User talk:Dispenser

From Wikimedia Commons, the free media repository
Jump to navigation Jump to search
Welcome to Wikimedia Commons, Dispenser!
Afrikaans | Alemannisch | asturianu | azərbaycanca | Bahasa Banjar | català | čeština | Cymraeg | dansk | Deutsch | Deutsch (Sie-Form)‎ | English | español | Esperanto | euskara | estremeñu | français | Frysk | galego | hrvatski | Bahasa Indonesia | interlingua | Interlingue | íslenska | italiano | Kiswahili | Kurdî | Latina | lietuvių | magyar | Bahasa Melayu | Mirandés | Nederlands | norsk bokmål | occitan | Plattdüütsch | polski | português | português do Brasil | română | rumantsch | Scots | shqip | sicilianu | slovenčina | slovenščina | Basa Sunda | suomi | svenska | Tagalog | Türkçe | vèneto | Tiếng Việt | Zazaki | Ελληνικά | беларуская | беларуская (тарашкевіца)‎ | български | Ирон | македонски | нохчийн | русский | српски / srpski | тоҷикӣ | українська | ქართული | Հայերեն | नेपाली | भोजपुरी | मराठी | हिन्दी | অসমীয়া | বাংলা | தமிழ் | മലയാളം | සිංහල | ไทย | ၽႃႇသႃႇတႆး  | မြန်မာဘာသာ | 한국어 | 日本語 | 中文 | 中文(台灣)‎ | 中文(简体)‎ | 中文(繁體)‎ | 粵語 | עברית | اردو | العربية | تۆرکجه | سنڌي | فارسی | +/−

Giggy (talk) 01:35, 6 November 2008 (UTC)[reply]


Ships by IMO number

[edit]

Hi Dispenser,

For categories in Category:Ships by IMO number, it would be interesting to have a page that links to various websites about ships that can be queried by IMO number, similar to booksources or the coordinates page. I made a request for a page similar to booksources in bugzilla, but it seems that it's unlikely to happen. Thus I was wondering if you would set one up on toolserver. --  Docu  at 07:43, 16 June 2012 (UTC)[reply]

As you may know, I don't particularly like GeoHack. Visitors are confused by the copious links, many of them second rate, and only use five out all the links. This mean the reset of the page is a waste. And then there's an ideological argument to prompt or limit links that meet a FLOSS definition.
Back to the topic, possibly the nicest way to implement this in JavaScript using &withjs= on the links. This would make editing, adding features, and maintenance easier and more reliable. Plus, it'll be implemented a lot sooner. You might need to pass the script off as a gadget to get it into the MediaWiki namespace. —Dispenser (talk) 03:22, 19 June 2012 (UTC)[reply]
Sounds good. Is there a sample I could build on? I checked Special:Preferences#mw-prefsection-gadgets, but there seems nothing similar. --  Docu  at 05:24, 23 June 2012 (UTC)[reply]
/* [[Commons:IMO sources]]
 * Quick and dirty for generic sources pages
 */
$(function() {
    // Retrieve and escape the replacement for MAGICNUMBER
    var mn_replacement = mw.util.getParamValue("imon");
    if(!mn_replacement) {
        // Null or blank value
        return;
    }
    if(!mn_replacement.match(/^\d+$/)) {
        // Invalid value
        return;
    }
    // Replace HTML with magic number text
    $("#content")[0].innerHTML = $("#content")[0].innerHTML.replace(/MAGICNUMBER/g, mw.html.escape(mn_replacement));
    // Remove the no JavaScript notice
    $("#nojs-notice").remove();
});
Drop that in MediaWiki:IMO_magic.js and open the specially crafted link. Dispenser (talk) 21:11, 23 June 2012 (UTC)[reply]
Excellent. Looks very compact. I made the corresponding edit request at MediaWiki talk:IMO magic.js. --  Docu  at 06:48, 24 June 2012 (UTC)[reply]

Mediawiki error lists

[edit]

Could you please create a pair of subpages and give the full lists of effected files for the errors you brought to the village pump? Thanks! Sven Manguard Wha? 15:27, 1 September 2012 (UTC)[reply]

All files with problems that I know of have been listed on the wiki. Unless you're referring to Image scaler problems which should be done within two weeks. Or the secret report on frozen GIFs. —Dispenser (talk) 22:52, 1 September 2012 (UTC)[reply]

Update

[edit]

Could you please update the list if this is no big deal. (I cleared some percent of all files and they could be deleted from this list "automatically"). --McZusatz (talk) 17:32, 7 September 2012 (UTC)[reply]

Done. —19:05, 7 September 2012 (UTC)

Hi Dispenser. What happened here? The image was working fine before, and File:The Naval Temple, at The Kymin, Monmouth 2.jpg was uploaded in the same minute yet doesn't exhibit any problem? Thanks. Mike Peel (talk) 19:15, 9 September 2012 (UTC)[reply]

I don't know. Likely an issue retrieving with the new storage engine (considering the horrible hack for "long" file name). Anyway, according to Bug 39615 WMFer User:Aaron Schulz is working to fix this. Dispenser (talk) 03:33, 21 November 2012 (UTC)[reply]

Broken Videos

[edit]
No. Error type
3 File:Dmitry Volodikhin.ogg
Exitcode: 1
2970   av_interleaved_write_frame()
33   Error while opening encoder for output stream
11   could not find codec parameters
3   Audio encoding failed
4   Unable to decode input.
2   Output file #0 does not contain any stream
10 Exitcode: 134
1 Exitcode: 136
34 Exitcode: 137
5 Exitcode: 139
3076 TOTAL

Do you know how to find all "broken" videos such as this one? --McZusatz (talk) 17:30, 20 November 2012 (UTC)[reply]

I don't have too much time for investigations this month. I'd try asking on IRC (#wikimedia-tech) if someone could just grep the database. Or with Commons having only 25,208 videos it would take ~8 hours to download each description page and check for class="errorlink". Whichever way, the transcoding queue is probably backed up for the next month and the complete error report with it. —Dispenser (talk) 04:46, 21 November 2012 (UTC)[reply]
Thanks for the help so far. What do you think is the best method to download each description page? --McZusatz (talk) 12:53, 22 November 2012 (UTC)[reply]
jira:TS-1570. —Dispenser (talk) 21:31, 23 November 2012 (UTC)[reply]
Is it likely to get a response? --McZusatz (talk) 13:55, 30 November 2012 (UTC)[reply]
I poked DaB on IRC and he quickly added it. Sorry about being slow in posting a report. I left 1,655 av_interleaved_write_frame() affected files (11%) as it might be an implementation issue. A few interesting tid bits: the software upscales 240p ogg to 360p webm, the transcoding timestamps can be used for load and CPU time graphs, 26,969 video files. —Dispenser (talk) 07:42, 17 December 2012 (UTC)[reply]
Thank you for the list. I will have a look at the stuff soon. --McZusatz (talk) 09:09, 17 December 2012 (UTC)[reply]
Btw. I do not think the software upscales 240p ogg to 360p webm. (See: https://upload.wikimedia.org/wikipedia/commons/thumb/1/14/Tmh.rev.test.ogv/Tmh.rev.test.ogv.360p.webm ) --McZusatz (talk) 22:16, 17 December 2012 (UTC)[reply]

Could you please do another update of your sandbox. I cleaned a lot of the files. --McZusatz (talk) 14:15, 24 May 2013 (UTC)[reply]

Done, related bug 46650. Dispenser (talk) 04:23, 26 May 2013 (UTC)[reply]
Thanks again. Jan Gerber requested the "av_interleaved_write_frame()"-errors to be reset-transcoded and I fixed the remaining files in your sandbox. I think it would be appropriate to update the list from time to time. Maybe every two weeks... What do you think? --McZusatz (talk) 14:44, 3 June 2013 (UTC)[reply]
It doesn't appear the extension author wants to acknowledge let alone address the transcoding issue. So I've moved the report to User:Dispenser/Transcode errors. The av_interleaved_write_frame() errors are still appearing despite bug 46795 claiming it was fixed. I usually run report whenever I feel like it, but I probably should implement something similar to Commons:Database reports. —Dispenser (talk) 15:06, 5 June 2013 (UTC)[reply]

Mismatched extensions

[edit]

Rillke did some great progress on the cleanup. Could you please update the list? --McZusatz (talk) 11:47, 18 December 2012 (UTC)[reply]

Done Dispenser (talk) 18:29, 18 December 2012 (UTC)[reply]

Autopatrol given

[edit]

Hello. I just wanted to let you know that I have granted autopatrol rights to your account; the reason for this is that I believe you are sufficiently trustworthy and experienced to have your contributions automatically marked as "reviewed". This has no effect on your editing, it is simply intended to make it easier for users that are monitoring Recent changes or Recent uploads to find unproductive edits amidst the productive ones like yours. In addition, the Flickr upload feature and an increased number of batch-uploads in UploadWizard, uploading of freely licensed MP3 files, overwriting files uploaded by others and an increased limit for page renames per minute are now available to you. Thank you. --McZusatz (talk) 14:09, 26 January 2013 (UTC)[reply]


Coordinates log

[edit]

Hi Dispenser,

Long time since I looked into this. It seems that finally someone else is working on them, see Template_talk:Location#10k_files_in_Category:Pages_with_malformed_coordinate_tags

BTW, the report at https://toolserver.org/~dispenser/logs/coord-commonswiki.log worked on the 26, but displays an error message today. Maybe it just needs re-starting. --  Docu  at 11:14, 1 March 2013 (UTC)[reply]

Kannst du die Liste bitte noch einmal aktualisieren? --McZusatz (talk) 19:11, 24 August 2013 (UTC)[reply]

Danke! Ich habe heute alle Videos repariert, soweit ich konnte. Falls du Lust und Zeit hast, würde ich mich über regelmäßige updates freuen. :) --McZusatz (talk) 17:33, 25 August 2013 (UTC)[reply]

This list looks really amazing. How did you create it? It seems for me that a lot of CPU cycles were consumed to achive this. --McZusatz (talk) 15:10, 27 August 2013 (UTC)[reply]

It works similar to GIF check, except a python script downloads the images and runs zip -9, jpegtran -copy all, and unrar l before printing the results. It takes about 4 hours to run the first time. I'll publish the source code later. We still need to figure out what's bloating the other files. —Dispenser (talk) 16:17, 27 August 2013 (UTC)[reply]
Some of the use weird compression but are likely not harmful. (File:Nevtor Oy - Logo (2012).jpg)
 Exif:
 Compression scheme 	Uncompressed
 Data arrangement 	chunky format
 Color space 	Uncalibrated
--McZusatz (talk) 13:23, 28 August 2013 (UTC)[reply]
The "harmful" ones likely have 0% zip compression and -80% to -100% jpegtran. Now a 100x26 image should be (100 wide)(26 tall)(3 bytes/pixel) = 7.6 KB uncompressed true-color bitmap. The other 560 KB appears to be data mostly unrelated to the image. That it was created with Adobe Photoshop CS3 Macintosh in 2009, Adobe XMP and Apple Property List XML junk, junk data repeated a few times, and lots of 0xFF bytes. Still a work in progress and nothing finished. Dispenser (talk) 20:10, 28 August 2013 (UTC)[reply]
So the half a megabyte of size is due to the embedded color profile, which is also copied to the smallest thumbnails. Color space support is bad outside of photography applications, even with the introduction of wide color gamut OLED phone displays. The image scalers should convert to sRGB below a certain pixel size like 64x64 (We'll need a bug for that). I've also got the report back working (didn't actually need unique img_name→page_id mapping), so expect an update soonish. Dispenser (talk) 16:43, 26 September 2013 (UTC)[reply]
Color profiles are important sometimes... File:Jigsaw-PNL.jpgDispenser (talk) 18:44, 29 September 2013 (UTC)[reply]
Yes, indeed. The automatic detection of the color profile failed sometimes. All instances should be fixed by now. --McZusatz (talk) 16:40, 30 September 2013 (UTC)[reply]
Since color profiles are copied to thumbnails I'm now recording file size in my thumbnail request script. Color profile estimates are now available for 2.1 million images larger than 48x48. This has reduced the false positives.
I've obtained a copy of Fireworks begun tagging images with multiple layers. However, I'm unsure what to do with those using it as an export tool (single layer, same as the output). --Dispenser (talk) 16:16, 16 May 2014 (UTC)[reply]

On the most recent version: What does the MDA mean? --McZusatz (talk) 19:13, 16 May 2014 (UTC)[reply]

Metadata Attributes (took a while to recall that, introduced in September). A count of metadata values extracted by MediaWiki in hopes of improving pre-selection, but MW isn't keeping it updated so likely it'll be dropped. —Dispenser (talk) 20:13, 16 May 2014 (UTC)[reply]

Uncompressed PNG

[edit]

Some programs and/or users save PNG files in a uncompressed way. (BMP-like compression). I would appreciate it, if you forged a query to find all PNGs which loose at least ~20% file size by recompression. --McZusatz (talk) 18:21, 23 September 2013 (UTC)[reply]

I've partially integrated it into the Absurd overhead, but I've limited to minimum of 750 KiB of estimated overhead until we better understand what data we're throwing away. —Dispenser (talk) 04:50, 1 October 2013 (UTC)[reply]

Do you know TweakPNG? You can use this tool to look into suspicious PNG files. From what I see none of these files is harmful or dangerous. They just contain some unused chunks. For example, this file was made with Adobe Fireworks and contains all information needed to recreate the image with Adobe Fireworks (which can be good and should not be dropped). A PNG optimizer drops this stuff. Our image scaler does the same. All non-original image sizes don't contain these chunks. Therefor it's pretty much a waste of time to optimize these images. Very few users are downloading the original image size anyway. --TMg 13:54, 4 October 2013 (UTC)[reply]

Looks useful. Too few Linux ports of these PNG utilities :-(. I've created {{Fireworks PNG}} to identify Fireworks files and will excluded them from the report. It's up to the community to implement this (debate on propriety formats/FLOSS religion and should be had at Commons:Village pump#Compression). To make it easier, I've already noted Fireworks files on the list. Also, the source file is sent when thumbnail width ≥ source file, which is happening to 2,000+ PNGs. —Dispenser (talk) 05:06, 5 October 2013 (UTC)[reply]
Did you saw the discussion about progressive JPEG files? It was the more or less same: A user tried to convert all these files to non-progressive. He was stopped because what he did wasted more bandwidth, space and CPU power than it would have saved in the next 100 years. I think it's the same here. Don't waste your time with this, please. Don't "fix" files that aren't broken. --TMg 22:21, 8 October 2013 (UTC)[reply]

TIFF check?

[edit]
Over 50 megapixel
Type Files
JPEG 13,849
SVG 1701
TIFF 316
PNG 237
DjVu 208
PDF 62
GIF 16

Hi Dispenser. This this is similar to your current GIF check, so I was wondering if you would be able to produce a report of TIFFs that fail to render previews because the file is above the current megapixel limit ($wgMaxImageArea is currently set to 50MP). This is affecting some of my uploads, and I'd like to see how widespread the issue is. Thanks! Dominic (talk)

Right now my thumbnail caching script reports 173 TIFFs give the The resolution of the source file is too large. No thumbnail will be generated error, with another 5 giving no error message at all. PNG and TIFF files currently ignore $wgMaxImageArea. The PNG limit has been lifted, although I still have trouble with File:WorldMap-A non-Frame.png. Tomorrow, I'll have a proper query. —Dispenser (talk) 06:32, 28 September 2013 (UTC)[reply]
filed bugzilla:54744 for the PNG file. --McZusatz (talk) 19:12, 28 September 2013 (UTC)[reply]
The caching script may have incorrectly marked some files as thumbnailable, I'll retry those above 50 MP. Do you need the file list or is a summary sufficient? Dispenser (talk) 18:05, 29 September 2013 (UTC)[reply]
I've posted the query to Category talk:TIF files affected by MediaWiki restrictions and User:Fae has tagged the remaining images. —Dispenser (talk) 04:34, 1 October 2013 (UTC)[reply]

DjVu files from archive.org

[edit]

Hello. Because of [Wikisource-l] archive.org switched to ABBYY FineReader 9.0, I'd like to have a list of all DjVu files on Commons (but also Wikisources if one wanted) which link to an archive.org item as their source, ideally split in two lists: 1) DjVu files used on some Wikisource, but with some Page: pages missing (this means the old text was not yet completely copied to wiki pages), 2) DjVu files not yet used on any Wikisource. Can you help with this? If I had a list of item,filename pairs I could rerun the OCR on archive.org and then update our copy of the files. --Nemo 07:45, 8 October 2013 (UTC)[reply]

To answer your question, yes, https://archive.org/details/VocabolarioDellaLinguaItaliana2 is the pattern I'm looking for. --Nemo 16:11, 9 October 2013 (UTC)[reply]

Do you know why File:Wikimania 2011 - Statistical analysis and MediaWiki development.ogv does not appear in the list? --McZusatz (talk) 22:31, 11 October 2013 (UTC)[reply]

I had it set to ignore empty errors. Its now fixed and I've reset the transcodes for the affected images. —Dispenser (talk) 06:18, 12 October 2013 (UTC)[reply]

About a tool

[edit]

Hi, just wondering about something, sorry if this isn't the ideal place to ask; it seemed like this is somewhere you're relatively active.

There's a tool over on toolserver, embeddedincount, that looks like it was yours. It's used in the {{high-risk}} templates on en.wn and en.wb. We've been switching such links over from toolserver to wmflabs, of course, but as best I can tell, that tool doesn't have a wmflabs equivalent. Do you know of an equivalent that exists, or that you or anyone else plans to create? --Pi zero (talk) 20:20, 3 May 2014 (UTC)[reply]

I asked the Foundation to allocate 24 Terabytes of non-backuped storage for tool improvements and they refused. The WMF made clear it does not value me nor the community. More concerned with soliciting donations, apparently. --Dispenser (talk) 16:55, 16 May 2014 (UTC)[reply]
As a Wikinewsie, I can appreciate bitterness about WMF priorities — since they've been leaving Wikinews out in the cold (seemingly hoping it would die of exposure) for years. They apparently don't realize that, just because they don't believe in the importance of what we do, doens't mean we don't believe in it so deeply we're willing to keep going despite them. (The same goes, come to think, for the social atmosphere at Wikipedia, which has turned rancid since I first arrived there but the project continues because lots of great people care about the work enough to do it anyway.) --Pi zero (talk) 13:28, 17 May 2014 (UTC)[reply]

Hi, could you please update the list. I am working on it, and a lot of files have already been renamed. Thanks in advance, Yann (talk) 23:27, 31 January 2015 (UTC)[reply]

Hi! I'm writing to let you know that I have implemented most of your suggestions from Special:Diff/228880645. I haven't changed coefficients for videos because current values IMO work quite well. Also I don't understand what exactly do you mean by "the function should be inverted". --jdx Re: 22:06, 14 January 2017 (UTC)[reply]

@Jdx: Right now it checks:
mime=="image/png" and size > Y,
this fails with mime="image/new_format", reverse the logic ("invert") to flag outside the valid scope (list what to accept instead of deny):
not(mime=="image/png" and size <= Y). Dispenser (talk) 22:37, 14 January 2017 (UTC)[reply]
Do you mean something like this:
[...]
  | (file_mediatype == "VIDEO"
   & ( !(file_mime == "video/webm"     & file_size <= int(136.8*1024*1024))   /* 84.1% */
     & !(file_mime == "application/ogg"& file_size <= int( 97.2*1024*1024)) ) /* 97.7% */
    )
[...]
instead of:
[...]
  | (file_mediatype == "VIDEO"
   & ( (file_mime == "video/webm"     & file_size > int(136.8*1024*1024))   /* 84.1% */
     | (file_mime == "application/ogg"& file_size > int( 97.2*1024*1024)) ) /* 97.7% */
    )
[...]
--jdx Re: 00:23, 15 January 2017 (UTC)[reply]
@Jdx: I was thinking simpler than that brain teaser:
[...]
& action == "upload"
/* Whitelisted mediatype, mime, and filesize */
& ! ( (file_mediatype == "BITMAP"  
   & ( (file_mime == "image/jpeg"     & file_size <= int( 26.5*1024*1024))   /* 99.9% */
     | (file_mime == "image/png"      & file_size <= int( 41.1*1024*1024))   /* 99.9% */
  [...]
Dispenser (talk) 00:57, 15 January 2017 (UTC)[reply]

Failed transcodes

[edit]

Hi, Could you please make a list of small (160p to 480p) failed video transcodes? Thanks in advance, Yann (talk) 20:08, 29 January 2017 (UTC)[reply]

@Yann: I stopped updating User:Dispenser/Transcode errors a few years ago as I realize the Foundation wasn't going to fix problems, just issue press releases. An example: Google announced to drop H.264 from chrome by January 2012. In 2013, WMF could've publicly pressured Google for not doing so. Today, the most open browser Firefox has H.264 support and our videos are rendered in JavaScript.

Anyway, the backlog is: 70,453 failed video transcodes for audio files, 316 failed audio transcodes for video files, and 991 video files with various failed transcodes. I was able to manually clear out some of these with the reset button. —Dispenser (talk) 05:34, 30 January 2017 (UTC)[reply]

Hi, Thanks for your message. I have opened several reports, and some bugs were fixed recently. Could you please make a list of 316 failed audio transcodes for video files, and 991 video files with various failed transcodes? BTW Special:TimedMediaHandler shows 354,846 failed transcodes, but it seems many of them are for audio files. Regards, Yann (talk) 09:30, 30 January 2017 (UTC)[reply]
@Yann: Special:Permalink/231572467Dispenser (talk) 15:14, 30 January 2017 (UTC)[reply]

Hi, It is useless to submit such files for transcoding (File:Awaara (1951).webm) as long as there is a bug blocking them. Regards, Yann (talk) 14:55, 6 February 2017 (UTC)[reply]

Figured I'd submit everything from the failed queue before somebody complains I didn't do enough. Dispenser (talk) 15:24, 6 February 2017 (UTC)[reply]

Facebook FBMD Reports

[edit]

Hi. You might be interested in generating reports for presence of pictures copied from facebook here on Commons. Seems that they have "Special instructions0000000FBMD\w*" in metadata (example). XXN, 20:51, 15 March 2017 (UTC)[reply]

@XXN: Interesting. Not much is know about FBMD, but is a suspected tracking ID. It looks like User:Gunnex has already generated some reports. But his SQL could be improved:
/* Images uploaded from Facebook by Newbies in last year
 * 
 * Runtime: 15 minutes
 */
USE commonswiki_p;
SELECT
  CONCAT("[[:File:", REPLACE(img_name,"_"," "), "]]") AS Filename,
  DATE_FORMAT(img_timestamp, "%Y-%m-%d") AS Uploaded,
  CONCAT("[[User:", img_user_text, "]] (", user_editcount, " edits",
    ", reg: ", DATE_FORMAT(user_registration, "%Y-%m-%d"), ")"
  ) AS User_info
FROM image
JOIN user ON user_id=img_user
JOIN page ON page_namespace=6 AND page_title=img_name
LEFT JOIN categorylinks ON cl_from=page_id AND cl_to IN (
  /* Exclude these categories */
  "Flickr_images_reviewed_by_FlickreviewR"
)
WHERE img_metadata LIKE "%SpecialInstructions\";a:1:{i:0;s:__:\"FBMD%"
AND cl_from IS NULL
  /* New users only */
AND user_editcount < 100
AND user_registration >= DATE_FORMAT(img_timestamp - INTERVAL 14 DAY, "%Y%m%d%H%i%S")
  /* Upload cut off date */
AND img_timestamp >= DATE_FORMAT(NOW() - INTERVAL 12 MONTH, "%Y%m%d%H%i%S");
In total, we have 57,275 (including images from Bureau of Land Management) and of those 15,397 were uploaded by newbies in the last 12 months. Keep in mind, one in eleven images is missing img_metadata. —Dispenser (talk) 22:51, 15 March 2017 (UTC)[reply]
Thanks. This should be publicized more in community probably, to discuss and find common issues and sollutions. XXN, 01:50, 16 March 2017 (UTC)[reply]

IRC #wikimedia-commons

[edit]

Hi, could you avoid running recent change bots on #wikimedia-commons? The channel is for discussion, and though manually driven bot queries are sometimes useful, it's not normally the right place for bots which may report several times per hour, as that could be more frequent than user discussion in a day. You may wish to think about promoting a recent changes channel.

If you think I'm getting this wrong, I suggest assessing consensus for it in-channel first, or running a poll for a few days about it. Thanks -- (talk) 14:46, 14 May 2017 (UTC)[reply]

@: The idea is to quickly squash the remaining Wikipedia Zero copyright violators in a final blow after decimating their RAR operation. I expected originally 4 messages/day, but I see it hit 20 today. I'm sorry about that and will only message #wikimedia-commons-admin now. I'm trialing the bot for 2 weeks (until May 19) then checking the data to see what effect it had. —Dispenser (talk) 20:31, 14 May 2017 (UTC)[reply]
Thanks for the explanation. I'm not against a RC style channel for highlighting odd account behaviour, and for new automated reporting, it would be very safe to create a temporary channel, then it could be discussed and merged into an existing one. The license review channel is a good example of a devoted channel for this sort of thing. -- (talk) 21:20, 14 May 2017 (UTC)[reply]

User notifications

[edit]

Please notify users about problems with their files. Help:Gadget-UserMessages is very helpful for this purpose. --EugeneZelenko (talk) 14:37, 18 May 2017 (UTC)[reply]

@EugeneZelenko: Thank you for the information, but these are WP0 abusers who make disposable accounts for pirating music and movies on Facebook Zero. The other admins will eventually get around to block the accounts. Dispenser (talk) 15:07, 18 May 2017 (UTC)[reply]

Gallica

[edit]

Hi, Could you make a list of all images from Gallica which are not in high resolution? i.e. adding Category:Images from Gallica with HR at source when they have gallica.bnf.fr or with {{Gallica}} in the source field and less than 4 Mpx. Thanks in advance, Yann (talk) 20:19, 17 August 2017 (UTC)[reply]

Special:Permalink/256563123 Done Just under 40% of files. I've included the query for generating the results. —Dispenser (talk) 03:20, 28 August 2017 (UTC)[reply]

Flagging new users to upload audio and video

[edit]

Hi Dispenser, could you flag Agarwaledx for upload? They're uploading videos of their courses. Thanks! --SJ+ 22:12, 30 September 2017 (UTC)[reply]

Resolved User:Guanaco had already flagged the user before I saw the message. Dispenser (talk) 14:50, 4 October 2017 (UTC)[reply]

A barnstar for you!

[edit]
The Technical Barnstar
The Technical Barnstar is awarded to anyone who has enhanced Wikimedia Commons through their technical work (programming, bot building, link repair, etc.).

Thanks for your tireless technical contributions regarding the wp zero issues & copyvio/spam upload prevention. Steinsplitter (talk) 19:05, 12 October 2017 (UTC)[reply]

Thank you. The Commons community has been very supportive, but the problem has grown outside it and at this point it more frustration than its worth. Brainstorming tasks that are not acted upon, writing documentation that isn't consulted, creating code that is ignored. If the returns aren't there, then my time is better spent where I can make a difference. —Dispenser (talk) 03:51, 13 October 2017 (UTC)[reply]

Files with dual extensions...

[edit]

Can I prompt you to run an update for this? User:Dispenser/Double extension

ShakespeareFan00 (talk) 09:26, 1 November 2017 (UTC)[reply]

Done 7,784 files The database server died last night breaking my automated update script, so I took the liberty of updating it with new extensions (.mp3, .opus, .webp) and several other changes. Dispenser (talk) 17:32, 1 November 2017 (UTC)[reply]

You're a life saver!

[edit]

Hi Dispenser, Thanks so much for the ping here, I actually had no idea why I was pinged and having searched my name and clicked on a file I've come across a whole bunch of images that I'd forgot where even uploaded!, So thank you so much! :), –Davey2010Talk 12:19, 1 January 2019 (UTC)[reply]

A kitten for you!

[edit]

Interesting analysis pages

Davidjessop (talk) 00:22, 24 February 2019 (UTC)[reply]

Hi Dispenser,

As you raised the issue of bitmaps embedded in SVG before, I just thought you might like to know that there are relief maps which are almost completely embedded bitmaps, e.g. File:USA_Hawaii_relief_location_map.svg. I added your template but was reverted with "there is no other way to have a relief map as a svg".

Cheers,
cmɢʟee ⋅τaʟκ 12:54, 11 July 2019 (UTC)[reply]

Weird summary

[edit]

Hi, This summary is weird. Two of these accounts are blocked, globally locked, and banned, and Dennis isn't an admin anymore. Any idea? Regards, Yann (talk) 04:42, 2 September 2019 (UTC)[reply]

Fixed I accidentally ran the script twice (first time bugged out, but apparently saved anyway). The query only looked at logs older than the start of the current month instead of only during the previous month, but why would somebody reupload a badly named file? [1] [2] [3] Dispenser (talk) 17:59, 15 September 2019 (UTC)[reply]

Hey Dispenser,

Can you run User:Dispenser/Double extension again? I know you normally do it monthly but I just fixed about 2,000 yesterday but didn't track which ones I fixed, so now I've blurred in fixed and non-fixed ones. ~riley (talk) 22:35, 1 November 2019 (UTC)[reply]

Hello. Is it possible for you to split User:Dispenser/Double extension? It's too big to actually navigate it in comfort. Probably sort it into subpages (months?) would help. --Minoraxtalk 01:20, 9 January 2021 (UTC)[reply]

Quarry / Missing and wrongly categorized Animated GIF files

[edit]

Hi Dispenser! I ran your SQL script on Quarry and moved a ton of files into the category of animated gif. Now I got a complaint about files which are not animated, but show up in the list created by the script. Here's one example: File:Isaac of Nineveh Mystic Treatises 001.gif. Do you have time to take peek? Cheers!! C(_) --Hedwig in Washington (mail?) 00:37, 9 December 2019 (UTC)[reply]

enwiki tools broken

[edit]

Hi Dispenser,

I see you are at least mechanically active here but not as recently on English Wikipedia. For the past several weeks, bug-reports have been piling up regarding dabsolver's URLs not working. Could you take a look at, for example, en:User talk:Dispenser#info.tm? DMacks (talk) 16:27, 12 July 2020 (UTC)[reply]

Template:Copy vio has been listed at Commons:Deletion requests so that the community can discuss whether it should be kept or not. We would appreciate it if you could go to voice your opinion about this at its entry.

If you created this template, please note that the fact that it has been proposed for deletion does not necessarily mean that we do not value your kind contribution. It simply means that one person believes that there is some specific problem with it.

Please remember to respond to and – if appropriate – contradict the arguments supporting deletion. Arguments which focus on the nominator will not affect the result of the nomination. Thank you!

King of ♥ 02:22, 23 July 2021 (UTC)[reply]


Please not that this refers to the redirect, not the template. Sinigh (talk) 12:13, 16 August 2024 (UTC)[reply]