Thank you for such as great project. Using Kstars and Ekos on RPI3 is improving. Some thoughts:

1. A stable release label and bleeding edge label to help control when to update (supporting already mentioned idea)
2. Imaging camera view with telescope control to help frame shots, without a focuser; helps to capture two objects when not just centring with plate-solving
3. No slewing option with scheduler - Celestron precise goto often can frame shot nicely, and would be good to just start sequences when it is a newly frame sequence.
4. Separate build of Ekos from Kstars (supporting already mentioned idea)
5. Ability to use another planetarium, like SkySafari. Currently can use both Kstars and Skysafari at same time as long as INDIserver is not stopped/restarted or such.

Cheers!

1: Polar alignment
I would like to see a quick polar alignment as in EQMOD/ASCOM in Ekos.
The use of the astrometry helps a bit for polar alignment but just to make an accurate alignment.
The polar alignment with astrometry would be more accurate if it could exists a graphical representation, like a target, of corrections done with the adjustement knobs like a target and points for each exposure done for the process.

2: Indigo
I had an overview of indigo and this looks to be a real improvement of INDI.
Hotplug of the devices become possible. This is really nice.

3: More stable versions
With possibility to go back to a previous version.

I definitely second this. It seems to be easily implementable in main telescope driver.
I have had a discussion with a friend of mine using EQMOD/ASCOM and he convinced me that polar alignment with this approach is 10-15 minutes exercise. Plate-solving technique currently implemented in Ekos takes ages to achieve the same goal. What we have now is great for home observatory (so I use it) but does not meet road warriors' needs.

Jasem I would love to see Ekos isolated from Kstars. as a separate executable

I usually use my polar alignment scope for the polar alignment and then I use plate solving for the alignment and then I am good to go. This whole process takes me less than 10 minutes. My alignment is probably not perfect, but it is good enough. I Are you referring to this tool at the lower right corner of the Align Module? I have never used that.

That's it. And by the way, polar alignment scope is designed for... polar alignment BUT how do you know exact Polaris hour angle? How do you set it up on your scope by rotating RA axis to Polaris hour angle?
The requested alignment tool is all about these two issues. Instead of using any external tool to find our Polaris HA (e.g. Polar Finder) and relaying on your RA axis rings and markers: (1) you can just click a button and your mount will slew (in RA only) to exact position of Polaris, track it and all of these for your specific location and time. (2) Take a look through your polar alignment scope and set the Polaris in the properly identified position.

The possibility to save a video.

Yes this is it. And I don't agree with the fact that an approximative polar alignment is enough. The better the polar alignment is the more efficient the autoguiding is.
And the less exposures you have to throw.
I don't use EQMOD polar alignment with my setup. But I use it with the setup of my astro club. And it is really terribly effective and so simple and accurate and so fast too.
Sure the plate solve is effective on a fixed station when you do it twice a year. But when you have to install your hardware at each session it is not so easy.
Furthermore on AZEQ6 the big knob that rules the altitude is so imprecise that a blind adjustment with the plate solving make me loose a lot of time.
How I proceed:
- first I align at polar scope whith the setting circles. It takes less than 5 mn to do it.
if I am lucky and if I do a good job I can obtain around a minute of accuracy on each axe. Generally I get few minutes.
- twice I use plate solving align. It takes at least 20mn to improve the alignment.

EQmod alignment need less than 5mn for a really good alignment for less than a minute on each axe. So plate-solving alignment can be optional.

What I hope to find in EKOS is the equivalent of EQMOD alignment and a more visual approach of the best alignment with plate solving.

I don't really know how ASCOM is proceeding, but I was never able to use the SynScan polar realign feature properly (firmware perhaps?).
I second your proposal, and I'd be interested in a method that DOESN'T rely on seeing Polaris to adjust (that would improve the available mount locations in my garden by 70% ). I'd start with a rough north alignment, and let the algorithm repeatedly tell me how I should adjust Az/Alt on a portable device. I'd say nearly-automated Bigourdan method, in which coordinates are plate-solved, and star trails are detected, but there may be easier methods.

As a side note, asking the mount to point to the "exact" Polaris position cannot be "exact" as the mount isn't synced yet. Depends on where the user locked the knobs ; that's why the polar scope must remain manual, isn't it?

-Eric

The mount should rotate RA axis to the exact Polaris hour angle BEFORE you touch the knobs. This exact position sets the destination for Polaris in your polar scope. You get there by adjusting the knobs.

PS. Jasem, why don't we move this part of discussion to a separate topic? Same for any discussion on other features.

Just to fix the things: In EQMOD process, the polar finder must be accurately adjusted (I mean align with the mount mechanical axis). And it is necessary to set an intial position for polaris (generally 6h). And then you can run the EQMOD process.
What I understand about Polar Alignement and Sync:
There is no real connection between polar alignment and sync, Sync is used to teach the mount how to point at a precise point. Sync is used to build an equation of the sky.
So you can make sync with poor polar alignment. The offsets on each axis will be taken in account in the sky equation.

@Kackzorek: Maybe there is an existing topic ongoing somewhere in the list of topics.

I'll also add my thanks for a nice application. I just set up AstroBerry-Piface and have been playing with it as part of a remote/local combination to get things going. Hope to turn the Pi into an Observatory controller, but more on that in another thread at another time.

For Ekos, I would really like the ability to import a target list. Ideally from a spreadsheet (look at how Word does it for mail merge), but a comma/tab/other delimited text file is OK. In my case, the same camera controls could be used for all targets and the output file could be named in the input file or one could use a directory/target_name method

We are imaging double stars for Speckle/autocorrelation, as such wee often have a list of 200-2000 targets. The 16 12 27 file is what we use for capture and automated reduction. The other two are from DeepSky. "planning" function in KStars and CDC are not options. The " on every cell of the text file can be scrubbed as needed (python?). I haven't found a good Linux based planning tool with parametric searches so a generic file format is probably the most flexible.

So an import function, but think 1000 plus objects.

Great question, thank you again. Not sure the forum liked the excel files... trying again