. I didn't find any template that would break or split the contents into half and put the other half on right side[6]. This template will utilize the blank space and will make the contents table easily accessible without scrolling down. THANKS. Harryroger (talk) 12:24, 17 June 2008 (UTC)

AWB GUI

Really needs an overhaul

You've suggested some good ideas with the DB scanner, its just hard to work from the way thats down... Dont know if you want to try making some changes

The List Maker does need overhauling, and with the code backend of it having been overhauled mainly by myself with Kingboyk's help, a few of the obstacles from your proposed changes are sorted...

Any chance of you trying to improve it again? And we will try and implement it?

Thanks!!

—Reedy 21:37, 17 June 2008 (UTC)

I have been writing down things over the past two weeks. The database scanner was one of the first things I've gotten completed. I don't have too much time during the week to do work for Wikipedia, but over the weekend I did complete the ListMaker context menu refactoring. — Dispenser 03:41, 18 June 2008 (UTC)

Cool, patch applied in rev 2971. Thanks!! —Reedy 10:10, 18 June 2008 (UTC)

DB Scanner

Hi, I've applied that patch, and also made some more changes, tweaks, and improvements, completing some of the code and such to go with it. And also a new AWB specific scanner. See WT:AWB for more info. Give me a shout on my talk page/whatever if you need help with the code or anything. Keep up the good work! —Reedy 20:56, 21 June 2008 (UTC)

template:PDF & PDFbot

Is there something like this for TeX, LaTeX, PostScript, EPS files? 70.55.88.113 (talk) 12:24, 21 June 2008 (UTC)

Link checker error.

Hello, I've been told that there are errors with your tool. Have a look at this, click the save button, then click save page on the article. I got this message "User Riana (talk) deleted this article after you started editing it, with a reason of:", though it wasn't deleted. Could you look into this? Thanks, --Steve Crossin (contact) 05:43, 6 July 2008 (UTC)

If you just "re-create" it by checkmarking the box, then the edit will continue. Gary King (talk) 06:00, 6 July 2008 (UTC)

The tool to find links to disambiguation pages is wonderful!

Believeing that you are the creator of it, I wanted to say how much I love the dabfinder tool! As someone who works with disambiguation pages quite a bit, it is wonerfully convenient to have that search to be able to find all the links to disambiguation pages on a given page. I know the toolserver page says that it is "currently experimental and may disappear in the future.", but I certianly hope it stays around! -- Natalya 21:54, 8 July 2008 (UTC)

Re: undelete request

Undeleted and moved to your userspace. Sorry about any inconvenience. --MZMcBride (talk) 03:20, 9 July 2008 (UTC)

Help needed centre align

Would it be possible if you can centre align the template? Thanks. 144.214.102.23 (talk) 02:30, 15 July 2008 (UTC)

http://en.wikipedia.org/wiki/Template:McMaster

Help!

Why is it not checking this article?. See here.--SkyWalker (talk) 14:50, 21 July 2008 (UTC)

how do i run reflinks-svn.py

On a page called "aaa"? Nergaal (talk) 16:38, 24 July 2008 (UTC)

I see no references on that page. --Adoniscik^{(t, c)} 00:07, 8 August 2008 (UTC)

Thanks for the userpage tweak

Thanks for the userpage tweaks to the toolserver links, and for the helpful tools you've put up there. Sorry to bug you with another minor request, but do you know if there is any reason why these aren't specified in WP:IW or meta:Help:Interwiki linking? -Optigan13 (talk) 04:11, 25 July 2008 (UTC)

I need help with a template

I was working on a heavily-used template called {{Infobox Hurricane Small}} and noticed some major style issues[7]. If you look at the page in IE, you'll see that it is very different than in Firefox. I don't know enough about css to fix this. I've tried the Wikipedia help areas a couple of times, but received no responses to my posts. Can you help, or do you know someone who can?Potapych (talk) 16:39, 26 July 2008 (UTC)

Communication

A link checker at Peer review would be useful, but I think you've realised that the autogeneration of the page affects how that will work. Could you clarify for me the implementation? Thanks, Geometry guy 20:28, 17 August 2008 (UTC)

I assume your talking about the edit to {{PR/header}}, the program uses re.compile(r'<b><a href="[^"]*?/wiki/([^"]*)" title="[^"]*">', flags) to get the list of articles from the HTML. This was the only effective way of doing it on WP:TFA/R and HTML is the only general purpose way of articles out pages like WP:FAC. If you think their's a better way of implementing this or if it causes trouble from some other scripts let me know. — Dispenser 21:01, 17 August 2008 (UTC)

link checker & .pd extension

I've been using your excellent link checker tool on Geology of Somerset, which is a current GA candidate & have dealt with most of the broken links etc. However the link to Severn Estuary Barrage from the Environment Agency gives a 200 error "File extension (.pd) not in dictionary" although the link opens the PDF document fine. Any ideas or help?— Rod ^talk 18:30, 21 August 2008 (UTC)

User:Dispenser/Checklinks/json.js

I've been using this for some time after you recommended it; it's great, but any chance you could include the same information that the external tool has – especially publisher, accessdate, etc. that has been entered? Thanks in advance! Gary King (talk) 22:42, 25 August 2008 (UTC)

It makes the list easier to scroll through. However, I realize that the JSON tool and the web application version both serve different purposes and so I'm okay with only using the application version. Gary King (talk) 04:50, 2 September 2008 (UTC)

Speedy deletion of Template:S-fic

A tag has been placed on Template:S-fic, requesting that it be speedily deleted from Wikipedia. This has been done for the following reason:

The template is no longer in use and its use is contrary to the MoS.

Under the criteria for speedy deletion, articles that do not meet basic Wikipedia criteria may be deleted at any time. Please see the guidelines for what is generally accepted as an appropriate article, and if you can indicate why the subject of this article is appropriate, you may contest the tagging. To do this, add {{hangon}} on the top of the page and leave a note on [[Talk:Template:S-fic|the article's talk page]] explaining your position. Please do not remove the speedy deletion tag yourself, but don't hesitate to add information to the article that would confirm its subject's notability under the guidelines.

For guidelines on specific types of articles, you may want to check out our criteria for biographies, for web sites, for bands, or for companies. Feel free to leave a note on my talk page if you have any questions about this. Bazj (talk) 13:48, 28 August 2008 (UTC)

TfD nomination of Template:S-fic

Template:S-fic has been nominated for deletion. You are invited to comment on the discussion at the template's entry on the Templates for Deletion page. Thank you. Bazj (talk) 14:10, 28 August 2008 (UTC)

reflinks.py broken?

I am getting an error on tools:~dispenser/cgi-bin/reflinks.py?page=Manuloc&citeweb=on. Is it something with the websites, or the tool? Thanks LegoK^ontribsT^alkM 04:57, 1 September 2008 (UTC)

Template:GeoTemplate

I have no idea what you mean by "we always encourage XHTML since including has no drawbacks on HTML4 complaints", but it doesn't sound very NPoV. Please explain. Why do you think it acceptable to replace HTML examples with XHTML? Andy Mabbett | Talk to Andy Mabbett 18:32, 1 September 2008 (UTC)

1. XHTML is both backwards and forwards compatible with HTML
2. For people authoring pages in XHTML they do not need to add / into the empty tags
3. Both XML and SGML parsers can read the document

In short there are advantages for people to use XHTML and no hinderance to those wishing to remain using the HTML4 specification. Evidence to support this is that our site is in XHTML, and there are editors who convert the wikimarkup to be more XHTML like. — Dispenser 19:24, 1 September 2008 (UTC)

I find your arguments unconvincing. Whether HTML or XHTML are better is a matter of personal (some would say religious) belief. It is not our place to be proselytising for one or the other. On the above page, HTML has precedence by virtue of being there first (like GB vs. US English spellings). I've raised the issue on the templates talk page. Andy Mabbett | Talk to Andy Mabbett 19:34, 1 September 2008 (UTC)

So the argument that I can Copy and Paste easier isn't good enough for you? I should also point out that being first is an argument used as last resort to prevent edit waring between style. Here XHTML has nothing but advantages. I had assumed that with your ongoing battles to improve that markup here that you would understand. — Dispenser 19:44, 1 September 2008 (UTC)

No; your ease of copying does not convince me that the task should be made less convenient for me, and other HTML 4.01 users. Why do you think it should? XHTML has disadvantages, too; and HTML 4.01 its advantages; like I said, it's a quasi-religious matter and Wikipedia is not the pace to proselytise on it. "Being first" was used here by way of analogy between spoken and mark-up languages - just as you used that analogy in the mark-up of one of your earlier edits. Andy Mabbett | Talk to Andy Mabbett 19:50, 1 September 2008 (UTC)

What are the disadvantages in this instance? (The "/" pass validation, BTW). — Dispenser 20:22, 1 September 2008 (UTC)

I would have hoped you would have done research on that subject before claiming "nothing but advantages". As for your "I LIKE" edit summary, that seems to exemplify your position, does it not? Andy Mabbett | Talk to Andy Mabbett 20:32, 1 September 2008 (UTC)

I assume that you meant to link the "Coding In Paradise: XHTML Considered Harmful" blog entry as that is the only relevant page, all the other are specific to an implementation/method or refute it on the three pages. The stuff mentioned in the 3 year old blog entry harps on browsers and wanting his tag soup back. I still would like too see what disadvantages this change would have to HTML 4 authors. And the edit summary was aimed at what I precive as factless arguments. — Dispenser 21:08, 1 September 2008 (UTC)

Problem (perhaps transient as things move around?)

I just tried checklinks: History of the Philippines (1898-1946), witt the result that this is a redirect page which redirects to a page with the same name.

Typing History of the Philippines (1898–1946) into the search box navigated me to http://en.wikipedia.org/wiki/History_of_the_Philippines_(1898%E2%80%931946), which rendered expected page content. However, I notice that http://en.wikipedia.org/w/index.php?title=History_of_the_Philippines_(1898-1946)&redirect=no is a redirect to History of the Philippines (1898–1946).

It looks tome as if checklinks: History of the Philippines (1898-1946) is attempting to check the redirect page rather than the article page. I don't see a way to disambiguate that. -- Boracay Bill (talk) 05:22, 4 September 2008 (UTC)

Its functioning the way its designed, there are two pages: one using a long dash, the other using a short dash. If you use the short dash it will the short dash page as a redirect page and provide a link to where's its going. This work just the same as double redirect pages in MediaWiki. — Dispenser 21:57, 10 September 2008 (UTC)

I can understand that. Perhaps the naming of this article and its redirect is unsuitable for checklinks. I'll take a look at WP:Naming conventions#Special characters and try to figure out what ought to be done re the naming of this article. However,

• Typing "History of the Philippines (1898-1946)" into the WP Search bar goes to the short-dash article and takes the redirect to the long-dash article
• Typing "History of the Philippines (1898-1946)" into the checklinks box goes to the short-dash page and stops there.
• Typing "History of the Philippines (1898&ndash;1946)" into the WP Search bar goes directly to the long-dash article.
• Typing "History of the Philippines (1898&ndash;1946)" into the checklinks box causes checklinks to say, "NoPage error encountered (No textarea found)"

What do I need to type into the checkinks box using my US-ASCII keyboard in order to get checklinks to check the article? -- Boracay Bill (talk) 23:10, 10 September 2008 (UTC)

Thanks to your tool, I corrected a number of coordinates. I also added a series of missing type parameters.

There are many coordinates that just use "city" as parameter (instead of "type:city"), I don't intend to fix for now, but I was wondering if there is a way to detect other elements as parameters without type:/region:/etc prefixes, that may be worth doing so. -- User:Docu

The log probably answers my question partially. I will continue to work through it. -- User:Docu

Would you re-generate the log? Given that I made a series of repairs and was wondering how much remains. -- User:Docu

I've regenerated the log as you've asked. The size change from 99 K to 63 K. The log may contain less than all the errors due to the error correction work. The log is also in tab-separated values (.tsv format) for easy importing into spreadsheet and other applications. Hope that helps. — Dispenser 21:47, 10 September 2008 (UTC)

It does, thank you. I started working through the updated version.

If you check "Victory,_Saratoga_County,_New_York#Geography" will find that the article uses "city" instead of "type:city" added by [8]. It's an error I frequently came across. Is there a way to identify how many articles do the same? If people start mass conversions, it might be helpful to fix these at the same time. -- User:Docu

I've cleaned up, commented, and released the source. Its probably a good idea to have a separate bot to fix the rambot articles. There are 25445 of coordinates that omit "type:" and 105 are using "type_". Idealy leaving the coordinates in the infobox (reduced resolution 1/3,600) and using the hi resolution (1/1,000,000) in the title with extra parameters (city size, source, region, etc.) instead of the 4 or 5 coordinates in the articles right now. — Dispenser 18:47, 11 September 2008 (UTC)

Looks good, I will try to read it in detail. With the script, we might not need Template_talk:Coord#Maintenance_category_to_find_and_fix_excessive_number_of_input_fields and possibly Template_talk:Coord#A_degree_with_60_minutes,_a_minute_with_60_seconds as this can be checked with your log. On the other side, there is an advantage in having also a category like Category:Coor title dms template needing repair as it appears directly on the article/page being edited.

As to the rambot articles, they could indeed use a more general overhaul.

My next group of articles to fix seems to be the stubs on observatories that include empty coordinates templates (e.g. Miller_Observatory) -- User:Docu

In the meantime, I fixed all the observatories. To output a few more empty coordinates templates in article namespace, I thought it may be worth to add a few lines. Possibly, you want to try the code below.

Additions

	#after line 10 (prependCount = 0)
	semicolonCount = 0
	regionCount = 0

		#after line 34 (pagename = '')
		namespace = ''

		#Before line 53 (#Check to make sure there's atleast one number)
		namespace = pagename.split(':')[0]
		if pagename == namespace: namespace = ''

			#after line 55 (# typically templates and examples)
			if namespace=='':print('NoDigits \t%r\t[[%s]]' % (coord, pagename, ))

			#replace line 58 to fix formatting ( if pagename:print 'No N/S   
			if pagename:print 'No N/S   \t%r\t[[%s]]' % (coord, pagename, )

				#after line 68 (lon = float(coord[0].split(';')[1])
				semicolonCount = semicolonCount + 1

				#after line 145 (geotype = s.lower())
				if geotype != "city": print 'Typeless \t%r\t[[%s]]' % (s, pagename, )

				#before line 156 (if geotype:	geotype+='-'+s)
				print 'Typeless \t%r\t[[%s]]' % (geotype, pagename, )
				regionCount = regionCount + 1

	#afterline 185  (print '%d coordinates use "type_"' % prependCount
	print '%d coordinates skip region:' % regionCount
	print '%d coordinates use ";" to separate lat/long' % semicolonCount

-- User:Docu

WikiBlame

Hi Dispenser, I replied here. Best regards, --Flominator (talk) 12:13, 10 September 2008 (UTC)

source tag

Hi,

You added a source tag to my script. What does that do? Lightmouse (talk) 10:57, 11 September 2008 (UTC)

Syntax highlighting. —Reedy 11:20, 11 September 2008 (UTC)

Thanks. Lightmouse (talk) 11:56, 11 September 2008 (UTC)

Lost in translation

Hi, thanks for fixing up the spelling in List of Norwegian ministries. However, your semi-automatic editing was "spell checking" the Norwegian language versions of the department names, which uses departement (the correct spelling in Norwegian), and then "fixing" them into the English department. I have reverted the edit, but please be weary of this pit in the future. Thanks, and happy editing. Arsenikk ^(talk) 17:41, 13 September 2008 (UTC)

Email/Script

Hi Dispenser, I sent you an email about the use of your readability script. I would like to run the script for about 2000 articles starting from tonight. I hope it's ok with you. Thanks. --Ragib (talk) 00:58, 18 September 2008 (UTC)

Thanks for the tip about the scripts. I noticed that you have already deployed v 1.4 and v 2.0. However, the v2.0 gives a different result than the previous versions ... have you finished coding it?

It will be quite nice if we can use the readability script to test readability of a particular version of a page. Can your script do that now? Thanks again in advance. --Ragib (talk) 03:01, 24 September 2008 (UTC)

By the way, in v2.0, you refer to the syllable module, but I don't see the code for that under the resources directory. Where do I get that? --Ragib (talk) 03:01, 24 September 2008 (UTC)

References script

Hey, I've noticed lately, (such as here), that your script is trying to change dates, such as "|accessdate=2007-12-28 " to "|accessdate$1=$2$3-$4-$5 ". Thought you'd appreciate the heads up. SQL^{Query me!} 07:12, 21 September 2008 (UTC)

Additionally, this throws an error. SQL^{Query me!} 07:14, 21 September 2008 (UTC)

They appear to have been fixed (or, for whatever reason, are working now) Thanks! :) SQL^{Query me!} 02:26, 22 September 2008 (UTC)

Checklinks

It hasn't been working for me lately; it constantly gets stuck at the uploading part. Is this happening to everyone at the moment, or is it just me? Gary King (talk) 15:43, 28 September 2008 (UTC)

It is working for me.--SkyWalker (talk) 15:52, 28 September 2008 (UTC)

Okay, it works for me for smaller articles. For ones with many links, it doesn't work. I'll try giving it a bit longer to work, like a few minutes. Gary King (talk) 15:56, 28 September 2008 (UTC)

Works on, say, James Blunt, but not on The Pirate Bay. Gary King (talk) 16:00, 28 September 2008 (UTC)

It work on The Dark Knight. Which has more than 100 links.--SkyWalker (talk) 16:01, 28 September 2008 (UTC)

I tested with James Blunt has you requested. It works fine. Let me test some more articles.--SkyWalker (talk) 16:05, 28 September 2008 (UTC)

Have you tried The Pirate Bay? I'm thinking that the problem is caused by one of the links on the page, probably one that is timing out. Gary King (talk) 16:35, 28 September 2008 (UTC)

There has been some configuration changes by River over the past month. One of which has been to enable fastcgi for all PHP applications and a case by case basis for other languages. Presumably, some time constants have been changed for regular cgi applications. There isn't really much I can do other than lower internal delay constants which doesn't solve the issue. — Dispenser 16:50, 28 September 2008 (UTC)

Alright; anything I can do to The Pirate Bay to make the tool work for it? Gary King (talk) 16:53, 28 September 2008 (UTC)

It is working for TPB too.--SkyWalker (talk) 17:10, 28 September 2008 (UTC)

Well it looks to be working for me. If you can see the standard footer it mean the script wasn't killed. You can play with some of the settings by adding &debug=1 in the URL. I also confirmed when threading was disabled, completing the run in 157 seconds which is far beyond the kill time for fast-cgi scripts. It may be that your browser is timing out, but most are set to timeout at something like 4 minutes. — Dispenser 17:20, 28 September 2008 (UTC)

Ah it works; it just didn't redirect for me to the preview page automatically like it does sometimes. Gary King (talk) 17:21, 28 September 2008 (UTC)

I've since removed the review of webreflinks.py feature from mergeChanges.py. I'm switching to the DumZiBot's reflink.py so that titles will be tagged as bot generated. — Dispenser 18:58, 28 September 2008 (UTC)

Coordinates (2)

Now that most coordinates were converted to {{coord}}, would you help me by re-generating the coordinates' log? There shouldn't be many changes compared the previous templates, but I'd like to make sure. The remaining "old" templates are mostly on Rambot/Smackbot pages.

BTW if you want, I can post a copy of your script including the changes suggested at #Coordinates to a page on your user talk "namespace". These changes as such are not needed to verify the conversions. -- User:Docu

I have rerun the script with your some of you modifications in place. You probably also want the following errors which aren't in the log:

More errors

./geodbcompiler.py:231: Warning: Data truncated for column 'region' at row 1
./geodbcompiler.py:231: Warning: Data truncated for column 'scale' at row 1
./geodbcompiler.py:231: Warning: Data truncated for column 'type' at row 1
./geodbcompiler.py:231: Warning: Incorrect integer value: 'building' for column 'scale' at row 1
./geodbcompiler.py:231: Warning: Incorrect integer value: 'city' for column 'scale' at row 1
./geodbcompiler.py:231: Warning: Incorrect integer value: 'landmark' for column 'scale' at row 1

— Dispenser 06:46, 5 October 2008 (UTC)

Great thanks. I think I fixed quite a few of them. At WP:GEO, I left a few that still need repair.

Geobox Rivers generated a lot of typeless errors, I fixed those, but not the ones generated by other Geoboxes. Most "Number:" errors seem to come from geoboxes as well. I left them for now, as I haven't quite figured out how they work. -- User:Docu

Thank You

		The WikiChevrons
		For locating several dead external links within the articles USS Missouri (BB-63) and USS Wisconsin (BB-64) I herby award you the WikiChevrons. Thanks for your help, and keep up the good work! TomStar81 (Talk) 06:30, 6 October 2008 (UTC)

User:davidzundel

Thanks for {{tl}} and fixing the minor issues of <br> and underlines. Unfortunately the {{tl}} edit corrupted the formating of the shortcut box. You might have asked first. -- davidz (talk) 03:00, 13 October 2008 (UTC)

Dispatch

Dispenser, can I interest you in writing a Dispatch describing the Featured article tools? I'm interested in encouraging editors to 1) install the tools on talk page as I did at Talk:Tourette syndrome so I can watch for dead links, 2) use the dab and dead link checker more frequently at FAC and FAR, and 3) we need to decipher how to use the readability indeces. If you're interested, you could start a draft at WP:FCDW/TempTools, and others will help copyedit and tweak. Sample Dispatches are at {{FCDW}}. SandyGeorgia (Talk) 05:46, 15 October 2008 (UTC)

There appear to be some quirks with the readability calculators. Taking Exile's Valor as an example, it says 59 words and 82 syllables, but 1.55 syllables per word (82/59 = 1.39). It seems to correctly plot 155 syllables per 100 words on the Fry graph, but 19.67 words/sentence would mean 5.08 sentences / 100 words; on my browser it shows up on the 4.8 line. It's not just that this is a stub - these quirks appear on larger articles, too. Any ideas? Gimmetrow 01:45, 20 October 2008 (UTC)

By the way, I had starting writing up something about the tools. I was intending to surprise Sandy at some point when the dispatch was behind, but I quit when you removed the old readability tool. Also, I've noticed the checklinks tool converts some manually-formed citations to cite web - which may not be appropriate in an article which doesn't use cite web. Can you do anything about that? Gimmetrow 19:19, 21 October 2008 (UTC)

New bug in reflinks.py

Hi. I love reflinks, but it's been making an odd error lately. It is removing {{cite web from the beginning of properly-formatted refs. E.g., on Halle Berry it was trying to change

<ref>{{cite web|url=http://www.genealogy.com/famousfolks/halleb/d0/i0000001.htm#i1|title=First Generation}}</ref>

to

<ref>http://www.genealogy.com/famousfolks/halleb/d0/i0000001.htm#i1|title=First Generation}}</ref>.

Thought you might want to know. [ roux ] [x] 18:11, 17 October 2008 (UTC)

Timestamps on 'in use' and 'under construction' templates

I saw your comments about {{REVISIONTIMESTAMP}} but I didn't quite understand the technical issue about bugs. Frankly, I think 'time since last edit' is not as good as 'time since tag was placed'. I am actively editing your talk page now with my fingers on the keyboard so I might slap an 'in-use' tag on it if it were an article (I can't imagine myself actually doing that). My claim that your talk page is 'in-use' starts now and the clock that expires my claim starts ticking now. The logic of fingers-active-on-keyboard depends on the activity immediately following application of the tag, not any subsequent edits. My claim for 'in-use' does not even depend on any subsequent edit that I might make because that could just be a routine edit.

I have the impression that many people are nervous to remove these tags on active pages even though all the recent edits were routine.

Thus it may not matter that 'time since last edit' is technically difficult. Do you think that it would be technically easier to work with 'time since tag was placed'? Regards Lightmouse (talk) 21:32, 21 October 2008 (UTC)

Links to newspapers in refs

Hi, Dispenser. When you were running Doctor Who missing episodes through your dead links bot, I see that you also removed some internal links to media sources in the references (e.g. changing "The Guardian" to "Guardian"). Is that standard usage? I had thought that it was considered informative to readers who might not be familiar with a given media source to link to our page on it. I don't much care about the use of the article ("The"), but why delink? Just curious. —Josiah Rowe (talk • contribs) 18:37, 24 October 2008 (UTC)

I look at it as improving the signal to noise ratio to both the page and backlinks. WP:OVERLINK seems to support this. The software will only unlink from an internal list of common publishers. You may wish to comment on the barrowing of functionality at a bot request while I continue to develop the script. — Dispenser 18:09, 27 October 2008 (UTC)

I'm not sure what you mean by "barrowing of functionality". I can see the argument for delinking sources like BBC or The New York Times, which are well-known internationally, but I'm not sure that (for example) The Observer is sufficiently well-known outside of the UK to justify delinking. I've asked at WT:OVERLINK#Names of common publications in references whether there has been any previous discussion of this, and whether there's an established consensus.

I see that on the bot request page the "unlink common publisher" function has been stricken out. Does that mean that you're not putting that into the final bot? (If there isn't really a consensus supporting it, that's probably a good idea.) —Josiah Rowe (talk • contribs) 18:44, 27 October 2008 (UTC)

Reflinks spelling mistakes

Hi, this isn't a major problem by any means, but Reflinks has a spelling mistake that bugs me every time I use it. When running it, the line ""title" yeilds: [u'THE WORLD ECONOMY']" has a spelling mistake; "yeilds" should be "yields" :) Gary King (talk) 21:41, 24 October 2008 (UTC)

AWB Feature requests

A changed a word in your new entry from sounding to surrounding - [9]. Not sure if that's what you meant? --Tagishsimon (talk) 21:06, 13 November 2008 (UTC)

Link removal

Hiya... regarding the link removal from the Public Register, the GG's site underwent a redesign a while ago, changed all their URLs. I was just too lazy last night to hunt down the new version, heh. //roux   22:45, 20 November 2008 (UTC)

Thanks!

for the link fixes in my userspace. - Dan Dank55 (send/receive) 21:02, 27 November 2008 (UTC)

Barnstar!

I'm not sure how you feel about these, but...

		The da Vinci Barnstar
		For the amazing webreflinks.py tool, which has simplified the life of many an editor. It's an excellent tool that should be rolled into the MediaWiki platform. Well done, and thank you very much for coding and maintaining it. //roux   09:16, 29 November 2008 (UTC)

Checklinks & infoboxes

Thanks for the tool. I find very useful and it saves a lot of work. One thing I notice and I believe can be improved is that it does not seem to pickup links in infoboxes (within templates?). Try it on Pyramid Peak (Colorado). In the infobox template there are two links that are missed. Thanks again. --DRoll (talk) 04:33, 2 December 2008 (UTC)

PS4

Due to PlayStation 4's deletion, I feel that this is the only way to define the intrigue of the PS4's release. Therefore, as I don't think the "PS4" search commonly refers to Phantasy Star IV, the PS4 article should reinstate my exclusive information. Also, this is official details about the next PlayStation, so why not? Redirects do not help Wikipedia expand. Please take this into consideration.  Mike |talk 20:02, 7 December 2008 (UTC)

I have userfied your entry at User:Trap The Drum Wonder/PlayStation 4. When the article meets notability guidelines and cities reliable published sources you may ask an administrator to move it to PlayStation 4. You may find Wikipedia:Your first article to be helpful in construction. However, a search revivals other reveals PlayStation 4 including User:AlexanderTG/PlayStation 4 which shows out of date predictions (PS2 compatibility, controller inputs). — Dispenser 20:28, 7 December 2008 (UTC)

Thank you, but as I have not humorously written about the PS4, but have simply given original information that people are bound to be interested in. Could I please create a disambiguation by the name of "PS4 (Console)"? As it is not interfering in your redirected article.  Mike |talk 20:39, 7 December 2008 (UTC)

Reflinks tool

I use your tool to edit bare references to cite web format. Thank you for the tool, it does a lot of editing automatically that I had been doing manually. I have a suggestion. I would ask if it is possible to take what currently automatically fills into the "publisher" field and have it fill into the "work" field instead? Usually that information doesn't truly apply to the publisher field. The name of a newspaper or a website, for example, are normally what would be considered to be the name of the work and the actual true publishing company is some other entity. It also renders the field in italics format, which looks a lot better.--2008Olympian^chitchat 11:32, 8 December 2008 (UTC)

PDFbot

Hi Dispenser. Is PDFbot still active? I've always found it very useful, and was hoping it would go through some articles. Jayjg ^(talk) 05:21, 26 December 2008 (UTC)

Well, I'm glad to hear it's still alive; I use it all the time for my articles, it's very helpful. Jayjg ^(talk) 17:55, 28 December 2008 (UTC)

question what is going on with PDF links for NRHP sites

In this diff on the Chautauqua Institution article and others on my watchlist, your PDFbot has been changing PDF links. I don't quite understand what is being done. I have cumulatively added about a zillion PDF links to National Register documents, and I am adding more as time goes by, and I maintain wp:NRHPMOS and provide other guidance to other NRHP editors. I'd like to understand, so at least I could revise my own practices and update my guidance. By the way, i was on a 2 month or so wikibreak so you may have done a lot that i did not notice already. doncram (talk) 21:37, 30 December 2008 (UTC)

For example in this edit of Falls of Clyde article, you revise a footnote that had 2 PDF links to now show 2 different types, one being a url link with a PDF symbol following, one being a PDFlink with a PDF file size showing. I don't grok why you'd want to mix 2 ways of linking to PDF files in the same reference. I have created thousands of these references that have 2 PDF links, one to a text PDF document and one to accompanying PDF photos document. I notice the bot is hitting more articles on my watchlist since i asked question above. Can you please stop and discuss? doncram (talk) 21:56, 30 December 2008 (UTC)

stop the bot please

I judge that the bot is malfunctioning. Please stop and discuss. I posted at Wikipedia:Administrators' noticeboard/Incidents#admin help needed to stop PDFbot or some such title, asking for an admin to hit the big red stop button. doncram (talk) 22:08, 30 December 2008 (UTC)

bot was "blocked" by an admin there, who offers to unblock upon request. doncram (talk) 22:18, 30 December 2008 (UTC)

Hi Dispenser. I assume u had other business to attend today. I know your bot was making some improvements, and I expect that the NRHP references involving 2 PDF links can be improved from the model i and some others have been implementing widely, but I do think the current bot was not addressing these properly. Not aware of what would be right forum to discuss proper formatting in references for PDF links. Would be happy to consult any guidelines, any forum you suggest. doncram (talk) 04:35, 31 December 2008 (UTC)

Thanks for responding at my talk page. I just asked at Wikipedia talk:Featured article candidates#Help needed on reference PDF formatting question for opinions on what should be done with the semi-complex NRHP references involving 2 PDFs, referring to the copy of discussion you put in at my talk page. Will you watch wikipedia talk FAC and/or my talk page? Thanks doncram (talk) 18:03, 31 December 2008 (UTC)