Graphic Parsing

Please post all questions and comments regarding Help & Manual 7 here.

Moderators: Alexander Halser, Tim Green

Post Reply
Tonea Morrow
Posts: 54
Joined: Tue Oct 31, 2017 3:20 pm

Graphic Parsing

Unread post by Tonea Morrow »

For those of you who don't know, the United States Access Board (USAB) has update the 508 standards (handicap access) and they go into effect January 18, 2018.

I have to comply with those standards. So, my quick jump graphic of the home screen of my software must have a parallel, non-visual presentation (36 CFR Part 1194, Appendix C, 301.1 Vision). :gottit: Back when I coded web pages in the '90s, I could break a large image apart, put tool tips on each element, and then reassemble the image as a table that was invisible to the viewer. Thus a mouse over a button would allow the button to have a tooltip saying what this button was while still allowing the button to function as hyperlink. I'm thinking that is what I need to do. [scratching my head] I'm just not sure that is the best route. (of course, I could just be dreading the task :roll: )

I have to be able to provide the same manual in three formats: electronic/interactive from within the software (they have yet to tell me if that is chm, html, or exe), PDF interactive/color, and PDF printout/BW.

Any guidance?

Tonea Morrow
Dave Gehman
Posts: 575
Joined: Sat Sep 23, 2017 9:05 pm

Re: Graphic Parsing

Unread post by Dave Gehman »

What is a parallel non-visual presentation? Computer-generated speech?
Tonea Morrow
Posts: 54
Joined: Tue Oct 31, 2017 3:20 pm

Re: Graphic Parsing

Unread post by Tonea Morrow »

Yep, pretty much. The electronic version doesn't have to provide speech (though it should), but it does have to support text readers that read the page to the visually impaired. So, any graphics have to be "readable" by the reader software. This is done through tool tips. The reader software reads the tooltips as it reads the rest of the text on the page. At least, that is my understanding.

A home screen has a lot of elements on the page. I can provide a link to a text description of what a viewer would see (don't see how that would work for a PDF), or, I can make a really long tool tip that provides the same description (poor choice). Or, I can break up the graphic into parts (parsing) and provide tool tips for each part so that the reader software can read the parts as it moves across the graphical area. At least, that is my understanding. If anyone sees the solution differently, please speak up.

Right now, I have broken the grahic into columns. Each column is in a cell of a table with zero padding. Then the collumns have been spliced into lines (non-table rows). Menu on the left-hand side is the cell on the left-hand side. Main viewing area...well, it was divided into 3 columns, but now that I think about it, it should be one cell divided into lines of pictures. Assuming the reader reads one cell at a time, the main area should be read as a whole, not as three columns. :roll: Great, I've got to rework what I did Friday. (sigh)

The final product has to be readable as both an electronic page and a PDF page viewed electronicly--and a PDF printout. I've got 25+ years of publishing experience, but that is all in printed media.

Guidance is sought on: pitfalls, things to keep in mind, lack of understanding, lack of knowledge in how H&M works, lack of knowledge in how the output will work, etc.

Tonea Morrow
Tonea Morrow
Posts: 54
Joined: Tue Oct 31, 2017 3:20 pm

Re: Graphic Parsing

Unread post by Tonea Morrow »

I forgot to mention that all of this has to be scalable. So, when the user resizes the window to resize the graphics, they have to scale together so the non-visually impaired person doesn't see any discrepancies. I love the "Display Size: % of page width, maximum is physical size."

Scaling graphics and text allows a moderaterly visually-impaired person to enlarge everything to viewable size. Also, it has to be scalable because some visual handicaps require the opposite--shrinking to make more area visible.

Whatever I do has to support scaling.

Tonea Morrow
Dave Gehman
Posts: 575
Joined: Sat Sep 23, 2017 9:05 pm

Re: Graphic Parsing

Unread post by Dave Gehman »

Does the group or groups behind the standards have any examples? If they have annotated examples, there might be some guidance around how all this can be done.

Your requirements make me wonder if this was legislation done without input from computer experts vis-à-vis feasibility...

In pursuit of simplicity, could your home/cover page be tiled, rather than being a single image that needs to have tooltips on segmented areas?

And: would it be possible to have two projects, one for sighted & moderately impaired, and one for blind / greatly impaired? It might be possible to make the latter be all text.
Tonea Morrow
Posts: 54
Joined: Tue Oct 31, 2017 3:20 pm

Re: Graphic Parsing

Unread post by Tonea Morrow »

Does the group or groups behind the standards have any examples? If they have annotated examples, there might be some guidance around how all this can be done.
Good point. I will look at what they provided and see if I can find any inspiration. (duh! to me) After reading the whole CFR, I didn't look at the references because I felt I needed to get some work done. But you are right, I should now look.
Your requirements make me wonder if this was legislation done without input from computer experts vis-à-vis feasibility...
The whole CFR (see original post) is about 90 pages of description on how these changes came about. How they sought input over years of time and who responded, etc. The actual codes that replace the old codes are Appendices A-C.
  • Appendix A is chapters 1&2 for 508 update (software)
    Appedix B is chapters 1&2 for 255 update (hardware)
    Appendix C is chapters 3-7 for both 508 and 255 update
    Appendix D is a relisting of the previous requirements for those who will be grandfathered (don't have to update)
In pursuit of simplicity, could your home/cover page be tiled, rather than being a single image that needs to have tooltips on segmented areas?
Uhm, what is tiled? I'm not following.

Tonea Morrow
Dave Gehman
Posts: 575
Joined: Sat Sep 23, 2017 9:05 pm

Re: Graphic Parsing

Unread post by Dave Gehman »

Tonea Morrow wrote: Uhm, what is tiled? I'm not following.
Something like the original Windows 10 tiled interface. It was met with resistance and/or indifference (largely because Redmond wanted us to throw away our monitors and buy touchscreens), but it's still there.

Image
Tonea Morrow
Posts: 54
Joined: Tue Oct 31, 2017 3:20 pm

Re: Graphic Parsing

Unread post by Tonea Morrow »

Hmm, well if I was doing software it might be possible. But I am doing the documentation on the software--the help system for it. My style may be dated. I'll think about tiling. Right now, though, I don't see it being helpful. The idea of replacing a traditional TOC with a tiled TOC is interesting, but not an option for this project. The goal with the graphical TOC is to ...actually, I think I will spend some time looking at your idea. If I strip out the white space, I'm just left with the side menue and all the buttons on the main area of the home screen. I can see how your idea would be cleaner than my idea.

I have that same graphic at the start of almost every topic. At one time, many years ago, I taught high school math. You always start with something people know (the home screen) and move them along the new topic (help on a particular area of the home screen). Given the new access rules (the CFR 1194 appendices) include support for cognitive limits (simple approaches), this is a good strategy for the help files. So, currently that home-screen graphic is a snippet. Once I made it friendly to the visually impaired, it would not only be in the visual TOC page, but also updated throughout the snippet's locations. I hope my description makes sense. Regardless of how I do the visual TOC, I still need to make the graphic "visible" to the blind for all the other places I use it.

Tonea Morrow
User avatar
Tim Green
Site Admin
Posts: 23143
Joined: Mon Jun 24, 2002 9:11 am
Location: Bruehl, Germany
Contact:

Re: Graphic Parsing

Unread post by Tim Green »

Hi Tonea,

It's important to remember that every graphic can only have one tooltip. You can add additional tooltips with hotspot links, but they currently only have the tip in a title= attribute, not in an alt= attribute (let us know if screen readers need that as well and we'll add it). However, if you add hotspot tooltips the hotspots are not really part of the the image. They are in a special "image map" that positions their clickable areas over the image. If you are going to use hotspots to implement multiple tooltips in a single graphic it would be important to test it out with current screen readers first to see how they handle them -- they would need to read the image maps in the "correct" order and respond accordingly.

On image sizing: For people with some residual vision it might be helpful to not limit the expansion of the image to its original size. In some cases it might be better for them to have a much larger image, even if it is a little "jaggy".
Regards,
Tim (EC Software Documentation & User Support)

Private support:
Please do not email or PM me with private support requests -- post to the forum directly.
Tonea Morrow
Posts: 54
Joined: Tue Oct 31, 2017 3:20 pm

Re: Graphic Parsing

Unread post by Tonea Morrow »

It's important to remember that every graphic can only have one tooltip.
I'm with you.
You can add additional tooltips with hotspot links, but they currently only have the tip in a title= attribute, not in an alt= attribute (let us know if screen readers need that as well and we'll add it).
Yeah, lost me. I have a fuzzy idea what you mean, and did think along those lines, but couldn't see how that would be done in H&M. As for what screeen readers need, I don't know. I don't have one. (Sigh) I'm just going off what I remember from when a friend of mine was helping to make the University of Oklahoma web site readable back in the 90's. I should do some web browsing and see what I can find out.
However, if you add hotspot tooltips the hotspots are not really part of the the image. They are in a special "image map" that positions their clickable areas over the image. If you are going to use hotspots to implement multiple tooltips in a single graphic it would be important to test it out with current screen readers first to see how they handle them -- they would need to read the image maps in the "correct" order and respond accordingly.
I remember and agree completely. I really wish I had access to software for testing that.

If I broke the image into two columns (one the menu and the other the viewable area), then I think it will read the map in the correct order. I wouldn't have to parse the graphic into individual pieces, just into the two columns with the multiple links on a single graphic within each. The menu would appear as a list and be ordered automatically. The viewable area doesn't need to be in a particular order, it just needs to make sense to the user where to click. I don't know how a reader software does that.
On image sizing: For people with some residual vision it might be helpful to not limit the expansion of the image to its original size. In some cases it might be better for them to have a much larger image, even if it is a little "jaggy".
Yes, I have made a formal arguement to requisition image editing software that can upscale digital images. I currently have none. If I did have it, I could upscale the screen shots, sharpen them to reduce bluring, and hopefully use them for enlargements and for laser prints (which need 300-600 dpi instead of 96 dpi). The governement is not fast, so I can't plan on it being available for this project.

I really am finding all this commentary helpful. Thanks!

To review:
1> check the CFR 1194 references and see if it is helpful
2> check the web for information on how the readers work and "best practices"
3> consider html attributes (which I'm hoping you'll clarify)

Tonea Morrow
Dave Gehman
Posts: 575
Joined: Sat Sep 23, 2017 9:05 pm

Re: Graphic Parsing

Unread post by Dave Gehman »

FYI - while waiting for the government - Fotografix (free) has very fine image resizing capabilities. By Madhavan Lakshminarayanan, https://lmadhavan.com/fotografix/. Here's an unedited example of a 96 dpi screen 3x original:

Image
User avatar
Tim Green
Site Admin
Posts: 23143
Joined: Mon Jun 24, 2002 9:11 am
Location: Bruehl, Germany
Contact:

Re: Graphic Parsing

Unread post by Tim Green »

Hi Tonea,

Sorry for not being more specific about the hotspots. These are clickable areas that you can add to your images with the hotspot editor, which is accessed from the dialog displayed when you double-click on an image in the HM editor ([Hotspots...] button on the lower right). See this chapter in the help for details:

https://helpandmanual.com/help/index.ht ... tspots.htm

On images broken into columns: to do that you would need to put the images in a table, This will work OK on desktop browsers, but tables tend to break the layout on mobile browsers.
Regards,
Tim (EC Software Documentation & User Support)

Private support:
Please do not email or PM me with private support requests -- post to the forum directly.
User avatar
Martin Wynne
Posts: 2656
Joined: Mon May 12, 2003 3:21 pm
Location: West of the Severn, UK

Re: Graphic Parsing

Unread post by Martin Wynne »

Tim Green wrote:On image sizing: For people with some residual vision it might be helpful to not limit the expansion of the image to its original size. In some cases it might be better for them to have a much larger image, even if it is a little "jaggy".
Hi Tim,

They can do that by zooming the browser, even for images set to max-width.

If you don't set a max-width, users on widescreen monitors see a much enlarged fuzzy image for no benefit.

regards,

Martin.
Post Reply