Addons and Developers Portal

[mnesarco]

Proposal description

A big part of the FreeCAD Ecosystem lives in Addons developed by external contributors. Over time, Addons have been adding impressive capabilities to FreeCAD but unfortunately there is no dedicated web portal where Users and Developers of addons can discover and rate such Extensions. Also a dedicated developers portal can give accurate and updated information to Addon developers to keep its Addons compatible and reasonable well structured.

This proposal is to develop an Addons and Developers Portal where users and developers can discover Addons, learn about Addon ecosystem, Addon development, etc.. and also see important statistics like number of downloads, likes, and quality points based on static analysis of the published code. Also the proposed backend system manages all the publishing process of the addons adding automated analysis and QA assesments, so the AddonManager can focus on downloading and installing the curated Addons and its dependencies in the user's local environment.

Similar sites that inspire this idea:

• https://pub.dev/packages
• https://pypi.org/
• https://marketplace.visualstudio.com/vscode
• https://extensions.blender.org/
• https://inkscape.org/gallery/=extension/

Deliverables

First a general idea of the architecture of the final product:

The project involves several Modules, so there are different ways to split into deliverables, As the scope is wide, I suggest to start with the design stages and reevaluate after that. Trying to estimate everything at this moment feels too unrealistic.

In general a global view of expected big-box deliverables:

1. Functional specification
   Specify all the use cases to be implemented like Search, Rating, Analysis metrics, authentication, displayed info per addon, organization by developer/group, etc...
2. Architectural design
   Specify all the technical aspects of the solution, basically a python based backend and reactjs based frontend, everything well containerized with async io for good performance and all the details...
3. Backend Implemetation, probably can be divided by pipeline component: Fetcher, Extractor, Analyzer, .... plus its infra (docker)
4. Frontend Implementation: a reactjs app and all its associated infra (docker)

The links supplied as inspiration provides a good expectative of the final visible product (fronend). And the backend expectative is illustrated in the diagram.

Timeline

This is not a tiny project. My estimation is that 6 months may be enough to have the first version working as I cannot work on this on a full time basis. The heavy part will be the backend as it has multiple non trivial modules dealing with external tools like git, log analyzers, static code analyzers, databases, cron jobs, etc... and having good unit tests. If I start in October and excluding December, April may be the goal. Serveral alpha and beta versions can be achieved during February or March.

Risks and mitigation

This is a self-contained project, so the main risk is to underestimate the time and effort required. During the specification and design phases we can have a better estimate and do better risk planning, but overall i feel that 6 months may be ok.

Compensation

Keeping in mind the 6 months estimate I think I can do this for USD$8000. It may be divided into 3 payments: 2k one after specification/design (between 1st and 2nd month). 4k after initial working backend (between 2nd and 4rd month), 2K after frontend (between 4th and 6th month).

After the initial release, based on users feedback we can start a new grant with a new scope.

About you

My name is Frank Martínez.
In forum, discord and in many other communities my nickname is mnesarco

About my experience, I have been working on backend development for at least 20 years, here is my Linkedin profile: https://www.linkedin.com/in/mnesarco/

I have been involved in Telco, Content, ERP, Infrastructure, Crypto, Engineering dealing with data processing, pipeline based systems, code and database optimization, and systems design in general.

I have been contributing to FreeCAD community since 2020 when I started the Marz Workbench for Guitar design, over time I have also contributed other workbenches like Svg Workbench, Channels Addon (Blender bridge), Vars Addon, and the fcapi among others. Also I have contributed a few direct small fixes and features to the FreeCAD Core.

I work on projects like this all the time, from small to medium size companies typically on intranet services, production pipelines and custom engineering software. My typical roles are software architect, technical leader and/or senior developer.

[Reqrefusion]

I thought of it for something like this at the time, in fact it still stands, it was something simpler, static php script that simply turned the plugin json into something like a gallery. I was going to use telemetry data in the ranking. It wasn't as cool as this one, xD. It looks very nice, it seems like there is a need for such things since we now handle the entire distribution of the current addonmanager. But I don't know.

As we always say, the goal of FPA is always to give grants to more people and we have certain budgets for each period, which is 20 thousand euros, the proposal here occupies a significant part of this budget.

[Reqrefusion]

As someone who has thought about scoring systems for many years, I think it's all unnecessary. Personally, the most reasonable number of votes should be the number of downloads or something like that, people should express their will with their actions. However, if there is to be a scoring system, I think it should be done with a two-option system and depending on time. So I simply liked it or didn't like it. The vote over the past two weeks should be shown as follows: Mostly negative-Negative-Neutral-Positive-Mostly positive. Because I don't think the star system is being used correctly by anyone.

[chennes]

Thank you for applying for an FPA developer grant. Your grant was submitted in the window for Q3 2025, and is now under committee review. After the committee makes its recommendation to the FPA, a voting window will open, with decisions expected on 1 October 2025. Committee members and the FreeCAD community at large are invited to continue discussing this grant application with you here on this GitHub issue.

[mnesarco]

As someone who has thought about scoring systems for many years, I think it's all unnecessary. Personally, the most reasonable number of votes should be the number of downloads or something like that, people should express their will with their actions. However, if there is to be a scoring system, I think it should be done with a two-option system and depending on time. So I simply liked it or didn't like it. The vote over the past two weeks should be shown as follows: Mostly negative-Negative-Neutral-Positive-Mostly positive. Because I don't think the star system is being used correctly by anyone.

I think the scoring system should be more focused on quality than votes, voles (likes) may be there but not the most important.

This is how flutter community scores its packages:

[Reqrefusion]

I think the scoring system should be more focused on quality than votes, voles (likes) may be there but not the most important.

This is how flutter community scores its packages:

This has its own problems. Automated analyses are always more problematic when it comes to showing the full value of something.

[mnesarco]

This has its own problems. Automated analyses are always more problematic when it comes to showing the full value of something.

Some sort of automated analysis is required to provide basic info. All of this is informative, not restricting the Addon in any way. Additional human reviews would be good when available, but highly difficult to provide human reviews for all Addons and keep it updated on each commit.

[yorikvanhavre]

I am of the same opinion as reqrefusion here, a "technical quality" info would not help users much when deciding which adon to install (an addon could be embrionary but still technically perfect).

Generally speaking, though, this proposal is a great idea!

[mnesarco]

The developers portal is to build a community around addon developers, to help them build better addons, they appreciate the technical info.

Users want documentation about how to use the addons while devs want documentation about how to create good addons. We need to serve both.

We can define different sets of metrics, thenical quality metrics are just one part of the pie. General metrics are harder to automate so they may be human driven like usability, translations quality, user documentation quality, stability, TNP, ...

Others are statistical: downloads, likes, number of issues per month, number of commits per month, number of developers, etc...

[Reqrefusion]

The FPA's voting on grant proposals for the Quarter 3 2025 has been completed, and I regret to inform you that this grant funding was not selected.

Nothing negative was said about the proposal itself; however, it was mentioned that the new restructuring for the addon ecosystem should be awaited until it is completed. Additionally, questions arose regarding your involvement in this restructuring and what your responsibilities would be.

The rejection of the funding of a proposal for a quarter by the FPA does not mean that the relevant project has been rejected by FreeCAD management. If you wish, you can proceed with your proposal without funding. Additionally, the rejection of your proposal for this period does not prevent you from submitting the same or a similar proposal in future periods, nor does it create any bias against its acceptance.