Direction of Pidgin development

[Dale Worley]

On Thu, 2007-05-03 at 23:16 -0400, Andrew Roeder wrote:

It sounds to me that you are incorrect in a superficial way, and correct
in a deep way.  Keeping those two facts straight is the way to make
progress.

> After reviewing a lot of the complaints coming with 2.0.0 and its releases 
> and the solutions that are/will be offered by the developers.  Am I wrong is 
> the thought process that it seems Pidgin is moving towards a focus a 
> userbase of more un-intelligent, uninformed, (if you'll forgive the bad 
> analogy) Window's type users.  Catering to people who don't know what their 
> software does, how it works, or what it can do.

The bitter reality is that if your software is going to be widely
adopted, the average user isn't going to be very knowledgeable.  Some of
them are stupid, of course, but most of them just don't have the time to
study up on the software, especially on how it works.  As a rule of
thumb, every 10 IQ points you shave off the threshold needed to operate
the software quadruples your market share.  Or let us say, if Pidgin is
unsuccessful, its user base will be a few dozen developers and their
friends, but if it is successful, its user base will be hundreds of
thousands (millions?) of high-school students.

        Computer-literate people have highly complex mental models of
        how
        computers and operating systems do what they do, and geekly
        conversation is made up mostly of an exchange and comparison of
        these models.
        
        Users do not want a complex mental model of their dishwasher and
        fiercely resent attempts to instill one.  It makes them feel
        put-down for not having one and yet, at the same time, they know
        they neither want nor need one.  All they want to know is what
        to set the knob to. They are appliance users and they want the
        computer to act like an appliance.  [...]
        
        As Tim Hunkin points out in his TV series, "The Secret Life of
        Machines," an appliance interface is the most difficult and
        sophisticated interface that can be constructed.  To allow
        people
        accomplish a complex task without possessing a mental model of
        the
        process is extremely difficult.
        -- "Ask Mr. Protocol" by Michael O'Brien
        
> Back the point of this, with, users wanting to fulfill Pidgin's potential 
> now have to drudge through plugins and popup a window to set individual 
> plugin settings, this is not "easier", hiding these options in such a 
> fashion protects the ignorant users but merely hinders the time it takes to 
> explain/setup Pidgin yourself, or for someone else who is looking for these 
> features.
> 
> I for one feel no loss of usability with Pidgin yet, but I do feel that I'm 
> using a program no longer actually built for me, it is built for the "dumb 
> wanting easy" which is what every opponent messenger seems to target.

The deep problem is that the plugins are affecting the configuration UI
in a way that makes it harder to navigate.  Or you perceive that, but in
this sort of issue, the perception *is* the reality.  That suggests that
not enough study has been put into the various subordinate windows, how
they are organized, and how they interact with each other.  It's a tough
task (but vital) that when a plugin is added, its interface elements get
integrated into the interface in a way that makes deep sense relative to
the logical structure of the interface.  I suspect that right now, the
added interface elements are organized based on which plugin supports
the element, which may not correspond to the overall *logical* structure
of the settings.

Dale