Template talk:Multiple image/Archive 1

Template not working
The box width parameter (It neither controls the width of the borders nor works as a global image width command) is not working:

You may need separate parameters for these issues.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTD) 14:20, 15 April 2008 (UTC)


 * width is not for the width of the box, but for the images. If you omit width1, width2, etc, width will be used. Since you have provided width1 = 180, the box will be sized according to that. Is there any reason you want to have a different width for the box? eDenE  14:32, 15 April 2008 (UTC)


 * If I omit the 180s, the box does not work in vertical alignment.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTD) 21:01, 15 April 2008 (UTC)


 * In fact the following code does not work either:


 * Ahh... you actually need to remove the entire line. Sorry for the lack of clarification as well as non-intuitive template usage. My question is, what is the expected result for you, if both width and width1 are given? I'll update as soon as possible based on your preference. eDenE  21:12, 15 April 2008 (UTC)

I don't know general programming conventions about whether global parameters override local ones or vice versa. However, I would think it would be easier to work if the global command controlled.--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTD) 22:08, 16 April 2008 (UTC)


 * Usually local ones would override global variables, but Wiki's not for programmer ;) I updated so that width will overrides the others. Also not removing width= will not produce any error. eDenE  23:03, 16 April 2008 (UTC)

Alginment
I would like to post my request here for anyone who may see this template. It would be helpful if the global image size controlled width in vertical alignment and height in horizontal alignment. Is this possible?--TonyTheTiger (t/c/bio/WP:CHICAGO/WP:LOTD) 14:25, 15 April 2008 (UTC)


 * Panorama simple does that— you have to enter the height and width of the original image and the height of the desired image:
 * px
 * --— Gadget850 (Ed)  talk  -  13:27, 17 April 2008 (UTC)


 * That's far too complicated a solution (in fact, I don't even understand what you are saying), and it's not quite what I would like to see. I am having the same problem as Tony with a set of images as well: Sympathomimetic drug. I don't want to resize any of the images, I just want to have the images align on the bottom instead of from the top. This way the caption texts line up nicely instead of looking staggerred as they are now. I envision this happenning with a verticalalign parameter or something similar. Chaldor (talk) 20:19, 9 August 2008 (UTC)

image height
Why not have a parameter for specifying image height, rather than inserting a cumbersome formula to calculate the width from the desired height?--Taylornate (talk) 18:45, 4 August 2011 (UTC)
 * Because the MediaWiki image syntax is based around image widths. I wanted to move the calculation inside the template, and so allow the use of a universal height parameter. Unfortunately this still needs the editor to provide the original dimensions of the image, because apparently there's mo way of obtaining this in software. -- Red rose64 (talk) 19:06, 4 August 2011 (UTC)

If that is true then why am I able to specify height when placing individual images?--Taylornate (talk) 19:42, 4 August 2011 (UTC)
 * On the basic image syntax like  that code is sent straight to the MediaWiki parser which knows how to obtain the image aspect ratio in order to draw a suitably-sized box around the image. But where we want two or more images inside a single box, which is the main function of, something needs to work out the overall width of the enclosing box. So, we need to give the individual image widths, not just to know how wide to draw each individual image, but also so that frames are drawn around the individual images, and thirdly so that the outer border has the correct dimensions. The frame calculation comes down to:
 * 2 + w
 * where w is the width of that particular image; and the outer border calculation comes down to:
 * w1 + w2 + 4 * (n - 2) + 16
 * where n is the number of images (for example 2); w1 and w2 are the individual image widths. Obtaining w1 and w2 from one single height just isn't possible. -- Red rose64 (talk) 20:15, 4 August 2011 (UTC)
 * BTW this has previously been covered at above. -- Red rose64 (talk) 20:17, 4 August 2011 (UTC)

I submitted a feature request --Taylornate (talk) 05:07, 9 August 2011 (UTC)


 * Someone replied to my feature request pointing me to this extension that has a function to return image height. If we can get this extension approved for inclusion in a release of MediaWiki or installed on Wikimedia then we can do this.  Thoughts?--Taylornate (talk) 23:08, 9 September 2011 (UTC)
 * Ah, you mean . That would certainly be just what is needed; it definitely isn't installed at the moment, because   should return 297, but instead returns . -- Red rose64 (talk) 18:21, 10 September 2011 (UTC)
 * Any idea what it might take to get it included?--Taylornate (talk) 22:53, 10 September 2011 (UTC)
 * No, so I've raised a thread at WP:VPT. -- Red rose64 (talk) 23:56, 10 September 2011 (UTC)

Tool to support Horizontal placement: matching image heights
For anyone interested, a web page which does the required calculations is at http://www.pxc.me.uk/multipleimage/. (An admin could add this information to the documentation, if thought appropriate.) Peter coxhead (talk) 12:22, 18 February 2012 (UTC)


 * Having got fed up with calculating, even with a tool, I've now created Auto images which automatically resizes a set of horizontal images. Peter coxhead (talk) 21:07, 21 February 2012 (UTC)

Would appreciate some advice
Hi, I'm wondering if someone here could help me with the horizontal band of images across the top of Portal:Animal rights. They are different sizes and so look pretty untidy. I'm trying to follow the instructions here for horizontal images with different aspect ratios, but I can't get it to work -- the images are reproducing in their full size when I use that template, so I'm obviously doing something wrong.

Would someone be willing to do one image for me, so that I can see how to do the rest? SlimVirgin (talk) 03:17, 29 May 2012 (UTC)
 * I switched the image sizes to use a fixed image height, rather than a fixed image width. I hope this is what you wanted.  Thanks! Plastikspork ―Œ (talk)  03:28, 29 May 2012 (UTC)
 * That's brilliant and exactly what I wanted. Thank you! SlimVirgin (talk) 04:03, 29 May 2012 (UTC)
 * I devised the technique described at Template:Multiple image because doesn't allow image dimensions to be specified in any form that is not a simple width. This trick is specifically for the  template, which you didn't use at Portal:Animal rights. Where images are included using the Extended image syntax (as at Portal:Animal rights), it's possible to specify a height directly, which is what Plastikspork did. -- Red rose64 (talk) 07:07, 29 May 2012 (UTC)
 * Hi Redrose, I did try to use the template (I didn't save, I just used preview), but the first image I tried it with was reproducing at full size, so I had to abandon it. Thanks for writing that template, by the way, it looks very useful. SlimVirgin (talk) 21:56, 29 May 2012 (UTC)
 * OK, if you want to use it in future, here it is:


 * This code replaces everything from through to, inclusive of those two tags. It looks almost the same, except that the margin around the outside is slightly broader (the margin between the images is the same), and there is a faint grey border (the border color is ). To suppress that border, amend #E3DAC9 to #E3DAC9; border: none; -- Red rose64 (talk) 13:13, 30 May 2012 (UTC)


 * It's really kind of you to do this, Redrose, thank you. SlimVirgin (talk) 16:39, 30 May 2012 (UTC)

Alternate approach?
Everyone wants this feature (it is necessary to make horizontal layouts look good without effort), and various approaches have been suggested (such as external tools, alternate templates that still require inputing the raw image dimensions, even the idea of changing the underlying wiki engine to expose more data for us to calculate from). All of these suggestions presume we need to calculate the ultimate image widths ourselves, and then pass it upstream. But the standard functions for displaying an image do already have the capacity to accept only the desired height (as exemplified in the image at right). I'm not convinced we ever need to figure out what width the images will be (and hence I don't think the template needs to know the original dimensions nor aspect ratio either).

Why don't we just change this template to avail ourselves of this existing option? To wit, all we need is for the template to prepend an "x" character under appropriate circumstances.. Cesiumfrog (talk) 02:46, 21 August 2012 (UTC)
 * Unfortunately, no, the thumb layout is a div, and therefore needs an explicit width, otherwise all thumbs are as wide as the caption that is beneath it. —Th e DJ (talk • contribs) 08:44, 21 August 2012 (UTC)

This picture is really tall and narrow.
 * I'm not certain of the details, but it sounds like precisely the kind of thing that html etc is capable of. Cesiumfrog (talk) 12:11, 21 August 2012 (UTC)
 * Unfortunately, display:table does not work on IE6 and IE7. —Th e DJ (talk • contribs) 14:31, 21 August 2012 (UTC)
 * I'm sure I've explained this before; an examination of this page will show that I'm one of the more prolific help-assistants. But I'm not going to fix this one: in eight edits between 00:50, 21 August 2012‎ and 06:07, 21 August 2012‎, the page has changed enormously - sections have been moved about and renamed, and in six of the eight edits there was an associated non-trivial increase in page size which suggests that a new posting has been made at the same time. I could revert the lot as a violation of WP:TPO, but I won't; I therefore decline to respond to any postings made between 00:50 and 06:07, because I can't work out what's new, what's been amended, what's been removed, and what's unchanged. I've chosen this section to post in because it's the one containing the most recent edit: TheDJ has the sense to not rearrange the page every post. -- Red rose64 (talk) 14:59, 21 August 2012 (UTC)


 * DJ, I'm no CSS expert, but during my brief googling I encountered 2-3 alternative techniques that are supposedly able to give the same effect. Can't help wondering simpler might be a plain html table..


 * Redrose, apologies for beleaguering you, but I think some refactoring was warranted since a small number of topics kept spawning duplicate threads (possibly exacerbated by headings insufficiently descriptive for newcomers), which was making it difficult to assess which are the primary concerns among editors and how much is already known about addressing them.Cesiumfrog (talk) 01:39, 22 August 2012 (UTC)
 * Cesiumfrog, I'm not sure when you qualify to be a CSS expert, but I have been involved with CSS and JS on Wikipedia for over 5 years and am intimately aware about the many browser compatibility issues that exist in this area. If you want to fix the template, it has a sandbox and testcases, that you can validate against all the major browsers. If I knew of a way to make it work, I would, but I don't. You claim to have found three ways by googling, so I invite you to try and validate them. —Th e DJ (talk • contribs) 07:30, 22 August 2012 (UTC)


 * Ok, what about this one then? (Also, zooming out on firefox seems fine.) Cesiumfrog (talk) 10:49, 22 August 2012 (UTC)
 * This approach has two large down sides, 1 it uses tables, which we try to avoid (though mbox'es and infoboxes use them due to similar sizing issues, so the door for exceptions is open.). 2 and more importantly, it is significantly different from the 'standard' thumb HTML structure, which is the layout we try to emulate. Using this HTML makes it impossible to directly reuse the core styling classes for thumbnails. —Th e DJ (talk • contribs) 11:43, 22 August 2012 (UTC)
 * Now? Cesiumfrog (talk) 12:37, 22 August 2012 (UTC)

2015: modern browsers and fluid design
Soo... it's 2015 now, IE6 and IE7 should not be a concern anymore, and fluid/responsive layouts are all the hype. With that in mind, have any previously 'unacceptable' solutions become 'acceptable', so that we can finally add height parameters here? In the worst case, we could even use a conditional clause that renders the template completely differently when the height parameter is used, if that's what it takes. Sygmoral (talk) 22:25, 28 April 2015 (UTC)

"thumb"
As I posted at Template talk:Double image, it would be nice to be able to use "thumb" or "frameless" as the image width parameter so that user default preferences could take over. Any help on this? Cheers, Rkitko (talk) 01:16, 24 November 2008 (UTC)

Bad behavior for small widths
The main box is wider than it should be. --NE2 02:15, 7 January 2009 (UTC)
 * This is caused by a rule in MediaWiki:Common.css, after discussion at [#.thumbinner_min-width WP:VPT] in October 2007. I suggest you bring it up at MediaWiki talk:Common.css —Ms2ger (talk) 09:21, 20 June 2009 (UTC)

Added parameter?
Hello. Some weeks ago I created some portings from this template. But now the translated versions have an added parameter, which makes possible to define the background for the little space around the images. It works in 5 versions of the template: de, dsb, eo, hsb, sk. Greetings --Tlustulimu (talk) 18:32, 27 May 2009 (UTC)
 * Hello. I just added the new parameter to the Russian template, also: ru:Шаблон:Multiple image. It works very fine there. Greetings --Tlustulimu (talk) 18:52, 28 May 2009 (UTC)
 * Hello. I just added the new parameter to the template and also added an example to its documentation. Greetings --Tlustulimu (talk) 18:08, 21 June 2009 (UTC)

ALT text
If the captions are left empty, on mouseover, the tooltip yields "". Could we have an ALT field or that if captions are empty, the mouseover tooltip yields the filename instead? Ref: Battle of Bosworth Field Jappalang (talk) 15:56, 11 June 2009 (UTC)
 * I second that, could it please be fixed soon?--Adi (talk) 19:30, 19 June 2009 (UTC)
 * Just found out that we can add alt text to images by using ! as shown in Image; however, the tooltip yielding " issue remains. Jappalang (talk) 01:27, 9 July 2009 (UTC)
 * Okay, removing s resolves the issue. Perhaps box_caption is a feature to be implemented, but right now it is causing accessibility issues, so I took the bold step of removing it.  Jappalang (talk) 01:52, 9 July 2009 (UTC)

Merging this template with Template:Image frame
I suggest perhaps merging Template:Image frame, a similar but not as well documented template, into this one. My discussion is at Template_talk:Image_frame; please participate if you can, thanks. Gary King ( talk ) 06:39, 20 September 2009 (UTC)


 * With less than a hundred transclusions of that template, I dare say we'd be as easy just migrating the transclusions here and then deleting it. Chris Cunningham (not at work) - talk 11:37, 15 October 2009 (UTC)

More parameters please
This display could be improved very easily with two more parameters:

1. Option of alignment of images by height and not width.

2. Provision for a wider margin to be set between images. This is particularly key for the display of art works, which should not be jammed up against each other.

If this complicates the current template, it would be very helpful if an alternative template please be provided with the above parameters available. Thanks.

 Ty  07:41, 20 September 2009 (UTC)


 * The reason width is required, is because the thumbframe requires a known width. So if you specify height, then there is no way to calculate the final width of the thumb. —Th e DJ (talk • contribs) 12:16, 20 September 2009 (UTC)


 * i want the wider margin option too! it's needed!--Aaa3-other | Talk | Contribs 15:31, 9 January 2012 (UTC)

How to get it to work on my wikimedia page?
is there any docs on how to get this template to work on my wikimeida page? http://tourismatiguas.com/index.php/Template:Doc —Preceding unsigned comment added by 190.212.200.197 (talk) 21:37, 3 February 2010 (UTC)
 * You don't seem to have the parserfunctions extension enabled. —Th e DJ (talk • contribs) 22:34, 3 February 2010 (UTC)

Formatting problem
Why does the last image fall to the next line? This also happens when there are 5 images (see Correctional Service of Canada). -- P 1 9 9 • TALK 15:04, 3 March 2010 (UTC)
 * It doesn't, what browser are you using ? —Th e DJ (talk • contribs) 17:16, 3 March 2010 (UTC)
 * IE6. -- P 1 9 9 • TALK 17:40, 3 March 2010 (UTC)
 * Maybe you have a really small resolution so your browser is pushing the last image to the next line? It looks fine for me, too. Gary King  ( talk ) 17:50, 3 March 2010 (UTC)
 * Likely this is the Internet_Explorer_box_model_bug. —Th e DJ (talk • contribs) 18:40, 3 March 2010 (UTC)

It shows fine in the Firefox browser... -- P 1 9 9 • TALK 01:15, 4 March 2010 (UTC)
 * In Firefox 3.5.9 in default resolution it looks fine, but if I zoom out 1 or 2 levels then the last picture is pushed. -- Egmontaz  ♤  talk  17:58, 22 May 2010 (UTC)
 * Using Firefox as well. Anything less than 100% causes the last image to get pushed. SharkD   Talk  19:35, 2 July 2010 (UTC)


 * I can't confirm this. What version of Firefox? Got a screenshot? Chris Cunningham (not at work) - talk 11:21, 6 July 2010 (UTC)
 * Firefox 3.6.6 at least; start at zoom level 0 (Ctrl+0 or Alt,V,Z,R) and the above group of 4 looks OK. Then try Ctrl+- (or alt,V,Z,O) and it changes to a row of 3 and a second row of 1. -- Red rose64 (talk) 11:41, 6 July 2010 (UTC)


 * Ah, sorry, out. I misread. Yep, verified. Hmmm: did this always happen, or is it new? Chris Cunningham (not at work) - talk 13:10, 6 July 2010 (UTC)


 * It also happens for me, even with two images (such as the ones in Picture_tutorial, which is really confusing), zooming out just once from the default level with Firefox 3.6.8 on Ubuntu 10.4. I never browse at the default zoom level, with any browser on any machine.  Please make this work ... Rp (talk) 12:37, 30 July 2010 (UTC)

Breaking
I run Firefox on a small 10.1" display, and the template always breaks into two columns on it, making vertical and double image perferred. Can this be fixed? Res Mar 03:23, 11 October 2010 (UTC)
 * I trust you mean, other than getting a larger display ? Are you saying that the space is there and that it just isn't used ? I have no such issues using Webkit btw. I just tested the documentation page. Are you changing default textsize perhaps ? Because that can have undesired effects at time icw this template. —Th e DJ (talk • contribs) 12:34, 11 October 2010 (UTC)

Horizontal alignment not working properly in Firefox 6 beta
According to a friend, the following code on the Seacology article is not displaying correctly in his copy of Firefox 6 beta:

He sent me a screenshot taken of Firefox in Windows displaying the box with the images stacked vertically and a large whitespace (approximately 250 width) to the right of the images. The caption displayed normally at the correct width (had the image been aligned horizontally). If anyone needs to see what I'm talking about, email me through Wiki and I will forward the screenshot. I just wanted people to be aware of this if reports start popping up... –  VisionHolder « talk » 22:44, 17 August 2011 (UTC)
 * Aha! This sounds like above. Does it sort itself if the zoom level is changed? In FF under Windows, the keys are Ctrl + to zoom in, Ctrl - to zoom out, and Ctrl 0 to return to default zoom level. -- Red rose64 (talk) 23:39, 17 August 2011 (UTC)


 * This now looks fine on my Mac in Firefox 6.0 (not beta); possibly the bug has been fixed in the release version of Firefox 6. Ucucha (talk) 14:56, 18 August 2011 (UTC)


 * My friend tried your suggestion (about returning to the default zoom level), and that seems to fix the problem. –  VisionHolder « talk » 18:16, 18 August 2011 (UTC)

Broken
Doesn't do horiz properly any-more. Rich Farmbrough, 23:51, 29 January 2012 (UTC).


 * Example? -- Red rose64 (talk) 23:53, 29 January 2012 (UTC)


 * Ok, here's an example from the template's documentation. I added horizontal even though the default is supposed to be horizontal. Also see §Horizontal placement: matching image heights in the documentation—shouldn't the images there be side-by-side rather than one above the other?


 * —Trappist the monk (talk) 14:10, 20 August 2012 (UTC)
 * the images are side-by-side in my browser. Frietjes (talk) 14:49, 20 August 2012 (UTC)


 * Chrome (21.0.1180.79) has yellow above red. Very old versions of MSIE and Opera show yellow left and red right. So, it appears that there is an issue there.


 * —Trappist the monk (talk) 15:08, 20 August 2012 (UTC)
 * Chrome 21.0.1180.79 has yellow to the left of red, as do Firefox 14.0.1, Internet Explorer 7.0.5730.13, Opera 12.01, and Safari 5.1.7 (7534.57.2). However, all these are at normal zoom  - if I zoom out using, Chrome and Firefox show yellow above red at one or more levels of zoom, as does Safari at two or more levels. This is the problem described above at . IE7 and Opera are the only ones to show them yellow to the left of red at all zoom levels, although IE zooms out by simply reducing the rendered page, pulling the right margin to the left: it doesn't take advantage of the "increased" horizontal space. -- Red rose64 (talk) 18:19, 20 August 2012 (UTC)


 * Yep, that's it. I assumed that WP had recently made a significant change to the CSS—main headings were no longer bold, smaller text in infoboxes, much smaller citation  text in reference sections, yet main article text wasn't that different.  Never occurred to me that I had somehow set my browser to a different zoom level.  Thanks.


 * —Trappist the monk (talk) 19:12, 20 August 2012 (UTC)

Solution?
This issue seems to be documented in such places as. So far I've not looked enough to tell whether technically (prescriptivism) the browser or the html source is more at fault, but either way it does seem likely that it should be resolvable by alteration to the source (hence the template). Cesiumfrog (talk) 00:50, 21 August 2012 (UTC)
 * In fact, as far as I can tell the examples at Template talk:Coin image box 2 singles are immune to the problem, which again suggests solution is possible. It might be worth addressing at the same time the feature request for control over borders and of the gap spacing between images. Cesiumfrog (talk) 06:07, 21 August 2012 (UTC)
 * This is a FF issue. There is more about it here and testcases here. Setting box-sizing does indeed make the problem more rare, though that CSS property reportedly has some issues with min and max width and height according to quirksmode.org. The template has a sandbox you know, go right ahead :D —Th e DJ (talk • contribs) 15:51, 21 August 2012 (UTC)
 * You'd have to find a proper solution for IE6/7 however. boxsizing compatiblity matrix —Th e DJ (talk • contribs) 15:55, 21 August 2012 (UTC)

PDF
In pdf version the formating seems broken. Can this be fixed? -- Egmontaz  ♤  talk  18:00, 22 May 2010 (UTC)
 * These are known limitations of the PDF renderer. —Th e DJ (talk • contribs) 19:05, 22 May 2010 (UTC)

Seven images
Would it be possible to make this template work for seven images? Or, if it is undesirable for this template to do so, can a special template which holds up to seven images be created? Thanks for any help. ÷seresin 00:25, 7 June 2010 (UTC)
 * Nevermind. Jack Merridew helped me out and I got what I needed. ÷seresin 08:23, 7 June 2010 (UTC)

Stack vertically and place side-by-side?
Is there a way with this template to, say, stack two (e.g. smaller) images on top of each other, and a third (e.g. taller) image to the side? Are there any other templates with which to achieve this? SharkD  Talk  07:27, 24 June 2010 (UTC)
 * Not with this template as it stands; and I don't know of existing templates that might do it. If you need to do this soon, it might be easier to construct a table; note the use of "rowspan=2" on the upper-left cell:


 * -- Red rose64 (talk) 12:20, 24 June 2010 (UTC)
 * See Double image stack as listed in the See also section of this template. —Th e DJ (talk • contribs) 19:25, 24 June 2010 (UTC)
 * That stacks two vertically; but how does it also provide for a third to be placed at the side? -- Red rose64 (talk) 20:47, 24 June 2010 (UTC)
 * Using a table would work, but would mean having to do without the normal styling associated with image containers. SharkD   Talk  01:47, 30 June 2010 (UTC)

Frameless?
Is it not possible to set image sizes to ? PC78 (talk) 12:23, 15 August 2010 (UTC)


 * Heh. I've actually been working on this in the sandbox without seeing this note; try . Check the sandbox code and the test cases, which work well enough for me: it needs stress-tested before being deployed though. Chris Cunningham (user:thumperward: not at work) - talk 10:43, 18 August 2010 (UTC)
 * Any updates on this? d'oh! talk 09:11, 29 August 2010 (UTC)

ALT not working properly
As seen here, even if both alt parameters set, actually only the caption of second image displays as alt2 and alt1 does not display at all (at least in Firefox 3.6, Win XP). --Kozuch (talk) 08:13, 21 August 2010 (UTC)
 * Aren't you confusing title elements of links with alt text for images ? —Th e DJ (talk • contribs) 12:52, 21 August 2010 (UTC)
 * I have examined the generated HTML (Ctrl+U in Firefox): both images have alt text - it's in the  attribute of both   elements:
 * However, in Firefox, the lower image has a tooltip when you hover the mouse pointer over it; the upper does not. This is because the tooltip does not come from the alt text, but from the individual image captions, which are placed in the  attribute of the   elements:
 * Only the second of these has the  attribute. Examination of the wikicode shows that caption2 has been filled in, but caption1 is blank. If the caption is intended to apply to both images, it should be placed in footer instead. If you want both images to have tooltips, fill in caption1. -- Red rose64 (talk) 15:15, 21 August 2010 (UTC)
 * I've observed a similar issue with this edit. Each image has a different alt text and caption. With the pointer hovering over each image, I'm seeing the respective caption text rather than the alt text. This needs verifying with a variety of browsers. --trevj (talk) 11:34, 15 February 2011 (UTC)
 * Only the second of these has the  attribute. Examination of the wikicode shows that caption2 has been filled in, but caption1 is blank. If the caption is intended to apply to both images, it should be placed in footer instead. If you want both images to have tooltips, fill in caption1. -- Red rose64 (talk) 15:15, 21 August 2010 (UTC)
 * I've observed a similar issue with this edit. Each image has a different alt text and caption. With the pointer hovering over each image, I'm seeing the respective caption text rather than the alt text. This needs verifying with a variety of browsers. --trevj (talk) 11:34, 15 February 2011 (UTC)
 * Only the second of these has the  attribute. Examination of the wikicode shows that caption2 has been filled in, but caption1 is blank. If the caption is intended to apply to both images, it should be placed in footer instead. If you want both images to have tooltips, fill in caption1. -- Red rose64 (talk) 15:15, 21 August 2010 (UTC)
 * I've observed a similar issue with this edit. Each image has a different alt text and caption. With the pointer hovering over each image, I'm seeing the respective caption text rather than the alt text. This needs verifying with a variety of browsers. --trevj (talk) 11:34, 15 February 2011 (UTC)


 * The phenomenon is not specific to, but occurs for any non-thumbnail image. Consider the image at right, which is constructed like this:
 * If you hover the mouse over that image, you may see a tooltip, which will be either "This tooltip is taken from the image caption" or "This tooltip is from the image Alt text". My own observations show that Firefox 3.6.13, Google Chrome 9.0.597.98 and Opera 11.01 all show the caption text as a tooltip when hovering the mouse, whereas Internet Explorer 7.0.5730.13 shows the alt text as a tooltip.
 * It should be noted that the primary purpose of the alt text is not to set the tooltip, but to serve as a textual equivalent to the image, for situations when the image cannot be displayed. These include text-only browsers; users who have disabled the image display in their browsers (for speed, perhaps); slow connections causing image download to time-out (this actually happened to me several times last week when they were trying to roll out MediaWiki 1.17); and not forgetting screenreader software for the visually impaired. See also WP:ALT. -- Red rose64 (talk) 15:44, 15 February 2011 (UTC)
 * Still going on, as seen at Bhutan. I understand what Red is saying, but it's still silly. They should force the alt text to display where the code is just reading the captions at the moment. — LlywelynII  00:08, 29 September 2011 (UTC)
 * This is a feature that is shared by all images on Wikipedia which have both caption and alt text, and is not confined to those displayed by . If you need to get image handling altered for Wikipedia, it's the MediaWiki devs who handle it. I suggest asking at WP:VPT but they're likely to direct you to bugzilla. As far as the inconsistency between Internet Explorer (tooltip displays the alt text) and the other four (tooltip displays the caption) is concerned, we have absolutely no control over this, and you would need to take it up with the browser's own support departments. -- Red rose64 (talk) 13:56, 29 September 2011 (UTC)
 * Thanks for clearing this up, Redrose. I too was confused by the intent of the tooltip; as long as the accessiblility options (pictures off or screenreader in use) work as intended, I'm good. FYI: I'm running IE8, and it shows (almost) similar behavior to Firefox: the tooltip displays caption instead of alt, unless alt is absent, in which case the tooltip is also absent. JustinTime55 (talk) 20:58, 6 December 2011 (UTC)
 * Thanks for clearing this up, Redrose. I too was confused by the intent of the tooltip; as long as the accessiblility options (pictures off or screenreader in use) work as intended, I'm good. FYI: I'm running IE8, and it shows (almost) similar behavior to Firefox: the tooltip displays caption instead of alt, unless alt is absent, in which case the tooltip is also absent. JustinTime55 (talk) 20:58, 6 December 2011 (UTC)

Fix for alternate text (Edit request on 12 August 2012)

Mistake to defer the blame upstream.

 

Like you say for a single image, when we use the thumb type, everything works: we automatically get a rectangular border containing a text caption, and the alt text is used as the tooltip on the image. But in this multiple image template we do not use the thumb option, because we want the template to manually take care of the rectangle and captions itself; when the thumb type is omitted there is no automatic border and no automatic caption. So what happens if you still supply a caption argument when we are not using the thumb type? Commonly in that circumstance you would want the caption to take precedence (and override the alt text in sighted people's tooltips as is the already noted tendency), but that is redundant for us since the caption still is already being printed on the screen. Therefore, this template should not pass on a caption text when it instructs for an image to be displayed, and that way the alt text will be used for the tooltip. Above I've put a strike through the part of this template that needs to be removed (which should be repeated for all ten of the possible images described by this template).

How do we get this change implemented? Cesiumfrog (talk) 04:06, 12 August 2012 (UTC)


 * Diff (confirmed using sandbox and testcases). Fixing a long lamented bug, this will make the alternate-text parameter work the way it is supposed to, the way it already does for individual thumbnail figures. The big improvement for the encyclopedia is that this change allows many sighted-readers to check that alternate text for any image is appropriate, just by hovering over with the mouse (the same regardless of whether the image appears in a figure alone or among multiple others).


 * Cesiumfrog (talk) 06:36, 12 August 2012 (UTC)
 * Red information icon with gradient background.svg Not done: please establish a consensus for this alteration before using the template. I don't think that this is right. At Template:Multiple image/testcases, I see differences between browsers.
 * {|class=wikitable

!Browser !!Original Example !!Original Example 2 !!Sandbox Example !!Sandbox Example 2 Google Chrome 21 Opera 12 Safari 5 ||Example ||Example 2 || ||
 * Firefox 14
 * Firefox 14
 * Internet Explorer 7 ||Alt1 ||Alternate text 2 ||Alt1 ||Alternate text 2
 * }
 * blank cells indicate that no tooltip is displayed. An examination of the HTML shows that the four elements all bear an appropriate   or   attribute, but there is a change in the </a> elements which enclose these. For the two "Original" cases, the  tag has the attribute   or   as appropriate, but for the two Sandbox cases, there is no such attribute on the  tag.
 * My conclusion: if there is no difference in IE, but other browsers are worse off, this should not be done. -- Red rose64 (talk) 14:38, 12 August 2012 (UTC)
 * Would you mind checking again? Redrose, I think you made a mistake in your table: Test 1 (sandbox version) does work with firefox 14. Now it is true that test 2 behaves as you describe, but so it should (note its wiki code); remember, this is a long-standing bug in this template and so the test cases were always written not to test the alternate text behaviour. Thus, test 2 does not supply any alternate text, and so test two should not be expected to display any alternate text in a tooltip. Now, the buggy version of the template does still display the caption in the tooltip, but that is undesirable for two reasons: i) the caption is already displayed below the image and so the tooltip is not needed, and ii) it obscures a loss of accessibility (preventing the sighted reader such as yourself from quickly recognising when the alt-text is not merely redundant but actually missing).
 * Would you mind checking again? Redrose, I think you made a mistake in your table: Test 1 (sandbox version) does work with firefox 14. Now it is true that test 2 behaves as you describe, but so it should (note its wiki code); remember, this is a long-standing bug in this template and so the test cases were always written not to test the alternate text behaviour. Thus, test 2 does not supply any alternate text, and so test two should not be expected to display any alternate text in a tooltip. Now, the buggy version of the template does still display the caption in the tooltip, but that is undesirable for two reasons: i) the caption is already displayed below the image and so the tooltip is not needed, and ii) it obscures a loss of accessibility (preventing the sighted reader such as yourself from quickly recognising when the alt-text is not merely redundant but actually missing).


 * Examination of the html is revealing; it shows the mechanism upstream by which my change works. Yes, in all cases that an alternate text is supplied to the template, it is already given in the alt parameter of the image tag (which is why we understood it to already be working for screen readers). As you recognise, the difference is in the title parameter of the anchor tag. What ought be the appropriate use of the title parameter for a graphic linked to its file page? The proper purpose is to give any further new information about where the link will direct the user to. (If I were a prescriptivist, I might say that the "technically correct" use, in all single- and multi-image figures, would be text like "Image file page" since that adds information about where the link goes without duplicating whatever content is already provided.) Unfortunately the parameter is commonly misused (in a misguided attempt to game search rankings) by duplicating already-provided text; redundant duplication with the title parameter serves no accessibility benefit (moreover, screen readers hence tend to ignore title text entirely). Regardless, typical browsers display the title text as the tooltip. For example, if I give you a link to blocking improvements then the title parameter can be used to inform the name of the page which that link directs to.


 * A pragmatic convention has already been established in single-image figures (thumbnails), which is that the title parameter should be absent for images in figures. (Obviously, this avoids redundantly overriding the tooltip with the caption, and facilitates sighted readers to keep a check on the appropriateness of the alternate text, to great advantage.) For consistency we too should follow this convention, and so this change should be implemented. Cesiumfrog (talk) 01:23, 13 August 2012 (UTC)


 * Consensus, renewed every few months (e.g., Kozuch, trevj, Llywelynll above), is this change should be made. Even if the sole dissenting argument is valid, the only percieved risk is to an aspect of the status quo (caption-duplication) nobody ever supported anyway. So do we need to elect a voting committee now or something? Cesiumfrog (talk) 05:02, 13 August 2012 (UTC)

So summary.
 * 1) There is an inconsistency in the usecase of two images with captions. 1 can have a title attribute on the link mirroring the caption, the other doesn't.
 * 2) Some browsers use the alt attribute for the tooltip, other browsers the title attribute.
 * 3) Normally (MediaWiki default), when adding an image without thumb layout, the purpose of this title attribute is to provide the information that the 'caption' provides when adding a thumb to a page.
 * 4) Normally (MediaWiki default) on thumbnails with captions, we do not use a title attribute on the link of the image.
 * 5) Your change will remove the title attribute from the images, where appropriate, exactly as you have stated
 * 6) Redrose's argument is that he now 'has less tooltips on certain browsers'
 * 7) Your argument is 'that is exactly what it is supposed to do, because some of those tooltips were redundant and not following the conventions that we were trying to emulate with this template (the conventions of the MediaWiki thumbnail).
 * That looks like a proper rationale to me (see how a summary can be much more clarifying than pointing fingers ? ). ✅ —Th e DJ (talk • contribs) 12:41, 13 August 2012 (UTC)

Help desk
Folks, would someone take a look at this question from the helpdesk relating to the display of images at Entranceway at Main Street at Roycroft Boulevard? Thanks. – ukexpat (talk) 16:33, 9 September 2010 (UTC)
 * This edit apparently fixed that. I'm not sure but I'd guess it was caused by the align parameters with multiple values... --Waldir talk 17:44, 9 September 2010 (UTC)
 * Update: it was the empty  parameter. I tweaked the template to prevent this issue from happening in the future. --Waldir talk 17:59, 9 September 2010 (UTC)

Edit request from FleetCommand, 6 February 2011
Currently, when user specifies no width parameter, the template generates junk. I propose that in the event no width is specified, each image must have a default width of 200 pixels. (Default width of an image with "thumb" parameter is 200 pixels.)

Fleet Command (talk) 11:15, 6 February 2011 (UTC)


 * It's only 200 if the user doesn't have it overridden in their prefs. In any case, not specifying width is legal: it just means that width1, width2 etc. must be specified individually. -- Red rose64 (talk) 15:14, 6 February 2011 (UTC)


 * I'm tling the editprotected, per Redrose64 above. --Waldir talk 18:26, 6 February 2011 (UTC)


 * That is not a plausible reason for rejection; if we can prevent a template from generating junk (graceful error recovery) then we must do so, especially when doing so imposes no additional load on Wikipedia servers. Defaulting to a non-flexible value is orders of magnitude better than generating junk. Fleet Command (talk) 07:17, 3 June 2011 (UTC)
 * That sounds reasonable to me. But could you put a working version in the sandbox and then get consensus before reactivating the request? &mdash; Martin (MSGJ · talk) 13:44, 3 June 2011 (UTC)
 * Done! Test case is ready! Here it is. Check out test results. Fleet Command (talk) 15:23, 3 June 2011 (UTC)
 * And done. Thanks! Plastikspork <sub style="font-size: 60%">―Œ <sup style="margin-left:-3ex">(talk)  23:15, 4 June 2011 (UTC)
 * Thank you too. Fleet Command (talk) 17:28, 5 June 2011 (UTC)

Coin image box 2 singles
It seems like Template:Coin image box 2 singles could be replaced by this one if we add a "border" option. 198.102.153.2 (talk) 20:00, 2 June 2011 (UTC)
 * Good idea. Anyone want to take a stab at it?  I think it's just a border around the captions in the horizontal orientation.  Thanks! Plastikspork <sub style="font-size: 60%">―Œ <sup style="margin-left:-3ex">(talk)  23:15, 4 June 2011 (UTC)

Stop making it so difficult to import templates
Wth is it with wikipedia making billions of templates? Can't you just place it on one dam template so I can import it into wikis? I just want to set some cool parameters like background color and stuff without having to reference back to Help:Wiki Markup.

For freakings sake,  just make one template and list the parameters. 216.121.189.35 (talk) 20:59, 27 August 2011 (UTC)
 * isn't a template, it's a conditional expression and as such should be built in already. -- Red rose64 (talk) 21:05, 27 August 2011 (UTC)
 * What is this guy talking about? Fleet Command (talk) 02:27, 28 August 2011 (UTC)
 * He's complaining about Wikipedia making it slightly more difficult for him to build content farm copies. — LlywelynII  00:10, 29 September 2011 (UTC)
 * Actually that's not necessarily the case. When using a standard en:WP template on other WP projects or Wikia projects one has to recurse through layers of called templates. It is a royal pain, especially if you get a less well informed denizen deleting the templates. Rich Farmbrough, 23:50, 29 January 2012 (UTC).

Text wrap
Not available or just not mentioned in the documentation? — LlywelynII  00:10, 29 September 2011 (UTC)
 * Nevermind. It was just an effect of centering the image. — LlywelynII  00:11, 29 September 2011 (UTC)

Align = none
could we have an "align = none" option? currently, i can float an image either (a) left, (b) right, (c) center, or (d) none. however, this template does not support none, but would be useful to have this as a feature. until then, i have been wrapping it inside a table. thank you. Frietjes (talk) 20:05, 1 November 2011 (UTC)
 * Where would this be of use? -- Red rose64 (talk) 20:53, 1 November 2011 (UTC)
 * it would be of use everywhere that you don't want text to wrap around it, and you don't want it to float to the center, to allow other floating items to float next to it. thank you. Frietjes (talk) 15:57, 2 November 2011 (UTC)
 * it's a very minor change, please add "none" to the first switch statement, like this. thank you. Frietjes (talk) 16:02, 2 November 2011 (UTC)
 * I fail to see how that differs from center in its effect. Please demonstrate in Template:Multiple image/testcases that your change works as intended. -- Red rose64 (talk) 16:08, 2 November 2011 (UTC)
 * there is a second switch further down in the template, which sets the margins for center (and see the testcases. If you want another example of where this is being used, see Template:Location map+. Frietjes (talk) 16:23, 2 November 2011 (UTC)
 * Yes check.svg Done -- Red rose64 (talk) 17:23, 2 November 2011 (UTC)

doc
this template refers to perhaps its a good idea to change that to  --DerekvG (talk) 23:43, 14 November 2011 (UTC)
 * Why bother? It's WP:NOTBROKEN. -- Red rose64 (talk) 14:33, 15 November 2011 (UTC)

Box border
An additional parameter option to make the box border disappear would be nice in my opinion. A set of pictures squeezed into a box appears cumbersome to me, as it adds four grey lines to your article that are not required. Sometimes a box is below a header - more lines. JMK (talk) 22:51, 20 March 2012 (UTC)


 * We have a - that would need to be suppressed and a new class added. It's possible to override the styles that it sets by misusing the background color parameter. Try adding ;border:none; - note the semicolon before the word "border". -- Red rose64 (talk) 23:40, 20 March 2012 (UTC)
 * Excellent, works very well. Thanks! JMK (talk) 16:31, 2 April 2012 (UTC)

Custom placement
hello,

it seems like this template does not have the feature to place the pictures like I need it, but you can only choose between either horizontal or vertical placement. But I want it like that: 3 pictures in a row, and one picture below the second image. Regards.-- GoP <sub style="color:#8EE5EE;">T <sup style="color:#8EE5EE;">C <sub style="color:#8EE5EE;">N 16:39, 12 April 2012 (UTC)


 * Assuming that you have a very good reason for such a layout, I would try a table: two rows, three cells per row. Peter coxhead (talk) 17:14, 12 April 2012 (UTC)
 * We've had this requested before, see above. It's not very easy to amend this template to cover all potential layouts. -- Red rose64 (talk) 17:16, 12 April 2012 (UTC)


 * I saw it, but it is not the design I am looking for. Look at the right table. I want to add three more women with which Dostoyevsky had affairs; they should be below the three, with captions. -- GoP <sub style="color:#8EE5EE;">T <sup style="color:#8EE5EE;">C <sub style="color:#8EE5EE;">N 20:12, 12 April 2012 (UTC)
 * How about this? Not knowing who the other three are, I've repeated the first three. -- Red rose64 (talk) 20:50, 12 April 2012 (UTC)
 * Better, but there is too much space between the triplets.-- GoP <sub style="color:#8EE5EE;">T <sup style="color:#8EE5EE;">C <sub style="color:#8EE5EE;">N 09:13, 13 April 2012 (UTC)
 * I've amended the example at right to have zero padding and zero margin. It's certainly smaller in Firefox, but I wouldn't like to say what IE does - that thing is a law unto itself when it comes to table measurements. -- Red rose64 (talk) 16:58, 13 April 2012 (UTC)


 * Ah, when you originally wrote "one picture below the second image", I thought you meant 3 pictures in the first row and one only in the second row below the middle one. If you want three in each row, how about this, using Auto images rather than Multiple image? Peter coxhead (talk) 10:09, 13 April 2012 (UTC)
 * Not bad, but is it possible to remove the break line and the grey header?-- GoP <sub style="color:#8EE5EE;">T <sup style="color:#8EE5EE;">C <sub style="color:#8EE5EE;">N 13:24, 13 April 2012 (UTC)
 * Using this template, you can remove the header (as per the current version), but not the line between the two rows. Actually, I think there should be a header. Wikipedia isn't an image repository and there needs to be a good reason for a set of images; being able to provide a meaningful header is a test of whether the images are appropriate. Peter coxhead (talk) 14:10, 13 April 2012 (UTC)

Line break
Is it possible to force a line break after the template? I find that the next header or the following text creeps into the very small space between the template and a taxobox / infobox for instance. JMK (talk) 15:43, 16 May 2012 (UTC)
 * I assume that you're describing . The template is designed to be a floating object: thus, text will be shown alongside, just like it is with any ordinary image. To force text to appear below instead, put a  immediately after the . -- Red rose64 (talk) 17:08, 16 May 2012 (UTC)
 * or none. Frietjes (talk) 17:24, 16 May 2012 (UTC)
 * Both suggestions work. Thanks for your quick replies. JMK (talk) 13:10, 29 May 2012 (UTC)

The template is broken
Why is there spacing?  JC  &middot; &#32; Xbox  &middot; &#32; Talk &middot; &#32; Contributions 06:40, 10 November 2012 (UTC)


 * It was because you set the direction to "verticle" instead of vertical. ;-) Cesiumfrog (talk) 08:32, 10 November 2012 (UTC)

Requested move
<div class="boilerplate" style="background-color: #efe; margin: 2em 0 0 0; padding: 0 10px 0 10px; border: 1px dotted #aaa;">
 * The following discussion is an archived discussion of the proposal. Please do not modify it. Subsequent comments should be made in a new section on the talk page. No further edits should be made to this section. 

The result of the proposal was no consensus. --BDD (talk) 17:08, 6 February 2013 (UTC)

Template:Multiple image → Template:Multiple images – Why not fix the spelling mistake? theFace 19:41, 10 January 2013 (UTC)
 * Oppose what spelling mistake? This is a multiple image template so can be called Template: multiple image. double image isn't called double images ; triple image isn't called triple images -- 76.65.128.43 (talk) 23:40, 10 January 2013 (UTC)
 * Oppose. Renaming a template would confuse the issue unnecessarily. Andrew327 06:08, 11 January 2013 (UTC)
 * Support, "multiple images" in the plural sounds more natural to me. We're not talking about other templates here, but this one. Also, other templates with much higher usage have been renamed, so this shouldn't be a problem. --The Evil IP address (talk) 16:45, 11 January 2013 (UTC)
 * support, assuming we keep the redirect, and don't actively orphan the redirect. Frietjes (talk) 18:21, 11 January 2013 (UTC)
 * The above discussion is preserved as an archive of the proposal. Please do not modify it. Subsequent comments should be made in a new section on this talk page. No further edits should be made to this section.

Border color
Hi, I really like this template, nice job. I came here wondering if I could change the color of the borders around the images? I saw the "Frameless?" and "Box border" discussions above, and so was able to remove one border, but I'd still like to give the appearance that there is no border at all. The reason for this is the strange padding between the bottom of an image and the bottom of the border, which I thought was a browser issue, but isn't. This extra padding seems to be used to bump the captions down just a wee bit, which is great. So rather than removing it, I'd simply like to make it "disappear" with a border_color option. Thank you! – Kerαu noςco pia <sup style="color:#A60000;">◁ gala xies 19:42, 25 March 2013 (UTC)
 * As with above, the way to do this is to misuse the background color parameter. Try adding ;border-color:transparent; - note the semicolon before the word "border". That will only affect the outer border, not the ones around the individual images. -- Red rose64 (talk) 21:18, 25 March 2013 (UTC)


 * Hi, thanks for your help here and above. I guess I didn't make myself clear. Your suggestion above (";border:none;") worked fine for removing the outer border. Now I'm trying to tackle the inside borders around the individual images. Not a fan of the "look". – Kerαu noςco pia <sup style="color:#A60000;">◁ gala xies 01:17, 26 March 2013 (UTC)
 * It's built into the  class as   - for consistency with thumb images - and there is no means for overriding that within . -- Red rose64 (talk) 14:31, 26 March 2013 (UTC)


 * I wondered if it wasn't some built-in thing like that; I couldn't find anything about a border in the code, but I can't read code anyway (which is probably apparent since I couldn't even find anything about the outside border). I appreciate your help. I solved the issue by using this template to do the re-sizing math and page placement for me, then replaced the template with a simple table. – Kerαu noςco pia <sup style="color:#A60000;">◁ gala xies 16:59, 26 March 2013 (UTC)

"Location map many" inside "multiple image" causes error
What causes the File:etc. at top? I can't get it out. Benjamin Trovato (talk) 03:25, 18 May 2013 (UTC)


 * That's odd. Maybe it's because multiple-image is expecting a raw image file, but instead you're trying to insert a fully formed floating figure. You could try a hack of faking the multiple-image (by wrapping your maps in tables and appropriate div tags) but it still won't look right unless you can change the "Location" template so that it doesn't pre-wrap a border and caption around the graphic... How do you like this?Cesiumfrog (talk) 03:56, 18 May 2013 (UTC)
 * It's not odd at all. The template is designed to arrange images, and not box-shaped objects in general. Since it needs the various width values in order to calculate the dimensions of the enclosing boxes, it needs those values to be specified outside of the image markup. To avoid redundancy (and possible errors due to inconsistent values) the template assembles its own image markup using values passed in the various parameters. It therefore operates on bare image names, not images that have already been marked up (whether by  image markup or by another technique such as templates). Accordingly, the imagen parameters cannot contain any templates or other markup. See any of the examples, such as Template:Multiple image which uses 60 Yellow card.svg Yellow cartouche. When fully expanded, the template yields   Now imagine that instead of the image filename being specified as Yellow card.svg it was instead image1. Unless that template yielded an image filename and nothing else, you'll end up with one huge mess, as shown at the top of this thread. -- Red rose64 (talk) 09:34, 18 May 2013 (UTC)
 * Perhaps (as one of those priviledged elite not prevented from effecting improvements) you could comment further in a section above where a fix was suggested (to one of this template's most frequent inconveniences) which could also readily (say, by a notafile option) permit more general input (as I demonstrated by solving BT's problem here at right).Cesiumfrog (talk) 02:11, 19 May 2013 (UTC)
 * It would not be possible for this template to work if given only a height, as I explained at (probably elsewhere, since a lot of threads were chopped up and mixed around on 21 August 2012). -- Red rose64 (talk) 07:49, 19 May 2013 (UTC)


 * Why do you still say it's not possible, when I've already done it? Cesiumfrog (talk) 08:02, 19 May 2013 (UTC)
 * You can close this whenever you want. I don't much care, but I thought you code people would be interested. Benjamin Trovato (talk) 02:39, 23 May 2013 (UTC)

Zoom not working
I see this may be related to an above discussion. Zoom breaks the horizontal layout in FF (21.0), MSIE (10.0.9200.16466), Chrome (27.0.1453.110 m) and Safari (5.1.7), but not in Opera or Netscape Navigator. I used Ctrl+[-] on Glenn Robinson III after noticing a problem at Roy Lichtenstein.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 16:05, 13 June 2013 (UTC)--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 16:05, 13 June 2013 (UTC)
 * P.S. above only discusses FF as far as I can tell.--TonyTheTiger (T/C/BIO/WP:CHICAGO/WP:FOUR) 16:09, 13 June 2013 (UTC)
 * Yes, discusses IE6 and Firefox, because at that stage, nobody had complained about breakage in other browsers. But further down, at, I described the behaviour of five browsers. Continued inconsistency between browsers doesn't surprise me, nor does correct behaviour in Opera. -- Red rose64 (talk) 17:14, 13 June 2013 (UTC)

Not working in Chrome?
Someone replaced a couple of these templates with html galleries here with the comment "Multiple image doesn't work (in Chrome at least)." Maybe worth investigating? Kendall-K1 (talk) 16:54, 11 October 2013 (UTC)
 * I see two problems there - but neither is specific to Chrome, since Firefox behaves similarly. First, on my screen (1280 px wide) the total width of the four images is greater than the page as a whole. That is at normal zoom; the second problem comes in if zooming out one level, when the fourth image moves to a second row: this is the problem that we already know about. -- Red rose64 (talk) 17:46, 11 October 2013 (UTC)

shortcut
Can someone add to this page.--TonyTheTiger (T / C / WP:FOUR / WP:CHICAGO / WP:WAWARD) 19:58, 15 December 2013 (UTC)

Double image versus multiple image
Supposedly Template:double image is deprecated. Unfortunately the newTemplate:Multiple image is not supported across all languages of Wikipedia. Until it is I am not supportive of medical articles switching over to Multiple image as it will interfere with my work here WikiProject_Medicine/Translation_task_force. Is anyone supporting switching dealing with the issue of cross language compatability? Doc James (talk · contribs · email) (if I write on your page reply on mine) 14:44, 6 February 2014 (UTC)
 * You can still use and ; although these are now wrappers for, the old syntax still works. -- Red rose64 (talk) 15:27, 6 February 2014 (UTC)
 * Great. As long as people do not start coming around trying to change the markup from double image to multiple image. Doc James  (talk · contribs · email) (if I write on your page reply on mine) 15:46, 6 February 2014 (UTC)

LUA version
I started a LUA version as Module:Multiple image and currently in Template:Multiple image/sandbox2. it is nearly output identical to the template with a few exceptions: you can see any differences in the testcases. This does reduce the single image bug (see Test 10 and Test 11). The output is much closer to the output for standard thumbnail images. There are also a few additional features have been imported from Template:Auto images:
 * 1) I set the width in the thumbinner div, rather than the outer div (how it is done with standard thumbnail images)
 * 2) I used class=center for center alignment, instead of margin:0 auto (how it is done with standard thumbnail images)
 * 3) The overall width for N images is (width1 + width2 + ... + widthN) + 4 * (N - 1) + 12 for horizontal and max(width1, ..., widthN) + 12 for vertical, which means for a single image you should get the same overall width for both direction=vertical and direction=horizontal
 * 1) You can choose the alignment of the captions with caption_align
 * 2) If you set total_width and horizontal and heightX then it will rescale the widths so that (1) the overall width is   and (2) the image heights are all the same
 * 3) If you set total_width and horizontal and the   values are not specified, it will rescale the widths so that the overall width is
 * 4) If you set total_width and vertical then it will use   as the value for width

Note that by switching to LUA, there is no enforced limit on the number of images, and we can now consider other layouts as an alternative to horizontal and vertical (not currently implemented, but certainly possible). For example, we could have image1-1, image1-2, image2-1, image2-2 for a 2 by 2 array of images (again, not currently implemented, but certainly possible).

Comments, bugs, suggestions? Frietjes (talk) 20:10, 24 April 2014 (UTC)
 * Maybe it would be possible to add sometring like per row (like in Gallery)? -- Edgars2007  (talk/contribs) 17:54, 7 May 2014 (UTC)
 * good idea. note, even more fine-grained specification of images perrow was suggested at template talk:auto images. Frietjes (talk) 23:42, 7 May 2014 (UTC)

Vertical alignment of unequal-height images (take N+1)
The issue from 6 years ago is still present. The case is not that we can't figure out the aspect-ratio or native height in order to re-scale each one and match the heights of the images when displayed in a row. But instead, we want to have the images retain their different heights (each retained at the same scale, rather than each image scaled differently), while still getting all of them aligned by their bottom edges rather than tops. For technical diagrams, scaling some more than others over-emphasizes them in the display or makes it harder to compare to see small differences if the diagrams themselves change size. And it does look kinda crappy IMO to have the captions all over the place. What about passing a to the  container around each /? That way we can "push down" each image+caption by whatever amount necessary to get their bottoms where we want them. DMacks (talk) 17:45, 4 June 2014 (UTC)

Add an option to remove the background called "background"
Like template "file" or "image" that if he puts the "thumb" parameter shows no background, please. Reranian (talk) 03:00, 13 July 2014 (UTC)


 * I've sandboxed a possible way, see right. -- Red rose64 (talk) 12:00, 13 July 2014 (UTC)

Template-protected edit request on 15 November 2014 (template breaks mobile view)
Hello together,

please refer to this bug report on bugzilla. This template breaks the view of mobile pages, because captions of images are too long. The problem can be solved by change the "width" inline style property to be "max-width". I have no template edit right on enwiki, so i request to change this value :)

Florianschmidtwelzow (talk) 21:10, 15 November 2014 (UTC)
 * Implemented in the sandbox. All the best: Rich Farmbrough, 00:20, 16 November 2014 (UTC).


 * done. Frietjes (talk) 17:14, 16 November 2014 (UTC)
 * Thanks! --Florianschmidtwelzow (talk) 12:43, 17 November 2014 (UTC)
 * This is an incorrect fix, you are breaking older browsers that don't support max-width. I've made some further improvements in the Sandbox. We rly should finish that lua version of it... as you can see, this becomes butt ugly without variables. —Th e DJ (talk • contribs) 13:55, 17 November 2014 (UTC)
 * TheDJ and Florianschmidtwelzow, as far as I know, the lua version is ready. I added it to the testcases as sandbox2.  if no one sees any problems, I will update the code to use it instead. Frietjes (talk) 15:46, 17 November 2014 (UTC)
 * I added my changes to it, and I think we can use that version. I don't see any ugliness with it. —Th e DJ (talk • contribs) 17:26, 17 November 2014 (UTC)
 * now updated, hopefully there are no side effects. Frietjes (talk) 18:41, 17 November 2014 (UTC)
 * Thanks for this change :) It looks good in mobile and i haven't see any side effect for now. Thx
 * With flex box model, i now improved the mobile compatibility even further, with full responsiveness. You can find the code in Module:Multiple image/sandbox and the testcases show this version as Sandbox2 now. —Th e DJ (talk • contribs) 23:02, 17 November 2014 (UTC)

Dudemanfellabra, it would be good to figure out if there is a fix for both problems. can you confirm for me if the following appears as three images in a row, or one wrapped below: Adoxa (Adoxa moschatellina) if it reproduces the problem, can you tell me if increasing the 'max-width' value for the outer div helps? Frietjes (talk) 00:51, 19 November 2014 (UTC)
 * Please take notice that due to the Lua-isation of this template, I am no longer able to offer technical support. -- Red rose64 (talk) 20:56, 17 November 2014 (UTC)
 * Duly noted. Thank you for the incredibly amount of work you've been doing lately btw. —Th e DJ (talk • contribs) 23:02, 17 November 2014 (UTC)
 * This change has broken the layout on WP:NRHPPROGRESS. There are three images side-by-side on that page wrapped in a scrollable div. Now that this "max-width" parameter has been set, the third of these images is displayed below the other two. How can I make these three images display side-by-side again, even if they go off the side of the page (thus the scrollbar)?
 * Actually it's a more widespread issue; it even affects the examples in the documentation for me. It may be that I'm using an old browser, though. I'm still on OS X 10.5.8 Leopard, and the latest version of Firefox for my OS is 16.0.2.--Dudemanfellabra (talk) 23:47, 18 November 2014 (UTC)
 * This appears as one wrapped below. I increased the max-width of the outer div all the way to 700px, and nothing changed on preview.--Dudemanfellabra (talk) 01:28, 19 November 2014 (UTC)
 * However, if I change the width as well--just increasing it by 2px to 494px--everything renders correctly.--Dudemanfellabra (talk) 01:33, 19 November 2014 (UTC)
 * okay, so perhaps we just bump up the width by 2px and everyone will be happy, or will this cause uneven padding on each side? Frietjes (talk) 01:47, 19 November 2014 (UTC)

Adoxa (Adoxa moschatellina)
 * You tell me. It looks even on my machine, though there may be a 2px difference. 2px is so small I would think the difference would be almost indistinguishable.--Dudemanfellabra (talk) 01:51, 19 November 2014 (UTC)
 * That's uneven padding in other browsers. In my opinion, old browsers should be allowed to 'fail' as long as all the content stays visible (progressive enhancement over backwards compatibility). If someone really wants to try, you could probably attempt with box-sizing tricks, and of note is that my recent changes in the sandbox using the flex box model remove the need for this max-width parameter, so perhaps that will help with this case as well. —Th e DJ (talk • contribs) 09:39, 19 November 2014 (UTC)
 * I just installed FF 16 on my mac btw, and I don't see the problem... not with monobook, not with vector... Do you have any gadgets or user scripts installed perhaps ? —Th e DJ (talk • contribs) 10:00, 19 November 2014 (UTC)
 * This is apparently like several previous threads such as, many of which I commented on. These showed that the degree of severity is different for various browsers. Search for my comment of 09:43, 14 December 2013 (UTC); in that post I link to a change that I made nine days earlier that increased the width by 1px and which fixed the zoom problem for several browsers - or at least reduced it by a significant amount: in the eleven months since I made that change, the complaints ceased until this thread was raised. -- Red rose64 (talk) 15:02, 19 November 2014 (UTC)

, good point that there were no reported problems for 11 months. I believe, the following is equivalent to what was generated by that version: Adoxa (Adoxa moschatellina which appears to have the same wider format, and slightly uneven padding, as the version that works for Dudemanfellabra (although not as uneven). in other words, the difference is between 3*(162 + 2) = 492 for the inner div (which isn't working) and 494 for the inner div (which is working) and 501 for the outer div (which is working?)?  I recall something like 3px padding for thumbinner, which would mean the inner div should be 3*(162 + 2) + 6 = 498, or I am probably missing something? Frietjes (talk) 18:29, 19 November 2014 (UTC)
 * This latest version is more uneven than the second one, although both display with the three images in line horizontally as expected unlike the original one. If I could choose, the second setup is best for me. As far as the user scripts/styles, I have quite a few scripts installed but none that I would think affect this situation directly. I do have some CSS styles applied, the only one of which I would think affects this would be, which shrinks all the body content to 85% font size ever since Wikipedia changed its font a while back and made it larger than it was previously. I just removed that line from my CSS and checked, and nothing changed, so I restored it. I also discovered it's apparently not a browser issue because for me the first setup shows the third image below the first two even on my iPhone, which is required to run the latest version of mobile Safari. Furthermore, if I log out and view this page, everything looks fine on both devices. So it has to be a script or some other user setting, but I'm just not sure which. I'll do some investigation by sequentially removing scripts and figuring out the culprit. I have to work 2:30-6:30 (CST here), though, so I won't be able to do any of that until later tonight. I'll report back if I find out anything.--Dudemanfellabra (talk) 20:02, 19 November 2014 (UTC)
 * I've found the culprit! This problem was not due to any personal javascript or CSS; I blanked both my vector.js and vector.css to check that, and nothing changed. The only other option would then be a user setting/beta feature. I looked through everything I have set, and I found that the third image dropping down only occurs when the option "New image thumb design, and other related CSS tests (TOC, categories, etc.)" is set under "Testing and development" on the gadgets page. Unfortunately, I don't know where the discussion/code for this "new" design is since there isn't a link, so I don't know who to talk to about the issue. It would be helpful, though, if you or other editors could enable this gadget and see if it reproduces the error above. Like I said, this also affects the layout on my iPhone, which definitely doesn't have an old browser, so it's probably a widespread issue.--Dudemanfellabra (talk) 16:28, 21 November 2014 (UTC)
 * Dudemanfellabra, after turning on that beta feature, I see the problem as well. checking the CSS, it looks like this beta feature changes the padding defined in the thumb, thumbinner and thumbcaption classes, so it's not surprising that math for computing the width no longer works out.

Adoxa (Adoxa moschatellina)
 * for the beta-feature version, it looks like there is 0 padding on the individual images, but 4px on thumbinner, which would be (160*3)+4+4=488px. tested above. Frietjes (talk)
 * This last one looks the best with the gadget turned on; however, it looks uneven with the gadget turned off. Can you think of any way to make it look decent in both situations? To make the thumbinner div centered with respect to the container div in both situations?--Dudemanfellabra (talk) 17:44, 21 November 2014 (UTC)

Broken template(s)
(Copied from User talk:Redrose64)


 * Hi Redrose64,
 * I noticed a number of instances of what looks like broken templates on United Arab Emirates. I looked at the code, chased the problem to the template pages, then got stuck. To be honest, I'm having difficulty even describing the problem adequately, so I thought I'd take this snapshot of the page and show it to you, as I thought you'd be able to understand it. Any help or information would be appreciated. I'm on Windows 7, Internet Explorer 11. Regards,  nagual  design   20:11, 17 December 2014 (UTC)
 * You need to take it to the template's talk page. As I noted some weeks ago, somebody else has converted the template to Lua, and I can therefore no longer maintain it (which includes no longer being able to explain why it does what it does). -- Red rose64 (talk) 20:53, 17 December 2014 (UTC)
 * Ah.. Fair dinkum, mate. I had tried to wade through that talk page, but the more I read the more I realized I didn't even know how to characterize the problem. I'll copy this to there. Cheers.  nagual  design   21:19, 17 December 2014 (UTC)

(/copy)

Please note that this problem (or similar) also affects Template:Double image, as seen in the second example in that snapshot of United Arab Emirates captioned, "Left is the Dubai Marina and a oil well on the right. UAE is one of the world's largest exporters of oil."  nagual  design   22:26, 17 December 2014 (UTC)
 * this is the same problem as the last few comments in the above section: "New image thumb design, and other related CSS tests (TOC, categories, etc.)" is set enabled under "Testing and development" on the gadgets page. This gadget changes the styling of thumbnails in a way that does not match the expectations of this template. There is no good way around this, without degrading the quality for all the other users. That's part of the reason why that gadget is in the section "Testing and development". —Th e DJ (talk • contribs) 12:18, 18 December 2014 (UTC)

Photo position
In this article, the photos appear one above the other, not left and right as indicated by the caption. I enquired about this on the article's talk page. The article's lead author advised me to report the matter here. (I use Firefox with Windows 7). <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 14:25, 17 February 2015 (UTC)
 * Do you by any chance have the "New image thumb design" gadget enabled? One small bug is that it pushes the last image to the next row. I though I fixed that, but you may want try purging the page, or clearing your cahce.  14:32, 17 February 2015 (UTC)
 * I don't know what the "New image thumb design" gadget is. The purge description is somewhat confusing and looks quite tedious. Clearing the cache does not make any difference. For what it's worth, after switching off my computer earlier today and switching on again now, the photo appearance remains one above the other. <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 23:11, 17 February 2015 (UTC)
 * User:Axl, (1) click on the 'Preferences' link at the top of the page (between the Sandbox and Watchlist links), (2) click on the 'Gadgets' tab, (3) scroll to the bottom to find "New image thumb design, and other related CSS tests (TOC, categories, etc.)" in the "Testing and development" section, (4) make sure this option is not selected. if this doesn't fix the problem, please let us know your operating system and web browser. Frietjes (talk) 01:24, 18 February 2015 (UTC)
 * There is a "Testing and development" section, but there is no entry for "New image thumb design, and other related CSS tests (TOC, categories, etc.)". (There is an entry for "Mobile sidebar preview - Show page in mobile view while browsing the desktop site.") <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 02:14, 18 February 2015 (UTC)
 * I think you're using Monobook then.  09:01, 18 February 2015 (UTC)
 * Yes, at, the "Testing and development" section has only one entry in all skins except Vector, where it has three. See Special:Gadgets, where two entries under "Testing and development" show "Available on the Vector skin." -- Red rose64 (talk) 13:27, 18 February 2015 (UTC)
 * one thing to check would be if the problem persists when you are logged out. Frietjes (talk) 16:51, 1 March 2015 (UTC)
 * Sorry for the delayed response. (After a week without further response, I stopped checking regularly.) The problem is not present when I am logged out. <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 14:49, 13 March 2015 (UTC)
 * Axl, okay, so can you confirm that are using monobook (it will be in the appearance section of your preferences). and which browser/operating system? Frietjes (talk) 15:09, 13 March 2015 (UTC)
 * Yes, MonoBook. I use Firefox on Windows 7 at home. However the problem also occurs when I use a PC at work with Internet Explorer. This seems to imply that the problem is with my personal settings rather than the browser/OS. <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 17:21, 13 March 2015 (UTC)
 * Axl, I tried switching to monobook and have been unable to reproduce any bugs in either Windows 7 or Linux. you could always try the 'Restore all default settings (in all sections)' button in your preferences, assuming that you can keep track of any important preferences you may have set.  Frietjes (talk) 19:00, 17 March 2015 (UTC)
 * I am not keen on resetting my preferences. I use a green-on-black colour scheme, and I wonder if that might be the cause.
 * Dodoïste, what do you think? <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 14:22, 21 March 2015 (UTC)
 * Hi User:Axl ! :-)
 * The problem comes from the white on black skin. It adds a 1px green border around images. Since it widens the image in the template Multiple image, it causes the second image to move under the first image.
 * If you want to I can solve this problem easily. I only need to change a small line of code in the green on black skin.
 * However this template creates several accessibility issues. I'm going to comment on that in another section. Dodoïste (talk) 10:12, 24 March 2015 (UTC)

Orphaned subtemplate
This template's conversion to Lua resulted in its subtemplate, /numImgs, being almost entirely orphaned (it's still transcluded from a few sandboxes and a couple talk pages). I'm not sure what's normally done about these templates, which is why I'm posting here instead of handling it myself. Cheers! 「 ディノ 奴 千？！ 」? · ☎ Dinoguy1000 04:47, 3 March 2015 (UTC)
 * the sandbox still uses it, which I believe we are (were) keeping around for comparison with the lua version. Frietjes (talk) 14:22, 3 March 2015 (UTC)
 * Putting Deprecated template in a noinclude block is one of the better ways in my opinion to deal with stuff like this. —Th e DJ (talk • contribs) 15:06, 3 March 2015 (UTC)

Support for perrow
I modified the version in module:multiple image/sandbox to support perrow. the syntax is 2 for two images per row, or 2 / 1 / 3 for two in the first row, 1 in the second row, and 3 in the third row. the module will adjust the value passed to perrow if it doesn't make sense. for example, if there are only 3 images, but you use 2 / 1 / 3 it will use 2 / 1 instead. or, if you say 2 / 1 / 2 and there are 9 images, it would use 2 / 1 / 2 / 2 / 2, repeating the last number until it reaches the total number of images. you can use any non-digit as the delimiter, so 2, 1, 2 works as well. the code is still a bit messy, so I may try to clean it up a bit more. please let me know if you have any comments or suggestions or see any bugs or other issues. you can see some of the results in the testcases (scroll down to around test 12), and look at the results from sandbox2. thank you. Frietjes (talk) 16:15, 4 March 2015 (UTC)


 * This seems to work well, as far as I've checked it. I like the ability to be flexible about the number of images per row, as in the old Auto images template using cont, which was always a bit of a "workaround".  It certainly means that this template is not replaced by the new facilities added to gallery tag, which although useful, are less flexible. Excellent work. Peter coxhead (talk) 10:24, 5 March 2015 (UTC)
 * How about small screens ? —Th e DJ (talk • contribs) 10:56, 5 March 2015 (UTC)
 * User:TheDJ, this template never really worked the same in mobile view, as you can see in [//en.m.wikipedia.org/wiki/Template:Multiple_image/testcases the mobile view version of the testcases]. it would be nice if the gallery tag syntax were expanded enough to completely cover all the formatting provided by this template.  the fact that gallery now takes style and class means we are getting closer.  someone should try to make a version of this template that is a frontend for <gallery ></gallery>.  for vertical, you would just use perrow=1, and for horizontal, you could use packed.  however, you probably wouldn't be able to get the image borders (yet)? Frietjes (talk) 15:07, 5 March 2015 (UTC)
 * Well that's why I did the proposal to use flex box —Th e DJ (talk • contribs) 11:46, 31 March 2015 (UTC)
 * I have updated the module to enable the perrow parameter as described above, now in use in Kyrias family. Frietjes (talk) 00:00, 15 May 2015 (UTC)

Accessibility issues
Hi. This template creates several accessibility issues (see WP:ACCESS for general reference).
 * 1) When several images are presented and there is only one caption (regardless of alt text). After the first image, a non-sighted user might be confused to not find the caption. And to find a caption after the second image that still applies to the first. It seems obvious to sighted users but I believe this is not universal. Bad example : Abortion. Good example : Abraham Lincoln.
 * 2) The caption is often contains reference to the visual position of images on the screen. Example : "the image on the left...". Blind users cannot use this information. Bad example : Abortion and Wordless novel.

See for reference the WCAG 2.0 guideline : F14: Failure of Success Criterion 1.3.3 due to identifying content only by its shape or location.

Thanks, Dodoïste (talk) 10:27, 24 March 2015 (UTC)


 * I leave it to you to decide on the best course of action to make the images as clear as possible to as many people as possible. Thank you. <b style="color:#808000">Axl</b> ¤ <small style="color:#808000">[Talk] 16:31, 24 March 2015 (UTC)


 * The problem isn't with the template, though, but with how it's used by editors, which can be poor with single images or galleries. Peter coxhead (talk) 16:34, 24 March 2015 (UTC)
 * The template allows editors to create one caption for several images, isn't it ? Then can we consider that the template encourages users to create accessibility issues ?
 * I would like to add that this template is included more than 14'000 times. It's very difficult to rewrite that many captions manually. The only option I can envision is to change the template. Dodoïste (talk) 22:24, 25 March 2015 (UTC)
 * Actually, strictly speaking, it doesn't – each image has its own caption. But it does allow a single footer, and if this is used as a kind of global caption without taking care over accessibility, then I can see that there could be an issue. This should be explained in the documentation; in particular a footer should not substitute for captions and alt texts. Peter coxhead (talk) 22:38, 25 March 2015 (UTC)
 * Explaining in the documentation is not enough. From my experience, most Wikipedians use templates by looking at how they are used in one article, and they copy-paste that. You have to fix every bad use of the template to solve a problem, otherwise it will come back. Dodoïste (talk) 21:02, 31 March 2015 (UTC)
 * Also something like: "the image on the left..." is not wrong. You have two images, and the whole point of these two images is (usually) to make a side by side comparison. It is ok, to tell a screenreader user all this information. Actually, I think it might be better. The assumption that the spatial information is 'incorrect' information for a user because he cannot see is a flawed argument, because even if he cannot visually experience it, he can have a 'mental model' of this spatial information (this is why voiceover on an iPhone screen works). In this case, he can have that model, because the screenreader will actually tell him that there are two images preceding this block of text (image, image, text). You could argue that saying: "The first image" is better than saying "left", but that is probably more 'correct' than a screenreader user would even expect. Be careful not to translate the 'generic' WCAG criteria to all usecases that u feed it, that is not how you are supposed to use WCAG (although it is how many 'review' companies use the criteria). The intent and the scope of the advise are very important in WCAG 2.0. —Th e DJ (talk • contribs) 11:43, 31 March 2015 (UTC)
 * This is precisely the time when we would need an expert. When we disagree on the interpretation of the WCAG. We are both volunteers without expertise on accessibility after all. I'm not convinced by your reasoning, I need more proof and evidence. And I do not feel confident enough to convince you.
 * At some point it always comes to this. The WMF should hire an accessibility expert. Dodoïste (talk) 21:02, 31 March 2015 (UTC)
 * I'm not saying that some of those bad cases that you point out are correct. I'm just saying that breaking functionality that people use is not a good idea and not required because you can simply rewrite it in a prose style that would not have these problems. In cases like this 'hiring an expert' isn't going to work. This is one of our biggest challenges, this is content... The only way we are ever going to get all content accessible, is by making every single content contributor an expert on accessibility. This is not gonna happen. That's why our content will never ever be able to pass a WCAG test. The criteria for WCAG are too high, and the methodology to author accessible content is too complicated and requires too much human judgement and consideration to fit the 'anyone can edit'-mantra. We can only strive to make it as good as possible, and cleanup after those who do no understand. What will help here, is converting the thumbnail to make use of new HTML5 semantic elements like: and (something that is being worked on right now and already done for VE). Once the screenreaders catch up... —Th e DJ (talk • contribs) 22:45, 31 March 2015 (UTC)