> On 02/15/[masked]:21 PM, Mark Nyon wrote:
>> Looking through this thread, a thought comes to my mind; are the
>> days of the Requirements Document gone? Between the Original
>> Poster's request, and sharing stories with other developers in the
>> field, it seems that when it comes to a lot of web
>> sites/applications, the cart keeps getting put before the horse,
>> and no one's applying the 5 Whys or another technique to determine
>> the reasons for the decisions made before selecting a technology
>> solution, or a particular path.
>> I'd love to hear that my experience is anomalous, but I worry that
>> it isn't.
Perhaps some relevant thoughts on this very subject:
Why we should build software like we build houses
A big issue in what is passing as technology products
lately is that the 'inventors' are operating in a very
improvisational manner; no body wants to lay down real
specs on something that is being invented as it comes
out of their mouth as a description. This is a LOT of
what I am seeing when it comes to app development, for
You get somebody who has an idea about 'market disruption'
in some retail field, attempts to get their head around a
bunch of features they think are going to make them another
Zuckerberg, then start spouting this list of constantly changing
and varying features to 'some developer' so they can implement
their sometimes ill-defined dream. Fun.
I also think it's cultural, perhaps even generational.
When I first worked in hardware design, I was trained by
military types, real 'mil-spec' types with backgrounds
in space certification, that kind of thing. EVERYTHING was
predefined, the types of tools, how to assemble, how to
re-wire, how to dress wires, how many wraps in a wire wrap
(yes, that far back).
You definitely need to approach things in hardware design
a fair amount with less improvisation than software because
hardware is more difficult to 'redo' ad hoc than software.
However, even software design today sees a lot of 'make it
up as we go along' think.
Maybe this has to do with the cost involved. Real, hard, up
front engineering is a big part of the cost to engineer
something. I hear this a lot from friends and clients building
technology for medical and bio med systems (robots, medical
But I also think that the 'drop out of school and start a startup
instead' culture seems to implicitly or even explicitly say
'old ways don't work, don't listen to the man, do things
/this/ new fangled way and live a better life.'
The thing is, when I sit down with this kind of engineer,
I often get the sense that they are far less prepared to
analyze things in any depth and tend to fall down more when
subtle effects emerge in an engineered system.
Well, in any case, there you go...