• Post Reply
  • Bookmark Topic Watch Topic
  • New Topic

Generative Techniques (or Tools) and Process

 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Does anyone know if there can be a conflict between the two ? I suppose so , if either out-dates the other.
Most recurring process problems can have a sets of tools that can be used so that no one has to re-invent the wheel. Generative techniques means no intuition is required other than do A , then B but not how A or B is done and can be improved.
So if it no longer is a good solution as the problem has changed shape how does one recognise that and stop to study the problem again ?
Should one study each and every tool to see why you are doing things a particular way?
Or should Process dictate the tool?
I'd be very uncomfortable with the thought that a tool dictated the Process.

"Right Tool for the Job" rather than "Right Job for the Tool".

regards
[ October 16, 2003: Message edited by: HS Thomas ]
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
I think I don't fully understand what you are getting at.
Can you give a real live example?
 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
I think this is an appeopriate live example.
Upgrading from WebSphere 4 to 5 (WSAD). You would expect a significant Process change.
I think I was asking which comes first, ideally , Tool or Process.
A chicken and egg situation.
WSAD, probably , is too big to be be a good example.
I need a smaller example.
regards
[ October 17, 2003: Message edited by: HS Thomas ]
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Originally posted by HS Thomas:
I think I was asking which comes first, ideally , Tool or Process.
A chicken and egg situation.

And the answer is the same: chicken and egg evolved together.
So, ideally, you have a holistic view - identify the problems you want to solve and think about how processes and tools can complement each other to do so.
 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Thanks Ilja.
I am not sure it is that straight forward.

regards
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
I didn't want to imply that it's straight forwared. In fact I think it's quite a complex topic. That's why there isn't a simple answer like "first do x, then do y". There is no cookbook, only common sense and experience.
 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Thanks Ilja.
The common sense needed would have to be very broad.
Experience , too.
By trial and error may be too narrow a approach unless used within XP as
development is in small iterations.
Does XP have anything on this subject?

regards
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
An XP approach would be along the lines of:
Think about the Simplest thing that could possibly solve your problem and works in alignment to the values of as direct Communication as possible and the whole team collaborating. Have the Courage to try it and get Feedback on how well it is working (for example via specific metrics). Adjust accordingly.
 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
for example via specific metrics


Metrics to show user satisfaction ?
I've yet to come across any metrics that covers the whole range of user satisfaction. It's usually quite limited to just bug-free and performance kind of measurements.

Well, I guess the higher-ups would have decided on issues like :
* high cost - there goes the users' Christmas bonus
* uncertain safety - ergonomics
* usage restrictions - poor documentation or too much
* lack of necessity - no particular "must have's" but a few few "good to have's".
* I think it will succeed -

(I pinched this list from a document describing the Segway but I think it applies equally well to software design and development)
XP mitigates all the above, of course.
But then you have to learn a framework (or use the parts that apply )and find the appropriate tool which I suppose can be worked within XP.
regards
[ October 20, 2003: Message edited by: HS Thomas ]
 
Ilja Preuss
author
Sheriff
Posts: 14112
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Originally posted by HS Thomas:
Metrics to show user satisfaction ?

I didn't notice that we were concentrating on user satisfaction...
 
HS Thomas
Ranch Hand
Posts: 3404
  • Mark post as helpful
  • send pies
  • Quote
  • Report post to moderator
Oops! I've made same changes while you were posting, Ilja.
Other than metrics of functions/requirements implemented in as short a time and bug-free as possible wouldn't you agree most processes would be
geared to customer/user satisfaction ? I thought XP would be more so?
Robert Martin in ASD PPP says to use free tools as much as possible but I'm not sure how practical that is on a commercial project. It's okay when learning. OK, he does say "use it until you can demonstrate you have outgrown it".

regards
[ October 21, 2003: Message edited by: HS Thomas ]
 
  • Post Reply
  • Bookmark Topic Watch Topic
  • New Topic