Re: measuring the cost-benefit analysis of standards compliance

On Wed, Jul 10, 2002 at 11:26:24AM -0500, Austin Govella wrote:
> 
> I've been trying to think about how we could quickly and efficiently come 
> up with some figures to use in such a CBA. I don't think it's practical to 
> expect a good number of designers to measure their time designing and 
> maintaining non-compliant websites and then to compare that with the time 
> spent on compliant websites.

This occurred to me - and to some extent it depends on the kind
of questions one asks. For example, "What proportion of your time per 
week do you spend on (such a such a task)?" is possibly a better
question than "How many hours do you spend ...".

It is easier to get a feel for the subdivisions of time rather than 
exact number of hours.

I do not know if it's reasonable to only target designers.
The cost of a project is spread over more than just designers,
so if there's a project manager, that would be the best person
to be asking these questions (one hopes). 

Thoughts ?

> While driving to work today, it occured me, we can use his equations for 
> "chart junk" (the amount of unnecessary clutter in a graphic) to determine 
> actual numeric ratings for "code junk" in a website.
> 
> Compliant websites have less code junk, and it would be simple to connect 
> the amount of code junk on a page, and the time required to edit, maintain,
>   design, or serve that page over a server. And time is money.
>
> Maybe a basic equation like this (and this may be way wrong. I'm just 
> brainstorming here, and I have no background in math or statistics):
> 
> actual weight of content / mark-up = code junk rating
> What do you think? Bad idea? Misapplied? I'm curious for your responses.

Very interesting idea.  How would you measure/establish the actual weight
of content and the mark-up (hm, what does this mean?)?

If we can assign a code-junk rating to a site, it might be an 
'encouragement' factor for someone to overhaul their site.

Hehe, we still need to know the average time taken to edit/maintain/
design. :) If we have statistics from the metrics I have been proposing,
this should be achievable.

I think you can only calculate an average time to work on a page
for a particular site, because it may depend on what technology
they use, how the site is structured - dynamic content or not? 
existing CMS? and so forth.  I don't know if this is a reasonable 
assumption or not!

I'm very open to ideas. :)

cheers,
-steph
random web dudette
http://unadorned.org/

Received on Wednesday, 10 July 2002 18:55:55 UTC