From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from lists.gentoo.org (pigeon.gentoo.org [208.92.234.80]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (2048 bits) server-digest SHA256) (No client certificate requested) by finch.gentoo.org (Postfix) with ESMTPS id 95DF215803E for ; Thu, 4 Jan 2024 00:17:52 +0000 (UTC) Received: from pigeon.gentoo.org (localhost [127.0.0.1]) by pigeon.gentoo.org (Postfix) with SMTP id AC2D62BC03B; Thu, 4 Jan 2024 00:17:48 +0000 (UTC) Received: from smtp.gentoo.org (woodpecker.gentoo.org [140.211.166.183]) (using TLSv1.3 with cipher TLS_AES_256_GCM_SHA384 (256/256 bits) key-exchange X25519 server-signature RSA-PSS (4096 bits) server-digest SHA256) (No client certificate requested) by pigeon.gentoo.org (Postfix) with ESMTPS id 396F22BC01E for ; Thu, 4 Jan 2024 00:17:48 +0000 (UTC) Received: by mail-yw1-f174.google.com with SMTP id 00721157ae682-5f07f9d57b9so28067b3.1 for ; Wed, 03 Jan 2024 16:17:47 -0800 (PST) X-Gm-Message-State: AOJu0YwrkD+FRxftKGOJeHcocePxFKaqpv5Lh5GRpl7YHGm7+el0ofkv ywjz4JOyNVSMApXQHaMQqj6neolPcw9IZl2/TYc= X-Google-Smtp-Source: AGHT+IHUJ9glI7z7WuYXdLOYCViX7tEM6/sCZqCum3tDlVHahDcwzLO961phgZ92VEAbFZ22ybKpumHEl3D8qVj6orw= X-Received: by 2002:a81:4428:0:b0:5ea:1bf4:11e5 with SMTP id r40-20020a814428000000b005ea1bf411e5mr12712399ywa.9.1704327465266; Wed, 03 Jan 2024 16:17:45 -0800 (PST) Precedence: bulk List-Post: List-Help: List-Unsubscribe: List-Subscribe: List-Id: Gentoo Linux mail X-BeenThere: gentoo-dev@lists.gentoo.org Reply-to: gentoo-dev@lists.gentoo.org X-Auto-Response-Suppress: DR, RN, NRN, OOF, AutoReply MIME-Version: 1.0 From: "Jason A. Donenfeld" Date: Thu, 4 Jan 2024 01:17:33 +0100 X-Gmail-Original-Message-ID: Message-ID: Subject: [gentoo-dev] default USE flags, binpkgs, and why developers should keep doing what they're doing To: gentoo development Content-Type: text/plain; charset="UTF-8" X-Archives-Salt: 2b812fc0-bbd9-4e0f-85bd-27a4d28ef6f7 X-Archives-Hash: 8ea0f6e6dda45610f3ac21d7d594a5ec Hi folks, I'd like to share some thoughts regarding the new binpkg builders dilfridge & co have been working on. First of all, that's super exciting. Congratulations on launching that. I look forward to seeing ~ builds in the future and also the positive effect this can have on our ecosystem. But I want to address a *potential* negative impact on our ecosystem. As this takes off, users will quickly find that sticking with default USE flags means lightning fast "builds", because they don't have to build anything; they just get it from the binhost. This in turn creates some incentive structure for users not to change the default USE flags, in a sense penalizing customization. Obviously it's not a real penalty though; nothing has changed with building from source, and just the default USE flag emerges got a massive speedup. So while on the surface, the customization negative pressure angle is kind of concerning, it's not really that bad when you think about it. But what the speedup of default USE flags will do, however, is naturally apply a positive pressure on *developers* to enable more USE flags by default in ebuilds. And this is what concerns me. I've long considered the default USE flags of packages to be sort of the "Gentoo wisdom" of a default setting. It's not overly minimal, but it's most certainly not bloated either, like full on binary distros that always build the whole kitchen sink. It's kind of always been this nice middle ground that developers pickup on when they start and carry through as it's always been, learning what's more of an art than a science. What I fear is that this artform will be gradually lost, as users ask for this or that default USE flag to be changed in this or that package, and a developer reasonably thinks, "oh sure, not a big deal, I can always turn it off in my personal builds." And then gradually all the default USE ebuilds become chunky, and the old art of Gentoo default USE flags is forgotten. This isn't the kind of thing that would happen overnight. But it does seem like there are mechanisms in place which might bring about this kind of gradual cultural shift. Maybe. So, this is all to say, everyone, let's keep doing what we've always been doing! Jason -- Jason A. Donenfeld zx2c4@gentoo.org