@gb I had to enlarge the system partition on my Nexus 5 to install everything, if I remember correctly. Maybe that's your issue? I think I followed this script https://github.com/plasma-mobile/plasma-phone-dev-setup/blob/master/usr/bin/resize-root-partition with minor adjustments, but caution I don't remember since it's been a while – do it on your own risk

@mariogrip Great work! One question, though: How would I evaluate if something is an issue with anbox itself or with the image?

@mariogrip Will you just upload new images to http://cdimage.ubports.com/anbox-images/ so that I can periodically check there or are they already there and I don't get it... (because it lists the Nexus 5 for example in the devices.json)?

@advocatux Thanks, indeed I should stop and wait for Marius to release an image!

@mariogrip Very nice – I was just trying without a custom image, but that failed at some point, so I'll skip the headache and wait for your image instead

I was wondering: What's blocking anbox from running on the Nexus 5. Is this something ubuntu touch needs to address or is it a cpu architecture that anbox needs to support first?

I'd be happy to help with anything I can, but would have no clue where to start

Edit: I guess it's that one:

In order to support the mandatory kernel subsystems ashmem and binder for the Android container you have to install two DKMS based kernel modules. The source for the kernel modules is maintained by the Anbox project here.

from https://github.com/anbox/anbox/blob/master/docs/install.md

Right?

Edit 2: So, I followed this manual https://github.com/anbox/anbox/blob/master/docs/install.md and came pretty far.

I had to resize the root partition with this script https://askubuntu.com/a/721296/717375 and mount it writable sudo mount -o rw,remount / to install the new kernel modules (of which I have no idea whether they work on ubuntu touch as well...)

Then I realized that snap doesn't seem to be included in this ubuntu touch image (at least for the Nexus 5 on 16.04). That's why I can't continue with installing anbox through the snap mechanism.

I guess the kernel needs to be adjusted to support snaps?
Now I'm hooked, but still can't see what could be a next step...

But there is a new answer by @advocatux below – let's see

Did you pursue this any further? I'm looking for a decent voip app for ubuntu phone and would not fear to develop a UI or packaged app myself, but I guess the backend library to talk to the sip/voip server would be the most important piece. And nobody wants to reinvent the wheel, right?

@doniks said in Dekko2 Developer Community Thread:

https://gitlab.com/dekkoproject/dekko

This is very cool – looking forward to (hopefully) fixing some bugs when I get to it in the upcoming weeks.
Email is (still) vital.

@UKPhil said in Idea for WhatApp on ubports:

From my understanding of the issues it is all due to the fast that facebook will not publish their whatsapp api's thus controlling the apps that use their service. If they did we could make something up. I might be wrong there so if I am please feel free to correct me.

Yes, you are absolutely right. This is not about the missing ability to create such an app, but about restrictions on Facebook's side that make it impossible to create a client app without reverse engineering an official client. At least this is my current state of knowledge. Hence my suggestion of the cumbersome, yet working, usage of the webapp together with a “server device”.

@guru said in Idea for WhatApp on ubports:

The communication stays by email.

And I'm guessing this didn't make you the most popular Dad in the group of parents You're absolutely right to do this and it is very idealistic and admirable – don't get me wrong! There is just the fact that WhatsApp is very convinient for many people, because they use it all the time and so they try to fit it to every use case they can find – and Facebook also tries to make the app adapt to these new use cases. Whereas you and me, not even “users” (“...and what is this Telegram, by the way?”), explaining the drawbacks and privacy implications of WhatsApp sound highly outlandish to the ears of many WhatsApp users.

– Thomas

Yes, despite all difficulties, I find this topic very interesting. It's also not that people use WhatsApp because they like it so much over competitors, but that it was the first to offer free, but SMS-like services in a few markets, then acquired by Facebook.

Since the app is effectively free, and chats are now supposed to be end-to-end encrypted, the question remains how Facebook wants to monetize WhatsApp. Maybe metadata is enough for them or they are just interested in holding a good market share in this area...

@UKPhil I totally get your point there, and it is very valid, in my opinion.

The question is, if ubports wants to aim at reaching these target groups, you describe, just now – at least in the near future ubports might want to focus on a user group inbetween, consisting of users like you and me, willing to join the discussion, make a sacrifice to use a different and flexible operating system on their mobile devices.

I could imagine the user base growing from “very geeky” to “very interested in tech and keen on experimenting” to “slightly techy” to a “mass market”. So, I don't think the first step should be to have all the shiny apps “for the masses”, but to have a nice platform for experts as well as enthusiasts.

But I guess everyone has slightly different expectations of ubports and its future and only time will tell who is “right”.

Edit: And you're right, the webapp will only work on one device at a time

– Thomas

Well, I was writing this after watching the latest Community Update. I simply don't agree with the statement: “moving everyone to Telegram is the only solution”.

Let me elaborate. I don't agree with Facebook's business model. It is not something one would strive towards. Hence, I wouldn't advocate that Facebook develops any apps for ubports. (Besides they probably wouldn't, as @Flohack explained. It took them a long time to create a client for Windows Phone and it still breaks the keyboard when composing messages.)

In contrast, what I want to promote is the idea of bringing “legacy” apps to ubports. They don't have to work in a great way, they just need to work. And my only requirement for that is for the respective service to offer a browser-based version. As long as something works in a browser, we should be able to make it work on ubports.

Once again, I don't think convincing everyone you ever want to communicate with to use your technology of choice is the way to go. (It is one way, though.) I think instead we should hack together smart bridges that allow us to merely use this – in my eyes – “legacy” technology in a basic, but reliable way.

For WhatsApp I think the webapp connection is feasible. @arubislander, I tried the webapp from the store, but turns out – surprise – the WhatsApp webapp is not fully responsive, that is, it doesn't fit the screen without scrolling. For now it kind of works, but I think maybe a custom stylesheet could fix this horrible experience.

I'd love to look into optimizing this at a later point and will keep you updated if I make any progress

@UKPhil, as much as I see the need for a working WhatsApp client, I have to agree that a native client simply won't happen. Firstly, everyone creating an inofficial client will face legal threats from Facebook and, secondly, the motivation of Facebook to create a client themselves is probably even lower than we think. We have to accept this and – this is my point of view – are able to work around this.

While it isn't great to force people through a more complex process than downloading an app and signing up, it is a solution that is only slightly more complex than installing ubports itself and ideally requires as good as no maintenance.

Also, the time might work in our favor here, because popular webapps get better every day.

– Thomas

I think the most efficient solution for the problem of the missing native WhatsApp app on ubports could be using the official webapp.

You would need two things for that:

• an old phone with an operating system supported by WhatsApp, e.g. Android, iOS or Windows.
• a wrapper app for the official WhatsApp webapp that can do the storage stuff needed to keep up the connection

The old phone can be used as a server, permanently connected, somewhere in your home. And the webapp can function pretty much in the same way as the native app (probably with some restrictions).

I tried this approach, but I'm not able (yet) to create a wrapper app for ubports and when I tried with the built in browser it would forget the session after every restart of the browser.

Maybe there is someone out there who could get the official webapp working reliably and wrap it into a native app?

– Thomas