This looks great. But a question. Does it have to be cloud based? Can it be hosted directly on the Raspberry Pi, for example? Put the Pi in hotspot mode, and my phone or tablet can connect to the Pi when I'm out in the middle of nowhere with no Internet access.

Agree with the idea of using it far from Internet access. My dark sites are dark in light, wifi and even cell phones. Truly off the grid.

This is from which branch? On Ubuntu?

Not yet, but it would be quite reasonable I assure you. I spent a lot of time in the development phase and now have to see how it actually scales along the associated costs.

Excellent question! For me, it would be very useful to use Ekos Polar Alignment Assistant tool directly without having to use TeamViewer..etc. Another use is for folks to monitor the progress of an ongoing astrophotography session while somewhere else. I've received stories like "My guiding aborted while I was an hour away and I couldn't do anything about it until I came back to my observatory". Well now with this, you can do such corrective actions from a distance! I'm sure there are other use cases, maybe for EAA as well (I plan on adding live stacking later).

I used iton Ubuntu.

I can see a use case for this. However, for the time being, I will work on the cloud deployment aspect. Maybe I can release a future version for offline use!

You need to use ekoslive branch:

git.launchpad.net/kstars-bleeding/log/?h=ekoslive

Will definitively give it a try, - even if I may not need it!

Thanks btw for all the updated tutorials on YouTube which are are great reference!

Ok, so I'm still working on the development part, but I'd like to get some feedback on the "Cloud Storage" idea.

So how would Cloud Storage be useful? What are use-case scenarios for it? Some might think that "Cloud Storage" is a solution looking for a problem, but I assure you it is not. Most folks, myself included, use some processing software to work on the image. Personally I've been using PixInsight for a couple of years and love it. Therefore, most of the data are calibrated, registered, and then later cropped..etc before the serious work begin on them, so what would cloud storage add to this?

Keep in mind that the storage capacity required to save such usually large images can ramp up to be in the gigabytes easily, and while online storage services are becoming cheaper these days, it still can get pretty expensive to save all these images online especially when they're usually used once. This is why I decided to check on the community before moving forward with this idea. Besides discarding the cloud storage idea altogether, I'm thinking of two approaches:

1. Store everything in the cloud
So raw images (FITS/CR2..etc) are stored online along with e JPG preview, thumbnail, and metadata. Ekos Live users can see a "gallery" of images (thumbnail size), and when they click on an image they get the full-sized JPG preview, and at their option, they can click on some button to download the "raw" image (FITS/CR2..etc).

This would act a backup online for all captured images.

Pros:
1. All captured images are saved online and accessible to download from anywhere and anytime, with possibility to share the link with other users and/or social sites.
2. Ability for advanced search. For example, the user can filter images to only show those images taken between date X and Y, in the H_Alpha filter.

Cons:
1. Storage size can ramp up steeply only after a session or two. Even if each user is given 10 GB of storage space, it's going to be filled up pretty quickly given the size of raw images.
2. Requires a VERY reliable and fast internet connection and bandwidth to upload the raw images immediately after capture.

2. Store only JPG in the cloud

So in this option, the raw files are never uploaded, only the smaller JPG (with thumbnail) is stored on the cloud.

Pros
1. Storage requirements are now less demanding and more reason and can sustain the user for a very long time.
2. Bandwidth requirements on the user side are less demanding as a result.

Cons
1. No access to raw data, just the rendered previews from Ekos.

So what do you guys think?

Option 1 : is an interesting idea, but I agree with the fact that you can saturate quickly the quota of storage allowed.
Option 2 : Personally I don't see any interest in this option.

Might it exists a non destructive compression for raw or fits files. that might give a better interest to option 1 ?

One idea, very often the digitization of the light signal is done on 10, 12 or 14 bits so there is 6 to 2 leadings bits usefulness. Why not trying to pack the datas ?

I use fpack & funpack FITS image compression programs .