maemo.org - Talk

maemo.org - Talk (https://talk.maemo.org/index.php)
-   MeeGo / Harmattan (https://talk.maemo.org/forumdisplay.php?f=45)
-   -   MeeGo Reconstructed - a plan of action and direction for MeeGo (https://talk.maemo.org/showthread.php?t=78462)

w00t 2011-10-06 09:36

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by abill_uk (Post 1103922)
But but how on earth can an OS be designed around CLOSED components???

A lot of them do, because they *have* to, under the conditions of their license. The Linux kernel is GPLv2, which means that if you distribute binaries, you must also distribute your source.

Furthermore: not all devices require closed components at all. Think commodity hardware.

gerbick 2011-10-06 09:41

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
When will we see device ports outside of x86 and the Nokia N900? I would love to see this on a tablet.

abill_uk 2011-10-06 09:43

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by w00t (Post 1103924)
A lot of them do, because they *have* to, under the conditions of their license. The Linux kernel is GPLv2, which means that if you distribute binaries, you must also distribute your source.

Furthermore: not all devices require closed components at all. Think commodity hardware.

I really do appreciate your explanations because i am just not able to understand why the meego adaption has not made pace and what the problems were.

I am not trying to be unsulting in any way or form please realise this because it is purely lack of understanding why and what the meego.ce was actually meant to be.

smoku 2011-10-06 09:44

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by gerbick (Post 1103927)
When will we see device ports outside of x86 and the Nokia N900?

http://www.raspberrypi.org/?p=201

w00t 2011-10-06 09:48

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by abill_uk (Post 1103930)
I really do appreciate your explanations because i am just not able to understand why the meego adaption has not made pace and what the problems were.

Talking about MeeGo here - but to be clear - just the core.

MeeGo was not encouraging to device hackers. It had draconian binary requirements which made it very difficult for tinkerers to get it running on devices they had lying around. One more public example of this was in the SSSE3 requirement for x86, effectively excluding any older AMD hardware, along with quite a bit of older Intel hardware.

Mer is a lot more relaxed about this: there's a stock i486 port which should run on just about anything, and on the ARM side, I believe there's an armv6 port, which also widens the door to a hell of a lot more devices (like the n810).

So that's one aspect of where MeeGo failed: running on everything.

As for CE, its aim is to be a community handset UI, day to day usable. I don't know what your complaints with it are (or were) so I can't really address them, and nor do I think they're quite on-topic in this thread, but I do encourage you to get an SD and try the 1.3 release. Things have come a long way, and we've generally got a lot of good feedback, although there's clear room for improvement, and we're going to work our asses off to make that happen. Pop onto #meego-arm on freenode if you need help trying it on an n900.

abill_uk 2011-10-06 09:53

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by w00t (Post 1103935)
Talking about MeeGo here - but to be clear - just the core.

MeeGo was not encouraging to device hackers. It had draconian binary requirements which made it very difficult for tinkerers to get it running on devices they had lying around. One more public example of this was in the SSSE3 requirement for x86, effectively excluding any older AMD hardware, along with quite a bit of older Intel hardware.

Mer is a lot more relaxed about this: there's a stock i486 port which should run on just about anything, and on the ARM side, I believe there's an armv6 port, which also widens the door to a hell of a lot more devices (like the n810).

So that's one aspect of where MeeGo failed: running on everything.

As for CE, its aim is to be a community handset UI, day to day usable. I don't know what your complaints with it are (or were) so I can't really address them, and nor do I think they're quite on-topic in this thread, but I do encourage you to get an SD and try the 1.3 release. Things have come a long way, and we've generally got a lot of good feedback, although there's clear room for improvement, and we're going to work our asses off to make that happen. Pop onto #meego-arm on freenode if you need help trying it on an n900.

I think your explanations are VERY encouraging and makes a few things a lot clearer and i thank you muchly for your time to answer me.

erendorn 2011-10-06 10:00

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by abill_uk (Post 1103918)
Not much of this actually makes sense to me because it seems all self inflicting.

Meego is common core you say but what "core" was it? was it a UI/GUI or was it an OS?.
<snip more nonsense>

Read this .
Quote:

Device drivers are either integrated directly with the kernel or added as modules loaded while the system is running.
In Mer, device drivers are not integrated directly.
In Hardware adaptation, they are, for the Hardware they are adapted for (why doesn't that make sense to you?)

Quote:

The graphical user interface (or GUI) used by most Linux systems is built on top of an implementation of the X Window System.
In Mer the GUI is not. (but the X Window server is implemented)

Mer is all the rest. The OS core. What is not device or UI dependent.

Maybe you just don't realize that there is a enormous lot of code between the GUI and the drivers?

abill_uk 2011-10-06 10:10

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by erendorn (Post 1103945)
Read this .

In Mer, device drivers are not integrated directly.
In Hardware adaptation, they are, for the Hardware they are adapted for (why doesn't that make sense to you?)


In Mer the GUI is not. (but the X Window server is implemented)

Mer is all the rest. The OS core. What is not device or UI dependent.

Maybe you just don't realize that there is a enormous lot of code between the GUI and the drivers?

Oh i realise a lot more than you think i realise ;).

I only wish this to come together and yes i very know the amount of work involved even in making a basic device driver and really maybe you no not realise the appreciation people will get from a polished UI that is more user friendly than the one we got at the moment from Nokia on the Maemo OS of the N900.

Keep up the great work and try to have the gentlemany qualities that Woot has because he is really good at explaining almost everything ;) (you do have them from time to time).

erendorn 2011-10-06 10:23

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by abill_uk (Post 1103951)
Keep up the great work

I'm affraid none of this is my work, but I agree it's great ;)

Quote:

Originally Posted by abill_uk (Post 1103951)
and try to have the gentlemany qualities that Woot has because he is really good at explaining almost everything ;) (you do have them from time to time).

I'll try.

danramos 2011-10-06 19:19

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by abill_uk (Post 1103922)
But but how on earth can an OS be designed around CLOSED components???.

Do you really think that manufacturers will give you all source code to the devices they have put in there particular design? because without this there can not be an OS of any kind.

Or are you going to start from scratch with data sheets of every component?.

They, more or less, already do it for Android. If this works out well and adoption starts SOMEWHERE, I would expect that there's no reason not to expect similar support.

danramos 2011-10-06 19:32

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by smoku (Post 1103931)

Oooo... this is good. I'm encouraged by this. ...And you know what a cynical grumpy son of a b*tch I am. :)

smoku 2011-10-08 12:18

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by SD69 (Post 1102460)
Great, have you run this on your Ideapad?

http://www.youtube.com/watch?v=ZT_jVhuOeJQ
Downloadable from http://codex.xiaoka.com/pub/mer/

andrewfblack 2011-10-09 01:10

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
You know I still have all my old Mer Stickers, and Posters and other stuff I took to Linux Converances.

SD69 2011-10-10 12:31

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by w00t (Post 1103935)
Talking about MeeGo here - but to be clear - just the core.

Thanks for taking this initiative and posting about it on tmo. Since a week has passed, I have seen the proposal of an independent core met with some discussion and a proposal to include integration with active plasma.

I have read the meego mailing list. There is to be a decision on build system and how to make the project widely of interest.

Will you still be work with maemo community? and with Cordia project? You are certainly welcome to. We have funding, infrastructure, etc., and of course we were the original home of Mer. I encourage you to read our revised freedom in maemo for software development

w00t 2011-10-10 12:47

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by SD69 (Post 1106521)
Thanks for taking this initiative and posting about it on tmo. Since a week has passed, I have seen the proposal of an independent core met with some discussion and a proposal to include integration with active plasma.

The way things are looking, Plasma will indeed be building on top of a Mer core, as CE (and Cordia) are doing.

If all that pans out, users will be able to install Plasma/CE/Cordia on their devices [with Mer running under the hood, as an implementation detail]

Quote:

Originally Posted by SD69 (Post 1106521)
Will you still be work with maemo community? and with Cordia project? You are certainly welcome to. We have funding, infrastructure, etc., and of course we were the original home of Mer.

I personally do certainly see our goals (and more specifically, that of CE's) aligning with that of maemo.org, that is, a hackable free software stack usable on Nokia's devices.

We're not entirely bound within that space, though, so I don't think it's completely appropriate to just jump in and make ourselves at home: we aren't limiting ourselves to just the Nokia devices, and a lot of the software we're talking about differs.

In terms of infrastructure, etc, a lot of details still need to be worked out (i.e. how exactly Mer's repositories are going to be shared, etc), but currently things are looking like they can be managed quite well on a shoestring budget and ported around pretty easily, so I don't expect we have any needs (if any) - but thanks for the offer, it's nice to know we have alternatives if need be.

SD69 2011-10-10 18:53

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by w00t (Post 1106528)

I personally do certainly see our goals (and more specifically, that of CE's) aligning with that of maemo.org, that is, a hackable free software stack usable on Nokia's devices.

We're not entirely bound within that space, though, so I don't think it's completely appropriate to just jump in and make ourselves at home: we aren't limiting ourselves to just the Nokia devices, and a lot of the software we're talking about differs.

You'll probably see more of this in the months to come, but I think the current trajectory of the maemo community is to become more independent of Nokia and not limit itself to just past Nokia maemo devices.

Quote:

Originally Posted by w00t (Post 1106528)
In terms of infrastructure, etc, a lot of details still need to be worked out (i.e. how exactly Mer's repositories are going to be shared, etc), but currently things are looking like they can be managed quite well on a shoestring budget and ported around pretty easily, so I don't expect we have any needs (if any) - but thanks for the offer, it's nice to know we have alternatives if need be.

We would like to see Mer implemented and easy porting as well.

danramos 2011-10-11 00:22

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
Quote:

Originally Posted by SD69 (Post 1106701)
You'll probably see more of this in the months to come, but I think the current trajectory of the maemo community is to become more independent of Nokia and not limit itself to just past Nokia maemo devices.

We would like to see Mer implemented and easy porting as well.

Huge, strong YES to both of these points. My God, man.. Yes, please!

demolition 2011-10-11 00:50

Re: MeeGo Reconstructed - a plan of action and direction for MeeGo
 
If things come back to maemo and move forward as Mer, a lot of momentum will surely build. I'm sure a lot of the "i'm leaving" brigade peek in, through the one-way way glass of not signing in. If things are really stiring, then this could be quite an exiting place to be ... and may draw back old talent as well as new ... hopefully! Perhaps Meego folding was a good thing for maemo.org?

For those who'd like to know more and potentially offer some input, where's the best place to head at the moment? Everything seems quite fragmented at the moment so apologies if that info's clearly written somewhere!

Lastly, although some contributions to this thread have seemed quite ranty-shouty, getting the responses from the Mer team (and others) out in black and white, for all us maemo folks to see has been quite useful. Again, apologies if the re-iteration has given you sore throats. It was worth it, for the benefit of all the rest of us. Thank you.


All times are GMT. The time now is 00:03.

vBulletin® Version 3.8.8