Re: Mer-based OS on a Tablet: what would you like to see?
 

I tried to open a similar discussion there : http://talk.maemo.org/showthread.php?t=96195&page=2
But It didn't had a lot of success...

Re: Mer-based OS on a Tablet: what would you like to see?
 

Well, but the difference here is that I have no interest in dealing with corporations for the purposes of this distribution (if distribution it will be). This is simply a question of what would it take for someone (for example, me ;) ) to cobble together a mobile Linux/Mer/etc. OS by themselves. What would be absolutely necessary? (And yeah, the more work involved, the less likely it is that a feature will appear...)

Re: Mer-based OS on a Tablet: what is needed?
 

Actually it is nothing to be joked about, having a proper terminal is the most desirable feature of a handheld, so why not for a tablet device too. Of course it needs a proper keyboard too; fingerterrm and like is usable but for example on my Jolla device I use TOHOKBD almost exclusively.

Basically this can be avoided by just shipping the device without GUI at all, everybody can then build the GUI they need and want.

Re: Mer-based OS on a Tablet: what is needed?
 

Lucky you. I was all excited when I got mine but after the initial novelty wore out, I found it less useful than expected. That has nothing to do with TOHKBD itself, it's the Sailfish UI that does not lend itself to a convenient use with a keyboard. I ended up reverting back to using my TOHOLED and my TOHKBD is gathering dust. I was considering selling it but seeing how many are on offer for months with little interest, I gave up.

I would prefer "install from repository" to "build". That of course implies that one can use standard repositories and avoid the blunder all mobiles make, including Jolla - trying to lock the user in with their own proprietary "app store".

Re: Mer-based OS on a Tablet: what is needed?
 

Baytrail is utter rubbish:
You may not want that Windows Bay Trail tablet after all

No one has succeeded in getting linux running completely on
any Bay Trail hardware - there are far too many driver issues.

AdamW is the stuff of legend in the linux industry,
and is due a great deal of respect. :D

I was answering the previous post but got interrupted:
Hmmm,
without dropping names or mentioning privy discussions,
I cannot provably demonstrate that Fedlet
and anything it represents is probably not going anywhere,
but you read what you might into that statement.

Just recognize that Fedora is driven by RedHat and
they have a very focused attention on where Fedora is headed
and any detours from that particular corporate toll road
would require permission never found inside their ecosystem.
Several someones at RedHat did apologize to me at one point:
they seemed to have problems with corporate direction.
From those discussions I do understand how difficult it is to keep the RHEL boat
moving in the direction of the bow without someone asking
for things which are far outside their foundry, so to speak.
But nevermind all that water under the bridge....

There are other things I know about tablets.

I am somewhat hopeful that there will eventually be an intel cpu
which is not yet another ATOM SoC (Chit ona Ship)
so we can get off the Android sludge-way.
They have not yet gotten the message yet but there is hope.
Doing anything with Android hardware is just plain ridiculous.
People who do that must be masochistic somehow.
This also applies to the SoC architecture
Intel keeps tossing on the market,
much like I did when feeding the pigs back on the farm:
It is a bucket - got something mushy in it,
the hogs are hungry so just tip it over the top rail
and watch them go at it !
That is the consumer device market - they almost do not care
what they get as long as the advertising is in color and on 24/7.


Give me a computer and a USB stick or a dvd drive
and I can generally find some way to install something on it,
regardless of whether it is ancient or leading edge.
That is strictly impossible with Android hardware.

To do anything with Android the first requirement is that you
have a working computer properly set up to juice the Arm
from outside, much like all the special tools needed to construct a
ship in a bottle
It is not impossible,
I just regard it as verging on being stupidly subservient
to people who could just as easily build proper bootloading
into the hardware.

for a more eloquent discussion of totally stupid Android is,
just dive into the horses mouth:
https://community.arm.com/groups/and...r-arm-platform

even better is this excellent explanation of why ANDROID
hardware is so blindingly stupid to work with:
http://android.stackexchange.com/que...om-a-usb-drive.

Android is not just stupid,
it is just beyond an order of magnitude more stupid.
It is the biggest reason why their is no working fremantle
ported to any working Android hardware - too much fiddling around pumping this and that into an image and retry-recompile.

but, keep in mind, Intel is not far behind with the ATOM:

Re: Mer-based OS on a Tablet: what is needed?
 

Care to elaborate?

That link is a little light on details really, the USB issue is an issue with USB, and not really specific to Baytrail. 500ma is likely not enough to run the tablet, and shorting the data-lines is the standard way to signal to the tablet that it can draw more than 500ma, maybe USB3/C will improve this once we see more devices with that standard.

The GPS chips they put in them is a bit unfortunate, but not a showstopper for me.

That link is also 2 years old, has anything changed in this time regarding Baytrail devices?

Re: Mer-based OS on a Tablet: what would you like to see?
 

I was wondering. Why not build on work that has been done using chroot environment to put linux on top of (or coexist with) android. For example

http://linuxonandroid.org/

or maybe better (not sure...)

http://whiteboard.ping.se/Android/Debian

Some advantages are that there is little or no performance sacrifice. You can still use android apps, make calls use all social media apps that folks like on android and so forth while simultaneously having, say, debian linux or your favorite distro at your fingertips. Also, you will be able to take advantage of more of the latest hardware out there. Seems like if 5 to 10 people from maemo put effort into this great strides could be made in the short term. Then, maybe long term, try Mer-based OS or some approach that is more aesthetically pleasing from the ground up.

Re: Mer-based OS on a Tablet: what would you like to see?
 

Hmm. I kind of thought that Hybris was already following this concept, only at a lower level. :)

Well, it doesn't sound like this project is doing much better than Sailfish, though. This is from one of the latest blog entires (from back in February this year):

:(

This one seems interesting... I may give it a try, once I acquire an Android device to hack on. :)

Re: Mer-based OS on a Tablet: what would you like to see?
 

In regards to hybris I guess it is a question of where you want to start...

While significant progress was made over the years, developer interest has waned for Linux on Android app. Regardless you can access android apps while running a linux distribution. I actually have several images (ubuntu, fedora and archlinux) stored on my phone's microsd card that I fire up once in a while.
Here is an screenshot from my phone where I'm running ubuntu with LXDE. At the same time, I can open a floating window for the android dialpad from which I can make a call. Also you can connect, via micro usb to hdmi, to bigger screen so I can have a linux desktop when coupled to a bluetooth keyboard. But there are problems that need to be addressed. For example LXDE is not finger friendly (I use Mate DE for fedora and that is a little better). Drivers are missing so I can't, from the Ubuntu side of things, play music. Instead I would have to use an android app player .

The last option I mentioned is something I would like to take a closer look at as well. Just have to free up some time!

Re: Mer-based OS on a Tablet: what is needed?
 

[OT]to this thread, but this is exactly (ofc just one ;)) reason why I still use N900![/OT]

Re: Mer-based OS on a Tablet: what would you like to see?
 

Well, I've now gotten myself something of a Sailfish tablet, though very little works. :) (This is a Nexus 7, with Sailfish installed using the instructions found here.)

Next, I need to see if I can get Nemo working, and move up from there...

Re: Mer-based OS on a Tablet: what is needed?
 

I assume its why Hildon was so minimalistic. You had desktops, you configured them the way you wanted. I remember the thread with everybodies different home screens

Re: Mer-based OS on a Tablet: what would you like to see?
 

Hey, typing this now from Nexus 7 / Mer / Plasma Active 4, as per ruedigergad's instructions. I'll try to add a link later. Runs very nice! So, I've gotten about as far as 2013, now for the hard work of trying to get this guy running an up-to-date instance of Mer, if I can manage it. :)

EDIT: Link to ruedigergad's site. Plus, some pics. :)

Re: Mer-based OS on a Tablet: what is needed?
 

Okay, I gotta ask... when you say "most desirable" are you talking about the folks here or the current consumers? If the latter, I'd disagree. If the former, I agree 100%.

Agree.

So... lemme ask. Why hasn't this happened yet? Or has it and I'm just that unaware (very possible)

Re: Mer-based OS on a Tablet: what is needed?
 

Well, having spent a day or two now digging into the Mer documentation (such as it is), I've gotta say yeah, Mer is very nearly exactly that: everything you need to set up your own mobile device, except for the GUI. :) Really, I'm kinda surprised there haven't been more Jolla-like entities popping up, as Mer really does hand you a significant amount of power seemingly without constraining you to a particular type of UI or a particular corporate entity...

Re: Mer-based OS on a Tablet: what is needed?
 

I'm of course talking about my own needs, I couldn't possibly know or guess what other people need.
(and why should I care either, hmm? :D)

"current consumers" are 100% certified iSheep anyways.

Re: Mer-based OS on a Tablet: what is needed?
 

My tone tends to be stuck in "Devil's advocate" mode while here, but I'd have to say it's wise to know what the rest of the folks are doing; especially in tech. The majority is unfortunately dictating what options we, the very vocal minority of people (read: we don't need no stinkin' fart apps, give us terminal instead) will have access to unless we somehow find pockets of like-minded people... like here at TMO.

Read above... and for reasons ;)

And those sheeple are unfortunately who the majority of R&D and marketing and products are being pushed towards and we're piggybacking whatever is out there and shoehorning our wants in a very incomplete manner on existing hardware.

My point I want to really make is that without options that fit the minority in a grander manner (read: geeks want terminal, geeks get terminal) but in a more uniform manner lest we build a fragmented approach to handling our geekery and it'll never get addressed and the sheeple will be totally fulfilled. Can't sell a properly R&D device to just one person. That's why you should care and get other likeminded people that would actually buy the damn thing to care.

Come together on something, or keep bickering about (take your pick: shell, GUI, encryption cipher, distro, kernel, package management, etc) and ignoring everybody else and you will be left out.

That's what is happening. It's great to be an individual. But companies need to make profit. A whole bunch of folks not caring about each other reduces the chances of profitability to near zero.

Comes off soapbox...

I personally think that Copernicus is onto something. Mer is indeed powerful. And it's something that could be shoved onto many other devices and a ton of UI's slammed on top of it. Viola! Instant geek tablet.

But who'd buy it still remains. I, for the life of me cannot figure out any use cases for terminal only on a tablet - all of the virtual keyboards I've seen suck and slow me down and if I add a bluetooth keyboard, I might as well use my laptop.

tl;dr

I'd like to see a geek tablet come out though. I just question if it would sell. And I'd say with confidence, no it would not because geeks are just an unhappy bunch about much too damn much.

Re: Mer-based OS on a Tablet: what is needed?
 

Why, thank you!

But now, I'm confused. Maybe it's because I grew up with Linux; for most of the lifetime of Linux, there was never a device sold running the OS. In fact, there was never a device even intended to be usable with Linux. No, if you wanted a Linux PC, you purchased a PC designed for some other purpose, performed a lobotomy on it, and inserted the Linux OS yourself.

That is my intention here. I'm taking this nice old Nexus 7, lobotomizing it, and installing Mer. I'll probably play with a few different UIs, and maybe throw together something of my own. Eventually, I'll get another mobile device, lobotomize it, and carry on with more Mer goodness.

And, ultimately, I think this is the right way to go. Throughout the 80s and 90s and much of the 2000s, companies like Apple and Microsoft dumped millions (billions?) of dollars into operating systems for PCs, and made a fortune on them. But those OSs (DOS, Windows 3.1, Windows 9x, MacOS earlier than OS X) are all dead and gone now. Linux (and other open Unix-like OSs) continue to move ahead, despite not making a fortune. And while a lot of that is due to their open nature, much is also due to their not being tied to any particular architecture -- various machines come and go, but these OSs continue to adapt to new devices while still being able to run their standard software.

tl;dr: No need to wait for a geek tablet to come out, or to worry about how much geeks squabble with each other; every tablet is a geek tablet, so long as you can get an open OS onto it.

Re: Mer-based OS on a Tablet: what is needed?
 

See, sometimes I'm slow but I do get your intention a lot clearer now and yeah... I've been around Linux since Slackware 3.1 or so but I've always wanted devices that would remove some of the guesswork around drivers (especially wifi).

But you're right... the likelihood of something being Linux from the start on a device is relatively low.

However it's changing.

Re: Mer-based OS on a Tablet: what is needed?
 

Thanks. I need to look into this more. Has anyone got this to work on Samsung devices as well? Btw, I happen to have the Jolla launcher for android and thought I'd try it on my Galaxy Tab Pro. While a little buggy it basically worked. Too bad Jolla didn't continue with this. I actually like the transitions when you put an app in background better with this launcher than the current version of Sailfish. Also like the app drawer better. Can also open floating resizable apps with a side launcher on the tablet. Not bad but not the real thing if you know what I mean!

Re: Mer-based OS on a Tablet: what is needed?
 

I think you've answered your own question there. Nobody wants to be a 'Jolla-like' entity. That's a lot of work and evidently not sustainable.

Re: Mer-based OS on a Tablet: what is needed?
 

I wouldn't say that it's not sustainable. I think that it's quite the challenge if your name isn't Apple or Google right now.

But... you might also be onto something as well. How to change that?

Re: Mer-based OS on a Tablet: what would you like to see?
 

My ideal mobile (which also applies to tablets) is much like Sailfish with a few additions. No wasted real estate with a navigation bar and status bar rubbish, non-terrible scaling for different resolutions the default for the nexus 5 is soo huge and a massive waste of the resolution. (Would also prefer if the Sailfish UI was scaled a bit better too. The app covers I would prefer if they had both v1 and v2 features swiping and buttons for extra functionality. Swiping from the edge could also be enhanced, top half can do one thing and the bottom half could do something else, swipe to go back to the homescreen and go to the next/previous running app. These extra swipe gestures can work if the phone is big enough, so it could work that way for tablets and for phones with a pressure sensitive screen it could work with that instead.


A hard press could do a right click action. Since there's more space on a tablet you could have two fingers on the screen and tap with the second (have a way to detect that) and have that as the right action.

Well that's my ideal mobile OS, I have a few commitments such as my messaging app, the Android apps on Sailfish , and work. But I should have time at some point soon TM where I can help with this project eg. Implement encryption algorithm chacha20-poly1305 or the more conventional AES or something else entirely. Although I do concede that my implementation may not be the most optimal and the security aspects of the the implementation may also not be completely locked down. But it will be a start and can be improved later.

EDIT: I wrote this post with my phone and so may have some grammatical/spelling (and some things may just not make sense) which I blame entirely on the swipe keyboard. :) (and rightfully so)

I won't have access to a computer to amend this until the last week of the year.

EDIT2: Missed a feature that I would like. Widgets screen on the right from the main screen.

Re: Mer-based OS on a Tablet: what is needed?
 

Of course. At some point a startup has to start earning money. Jolla seem to have bet the company on selling OS licences with very few other income streams. No app store income. No accessories. No updated phones. No paid updates.

Re: Mer-based OS on a Tablet: what would you like to see?
 

I will be implementing chacha20-poly1305 for my messaging application in C++ at some point, that same code, if not slightly modified could also work for this project.

Re: Mer-based OS on a Tablet: what is needed?
 

Your news is encouraging,
hope you can let us know sometime a bullet list
of what works vs what fails
( nexus 7 should mostly work by now ?)