Discussion: More types!

[martindurant]

What do people use to operate on projects?
What contents or artefact types have we not yet considered?

That there are some stubs in the code, such as projspec.proj.helm, showing our intent to eventually implement something we haven't got to yet. These stubs are to be listed here, and we hope people will be interested in creating their parser classes.

[martindurant]

• golang projects

[martindurant]

• azure apps (e.g., https://github.com/Azure-Samples/azure-search-openai-demo )

(other cloud platforms like cloudflare and google-app-engine ?)

[martindurant]

• traditional configure/make projects, C/C++ (artefact is a binary library, maybe executable)

[martindurant]

From offline brainstorming:

• pre-commit (we already mention linters and GHA, but dontt implement anything)
• makefiles or even docker as an artifact (i.e.: productionise this project)
• snyk commands specialised to the language/runtime
• semgrep, owasp and other security scanners
• ansible, kube (helm?), docker-compose and other deployment infra
• port specification for webapps and deployments (cf openVAS)
• mlflow (cf prefect, airflow et al)
• dvc (cf lakeFS, xet et al, which are not project scoped, but may have local mounts)
• dbt pipeline https://docs.getdbt.com/reference/dbt_project.yml
• directories that are definitely data (parquet, zarr etc)
• identify major AI components: ray, spark, TF, openCV, torch, etc ?
  • find resource and cluster requirements

[martindurant]

• refine specific nodejs use-cases. We already have jlab extension, but what about, for instance, vscode extensions (cf VSCode integration #41) so that we can build using our own tool?

[martindurant]

Maven: https://maven.apache.org/guides/getting-started/maven-in-five-minutes.html (pom.xml)

[martindurant]

• AI discovery text, skills and such

[martindurant]

See #86 for a very bot-comprehenside list. That should be enough for at least the medium term