I'm thinking we should take a different approach, partially motivated by the "Getting Real" approach.
1. Let's make this version of hop into "opinionated software". We don't need to please everybody. In particular, I'd like to include something like 'trac' or 'bugzilla', but make it both non-linear and scrum inspired (i.e. make it in the form of a backlog (a simple ordered list of tasks), etc.)
2. The main theme of hop is "a non-linear source code control system". We want to support atomic change units but allow them to live in a graph of dependencies rather than on a branched sequence.
3. Let's underdo our competition. We don't need all the bells and whistles. Let's stick to the core.
4. Let's ditch cvs and just use a free database and the file system. Disk is cheap.
5. Maybe we can call it SHOP? Scrum + HOP = SHOP? Or else: rhop, hopr, chaos, warp, ????
6. I really like the ability to work off-line that CVS has. Let's do that too.
7. Maybe we can allow people to "insert" changes into the middle of a change graph. This would be really cool for patches, etc.
8. Does it make sense to incorporate TDD or BDD as a first-class activity with its own artifacts? In other words, should unit tests or behaviour specifications be highlighted or treated as their own first-class entities in some way? (another opinion!)
9. Rather than do it on sourceforge or rubyforge, I think it would make sense to try to boostrap SHOP in itself from the beginning as much as possible. How can we make our change zero as small as humanly possible?
If you would like to refer to this comment somewhere else in this project, copy and paste the following link:
Hi all. I'm creating this thread to discuss the next generation of Hop. I think it's time to re-invent the wheel!!! :)
Some opening thoughts in random order:
I'm thinking we should take a different approach, partially motivated by the "Getting Real" approach.
1. Let's make this version of hop into "opinionated software". We don't need to please everybody. In particular, I'd like to include something like 'trac' or 'bugzilla', but make it both non-linear and scrum inspired (i.e. make it in the form of a backlog (a simple ordered list of tasks), etc.)
2. The main theme of hop is "a non-linear source code control system". We want to support atomic change units but allow them to live in a graph of dependencies rather than on a branched sequence.
3. Let's underdo our competition. We don't need all the bells and whistles. Let's stick to the core.
4. Let's ditch cvs and just use a free database and the file system. Disk is cheap.
5. Maybe we can call it SHOP? Scrum + HOP = SHOP? Or else: rhop, hopr, chaos, warp, ????
6. I really like the ability to work off-line that CVS has. Let's do that too.
7. Maybe we can allow people to "insert" changes into the middle of a change graph. This would be really cool for patches, etc.
8. Does it make sense to incorporate TDD or BDD as a first-class activity with its own artifacts? In other words, should unit tests or behaviour specifications be highlighted or treated as their own first-class entities in some way? (another opinion!)
9. Rather than do it on sourceforge or rubyforge, I think it would make sense to try to boostrap SHOP in itself from the beginning as much as possible. How can we make our change zero as small as humanly possible?
Hi all,
I've created a basecamp project for hop, to see if we like it for bootstrapping or managing hop. It's at
http://sbrew.updatelog.com