[Sigia-l] Site Counters - good or bad

[d.brodale]

On Wed, 4 Jun 2003, John O'Donovan wrote:

> [Site counter are] up there with having comments such as:
>
> "This homepage was last updated on 12/11/1999"

Here, I would disagree on ranking, as the "freshness" of Web documents in
any number of cases is of clear significance to the end-user. Moreso that
the number of visitors that may or may not have "hit" a given document -
both in terms of data clarity and task relevance (date vs. an
undifferentiated "count") and the number of applicable instances.

Although I agree wholeheartedly that visible site counters usually
advertise lack of traffic more than anything else to end-users and note
how freshness dating could similarly devalue content presentation, I don't
believe in the latter case it negates the utility of document dating in
many cases. I would suggest that managers of stagnant sites that offer
self-dating documents ought to either halt the practice or simply take
their lumps. This is not to say that all documents need to be explicity
date-tagged, but rather that the practice often deepens the context in
which a given Web document happens to be placed.

Given the prevalence of server-generated ("dynamic") content generated
upon demand or delivered with improper response headers, conscientious,
in-document self-dating is commonly the most reliable means of conveying
temporal relevance to end-users. In an ideal world, such information could
be conveyed to browsers for display upon demand within the requesting
application's UI (to clear page clutter but offer users access to the
data), but it just isn't so and Web authors would likely treat any such
mechanism as they do response headers now: sloppily.

-don