Article 6274 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!newspeer.monmouth.com!uunet!ffx.uu.net!ffx2nh5!not-for-mail From: Tim Shoppa Newsgroups: comp.protocols.tcp-ip.domains,alt.folklore.computers,alt.sys.pdp10 Subject: Re: reasoning for naming restrictions Date: Sat, 22 Jan 2000 21:55:21 -0400 Organization: Trailing Edge Technology Lines: 71 Message-ID: <388A2749.46262B49@trailing-edge.com> References: <388A00D9.FC6DAEDD@ehsco.com> <86ddv6$ktn$1@ns1.vrx.net> <388A46B9.DBF3A357@ehsco.com> <86di3c$mlg$1@ns1.vrx.net> <388A506D.354CEDA6@ehsco.com> NNTP-Posting-Host: timaxp.trailing-edge.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: ffx2nh5.news.uu.net 948596134 28996 63.73.218.130 (23 Jan 2000 02:55:34 GMT) X-Complaints-To: news@ffx2nh5.news.uu.net NNTP-Posting-Date: 23 Jan 2000 02:55:34 GMT X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) Xref: news3.best.com comp.protocols.tcp-ip.domains:23150 alt.folklore.computers:149360 alt.sys.pdp10:6274 Eric A. Hall wrote: > > > It's no accident that domain name syntax is the same > > as C variable syntax. > > That's believable. Any evidence? It's a bit misleading to offer C as the language that drove the standards used for the Internet. Many of these standards have been inherited (often with great revision) from the early days (early 1970's) of ARPAnet, and long predate Unix or C. Think of TENEX and Multics and Fortran and COBOL, not of C and Unix. C and Unix were complete non-players in setting these standards. The early recognition of the need to hook widely different computers into some grand naming scheme - had to take into account many factors: 1. Machines used different character sets. Yes, ASCII was one of those character sets, but EBCDIC, truncated versions of ASCII, and other "proprietary" character sets were in wide use. If you wanted a name to be universal, it had damn well better be typable on all the machines in the network. And it's just plain wrong to think of today's interpretation of ASCII, because in the late 60's and early 70's many of the ASCII codes were still being revised by the committees. 2. Even if a user was accessing a machine that was capable of upper and lower case, they may be accessing through a terminal (like a Model 33) without lower case characters. So case sensitivity is a no-no. 3. Between character sets, you can count on letters, numbers, and some *very common* punctuation being available. The only punctuation codes that were widely available on all machines - when you look at what's in common between commercial and scientific character codes - you're limited to "-", "/", ".", "$", and ",". 4. The "variable name must begin with letter, not a number" scheme was in use long before the 1970's. Early Fortran (and I'm talking early 1960's) did this, and it wouldn't surprise me if there are earlier languages and/or assemblers that did it as well. So why was "-" allowed, but not those other 4 characters? I dunno. Many technical standards were chosen not for technical reasons, but because they were convenient or just because *something* had to be chosen. Why are the AC electrical systems around the world dominately 50Hz and 60Hz? You can go into long explanations about gearing ratios and subdividing circles and transformer weight and motor cooling, but the fact is that some number had to be chosen as a standard, and those are what was agreed on, even though they certainly weren't the only choices available, and possibly not even good choices for some factors. And just like many AC power frequencies haven fallen into disuse (it wasn't that long ago that 25Hz power was used in many places), many internetworking naming conventions have fallen out as well. It was only a few years ago that everybody talked about bang paths and knew that in the UK that the endianess of the components of a name was reversed, but you're hard pressed to find anyone remembering those days now. Heck, ten years ago all the major archive sites on the Internet were PDP-10's, and today you're hard pressed to find a FTP client that knows about TENEX mode. Things change, and it wouldn't surprise me if the current host naming scheme was junked and forgotten just as quickly as those other conventions. -- Tim Shoppa Email: shoppa@trailing-edge.com Trailing Edge Technology WWW: http://www.trailing-edge.com/ 7328 Bradley Blvd Voice: 301-767-5917 Bethesda, MD, USA 20817 Fax: 301-767-5927 Article 6277 of alt.sys.pdp10: Path: news3.best.com!news2.best.com!news.maxwell.syr.edu!nntp.abs.net!uunet!ffx.uu.net!ffx2nh5!not-for-mail From: Tim Shoppa Newsgroups: comp.protocols.tcp-ip.domains,alt.folklore.computers,alt.sys.pdp10 Subject: Re: reasoning for naming restrictions Date: Sun, 23 Jan 2000 08:34:52 -0400 Organization: Trailing Edge Technology Lines: 66 Message-ID: <388ABD2C.501D3305@trailing-edge.com> References: <388A00D9.FC6DAEDD@ehsco.com> <86di3c$mlg$1@ns1.vrx.net> <388A506D.354CEDA6@ehsco.com> <388A2749.46262B49@trailing-edge.com> <86e0vr$is$1@ns1.vrx.net> NNTP-Posting-Host: timaxp.trailing-edge.com Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit X-Trace: ffx2nh5.news.uu.net 948634494 26457 63.73.218.130 (23 Jan 2000 13:34:54 GMT) X-Complaints-To: news@ffx2nh5.news.uu.net NNTP-Posting-Date: 23 Jan 2000 13:34:54 GMT X-Mailer: Mozilla 3.03Gold (X11; I; OpenVMS V7.0 DEC 3000 Model 300L) Xref: news3.best.com comp.protocols.tcp-ip.domains:23159 alt.folklore.computers:149384 alt.sys.pdp10:6277 Richard J. Sexton wrote: > > In article <388A2749.46262B49@trailing-edge.com>, > Tim Shoppa wrote: > >Eric A. Hall wrote: > >> > >> > It's no accident that domain name syntax is the same > >> > as C variable syntax. > >> > >> That's believable. Any evidence? > > > >It's a bit misleading to offer C as the language that drove the > >standards used for the Internet. Many of these standards have > >been inherited (often with great revision) from the early days (early > >1970's) of ARPAnet, and long predate Unix or C. Think of TENEX and > >Multics and Fortran and COBOL, not of C and Unix. C and Unix were > >complete non-players in setting these standards. > > DNS came out in the mid 80's the original poster asked about > domain names, not (arpa) host names. Oh yeah, right, the character set for DNS host names has nothing at all to do with ARPANET names, that's why the early DNS related RFC's contained statements like The labels must follow the rules for ARPANET host names. (Above quote appears in both RFC 1035 and RFC 883). Come on, Richard, if you want to put up a fight at least put one up that isn't easily disproven by looking at the DNS RFC's. In any event, in the mid 80's all the character-set compatibility issues I specifically mention were still relevant. The Internet was still dominated by TENEX, ITS, and other non-Unix and non-C machines. Just look at the HINFO fields from any mid-80's host database and you'll see few (if any) Unix machines. For instance, of the "standard" HINFO names from RFC 883, only two were capable of running a Unix at the time: PDP-11/70 C/30 C/70 VAX-11/780 H-316 H-516 DEC-2060 DEC-1090T ALTO IBM-PC IBM-PC/XT PERQ IBM-360/67 IBM-370/145 and indeed in the OS field these are the suggested choices: ASP AUGUST BKY CCP DOS/360 ELF EPOS EXEC-8 GCOS GPOS ITS INTERCOM KRONOS MCP MOS MPX-RT MULTICS MVT NOS NOS/BE OS/MVS OS/MVT RIG RSX11 RSX11M RT11 SCOPE SIGNAL SINTRAN TENEX TOPS10 TOPS20 TSS UNIX VM/370 VM/CMS VMS WAITS Unix and C simply weren't major players at the time the standards were set. Anyone claiming otherwise is itching for a fight. -- Tim Shoppa Email: shoppa@trailing-edge.com Trailing Edge Technology WWW: http://www.trailing-edge.com/ 7328 Bradley Blvd Voice: 301-767-5917 Bethesda, MD, USA 20817 Fax: 301-767-5927