[draq]

I think this is another example of the QWERTY effect. Obviously, there are many pieces of word processing software that are superior, but he is so used to his system that, to him, the transaction cost outweighs the opportunity cost.

[GuiA]

Superior in what way?

It seems that all he cares about is writing text, period. A typewriter is not good enough because editing is a pain, but Wordstar 4.0 seems to fit his needs perfectly. How is there anything wrong with that?

A lot of software is novelty-driven, because you need to sell your users on the next version. Sadly, this is often at odds with the goals software should aim to fulfill as a tool. Did we ever need a v2 for the hammer, or drafting pencil?

[dragonwriter]

> Did we ever need a v2 for the hammer, or drafting pencil?

Uh, yeah, both the pencil and the hammer not only have gone through multiple iterations over history, but have a branching family tree with real improvements for specific roles over time.

Sure, there's a lot of fluff in software (and all kinds of other products), but technology actually does bring improved tools.

[stcredzero]

A lot of software is novelty-driven

Some of it is driven by available technology. It would have been very hard to have a great spell-check functionality on early home computers, like my 48k Apple II+, simply because of the hardware.

Did we ever need a v2 for the hammer, or drafting pencil?

Hammers did indeed evolve quite a bit from their stone age original form, though the general shape and principle have remained the same.

There is a modern claw hammer halfway up the left side in this illustration from 1514:

http://en.wikipedia.org/wiki/File:Melencolia_I_(Durero).jpg

Drafting pencils have been around from the late 1500s:

http://leadholder.com/main-history.html#leadholder_origins

It's arguable that truly modern ones didn't appear until the 20th century, however.

[grayclhn]

Just to be a dick: yes, drafting pencils that are deliberately shaped to not roll off tables are nice, and probably not the absolute first version produced.

[chops]

I dunno. Did the first hammers have a claw? Because that's a pretty handy innovation.

[GuiA]

There have been clawed hammers since at least 1514: http://en.wikipedia.org/wiki/Claw_hammer

Clearly, innovation is good - the modern (1514) hammer is better than the big stone our Homo Abilis ancestors used. But at some point, you reach an upper bound.

[mpyne]

It was a separate innovation, followed soon after by a different innovation to combine a claw tool and a hammer tool into a combined tool.

But there are other things you can put on the other end of a hammer besides a claw, none of which made the original hammer a bad idea if all your problems really are nails that you don't need to claw back out.

[jmcphers]

There may be instances where people stick with something inferior because of inertia or transaction cost, but QWERTY doesn't really deserve to be the poster child for this phenomenon; the often-cited example of market failure in the Dvorak case is not what most folks assume it to be.

Well worth a read (1996): http://reason.com/archives/1996/06/01/typing-errors

[jblow]

"Obviously"? Really, is it obvious? You aren't drawn to question the supposed superiority of this software even for a second?