Re: designers as users etc.

["Simon St.Laurent" <simonstl@simonstl.com>]

On 12 Jun 2001 08:25:20 -0700, David Brownell wrote:
> > Sadly (to me), the software frameworks people are designing around
> > markup aren't nearly as flexible.  Developers seem to be writing code
> > that couples tightly to data based on very limited possibilities for
> > what that code might contain, though many larger frameworks offer
> > "escape hatches" where they suspect such things are necessary.
> 
> I think that the early stages of the web have necessarily been about
> the less-techie crowd learning about what can be done, which has
> meant limiting options for much the same reasons that any intro level
> course doesn't dwell on advanced features.

Certainly.  But xml-dev is pretty much the (public) AI Lab of the XML
universe, and I don't see much discussion of these 'advanced features'
here either. 

> The ITS ("Incompatible Timesharing System") was explicitly not an
> intro level effort, so different rules had to apply there.  

It wasn't an intro level effort, but computing's also developed a long
ways since ITS.  You don't need to understand the ins and outs of
bumming assembly language instructions or how to hardwire new
instructions into the system.

I have to say I was impressed by the "learning by watching" story in
Hackers.  Maybe assembler was easier to learn when there were just six
instructions?

> I suspect that
> in context, it was still simpler than the innards of many web sites that
> readers here use on a daily basis.

And it had even less security!

> The design process for ITS was
> in many ways like a single-campus version of many open source
> (or Free software) development projects of today.

Agreed in some respects.  The Internet (they were connected to ARPAnet)
provides a campus-unifying layer of communications today, as does the
fact that we've got lots of computers (I've got five right now) instead
of just one PDP-6.

On the other hand, I find it interesting that they didn't push hard for
ITS to reach a wider audience.  Levy suggests that they didn't find it
an interesting project, and that the world was too hostile for the open
approach they'd taken anyway.  The missionary side of open source didn't
appear until much later.

Another problem I find with many open-source projects is that the
user/designer distinction re-emerges over time.  (Lack of) documentation
and code styles become barriers to entry fairly rapidly.

That's something I like about XML - to a certain limited extent, it's
more self-documenting than code, and it's easier for people to walk up
to it and have a chance of doing _something_ with it.

> When folk take off their "dot com blinders" and stop viewing the
> web primarily in terms of commerce, I think those "escape hatches"
> may become the primary targets of system architecture ... rather than
> unavoidable artifacts of large scale cash-extraction systems.  (Which
> is not to say that flexible systems can't extract cash ... :)

That's what I'm hoping for, though my patience ran out long ago.