SharpCap Forums

I apologize for repeating myself, but if the documentation is going to be updated later, the manual itself will not make sense.

Wouldn't one strategy be to clearly state on the front page of the manual that new features that are not in the manual are explained in a dedicated thread in the forum?



Watanabe

Robin,

The last Sharpcap user manual update was in 2023. That is quite a long time between updates. So many new features have been added to Sharpcap since then, the present documentation is no longer nearly as useful as it should be. Since I am more of a casual Sharpcap user, I am quite often frustrated at trying to figure out some of these new features and how I might implement them during my observing. I spend a lot of time looking through old posts on the forum trying to understand what they are and how to use them. It would certainly be more time efficient, at least for me, if they were in the manual. So ... I was wondering if you have any plans to update the user manual to 4.2? I know documentation is not a fun thing to do but it is sort of a necessary part of life, at least it is for me in my engineering work. Perhaps in the future, it might be feasible to document little by little as new features are added. Thanks again for your wonderful code. I want to use it well.

Best, Bill

Hi Bill,

you are absolutely right - the manual is out-of-date and I am aware that it really needs updating. I'd like to spend some time on that after I get the release of 4.2 sorted (basically stop doing new features at that point for 6-8 weeks, write some documentation while fixing bugs that come up). There are some issues with that plan though...

* There are always good new ideas for ways to improve the software
* Adding a good new feature is probably a more profitable use of time that documenting what is already there - lots of people will potentially use the new feature and benefit from it, fewer will read the documentation
* I much prefer writing code to writing documentation.

Secretly, I am hoping that AI reaches the point where it will write the documentation for me

cheers,

Robin

>>Secretly, I am hoping that AI reaches the point where it will write the documentation for me <<

I think it's already there.

In any case, I think my AI has just accepted the challenge.....

"updating the SharpCap manual is exactly the kind of task AI is reasonably well suited to handle, particularly because there's already an existing manual to work from.

Rather than starting from scratch, we could:
1. Use the existing SharpCap manual as the foundation.
2. Identify features introduced since the manual was last updated.
3. Compare the documentation against SharpCap 4.1 and the current 4.2 releasegiving an AI agent access to a Windows computer running SharpCap 4.2, along with the existing manual.

The important distinction is that AI wouldn't need to invent the documentation. It could help consolidate what's already available in the manual, release notes, SharpCap forums, and other official resources.


The biggest challenge wouldn't necessarily be writing the documentation. It would be verifying that the instructions accurately reflect how the current software actually works."

As to the last part - verifying the instructions accurately reflect how the current software actually works -

Current computer-use agents can already interact with graphical interfaces, including clicking controls, entering text, navigating menus, and inspecting screenshots.

So.....

Give an AI agent access to SharpCap.

The agent's assignment might be:

"Read the section of the manual describing Polar Alignment. Open SharpCap 4.2, follow those instructions, and identify any discrepancies between the documentation and the current user interface. Rewrite the instructions to reflect the current software, including screenshots."

The agent could then:
1. Read the existing instructions.
2. Launch SharpCap.
3. Navigate through the menus and settings.
4. Compare the interface against the manual.
5. Identify missing, renamed, or relocated controls.
6. Capture screenshots.
7. Rewrite the relevant documentation.
8. Follow its newly written instructions to check whether they work..
4. Draft updated sections explaining the newer features, settings, and workflows.
5. Flag anything requiring Robin's technical input or verification.
6. Produce an updated manual for review.

Perhaps it wouldn't all be doable by an AI agent, but much of it, in theory, would seem to be with "test cameras" already built in to SharpCap. No need to connect a telescope/camera, wait for nightfall, or, even, clear skies.

You, Robin, lay on a beach somewhere sipping Mai Tais while the AI agent slaves away.

All true but ...

* There are always good new ideas for ways to improve the software.
Delaying a new idea 6 to 8 weeks should not cause too much anxiety for anyone. We wait all year for Christmas don't we?

* Adding a good new feature is probably a more profitable use of time that documenting what is already there - lots of people will potentially use the new feature and benefit from it, fewer will read the documentation
Lots of people won't benefit from it if they don't know how to use it properly or realize it exists. If people don't read the documentation then they are just lazy and I feel sorry for them. If you have a great piece of software, why would you not want to spend some time learning how to use it?

* I much prefer writing code to writing documentation.
That's part of the game. Everything in life is not always fun. Maybe the AI idea will work.

Best, Bill