Next meeting
- Vous devez vous identifier ou créer un compte pour écrire des commentaires
When will the next Trisquel meeting be? No time has been decided on, so maybe people should join at the same time as last (Tuesday at 17:00 UTC).
I'll stay on IRC and see if Ruben is available for this Tuesday (the 22nd). I am hoping I will be able to make it then.
I am a member!
I'll try and make it Tuesday the 22nd at 12:00 noon eastern standard time (17:00 UTC).
I haven't been able to contact Ruben as of yet. Some of us could still talk on IRC if Ruben can't make it, if others still want to.
Otherwise, if I can't get hold of Ruben then maybe we should try for the 29th of January.
Since I'll be in Mexico City conducting a public training at that time, it's unlikely I'd be able to drop in. But I hope to read any minutes that may be taken. :)
I haven't been able to contact Ruben via IRC or email in the past few days. He could be offline or busy enough to not read his email.
Again, we could still meet on IRC in 7.5 hours time, but I'm not sure if we will achieve as much without Ruben. If anyone can confirm in the next hour that they still want to meet up *without Ruben* then I will make it.
Otherwise, please keep the 29th free! I'm confident that with more notice we might be able to get Ruben and more other users to join.
Okay, it's probably looking like this meeting won't be on. Apologies if that has inconvenienced anybody.
Hopefully the 29th will be okay.
>Apologies if that has inconvenienced anybody.
On the contrary. Trisquel needs helping hands, like yours.
I am a member!
It looks like two people showed to the meeting. At 12 after I and one other were in attendance.
As mentioned, I am delivering a training, and unable to make an appearance. Hopefully it happens next week, and then I'll be able to attend. :)
> It looks like two people showed to the meeting. At 12 after I and one other were in attendance.
Did you talk about much?
> As mentioned, I am delivering a training, and unable to make an appearance. Hopefully it happens next week, and then I'll be able to attend. :)
Okay, cool. :-) Next week I will definitely attend whether Ruben is able to attend or not. Hopefully with more notice more people will attend on the 29th as well.
I am a member!
We really need Rubén to attend the meetings. He is the one who can administer any changes, answer questions, etc.
I dropped him an email earlier today. I'm hoping that he will reply on the weekend. I am also checking on IRC most days to see if he is online.
I am a member!
IRC is probably best to reach him. His inbox is overflowing.
I caught Rubén on IRC and he will be sending an email to the trisquel-devel mailing list about the next meeting soon.
He also said Abrowser 18 would be coming out soon for all Trisquel versions.
I am a member!
Start one.
Add these to the agenda:
1. Question one for Rubén is what to expect in regards to support for ath9k_htc firmware/kernel patching after 6's release. I don't see why the patch couldn't be done now and an update for when the firmware is released under a free software license seeing that the free firmware is coming.
2. Release date for 6? What is holding things up? Is there any idea when a release might happen?
Also- we have stuff we should be covering here:
http://trisquel.info/en/wiki/proposed-policies-procedures-solutions
Hopefully I'm not forgetting anything and we actually added anything and everything to that above page.
Reagrding ath9k_htc:
Correct me if I'm wrong but to my knowledge no one from Qualcomm Atheros has provided guidance as to when this firmware will be released. It could be tomorrow, it it could be eight months from now.
Based on not knowing when it'll happen, I think it makes sense to leave the kernel as-is for now for a few reasons.
One is that it means having a feature that, until the release of the freed firmware, is only usable for proprietary software. No need to have such a thing in Trisquel.
The second is that the freed firmware will need to be packaged up and included in the Trisquel repository. That package should probably be included in the ISO and installed by default, to avoid a chicken-and-the-egg problem where someone might have this device as their only network connection.
That freed firmware can't be placed into the Trisquel repositories (and then the ISO) until it actually exists. Knowingly including non-free software in the repositories would be against Trisquel policy and potentially even put their FSF endorsement in jeopardy.
So this leaves us with:
* If the freed firmware is released before Trisquel 6, package it up into the Trisquel repo, add it to the ISO, and modify the kernel such that it'll load the firmware.
* If they freed firmware is not release before Trisquel 6, leave everything as-is and then do the above once the firmware is available. This will mean generating new ISO images so that the new kernel is included but they'd need to have been regenerated so that the firmware could included anyway to avoid the chicken-and-the-egg problem if someone has this as their only network connection.
I am a member!
That sounds like a good plan of action. I haven't talked to Rubén so much of this is me thinking out loud here.
The ath9k_htc firmware not being free software yet is correct. You are also correct that there is no time line for its release under the GPL license. The reason though is because Qualcomm is waiting on the third party developers to finish cleaning up the code. Third party volunteers are cleaning it up now. They have the source code under NDAs. However they don't have the permission/authority to release that code. However there is a contract which states that Qualcomm will release the code upon the completion of the clean up. Not to mention that all the people who needed to sign off already have. I believe once the cleaned up code is submitted back to Qualcomm it'll be released shortly thereafter. My understanding is nothing is holding it back other than the volunteer developers having the time to complete the clean up.
As far as modifying the kernel now or waiting shouldn't make any difference. It is merely a loading mechanism for the firmware. The code in the driver is already free and it is only the firmware which are not- but the free firmware pieces are imminent anyway. I don't see how this would violate any principle, requirement, rule, or jeopardize endorsement by the FSF. It is merely preparing for the inevitable at this point.
If the loading mechanism was a non-free bit within the driver that would violate the the requirement that everything be free software. Obviously just because it can load it doesn't mean it would be loading anything given that the non-free firmware would still be missing from the repository / distribution. It would simply report an error about missing firmware if the user connected a device with a compatible chipset. We aren't even planning to offer this adapter until the firmware code is available under the GPL. We will offer adapters though as soon as the code is available even if Trisquel doesn't support it for a while. If there is concern that users might accidentally download the firmware not realizing tha it is non-free then the message should simply be adjusted to make people aware the free firmware is forthcoming and to check a particular web site for details on its release.
It doesn't really matter how/when it is done though and your suggestion is still wise. My main thought on this was if it is easy/quick time wise for Rubén to make the changes to the kernel and release an update with the newer firmware after the distributions release to do that. If not then lets modify it now so we don't have to do extra work later. It is a small change though as you and others have stated. However it will be a kernel update still. So there are downsides to doing it that way even if it is easy for Rubén to do. It would mean more bandwidth used, the generation of a new ISOs, and even possible confusion unless the number updated too (6 is released and then updated to 6.01 with the new firmware).
So in summary, the kernel in 12.04 and above already has the free driver in it and doesn't need direct modification in the future. It's just the .deb with the system firmwares needs to be customized to include the firmware when it is ready.
When the firmware is ready, the Trisquel firmware .deb will get a point release with the Atheros firmware. Until then, it is blocked and/or left out.
It would be cool to get the free firmware added to the main Debian and Ubuntu repos too down the road.
I am a member!
I think you have said this wrong.
1. Ubuntu has the non-free firmware currently.
2. Ubuntu has a kernel which is a derivative of the mainline kernel. It loads the non-free firmware. Most kernels derived from the mainline kernel include non-free bits and already would load the free firmware if it were released. The driver portion for the ath9k_htc (and this chipset) is already free software. It however loads a non-free firmware provided it is present.
3. Trisquel uses a kernel derived from the mainline kernel, and I believe Ubuntu kernel, except with the non-free bits removed. It apparently is also patched to disable the free driver from loading ANY firmware. Free or not it doesn't matter. This is because there is only non-free firmware currently available. The only firmware a user would be loading right now is non-free.
4. Trisquel could re-enable the bit in the ath9k_htc driver code within the kernel to load its firmware and the distribution would still be 100% free. The reason Trisquel appears not to be doing that currently is there is no free firmware in existence and because the non-free firmware is obviously excluded from the distribution. If this bit was changed back to load firmware it would simply produce a message indicating that it is missing firmware when a card was inserted. This could be seen as encouraging users to install the non-free firmware. This message could be changed to avoid that I think. However there appears to be no good reason to pre-patch the kernel to load firmware until the free firmware is available and also included in the distribution since it is easy to make this change at the same time the free firmware is added to the distribution.
As far as debian goes and other distributions including it is almost certainly going to replace the non-free firmware. What will happen is the old firmware will be discarded as improvements get made to the GPL version of the firmware that can't be included in a non-free version. Once this happens there is no reason to maintain a non-free firmware at all.
I'd like to quickly bring up the Drupal wiki permissions for showing detailed revision logs (I think they are currently admin-only).
Reasons:
* Users can track changes by other users on the wiki.
* Also possibly better for translators who need to track changes.
I am a member!
Please feel free to add here : http://trisquel.info/en/wiki/2013-01-29-meeting your questions for the next meeting.
Mailing list announcement:
http://listas.trisquel.info/pipermail/trisquel-devel/2013-January/000714.html
I am a member!
This is a call for a development meeting:
Tuesday 29, at 1700UTC at #trisquel-dev
Topics to discuss:
* Toutatis release
* Kernel and security updates maintenance
* Spam/Trolling control
* Jabber service announcement
* LibrePlanet
Announcements:
* All packages for all Trisquel versions up to date
* Websites in track
See you there!
---- <<< from list...
If anybody can make it this is:
USA East Coast: 12 NOON EST Tuesday
West Coast: 9AM PST Tuesday
Anybody do the conversion for Europe?
I am a member!
I think this might be relevant > https://en.wikipedia.org/wiki/List_of_UTC_time_offsets
Meeting minutes and a full log have been added at https://trisquel.info/en/wiki/2013-01-29-meeting-0
Thanks to everybody for your help!
I am a member!
Sorry to have missed such a productive meeting; good work, all! Am
especially pleased to have dates for the candidate and final release.
Make sure new, improved captcha is accessible to screen reader users.
Will forum and email list stay synched? Maybe we need better
authentication mechanism for mailing list subscriptions?
Cheers,
Dave H.
Those lines from daroal and quidam motivated me:
18:14:01 magic_banana: oh, you mean a general XMPP manual (not specific to this server)?
18:14:05 magic_banana: feel free to add one!
I created this manual. Feel free to improve upon it:
- Correct the language (I am not a native English-speaker);
- Correct the errors (I am not a specialist in instant messaging protocols);
- Add useful information that would be missing;
- Add a new manual (and a link in this manual) for the specific use of Pidgin (similar to this one about IRC);
- Translate the manual to another language.
I also filled a feature request named "Non-English users should be redirected to the English documentation when inexistant in their language". Feel free to comment on it (not in this forum but in the "issue" itself).
Yeah, I was either eating or working on a project and didn't notice the time. From my perspective, I'm glad that jxself is handling the Linux Libre kernels for Trisquel. As stated from our previous discussion(s), it would be nice if he also mirrored the kernel in a launchpad PPA so the kernels that matched the Ubuntu version weren't only for those who have access to the Trisquel repositories. Share the luv. :-P
I am a member!
We need to go through the meeting notes and update the wiki.
... any volunteers?
* not sure if somebody did this already... mentioned it in the IRC channel after the meeting though.
I am a member!
>AGREED: to publish Toutatis RC on feb 8 and final release on feb 15 (quidam, 18:56:26)
Thank you based Ruben, and all those involved in the development meeting
I am very sorry I didn't make it to the meeting! I didn't have internet access, I'm guessing because of power outages in the area. :-(
I just started reading the meeting log - it looked very productive!
I am a member!
It was a LONG meeting!
I'm not sure how productive it actually was. I'd like to see results from it. I guess Rubén is going to give jxself access to at least some pieces. This is a good move.
I also did get two questions answered (although one was partly answered) about linux-libre / ath9_htc and the other about back porting pieces (not that this is going anywhere, and technically it was after the meeting).
- Vous devez vous identifier ou créer un compte pour écrire des commentaires