Less readable to be more accessible?
September 6, 2004 at 6:51 PM
So I read a post about this new standard being formed for text email newsletters. It is designed to increase usability and accessibility of plain text email newsletters. It's called Text Email Newsletter Standard (or TEN for short). I thought it was a worthy cause (hey, anything for improved usability and accessibility) so I looked into the details. After reading through the summary and specs I decided I'd take the last newsletter emailed to the Digital Web Magazine readership and convert it to what it would look like using TEN. Well, I quickly came to the conclusion that while this standard may make the newsletters more usable and accessible to screen readers, it certainly made it far less usable to standard email readers, much less readable to actual people who are reading the emails. Make no mistake, this is a good effort and a worthy cause, but I think the implementation is too far from practical. What are your thoughts on this? [from 456 Berea Street]
unless i'm missing something, it doesn't make it more "usable and accessible to screen readers". screen readers treat it like any other piece of simple text. it's not that it triggers any special behaviour in the assistive technology. what it does do, however, is define a format that users of screen readers can listen out for...basically, a way of codifying semantic meaning in an inherently unstructured file format (plain text). if you feel it's making it less readable, you could offer it as an alternative that the user can select. most email newsletters give you the choice between HTML and text; you could extend it to include HTML, text and TEN formatted text (with a helpful little link explaining what TEN is all about). let the users decide if they find it more or less readable and usable.
The thing is, even a properly formatted HTML document-- i.e., one that uses structural markup that defines the purpose, not just the look, of each element-- would be accessible to screen readers (with appropriate style sheets to make it look good to sighted users, of course). There are some e-mail newsletters I get where that is, in fact, the case. Unfortunately, that's not the case for a majority of them. In the worst case-- and we're talking major, million-dollar e-commerce sites here!-- I've seen HTML e-mails that consisted solely of hotlinked images with no alternative text.
I'm a user of both Windows and Linux, as well as portable devices using WinCE and Palm O/S. When I am connected to my high-speed connection in my office, I won't get upset over receiving an HTML-formatted email, but still prefer receiving text-based emails by choice. Why? Because, with a text-based email (speaking in terms of subscribed newsletters) I can quickly scan the topic headings within and choose to click on a link or move on to whatever other mail I have. When I am traveling and having to rely upon 56K dial-up connections, waiting for HTML-formatted email to finish downloading is intolerable. Then there is the dangerous aspect of HTML in emails: viri and dangerous scripting. Linux users can view HTML-formatted, but my experience is that most filter any emails with HTML content to the their trash bin or delete at the source. There are millions of users of Linux: a number quickly growing as each day passes. Many Windows users are being taught to do the same thing. How effective is HTML-formatting when those you wish to target or reply to are trashing your emails before they take one look at it? This is too much fuss over what doesn't matter. HTML in email is effective, but is not acceptable to many viewers. It's inherently dangerous, easily exploited, a bandwidth hog, and depends upon an outside source (the sender's servers and external links) to complete its mission. TEN sounds good in theory, but I don't see why learning a new language is better than giving a choice over two that everyone can already understand and work with: HTMl or Plain Text. Or is this merely about increasing convenience to the content provider? Instead of worrying about a universal deployment, stick with the dual-deployment method now used and concentrate efforts on making the content worthwhile. Let the subscriber choose. It worked before. It works now. There's no need to fix what isn't broken.
I've been using the TEN format as a navigation aid for screenreaders in my Web Design Update Newsletter for over two years now. User response to the format has been quite positive. The newsletter even won an award in 2003. Regarding the TEN format, people have commented that they appreciate: 1. "Attention to accessible-friendliness." 2. "Clearly distinguished separate sections and articles." 3. "Summary of contents at the top of each issue." 4. "Spacing between the heading and articles." 5. "Consistency and clear writing." 6. "Organization and format." 7. "The simplicity." 8. "Attention to accessibility issues." To date I haven't received any negative feedback. Web designers and developers are invited to join the webdev listserv and receive the Web Design Update. For information on how to subscribe visit: http://www.d.umn.edu/goto/webdevlist
Paul wrote: "TEN sounds good in theory, but I don't see why learning a new language is better than giving a choice over two that everyone can already understand and work with: HTMl or Plain Text." Paul, first of all it's not a new language, only a shorthand convention. Secondly, it does serve a valuable purpose because plain text is inherently non-structural, and if you have to listen to it or scan it line by line on a braille display, it's difficult to understand what is a heading, etc. You can't tell those users "why don't you just use the HTML version?", because - as well as using assistive technology - they may just as well be on a slow dialup connection.
additionally, Paul: "Instead of worrying about a universal deployment, stick with the dual-deployment method now used and concentrate efforts on making the content worthwhile." did you actually read what TEN is about? it's not meant to be a replacement for HTML emails, but a way to better format plain text emails to still convey a minimal ammount of structure. it's not "universal deployment", merely a (proposed) enhancement to the plain text side of the "dual deployment"...
My comments originally at Andy Budd's post Text Email Newsletter Standard. Here's a quick excerpt: At first I was a bit skeptical, and found it to be a bit clumsy to look at. I
Laura: not to get technical but 1 and 8 are the same (or at least not defined clearly enough), 2 and 5 are the same, 2 and 6 are very simular or at least closely related, 4... isn't this what the point was in 2? Which leaves us with a few points covered and then number 3 which is what most well designed newsletters are already doing. I agree with Derek's comment on Andy Budd's thread... how did the group come to these conclusions? Where are the findings from? Despite Dereks warm optimisim, I believe the + structure is horrible for reading... even if the reader can put a blinder on for them... I mean, we can say the same for ads on web pages... but does that mean they really add value if they user's not seeing them to begin with??
Good point, Nick. I somehow fail to see any extra value in extra markup. Keep your plain text simple and organized and you'll get same benefits without extra pluses and digits. Another confusing point is about sticking to ASCII. What've happened to encodings, not to mention Unicode? The world speaks not English only, you know...
Nick, The 1-8 list is verbatim comments from my newsletter readers. Laura