<?xml version="1.0" encoding="UTF-8"?><!-- generator="wordpress/2.2.3" -->
<rss version="2.0" 
	xmlns:content="http://purl.org/rss/1.0/modules/content/"
	xmlns:dc="http://purl.org/dc/elements/1.1/"
	>
<channel>
	<title>Comments on: HTML Accessibility</title>
	<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/</link>
	<description></description>
	<pubDate>Wed, 09 Jul 2008 12:13:44 +0000</pubDate>
	<generator>http://wordpress.org/?v=2.2.3</generator>

	<item>
		<title>By: jm</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16213</link>
		<dc:creator>jm</dc:creator>
		<pubDate>Tue, 25 Mar 2008 00:51:30 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16213</guid>
		<description>You may want to try the following LiveCD: http://clcworld.net/moz_csun_2007_iso.zip

Details at: http://lists.becta.org.uk/pipermail/oats-sig/2007-March/000919.html

It uses the following:
1. FireVox - the screen reader add-on to Firefox giving access to both browser and the web
2. CLiCk, Speak - the add-on that speaks and highlights the web under user direction
3. NVDA - the open source windows screen reader.</description>
		<content:encoded><![CDATA[<p>You may want to try the following LiveCD: <a href="http://clcworld.net/moz&#95;csun&#95;2007&#95;iso.zip" rel="nofollow">http://clcworld.net/moz<em>csun</em>2007_iso.zip</a></p>

<p>Details at: <a href="http://lists.becta.org.uk/pipermail/oats-sig/2007-March/000919.html" rel="nofollow">http://lists.becta.org.uk/pipermail/oats-sig/2007-March/000919.html</a></p>

<p>It uses the following:
1. FireVox - the screen reader add-on to Firefox giving access to both browser and the web
2. CLiCk, Speak - the add-on that speaks and highlights the web under user direction
3. NVDA - the open source windows screen reader.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Ian Bicking</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16209</link>
		<dc:creator>Ian Bicking</dc:creator>
		<pubDate>Mon, 24 Mar 2008 17:27:30 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16209</guid>
		<description>Bat: thanks a lot for all the links.  I gave NVDA a try, though only with limited success.  It seemed to have performance issues that caused it to delay for long periods of time with no feedback, or start reading something that had moved from the screen a while ago.  It ended up swallowing keys from the right-hand side of the keyboard, even though its speech claimed to be typing those keys.  Luckily I at least was able to see the page to realize this was happening.  I tried a couple times to navigate with the screen turned off, but it was quite unsuccessful.

Another issue I had was with links, which sometimes NVDA would read out in full (the href -- a long, tedious, and unhelpful link).  I don't know why it chose to do that sometimes and not other times.  There seem to be a number of toggles which I imagine you'd turn on or off depending on the page.  But you have to be familiar with the tool to know what to do when.

In the end I got [a couple ideas](http://www.openplans.org/projects/opencore/blog/2008/03/24/accessibility-on-our-site/) of small things we could improve, but only small things.  Actually getting into the site revealed problems with NVDA that I didn't know how to get past.  I'll be giving some other tools a try as well.</description>
		<content:encoded><![CDATA[<p>Bat: thanks a lot for all the links.  I gave NVDA a try, though only with limited success.  It seemed to have performance issues that caused it to delay for long periods of time with no feedback, or start reading something that had moved from the screen a while ago.  It ended up swallowing keys from the right-hand side of the keyboard, even though its speech claimed to be typing those keys.  Luckily I at least was able to see the page to realize this was happening.  I tried a couple times to navigate with the screen turned off, but it was quite unsuccessful.</p>

<p>Another issue I had was with links, which sometimes NVDA would read out in full (the href &#8212; a long, tedious, and unhelpful link).  I don&#8217;t know why it chose to do that sometimes and not other times.  There seem to be a number of toggles which I imagine you&#8217;d turn on or off depending on the page.  But you have to be familiar with the tool to know what to do when.</p>

<p>In the end I got <a href="http://www.openplans.org/projects/opencore/blog/2008/03/24/accessibility-on-our-site/">a couple ideas</a> of small things we could improve, but only small things.  Actually getting into the site revealed problems with NVDA that I didn&#8217;t know how to get past.  I&#8217;ll be giving some other tools a try as well.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Kumar McMillan</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16207</link>
		<dc:creator>Kumar McMillan</dc:creator>
		<pubDate>Mon, 24 Mar 2008 15:33:56 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16207</guid>
		<description>Hey Ian, I was a bit surprised by your heckler but you seemed to perk up, as if challenged.  If it were me I doubt I'd be able to handle it so smoothly.  Good job ;)  PyCon isn't well suited to having a "discussion" during a talk but, then again, that may or may not have been useful.  

There was an important message I got from your talk that deserves mention: the web is filled with bad HTML and we are stuck with it, so let's learn to deal with it instead of despise it.  Accessibility purists have a hard to time coming to terms with this.  

On a personal note, I am highly grateful for your work on lxml.html because it has been the only reliable way for me to apply xpaths to ebay HTML descriptions (for a project where I want to analyze thousands of auctions).  Anyone who is familiar with the site knows these descriptions are created by sellers and quality wise are on par with geocities markup.  Yet for my purposes contain information I want to parse and there is nothing I can do about their abuse of HTML.  So, thanks!</description>
		<content:encoded><![CDATA[<p>Hey Ian, I was a bit surprised by your heckler but you seemed to perk up, as if challenged.  If it were me I doubt I&#8217;d be able to handle it so smoothly.  Good job ;)  PyCon isn&#8217;t well suited to having a &#8220;discussion&#8221; during a talk but, then again, that may or may not have been useful.  </p>

<p>There was an important message I got from your talk that deserves mention: the web is filled with bad HTML and we are stuck with it, so let&#8217;s learn to deal with it instead of despise it.  Accessibility purists have a hard to time coming to terms with this.  </p>

<p>On a personal note, I am highly grateful for your work on lxml.html because it has been the only reliable way for me to apply xpaths to ebay HTML descriptions (for a project where I want to analyze thousands of auctions).  Anyone who is familiar with the site knows these descriptions are created by sellers and quality wise are on par with geocities markup.  Yet for my purposes contain information I want to parse and there is nothing I can do about their abuse of HTML.  So, thanks!</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Simon Willison</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16205</link>
		<dc:creator>Simon Willison</dc:creator>
		<pubDate>Mon, 24 Mar 2008 14:04:15 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16205</guid>
		<description>Screen readers do indeed execute JavaScript (or rather run on top of browser DOM trees that have already executed it for them) - they have to, or a huge number of sites would be completely useless. That doesn't mean that unobtrusive scripting can't help with accessibility though. The principle accessibility problem facing heavy JavaScript sites today is that there's no reliable hook for telling a screen reader "that part of the page was just dynamically updated". If you submit a form by Ajax and then refresh the contents of a div with a thankyou, a screen reader user won't hear anything - they'll have to manually rescan the entire page to figure out what just happened.

With unobtrusive scripting, you can provide an "accessible version" which is actually just the regular site but with all of the window.onload style handlers disabled. A simple checkbox (hidden off screen with CSS if you like) which sets a cookie and you're done.</description>
		<content:encoded><![CDATA[<p>Screen readers do indeed execute JavaScript (or rather run on top of browser DOM trees that have already executed it for them) - they have to, or a huge number of sites would be completely useless. That doesn&#8217;t mean that unobtrusive scripting can&#8217;t help with accessibility though. The principle accessibility problem facing heavy JavaScript sites today is that there&#8217;s no reliable hook for telling a screen reader &#8220;that part of the page was just dynamically updated&#8221;. If you submit a form by Ajax and then refresh the contents of a div with a thankyou, a screen reader user won&#8217;t hear anything - they&#8217;ll have to manually rescan the entire page to figure out what just happened.</p>

<p>With unobtrusive scripting, you can provide an &#8220;accessible version&#8221; which is actually just the regular site but with all of the window.onload style handlers disabled. A simple checkbox (hidden off screen with CSS if you like) which sets a cookie and you&#8217;re done.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: roberthahn</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16204</link>
		<dc:creator>roberthahn</dc:creator>
		<pubDate>Mon, 24 Mar 2008 13:14:05 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16204</guid>
		<description>To your point about freely available screen readers, I can testify that the screen reader built into Mac OS X is very usable, and I'm told there's screen reader infrastructure built into Windows XP.  

I'm all for empirical testing of the accessibility of web pages, but I suspect you'll find that after you play with the screen reader on a representative sample of code, you'll gain an intuition on what works and what doesn't very quickly, and will use the software less over time.</description>
		<content:encoded><![CDATA[<p>To your point about freely available screen readers, I can testify that the screen reader built into Mac OS X is very usable, and I&#8217;m told there&#8217;s screen reader infrastructure built into Windows XP.  </p>

<p>I&#8217;m all for empirical testing of the accessibility of web pages, but I suspect you&#8217;ll find that after you play with the screen reader on a representative sample of code, you&#8217;ll gain an intuition on what works and what doesn&#8217;t very quickly, and will use the software less over time.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: The Bat!</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16200</link>
		<dc:creator>The Bat!</dc:creator>
		<pubDate>Mon, 24 Mar 2008 10:04:28 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16200</guid>
		<description>I'm a blind screen reader user, and try to keep my own web site as accessible as I can, whatever 'accessible' means to any particular person. 

I don't understand all the technicalities being discussed here, but maybe a few comments would be helpful.

To me, a good web page is just a well-structured page, with content that I actually want to read. Get the structure right and most other things will follow.

I visited a friend's site because he wanted accessibility comments. It was great - the screen reader told me everything I needed to know, and it did it without any glitches.

But when I ran the code through an HTML validator, the validator moaned about various supposed accessibility problems. These amounted to stylistic niceties. The id attribute for some of the paragraphs wasn't explicit enough. Well, big deal, who cares? It didn't stop the page reading fine.

It's hard to imagine anyone getting heated over the longdesc tag. Don't use it, is all I'd say. Use succinct alt tags instead and nobody needs to break their brains trying to make graphics accessible. Alt tages being absent altogether can be a major problem, however, if you're expected to hit buttons that just look like a series of blobs - something you often come across on the Web, and in desktop programs, for that matter. 

There are freely available screen readers, and some are getting very good. NVDA at http://www.nvda-project.org/ is possibly the best of these. It is fairly JAWS-like in some ways, but costs a thousand dollars less, and doesn't do horrible things to your video display or your registry.

There's also System Access to Go, which has been made available free as a web based screen reader at http://www.accessibilityisaright.org/

Then there's Thunder, free from www.screenreader.net. Arguably Thunder is the weakest of these screen readers, but whatever floats your boat... It does seem to go down well with less expert computer users.

What do blind people actually use? Well, moot point. Most don't use anything, and don't go near computers, which are too alien to deal with. At the other end of the market, the elite of screen reader users are relatively well catered for, not least because many get access tech paid for so they can do a job, and get subsidies from Governement agencies. That's the only reason that things like JAWS can exist. The point, though, is that accessibility has been pretty much formulated for screen reader users, and other people with disabilities don't get anything like so much of the action. I have an axe to grind as a once partially sighted computer user. Then there are dyslexics, colourblind people, those with minimal dexterity, deaf people and on and on, who also deserve a slice of the cake. But in the end, so long as web designers don't put up unnecessary barriers, they won't cause any of those people too much grief.

Less well known, and cheaper, screen readers are Window Eyes, Hal, and Serotek System Access (paid version). Although JAWS is the well-known one, it doesn't seem to rule the place like it used to.  Every screen reader has its advocates, but then, once you'd paid up and mastered all those dastardly 4-finger keystrokes, you'd better believe you're using something really good! At the moment, it seems that the jury's out on which screen reader will prevail longer term.

I can't believe JAWS cam't recognise paragraph markup! NVDA does it fine, but you need to spend a minute setting up the config so that it does it. JAWS, I understand, has a jungle of verbosity settings, and with enough fiddling about, I'm sure it can be made to treat the markup of pages in many different ways.

Worth mentioning that JAWS and the others I mention are confined to the Windows platform. Voiceover on the Mac is now a serious contender, not least because it's included in the price of the computer. Then there's Linux........ 

Quick mention of CSS - a great way for sighted people to have their eye candy and for blind people to eat it too.

I don't like the sound of a separate layer for blind people. I don't know what it means technically anyway, but let's stay mainstream, please.</description>
		<content:encoded><![CDATA[<p>I&#8217;m a blind screen reader user, and try to keep my own web site as accessible as I can, whatever &#8216;accessible&#8217; means to any particular person. </p>

<p>I don&#8217;t understand all the technicalities being discussed here, but maybe a few comments would be helpful.</p>

<p>To me, a good web page is just a well-structured page, with content that I actually want to read. Get the structure right and most other things will follow.</p>

<p>I visited a friend&#8217;s site because he wanted accessibility comments. It was great - the screen reader told me everything I needed to know, and it did it without any glitches.</p>

<p>But when I ran the code through an HTML validator, the validator moaned about various supposed accessibility problems. These amounted to stylistic niceties. The id attribute for some of the paragraphs wasn&#8217;t explicit enough. Well, big deal, who cares? It didn&#8217;t stop the page reading fine.</p>

<p>It&#8217;s hard to imagine anyone getting heated over the longdesc tag. Don&#8217;t use it, is all I&#8217;d say. Use succinct alt tags instead and nobody needs to break their brains trying to make graphics accessible. Alt tages being absent altogether can be a major problem, however, if you&#8217;re expected to hit buttons that just look like a series of blobs - something you often come across on the Web, and in desktop programs, for that matter. </p>

<p>There are freely available screen readers, and some are getting very good. NVDA at <a href="http://www.nvda-project.org/" rel="nofollow">http://www.nvda-project.org/</a> is possibly the best of these. It is fairly JAWS-like in some ways, but costs a thousand dollars less, and doesn&#8217;t do horrible things to your video display or your registry.</p>

<p>There&#8217;s also System Access to Go, which has been made available free as a web based screen reader at <a href="http://www.accessibilityisaright.org/" rel="nofollow">http://www.accessibilityisaright.org/</a></p>

<p>Then there&#8217;s Thunder, free from <a href="http://www.screenreader.net." rel="nofollow">www.screenreader.net.</a> Arguably Thunder is the weakest of these screen readers, but whatever floats your boat&#8230; It does seem to go down well with less expert computer users.</p>

<p>What do blind people actually use? Well, moot point. Most don&#8217;t use anything, and don&#8217;t go near computers, which are too alien to deal with. At the other end of the market, the elite of screen reader users are relatively well catered for, not least because many get access tech paid for so they can do a job, and get subsidies from Governement agencies. That&#8217;s the only reason that things like JAWS can exist. The point, though, is that accessibility has been pretty much formulated for screen reader users, and other people with disabilities don&#8217;t get anything like so much of the action. I have an axe to grind as a once partially sighted computer user. Then there are dyslexics, colourblind people, those with minimal dexterity, deaf people and on and on, who also deserve a slice of the cake. But in the end, so long as web designers don&#8217;t put up unnecessary barriers, they won&#8217;t cause any of those people too much grief.</p>

<p>Less well known, and cheaper, screen readers are Window Eyes, Hal, and Serotek System Access (paid version). Although JAWS is the well-known one, it doesn&#8217;t seem to rule the place like it used to.  Every screen reader has its advocates, but then, once you&#8217;d paid up and mastered all those dastardly 4-finger keystrokes, you&#8217;d better believe you&#8217;re using something really good! At the moment, it seems that the jury&#8217;s out on which screen reader will prevail longer term.</p>

<p>I can&#8217;t believe JAWS cam&#8217;t recognise paragraph markup! NVDA does it fine, but you need to spend a minute setting up the config so that it does it. JAWS, I understand, has a jungle of verbosity settings, and with enough fiddling about, I&#8217;m sure it can be made to treat the markup of pages in many different ways.</p>

<p>Worth mentioning that JAWS and the others I mention are confined to the Windows platform. Voiceover on the Mac is now a serious contender, not least because it&#8217;s included in the price of the computer. Then there&#8217;s Linux&#8230;&#8230;.. </p>

<p>Quick mention of CSS - a great way for sighted people to have their eye candy and for blind people to eat it too.</p>

<p>I don&#8217;t like the sound of a separate layer for blind people. I don&#8217;t know what it means technically anyway, but let&#8217;s stay mainstream, please.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: cariaso</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16194</link>
		<dc:creator>cariaso</dc:creator>
		<pubDate>Mon, 24 Mar 2008 03:24:59 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16194</guid>
		<description>Would it be possible for the blind to read through an annotation layer similar to Ping's Crit
http://zesty.ca/crit/
in which they and others with poor vision could collaboratively edit a screen reader appropriate layer?</description>
		<content:encoded><![CDATA[<p>Would it be possible for the blind to read through an annotation layer similar to Ping&#8217;s Crit
<a href="http://zesty.ca/crit/" rel="nofollow">http://zesty.ca/crit/</a>
in which they and others with poor vision could collaboratively edit a screen reader appropriate layer?</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: David Boddie</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16191</link>
		<dc:creator>David Boddie</dc:creator>
		<pubDate>Mon, 24 Mar 2008 01:37:07 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16191</guid>
		<description>I don't have the answers to your many questions about accessibility and the often poor support on different platforms, but I attended a presentation last year that I found enlightening:

http://www.nuug.no/aktiviteter/20071211-accessibility/

The resources for the presentation, including the slides and video, are available.</description>
		<content:encoded><![CDATA[<p>I don&#8217;t have the answers to your many questions about accessibility and the often poor support on different platforms, but I attended a presentation last year that I found enlightening:</p>

<p><a href="http://www.nuug.no/aktiviteter/20071211-accessibility/" rel="nofollow">http://www.nuug.no/aktiviteter/20071211-accessibility/</a></p>

<p>The resources for the presentation, including the slides and video, are available.</p>
]]></content:encoded>
	</item>
	<item>
		<title>By: Gary Bishop</title>
		<link>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16190</link>
		<dc:creator>Gary Bishop</dc:creator>
		<pubDate>Mon, 24 Mar 2008 01:19:02 +0000</pubDate>
		<guid>http://blog.ianbicking.org/2008/03/23/html-accessibility/#comment-16190</guid>
		<description>Thanks for the thought provoking post.</description>
		<content:encoded><![CDATA[<p>Thanks for the thought provoking post.</p>
]]></content:encoded>
	</item>
</channel>
</rss>
