Hello @brentarias! It's great that you want to contribute to documentation, your help will be highly appreciated. We are going to release BenchmarkDotNet v0.11.0 on the next week with an updated documentation. After the release, I will answer all your questions and provide the corresponding links.

@brentarias, v0.11.0 is released! In this version, we migrated to DocFX and improved many parts of the documentation. It's still not perfect, but now it should be easy to modify it.

Below you can find some links which are relevant to your questions

There is no documentation about how to contribute to documentation. This is why I'm posting this issue here.

https://benchmarkdotnet.org/articles/contributing/documentation.html

launchCount, warmupCount, targetCount and invocationCount. The documentation nowhere defines these. I've been figuring it out simply through experimentation.

https://benchmarkdotnet.org/articles/configs/jobs.html

The BenchmarkSwitcher has no documentation of what command-line switches it acknowledges. The sample documentation doesn't use an intuitive example.

https://benchmarkdotnet.org/articles/guides/console-args.html
https://benchmarkdotnet.org/articles/configs/filters.html

Aside from the BenchmarkSwitcher, there is no documentation that indicates - one way or another - if the BDN console app is capable of accepting other command line switches.

https://benchmarkdotnet.org/articles/guides/how-to-run.html

There is no commentary about how or if test execution success or failure is taken into account. For example, if roughly 10% of [benchmark] executions result in an exception...how does BDN report this? Does it even have the ability to include this information in the results?

It's not supported for now, see #373

There is no discussion of "number of simultaneous users/requests." The documentation should state if there is such a capability and, if yes, how to use it.

If you are talking about multithreaded benchmarks, it's also not supported for now, see #147

Console output indicates events such as "Pilot N:" and "IdleTarget N:", but documentation does not explain these. Likewise, the documentation does not explain the difference between "IdleWarmup N:" and "MainWarmup N:" or the difference between "MaintTarget N:" and "Result N:".

https://benchmarkdotnet.org/articles/guides/how-it-works.html

Hi all!
My first contribution to documentation was as straightforward as:

• read through https://benchmarkdotnet.org/
• notice a room for improvement
• click Improve this doc
• use GitHub web editor to edit the corresponding .md file and create a pull request

But when it came to obscure bugs with links (and I did have noticed a commit history of the ongoing fight with links) and it was no longer a matter of editing .md, I followed the instructions in https://benchmarkdotnet.org/articles/contributing/documentation.html to build and start docfx locally. That's when I noticed that the site https://benchmarkdotnet.org/ is deployed with content/scripts that are nowhere to be seen in dotnet/BenchmarkDotNet repository.
The bug is not present in locally built docs, but present in "production"!

Is there some manual process involved in preparing docs for deployment? Or I am just missing something...

Update: one example of the content that looks manual is https://benchmarkdotnet.org/styles/main.js

@AndreyAkinshin PTAL

@Maximusya you are right, I build and deploy docs manually. When we migrated to the docfx-based docs, I decided to introduce versioning. Unfortunately, docfx 2.x doesn't support versioning, so I use a pretty tricky manual way to deploy docs.

However, according to the Google Analytics reports, nobody read docs for the previous versions. I guess, we can remove the versioning from the documentation. It will significantly simplify documentation building and deploying. We can bring versioning back after BenchmarkDotNet 1.0 is released (docfx 3.x should support it out of the box).

@adamsitnik what do you think?

@AndreyAkinshin I agree, if nobody is using it then we should remove it and keep things simple