Subject:
|
Re: Track Designer Registry
|
Newsgroups:
|
lugnet.trains
|
Date:
|
Wed, 13 Mar 2002 00:59:04 GMT
|
Viewed:
|
2051 times
|
| |
|
|
"James J. Trobaugh" wrote:
>
> I too had though about a complete re-write, but it's so hard to get
> motivated to take on the project when you see how much stuff is packed into
> the existing TD. First I make a list of what I'd like to have in such an
> application, then I list what the current TD can do. Quickly the reqirements
> list grows to a point that I quickly realize that there wouldn't be much
> time for building.
Personally, I think a rewrite would be a bad idea, although some
reorganizing of the code might be good. (keep in mind I haven't seen
the code)
Typically what happens in largish software projects is they go along for
a few years, adding various features and fixing various bugs, and the
code ends up getting twisted around for various logistical reasons.
Then someone looks at the code and says "I could make this a lot
cleaner/better if I just rewrite it."
However, what that person is forgetting is that the original code is all
twisted around for a reason. In this case, the code is already at
version 2.0, has had a lot of features added into it, and probably has
many bugs ironed out of it. If someone was to try to rewrite all that
code they would spend all their time just trying to reinvent the
existing features, and in the process introduce a bunch of new bugs, or
worse, reintroduce the same bugs again.
If a new person is taking over the code base, it is usually a much
better investment of time for that person to figure out how the existing
code works rather than rewriting the code.
See this web page for a really good description on this (how netscape
screwed up with their rewrite of netscape 6):
http://www.joelonsoftware.com/articles/fog0000000069.html
Here is my suggestion, if Matthew would be willing: Release the source
publicly under an open source license such as the GPL. That way the
source would live on forever, and individual people could work on the
individual features they want. Then everyone would benefit from
anyone's work on the project, and all the work would not be one person.
Ideally the code would be put somewhere central, like sourceforge or
something, and a couple of people (maybe Matthew or someone else who
knows the code already) would be responsible for accepting new features
into the original code. That way there is someone to control what
changes get in the official version, but the code is still available to
everyone to play with and add new features.
Again, its really up to Matthew as its his project. He may want to take
it commercial someday. But many people would love to see the source
publicly available, it would be great for the community, and track
designer would become an even more amazing tool because of it.
Kevin
|
|
Message is in Reply To:
 | | Re: Track Designer Registry
|
| I too had though about a complete re-write, but it's so hard to get motivated to take on the project when you see how much stuff is packed into the existing TD. First I make a list of what I'd like to have in such an application, then I list what (...) (23 years ago, 11-Mar-02, to lugnet.trains, lugnet.cad)
|
52 Messages in This Thread:
 
 
- Entire Thread on One Page:
- Nested:
All | Brief | Compact | Dots
Linear:
All | Brief | Compact
|
|
|
|
