Date: Tue, 10 Feb 2026 01:17:14 +0300 From: Vadim Goncharov <vadimnuclight@gmail.com> To: "Poul-Henning Kamp" <phk@phk.freebsd.dk> Cc: Graham Perrin <grahamperrin@gmail.com>, freebsd-hackers@freebsd.org Subject: Re: FreBSD pkgbase MUST be ebandoned Message-ID: <20260210011714.57ed31c9@nuclight.lan> In-Reply-To: <202602092120.619LKVk4015515@critter.freebsd.dk> References: <0iqhe92aheNJohSnhh8-hXkXhQsaRG4D64nLTlTSIPgd6Iit07IwlMwmn-mIS-Qtp9KuZElphybTlYDIVTUDcVGpHWaUbQVGPKt53NSL5Jg=@proton.me> <1678741437.20260206163514@yahoo.com> <07af999d-3c7b-4fe2-8ed2-a37cf89b663b@quip.cz> <20260206230926.50c6afed@nuclight.lan> <aead1402-e403-4b61-943b-0938881d2cde@gmail.com> <20260209235855.7de86893@nuclight.lan> <202602092120.619LKVk4015515@critter.freebsd.dk>
index | next in thread | previous in thread | raw e-mail
On Mon, 09 Feb 2026 21:20:31 +0000 "Poul-Henning Kamp" <phk@phk.freebsd.dk> wrote: > -------- > Vadim Goncharov writes: > > > So, the pkgbase project, as it is in it's current form, MUST be abandoned > > (yeah, in a sense by RFC 2119) and something much serious and > > mission-critical should be made instead. > > Absolutely nothing prevents you from shutting up and coding something > better, so why dont you ? > > That would be a much more productive use of you time than standing > on a soapbox yelling demands of other people, over which you have > neither sway no authority. > > If you are right in your claims, you will out-compete pkgbase in no time. Dear Poul, Let me educate you about how human society (and, by consequence, opensource community), works. It works by a thing called "division of labor". Like in, "Let them sing in Grand Theater, and I will do surgery". That is, most effective usage of time of every society member is in areas where this member was trained, has competence, skills, experience, etc. In other areas it will be less effective, up to on order of magnitudes - this is important: by utilizing each member in proper area of expertise, the society built is much more better in comparison to when in non-trained. Open-source community works similar: every member contributes in area of best knowledge, contribution of one individual can be small but multiplied by numbers, complete ecosystem is built. Yes, ecosystem - it is not tied to one particular project or even operating system, it's much more broader [*]. Now it is quite easy to see ethical aspect here: if member X makes it best, and member Y makes it best, and everyone uses from everyone, - why should they allow member Z to make not even non-best (that's debatable) but actually bad and harmful things? Are they (X and Y) obliged to do Z's work *instead* of Z, where they will spend much more time (than Z) affecting that now they're doing less work on *their* parts? That is, nobody need to have sway or authority to say: - Don't be faggots! because it's just fundamental human right arising from ethics. Yes, don't break things for other people. FreeBSD always has been famous by being stable and reliable, and by ability to exploit and upgrade a box several major releases behind (just worked and forgotten, discovered by newly hired sysadmin and then upgraded without much pain, in contrast to linuxes in such situations). Which is degraded for years and years. They do delete ports without maintainers, they do delete packages for 12-STABLE exactly the day it's EOL - as if another 100 Gb on servers will make us into poverty, as if it costs as 100 Tb to Amazon! So it is NOT my responsibility to save your ass and spend my time rescuing after your fuckups. You broke things - you fix them, ethics here is as simple as this. > And should you fail, the rest of us will still have pkgbase, which > is clearly better than freebsd-update. > > Clearly a win-win sitution! With one "minor" detail making it just a lie: it's not better, it is broken and doesn't work. I have no time to extract and translate individual messages from chat for several days (interspsersed with other discussions), one can try e.g. LLM to automate this one-message-per-one-URL if curious. That man's misadventures seem to start from this message https://t.me/freebsd_ru/735736 : | # pkg upgrade | ld-elf.so.1: Shared object "libutil.so.10" not found, required by "pkg" which by itself is very funny in two lines debunking whole pkgbase marketing. In particular, from his messages, seems it not possible to go 15-R to 15-S or 16-C by pkgbase, and other bugs. [*] E.g. I'm currently working as IETF CBOR WG, and consider it more important for overall ecosystem than FreeBSD. -- WBR, @nuclighthome | help
Want to link to this message? Use this
URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20260210011714.57ed31c9>