Comments on: Accessibility First – A Novel Teaching Method http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/ Working together for standards Wed, 27 Mar 2013 12:19:03 +0000 hourly 1 http://wordpress.org/?v=3.3.1 By: Max Design - standards based web design, development and training » Blog Archive » Some links for light reading (21/3/06) http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-478 Max Design - standards based web design, development and training » Blog Archive » Some links for light reading (21/3/06) Sat, 29 Apr 2006 05:29:17 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-478 [...] Accessibility First - A Novel Teaching Method [...] [...] Accessibility First – A Novel Teaching Method [...]

]]>
By: Krishan Patel http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-161 Krishan Patel Thu, 23 Mar 2006 20:12:51 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-161 I don't see what is new in this... part of developing with standards is the approach of graceful degredation or progressive enhancement -- i.e. design your HTML (linear flow etc) first, then add CSS, and then add JavaScript... So, why is this revelation new??? I don’t see what is new in this… part of developing with standards is the approach of graceful degredation or progressive enhancement — i.e. design your HTML (linear flow etc) first, then add CSS, and then add JavaScript…

So, why is this revelation new???

]]>
By: Bruno Girin http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-145 Bruno Girin Wed, 22 Mar 2006 06:36:56 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-145 In my experience, it depends on the purpose of the web site or web application and to a certain extent the technology used. For instance,the area I work in is primarily web enabled applications. 90% of the work is Java server code (J2EE) that needs to connect to any number of back-end systems. The web interface is only one channel among many to access the system (others can include web services, messaging systems such as MQ-Series, file upload, etc.) As a result, when we come to the web interface, we are building a view that can represent and update the data we have in the system. And that data can be structured in extremely complex ways. So it is natural for us to design the interface around semantic data markup and efficient navigation. The design can come later and that design usually has to abide by very strict rules and cannot mess with the markup. At the end of the day, if it's not visually pretty, it doesn't matter, we're talking banking applications after all. So in practice, the typical approach, as I am doing on my current project, is to start by producing a front-end that only has little or no styling. In my case, I currently have 5 rules in my CSS: one for the body tag and one for each of the 4 structural div tags to define their positionning. Once the most complicated screens work as expected, we can start thinking about real styling. This sounds like an ideal situation to take into account accessibility and standards from the very beginning of a project. And I am hoping to do so on the current one. The caveat though is that in most situations, the developers on a project like this are Java/J2EE developers who only have a vague notion of HTML and are not aware of front-end issues. So, if the person who builds the front-end is one of the back-end developers, you usually end up with HTML tag soup; if the front-end is produced by web designers, they usually have a very hard time getting up to speed with all the custom tags required by the JSP/Struts/JSF pages they are dealing with that hook up the front-end to the back-end processes and you end up with designs that break when the back-end system unexpectedly returns data that is unusually complex. Now, I can also see why you would design a web site the other way round: starting with the visuals and fitting in semantics as an afterthought. I reckon this is more typical of sites that deal with unstructured data, such as weblogs, news or marketting sites. The main problem with that sort of sites is that the data you are trying to display has little structure or has a structure that is subjective and depends on who is writing the content. For instance, two people writing a news article will have a different style and a different way to structure the data contained in their article. As a result, it can be difficult to build semantics that will ensure good accessibility. All this to say that teaching wanabee web developers about semantics of the data they want to display before letting them loose on Photoshop is a good idea. In the long run, this should produce better designs because they will be based on an understanding of the data rather than a partial view of it. As someone famous said before me: the map is not the terrain, it is only a visual representation of it. If you understand the terrain (data and its semantics), you will be able to produce a better map (visual representation) of it. In my experience, it depends on the purpose of the web site or web application and to a certain extent the technology used.

For instance,the area I work in is primarily web enabled applications. 90% of the work is Java server code (J2EE) that needs to connect to any number of back-end systems. The web interface is only one channel among many to access the system (others can include web services, messaging systems such as MQ-Series, file upload, etc.) As a result, when we come to the web interface, we are building a view that can represent and update the data we have in the system. And that data can be structured in extremely complex ways. So it is natural for us to design the interface around semantic data markup and efficient navigation. The design can come later and that design usually has to abide by very strict rules and cannot mess with the markup. At the end of the day, if it’s not visually pretty, it doesn’t matter, we’re talking banking applications after all. So in practice, the typical approach, as I am doing on my current project, is to start by producing a front-end that only has little or no styling. In my case, I currently have 5 rules in my CSS: one for the body tag and one for each of the 4 structural div tags to define their positionning. Once the most complicated screens work as expected, we can start thinking about real styling.

This sounds like an ideal situation to take into account accessibility and standards from the very beginning of a project. And I am hoping to do so on the current one. The caveat though is that in most situations, the developers on a project like this are Java/J2EE developers who only have a vague notion of HTML and are not aware of front-end issues. So, if the person who builds the front-end is one of the back-end developers, you usually end up with HTML tag soup; if the front-end is produced by web designers, they usually have a very hard time getting up to speed with all the custom tags required by the JSP/Struts/JSF pages they are dealing with that hook up the front-end to the back-end processes and you end up with designs that break when the back-end system unexpectedly returns data that is unusually complex.

Now, I can also see why you would design a web site the other way round: starting with the visuals and fitting in semantics as an afterthought. I reckon this is more typical of sites that deal with unstructured data, such as weblogs, news or marketting sites. The main problem with that sort of sites is that the data you are trying to display has little structure or has a structure that is subjective and depends on who is writing the content. For instance, two people writing a news article will have a different style and a different way to structure the data contained in their article. As a result, it can be difficult to build semantics that will ensure good accessibility.

All this to say that teaching wanabee web developers about semantics of the data they want to display before letting them loose on Photoshop is a good idea. In the long run, this should produce better designs because they will be based on an understanding of the data rather than a partial view of it. As someone famous said before me: the map is not the terrain, it is only a visual representation of it. If you understand the terrain (data and its semantics), you will be able to produce a better map (visual representation) of it.

]]>
By: Richard Morton http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-135 Richard Morton Tue, 21 Mar 2006 12:20:42 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-135 I would agree with Keri Henare that these are such basic concepts that they ought to be the only way of starting out design and development. Unfortunately, it isn't the common approach that it should be. One of my bugbears is that schools and colleges etc. don't seem to have grasped this idea yet (and that is the ideal place to start if change is going to filter through). Does anyone have examples of Schools or Colleges where they do work with web standards, and accessibility in mind? I would agree with Keri Henare that these are such basic concepts that they ought to be the only way of starting out design and development. Unfortunately, it isn’t the common approach that it should be.

One of my bugbears is that schools and colleges etc. don’t seem to have grasped this idea yet (and that is the ideal place to start if change is going to filter through).

Does anyone have examples of Schools or Colleges where they do work with web standards, and accessibility in mind?

]]>
By: lloydi http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-132 lloydi Tue, 21 Mar 2006 08:41:32 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-132 @Keri - Common approach? Well, it depends who you ask. I sometimes start with photoshop and then move to CSS, sometimes it's a side-by-side approach ( a bit of CSS layout, some graphical additions, more CSS tweaks etc), but I've never built an entire site without having <em>some</em> design element at the start. Geoff mentioned 'white sites' and I've done something similar before by using basic wireframe-style sites and the final polish is added later. This I think is much more common, but the point here is that all development on the site I'm referring to is being carried out almost as if the screen doesn't exist - no UI distractions, get it working for screen reader users first (the whole site) and later deal with the visuals. I don't think there are many people who develop to that extreme. @Keri – Common approach? Well, it depends who you ask. I sometimes start with photoshop and then move to CSS, sometimes it’s a side-by-side approach ( a bit of CSS layout, some graphical additions, more CSS tweaks etc), but I’ve never built an entire site without having some design element at the start. Geoff mentioned ‘white sites’ and I’ve done something similar before by using basic wireframe-style sites and the final polish is added later. This I think is much more common, but the point here is that all development on the site I’m referring to is being carried out almost as if the screen doesn’t exist – no UI distractions, get it working for screen reader users first (the whole site) and later deal with the visuals. I don’t think there are many people who develop to that extreme.

]]>
By: Keri Henare http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-131 Keri Henare Tue, 21 Mar 2006 08:16:08 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-131 I thought that this was a common idea? Accessibility and Semantics are such basic concepts that they're the perfect way to introduce people to webdesign. The idea of learning to do everything one way, then un-learning all that and re-learning everything a different way... it never made sense to me. People seem to accept this method because it's how most of us learnt, but it's not the most logical way to learn. We want to make it as easy for people to learn webstandards as humanly possible. If we teach them one way, then tell them that everything that they know is wrong, you will loose some to 'the dark side'. Make it easy and make it quick. Teach accessibility and semantics as the foundations off all webdesign courses. I thought that this was a common idea?
Accessibility and Semantics are such basic concepts that they’re the perfect way to introduce people to webdesign.

The idea of learning to do everything one way, then un-learning all that and re-learning everything a different way… it never made sense to me. People seem to accept this method because it’s how most of us learnt, but it’s not the most logical way to learn.

We want to make it as easy for people to learn webstandards as humanly possible. If we teach them one way, then tell them that everything that they know is wrong, you will loose some to ‘the dark side’.

Make it easy and make it quick. Teach accessibility and semantics as the foundations off all webdesign courses.

]]>
By: Geoff Pack http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-130 Geoff Pack Tue, 21 Mar 2006 03:09:57 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-130 This is not new. Back in the old days we used to call these 'white-sites'. We'd mock up the site in plain html and get the navigation and all the includes working, then we'd work on the layout templates while the server-side guys got the back-end sorted. Very accessible. But then we'd bugger things up with table layouts, font tags, and image rollovers and headings Nowadays we just slap on a few wrapper divs and ids, and write the CSS instead, and avoid the buggering-up bit. This is not new. Back in the old days we used to call these ‘white-sites’. We’d mock up the site in plain html and get the navigation and all the includes working, then we’d work on the layout templates while the server-side guys got the back-end sorted.

Very accessible. But then we’d bugger things up with table layouts, font tags, and image rollovers and headings

Nowadays we just slap on a few wrapper divs and ids, and write the CSS instead, and avoid the buggering-up bit.

]]>
By: lloydi http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-129 lloydi Mon, 20 Mar 2006 23:16:41 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-129 Just to reply to Brian about the visual aspect. My point really was that having the design ethic turned on its head and the visual element added last, there is every good chance that it may not look all that great. And I strongly believe - as do many others - that we need *more* examples of fully accessible sites that look the business. So I really hope that taking this approach of making it work fantastically with screen readers first and then styling it last works well. I am not sure if I can name the site, unfortunately, nor who's behind it. Sorry about that. Just to reply to Brian about the visual aspect. My point really was that having the design ethic turned on its head and the visual element added last, there is every good chance that it may not look all that great. And I strongly believe – as do many others – that we need *more* examples of fully accessible sites that look the business. So I really hope that taking this approach of making it work fantastically with screen readers first and then styling it last works well. I am not sure if I can name the site, unfortunately, nor who’s behind it. Sorry about that.

]]>
By: BlindSurfer blogt » Accessibility First http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-128 BlindSurfer blogt » Accessibility First Mon, 20 Mar 2006 23:13:51 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-128 [...] Accessibility First Geschreven door Roel om 0u11 Gelezen op de vernieuwde website van The Web Standards Projects. [...] [...] Accessibility First Geschreven door Roel om 0u11 Gelezen op de vernieuwde website van The Web Standards Projects. [...]

]]>
By: Jake Archibald http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/comment-page-1/#comment-127 Jake Archibald Mon, 20 Mar 2006 22:56:40 +0000 http://www.webstandards.org/2006/03/20/accessibility-first-a-novel-teaching-method/#comment-127 Yes, but in my experience templates are usually used by non-developer (or a poor developer) and a copy of FrontPage :( Of course, my experience might not be represent the usual abuse of templates. Jake. Yes, but in my experience templates are usually used by non-developer (or a poor developer) and a copy of FrontPage :(

Of course, my experience might not be represent the usual abuse of templates.

Jake.

]]>