Catch up on stories from the past week (and beyond) at the Slashdot story archive

 


Forgot your password?
Close
typodupeerror
×
The Internet

Cynthia Says... Create Accessible Web Sites 35

Posted by michael from the or-else dept.
Kynn writes "The folks at ICDRI, in partnership with the Internet society and HiSoftware, bring us Cynthia Says, a free service to help you evaluate your Web pages for accessibility. In other words, it's roughly equivalent to what Bobby used to be, before it went commercial. It features what seems to be a cartoon version of my friend Cynthia Waddell, which is a bit creepy, but in all honesty it's a much better symbol than the old cartoon cop used with Bobby. I always thought there was an implied menace, as if the smiling chap would happily bludgeon you with his truncheon if you created an inaccessible Web site." If only.
This discussion has been archived. No new comments can be posted.

Cynthia Says... Create Accessible Web Sites More

Cynthia Says... Create Accessible Web Sites

Comments Filter:

  • Yet another Web Accessibility article (Score:4, Insightful)

    by BornInASmallTown ( 235371 ) on Friday March 14, 2003 @05:00PM (#5514596)
    I really hate to sound insensitive on this topic, but I'm tired of reading articles about web accessibility when it almost always boils down to one thing:

    Don't forget to use ALT tags!

    Ok, ok, so there's more to it than that. However, in my designs, I've begun to apply the following rule of thumb in regards to web accessibility:

    The page is accessible if it can be properly viewed and navigated using a text-based browser (i.e. Lynx).

    Lynx forces the page creator to use ALT tags liberally, and it reduces or eliminates the page's dependency on things like Javascript and Flash.

    What else, really, has to be considered outside of the limitations of a text-based browser? I'd love to read some comments from folks with more expertise in this area.

    • Blind users or users with very limited seeing. That is also what the ALT attribute (there is no such thing as an ALT tag) is used for, to provide information about the image to the screen reader they are using. So, yes, the ALT attribute is used for things other than text browsers.

      For example, why is Flash so bad for the web? Simple: say you have a blind user. How on earth are they supposed to navigate a Flash site when there are no ALT attributes to guide them and their screen readers can't "read" a .sw

      • For example, why is Flash so bad for the web? Simple: say you have a blind user. How on earth are they supposed to navigate a Flash site when there are no ALT attributes to guide them and their screen readers can't "read" a .swf file.

        Spot on as far as I am concerned. At the moment there is far to much emphasis on looks rather than useability, which is more important? Useability. By far, after all your olds looked after you why shoulded we look after them, and everybody elses'!!

        Nuf said and there is mu

      • How on earth are they supposed to navigate a Flash site when there are no ALT attributes to guide them and their screen readers can't "read" a .swf file.

        The correct way of embedding a Flash presentation into an HTML document is to use the <object> element. Alternative representations of the embedded object should be encoded as the contents of the <object> element. This is actually far more flexible than using an alt attribute.

        Unfortunately, browser bugs interfere with this quite a bit

    • Re:Yet another Web Accessibility article (Score:4, Informative)

      by JimDabell ( 42870 ) on Friday March 14, 2003 @06:09PM (#5515273) Homepage

      Don't forget to use ALT tags!

      I'd take advice like that with a pinch of salt, as the person dispensing it clearly demonstrates no understanding of the basic structure of an HTML document.

      There is no such thing as an "alt tag". There is an alt attribute, which is a completely different thing.

      The page is accessible if it can be properly viewed and navigated using a text-based browser (i.e. Lynx).

      That's a dangerous assumption. Take guiltless image use [stopdesign.com] as an example. Works fine in lynx, but fails miserably when you use a browser that renders CSS but does not display background images.

      Website accessibility is a complex topic, and there's no way you can automatically test something like this. The best you can do is provide hints on what to look for.

      I'm not particularly inclined to trust Cynthia, as the report document produced uses font sizes set at 12px and 10px verdana (!), and gives horizontal scrolling at 1024x768.

      One tool I have found to be of high quality is Accessibility Valet [webthing.com].

    • Re:Yet another Web Accessibility article (Score:2, Interesting)

      by Anonymous Coward
      What else, really, has to be considered outside of the limitations of a text-based browser? I'd love to read some comments from folks with more expertise in this area.
      • Pages depending on Java/Javascript (if you're testing with a text-based browser that supports JS, you might not realize the problem). No HREF tag should ever point to "#" or "javascript:...". Many sites use this for popups. You can add an ONCLICK element to a link to open a new window if you want, or use TARGET="_blank" (IIRC), but make sure

      • Re:Yet another Web Accessibility article (Score:4, Informative)

        by JimDabell ( 42870 ) on Saturday March 15, 2003 @10:11AM (#5519025) Homepage

        If your javascript function returns a specific value (possibly 0, but I don't remember for sure), the browser won't follow the link after opening the popup.

        You need to return false:

        <a href="alternative.html" onclick="dostuff(); return false;">...</a>

        Some sites look at your user-agent string and tell you to fuck off if they don't recognize that browser

        For instance, argos.co.uk [argos.co.uk] will refuse to handle gecko-based browsers. Idiocy.

        Although their web pages seem to render perfectly in Lynx/w3m/elinks/Mozilla/Konquerer/Opera, you can only log into the site with a user-agent of IE/Netscape.

        ...and of course, their logs show that people only use ie/netscape when visiting their site (because people using other browsers lie in their ua string, or just can't access the site at all). It's a vicious circle.

    • The problem is, it's not a matter of just allowing "text-only" browsers to correctly display your page; your pages should "make sense" from a semantical point of view, thus allowing semantic interpreters (such as browsers for blind persons and so on) to easily and correctly parse them. e.g. put the "title" attribute in your anchors, consider accesskeys, validate your xhtml, etc etc.

      For further things, take a look at Dive Into Accessibility [diveintoac...bility.org], a really good book.

    • The page is accessible if it can be properly viewed and navigated using a text-based browser (i.e. Lynx).

      A good start, but not strictly true. Lynx is a full-screen text browser, so you get a page by page view of a document. A better test would either be a line-mode browser (like the original W3 browser), or a speech browser like IBM's Homepage Reader.

      Apparently the RNIB (Royal National Institute for the Blind) have accessibility packs that includes a blindfold.

  • Not very usable (Score:2, Interesting)

    by Enrico Pulatzo ( 536675 )
    Cynthia throws the errors, but doesn't specify exactly what went wrong. For instance, the rule (paraphrasing) "every non-text element must contain an alt or longdesc tag" gets thrown, but doesn't say where the offense is coming from. In that same rule, Cynthia says that inputs must be inside forms. Why not break up the rules and show the user where they "went wrong". (by the way, I couldn't find in the page what she was complaining about--it checked out with Bobby and the validator)
    • You may want to have a go at the Bobby [watchfire.com] accessibility tests which throw not only errors and warnings but will show you what the page looks like without any style sheets and scripts, indications where the errors and warnings are, and links in the report to full-length explanations of what the errors mean. Almost all of the explanations provide samples of good and bad code, along with HTML coded samples.

      They limit the number of tests you can run via the web interface, but they also sell a standalone and serv

      • I use Bobby to test my site for content accessibility and made changes to as many pages on my site as possible in order to get almost every page to meet section 508 and/or WCAG Level A (or level 1).

        I hope you are not relying on the results of Bobby (or any automated accessibility checker) as the signal of compliance. It cannot be totally done automatically. For instance, the WCAG says that images must have meaningful alt attributes. A script can check that an alt attribute exists, but there's no way for

        • I do understand that it can only test for basic compliancy... I'm not saying that my site is 100% accessible to every single user agent in the world, but I'm using the results as a guideline to meet the requirements set out by Section 508 or WCAG.

          As far as alt attributes are concerned, I don't use a whole lot of images on my site... and yes, the alt and title attributes I use aren't the best ones.

          If the site was a government site or for a corporation, I would definitely not base the site's accessibility b

  • Avoid HiSoftware like the plague. (Score:1, Insightful)

    by Anonymous Coward
    I've had the pleasure of working with HiSoftware and all I can say is that it's a steaming pile of junk.

    It does NOT play very well W3C standards - so expect your nice XHTML Compliant webpage to be choke when it's run through the software. You can forget about CSS layouts "validating". You'll end up fighting the software instead of fixing your site.

    If you want an accessible website that is Section 508 compliant, use the W3C's Web Content Accessability Guidelines. If you pass those, you'll pass government r
  • In case you have a problem with using something (at least partially) from HiSoftware (I know some Assistive Tech. Specialists who do), you might be interested in using the WAVE. [webaim.org]

    Here's a Google [google.com] of some resources and info, as well.

    Ultimately, the biggest problem I have, is that too many web designers utterly rely on these validators. The problem is, they can only check for a few different parts of the standard. For instance, an automated validator may only be able to verify compliance with maybe half o
  • Bobby [watchfire.com]

    If you intend to follow the guidelines or not... reading the results is often interesting in either case =)

  • Her own medicine (Score:3, Interesting)

    by bertilow ( 218923 ) on Friday March 14, 2003 @09:35PM (#5516665) Homepage
    So, I checked my home page with Cynthia, and I got some complaints. They were reasonable. But then I saved the report Cynthia produced, and had her check her own code.

    Here it is:

    http://www.bertilow.com/div/cynthias_medicine/

    And here's her verdict:

    Verified File Name:
    http://www.bertilow.com/div/cynthias_medicine/
    Emulated Browser: Cynthia 1.0
    Date and Time: 3/14/2003 8:34:15 PM
    Failed Automated Verification
    Emulated Browser: Cynthia 1.0

    She failed! The reason is the crappy markup with loads of deprecated stuff. What were they thinking?
    • You devious chappy you!

      I did quick tests against a few big name sites
      www.sun.com - failed
      www.ibm.com - failed
      www.debian.org - failed
      www.redhat.com - failed
      www.slashdot.org - failed
      www.microsoft.com - failed
      www.w3c.org - failed
      www.opera.com - failed
      www.mozilla.org - passed

      The failure closest to passing was www.opera.org, which has a _single_ minor error, probably a typo, rather than policy.

      My web pages pass, but then again they're optimised for Lynx...

      YAW.

  • humans needed (Score:3, Informative)

    by bcrowell ( 177657 ) on Friday March 14, 2003 @10:55PM (#5517054) Homepage
    I run a web site [lightandmatter.com] that's small compared to Amazon.com, but large enough that it's a serious job maintaining all the nooks and crannies. The problem I had with Bobbie, and the problem I also have with Cynthia Says, is that they don't help me focus on any specific issues that would be real accessibility concerns. That makes it impractical for me to use their results, since I have a limited amount of time.

    For instance, I have a textual "home" link on every page that takes you to the site's home page. It also happens that I have made the graphical banner on my pages into a clickable link that will also take you to my home page. A blind person doesn't need to worry that there are two methods for getting to the home page -- there's one method that can be read aloud with speech-to-text software.

    On the other hand, there may be other things on my site that really are accessibility issues. The problem is, I can't tell from Cynthia's output what they are.

    It seems to me that the real need is for actual humans with disabilities to test web sites. Yes, I know that's expecting them to do something that they really shouldn't have to do, but I just don't think there's any alternative.

    I've been contacted once by a blind person who was having trouble using my site. The problem, however, was with my PDF files, not with my HTML. Bobby and Cynthia don't check PDF. And in fact, it wasn't something that I was able to solve, due to the realities of the way I created the PDFs.

  • I think I can sum up the feelings of the whole /. community with one uncreative statement: "Who the fuck cares?"

    • "Who the fuck cares?"

      Obviously not website designers, that's why governments around the world feel compelled to legislate website designers into doing the right thing.

      As Joe Clark [joeclark.org] points out, if you are making your website accessible because of legislation, then you are doing the right things for the wrong reasons.

      To pre-empt the "its only government sites that need to be accessible" crowd, here's a counter-argument to the South Western misruling [findlaw.com].

Slashdot Top Deals

Our OS who art in CPU, UNIX be thy name. Thy programs run, thy syscalls done, In kernel as it is in user!

Close