CHI-WEB Archives

ACM SIGCHI WWW Human Factors (Open Discussion)

CHI-WEB@LISTSERV.ACM.ORG

Options: Use Forum View

Use Monospaced Font
Show Text Part by Default
Show All Mail Headers

Message: [<< First] [< Prev] [Next >] [Last >>]
Topic: [<< First] [< Prev] [Next >] [Last >>]
Author: [<< First] [< Prev] [Next >] [Last >>]

Print Reply
Subject:
From:
Ash Donaldson <[log in to unmask]>
Reply To:
Date:
Sat, 17 Apr 2004 12:05:19 +1000
Content-Type:
text/plain
Parts/Attachments:
text/plain (126 lines)
I think "It depends." needs to be an opening sig for all my posts.

Quoting Marty Landman <[log in to unmask]>:
>What drop downs do imo when used judiciously is
>
> 1. reduce the probability of error when compared to the more common input
> text field which is a tabula raza if ever there was one

In the case of an uncommon operation (e.g. web user filling out a typically
designed form - focus on content), this may reduce errors.  In the case of a
repetitive operation (e.g. a data entry operator processing numerous forms -
focus on speed), this may tend to cause errors.

> 2. imo they may also answer questions of what constitutes valid data in
> case the visitor isn't sure cf. the old help desk joke about "where's the
> 'any' key?"

A well designed form will be unambiguous as to what data is required.  If there
are common permutations of data format for the input, they should be handled by
business logic.

> 3. dropdowns also offer the handy capability of representing the data
> differently as the program gets it compared to what the visitor sees, e.g.
> a country dropdown may list
>
> United States
> Zimbabwe
> Mexico
>
> but what the program gets in response to these options would more likely be
>
> US
> ZW
> MX
>
> which is something better done in the front end than get processed in the
> back end IMO.

Why only look at front and back when there's a middle? ;)  Again, the correct
business logic could accept and parse America, US, U.S., United States, USA,
U.S.A, etc.

> >again, I ask: "best for whom?"
>
> IMO it's best for the casual user. I would guess that in a data entry
> situation it'll slow folks down.

Yes.  Remember though, this is also dependent on the target market.  Being a
little extreme here, but could your "casual users" be predominantly bank tellers
that do data entry all day and are used to a common industry standard for data
input?

> >That was the problem with the form you cited -- implementing a drop
> >down is not the only solution.
>
> Please what are the other approaches that are possible in these situations?

What about separated text fields (separating year, month and day), or an open
text field?  Of course, both would require the appropriate validation and
business logic.

> >With clearly marked fields (such as above) ... what are the data on
> >user input error?
>
> That's perhaps the lion's share of the consideration, at least for a well
> travelled site. But I don't recall any discussion yet of the more mundane
> consideration of programmer's time, effort, and frustration. Anyone who
> hasn't attempted to code field validation logic should take a few minutes
> to do a flowchart or however they approach laying out logic. I've done this
> stuff almost a quarter century and it is time consuming, frustrating, and
> error prone. Adequate testing is arguably not even part of the design and
> coding of such algorithms and what it all boils down to is that the more
> that software relies on back end data validation the greater the expense
> and higher the probability of bugs in the field.
>
> So in that light I'd say drop downs have the potential of benefiting
> everybody using the product.

I can empathise with you here.  However, a set of standard objects or libraries
for common data input and validation only has to be done once per technology (if
coded correctly) - and the logic behind it, once per object.  Things such as
this example (to handle dates) are common tasks and so should be catered for.

In the end, our job is to do what you lament - to do the up-front thinking and
work to make it easier for our users or customers to achieve their goals.  Why
should they have to be hindered by poorly designed technology?

> >I consider myself an *expert* web user. And I rarely -- **rarely** --
> >try type selection because more than 50% of the time, it doesn't work
> >and the result of trying is more of a pain than simply conforming to
> >the programmer's mindset.
>
> I don't even know what type selection is. Could someone please post a link
> so I can see an example of this technique?

I think Kathy means tabbing to a drop-down, then pressing the first character on
the keyboard repeatedly to get to your selection e.g. since I'm in Australia,
whenever I come across a 'country' drop-down, I tab to it, and hit the "A" key
repeatedly as it selects: Afghanistan, Albania, Algeria, American Samoa,
Andorra, Angola, Anguilla, Antarctica, Antigua and Barbuda, Argentina, Armenia,
Aruba, Australia.  Of course, I could just hit it 13 times - but for some lists,
it's 3, some it's 8, some it's 13... :P

> As for mindset conforming, if there is to be communication this is a
> necessity. When the communication is to a computer then the rules become
> either totally strict or, ahem, reliably unreliable. Speaking as a
> developer though my mindset on a job ought to conform with the visitor's
> rather than vice versa. That's a vital part of what I'm being paid for.

"It depends."  It depends on how much consideration and effort is put into the
design up front.

Best regards,

Ash Donaldson
"It depends."
User Experience Designer

    --------------------------------------------------------------
        Tip of the Day: Email mailto:[log in to unmask]
               with any comments, questions or problems
     CHI-WEB: www.sigchi.org/web POSTINGS: mailto:[log in to unmask]
              MODERATORS: mailto:[log in to unmask]
       SUBSCRIPTION CHANGES & FAQ:  www.sigchi.org/web/faq.html
    --------------------------------------------------------------

ATOM RSS1 RSS2