Project

General

Profile

ContributorsMeetingJuly2019 » History » Revision 121

Revision 120 (Denis 'GNUtoo' Carikli, 07/30/2019 04:12 PM) → Revision 121/187 (Denis 'GNUtoo' Carikli, 07/30/2019 04:13 PM)

h1. Replicant contributors meeting the 27,28 July in France 

 h2. Date 

 The meeting will take place the 27 and 28 July 2019. 

 h2. Location 

 The event will take place in the "April":https://en.wikipedia.org/wiki/April_(French_association) "office":https://april.org/en/contact-us. April is an association for the defense and promotion of free software. 

 *Precise location*: April, 44/46 rue de l'Ouest, bâtiment 8, 75014 Paris (It's accesible through the "place de la Catalogne", on the left of the "Biocoop" supermarket). Select "April" on the intercomm. 
 *map:* https://www.agendadulibre.org/events/19754 
 *Phone number of the April office*: 01 78 76 92 80. 

 The April office is relatively close to the "Gare Montparnasse railway station.":https://en.wikipedia.org/wiki/Gare_Montparnasse 

 In addition to trains, Gare Montparnasse railway station has also, in the same building, there is a metro station where you can access the metro lines "4":https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_4, "6":https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_6, "12":https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_12 and "13":https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_13. 

 h3. Related events 

 On Friday, a dinner will take place. 

 *Dinner meeting time:* From 18h30 to 18h45 
 *Dinner meeting location*: "Near the Montparnasse tower":https://redmine.replicant.us/attachments/1614/Friday_dinner_meeting.jpeg . The Montparnasse tower is very close to the Montparnasse railway station. 

 At 18h45 we will try to find a Restaurant nearby. 

 h3. Airports 

 The two main airports near Paris are the "Roissy Charles de Gaulle International Airport":https://en.wikipedia.org/wiki/Charles_de_Gaulle_International_Airport and the "Paris Orly Airport":https://en.wikipedia.org/wiki/Orly_Airport. 

 h4. Roissy Charles de Gaulle International Airport <-> Gare Montparnasse railway station 

 To get to the Gare Montparnasse railway station from the Roissy Charles de Gaulle International Airport by public transportation, you can take the "RER":https://en.wikipedia.org/wiki/R%C3%A9seau_Express_R%C3%A9gional B metro line up to the "Denfert-Rochereau station":https://en.wikipedia.org/wiki/Denfert-Rochereau_(Paris_M%C3%A9tro) where you can take the "line 6":https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_6 up to the Gare Montparnasse railway station. 

 h4. Paris Orly Airport <-> Gare Montparnasse railway station 

 To get to the Gare Montparnasse railway station from the Paris Orly Airport by public transportation, you can take the "Orlyval":https://en.wikipedia.org/wiki/Orlyval up to the "Antony":https://en.wikipedia.org/wiki/Antony_(Paris_RER) stop, and from there take the "RER":https://en.wikipedia.org/wiki/R%C3%A9seau_Express_R%C3%A9gional "B":https://en.wikipedia.org/wiki/RER_B metro line up to the "Denfert-Rochereau station":https://en.wikipedia.org/wiki/Denfert-Rochereau_(Paris_M%C3%A9tro) where you can take the "line 6":https://en.wikipedia.org/wiki/Paris_M%C3%A9tro_Line_6 up to the Gare Montparnasse railway station. 

 h2. More information 

 "Wikivoyage":https://en.wikivoyage.org has a page on "Paris":https://en.wikivoyage.org/wiki/Paris and "France":https://en.wikivoyage.org/wiki/France which have many practical information such as: 
 * Visa and other border requirements 
 * How to stay safe 
 * The climate 
 * Information on the public transportation 
 * SIM cards 

 It also has pages on the "Roissy Charles de Gaulle International Airport":https://en.wikivoyage.org/wiki/Paris_Charles_de_Gaulle_Airport and the "Paris Orly Airport":https://en.wikivoyage.org/wiki/Paris_Orly_Airport with more details on the public transportation lines to use to go to Paris, which tickets to buy, etc. 

 h2. Contacting the organizers 

 You can use the Replicant [[PrivateContact|private contact mail address]] in advance to obtain the cellphone number of an organizer to be able to    use it in case of issue (like being lost, not finding the location of the meeting, etc). 

 h2. Confirmed presentations 

 |_. Topic |_. Time and dependencies |_. Status |_. Rationale | 
 | Replicant history: "WIP source code":https://git.replicant.us/contrib/GNUtoo/presentations.git/tree/Replicant_contributors_meeting_27_28_July_2019_Paris_France/history?h=GNUtoo/wip    | | | * Has some context that might be useful for new Replicant contributors 
 * Has information on the relationship between Replicant and GNU/Linux | 
 | "Porting AOSP for a new device":https://redmine.replicant.us/attachments/download/1618/Porting-AOSP.pdf | Should be in the beginning / before technical talks. Can be after Replicant history presention. | | | 
 | Replicant and bootloaders: "Source code":https://git.replicant.us/contrib/GNUtoo/presentations.git/tree/Replicant_contributors_meeting_27_28_July_2019_Paris_France/bootloaders    "pdf":https://redmine.replicant.us/attachments/download/1613/Replicant_and_bootloaders.pdf | | | * The Galaxy SIII (and similar devices) bootloader status is complicated and relevant to Replicant 9 | 
 | Replicant and modems:  
 * "Source code":https://git.replicant.us/contrib/GNUtoo/presentations.git/tree/Replicant_contributors_meeting_27_28_July_2019_Paris_France/modems 
 * "Introduction pdf":https://redmine.replicant.us/attachments/download/1615/Replicant_and_modems_Introduction.pdf 
 * "Samsung IPC PDF":https://redmine.replicant.us/attachments/download/1616/Replicant_and_modems_Samsung-ipc.pdf 
 * "oFono based Java RIL pdf":https://redmine.replicant.us/attachments/download/1617/Replicant_and_oFono-based_Java_RIL.pdf | | | * Meant to enable new contributors to work on the modem part | 
 | Discussions on the bootloader situation on the Galaxy SIII and similar smartphones | After the talk on the bootloaders | | | 
 | Discussions on minimal requirement to accept a device in Replicant: 
 * Do we still accept devices with modems that are not isolated? 
 * Do we have plan to require free software bootloaders?  
 * Do we require replacable batteries? | After the talk on modems and bootloaders | | | 
 | Discussions on the future of Replicant: 
 * Which devices do we target 
 * Do we continue focusing on devices with signed bootloaders 
 * Allwinner tablets, upstream Linux, and scalability 
 * Devices with non-replacable batteries | After the talk on Replicant history | | 
 | Discussion about Upstream components, design choices, and cultural re-appropriation of technology 
 * [[Upstream]] 
 * Issues when combining together different build systems (Example: Android build system with Kconfig) 
 * Sharing work with GNU/Linux to enable more political control and cultural re-appropriation of mobile device in the long run, and the risk associated with it 
 * Android upstream anti-features and political design choices 
 * How subjective security is, threat models, and the difference between free software and device maker point of view | Please take a look at [[Upstream]] before attending if possible. | | | 
 | Graphics acceleration on Replicant | Should optimally be before Paul leaves | | | 


 Known schedule constraints: 
 * On Sunday, the April offices will be closed at 16h, so we need to already have packed everything and left the room at this time. 

 h2. Still being defined, to be confirmed 

 Precise schedule: 
 *Saturday:* 10:00 -> 17:00 (possibly continuing to the night)? 
 *Sunday:* 10:00 -> 14:00? 

 *Poll for using the April offices*: https://framadate.org/ReplicantContributorMeetingJuly2019RoomOccupation 

 h3. Presentations: 

 * The Graphic stack in Replicant 9 
 * Porting Replicant to Android 9 by Joonas Kylmälä? 
 * A discussion about [[Upstream]] components, [[Upstream#Design-decisions|design choices]], and cultural re-appropriation of technology? 
 ** How subjective security is, threat models, and the difference between free software and device maker point of view 
 ** Sharing work with GNU/Linux to enable more political control and cultural re-appropriation of mobile device in the long run, and the risk associated with it 
 ** Android upstream anti-features and political design choices 
 ** Issues when combining together different build systems (Example: Android build system with Kconfig) 


 h3. Over a cup of coffee: 

 * Fil is available for informal chatting about RepWifi 

 h3. Hardware 

 h4. Devices 

 Good practices: 
 * It would be a good idea to put your name on a sticker on the device if possible 

 |_. Person |_. Hardware |_. Comments |_. Usage | 
 |/12. GNUtoo | Galaxy SIII (I9300) with the stock bootloader | | 
 | Galaxy SIII 4G (I9305) with the stock bootloader | | 
 | Galaxy SIII 4G (I9305) with u-boot | | 
 | Galaxy Nexus (I9250) with the stock bootloader | | 
 | Galaxy SII (I9100) with the stock bootloader | | 
 | GTA04 A3 | | | 
 | GTA04 A4 | | | 
 | Optimus black (P970) | | | 
 | GTA01 |/4. Used in a presentation about Replicant history | | 
 | GTA02 | | 
 | HTC Dream | | 
 | N900 | Testing the battery charger driver is still needed but require a heavy PSU | 
 |/4. Fil | Galaxy Note 2 (N7100) stock bootloader | [Working] Available for non-critical tests | * Test the upstream touchkey driver 
 => TODO: 
 * Bring a Parabola microSD (GNUtoo) 
 * Test the patch with an I9300 
 * Rebase the patch on master or linux-next | 
 | Galaxy SIII (I9300) stock bootloader | [Working] Test Subject available for any experiment | | 
 | Galaxy SIII (I9300) stock bootloader | [Bricked] available for hardware hacking | * Test "fixing the phone":https://media.ccc.de/v/34c3-8784-emmc_hacking_or_how_i_fixed_long-dead_galaxy_s3_phones | 
 | Galaxy Tab 2 7.0 (GT P3100) | Property of the Replicant Project | * -Add support for it in the [[BackupTheEFS]] instructions- | 
 |/2. Paulk | Galaxy Note (N7000) | | * Add support for it in the [[BackupTheEFS]] instructions | 
 | Galaxy Tab 2 10.1 (P5100) | | * -Add support for it in the [[BackupTheEFS]] instructions- | 
 |\2. Looking for a Galaxy Note 8.0 (N5100) | | * Add support for it in the [[BackupTheEFS]] instructions | 
 | | Add devices you can bring here or you want other to bring | Add comments, requirements, etc | Add what you want to do with the device or other people's device | 

 h4. Debug utilities 

 |_. Person |_. Hardware |_. Comments |_. Usage | 
 |/6. GNUtoo | Serial port cable with variable resistors |    | 
 | Multimeter |    | 
 | Simtrace 1: Can get the dialog between the modem and the SIM card in wireshark |    | 
 | SIM card that is not recognized in Replicant (STK related?) | Test on Replicant 4.2 and on Replicant 6.0 | 
 | SIM card + phone that can trigger the audio call issue | | 
 | Sigrok compatible adjustable power supply | Not sure to bring it (heavy) | 
 |/1. Fil | SIM card that is not recognized in Replican 6 | | 

 h2. Discussion results 

 |_. Topic |_. Impact |_. Result |_. Consensus in that meeting | 
 | Do we care about supporting devices with non-removable batteries? | This impacts the design decisions as supporting new devices through upstream code takes longer than hacking together the vendor code not to require nonfree libraries. Because of that either users, repair shops or Replicant device vendors will need to have a viable way to change the non-removable battery. If this is not the case it would force the Replicant project to race to try to support a device while it's still being sold new. If not users buying devices would have a battery that doesn't last enough when bought second hand. |    It's too hard, even for vendors like Technoethical which have some expertise in that area to upgrade non-removable batteries when the case is sealed. Practically speaking that could be done but the end result would not be as good as the new device, like the sealing would not be as good, making users damage the phone as they would expect the sealing to work. | The cost for Replicant would be too high, so participant thought that not caring about devices without non-removable batteries and prioritize to support devices in Replicant through upstream project was better | 
 | How to handle the various keys used to sign releases, and other related things and should the recovery check signatures    | * The installation instructions need to be as simple as possible => If users have to check the signatures only once it's better. So now users check the recovery only, which then checks the rest. 
 * The people involved in local Replicant install parties (in Paris) will help local Replicant users to update the recovery 
 * Technoethical previously stated that the users having bought devices already running Replicant through them are either able to install new recoveries or that they are able to help them do it |\2. * Keys usage in the short term: Developers sign with their keys 
 * Keys usage in the longer term (requires time to implement it): use a keyring with the public key of several Replicant developers, like it is done in Parabola with the parabola-keyring package  
 * Keep checking signatures by default in the recovery | 
 | Improve information for current and potential Replicant users |\3. |3\. * Fil Bergamo plans to work on an unified table that has various information on devices supported by Replicant: 
 * He plans to add in information about what hardware feature are supported or not 
 * He plan to add information about how likely are the devices to be supported in the next Replicant 6 and 9 releases | 
 | Improving the Replicant website |\3. |3\.    *It would be nice to make information for users more accesible on the main page (https://replicant.us) |  
 | Making it easier for anyone to contribute to Replicant |\3. |3\. * Fill plans to work on a supported | 

 h2. Video encoding 

 h3. Sound 

 The microphone was mono, with only a channel from the left. So we need to create mono audio files from the videos. 

 To make the sound go in both channel, do the following in Audacity: 

 * Create a new project with File -> New 
 * Import the video file with File -> Open 
 * Click on the audio channel -> Split Stereo to Mono 
 * Delete the second channel 
 * File -> Export -> Export Audio 

 h3. Video  

 The videos were recorded with some equipement that was lent to us: 
 * A camera that outputed mp4 files on an SD card 
 * One wireless microphone 

 => Next time ask if they have a second microphone available for questions. 

 * The videos are splitted in several mp4 files 
 * The part before the presentation has to be removed 

 Kdenlive is being used for that. 

 Using kdenlive with Nouveau (with default settings), xfce4 under Parabola manage to freeze the screen (there are some messages like "nouveau: kernel rejected pushbuf: Cannot allocate memory" that can be observed when launching kdenlive through SSH). 

 To workaround that the following was used: 
 * gnome3 (probably under wayland) 
 * pushbuffer enabled in the module parameters for nouveau 

 => It's still slow while zomming in the timeline but it manage not to freeze the whole graphics stack. 

 To edit a video with kdenlive: 
 * create a project, and select the right resolution (720p 25fps here) 
 * Import the videos files to use with project->Add clip 
 * Also import the mono audio files that you just converted with audacious 
 * Import the clips from the Project bin into the timeline with drag and drop 
 * Right click on the imported click in the timeline and select Ungroup 
 * Right click on the audio below, and select remove 
 * Import the mono audio file from the Project bin in the timeline with drag and drop 

 Once that is done you can remove the part before the presentation this way: 
 * In the timeline you can zoom at the maximum level  
 * Use the cisors to divide the zone in two, you only need to click once with the left click to separate two zones 
 * Select the arow again (on the left of the cisors) and right click on the zone to remove, and select remove 
 * Do that for both the video and the audio tracks, then move both to the left to remove the gap done by the removal if there is one. 

 h2. Files