docs: initial platform improvements #483
Conversation
|
📚 Documentation Preview ✅ A preview of the documentation changes in this PR is available for maintainers at: Last updated: 2025-10-23 15:26 UTC |
tristanls-tbp
requested review from
codeallthethingz,
jeremyshoemaker and
vkakerbeck
tristanls-tbp
added
documentation
jeremyshoemaker
left a comment
jeremyshoemaker
left a comment
There was a problem hiding this comment.
I left a suggestion, but otherwise, it looks fine to me.
| contributor: jeremyshoemaker | ||
| --- | ||
|
|
||
| If we (Injection #9 Integrate a second simulator that runs Python >= 3.8) and we **(Injection #16 Integrate a second simulator that runs on CPU)** and we (Injection #17 Integrate a second simulator that runs on Windows) and we (Injection #19 Integrate a second simulator that does not require conda), then we achieve the desired effect (141 Monty is decoupled from a specific simulator implementation). |
There was a problem hiding this comment.
I find these hard to read. Would it make sense to change these to links to the relevant pages and remove the "Injection #N" parts?
| If we (Injection #9 Integrate a second simulator that runs Python >= 3.8) and we **(Injection #16 Integrate a second simulator that runs on CPU)** and we (Injection #17 Integrate a second simulator that runs on Windows) and we (Injection #19 Integrate a second simulator that does not require conda), then we achieve the desired effect (141 Monty is decoupled from a specific simulator implementation). | |
| If we [integrate a second simulator that runs Python >= 3.8](TODO) and we **[integrate a second simulator that runs on CPU](TODO)** ... etc. |
If you do this, you could change the links to use Markdown's "reference links" feature where you only specify the URL once, since we have the list below under "Related Work".
A [link to injection 1][injection-1] and another [link to injection 1][injection-1]
[injection-1]: https://example.com/path/to/injection/1
There was a problem hiding this comment.
I'm hesitant. These are the names of nodes in the graph depicted in the image below. The sentence is a prose representation of the graph, intended to reinforce how the graph should be interpreted.
vkakerbeck
left a comment
vkakerbeck
left a comment
There was a problem hiding this comment.
Looks good :) just two small comments on providing a bit more of our current knowledge and context
|
|
||
|  | ||
|
|
||
| ## Related work |
There was a problem hiding this comment.
It could be worth including a bit more detail on explorations we already did into simulators that meet those requirements. For example linking to this thread https://thousandbrains.discourse.group/t/the-next-monty-3d-environment-simulator/288
| --- | ||
| title: "Injection #17: Integrate a second simulator that runs on Windows." | ||
| tags: platform, simulator, environment | ||
| skills: python, mujoco |
There was a problem hiding this comment.
The skill listed here is mujoco but none of the description mentions why. Maybe add a sentence saying that Mujoco seems to meet all these requirements
codeallthethingz
left a comment
codeallthethingz
left a comment
There was a problem hiding this comment.
LGTM, there are a couple other fields that you may wish to add but neither are required:
improved-metric: https://github.com/codeallthethingz/tbp.monty/blob/f4fcfb8ebad35cf4a8cb919f559ce0a5d6ed31e2/docs/snippets/future-work-improved-metric.md
|
This has been open for a while, and the last comment was back in October. It looks like it has the approvals it needs. Is there some follow up needed, or can it get merged? |
|
@jeremyshoemaker I'll come back to it once I've wrapped up the configuration work. |
5 participants
This pull request adds initial information on Platfrom improvements from the engineering future reality tree.
@codeallthethingz could you review and see if I added appropriate metadata for the anticipated future work widget?
@vkakerbeck and @jeremyshoemaker is the information content here sufficient for a future work page?