home *** CD-ROM | disk | FTP | other *** search
- Newsgroups: comp.lang.tcl
- Path: sparky!uunet!eco.twg.com!twg.com!news
- From: "David Herron" <david@twg.com>
- Subject: Object orientedness & The Customer (tm)
- Message-ID: <1992Aug18.170731.12063@twg.com>
- Sensitivity: Personal
- Encoding: 48 TEXT , 4 TEXT
- Sender: news@twg.com (USENET News System)
- Conversion: Prohibited
- Organization: The Wollongong Group, Inc., Palo Alto, CA
- Conversion-With-Loss: Prohibited
- Date: Tue, 18 Aug 1992 17:10:23 GMT
- Lines: 53
-
- The recent discussion about the limitations of TCL have been interesting.
- I've got to agree with most of the points. From an engineering perspective
- that is.
-
- However my point of view is not one which looks on this as a research
- proposition. Instead the concern is to make applications where the
- customer is not limited by the vision of the people selling the
- software. Or, as Sean said yesterday, to raise the least common
- denominator of software.
-
- But whatever I do with it must be usable by some version of the
- stereotypical End User. After all, we survive here on what we sell,
- not on how many papers we write ;-).
-
- As an engineer the comments about how an `object' paradigm to programming
- helps to control complexity are right on the mark. It puts into focus
- some of the problems I've had with my own TCL applications. So far I've
- been solving these with careful modularization of data & procedures,
- but there's only so far that can go.
-
- ..BUT...
-
- If one is trying to sell an application with an embedded configuration/extension
- language built in, you have to expect The Customer to play with it. Not
- all will, but some will. And the more who do play with it (to an extent)
- the better. Having an embedded extension language does not limit my responsibility
- for creating a good application, but regardless of how good a job an engineering
- team does it cannot predict all the things the customer will want to do with
- the software they build. So it is a win/win situation to write the front end
- using an interpreted language and let the end user be able to see and play
- with the front end if s/he wishes.
-
- For this to be of any aid to The Customer they must be able to understand
- whatever language we choose to use. Throwing an Object and Class paradigm
- into the language seems to raise the amount s/he must learn before being
- able to play with the applications front end. But that's just `intuition'
- and some user-factors research might prove me wrong. Especially if it was
- presented well with something to make the class structure obvious.
-
- I'm open to persuasion here ...
-
-
- BTW, I have to agree with most of Sean's comments. TCL/TK has been a breeze
- to work with and I am 100% sold on interactively and iteratively building
- the front end to applications. Now to sell the rest of my group ;-). Or
- maybe find a job somewhere that a new project is starting and they appreciate
- the value of the things I've touched on up there ... ;-)
-
-
- <- David Herron <david@twg.com> (work) <david@davids.mmdf.com> (home)
- <-
- <- "Just because I recreate the middle ages on weekends doesn't mean I have to
- <- at work." -- me
-