Hello everyone.
I suppose many of you know the handheld games console project based on EOMA68 (the provisional name is ZEOMA).
The website (it is outdated (the images are also outdated), when the project is more advanced I'll update) (thanks to Peter Bouda) is: http://www.ubrew.it/
The features of the console are: - 4.5 inch 480x854 IPS screen. - Resistive touch panel. - DPad + A B X Y buttons + R button + L button + 2 Analog triggers + 2 Analog Joystick (with push button) + Start + Select + Home + Vol +/- + 2 extra buttons. - 4000 mAh battery. - Stereo. - MicroSD slot. - USB 2.0 Host. - MicroUSB (for charging). - STM32F072 for controls. - AR9271 WIFI.
I have done a small block diagram you can see here: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
I have placed the datasheets of the components here: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/
I have no prior experience designing schematics and PCB, but I will try to do my best. If I have any questions, I will ask you.
I will publish all the schematics and PCB under GPLv3+.
If anyone have any suggestions or improvement, I appreciate it.
Thanks to Luke for all his help. Also thanks to Alexander for allowing us to store the files of the project in his server.
What kind of games are you planning to run?
Because this is very libre focused project so I assume emulators of proprietary hardware and proprietary games for that hardware out of the question.
GaCuest:
Hello everyone.
I suppose many of you know the handheld games console project based on EOMA68 (the provisional name is ZEOMA).
The website (it is outdated (the images are also outdated), when the project is more advanced I'll update) (thanks to Peter Bouda) is: http://www.ubrew.it/
The features of the console are:
- 4.5 inch 480x854 IPS screen.
- Resistive touch panel.
- DPad + A B X Y buttons + R button + L button + 2 Analog triggers
- 2 Analog Joystick (with push button) + Start + Select + Home
- Vol +/- + 2 extra buttons.
- 4000 mAh battery.
- Stereo.
- MicroSD slot.
- USB 2.0 Host.
- MicroUSB (for charging).
- STM32F072 for controls.
- AR9271 WIFI.
I have done a small block diagram you can see here: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
I have placed the datasheets of the components here: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/
I have no prior experience designing schematics and PCB, but I will try to do my best. If I have any questions, I will ask you.
I will publish all the schematics and PCB under GPLv3+.
If anyone have any suggestions or improvement, I appreciate it.
Thanks to Luke for all his help. Also thanks to Alexander for allowing us to store the files of the project in his server.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
Actually, I personally take a hybdrid approach to my libre software. I focus on the important components being libre, but if some high-level software isn't, I can live with that as long as I'm not dependent on it.
So with that in mind, I know I would play some emulators on there. Also, if we just focus the EOMA68 projects on fully-libre usecases, we will miss out on many mainstream users who are necessary if we want the standard (or whatever revision its reached by then) to be at all relevant in 20 years. I like the idea of 100% libre hardware and a 100% libre software stack, but with easy options for the user to choose to add non-free components if they wish (like the Debian non-free repository being only a couple of clicks away)
Thanks
-Ryan
On 10/15/2016 12:52 AM, FaTony wrote:
What kind of games are you planning to run?
Because this is very libre focused project so I assume emulators of proprietary hardware and proprietary games for that hardware out of the question.
GaCuest:
Hello everyone.
I suppose many of you know the handheld games console project based on EOMA68 (the provisional name is ZEOMA).
The website (it is outdated (the images are also outdated), when the project is more advanced I'll update) (thanks to Peter Bouda) is: http://www.ubrew.it/
The features of the console are:
- 4.5 inch 480x854 IPS screen.
- Resistive touch panel.
- DPad + A B X Y buttons + R button + L button + 2 Analog triggers
- 2 Analog Joystick (with push button) + Start + Select + Home
- Vol +/- + 2 extra buttons.
- 4000 mAh battery.
- Stereo.
- MicroSD slot.
- USB 2.0 Host.
- MicroUSB (for charging).
- STM32F072 for controls.
- AR9271 WIFI.
I have done a small block diagram you can see here: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
I have placed the datasheets of the components here: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/
I have no prior experience designing schematics and PCB, but I will try to do my best. If I have any questions, I will ask you.
I will publish all the schematics and PCB under GPLv3+.
If anyone have any suggestions or improvement, I appreciate it.
Thanks to Luke for all his help. Also thanks to Alexander for allowing us to store the files of the project in his server.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
Debian's approach of this isn't really ethical. Anyone has the freedom of install installing propitiatory software even non tech users.
Including non-free software in the repository is insisting/proposing users, especially non tech ones, to give up their freedom for simplicity without understanding the importance of them and the technical problems that non free/libre software brings (a good example of that is the game modding community).
Mainstreams users like you seem to refer to them are what makes software and hardware go in decadence. I don't say that they are directly concerned, but it is how their were treated like, that made them what they are now and ask the same bad products.
If you continue to give them what they are made of the project will slowly become like them and only enforce the already bad circle.
Their are already lots of free software games, seek the libre game wiki encyclopedia. https://libregamewiki.org/Main_Page
Note: emulators can be free software but the roms/blobs shouldn’t be included, f-droid already does that.
Le 15/10/2016 08:24, ryan a écrit :
Actually, I personally take a hybdrid approach to my libre software. I focus on the important components being libre, but if some high-level software isn't, I can live with that as long as I'm not dependent on it.
So with that in mind, I know I would play some emulators on there. Also, if we just focus the EOMA68 projects on fully-libre usecases, we will miss out on many mainstream users who are necessary if we want the standard (or whatever revision its reached by then) to be at all relevant in 20 years. I like the idea of 100% libre hardware and a 100% libre software stack, but with easy options for the user to choose to add non-free components if they wish (like the Debian non-free repository being only a couple of clicks away)
Thanks
-Ryan
On 10/15/2016 12:52 AM, FaTony wrote:
What kind of games are you planning to run?
Because this is very libre focused project so I assume emulators of proprietary hardware and proprietary games for that hardware out of the question.
GaCuest:
Hello everyone.
I suppose many of you know the handheld games console project based on EOMA68 (the provisional name is ZEOMA).
The website (it is outdated (the images are also outdated), when the project is more advanced I'll update) (thanks to Peter Bouda) is: http://www.ubrew.it/
The features of the console are:
- 4.5 inch 480x854 IPS screen.
- Resistive touch panel.
- DPad + A B X Y buttons + R button + L button + 2 Analog triggers
- 2 Analog Joystick (with push button) + Start + Select + Home
- Vol +/- + 2 extra buttons.
- 4000 mAh battery.
- Stereo.
- MicroSD slot.
- USB 2.0 Host.
- MicroUSB (for charging).
- STM32F072 for controls.
- AR9271 WIFI.
I have done a small block diagram you can see here: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
I have placed the datasheets of the components here: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/
I have no prior experience designing schematics and PCB, but I will try to do my best. If I have any questions, I will ask you.
I will publish all the schematics and PCB under GPLv3+.
If anyone have any suggestions or improvement, I appreciate it.
Thanks to Luke for all his help. Also thanks to Alexander for allowing us to store the files of the project in his server.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On 10/15/16, mdn bernardlprf@openmailbox.org wrote:
Debian's approach of this isn't really ethical.
they're doing the best that they believe they can do, but they _have_ been told. see joey hess's very public description of the Debian Charter as a "toxic document".
i've spoken to the FSF about this: from what i gather, the changes required are actually very very simple: all they have to do is add in a simple popup message whenever someone clicks the "nonfree" section, issuing a warning to the end-user that the consequences of their actions are leading them into unethical territory.
... how simple would that be to add?
the other parts (creating separate DNS names and different repositories for the nonfree sections) could be done transparently with HTTP rewrites and redirects (just like devuan seems to be doing) as an interim measure, then removed at some appropriate point after a couple of major releases.
it's really, really not very hard, and we'd end up with Debian - one of the world's leading Software Libre OSes - being RYF Compliant.
as it is, we have to fuck around forking tens of THOUSANDs of packages, with efforts to do so failing under the sheer weight of the task and the required resources.
i really really wish the debian group would wake up, just a little bit.
*sigh*.
l.
Le 15/10/2016 16:40, Luke Kenneth Casson Leighton a écrit :
On 10/15/16, mdn bernardlprf@openmailbox.org wrote:
Debian's approach of this isn't really ethical.
they're doing the best that they believe they can do, but they _have_ been told. see joey hess's very public description of the Debian Charter as a "toxic document".
Tanks for the reference
i've spoken to the FSF about this: from what i gather, the changes required are actually very very simple: all they have to do is add in a simple popup message whenever someone clicks the "nonfree" section, issuing a warning to the end-user that the consequences of their actions are leading them into unethical territory.
... how simple would that be to add?
Not a all, but that was a rhetoric question. I am looking forward to make pedagogic help for basic users, I was one myself and I know that even a bit of more help (more that just a popup) isn’t that hard to help, it's just very time consuming.
the other parts (creating separate DNS names and different repositories for the nonfree sections) could be done transparently with HTTP rewrites and redirects (just like devuan seems to be doing) as an interim measure, then removed at some appropriate point after a couple of major releases.
it's really, really not very hard, and we'd end up with Debian - one of the world's leading Software Libre OSes - being RYF Compliant.
as it is, we have to fuck around forking tens of THOUSANDs of packages, with efforts to do so failing under the sheer weight of the task and the required resources.
Agreed
i really really wish the debian group would wake up, just a little bit.
((trolling) same thing with systemd I wish debian would wake up)
Sorry I couldn't resist ^^
*sigh*.
l.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On Sat, Oct 15, 2016 at 03:40:10PM +0100, Luke Kenneth Casson Leighton wrote:
On 10/15/16, mdn bernardlprf@openmailbox.org wrote:
Debian's approach of this isn't really ethical.
they're doing the best that they believe they can do, but they _have_ been told. see joey hess's very public description of the Debian Charter as a "toxic document".
i've spoken to the FSF about this: from what i gather, the changes required are actually very very simple: all they have to do is add in a simple popup message whenever someone clicks the "nonfree" section, issuing a warning to the end-user that the consequences of their actions are leading them into unethical territory.
Debian and the FSF have agreed to differ on this: Debian folk have problems with GFDL with invariant sections, for example. Ask John Sullivan what the FSF posiiton is.
Non-free is NOT a part of Debian, nor is contrib - but they are provided as a covenience for users. It's also worth knowing that security updates for non-free are almost impossible.
BUT ... If you've got a Broadcom chipset, for example, you may have no option but to use proprietary software. Most Intel wifi chips also require firmware - what are you going to do when that's emebedded in a new laptop / nettop ?
They repostiories do have to be explicitly enabled: the question of whether you want to install non-free software is asked explicitly in the installer - so the notifications are there.
Ironically, if wifi adapters / Ethernet cards still came with burnt-in firmware, Debian would be a fully free distribution (and it's worth remembering that Debian was endorsed and funded by the FSF for a while).
If you want any architecture other than Intel / AMD as a primary supported architecture then your choice is prety much Debian from the mainstream distributions and Trisquel / GNewsense are forks which don't yet support all other architectures. So, if you want to do work to enable your project on a Cubietruck - you use Debian, probably.
Andy C
NOT SPEAKING FOR DEBIAN PROJECT AS A WHOLE :)
... how simple would that be to add?
the other parts (creating separate DNS names and different repositories for the nonfree sections) could be done transparently with HTTP rewrites and redirects (just like devuan seems to be doing) as an interim measure, then removed at some appropriate point after a couple of major releases.
it's really, really not very hard, and we'd end up with Debian - one of the world's leading Software Libre OSes - being RYF Compliant.
as it is, we have to fuck around forking tens of THOUSANDs of packages, with efforts to do so failing under the sheer weight of the task and the required resources.
i really really wish the debian group would wake up, just a little bit.
*sigh*.
l.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
Luke Kenneth Casson Leighton wrote:
they're doing the best that they believe they can do, but they _have_ been told. see joey hess's very public description of the Debian Charter as a "toxic document".
I've seen https://lists.debian.org/debian-devel/2014/11/msg00174.html where Hess makes this statement but I haven't seen anything written by Hess clearly explaining why the Debian Constitution is "toxic".
Where would I find something written by Hess clearly explaining why the Debian Constitution is "toxic"?
i've spoken to the FSF about this: from what i gather, the changes required are actually very very simple: all they have to do is add in a simple popup message whenever someone clicks the "nonfree" section, issuing a warning to the end-user that the consequences of their actions are leading them into unethical territory.
... how simple would that be to add?
It's entirely possible something has changed and I am not aware of relevant updates on this (I don't doubt you're in touch with them far more than I am). Please do reply to the list with updates to this situation.
But according to published documents I point to below, a popup might be quite simple to add but insufficient to allow Debian GNU/Linux to appear on the list of FSF Free System Distributions. I'll explain why I believe this to be true.
In https://www.gnu.org/distros/common-distros.html we find the following objection, "Debian also provides a repository of nonfree software. According to the project, this software is "not part of the Debian system," but the repository is hosted on many of the project's main servers, and people can readily find these nonfree packages by browsing Debian's online package database and its wiki".
John Sullivan went into more detail on the FSF's objection at Debconf2015:
So, in Debian's case, the lack of endorsement from us is primarily because of the relationship between official Debian and unofficial Debian -- the 'non-free' and 'contrib' repositories. And that relationship to us seems too close for our comfort. There are spots in the Debian infrastructure where those sections even though technically separate are integrated very closely with main. So, for example, in package searching, in 'recommends' and 'suggests' fields within packages that are displayed to users. So even though, in Debian, we have an idea that these are separate that's not always as clear to users on the outside and they can end up being sometimes inadvertently or sometimes just led to install nonfree components on top of the official distribution.
Source: http://meetings-archive.debian.net/pub/debian-meetings/2015/debconf15/Debian... (12m18s)
I believe the FSF is right to point out Debian's cognitive dissonance. Debian gets to:
- host repos containing nonfree software, - include UI with pointers to said repos in the installed repo list, - list packages from the nonfree repos as alternatives to free software packages, - and also claim that these repos are somehow "not part of the Debian system"?
I too believe that Debian is hosting nonfree software and integrating nonfree software with free software and this is indistinguishable from what other distros not listed do (such as Ubuntu's GNU/Linux).
If Debian wanted the FSF's approval Debian could remove the nonfree and contrib repos from Debian entirely, and remove mentions of packages from these repos from the free packages. Any packages one installs from Debian's repos post-installation would have the same restrictions too (thus addressing what Sullivan mentioned immediately after the above quote).
It was good of Debian to move the nonfree blobs to the nonfree and/or contrib repos in Debian 6.0 ("squeeze") in February 2011 but the OS installer makes the same kinds of recommendations the FSF objects to. I understand the consequences for users looking to most conveniently install Debian GNU/Linux plus whatever nonfree software to let the OS run on their hardware. But I don't see a popup fixing this. I see this as another convenience vs. software freedom tradeoff (wherein security is certainly on the side of software freedom too).
Repo redirects to sets of packages that only mention free software packages with no references to nonfree software could work but that still involves providing work for thousands of packages, as you say.
On Sat, Oct 15, 2016 at 04:47:05PM -0500, J.B. Nicholson wrote:
Luke Kenneth Casson Leighton wrote:
I've seen https://lists.debian.org/debian-devel/2014/11/msg00174.html where Hess makes this statement but I haven't seen anything written by Hess clearly explaining why the Debian Constitution is "toxic".
i've spoken to the FSF about this: from what i gather, the changes required are actually very very simple: all they have to do is add in a simple popup message whenever someone clicks the "nonfree" section, issuing a warning to the end-user that the consequences of their actions are leading them into unethical territory.
... how simple would that be to add?
Pick up the Debian netinst iso / the first Debian CD / the first Debian DVD.
You can install an entirely free system with no non-free components.
You can also install Debian without taking account of any recommends.
On (both) the Thinkpads in front of me, that would result in non-working wifi but everything else would work. I could plug in one of a few wifi dongles and have a fully free Debian.
On the Intel desktop machine away behind me I couldn't get hardware acceleration on the Nvidia card - I could care less.
On a Cubietruck / Pine64 / Chip / Raspberry Pi / Pi3 - I couldn't get functionality without non-free which I could get with Allwinner / Broadcom firmware. Debian doesn't supply "non-free" components: in each case you're using firmware distributed with the hardware. Without non-free firmware / forked kernels, all of the ARM hardware we have is pretty much unusable. I'm hopeful that you can prove differently Luke.
But according to published documents I point to below, a popup might be quite simple to add but insufficient to allow Debian GNU/Linux to appear on the list of FSF Free System Distributions. I'll explain why I believe this to be true.
In https://www.gnu.org/distros/common-distros.html we find the following objection, "Debian also provides a repository of nonfree software. According to the project, this software is "not part of the Debian system," but the repository is hosted on many of the project's main servers, and people can readily find these nonfree packages by browsing Debian's online package database and its wiki".
John Sullivan went into more detail on the FSF's objection at Debconf2015:
So, in Debian's case, the lack of endorsement from us is primarily because of the relationship between official Debian and unofficial Debian -- the 'non-free' and 'contrib' repositories. And that relationship to us seems too close for our comfort. There are spots in the Debian infrastructure where those sections even though technically separate are integrated very closely with main. So, for example, in package searching, in 'recommends' and 'suggests' fields within packages that are displayed to users. So even though, in Debian, we have an idea that these are separate that's not always as clear to users on the outside and they can end up being sometimes inadvertently or sometimes just led to install nonfree components on top of the official distribution.
Source: http://meetings-archive.debian.net/pub/debian-meetings/2015/debconf15/Debian... (12m18s)
Where would you suggest that Debian point users with unusable hardware - note (_users_ not developers) ?
It's very clear on the website and in documentation back to 1994
www.debian.org/CD/netinst - no mention of non-free
https://www.debian.org/CD/faq#official - unofficial CDs may contain additional hardware drivers, or additional software packages not part of the archive.
I believe the FSF is right to point out Debian's cognitive dissonance. Debian gets to:
- host repos containing nonfree software,
- include UI with pointers to said repos in the installed repo list,
- list packages from the nonfree repos as alternatives to free software
packages,
- and also claim that these repos are somehow "not part of the Debian system"?
I too believe that Debian is hosting nonfree software and integrating nonfree software with free software and this is indistinguishable from what other distros not listed do (such as Ubuntu's GNU/Linux).
If Debian wanted the FSF's approval Debian could remove the nonfree and contrib repos from Debian entirely, and remove mentions of packages from these repos from the free packages. Any packages one installs from Debian's repos post-installation would have the same restrictions too (thus addressing what Sullivan mentioned immediately after the above quote).
It was good of Debian to move the nonfree blobs to the nonfree and/or contrib repos in Debian 6.0 ("squeeze") in February 2011 but the OS installer makes the same kinds of recommendations the FSF objects to. I understand the consequences for users looking to most conveniently install Debian GNU/Linux plus whatever nonfree software to let the OS run on their hardware. But I don't see a popup fixing this. I see this as another convenience vs. software freedom tradeoff (wherein security is certainly on the side of software freedom too).
Repo redirects to sets of packages that only mention free software packages with no references to nonfree software could work but that still involves providing work for thousands of packages, as you say.
Genuinely: run through a Debian install from the netinst / CDs. Please point out to me where non-free software will be installed without an explicit action to include nonfree software on the part of the person installing. The screen mentioning non-free mentions that hardware drivers that may be required may be non-free but you have to opt in to install them.
All the best
Andy C
[still not speaking for the Debian project]
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
Andrew M.A. Cater wrote:
You can install an entirely free system with no non-free components.
You can also install Debian without taking account of any recommends.
But the recommends and suggests fields are still listing nonfree software, which was the FSF's issue. Not accepting the suggestions or recommendations doesn't address the issue the FSF raised in Sullivan's DebConf talk.
Where would you suggest that Debian point users with unusable hardware - note (_users_ not developers) ?
Developers are users too. But where I would point them doesn't matter. These are the FSF's requirements we're talking about. Although I don't speak for the FSF, I believe they'd point any computer user to the FSF's "Respects Your Freedom" hardware (such as what the FSF itself uses) and I believe they'd point out that sometimes freedom requires a sacrifice (as rms points out in all of his talks going back many years). One might not be able to use just any hardware with a Debian GNU/Linux system that satisfies the FSF's recommended distro list.
Genuinely: run through a Debian install from the netinst / CDs. Please point out to me where non-free software will be installed without an explicit action to include nonfree software on the part of the person installing. The screen mentioning non-free mentions that hardware drivers that may be required may be non-free but you have to opt in to install them.
Which suggests the nonfree software integration the FSF spoke of is in there. After all, like you just said, if it's an opt-in away to get the nonfree software the nonfree repos are listed but not enabled until one answers "yes" to activate the nonfree repos Debian hosts. If this isn't the case, and the FSF's requests are being met it's a simple matter for someone from Debian to submit the latest Debian GNU/Linux for a proper review and possible inclusion on the list.
On 2016-10-15 at 18:06:52 -0500, J.B. Nicholson wrote:
Andrew M.A. Cater wrote:
You can install an entirely free system with no non-free components.
You can also install Debian without taking account of any recommends.
But the recommends and suggests fields are still listing nonfree software, which was the FSF's issue. Not accepting the suggestions or recommendations doesn't address the issue the FSF raised in Sullivan's DebConf talk.
Suggests, yes, but Recommends to software in non-free shouldn't be there as they are forbidden by the policy
https://www.debian.org/doc/debian-policy/ch-archive.html#s-main
If you find one, please file a bug so that it can be removed (either because the Recommends wasn't really supposed to be there, or by moving the package to contrib, if it really needs non-free software to work)
Which suggests the nonfree software integration the FSF spoke of is in there. After all, like you just said, if it's an opt-in away to get the nonfree software the nonfree repos are listed but not enabled until one answers "yes" to activate the nonfree repos Debian hosts. If this isn't the case, and the FSF's requests are being met it's a simple matter for someone from Debian to submit the latest Debian GNU/Linux for a proper review and possible inclusion on the list.
I wish Debian and the FSF would work together to resolve this issue. It shouldn't be that hard to modify Debian so that `non-free` is only ever used based on an explicit user request (and to let the user specify that this explicit request only applies this one time).
Along the same lines, the `non-free` section should be split in two: `proprietary`, `non-dfsg`, where the `non-dfsg` part would only contain packages which the DFSG rejects as non-free but which many people in the Free Software world consider Free nevertheless (basically FSF's docs).
Stefan
On 2016-10-17 at 10:02:26 -0400, Stefan Monnier wrote:
I wish Debian and the FSF would work together to resolve this issue.
They are, more or less: there has been quite some activity a few years ago which lead to some changes, but work seems to have stalled (the `mailing list`_ isn't seeing much traffic lately)
.. _`mailing list`: https://lists.alioth.debian.org/mailman/listinfo/fsf-collab-discuss
I suspect that what changes could be agreed on have been done, while most other cases are points where they had to agree to disagree, such as the freedom status of the FSF docs and the existence of non-free.
I see that there has been a BOF_ about collaboration between Debian and the FSF at the latest Debconf, but I haven't seen the video, so I don't know what was said (yet, I may have just found something to watch in the near future)
It shouldn't be that hard to modify Debian so that `non-free` is only ever used based on an explicit user request (and to let the user specify that this explicit request only applies this one time).
It is, already. users already have to explicitely accept (in some cases that involve hardware support) or request (in all other cases) that non-free is enabled.
There is disagreement on how hard it should be to do so, with the FSF considering what Debian choose to do "too easy".
Along the same lines, the `non-free` section should be split in two: `proprietary`, `non-dfsg`, where the `non-dfsg` part would only contain packages which the DFSG rejects as non-free but which many people in the Free Software world consider Free nevertheless (basically FSF's docs).
If something is not-DFSG is by definition proprietary as far as Debian is concerned.
There have been talks about dividing non-free, however, splitting out the firmwares (that lots of people consider a necessary evil for another few years), documentation (for which some people including the FSF tend to have lower requirements) and everything else (the really evil stuff)
There was agreement on this split, but I suspect that it has been stuck in a lack of volunteer time.
On 2016-10-17 at 19:46:21 +0200, Elena ``of Valhalla'' wrote:
I see that there has been a BOF_ about collaboration between Debian and the FSF at the latest Debconf, but I haven't seen the video, so I don't know what was said (yet, I may have just found something to watch in the near future)
apparently I a) forgot the link, which is https://debconf16.debconf.org/talks/91/ b) forgot that I did try to watch that video, but it's missing the first 20 minutes or so of audio (it's in the known issues at http://ftp.acc.umu.se/pub/debian-meetings/2016/debconf16/README.txt )
Of course, you can install a fully free Debian system, but 1 single dialog in setup wizard is a bit too little.
I would rather have the tickbox to install non-free repos somewhere deep in preferences menu and I would certainly not host them on the debian.org domain.
Ideally, you would only add non-free repo by manually editing sources.list.
Andrew M.A. Cater:
On Sat, Oct 15, 2016 at 04:47:05PM -0500, J.B. Nicholson wrote:
Luke Kenneth Casson Leighton wrote:
I've seen https://lists.debian.org/debian-devel/2014/11/msg00174.html where Hess makes this statement but I haven't seen anything written by Hess clearly explaining why the Debian Constitution is "toxic".
i've spoken to the FSF about this: from what i gather, the changes required are actually very very simple: all they have to do is add in a simple popup message whenever someone clicks the "nonfree" section, issuing a warning to the end-user that the consequences of their actions are leading them into unethical territory.
... how simple would that be to add?
Pick up the Debian netinst iso / the first Debian CD / the first Debian DVD.
You can install an entirely free system with no non-free components.
You can also install Debian without taking account of any recommends.
On (both) the Thinkpads in front of me, that would result in non-working wifi but everything else would work. I could plug in one of a few wifi dongles and have a fully free Debian.
On the Intel desktop machine away behind me I couldn't get hardware acceleration on the Nvidia card - I could care less.
On a Cubietruck / Pine64 / Chip / Raspberry Pi / Pi3 - I couldn't get functionality without non-free which I could get with Allwinner / Broadcom firmware. Debian doesn't supply "non-free" components: in each case you're using firmware distributed with the hardware. Without non-free firmware / forked kernels, all of the ARM hardware we have is pretty much unusable. I'm hopeful that you can prove differently Luke.
But according to published documents I point to below, a popup might be quite simple to add but insufficient to allow Debian GNU/Linux to appear on the list of FSF Free System Distributions. I'll explain why I believe this to be true.
In https://www.gnu.org/distros/common-distros.html we find the following objection, "Debian also provides a repository of nonfree software. According to the project, this software is "not part of the Debian system," but the repository is hosted on many of the project's main servers, and people can readily find these nonfree packages by browsing Debian's online package database and its wiki".
John Sullivan went into more detail on the FSF's objection at Debconf2015:
So, in Debian's case, the lack of endorsement from us is primarily because of the relationship between official Debian and unofficial Debian -- the 'non-free' and 'contrib' repositories. And that relationship to us seems too close for our comfort. There are spots in the Debian infrastructure where those sections even though technically separate are integrated very closely with main. So, for example, in package searching, in 'recommends' and 'suggests' fields within packages that are displayed to users. So even though, in Debian, we have an idea that these are separate that's not always as clear to users on the outside and they can end up being sometimes inadvertently or sometimes just led to install nonfree components on top of the official distribution.
Source: http://meetings-archive.debian.net/pub/debian-meetings/2015/debconf15/Debian... (12m18s)
Where would you suggest that Debian point users with unusable hardware - note (_users_ not developers) ?
It's very clear on the website and in documentation back to 1994
www.debian.org/CD/netinst - no mention of non-free
https://www.debian.org/CD/faq#official - unofficial CDs may contain additional hardware drivers, or additional software packages not part of the archive.
I believe the FSF is right to point out Debian's cognitive dissonance. Debian gets to:
- host repos containing nonfree software,
- include UI with pointers to said repos in the installed repo list,
- list packages from the nonfree repos as alternatives to free software
packages,
- and also claim that these repos are somehow "not part of the Debian system"?
I too believe that Debian is hosting nonfree software and integrating nonfree software with free software and this is indistinguishable from what other distros not listed do (such as Ubuntu's GNU/Linux).
If Debian wanted the FSF's approval Debian could remove the nonfree and contrib repos from Debian entirely, and remove mentions of packages from these repos from the free packages. Any packages one installs from Debian's repos post-installation would have the same restrictions too (thus addressing what Sullivan mentioned immediately after the above quote).
It was good of Debian to move the nonfree blobs to the nonfree and/or contrib repos in Debian 6.0 ("squeeze") in February 2011 but the OS installer makes the same kinds of recommendations the FSF objects to. I understand the consequences for users looking to most conveniently install Debian GNU/Linux plus whatever nonfree software to let the OS run on their hardware. But I don't see a popup fixing this. I see this as another convenience vs. software freedom tradeoff (wherein security is certainly on the side of software freedom too).
Repo redirects to sets of packages that only mention free software packages with no references to nonfree software could work but that still involves providing work for thousands of packages, as you say.
Genuinely: run through a Debian install from the netinst / CDs. Please point out to me where non-free software will be installed without an explicit action to include nonfree software on the part of the person installing. The screen mentioning non-free mentions that hardware drivers that may be required may be non-free but you have to opt in to install them.
All the best
Andy C
[still not speaking for the Debian project]
"J.B. Nicholson" jbn@forestfield.org writes:
Luke Kenneth Casson Leighton wrote:
they're doing the best that they believe they can do, but they _have_ been told. see joey hess's very public description of the Debian Charter as a "toxic document".
I've seen https://lists.debian.org/debian-devel/2014/11/msg00174.html where Hess makes this statement but I haven't seen anything written by Hess clearly explaining why the Debian Constitution is "toxic".
Where would I find something written by Hess clearly explaining why the Debian Constitution is "toxic"?
Yes, it was the Debian Constitution he was referring to. I'm not really sure why this is relevant to the discussion of free software, but I suspect that Luke is conflating it with the Social Contract, and calling it "Debian Charter" which is ... not a thing.
I think Joey was saying that the constitutions existence has resulted in some people having endless discussions about the internal structures of Debian, rather than getting on with something useful instead.
It has absolutely nothing to do with what Luke seems to be suggesting.
As for the non-free thing and the FSF -- changing things would require Debian to consider that to be a good idea, which was certainly not the case in 2004:
https://www.debian.org/vote/2004/vote_002
I doubt that opinion has changed.
Claiming that is related to being unethical, rather than a result of people having differing proprieties, strikes me as rather childish.
On this laptop, I note that I have 4 packages installed from "non-free". One is firmware-iwlwifi, and the other 3 are GFDL licensed docs with invariant sections. I suppose I could buy another wifi card (perhaps one with the same chipset, with the same firmware, in a ROM?).
Then I could chuck the old card into landfill for an "ethical" outcome.
Cheers, Phil.
On 10/16/16, Philip Hands phil@hands.com wrote:
"J.B. Nicholson" jbn@forestfield.org writes:
Luke Kenneth Casson Leighton wrote:
they're doing the best that they believe they can do, but they _have_ been told. see joey hess's very public description of the Debian Charter as a "toxic document".
I've seen https://lists.debian.org/debian-devel/2014/11/msg00174.html where Hess makes this statement but I haven't seen anything written by Hess clearly explaining why the Debian Constitution is "toxic".
Where would I find something written by Hess clearly explaining why the Debian Constitution is "toxic"?
Yes, it was the Debian Constitution he was referring to. I'm not really sure why this is relevant to the discussion of free software, but I
i'm beginning to appreciate that everything we're doing boils down to the extremely rare combination of applying ethics to software. where we decide to "draw the line" on those ethics is where various groups involved in free software (and "open source") is where we differ.
suspect that Luke is conflating it with the Social Contract, and calling it "Debian Charter" which is ... not a thing.
yes. thanks for clarifying. so much to do, covering so many things, i can't possibly recall all the details at the time that they're needed, so thank you.
I think Joey was saying that the constitutions existence has resulted in some people having endless discussions about the internal structures of Debian, rather than getting on with something useful instead.
It has absolutely nothing to do with what Luke seems to be suggesting.
As for the non-free thing and the FSF -- changing things would require Debian to consider that to be a good idea, which was certainly not the case in 2004:
https://www.debian.org/vote/2004/vote_002
I doubt that opinion has changed.
Claiming that is related to being unethical, rather than a result of people having differing proprieties, strikes me as rather childish.
On this laptop, I note that I have 4 packages installed from "non-free". One is firmware-iwlwifi, and the other 3 are GFDL licensed docs with invariant sections. I suppose I could buy another wifi card (perhaps one with the same chipset, with the same firmware, in a ROM?).
or one where the source code of the firmware is entirely libre.
Then I could chuck the old card into landfill for an "ethical" outcome.
indeed. ha. i like the irony of throwing the old one away.
you could view that action (replacing the card) as being one of convenience. thinkpenguin have a stack of available cards (just for goodness sake get the right one.... there's a "standard" that isn't actually a standard..)
apologies for explaining this if you're already aware of it phil (i'm explaining for other people's benefit) but if you got one of those cards, then when you next come to upgrade, you like many people who buy thinkpenguin's products that "just work", any issues with the nonfree firmware being incompatible with the kernel as it was being upgraded (or other similar issues) would *not happen*.
this "software libre is actually about taking away the stress and inconvenience" is something that i really did not appreciate until chris explained thinkpenguin's business model to me.
chris worked for linspire as a QA engineer. he got to see first-hand that linspire's chances of ever being a pre-installed OS shipped out by default along-side (or instead of) Windows was utterly negligeable. that there was no chance whatsoever of winmodems working on linux-based distros, and so on.
thus he formed the idea to *pre-vet* hardware and *only* sell a comprehensive range of *pre-tested* products that have full libre firmware (if any is required at all). as a result, he continues to support 15-year-old distros to this day, and supplies *one percent* of the world's WIFI dongles, which is an amazing achievement for a company that only employs three people.
l.
On 2016-10-15 16:47 -0500, J.B. Nicholson wrote:
I too believe that Debian is hosting nonfree software and integrating nonfree software with free software and this is indistinguishable from what other distros not listed do (such as Ubuntu's GNU/Linux).
There is a difference: Ubuntu will install non-free software (firmware needed to make hardware work, and binary drivers) by default. Debian will not do so unless the user adds the non-free repository (SFAIK).
If Debian wanted the FSF's approval Debian could remove the nonfree and contrib repos from Debian entirely, and remove mentions of packages from these repos from the free packages.
They (we) could. Ironically a large fraction of the packages in non-free are FSF documentation. Most of the rest is firmware blobs (usually running on a different CPU from your main one).
Debian has taken the view for many years now that having this stuff available relatively painlessly is the right balance between usability and freedom. There have been suggestions made about putting firmware in a different categary, as that's the main reason people enable non-free, and once enabled you get all of it available, not just the one or two bits you needed, and it would be good if it was less all-or-nothing. Not sure where those changes got to.
I have 11 non-free packages. 5 of them are FSF documentation (gcc x2, make, cpio, emacs). 4 others are firmware for this thinkpad (bluetooth and wifi). In fact I even made and uploaded a non-free package: cpio-doc as the cpio docs were not available on Debian without someone doing that (have you tried using cpio without the docs? - it's hard going). The others are tools installed for interacting with other people, where free alternatives do not exist: unrar, nautilus-dropbox.
None of that is particularly unethical, except maybe the last two packages, but whilst I am a big supporter of free software (that's why I'm on this list), I think it's OK that Debian manages this stuff properly for when one needs it: it's much better than having to go find random binaries on line to install, for example.
I've just installed unrar-free, and removed nautilus-dropbox, as these days one can use some non-free software online to do the same job, when tiresome people send you things via that service. That's more 'FSF pure', but I don't think it really makes much ethical difference: Dropbox is proprietary however you access it.
Wookey
On 10/16/16, Wookey wookey@wookware.org wrote:
On 2016-10-15 16:47 -0500, J.B. Nicholson wrote:
I too believe that Debian is hosting nonfree software and integrating nonfree software with free software and this is indistinguishable from what other distros not listed do (such as Ubuntu's GNU/Linux).
There is a difference: Ubuntu will install non-free software (firmware needed to make hardware work, and binary drivers) by default. Debian will not do so unless the user adds the non-free repository (SFAIK).
appreciated you pointing out the distinction / differenec, wookey. i only became aware of the FSF's position from my conversations with josh gay, some eight or so months ago. i've mentioned them on here before: you may have missed them (i'm aware you're a lurker) so am happy to repeat them in this context.
the FSF figures that technically-competent people can look after themselves. "technically-competent" is defined as loosely fitting with "someone who has the capacity to take the initiative to seek out help online or from friends, where such help requires *explicit* following (and trust of) *specific* instructions, step-by-step without deviation or elaboration, usually at the command-line".
the FSF's position there covers *everyone else*, who, by definition, cannot trust or be trusted to follow explicit written or verbal instructions, cannot cope with a command-line prompt, cannot comprehend the consequences of their actions, does not understand or read "terms and conditions" and so on.
these are the people whom the FSF's position protects (from themselves) - they are the people who are extremely likely to go *click* synaptics package manager what's that it's not enabled *click* i wonder what nonfree is don't understand don't care oh well let's enable it anyway *click* oh look there's these extra packages i wonder what they do *click* and now they've opened up a means to compromise their computer and their privacy without *ever* encountering a warning that that was even possible.
*that's* what the FSF objects to about debian. it's not that the packages *are* separate, it's that it's *too easy* to install them without any warning of any kind, whatsoever.
we as technical people just go in and edit /etc/apt/sources.list and add "nonfree" to the end of the appropriate deb line. *non-technical* people run synaptics and its ilke, where there's a GUI-based no-warnings-whatsoever option *right there* in the menus / dialogs, to enable non-free repositories.
anyway. thank you for making me aware that FSF documentation is qualified as non-free, that really made my day.
l.
Luke Kenneth Casson Leighton lkcl@lkcl.net writes:
...
the FSF's position there covers *everyone else*, who, by definition, cannot trust or be trusted to follow explicit written or verbal instructions, cannot cope with a command-line prompt, cannot comprehend the consequences of their actions, does not understand or read "terms and conditions" and so on.
Right, so hardly Debian's target audience then.
It's all very well having something to cater to the non-technical folk, and I applaud the effort, but you'll note that almost all of the "Libre" OSs are actually Debian based, and if you made Debian unusable on most of the hardware that Debian developers actually use (or are paid by their employers to use) then all you'd do is make sure that they use something else, so you wouldn't have the same mindshare in Debian, and would end up with Debian being as poorly maintained as most of the "libre" things you apparently wish we'd emulate.
The fact that some of the "libre" OSs base themselves on Ubuntu strikes me as particularly deranged, given that Ubuntu is actually a step further away from what they want, but there you go.
So, sure, use a Libre OS of you like the compromises they make, but be aware that the main reason that you have the chance to do so is that Debian has made different compromises in order to be popular enough to be the default upstream for Linux, and thus has made it possible for someone to create the Libre OS that you are running.
Giving us grief about ethics will not make things better for you.
When I got into Free Software, the way you ran things was to spend three days recompiling GCC on your proprietary UNIX(TM) OS, followed by perl etc. -- How useful would it have been to be purist about things then?
The place where one can draw the line has been slowly pushed towards the hardware, but pretending that the masses are currently able or interested in running on truly free hardware does not make it true. It might even sabotage the effort to make it possible. After all, most people are firmly clutching their android devices, totally unaware that there's Free software inside, without even a temptation to look under the surface.
It seems to me that we're all progressing towards the same destination, even if via slightly differing routes. Reenacting "The Life of Brian's" Splitters scene is just a way of not getting on with something useful instead -- please give it a rest.
Cheers, Phil.
On Sunday 16. October 2016 12.14.11 Philip Hands wrote:
The fact that some of the "libre" OSs base themselves on Ubuntu strikes me as particularly deranged, given that Ubuntu is actually a step further away from what they want, but there you go.
I did find it rather odd that Trisquel had switched to using Ubuntu as their base rather than Debian: it makes wider architecture support a real problem because Ubuntu has narrowed its own support, presumably dropping non- lucrative/non-enterprise architectures, meaning that one presumably has to reactivate other architectures in Ubuntu to propagate and access the necessary content. I imagine that they also need to do a lot more filtering and rebuilding on what Ubuntu provides than they would had they stuck with Debian, but I didn't follow the decisions around them switching from one to the other.
I don't want to get into arguments about popularity, compromises, and so on, but I have an observation to make. If there were a more conservative base for Debian, meaning that certain controversial or unwanted content would be excluded in those base packages, then the derivatives wanting to preserve that conservatism would have an easier task branching out in their own direction, and it would probably even help the greater Debian distribution in terms of managing and maintaining the archive. I think that's what people are looking for from Debian.
I've looked into various libre distributions (as anyone reading this list might have noticed), and it is an annoyance that while considering how one might bootstrap one of the non-Debian-derived libre distributions on other architectures, Debian has supported such architectures all along. It really should be a matter of selectively obtaining packages already built by Debian and no more. Instead, it seems like auditing is still required, and this appears to be the time-consuming part. That said, I'm still familiarising myself with things like gNewSense, so I could be wrong, I suppose.
Paul
Late and off-topic, but:
On Sun, Oct 16, 2016 at 04:02:51AM +0100, Wookey wrote:
I've just installed unrar-free,
See also unar as a replacement for unrar-nonfree.
On 10/15/16, J.B. Nicholson jbn@forestfield.org wrote:
Luke Kenneth Casson Leighton wrote:
It's entirely possible something has changed and I am not aware of relevant updates on this (I don't doubt you're in touch with them far more than I am). Please do reply to the list with updates to this situation.
i'm not - they're extremely limited on resources and time, so i keep communication to a minimum.
In https://www.gnu.org/distros/common-distros.html we find the following objection, "Debian also provides a repository of nonfree software. According to the project, this software is "not part of the Debian system," but the repository is hosted on many of the project's main servers, and people can readily find these nonfree packages by browsing Debian's online package database and its wiki".
... which is why i proposed an HTTP-redirect-and-rewrite style split that would appear seamless and transparent (appearing to be a "single site") for many years. the wiki however would be a problem that would need careful and comprehensive review... but if steps are never taken, even small ones, zero progress will ever be made.
l.
El 15 de octubre de 2016 a las 16:28:55, mdn (bernardlprf@openmailbox.org) escribió:
Debian's approach of this isn't really ethical. Anyone has the freedom of install installing propitiatory software even non tech users.
Including non-free software in the repository is insisting/proposing users, especially non tech ones, to give up their freedom for simplicity without understanding the importance of them and the technical problems that non free/libre software brings (a good example of that is the game modding community).
Mainstreams users like you seem to refer to them are what makes software and hardware go in decadence. I don't say that they are directly concerned, but it is how their were treated like, that made them what they are now and ask the same bad products.
If you continue to give them what they are made of the project will slowly become like them and only enforce the already bad circle.
I understand you say.
I also prefer libre games, but the quality of these games are usually low (projects are very small and without money, I understand it and I'm not criticizing that games).
My idea is similar to the idea of Luke, when you go to download a proprietary game, you will be warned that it is a proprietary game and its consequences.
In my opinion, the problem of libre software is not the existence of proprietary software. The problem is that developers barely get economic benefits doing libre software. Maybe we should think about how developers can make profits doing libre software.
Le 15/10/2016 19:54, GaCuest a écrit :
El 15 de octubre de 2016 a las 16:28:55, mdn (bernardlprf@openmailbox.org) escribió:
Debian's approach of this isn't really ethical. Anyone has the freedom of install installing propitiatory software even non tech users.
Including non-free software in the repository is insisting/proposing users, especially non tech ones, to give up their freedom for simplicity without understanding the importance of them and the technical problems that non free/libre software brings (a good example of that is the game modding community).
Mainstreams users like you seem to refer to them are what makes software and hardware go in decadence. I don't say that they are directly concerned, but it is how their were treated like, that made them what they are now and ask the same bad products.
If you continue to give them what they are made of the project will slowly become like them and only enforce the already bad circle.
I understand you say.
I also prefer libre games, but the quality of these games are usually low (projects are very small and without money, I understand it and I'm not criticizing that games).
without money ? Why not ask some devs of some project to finance it via your platform ? Like a crowdfunding ? Their was freedom sponsor https://freedomsponsors.org/ I tried to contact some people because someone was proposing cmyk support for gimp and I wanted to participate for that but I had no response.
My idea is similar to the idea of Luke, when you go to download a proprietary game, you will be warned that it is a proprietary game and its consequences.
In my opinion, the problem of libre software is not the existence of proprietary software. The problem is that developers barely get economic benefits doing libre software. Maybe we should think about how developers can make profits doing libre software.
Imo the problem is the monopoly of Enormous entities like EA and more. These entities are the reasons why devs anc content creators don't have more economical benefits instead of the people who worked on it it goes to the investors.
Their is a solution to that and you can make libre software at the same time. I don't remember who said that but this (maybe someone at red-hat IDR) is what comes in my mind when you want to pay for free software. "You don't sell free software, you work around it"
This is a very interesting idea, instead of paying once a game you could make a small monthly payment to have the services (bug correction, more features etc..) , of course you need to have a lot of participant/customers in that.
Another one witch is also a compromise that I come up with. Is that since all video games are ephemera (1 or 2 years) what you can do is to sell the content of it and releases the software/sources under free/libre licence and when you have made/reached the estimate amount of money or more you can release under copyleft the content of the game (art etc...).
To my knowledge the last propitiatory game engine that was released under GPL in 2016 was serious sam (2001) https://github.com/Croteam-official/Serious-Engine
Witch is one of the very rare games who's licence changed to copyleft. It is sad to see such old software to be released just now and their are even older software that where never released and even lost.
For example the source code of "homeworld cataclysm" has been lost that is why their was no remake of it a few months ago.
The lifetime of a game also depends of the community, just look at the fallout community witch I participated myself a long time ago, mods are still made on that. The tools to correct bugs are not the best, it isn't really correction, but they managed to do it.
For example this mod witch is a obligation to play fallout 3 without much trouble: http://www.nexusmods.com/fallout3/mods/19122/? The number of corrections is just ludicrous.
Do you think this is normal that Bethesda didn't made all these corrections? Can you imagine if the modding community legally had the sources code, correct tools and could legally make changes ?
A living example of that is Open Morrowind the full-featured reimplementation of the Morrowind engine. https://openmw.org/faq/#do_i_need_morrowind
I have been part of the gaming community since 2003, I have stopped two years ago to concentrate my objectives on my migration on free/libre software.
I discovered the power of a community twice. First when I discovered free/libre software. The second is re-discovery of the real potential of the modding community in games.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On 10/17/2016 07:46 PM, mdn wrote:
Another one witch is also a compromise that I come up with. Is that since all video games are ephemera (1 or 2 years) what you can do is to sell the content of it and releases the software/sources under free/libre licence and when you have made/reached the estimate amount of money or more you can release under copyleft the content of the game (art etc...).
I had a similar idea, where written into the liscense for the game is a "self-destruct" feature. Basically X number of years after initial release the game code goes GPL and the assets CC (possibily GPL depending on how the copyright law works with re-using the assets in later games, especially where voice actors and celebrety cameos are concerned)
I can't help but think of Microsoft Train Simulator as a great example of why this should exist. It's a codebase old enough that it doesn't work well on modern systems, there's still a community interested in playing it, and Microsoft isn't doing anything with it. It's not available anywhere except for original retail copies getting resold and passed around amongst those who are actually interested in some old railroad simulator. If that game got GPL'ed Microsoft wouldn't miss out on a dime, but they won't bother doing so because "nobody's interested in that old thing," and they're Microsoft.
On 10/18/16, ryan rrryan@tds.net wrote:
On 10/17/2016 07:46 PM, mdn wrote:
Another one witch is also a compromise that I come up with. Is that since all video games are ephemera (1 or 2 years) what you can do is to sell the content of it and releases the software/sources under free/libre licence and when you have made/reached the estimate amount of money or more you can release under copyleft the content of the game (art etc...).
this happened with descent and descent 2. the first 3d game ever to have 6 degrees of freedom. you can look it up (on sourceforge i believe) and compile it up for DOS as well as linux. the data files are non-free (but still accessible) and there is a community around the engine creating their own maps.
the really *really* nice thing about descent is that it actually works well on 320x240 on 16mhz 386s all the way up to modern systems.
l.
Thanks I'll try them out :)
Le 18/10/2016 14:28, Luke Kenneth Casson Leighton a écrit :
On 10/18/16, ryan rrryan@tds.net wrote:
On 10/17/2016 07:46 PM, mdn wrote:
Another one witch is also a compromise that I come up with. Is that since all video games are ephemera (1 or 2 years) what you can do is to sell the content of it and releases the software/sources under free/libre licence and when you have made/reached the estimate amount of money or more you can release under copyleft the content of the game (art etc...).
this happened with descent and descent 2. the first 3d game ever to have 6 degrees of freedom. you can look it up (on sourceforge i believe) and compile it up for DOS as well as linux. the data files are non-free (but still accessible) and there is a community around the engine creating their own maps.
the really *really* nice thing about descent is that it actually works well on 320x240 on 16mhz 386s all the way up to modern systems.
l.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On October 18, 2016 5:28:42 AM PDT, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
the really *really* nice thing about descent is that it actually works well on 320x240 on 16mhz 386s all the way up to modern systems.
I almost find this incredible! Considering games like Doom ran on a few Mhz and NO 3D acceleration, I expect them to run comfortably on modern 1 Ghz computers, but in my experience Freedoom runs slow as a snail on an ARM board. Is there some factor I'm missing? Perhaps the SDL abstraction layer gets in the way of fast direct-architecture code?
I'm tempted to see about compiling that free Descent source and see what you mean. I have the data files in my possession somewhere.
On Oct 19, 2016 1:05 PM, "Eric Duhamel" ericxdu23@gmail.com wrote:
in my experience Freedoom runs slow as a snail on an ARM board. Is there
some factor I'm missing?
AFAIK, Freedoom is just a BSD-licensed IWAD (i.e. game content) that can be used instead of the proprietary IWAD with any number of source ports of the Doom engine. And since these source ports vary widely in the number of additional features and capabilities bolted on, it may be a question of which source port you're using.
I haven't messed with Doom on anything below 1GHz for some years, so I don't have any specific recommendations for lightweight/efficient source ports, but you might try several and see if it makes a difference.
Benson Mitchell
On October 19, 2016 11:35:17 AM PDT, Benson Mitchell benson.mitchell+arm-netbook@gmail.com wrote:
AFAIK, Freedoom is just a BSD-licensed IWAD (i.e. game content) that can be used instead of the proprietary IWAD with any number of source ports of the Doom engine. And since these source ports vary widely in the number of additional features and capabilities bolted on, it may be a question of which source port you're using.
I haven't messed with Doom on anything below 1GHz for some years, so I don't have any specific recommendations for lightweight/efficient source ports, but you might try several and see if it makes a difference.
Correct. I don't know why I just said "Freedoom". I've tried Freedoom with Vavoom and something called "Chocolate Doom" and was rather disappointed... maybe I should pick up the search again.
On Oct 19, 2016 2:57 PM, "Eric Duhamel" ericxdu23@gmail.com wrote:
I've tried Freedoom with Vavoom and something called "Chocolate Doom" and
was rather disappointed...
Maybe try PrBoom? IIRC that's what Rockbox's Doom port is based on, and that runs on MP3 players with 40 or 50MHz ARM CPUs.
Benson Mitchell
On October 19, 2016 11:17:00 PM PDT, FaTony fatony@fatony.net wrote:
Maybe try PrBoom? IIRC that's what Rockbox's Doom port is based on,
and
that runs on MP3 players with 40 or 50MHz ARM CPUs.
There's Prboom-plus in the official Debian repository.
Yeah, there it is. I just ran it smoothly on my Beaglebone Black (1 GHz). I've yet to try PrBoom on my Raspberry Pi B+ (700 MHz), but I'm guessing it will run well on the EOMA68-A20 with it's 1.2 GHz.
guessing it will run well on the EOMA68-A20 with it's 1.2 GHz.
Last I checked the A20 only goes up to 960MHz or so:
% cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies 144000 312000 528000 720000 864000 912000 960000 %
You can try to overclock it, of course.
Stefan
On October 20, 2016 2:19:52 PM PDT, Stefan Monnier monnier@iro.umontreal.ca wrote:
guessing it will run well on the EOMA68-A20 with it's 1.2 GHz.
Last I checked the A20 only goes up to 960MHz or so:
% cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies 144000 312000 528000 720000 864000 912000 960000 %
I must have misunderstood, then. The Crowdsupply page and other pages rate the A20 at 1.2 Ghz
look up the cpufreq page on linux-sunxi wiki.
On 11/3/16, Eric Duhamel ericxdu23@gmail.com wrote:
On October 20, 2016 2:19:52 PM PDT, Stefan Monnier monnier@iro.umontreal.ca wrote:
guessing it will run well on the EOMA68-A20 with it's 1.2 GHz.
Last I checked the A20 only goes up to 960MHz or so:
% cat /sys/devices/system/cpu/cpu0/cpufreq/scaling_available_frequencies 144000 312000 528000 720000 864000 912000 960000 %
I must have misunderstood, then. The Crowdsupply page and other pages rate the A20 at 1.2 Ghz
-- Eric Duhamel http://www.noxbanners.net/
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On November 3, 2016 7:48:05 AM PDT, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
look up the cpufreq page on linux-sunxi wiki.
linux-sunxi wiki says in "overclocking" section that 1.2 Ghz is "rather unrealistic". Well, I'll move forward assuming the A20 card will be about as fast as my Beaglebone Black (~1 Ghz) with double the RAM.
Le jeu. 3 nov. 2016 à 9:04, Eric Duhamel ericxdu23@gmail.com a écrit :
linux-sunxi wiki says in "overclocking" section that 1.2 Ghz is "rather unrealistic". Well, I'll move forward assuming the A20 card will be about as fast as my Beaglebone Black (~1 Ghz) with double the RAM.
Double the cores, quadruple the RAM.
Le 15/10/2016 19:54, GaCuest a écrit :
El 15 de octubre de 2016 a las 16:28:55, mdn (bernardlprf@openmailbox.org) escribió:
Debian's approach of this isn't really ethical. Anyone has the freedom of install installing propitiatory software even non tech users.
Including non-free software in the repository is insisting/proposing users, especially non tech ones, to give up their freedom for simplicity without understanding the importance of them and the technical problems that non free/libre software brings (a good example of that is the game modding community).
Mainstreams users like you seem to refer to them are what makes software and hardware go in decadence. I don't say that they are directly concerned, but it is how their were treated like, that made them what they are now and ask the same bad products.
If you continue to give them what they are made of the project will slowly become like them and only enforce the already bad circle.
I understand you say.
I also prefer libre games, but the quality of these games are usually low (projects are very small and without money, I understand it and I'm not criticizing that games).
Did you try 0ad ? minetest ? free orion ? free civ ? I admit that sometimes the interfaces isn't intuitive but dam some of them are nice.
My idea is similar to the idea of Luke, when you go to download a proprietary game, you will be warned that it is a proprietary game and its consequences.
In my opinion, the problem of libre software is not the existence of proprietary software. The problem is that developers barely get economic benefits doing libre software. Maybe we should think about how developers can make profits doing libre software.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
A game console based on emulators and ports can be really interesting even before adding non-free software.
For starters, you may be able to find public domain or even free software ROM for most of the emulators, and they could even be shipped with the device operating system as part of the free software included! Users looking for a device to play classic games will of course download what they want, but that base set of games will be there, pre-installed, free, and legal. It would take some searching and verification, but even a handful of free ROM would be good. [PD Roms](http://pdroms.de/) [Community Software]( https://archive.org/details/open_source_software)
RetroPie has an interesting project they call [Ports](https://github.com/retropie/RetroPie-Setup/wiki/Ports). Some of the engines and games they are porting are free software.
Lastly, it could be really helpful to provide a platform target for free software games developers. I'd guess that a device that can run/test games written in [Godot](https://godotengine.org/), [Löve](https://love2d.org/), or [Pygame](http://www.pygame.org/) would be rather attractive.
El 15 de octubre de 2016 a las 20:33:12, Eric Duhamel (ericxdu23@gmail.com) escribió:
A game console based on emulators and ports can be really interesting even before adding non-free software.
For starters, you may be able to find public domain or even free software ROM for most of the emulators, and they could even be shipped with the device operating system as part of the free software included! Users looking for a device to play classic games will of course download what they want, but that base set of games will be there, pre-installed, free, and legal. It would take some searching and verification, but even a handful of free ROM would be good. [PD Roms](http://pdroms.de/) [Community Software]( https://archive.org/details/open_source_software)
RetroPie has an interesting project they call [Ports](https://github.com/retropie/RetroPie-Setup/wiki/Ports). Some of the engines and games they are porting are free software.
Lastly, it could be really helpful to provide a platform target for free software games developers. I'd guess that a device that can run/test games written in [Godot](https://godotengine.org/), [Löve](https://love2d.org/), or [Pygame](http://www.pygame.org/) would be rather attractive.
Yes, our idea is to support all that. I still keep my GP32 and GP2X, I'm a fan of retro games and emulators.
As I said above, it may be interesting to launch a EOMA68-A20 completely libre with this type of software. Without proprietary drivers for the GPU to be completely free.
Thanks for your comments.
Hello everyone!
If anyone if interested, I’ve updated the block diagram with new changes: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
Any suggestions/corrections are appreciated. Thanks!
On 10/15/2016 06:21 PM, GaCuest wrote:
Hello everyone!
If anyone if interested, I’ve updated the block diagram with new changes: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
Any suggestions/corrections are appreciated. Thanks!
Do you have a chip picked out to provide 5V boost to the EOMA card yet? The AXP209 does not appear to provide it.
My plan for handheld computer was to use a TI brand LiPo charging IC and 5V boost. The AXP209 seems like a better idea than the TI charging solution since it has configurable regulator outputs and lower cost.
Also, I see you are not using SSD2828 for RGB to MIPI conversion anymore. This week I finished up a preliminary schematic for a SSD2828 testing/breakout board. I briefly started the layout and began to have second thoughts. It seems like too much work to incorporate it. I'm now looking for ~4 inch RGB LCD's instead.
take a look at the various schematics available for tablet reference designs etc. etc. you want something like a SY7208 see http://hands.com/~lkcl/eoma/TABLET162_20121110.pdf page 7 and make sure it's providing enough current.
please for goodness sake read the next update when it comes out, please for god's sake don't go searching on digikey, pick the first random convenient IC and expect to be able to have what you design made up in even a small volume at anything approaching a reasonable cost. it's very very easy to design ultra-expensive products.
take a look at frida's LCDs, we have the advantage of being in touch directly with marco, the possibility of group buys, etc. etc. datasheets all here: http://rhombus-tech.net/suppliers/shenzen/frida_lcd/
l.
On 10/16/16, Joseph Honold mozzwald@gmail.com wrote:
On 10/15/2016 06:21 PM, GaCuest wrote:
Hello everyone!
If anyone if interested, I’ve updated the block diagram with new changes: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
Any suggestions/corrections are appreciated. Thanks!
Do you have a chip picked out to provide 5V boost to the EOMA card yet? The AXP209 does not appear to provide it.
My plan for handheld computer was to use a TI brand LiPo charging IC and 5V boost. The AXP209 seems like a better idea than the TI charging solution since it has configurable regulator outputs and lower cost.
Also, I see you are not using SSD2828 for RGB to MIPI conversion anymore. This week I finished up a preliminary schematic for a SSD2828 testing/breakout board. I briefly started the layout and began to have second thoughts. It seems like too much work to incorporate it. I'm now looking for ~4 inch RGB LCD's instead.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
El 16 de octubre de 2016 a las 9:16:56, Luke Kenneth Casson Leighton (lkcl@lkcl.net) escribió:
take a look at the various schematics available for tablet reference designs etc. etc. you want something like a SY7208 see http://hands.com/~lkcl/eoma/TABLET162_20121110.pdf page 7 and make sure it's providing enough current.
The link is incorrect. Do you want to say this link?:
http://hands.com/~lkcl/eoma/kde_tablet/TABLET162_20121110.pdf
I have seen that this is more recent:
http://hands.com/~lkcl/eoma/kde_tablet/tablet5.pdf
What you recommend me?
please for goodness sake read the next update when it comes out, please for god's sake don't go searching on digikey, pick the first random convenient IC and expect to be able to have what you design made up in even a small volume at anything approaching a reasonable cost. it's very very easy to design ultra-expensive products.
take a look at frida's LCDs, we have the advantage of being in touch directly with marco, the possibility of group buys, etc. etc. datasheets all here: http://rhombus-tech.net/suppliers/shenzen/frida_lcd/
I think that most components are in large quantities in China:
- Linear potentiometer (trigger) and joystick -- Favor Union/Polyshine (a Hong Kong company). MOQ is higher, but price is very good. - TFT + RTP -- Frida. - Battery -- Kamcy (a Chinese company). A lot of chinese companies sells this battery. - EEPROM -- The same that you use on laptop/tablet. - STM32F072 -- The same that you use on laptop. - Audio IC -- The same that you use on laptop. - TS4408A button -- a lot of stores sell its in Aliexpress/Alibaba with good prices: https://aliexpress.com/af/4mmx4mmx0.8mm.html - SKRTLAE010 button -- you use it on laptop. Have you found a chinese replacement? - USB 2.0 -- you use it on laptop. Have you found a chinese replacement? - PMIC -- you use it on tablet. - MicroUSB, microSD and Audio Jack -- you use it on laptop (Runde). - XPT2046 -- a clone of ADS7843. A lot of chinese stores sell it on Aliexpress/Alibaba.
What do you change?
Thanks!
On 10/16/16, GaCuest gacuest@gmail.com wrote:
El 16 de octubre de 2016 a las 9:16:56, Luke Kenneth Casson Leighton (lkcl@lkcl.net) escribió:
take a look at the various schematics available for tablet reference designs etc. etc. you want something like a SY7208 see http://hands.com/~lkcl/eoma/TABLET162_20121110.pdf page 7 and make sure it's providing enough current.
The link is incorrect. Do you want to say this link?:
http://hands.com/~lkcl/eoma/kde_tablet/TABLET162_20121110.pdf
page 7. contains relevant step-up to 5V converter IC.
I think that most components are in large quantities in China:
- Linear potentiometer (trigger) and joystick -- Favor Union/Polyshine
(a Hong Kong company). MOQ is higher, but price is very good.
let me put the photos in front of runde i know they have something
- TS4408A button -- a lot of stores sell its in Aliexpress/Alibaba with
good prices: https://aliexpress.com/af/4mmx4mmx0.8mm.html
send me some photos (compressed-archive) by direct email message. i'll put them in front of runde when i visit them
- SKRTLAE010 button -- you use it on laptop. Have you found a
chinese replacement?
runde. again.
- USB 2.0 -- you use it on laptop. Have you found a chinese
replacement?
yes. runde. again. they have practically everything
El 17 de octubre de 2016 a las 14:17:16, Luke Kenneth Casson Leighton (lkcl@lkcl.net) escribió:
On 10/16/16, GaCuest wrote:
El 16 de octubre de 2016 a las 9:16:56, Luke Kenneth Casson Leighton (lkcl@lkcl.net) escribió:
take a look at the various schematics available for tablet reference designs etc. etc. you want something like a SY7208 see http://hands.com/~lkcl/eoma/TABLET162_20121110.pdf page 7 and make sure it's providing enough current.
The link is incorrect. Do you want to say this link?:
http://hands.com/~lkcl/eoma/kde_tablet/TABLET162_20121110.pdf
page 7. contains relevant step-up to 5V converter IC.
Yes, but in that document you use RT9266PE.
In this document (page 3): http://hands.com/~lkcl/eoma/kde_tablet/tablet5.pdf you use SY7208B. What is your recommendation?
Thanks.
On 10/16/16, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
take a look at frida's LCDs, we have the advantage of being in touch directly with marco, the possibility of group buys, etc. etc. datasheets all here: http://rhombus-tech.net/suppliers/shenzen/frida_lcd/
joseph update page i found the NT35510 datasheet it looks like it can do SPI (am checking with marco) which makes it perfect for the hybrid-phone i want to do, if you use this one as well FRD39751040V then we can do a group buy between the three projects. it's 800x480 and it's IPS.
Also, I see you are not using SSD2828 for RGB to MIPI conversion anymore. This week I finished up a preliminary schematic for a SSD2828 testing/breakout board. I briefly started the layout and began to have second thoughts. It seems like too much work to incorporate it. I'm now looking for ~4 inch RGB LCD's instead.
I've been looking on aliexpress and found some 3.97" LCD's based on the nt35510/nt35512. They are similar to the FRD39751040V but have different pinouts. Does Frida sell sample/single LCD's anywhere online?
https://www.aliexpress.com/item/IPS-4-0-inch-16M-HD-TFT-LCD-RGB-Screen-with-...
https://www.aliexpress.com/item/IPS-3-97-inch-61PIN-3-SPI-24Bit-TFT-LCD-LCM-...
The linux driver you posted appears to be for MIPI/DSI mode so probably not useful for us. I found this RGB24 initialization driver for u-boot/nt35510. It could be modified for RGB18, then I assume we could just use the sunxi-fb lcd linux driver?
http://lists.denx.de/pipermail/u-boot/2013-August/161006.html
On 10/17/2016 01:01 PM, Luke Kenneth Casson Leighton wrote:
On 10/16/16, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
take a look at frida's LCDs, we have the advantage of being in touch directly with marco, the possibility of group buys, etc. etc. datasheets all here: http://rhombus-tech.net/suppliers/shenzen/frida_lcd/
joseph update page i found the NT35510 datasheet it looks like it can do SPI (am checking with marco) which makes it perfect for the hybrid-phone i want to do, if you use this one as well FRD39751040V then we can do a group buy between the three projects. it's 800x480 and it's IPS.
Also, I see you are not using SSD2828 for RGB to MIPI conversion anymore. This week I finished up a preliminary schematic for a SSD2828 testing/breakout board. I briefly started the layout and began to have second thoughts. It seems like too much work to incorporate it. I'm now looking for ~4 inch RGB LCD's instead.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
Hello,
Just one piece of advise.
Look on TAOBAO as well. Site is completely in Chinese but it is much more relevant for sourcing electronic parts in China if you know little bit around. For example you can register on some of Taobao agents (like www.bhiner.com), you can copy/paste link from Taobao (and Aliexpress) and you will be able to see complete translation (fair one at least) as well they will be able to send goods to you for a small fee. Of course, long term it is the best to have direct access to factory or some help in China, however being able to buy on Taobao can be helpful.
Hrvoje
On Tue, Oct 18, 2016 at 4:34 PM, Joseph Honold mozzwald@gmail.com wrote:
I've been looking on aliexpress and found some 3.97" LCD's based on the nt35510/nt35512. They are similar to the FRD39751040V but have different pinouts. Does Frida sell sample/single LCD's anywhere online?
https://www.aliexpress.com/item/IPS-4-0-inch-16M-HD-TFT- LCD-RGB-Screen-with-Adapter-Board-800-480-MCU/32669105821.html
https://www.aliexpress.com/item/IPS-3-97-inch-61PIN-3- SPI-24Bit-TFT-LCD-LCM-Color-Screen-with-touch-panel/1764842496.html
The linux driver you posted appears to be for MIPI/DSI mode so probably not useful for us. I found this RGB24 initialization driver for u-boot/nt35510. It could be modified for RGB18, then I assume we could just use the sunxi-fb lcd linux driver?
http://lists.denx.de/pipermail/u-boot/2013-August/161006.html
On 10/17/2016 01:01 PM, Luke Kenneth Casson Leighton wrote:
On 10/16/16, Luke Kenneth Casson Leighton lkcl@lkcl.net wrote:
take a look at frida's LCDs, we have the advantage of being in touch directly with marco, the possibility of group buys, etc. etc. datasheets all here: http://rhombus-tech.net/suppliers/shenzen/frida_lcd/
joseph update page i found the NT35510 datasheet it looks like it can do SPI (am checking with marco) which makes it perfect for the hybrid-phone i want to do, if you use this one as well FRD39751040V then we can do a group buy between the three projects. it's 800x480 and it's IPS.
Also, I see you are not using SSD2828 for RGB to MIPI conversion anymore. This week I finished up a preliminary schematic for a SSD2828 testing/breakout board. I briefly started the layout and began to have second thoughts. It seems like too much work to incorporate it. I'm now looking for ~4 inch RGB LCD's instead.
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On 10/18/16, Joseph Honold mozzwald@gmail.com wrote:
I've been looking on aliexpress and found some 3.97" LCD's based on the nt35510/nt35512. They are similar to the FRD39751040V but have different pinouts. Does Frida sell sample/single LCD's anywhere online?
yes, but they have to make them by hand (if they're not part of an existing production run right at the time you ask) so give them a couple of weeks.
https://www.aliexpress.com/item/IPS-4-0-inch-16M-HD-TFT-LCD-RGB-Screen-with-...
https://www.aliexpress.com/item/IPS-3-97-inch-61PIN-3-SPI-24Bit-TFT-LCD-LCM-...
The linux driver you posted appears to be for MIPI/DSI mode so probably not useful for us. I found this RGB24 initialization driver for u-boot/nt35510. It could be modified for RGB18, then I assume we could just use the sunxi-fb lcd linux driver?
http://lists.denx.de/pipermail/u-boot/2013-August/161006.html
added.
El 18 de octubre de 2016 a las 16:34:37, Joseph Honold (mozzwald@gmail.com) escribió:
I've been looking on aliexpress and found some 3.97" LCD's based on the nt35510/nt35512. They are similar to the FRD39751040V but have different pinouts. Does Frida sell sample/single LCD's anywhere online?
https://www.aliexpress.com/item/IPS-4-0-inch-16M-HD-TFT-LCD-RGB-Screen-with-...
https://www.aliexpress.com/item/IPS-3-97-inch-61PIN-3-SPI-24Bit-TFT-LCD-LCM-...
The linux driver you posted appears to be for MIPI/DSI mode so probably not useful for us. I found this RGB24 initialization driver for u-boot/nt35510. It could be modified for RGB18, then I assume we could just use the sunxi-fb lcd linux driver?
http://lists.denx.de/pipermail/u-boot/2013-August/161006.html
Maybe you can also use the 4.5” 480x854 IPS LCD that we want to use.
I sent it to Luke, but I don’t know if he can use it.
The price is very similar to FRD39751040V. $8 without touch panel, and $9.5 with RTP. The problem is that this LCD hasn't entered into mass production now (now Frida only send samples), and we need to buy 5000 units to enter into mass production.
Here you can see the drawing that Frida sent me: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/...
And here you can send the info that Frida sent me (initial code): http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/...
On 10/19/2016 03:44 AM, GaCuest wrote:
Maybe you can also use the 4.5” 480x854 IPS LCD that we want to use.
I sent it to Luke, but I don’t know if he can use it.
The price is very similar to FRD39751040V. $8 without touch panel, and $9.5 with RTP. The problem is that this LCD hasn't entered into mass production now (now Frida only send samples), and we need to buy 5000 units to enter into mass production.
Here you can see the drawing that Frida sent me: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/...
And here you can send the info that Frida sent me (initial code): http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/...
Your datasheet is a bit hard to read, but I think it shows size as 100mm x 56.?mm. The FRD39751040V datasheet states 97mm x 57mm. They both seem quite close in size and I might be able to use the larger. I have not done a case design yet so I'm not dead set on anything.
El 20 de octubre de 2016 a las 16:34:35, Joseph Honold (mozzwald@gmail.com) escribió:
On 10/19/2016 03:44 AM, GaCuest wrote:
Maybe you can also use the 4.5” 480x854 IPS LCD that we want to use.
I sent it to Luke, but I don’t know if he can use it.
The price is very similar to FRD39751040V. $8 without touch panel, and $9.5 with RTP. The problem is that this LCD hasn't entered into mass production now (now Frida only send samples), and we need to buy 5000 units to enter into mass production.
Here you can see the drawing that Frida sent me: http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/...
And here you can send the info that Frida sent me (initial code): http://george.the-petries.co.uk/shared-write-access/eoma/Datasheets%20ZEOMA/...
Your datasheet is a bit hard to read, but I think it shows size as 100mm x 56.?mm. The FRD39751040V datasheet states 97mm x 57mm. They both seem quite close in size and I might be able to use the larger. I have not done a case design yet so I'm not dead set on anything.
The active area of FRD450 is 55,44mm x 98,64mm. The active area of the FRD3975 is 51,84mm x 86,40mm. Maybe you have seen something wrong.
Is your project a handheld laptop like OpenPandora/Pyra?
On 10/20/2016 09:49 AM, GaCuest wrote:
The active area of FRD450 is 55,44mm x 98,64mm. The active area of the FRD3975 is 51,84mm x 86,40mm. Maybe you have seen something wrong.
I was referring to the module size, not viewing area.
Is your project a handheld laptop like OpenPandora/Pyra?
Yes, a handheld computer but not clamshell. My plan is to make something more like a Blackberry/Peek/Nokia E71. Screen above qwerty keyboard.
El 20 de octubre de 2016 a las 17:05:03, Joseph Honold (mozzwald@gmail.com) escribió:
On 10/20/2016 09:49 AM, GaCuest wrote:
The active area of FRD450 is 55,44mm x 98,64mm. The active area of the FRD3975 is 51,84mm x 86,40mm. Maybe you have seen something wrong.
I was referring to the module size, not viewing area.
Then you have seen bad the FRD450. The size of FRD450 is 60mm x 109mm. The size of FRD3975 is 57.5mm x 97.20mm.
Good luck with your project!
El 16 de octubre de 2016 a las 2:57:51, Joseph Honold (mozzwald@gmail.com) escribió:
On 10/15/2016 06:21 PM, GaCuest wrote:
Hello everyone!
If anyone if interested, I’ve updated the block diagram with new changes: http://george.the-petries.co.uk/shared-write-access/eoma/Block%20Diagram%20Z...
Any suggestions/corrections are appreciated. Thanks!
Do you have a chip picked out to provide 5V boost to the EOMA card yet? The AXP209 does not appear to provide it.
My plan for handheld computer was to use a TI brand LiPo charging IC and 5V boost. The AXP209 seems like a better idea than the TI charging solution since it has configurable regulator outputs and lower cost.
I have little idea about electronics, so I try to use the schematic of the tablet that Luke was doing.
I have seen this: http://hands.com/~lkcl/eoma/kde_tablet/tablet5.pdf
So it seems, he uses SY7208 to provide 5V to EOMA68.
Yes, my choice of AXP209 is cost-effective and the possibility to reuse the work of Luke.
Also, I see you are not using SSD2828 for RGB to MIPI conversion anymore. This week I finished up a preliminary schematic for a SSD2828 testing/breakout board. I briefly started the layout and began to have second thoughts. It seems like too much work to incorporate it. I'm now looking for ~4 inch RGB LCD's instead.
As I mentioned earlier, I have little idea about electronics. If I use SSD2828, possibly I could never do it.
In addition, FRIDA has given me a good screen (IPS, 4.5inch, 480x854, and RGB 18-bit) with resistive touch panel for a good price ($9.5). Perhaps the problem is that the MOQ is 5000 units, but if I want to do ZEOMA, that should be the minimum selling amount, as the costs of injection molds for plastic are also high. Produce ZEOMA in lower amounts would do that the price was very high (as with Pyra).
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On October 15, 2016 12:47:14 PM PDT, GaCuest gacuest@gmail.com wrote:
Yes, our idea is to support all that.
That's great to hear.
As I said above, it may be interesting to launch a EOMA68-A20 completely libre with this type of software. Without proprietary drivers for the GPU to be completely free.
If the ZEOMA will "just plug it in" and work with an EOMA68-A20 cards on offer, meaning the drivers provided by the OS can run the screen, wi-fi, etc., then I think it is already over half-way done! For instance, I'd just need to install the Xorg Joystick driver or make it automatically launch EmulationStation. EmulationStation can be configured to launch any libre games practicable. I've been experimenting with Zeroinstall for fetching games or versions that are not in repos.
Solarwolf runs well on as low as 1 Ghz processors without any 3D acceleration. Freedink is probably a good candidate, too. I haven't tried many libre games on low-powered hardware so this list needs expansion. Also, games with no built-in joystick support will require extra accommodations.
Of course, someone would need to put in the work to derive a pre-configured gaming ISO from a Debian or Parabola base. This could be sold on eoma68 cards or put on a "specially prepared" SD card to flash eoma cards. If anyone can offer some expertise in this area, please chime in.
P.S. I'd like to thank and welcome GaCuest to this mailing list. I had heard the ubrewit-zeoma project was a little hard to reach. Collaboration between these two communities should be beneficial.
A game console based on emulators and ports can be really interesting even before adding non-free software.
For starters, you may be able to find public domain or even free software ROM for most of the emulators, and they could even be shipped with the device operating system as part of the free software included! Users looking for a device to play classic games will of course download what they want, but that base set of games will be there, pre-installed, free, and legal. It would take some searching and verification, but even a handful of free ROM would be good. [PD Roms](http://pdroms.de/) [Community Software]( https://archive.org/details/open_source_software)
RetroPie has an interesting project they call [Ports](https://github.com/retropie/RetroPie-Setup/wiki/Ports). Some of the engines and games they are porting are free software.
Lastly, it could be really helpful to provide a platform target for free software games developers. I'd guess that a device that can run/test games written in [Godot](https://godotengine.org/), [Löve](https://love2d.org/), or [Pygame](http://www.pygame.org/) would be rather attractive.
A game console based on emulators and ports can be really interesting even before adding non-free software.
For starters, you may be able to find public domain or even free software ROM for most of the emulators, and they could even be shipped with the device operating system as part of the free software included! Users looking for a device to play classic games will of course download what they want, but that base set of games will be there, pre-installed, free, and legal. [PD Roms](http://pdroms.de/) [Community Software]( https://archive.org/details/open_source_software)
RetroPie has an interesting project they call [Ports](https://github.com/ retropie/RetroPie-Setup/wiki/Ports). Some of the engines and games they are porting are free software.
Lastly, it could be really helpful to provide a platform target for free software games developers. I'd guess that a device that can run/test games written in [Godot](https://godotengine.org/), , or Pygame would be rather attractive.
-- Eric Duhamel http://www.noxbanners.net/
Debian's approach of this isn't really ethical. Anyone has the freedom of install installing propitiatory software even non tech users.
FWIW, any distribution which comes with a browser that doesn't do something like LibreJS suffers from the same problem (or worse): users will download and run proprietary software without even being aware of it just by going to their favorite web sites.
Stefan
On October 16, 2016 6:16:24 AM PDT, Stefan Monnier monnier@iro.umontreal.ca wrote:
FWIW, any distribution which comes with a browser that doesn't do something like LibreJS suffers from the same problem (or worse): users will download and run proprietary software without even being aware of it just by going to their favorite web sites.
This is true, although programs delivered via non-free Javascript are so ubiquitous on the web and the demand for them so high the general practice seems to be just letting the user have this Javascript as it is expected behavior.
I don't think much can be done except trying to push a user toward a js-off experience, but letting them turn it on easily when their website doesn't work. This is far from a comprehensive solution though.
Is this the thread about the handheld device? Talk about getting off-topic! :-P
FWIW, any distribution which comes with a browser that doesn't do something like LibreJS suffers from the same problem (or worse): users will download and run proprietary software without even being aware of it just by going to their favorite web sites.
This is true, although programs delivered via non-free Javascript are so ubiquitous on the web and the demand for them so high the general practice seems to be just letting the user have this Javascript as it is expected behavior.
I know. But similarly, most users expect to be able to use their wifi/gpu card regardless of where the firmware is stored ("in hardware" or "in a blob").
Both are problems w.r.t computer freedom and ethics. But for some reason, most Free Software advocacy focuses on the "blob" part and turns a blind eye to the JS part.
Stefan
This is indeed a problem a problem but I think it's because most inexperienced users (lambda users in french (sorry I don't know the work for that in english) have the habit that "it's just works".
This is one of the biggest problems that we face, normal users. I think that most people were forced to use computers and that is part of why they don't want to change and they just want it to work without thinking.
This is a pedagogical problem and like said before I am working on that subject to create new approaches to make users understand, to build bridges so that they can choose without mindlessly clicking next.
Le 17/10/2016 15:53, Stefan Monnier a écrit :
FWIW, any distribution which comes with a browser that doesn't do something like LibreJS suffers from the same problem (or worse): users will download and run proprietary software without even being aware of it just by going to their favorite web sites.
This is true, although programs delivered via non-free Javascript are so ubiquitous on the web and the demand for them so high the general practice seems to be just letting the user have this Javascript as it is expected behavior.
I know. But similarly, most users expect to be able to use their wifi/gpu card regardless of where the firmware is stored ("in hardware" or "in a blob").
Both are problems w.r.t computer freedom and ethics. But for some reason, most Free Software advocacy focuses on the "blob" part and turns a blind eye to the JS part.
Stefan
arm-netbook mailing list arm-netbook@lists.phcomp.co.uk http://lists.phcomp.co.uk/mailman/listinfo/arm-netbook Send large attachments to arm-netbook@files.phcomp.co.uk
On 10/17/16, Stefan Monnier monnier@iro.umontreal.ca wrote:
Both are problems w.r.t computer freedom and ethics. But for some reason, most Free Software advocacy focuses on the "blob" part and turns a blind eye to the JS part.
the FSF doesn't... FSF-Certified OSes afaiui require LibreJS to be running as well as privacy-violating default search engines to be removed. duckduckgo i've seen available as the default in Parabola browsers for example.
l.
On 10/15/16, FaTony fatony@fatony.net wrote:
What kind of games are you planning to run?
Because this is very libre focused project
it's an *ethically* focussed project of which it so happens, by way of the four freedoms being designed around the extremely rare and hard-to-understand *joint* combination of software development *and* ethical considerations, that libre software is a huge part *of* the project but it is not *the* focus of the project.
now, if people want to *ignore* those ethical considerations they are entirely free to do so and to experience the consequences of doing so (which, if they piss on anything that is part of this project, interfering with it or bringing it into disrepute in any way, then to say that those consequences would be bad for them would be a massive understatement).
so I assume emulators of proprietary hardware and proprietary games for that hardware out of the question.
they're not [out of the question]. they will not be able to receive an RYF Certificate, that's for sure, but that's nothing to do with EOMA68.
if you recall a couple weeks ago i began writing up the specification to incorporate the circumstances under which proprietary software is acceptable, as well as outlining the [rather large] burden of responsibility that proprietary software vendors will be taking on as a result.
if that proprietary software is installable by the end-user (over the internet) *after* the product ships, that burden is greatly reduced. if there isn't *any* choice *other* than proprietary software however, such that that proprietary software might as well already be on the device, i might however get a bit unhappy about that.
it's still all forming, basically, but the fundamental underlying rule is: ethical considerations FIRST. stop causing people pain and distress just because they're buying technology devices.
l.
El 15 de octubre de 2016 a las 7:53:40, FaTony (fatony@fatony.net) escribió:
What kind of games are you planning to run?
Because this is very libre focused project so I assume emulators of proprietary hardware and proprietary games for that hardware out of the question.
My idea is to offer a completely libre console that anyone can improve it and anyone can do whatever he wants with it.
My idea is that this freedom allows you to install what you want. For example, if you want, you can install a completely libre OS (like Parabola OS) or you can install a proprietary OS (like Windows, if there is a EOMA68 compatible in the future).
On the other hand, my idea is to focus on a EOMA68-A20 with GNU/Linux OS (Debian perhaps?). Would be sent without any proprietary software. When you turn on it (first time), the OS will ask if you want to download the proprietary drivers for the GPU and if you want to activate the repository with proprietary software (such as proprietary games or emulators).
I would also like to offer a EOMA68-A20 card with Parabola OS and other with Android OS. While we may not provide support for these cards (for lack of money and time).
arm-netbook@lists.phcomp.co.uk
-
Andrew M.A. Cater -
Benson Mitchell -
Elena ``of Valhalla'' -
Eric Duhamel -
FaTony -
GaCuest -
Hrvoje Lasic -
Isaac David -
J.B. Nicholson -
Joseph Honold -
Luke Kenneth Casson Leighton -
mdn -
Paul Boddie -
Philip Hands -
ryan -
Stefan Monnier -
Tzafrir Cohen -
Wookey