From mboxrd@z Thu Jan 1 00:00:00 1970 Path: news.gmane.io!.POSTED.blaine.gmane.org!not-for-mail From: Jean Louis Newsgroups: gmane.emacs.devel Subject: Re: NonGNU ELPA Date: Sun, 22 Nov 2020 02:19:25 +0300 Message-ID: References: <87v9eg4gm5.fsf@gnu.org> <87o8k7yt7n.fsf@gnu.org> <87ima56h1a.fsf@gnu.org> Mime-Version: 1.0 Content-Type: text/plain; charset=utf-8 Content-Transfer-Encoding: 8bit Injection-Info: ciao.gmane.io; posting-host="blaine.gmane.org:116.202.254.214"; logging-data="7840"; mail-complaints-to="usenet@ciao.gmane.io" User-Agent: Mutt/2.0 (3d08634) (2020-11-07) Cc: emacs-devel , Amin Bandali , Stefan Monnier , Richard Stallman To: =?utf-8?Q?Cl=C3=A9ment?= Pit-Claudel Original-X-From: emacs-devel-bounces+ged-emacs-devel=m.gmane-mx.org@gnu.org Sun Nov 22 02:34:56 2020 Return-path: Envelope-to: ged-emacs-devel@m.gmane-mx.org Original-Received: from lists.gnu.org ([209.51.188.17]) by ciao.gmane.io with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.92) (envelope-from ) id 1kgeHI-0001wy-8F for ged-emacs-devel@m.gmane-mx.org; Sun, 22 Nov 2020 02:34:56 +0100 Original-Received: from localhost ([::1]:53106 helo=lists1p.gnu.org) by lists.gnu.org with esmtp (Exim 4.90_1) (envelope-from ) id 1kgeHH-00070o-A5 for ged-emacs-devel@m.gmane-mx.org; Sat, 21 Nov 2020 20:34:55 -0500 Original-Received: from eggs.gnu.org ([2001:470:142:3::10]:33494) by lists.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1kgeG8-0006BM-Mv for emacs-devel@gnu.org; Sat, 21 Nov 2020 20:33:44 -0500 Original-Received: from static.rcdrun.com ([95.85.24.50]:52135) by eggs.gnu.org with esmtps (TLS1.2:ECDHE_RSA_AES_256_GCM_SHA384:256) (Exim 4.90_1) (envelope-from ) id 1kgeG6-0000a6-4C; Sat, 21 Nov 2020 20:33:44 -0500 Original-Received: from localhost ([::ffff:41.202.241.56]) (AUTH: PLAIN admin, TLS: TLS1.2,256bits,ECDHE_RSA_AES_256_GCM_SHA384) by static.rcdrun.com with ESMTPSA id 00000000002C000B.000000005FB9BFF2.00003AA9; Sun, 22 Nov 2020 01:33:38 +0000 Content-Disposition: inline In-Reply-To: Received-SPF: pass client-ip=95.85.24.50; envelope-from=bugs@gnu.support; helo=static.rcdrun.com X-Spam_score_int: -18 X-Spam_score: -1.9 X-Spam_bar: - X-Spam_report: (-1.9 / 5.0 requ) BAYES_00=-1.9, SPF_HELO_PASS=-0.001, SPF_PASS=-0.001 autolearn=ham autolearn_force=no X-Spam_action: no action X-BeenThere: emacs-devel@gnu.org X-Mailman-Version: 2.1.23 Precedence: list List-Id: "Emacs development discussions." List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: emacs-devel-bounces+ged-emacs-devel=m.gmane-mx.org@gnu.org Original-Sender: "Emacs-devel" Xref: news.gmane.io gmane.emacs.devel:259570 Archived-At: * Clément Pit-Claudel [2020-11-22 01:22]: > On 11/21/20 4:18 PM, Stefan Monnier wrote: > > I don't expect it will become significantly larger than the actual ELPA > > archive itself: > > I think the right metric would be the MELPA archive: I don't know how bit a complete checkout is of all MELPA packages. It is about 14 GB. If I change it slightly to --depth 1 it is about 10-11 GB. Git is in my opinion not for releasing software, it is for collaborative development. Releases of any software from git or other version control systems should be packed and contain only what is necessary for the user who receives such package. This is also because authors or maintainers are deciding what is development version and what is stable version. Git sources need not be stable and they do not represent "release" and should not be regarded as release how MELPA is accepting them. The fact that many git repositories are online accessible does not make them software releases. Author's opinion on what is release and what is not shall be respected. But people did start going into direction that git is automatically stable version which puts many people and their data at stake. Beside the git download size, when packages become packages after building they are not so large, if I remember well just under 600 MB. I am doing review of MELPA packages. There are many useless packages and many unsafe and not polished and those repeating functions which already exists. I would not include such. There are those where author's name is not known as it is written only as a nick. For me it would be legal problem as there is no truthful authentic relation between the author who is not legally named "zack" (example) and the receiver of software. Receiver would not know from which entity or person did receive get the license, or both parties would not have any option of defense or enforcement by the law. > The main problem would be cases in which an emacs mode exists as > part of a larger repo (like llvm-mode, which is part of lplvm — it > was removed from MELPA because it took too long just to close the > repo). Isn't it not so that Emacs packages shall be either .el or .tar files? Those packages that do not provide such releases and are useful can be anyway packaged in non-GNU ELPA, why not? There is no need to replicate git repositories, but rather actual packages regardless if such are part of git repository or not.