X-Authentication-Warning: delorie.com: mail set sender to djgpp-bounces using -f X-Recipient: djgpp AT delorie DOT com MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit Message-ID: <23003.13580.341954.371668@turnbull.sk.tsukuba.ac.jp> Date: Mon, 9 Oct 2017 17:36:28 +0900 From: "Stephen J. Turnbull" To: djgpp AT delorie DOT com Subject: Re: DOSBox? Mail-Followup-To: "Stephen J. Turnbull" Newsgroups: comp.os.msdos.djgpp In-Reply-To: <59d767c6$0$9406$426a74cc@news.free.fr> References: <59d767c6$0$9406$426a74cc AT news DOT free DOT fr> X-Mailer: VM 8.2.0b under 21.5 (beta34) "kale" 8dea4d0549ac XEmacs Lucid (x86_64-apple-darwin16.7.0) X-SA-Exim-Connect-IP: X-SA-Exim-Mail-From: steve AT turnbull DOT sk DOT tsukuba DOT ac DOT jp X-SA-Exim-Scanned: No (on turnbull.sk.tsukuba.ac.jp); SAEximRunCond expanded to false Reply-To: djgpp AT delorie DOT com Errors-To: nobody AT delorie DOT com X-Mailing-List: djgpp AT delorie DOT com X-Unsubscribes-To: listserv AT delorie DOT com Precedence: bulk tl;dr RFCs classified as Proposed Standards are indeed standards for practical purposes, although there is no enforcement agency. An Internet Standard is just a relabeling of an existing RFC. The status of the cited RFC 1855, however, is Informational, not Proposed Standard. The long version includes historical rationale for why it works this way. Please reply to me if you want to discuss; Reply-To set (but this list may reset it; please check the addressees if you do reply). Mateusz Viste writes: > 1. RFC are no standards, otherwise they wouldn't be called "RFCs" > in the first place That's not a realistic description of actual practice. RFCs are technically classified as "Proposed Standards", "Best Current Practice", and "Informational". "Internet Standards" are simply redesignations of particular Proposed Standards as Internet Standards, without changing any text. Neither Proposed Standards nor actual Internet Standards are enforced, except by the acceptance of the community and refusal to communicate with nonconforming applications. In that sense you might consider both to be "non-standards". However, as formal descriptions of communication processes that most of us strive to conform to, Proposed Standards are standards indeed. It is acceptable to criticize an implementation that claims conformance to a Proposed Standard on the grounds that it does not implement all of the REQUIRED behaviors, as well as all of the RECOMMENDED behaviors where a good reason for the lack is not presented. It is acceptable, though often impractical (think Yahoo! mail or Gmail, or Reply-To- munging mailing lists like this one ;-), to refuse to communicate with a badly implemented or non-conforming application. The original intent was that Proposed Standards might be superseded by a better proposal of a similar protocol, which would be elevated to Internet Standard instead of the original. In practice it turns out that the "working code and rough consensus" process produces products robust enough, and takes long enough, that Proposed Standards attract near-universal conformance in most cases, and rarely are actually elevated to Internet Standard. > 2. The very same RFC you mention does say: > > "this Guide offers a minimum set of behaviors which organizations and > individuals may take and adapt for their own use." > ^^^ Indeed, RFC 1855 is also explicitly classified as "INFORMATIONAL" in the heading, which means the same thing. For completeness, "Best Current Practice" RFCs describe emerging consensus on - implementation techniques, - on local handling of data received or to be sent (Proposed Standards define behaviors needed for inter-system interoperability, and thus should not discuss purely local behavior), and - about the best configuration of optional features (and sometimes changed opinion about whether to implement recommended features by default). Hope this helps! Steve