c'est illégal à cause de lois sur le droit d'auteur particulièrement absurdes Qu’on considère une loi absurde ou non ne change pas grand chose à sa validité dans la pratique ;)
Soyons clair : je ne condamne pas plus la pratique de l’abandonware que celle plus générale de la distribution sous le manteau de jeux vidéo. Et je ne suis pas non plus satisfait par les lois actuelles sur le droit d’auteur, en particulier dans le domaine du logiciel (les durées sont clairement fantaisistes).
Mais je trouve que ce n’est pas très honnête de différencier l’abandonware des autres pratiques de partage non-autorisé. Et je trouve dommage que certaines personnes (je ne parle pas de toi en particulier) finissent par croire à la fable qu’on aime se répéter, et se retrouvent à penser que l’abandonware est une pratique légale.
Mais d’un autre côté il faut bien comprendre que ça a exactement le même niveau de légalité que ThePirateBay et autres sites dans le genre (dont l’Internet Archive, malgré certaines prétentions affichées).
Donc si le but est simplement de te procurer gratuitement des jeux, il existe plein d’alternatives tout aussi illégales qu’Abandonware France ;)
Poster un message similaire dans la section francophone des forums GOG aboutit à un signalement par un utilisateur (pas étonnant, cette section des forums est turbo-réac, ils sont probablement en deuil), mais surtout, beaucoup plus décevant, la suppression du message et un avertissement de la part de la modération du forum (des employés de GOG, pas de simples clients de la boutique).
Ma réponse à leur modération se conclut tout simplement sur : I understand that celebrating the death of a well-known Nazi is not welcome on GOG, and won’t do that again.
Bah, au moins ça ne casse que le système, le genre de chose qui se répare par une simple réinstallation.
Steam fait beaucoup mieux, en supprimant spécifiquement les fichiers qu’on ne pourra pas récupérer : « Moved ~/.local/share/steam. Ran steam. It deleted everything on system owned by user. » — https://github.com/valvesoftware/steam-for-linux/issues/3671
We got a slightly similar experience with @Hetzner_Online@social.cologne in 2021, following a bogus claim by a copyright troll ("DMCA Force", seemingly working for the game development studio Cyan Worlds).
At no time did Hetzner try to check if the claim was legit, and forced us to take down all of our websites (at least they did not deleted them with no prior warning). They immediately sided with the copyright troll, against their legit paying customer.
We managed to get an apology from Cyan Worlds, who promised they are going to be more careful in the future with who they work with, and made the copyright troll retract their claims. But nothing from Hetzner. Not even an informal apology. Not even an acknowledgement of the claims retraction.
In our case we moved to self-hosting to ensure this would never happen again, but this might not be a good solution for you if you have much bigger hosting needs.
But there is still something that is really lacking: support for the RPM packages used by many distributions like Fedora, OpenSuse or Mageia.
If you happen to be an user of a distribution based on RPM packages, like to play video games but do not like the use of DRM to restrict access to games you bought, you’re the perfect person to help us fix this glaring problem with the current state of ./play.it!
So please get in touch with us if you would like to work on RPM support.
No prior software development experience is required, as this is something we would do together. We do not expect anyone to provide a patch by themselves, what we need is RPM users who would be willing to share with us the experience they have with this packages system, and together we could turn that into a ./play.it update that would allow the generation of RPM packages for DRM-free video games.
These will not become the canonical repositories, but their purpose is to provide new contributors a way to use some interface they are already used to, so they can send patches our way until they gain enough trust to get a direct access to the canonical repositories served directly from our server.
As a reminder, the canonical repositories that are replacing the forge.dotslashplay.it ones are the read-only repositories we self-host using cgit: https://git.dotslashplay.it/
The ReadMe files are not updated to reflect that yet, but this will be done, alongside the addition of a contribution guide, shortly after the release of the incoming 2.28.0 feature update. From this moment, the repositories hosted on forge.dotslashplay.it will probably be slowly phased out in favour of an e-mail based workflow.
Bugs and requests tracking are going to stay on forge.dotslashplay.it for now, but will probably be migrated to another system too at some point in the future. For that we still need to find a good bugs tracker (must be a bugs tracker only, with no support for extra things) that can be easily installed on Debian.
We have now been thinking about dropping the GitLab merge request workflow in favour of something far less standardized but at the same time giving much more freedom to the potential contributors: - Host your fork anywhere (including on our GitLab instance for now); - Work however you wish; - Send us an e-mail or a message on IRC, or even a message here on the Fediverse, asking us to integrate the changes from your fork.
The main downside would be the loss of the public discussions on merge request diffs, but it is not like these usually involve more than two persons (so could have been done by e-mail just as well).
Your comments on this suggested workflow are welcome, whether or not you are already a ./play.it contributor, and whether or not you were planning to become one in the future. Nothing has been really decided yet, so all feedback can help us going choosing the best way to handle contributions.
As usual the packages in the Debian repositories are going to be updated soon, for other distributions you should get in touch with your packagers/maintainers.
This update focuses mostly on correct errors handling:
* Ensure that a missing required extra archive stops the script execution. * error_icon_path_empty - Fix showing the error message when LANG is not set to fr_\* or en_\*. * launcher_target_presence_check - Ensure that the script execution stops if the binary path is not set. * unity3d_icon_path - Throw an error if no application type is found. * unity3d_application_exe_default - Throw an error if no binary could be found.
* Support for the following deprecated functions is dropped: - context_specific_value - icons_linking_postinst - organize_data - use_archive_specific_value * Support for the following message functions is dropped, see "New wrapper for messages display" for more details: - print_warning - print_error * archive_get_type is deprecated, archive_type should be used instead. See "Changes related to archives" for more details. * The legacy global variables ARCHIVE and PKG should no longer be used. See "Improvements of the context system" for more details. * The functions context_archive and context_package are deprecated. See "Improvements of the context system" for more details. * The variable APP_xxx_TYPE_VARIANT is no longer supported, GAME_ENGINE should be used instead. It can usually be omitted, like with APP_xxx_TYPE_VARIANT.
* A new function is provided to display all messages: print_message. It expects a priority level as its first argument: - print_message 'error' $message replaces print_error ; printf $message - print_message 'warning' $message replaces print_warning ; printf $message - print_message 'info' $message replaces printf $message
* The presence of an optional archive can be checked using a dedicated function: archive_is_available * The type of an archive is retrieved using a new function: archive_type. Unlike the previous function (archive_get_type), this will not trigger an error if no type is set for the given archive.
* A new function is provided to get the full path to an icon file: icon_full_path * Most functions related to icons extraction now take an icon identifier instead of the path to an icon file. * Reliance on the global variable WRESTOOL_OPTIONS is dropped.
* A new function is provided to set the current archive: set_current_archive * A new function is provided to set the current package: set_current_package * The functions used to get the current context have been renamed: - context_archive → current_archive - context_package → current_package
* A game script can rely on Visionaire engine support by setting GAME_ENGINE='visionaire' or by setting a value to VISIONAIRE_NAME. Since the engine value falls back to visionaire when VISIONAIRE_NAME is set, GAME_ENGINE can usually be omitted. * Default values are set for multiple variables: - APPLICATIONS_LIST - APP_xxx_EXE - CONTENT_LIBS_BIN_PATH - CONTENT_LIBS_BIN_FILES - CONTENT_GAME_BIN_FILES - CONTENT_GAME_DATA_FILES - CONTENT_DOC_DATA_PATH - CONTENT_DOC_DATA_FILES - PACKAGES_LIST - PKG_DATA_ID - PKG_DATA_DESCRIPTION - PKG_BIN_DEPS - PKG_BIN_DEPENDENCIES_LIBRARIES * For native Linux games, the used of system SDL is forced. * For WINE games, SDL_VIDEODRIVER is prevented from taking the value wayland.
This was a tricky one, a symbol conflict with modern libsdtc++.so.6. But since one of our contributors, Phil Morrell, already fixed a similar issue with the Unreal Tournament (1999) engine a couple years ago and documented the process, we had good hints on how to fix the similar conflict for Neverwinter Nights.
As usual with ./play.it-related things this ActivityPub instance is self-hosted, using snac2 — https://codeberg.org/grunfink/snac2 — on top of a Debian Trixie.