alt="" vs name=""??? (1 Viewer)

[Denny Oubidoux]

im doing this project on web accessibility in relation to this one site that uses gifs for navigation. all the alts are left blank but they all have name="blah". i presume the "name" is no good to anyone who relys on alts to get what the image says, right? the "name" attributes are all fairly meaningless anyway as are the urls so...

i cant find any mention anywhere of name being an acceptable alternative to alt but i thought i'd ask before starting the diss on it. i'll probably be back with more such questions

 

[plug]

as far as I'm aware the name attribute is purely there in order to provide a unique reference to a particular object, (ie: in order to give an image or a table etc. a unique name for scripting purposes etc.), but as far as I'm aware it does not work and is not intended to be used as an alternative to the alt attribute. so.. go tear that site a new asshole!!!

 

[krystal]

nope, name = bad unless you're talking forms. You must always use alt - if they've used alt="blah" throughout the site they should be shot.

 

[Denny Oubidoux]

nope, name = bad unless you're talking forms. You must always use alt - if they've used alt="blah" throughout the site they should be shot.

excellent. do you mind if i reference that? its not actually alt="blah" though, just alt="" which isnt anymore helpful i suppose.

 

[krystal]

no, but actually alt=" " has its uses.

You'd use that if you were putting in a spacer graphic or somethng, which doesn't provide any information and so doesn't need alt text. But if you provide the alt=" " then a screen reader will just read it out like 'image is blank' as opposed to 'no information provided about image' which would leave a user wondering if they'd missed out on important information on the page. It's good manners basically. Plus you won't pass a bobby test without giving all images tags, blank or otherwise.

But that said, you shouldn't be using hacks like that in pages unless you really have to - formatting should be taken care of outside of the document structure in the stylesheet.

excellent. do you mind if i reference that? its not actually alt="blah" though, just alt="" which isnt anymore helpful i suppose.

 

[Denny Oubidoux]

no, but actually alt=" " has its uses.

thanks for the help! so do you need the gap in alt=" " or is alt="" as good?

by the way, im going through these WCAG guidelines on accessibility (just the priority ones) and am so far on point 1.1 so perhaps you can see where this thread may be going... !zed

 

[krystal]

I always put the space. But I don't know for sure if you do have to or not. Ah sure I just do.

There are some really good books out there which help make sense of the WAI guidelines - Joe Clarke's Building Accessible Websites and Jeffrey Zeldman's Designing with web standards is my personal bible.

thanks for the help! so do you need the gap in alt=" " or is alt="" as good?

by the way, im going through these WCAG guidelines on accessibility (just the priority ones) and am so far on point 1.1 so perhaps you can see where this thread may be going... !zed

 

[krystal]

have just remembered - name is used in image declarations for javascript. Was there javascript used for rollovers or something? However, they're usually unique, so you'd have a different name per image or else they'd all rollover to the same thing.

nope, name = bad unless you're talking forms. You must always use alt - if they've used alt="blah" throughout the site they should be shot.

 

[ernesto]

NERRRRRRRRRRDSSS!!!

 

[Denny Oubidoux]

NERRRRRRRRRRDSSS!!!

shut up

i have that zeldman book too, i read most of it and found it good and interesting though its not much good as a reference if theres something youre not sure about and want to find out the right way to do it. the index is a bit shit. all makes good sense though. the joe clarke one is better for instructions if thats what you need, i need the instructions. ive been reading the free online version.

there is javascript rollovers alright so thats that solved.

 

[ernesto]

shut up

i have that zeldman book too, i read most of it and found it good and interesting though its not much good as a reference if theres something youre not sure about and want to find out the right way to do it. the index is a bit shit. all makes good sense though. the joe clarke one is better for instructions if thats what you need, i need the instructions. ive been reading the free online version.

there is javascript rollovers alright so thats that solved.

ok.... ye big bollix

 

[Denny Oubidoux]

anybody any thoughts on this one...

3.2 Create documents that validate to published formal grammars.

What are the accessibility implications of this?

this is fun isnt it....

 

[krystal]

It means you should use XHTML, either transitional or strict. Use MathML if you need to, change the lang atttibute within documents if the language changes, use proper doctypes etc.

Getting your code to validate is the single easiest way of ensuring that you'll be able to provide similar user experiences across browsers, platforms and devices. Good firms make software for devices that conforms to standards, so creating standardised, validated code means whatever you create will work on those devices and programs.

use the markup validation service to check your XHTML and CSS documents

http://validator.w3.org

anybody any thoughts on this one...


What are the accessibility implications of this?

this is fun isnt it....

 

[Denny Oubidoux]

this one about tables is a bit confusing. guideline 3.3 says "Use style sheets to control layout and presentation", grand, its all done with tables so i can knock them for that, but then 5.3 goes and says "do not use tables for layout unless the table makes sense when linearized. Otherwise, if the tables do not make sense provide an alterntaive equivalent (which may be a linearized version)" and 5.4 says "If a table is used for layout, do not use any structural markup for the purpose of visual formatting".

is 5.4 not a contradiction in terms? i thought tables were structural mark-up, used for structuring tabular data. i wish the w3c would get more dogmatic about these things and be more black or white, more heaven or hell.

tables are a pain in the hole.

 

[krystal]

You can use tables for layout alright. IE hasn't gotten to the stage of firefox and the mozilla based browsers in terms of its standards support so sometimes using a sneaky table is about the only way you can get a particular layout to work across browsers. Now - you can use them for simple layout, but you shouldn't be using anything within the XHTML document structure to control how they appear on screen. So this means you need to give all your layout tables ids and assign values in the CSS. So in the XHTML document you might have a simple table for your navigation buttons - you mustn't specify within the XHTML code how wide it's to be, or where it's positioned, or anything like that. You need to do that in the CSS.

The reason there are guidelines on table usage is that people use them all the time for positioning hacks. I lay out my documents using sequential divs and sometimes resort to a table for navigation buttons. That's ok in my book. And it passes muster. If you're laying out your whole document as a table then that's not so good - think about how a screen reader will read out the data to a user - that's what is meant by the lineraized bit. So if you have a table that's actually just 4 rows and each cell contains sequential information for the page, then that's ok. It's using tables for layouts that have many colspans="3" etc that will cause a problem. And it's not ok.

Now, handling tabular data within documents is another ballgame. People use tables all the time because they give a graphic design feel to a webpage and make controlling text positioning easy. So you need to think about how you present data on pages - you should only use tables if there is a genuine need for one. The table must be summarised, and given header values which are relevant to rows and/or columns in which the data will reside.

this one about tables is a bit confusing. guideline 3.3 says "Use style sheets to control layout and presentation", grand, its all done with tables so i can knock them for that, but then 5.3 goes and says "do not use tables for layout unless the table makes sense when linearized. Otherwise, if the tables do not make sense provide an alterntaive equivalent (which may be a linearized version)" and 5.4 says "If a table is used for layout, do not use any structural markup for the purpose of visual formatting".

is 5.4 not a contradiction in terms? i thought tables were structural mark-up, used for structuring tabular data. i wish the w3c would get more dogmatic about these things and be more black or white, more heaven or hell.

tables are a pain in the hole.

 

[egg_]

Minimal tables makes integrating php way easier too

 

[Denny Oubidoux]

Minimal tables makes integrating php way easier too

how so? ive been trying to learn that too, does it not just output whatever html you tell it to?

 

[Denny Oubidoux]

Can anyone save me the bother of doing endless research by telling me the legal situation in ireland with regard to web accessibility? I know all government department websites are meant to be AA compliant but is that just an aspiration or a law? is there EU interference?

.|..|

 

[taubstumm]

Can anyone save me the bother of doing endless research by telling me the legal situation in ireland with regard to web accessibility? I know all government department websites are meant to be AA compliant but is that just an aspiration or a law? is there EU interference?

.|..|

it's not law. if it was, websites would ostensibly become 'illegal' as soon as someone added a new picture and forgot to add an alt attribute. however, as far as i know, accessibility (as a design practice, not a buzzword) is a requirement for all new government websites.

american government websites are meant to adhere to section 508 requirements. they're based on the w3c requirements but they have a few differences. i think, so far, the only website to have been successfully sued for being inaccessible was the one for the sydney olympics.

the whole 'a'/'aa'/'aaa' compliance thing is a bit of a joke when you get into the nitty-gritty of it - beware. they're just points along a scale of adherence to accessibility conventions, and even those conventions and assumptions change over time. it's very easy to hack a website into adhering to aa, and quite often aaa, and for it still to be a donkey. web design and web accessibility are the same thing, and websites should be built from the beginning with accessibility and usability in mind as much as style and content.

 

[Denny Oubidoux]

spfbuiaasfd. im sick of this shite now. why does there have to be disabled people? (only messing). Thanks beetle...