W3C home > Mailing lists > Public > w3c-wai-ig@w3.org > January to March 2002

Re: FAQ - where are the FAQs for this list?

From: Al Gilman <asgilman@iamdigex.net>
Date: Fri, 22 Feb 2002 13:24:45 -0500
Message-Id: <200202221824.NAA142286@smtp2.mail.iamworld.net>
To: "Charles F. Munat" <chas@munat.com>, WAI IG <w3c-wai-ig@w3.org>
At 11:31 AM 2002-02-22 , Charles F. Munat wrote:
>I'm half in agreement with you, Al, but only half.
>

Yes, this is a point worth looking on both sides of.

I actually come up with something more like 60% agreement in my current estimate, based on the following rough guesses:

Munat would balance the FAQ somewhere around 15% street language, 85% tech lanaguage.

Gilman would balance the fAQ somewhere around 55% street language, 45% tech language.

In my calculator that is 15 + 45 = 60% of the emphasis that we would allocate alike, and the other 40% lies in disputed territory.  YMMV.

>Yes, the FAQ should point out that alt attributes are often -- 
>erroneously -- referred to as "alt tags." And it should explain why they 
>are *not* "tags" (and they are *not*). But it should then try to 
>encourage use of the correct terminology.
>
>I am not saying this because I believe that we should be prescriptive 
>about terminology (like that would work!). Nor do I disagree that "tag" 
>-- in a wider sense -- can be applied to the alt attribute. I am saying 
>this because I've spent a lot of time teaching people to write quality 
>HTML and I've found that confusion between elements and attributes is 
>widespread. It does not help when we apply terminology so loosely.
>
>For most people (IMO), this is a tag: <tag>
>

But not for most people _who need the FAQ_.  The FAQ should be tilted in favor of the newbie.

>This is not: attribute=""
>
>Remember also that the FAQ is really an introduction to the WCAG, where 
>terminology is (usually) used more precisely. It should be warming 
>people up for that.
>


This view should not be accepted without some question.  Something like 80% of the people who consult the FAQ should get what they need there without recourse to the definitive, technical stuff such as the guidelines Recommendation.  If it's working, it is more self-contained than dependent on the long form, in terms of successful trips through it.

People who apply the FAQ advice with success will come back and read more deeply.  This is your audience for technospeak.  Not first-time visitors to the FAQ.  We need to be able to deliver the goods with the minimum detour into technospeak to make the FAQ fill its highest and best role.  [But on the other hand, see what I do in the example below.]

So I stand my ground.  The FAQ should be majority invested in solving the actual problems that people find they need to ask about, as quicly as we can.  And minority invested in inducting them into the corps of the cognoscenti.  [Note I thump this point a little extra hard because _my_ explanations do tend to run on and it takes me several passes to get anything short.]

Form fields in tables is a good example.  The technical situation is a long, sad story.  But the FAQ entry needs to start short and sweet:

Q: How do I make form fields laid out in a tabular form accessible?
A: Us the TITLE attribute on the INPUT element itself.

The preponderance of FAQ-consulters will get what they need from that.  We need to write in a journalistic style where there is a succinct summary and the rest of the story is optional.

Al

More:  For a concrete example of what I did to try to make a FAQ effective, consult the time machine trip below.  The information is outdated, but the tone and flow are representative of principles that I expect would be effective. (long URL warning)


http://web.archive.org/web/19981203154220/http://www.access.digex.net/~asgilman/lynx/FAQ/Als_picks.html

>So I agree with Ineke. While the term "alt tag" should be mentioned 
>clearly in the FAQ (and perhaps in multiple places), the difference 
>between attributes and elements should be clearly explained also, and 
>the term "alt tag" should be clearly labelled as imprecise.
>
>(One benefit to using "alt tag" somewhere in the FAQ is that people 
>searching for "alt tag" will also pull up our FAQ!)
>
>Charles F. Munat
>Seattle, Washington
>
>Al Gilman wrote:
>> At 03:34 PM 2002-02-18 , Charles F. Munat wrote:
>> 
>>>2. It would be nice if we were careful to use correct terminology. For
>>>example, I ran across a couple of references to "alt tags." Here is an
>>>opportunity to correct those misaprehensions.
>>>
>>>
>> 
>> While this point is likely to gain consensus backing on _this list_ this illustrates why this list is not recognized in the workplan of the WAI as a consensus body.
>> 
>> This is a gold-plated bad idea for FAQs.
>> 
>> One of the key principles in FAQs is to respect and bond with the street language by which people with questions express their questions.
>> 
>> In this case, "ALT tags" is the incumbent street language, and there is no reason to try to 'correct' it.  The ALT attribute is a 'tag' in a natural language sense that is more important to bind to in the FAQs than to fight against.  Yes, the FAQ should explain that to find this question in the technical literature one will need to use the language "ALT attribute" so that when the reader searches for more information they can dig up the technical discussions as well as those who have taken the trouble to meet real people on their own terms.
>> 
>> In practical terms there is no problem with "ALT tags" because if you Google for "ALT tags," you get Alan Flavell's excellent discourse that explains the technical niceties without talking down to the reader.  If you search with this street language you get better-filtered tutorial literature than if you use the technically 'correct' dialect.  Go figure.
>> 
>> Al
>> 
>> For more info:
>> 
>> read why RTFM.mit.edu was insanely great.  Archived at
>> 

>>   http://www-unix.gridforum.org/mail_archive/gce-wg/msg00134.html
>> 
>> 
>> 
>> 
>> 
>> 
> 
Received on Friday, 22 February 2002 13:24:39 GMT

This archive was generated by hypermail 2.2.0+W3C-0.50 : Tuesday, 19 July 2011 18:14:00 GMT