IMHO: Using side effects to implement the core logic challenges the
readability of the code. (The present example has provided evidence of
that.) Yet, readability is of primary importance in a large group effort.
The opinion of many is that one should choose readability over
economy/cleverness/performance unless the specific tradeoffs in the case at
hand are compelling. One of those "many" is quoted below.
"The main idea is to regard a program as a communication to human beings
rather than as a set of instructions to a computer. "
- Donald Knuth (http://www-cs-faculty.stanford.edu/~knuth/cweb.html)
-J
-----Original Message-----
From: Doug Melvin [mailto:[EMAIL PROTECTED]]
Sent: Tuesday, January 30, 2001 10:51 AM
To: [EMAIL PROTECTED]
Subject: Re: Re[2]: [Dynapi-Dev] [Bug #130357] small typo in list.js
I've just had that conversation with a co-worker.
Can anyone tell me why this is not proper?
The best my co-worker could come up with was 'it doesn't feel right'
....
> Either way, we shouldn't be using if (this.selected=b). That's not
> very proper.
>
> --
> // Robert Rainwater
>
> On 1/30/2001, 8:47:21 AM EST, Raides wrote about "[Dynapi-Dev] [Bug
#130357] small typo in list.js":
>
> > Michael Pemberton wrote:
> >>
> >> please put the "typo" back. What is does is set the value of
this.selected to
> >> b and then evaluates the returned value (the new value of
this.selected). It
> >> is needed to change the selected state of the item.
> >>
>
> >> > if (this.selected=b) {
> >> >
> >> > should be
> >> >
> >> > if (this.selected==b) {
> >> >
> >> > right?
>
> > I will answer both for NS6's sake:
>
> > This is indeed a bug in NS6 and prevents normal execution of the
rest of the
> > code. It should read:
>
> > this.selected=b
> > if(this.selected){
> > ...
> > }else{
> > ...
> > }
>
> > I have done this kind of changes all around my local dynapi
distribution and
> > those pesky errors that NS6 throwed at me and their side effect of code
not
> > executing in that function anymore disappeared. Other special
perso-NS6-ality of
> > this browser is that if you want to dynamically resize an image, you
have to do
> > it TWICE, using a setTimeout to delay appropiately (30 millisecs are
enough) the
> > effect. The code I use to test NS6 comes with this mail. It is spanish
code and
> > HTML, poorly documented but that can be executed in IE4.0 and above with
no
> > changes at all and some errors due to their different underlying model.
>
> > Raides J.
>
>
> ----------------------
> DynAPI Snapshots: http://dynapi.sourceforge.net/snapshot/
> DynAPI Homepage: http://dynapi.sourceforge.net/
>
>
>
> _______________________________________________
> Dynapi-Dev mailing list
> [EMAIL PROTECTED]
> http://lists.sourceforge.net/lists/listinfo/dynapi-dev
_______________________________________________
Dynapi-Dev mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/dynapi-dev
_______________________________________________
Dynapi-Dev mailing list
[EMAIL PROTECTED]
http://lists.sourceforge.net/lists/listinfo/dynapi-dev
