Slashdot Mirror
Google Accessible Search Released
Philipp Lenssen writes "Google today released Accessible Search, a Google Labs product aiming to rank higher pages which are optimized for blind users. Google asks you to adhere to the W3C's Web Content Accessibility Guidelines if you want to make sure your pages are accessible (and thus, rank better on Google Accessible Search). I wrote a small tool to compare results of default and accessible results."
From the Accessible Search FAQ:
How can sites make their content more accessible to the blind?
Some of the basic recommendations on how to make a website more useable and accessible include keeping Web pages easy to read, avoiding visual clutter -- especially extraneous content -- and ensuring that the primary purpose of the Web page is immediately accessible with full keyboard navigation
I wish more sites where like that. Do you want info? You get it right there, without all the mumbo-jumbo associated with most current websites.
Many aren't at present, that's the whole point of making it easier to find those that are. Of course, making pages as device independent as practical helps many others as well as the blind.
Blind people can use screen readers, but if you bury your content in flash or in images without alt text, it makes it very difficult (impossible really) for the screen reader to know what to say. Also, if you position everything in tables and it loads in a weird order but looks right when it hits the screen, it presents problems, since the reader won't know that the 5 rows down actually explains the 3 columns over.
.html (or .php or whatever) source files nice, clean and logical. A simple test is, if you can read your source file easily (ie in notepad or vi or whatever) then you're probably ok.
CSS can help with this, as it keeps the formatting away from the content, but you still have to keep your
I somehow think that a search for "Adobe", "Photoshop", etc. will still give you adobe.com as the #1 result, despite its accessibility problems.
FYI: Slashdot.org is an "accessible page" http://www.google.com/u/accessible?cx=accessible!& q=slashdot&btnG=Search
So is digg.com:
http://www.google.com/u/accessible?hl=en&lr=&ie=IS O-8859-1&cx=accessible!&q=digg&btnG=Search
But microsoft.com is not:
http://www.google.com/u/accessible?hl=en&lr=&ie=IS O-8859-1&cx=accessible!&q=microsoft&btnG=Search
There has been standards for blind web browsing for many years! The problem is that hardly anyone use them, even a lot of government web sites which by law must be accessible.
The amazing thing is that google, by page-ranking these pages higher, I believe it will do more to improve web accessibility than any law or standards organisation could.
--jeffk++
ipv6 is my vpn
I didn't get it initially but this is one of the best tool google ever gave us, most spam sites designer do not care for standards and are left out of the 'accessible' results. I think I'm going to switch to the new sevice soon.
With Google's "Accessable search", sites will be able to tell from their weblogs how many visitors are coming in via the "accessable search" route. So it will be possible to figure out the financial benefit of web accessability. If it turns out to be low, even for pages that Google thinks are "accessable", there's a business case for not bothering with "accessability".
Finally I can start using Links2 or even Lynx to start browsing or any other curses based browser... mmm... gopher. Will work perfectly with my pine mail and telnet based IRC.
I really commend google for providing us and maybe even forcing webdevelopers to use decent, W3C compatible standards. This means soon enough, we'll have websites that aren't IE compatible.
Custom electronics and digital signage for your business: www.evcircuits.com
It seems I'm not being served any sponsored links on this one. Nice. I also note that searching for "white house" will put the Wikipedia article above the official home page, as opposed to what happens on the regular search.
blind people. i think we all can benefit from this simple and straightforward representation of search results. i actually like the new accessible interface. it was a kind decision by google to not put any ads in their accessible search. i think more people will use the google accessible search than just the blind people.
so, has anyone tried using any of the screen reading software to test whether search results are actually readable without looking at the screen?
and also what about keyboard shortcuts? since blind people can't use mouse, there must be some other way for them to find a search box / search result, right?
Check out Pandora by Music Genome Project
Perhaps you need to open your eyes then.
A well designed standards based website is built in a uniform standard way and contains all the hints required for a screenreader to pickup on.
Badly designed sites use lots of custom content and stupid user interface elements which make it difficult to access (both from a blind screen readers perspective and usually from a normal users view.
liqbase
http://validator.w3.org/check?uri=http%3A%2F%2Flab s.google.com%2Faccessible%2F
(neither is any of google's web pages).
I don't want a signature.
It depends. If you actually talk to a blind person, tables aren't as annoying as people using characters between links. For instance, doing navigation with a pipe (vertical bar) or some other character sounds strange to them. I worked for someone who was blind and owned their own business for a time. He said cnet was one of the most annoying web properties for him to use. At the time their navigation sounded terrible and was quite randomly placed. He also didn't like navigation at the top as much except on the first page. Worst of all, he hated hearing. about us vertical bar products veritcal bar etc...
Oh and he used Internet explorer. His software tied in with that so it looked like an ie hit and even tried to load all the crap everyone else deals with in IE. He did IE on windows using a dell.
MidnightBSD: The BSD for Everyone
Tables should be used to mark up truly tabular information ("data tables"). Content developers should avoid using them to lay out pages ("layout tables"). Tables for any use also present special problems to users of screen readers (refer to checkpoint 10.3).
Even Google's ultra-simple front page violates this guideline, despite zero need to do so.
Point 3 of the guidelines says this:
Mark up documents with the proper structural elements. Control presentation with style sheets rather than with presentation elements and attributes.
But if you dig into the source of google.com, you see cruft like this:Google fails rather dramatically to implement any web standard, not even including a doctype. These problems aren't limited to their front page, either. news.google.com is just as bad or worse.
This is really a shame. The content that google presents is lightweight and free of the layout challenges that can sometimes make web standards difficult to follow: Google should be the perfect test case for perfect standards and accessibility. Instead, it's a throwback to 1996 web design. That they're launching a tool to test accessibility to the blind is incredibly ironic.
I stole this sig from someone cleverer than me.
My son is blind. He uses the web with a reader, and loves google. He searches for music, among other things. It's pretty amazing how effecient he is with it, given he can't see. Some sites work better than others. It is possible to optimize a site for the blind. You're ignorant.
http://cubemonkey.net/quotes -- fortune-mod quote generator
From what I've read of the guidelines, my website gopher://gopherrulez.org just won't cut it!
If someone says he and his monkey have nothing to hide, they almost certainly do.
That's utter nonsense. First off, you think blind people can't process information? WTF?
Second, it's not even about 'optimising for the blind' so much as simply 'using (rather than abusing) the web.' The web was designed from the start to deliver information in a neutral format so that the user-agent (browser) could then deliver that information appropriately. This may mean laying it out on a screen (of unknown dimensions and capability,) or it may mean speaking it aloud, or whatever. Proper web design is accessible to everyone. The errors that make sites inaccessible to the blind are the same errors that make them annoying and sometimes unusable to the rest of us as well.
Keep in mind that you cannot dictate layout and use html properly and you'll have no problem. Ignore that fact and you shut out a lot of people, not just the blind.
=-=-=-=-=-=-=-=-=-=-=-=-=-=-
Friends don't let friends enable ecmascript.
Has anyone ever had the problem of trying to find song lyrics and being bombed with pages full of nonsence?! First time I tried with this interface I got the exact results I was wanting. So I know now that this is better for lyrics... what else would it be better for?
Flash's accessibility features are not terribly helpful. They don't adhere to any standards, and require you to run particular software (they rely on Microsoft Active Accessibility, for example) to be compatible with them.
No, before you go around spouting "STFU" at other folks, make sure you're able to back up your claims. Just becuase a company says "our stuff has accessibility features" doesn't mean they DO, or that they are useful to people who would need them.
PS: I just tried to find out more about their accessibility features, but they use Flash to explain that, and I don't have sound on this machine... kind of useless.
The w3c check is not an acurate measure on how thruly accessible the pages are
I now this because I have friends who are blind and frecuently use screen reader applications
If you really want to make sure your pages are accessible then download the trial version of Jaws for Windows wich is the defacto standard screen reader. This trial is limited to 40 minutes per session, but those 40 minutes should be enough to test your webpages.
As mentioned in posts above, make sure the content can be reaced quickly by readers.
Which raises an interesting point. What would happen if Google were to rank pages with valid html slightly higher that pages that don't validate properly?
XP is basicly 98 with a lot more extra features to hunt down and disable. --Dram
When I worked for Albertsons.com, we accidently created a site that worked well for the blind, and even received some recognition for it. After the initial release of the site, and finding out how some blind people were using it, we actually had a lady talk to our develoment staff about optimizing the site even further, which we did.
It is a VERY good feeling to have, when someone that is in their 40s tell you how grateful they are for your service, since it is the first time in her life, she was able to actually shop for groceries. Adding alt text to images, that TELLS people what the image is, is a huge help. It is very simple things, that you do in HTML to make it possible. If you go in aiming to create well-formed HTML, you are about 80% there, by default.
A common misconception is that website accessibility only benefits blind people. Accessible, semantically-structured websites are good for many things, most notably Search Engine Optimisation and cutting down on bandwidth. How's that for financial benefit?
The Olympics was an Australian case. Target was not "successfully sued"; that's still pending. Southwest Airlines won a case over that issue, Access Now vs. Southwest.; their "virtual ticket counter" does not have to be handicapped-accessable. Access Now appealed, and the Court of Appeals for the Eleventh Circuit rejected the appeal. That's the only final US court decision on the subject to date.
As the court put it, it's up to Congress to change the ADA if Congress wants it to apply to the Internet.
Let's take a hypothetical situation. You live in a communist country (China), and are trying to do some research on a paper for your studies. You need to pull up some information on a completely legitimate topic (as per China's regulations). What would you use to find this information? Would you, as a Chinese person, whine and moan and boycott Google because they were censoring your results? Would you refuse to use them, because you know full well that if they didn't censor their results, there'd be no Google for you at all?
I'm not trying to say Google is justified here. But, at some point you have to draw the line. China is who you should be griping at, not Google. Google is still trying to help the people of China, they just aren't legally allowed to help them as much as they can help people in other countries.
So, I'll ask again. Isn't some of Google better than none?
The reason is of course not that Google cannot make a page without the use of tables or the font tag, but that the pages work with even the oldest of browsers (except IE 2.0 which doesn't do tables). You can't DIV or CSS your way around that. Big sites like Google and Yahoo have always been breaking rules, either for speed (not using quotes etc.) or for backward compatability. W3C is nice but doesn't earn you money.
Actually, many modern screen readers have the capability of analyzing the HTML document and creating an alphabetical list of the links on the page that they can navigate.
Here's a list of Features and Enhancements for the latest versions of JAWS (a widely used screen reader). Reading through them will give you some idea of how screen readers operate. There's a lot more to it than you think.
Here's a clip from one of them:
HTML and the Internet
Improved Performance on the Web
With the new Internet and HTML support in JAWS 7.10, you will experience increased accuracy, improved navigation, and better text recognition when reading Web pages or other virtual documents.
Visible Focus Rectangle
When you press TAB or SHIFT+TAB to move to links and buttons in Internet Explorer and Firefox, the focus rectangle is now visible. This rectangle is useful to sighted users because it visually indicates the location of the cursor. The focus rectangle does not move when you use the arrow keys or Navigation Quick Keys to read.
Route the Virtual Cursor to the JAWS Cursor
You can now press INSERT+NUM PAD PLUS in Internet Explorer, Firefox, and virtual documents to route the Virtual Cursor to the mouse pointer. Using this command moves the Virtual Cursor to the current location of the mouse pointer and can help sighted users navigate Web pages. In addition, routing the JAWS Cursor to the Virtual Cursor (INSERT+NUM PAD MINUS) is much more accurate and moves the mouse pointer to the exact character the Virtual Cursor is located on.
Route the PC Cursor to the Virtual Cursor
You can now press CTRL+INSERT+DELETE in Internet Explorer, Firefox, and virtual documents to route the PC Cursor to the Virtual Cursor. Using this command moves the PC Cursor (and the application focus) to the current location of the Virtual Cursor. Visually, the page will scroll so that the area containing the Virtual Cursor is visible on the screen. This command is the opposite of the Route Virtual to PC Cursor command (INSERT+DELETE).
One Setting for Controlling Page Refreshes
There is now only one setting for controlling page refreshes. Previously there were two settings, one for controlling page refreshes caused by the browser, and another for controlling refreshes caused by embedded ActiveX controls, such as Macromedia Flash. These have been consolidated into a single option for controlling both since it is often not apparent which is causing the page to refresh.
Improved Detection of Dynamic Page Updates
Previously, if script code was used on a page to control visibility without the user actually interacting with the page, JAWS would not detect the page update and would either show content that was not really there or not show content which was made visible. This should no longer occur.
Enhanced Screen Tracking
The screen no longer scrolls up or down erratically while you are using the Say All command or navigating by other means. The screen only moves when the content about to be read is not visible.
Document Presentation Mode Line Length
You can now define how long a single line will be when viewing an HTML page in Document Presentation Mode. This can help you read lengthy tables easier because all the content from each row in the table can fit on a single line. The increased line length stops JAWS from rendering rows across multiple lines. When you exit Document Presentation Mode, JAWS will render the page using the normal maximum line length.
The default line length in Document Presentation Mode is 400 characters, which is enough to fit most table rows on one line. To change the line length, open the Utilities menu and choose Configuration Manager. Then, open the Set Options menu and choose HTML Options. Enter a new line length in the Document Presentation Mode Maximum Line Length edit box located on the Text tab.
'Yes, firefox is indeed greater than women. Can women block pops up for you? No. Can Firefox show you naked women? Yes.'