FreeCAD enhancements proposals

[looo]

For python there are pep's: https://www.python.org/dev/peps/
Ipython has IPEP's: https://github.com/ipython/ipython/wiki ... -Proposals
Conda-forge has cfep's: https://github.com/conda-forge/conda-fo ... -proposals

What do you think about creating something similar for FreeCAD?

In my eyes something like this would be very handy. Often ideas need the input of multiple developers and the forum is not the ideal place to create such proposals. The ability to create PR's and discuss on the differences will simplify the process of finding feasible solutions.

Some examples might be:
- releases (date, diff, TODO's)
- rules for python API
- topologic-naming solutions
- maintenance
...

[sgrogan]

What do you think about creating something similar for FreeCAD?

I think it's worth a try. It could be considered as an upgrade to https://www.freecadweb.org/wiki/Development_roadmap which is hard to maintain.

How do you envision it working? A read.me in a git repo, comment on pull requests? How about issues/feature requests to coordinate with mantis.
Maybe how it works is the first proposal?

[looo]

What do you think about creating something similar for FreeCAD?

I think it's worth a try. It could be considered as an upgrade to https://www.freecadweb.org/wiki/Development_roadmap which is hard to maintain.

How do you envision it working? A read.me in a git repo, comment on pull requests?

I guess it's best to look at the other examples. First, we need a template and define what we want to address with these proposals.

00: Template
01: FEP Purpose and Guidelines

How about issues/feature requests to coordinate with mantis.
Maybe how it works is the first proposal?

Not sure if referencing the mantis issues in the proposals would be sufficient. Any other ideas?

[Joel_graff]

I think it's worth a try. It could be considered as an upgrade to https://www.freecadweb.org/wiki/Development_roadmap which is hard to maintain.

How do you envision it working? A read.me in a git repo, comment on pull requests? How about issues/feature requests to coordinate with mantis.
Maybe how it works is the first proposal?

I definitely think it's a great idea. Maybe use github's project kanban boards or something to provide a little better tracking / accountability?

[looo]

I definitely think it's a great idea. Maybe use github's project kanban boards or something to provide a little better tracking / accountability?

Not sure if this is really necessary. Too many tools will make it more difficult to track all the stuff going on. I guess keeping this as simple as possible will help to get involved. This is a simple repo with enhancement-proposals as markdown files, PR's and issues. That's all. But for sure we can try some more advanced stuff too.

[Kunda1]

Another example of a proposal enhancement repo is qgis:
https://github.com/qgis/QGIS-Enhancement-Proposals

[Joel_graff]

Not sure if this is really necessary. Too many tools will make it more difficult to track all the stuff going on. I guess keeping this as simple as possible will help to get involved. This is a simple repo with enhancement-proposals as markdown files, PR's and issues. That's all. But for sure we can try some more advanced stuff too.

I agree. If nothing else, this should completely replace the development roadmap in the wiki.

My thought with the project board might be to provide a layer of accountability for these enhancements. The issue tracking would allow us to target milestones, identify the project leader, etc. I'd really just look at it as a way to hold us accountable for the enhancements we're proposing and think about them more often. Maybe force us to ask questions like, 'is this really necessary?', 'is person x really working on this?', 'is the timeframe reasonable?', etc. If you're already tracking issues, the project board is a small step, really.

Granted, it does require maintenance effort, but I don't think it'd take a lot to occasionally visit the board, review the enhancements that are targeted for a particular milestone, ping the responsible devs for a status update, post in the forums on the need for the enhancement, etc. A monthly status update, maybe. I don't know.

Anyway, @kunda1 already does this on the forums in various ways. And, if there are other things we're using to provide this level of accountability / tracking, maybe they can be eliminated with this. I might even be willing to pitch in, but only if it seems like it'd actually be valuable to the community.

[sgrogan]

I guess it's best to look at the other examples.

The conda-forge model seems to be the simplest. I like the Ipython the best.
I'd suggest using the conda-forge model and see where that takes us.

[looo]

The conda-forge model seems to be the simplest. I like the Ipython the best.
I'd suggest using the conda-forge model and see where that takes us.

Yes I also think the conda-forge example is most suited to the freecad-politics. Ipython uses the github wiki which I think is not ideal, as it doesn't allow to make PR's.
In my eyes it's important to allow people making pull-requests on proposals as everyone should be able to modify the proposals. So if someone is not happy with a statement in a proposal he/she can rework the proposal. If these changes make sense and are accepted by the majority it's going to be merged.

My thought with the project board might be to provide a layer of accountability for these enhancements. The issue tracking would allow us to target milestones, identify the project leader, etc. I'd really just look at it as a way to hold us accountable for the enhancements we're proposing and think about them more often. Maybe force us to ask questions like, 'is this really necessary?', 'is person x really working on this?', 'is the timeframe reasonable?', etc. If you're already tracking issues, the project board is a small step, really.

I agree this would be an interesting extension. But we need to take care of the open-source character of this project. We are all aiming to make FreeCAD better but in my eyes, motivation shouldn't be forced. For sure things shouldn't be done twice (or 4 times like assembly ) but we shouldn't be too strict about this (better 4 assembly-wb than 0 )

[freedman]

I use 3 or 4 macros most all the time and I presume that most of the long-term users do the same. Maybe we could start a listing of what folks use and then determine what they like most. Many maros or code could be being used only by the creater so they could describe what they are doing and why they need it. Maybe this is a program direction indicator, not sure.