Geeks With Blogs
No Fun Intended Shoo! You are debugging the crap outta me!
It has been a long process determining what we have to build for the first release. It is also more complicated than I had originally thought. When I look at a segment, I use the following criteria when evaluating existing os projects:

  1. What product is most popular
  2. What product offers the most functionality
  3. How easy it will be to integrate the product into NTeam
  4. How much work is involved in reproducing the product
  5. What license the product is bound to.
As an example, take NCover using this criteria.
  1. NCover is a good product and has a pretty mature feature set
  2. NCover is probably the best open source code coverage tool available for .NET
  3. NCover easily integrates into a build cycle
  4. NCover is not easily produced.
  5. NCover is under a GNU license.
I am still battling the plethora of open source licenses but NTeam is under the Mozilla license. If my understanding of GNU is correct, if you are depending on a product using GNU, your project also becomes GNU. I really need to consult a lawyer for the distinction on what is acceptable and not acceptable for use of a library but I don't like the way it looks.

Anyway, back to the matter at hand, what we are and are not producing.

1. We are not reproducing NAnt, NUnit, or FxCop.
2. We are not creating our own source code control provider. This is still a black box. We are not forcing people to use a certain SCC provider but it is difficult to get a system to work with many different scc providers.
3. We are not reproducing an issue management system but we are all ready working on work item tracking that will (hopefully) integrate with a couple of different open source issue management solutions such as BugNET, btnet, etc. This list has not been finalized so those two are a guess but we will have to wait and see. Our hope is that the community will make interfaces to other popular systems (like BugZilla) available separate from NTeam. There are simply too many for us to hope to hit even 10% of them.
4. We are going to create a continuous integration system but it should be easy enough for users to plug-in Draco or others.
5. We are not reproducing an IDE. I am not sure how it circulated that we were but this is just not true. We are going to target VS.NET 2003, VS.NET 2005, Delphi, #Builder, and eventually the mono ide.
6. We may have to produce our own project management solution simply because there are not that many on the market. I found two only: dotNetProject and UserStory.NET. I am not sure if dotNetProject is being maintained or not. There doesn't seem to be much action on that front. And, of course, not everyone uses XP so UserStory.NET may/may not be included. Again, hopefully the community will step up and provide interfaces to connect with other systems. I will try to post a more in-depth analysis of how the products will integrate later on. Today is a busy day. Posted on Monday, May 2, 2005 8:36 AM .NET , Open Source , My Projects | Back to top


Comments on this post: What NTeam will produce

No comments posted yet.
Your comment:
 (will show your gravatar)


Copyright © Jason Bentley | Powered by: GeeksWithBlogs.net