COBOL Trends
in [Cobol]
|
From: Richard on 23 Oct 2009 15:00 On Oct 23, 5:44 pm, Paul <paul-nospamatall.rauler...(a)mac.com> wrote: > On 2009-10-22 13:23:05 -0500, Richard <rip...(a)Azonic.co.nz> said: > > > > > On Oct 22, 6:11 pm, Paul <paul-nospamatall.rauler...(a)mac.com> wrote: > >> On 2009-10-20 15:48:39 -0500, Ken <klsha...(a)att.net> said: > > >>> On Sep 17, 7:45 pm, Paul <paul-nospamatall.rauler...(a)mac.com> wrote: > >>>> On 2009-09-16 11:40:25 -0500, Michael Wojcik <mwoj...(a)newsguy.com> sai > > d: > > >>>> E-mail was certainly one of the first client/server applications that > > ent > >>> ered > >>>> in the general awareness of people. Decades ago. And believe me, > >>>> when it was developed and first started being used, it was striking, > >>>> innovative, > >>>> *and totally non obvious to most users*. I was there. :) > > >>> Silly me. I thought that E-mail originated as a "store-and-forward" > >>> application. That's why they called it "mail", right? Just like the > >>> USPS still does today with their paper letters - collect them, store > >>> them, sort them, and forward them. > > >>> Wasn't the first widespread "electronic transport" for E-mail the uucp > >>> (unix-to-unix copy program, I believe it is) protocol? Didn't uucp pre- > >>> date TCP/IP and Arpanet? > > >>> Store-and-forward seems to me to be much more closely aligned to > >>> Michael's peer-to-peer than client/server. Who was the client and who > >>> was the server on Usenet? > > >> Actually, the first e-mail was sent between two PDP-10 computers over > >> the very young > >> but very much existent ARPANET. Ray Tomlinson is usually accepted as > >> having sent > >> the first e-mail when he worked at BBN, in 1970 or 1971 I think. You > >> can google it > >> easily enough and probably get a ton of references. > > > Such as this one perhaps: > > > """Email is much older than ARPANet or the Internet. It was never > > invented; it evolved from very simple beginnings. > > > Early email was just a small advance on what we know these days as a > > file directory - it just put a message in another user's directory in > > a spot where they could see it when they logged in. Simple as that. > > Just like leaving a note on someone's desk. > > > Probably the first email system of this type was MAILBOX, used at > > Massachusetts Institute of Technology from 1965. Another early program > > to send messages on the same computer was called SNDMSG.""" > > > ARPANET was not a public system, even when they changed the protocols > > from NCP to TCP it only slowly turned into the internet for general > > use. > > > """public access and hobbyist networking systems grew in popularity, > > including unix-to-unix copy (UUCP) and FidoNet""" > > > So Michael was correct that the "first _WIDESPREAD_ "electronic > > transport" for E-mail" was uucp. > > Are you talking about e-mail? You know, the stuff defined in the dictionary > as: > > a system for sending messages from one individual to another via > telecommunications links between computers or terminals. http://en.wikipedia.org/wiki/E-mail """E-mail started in 1965 as a way for multiple users of a time- sharing mainframe computer to communicate.""" Yes, they do indeed seem to be talking about e-mail and "sending messages from one individual to another via telecommunications links between computers or terminals". > ARPANET was not open to general public, no, but I certainly had an > arpanet account. So did a heck of a lot of college kids, and a lot of > people in defense and aerospace. It got a heck of a lot of use. Bitnet > was about the only other comparable system, and it was far more > restricted. > > The ARPA plan for ARPANet was submitted and approved in mid 1968 - > about one year before we walked on the moon. > > UUCP protocols usually required a map to find where they were going, as > in "bang addresses" (which is why most of us from that time still call > the "!" character a bang...) Note that in those days, the "internet" > was small enough most people kept entire maps of the net on their host > computers. Bill Joy and DNS made all that irrelevant, and I think > Tomlinson is credited with coming up with the at address style. (I may > be wrong there.) > > UUCP is a protocol, layered over a communications network. It was > adapted for use with modems, which spread the popularity. But that > didn't happen until the late 1970's. The foldest UUCP map I can > remember would show wolfvax and ucbvax connected by at least 10 or 12 > connections. Most of them would have been UUCP over network, though the > long hauls would have been over dialups. There were certainly > connections to ARPANet, Berknet, and others on that map as well. > > As late as 1984 and 1985, I was pulling e-mail and newsgroups to a > small AT&T 3B2 computer in my home office, and allowing folks to dial > in and use it from the local area. I believe my primary connection was > to Rutgers at that time. > > All this to point out that e-mail was in pretty good use before UUCP > became the de-facto standard. It was not the the first widespread > system, though it did become the most used, at least for a period. > > > > >> But besides that, email has always been transmitted server <---> > >> server, and read > >> with a client. Not sure what your point is. Client programs usually > >> accounted for > >> more processing cycles than the server transmissions. I'm sure there > >> were exceptions. > > > 'Clients' such as 'more' or 'cat', yes those really chewed up the > > cycles. > > Clients more long the lines of mail, pine, or emacs. > > > > >> I hate to point out that the entire network is based on a store and > >> forward concept. > >> This router grabs a packet, decodes it, stores it in local memory, > >> figures out what to > >> do with it, and sends it on along. In fact, switches act that way > >> today, as they do a lot > >> of routing, especially between VLANS. The "store" time may be very > >> small, but it is there. > > >>> Remarkable that Michael is getting such resistance when the very Forum > >>> that we are using has a Usenet legacy; Usenet discussion groups > >>> likewise were originally store-and-forward. > > >>> For those old enuf or dumb enuf to care. :-) > > >>> And for those who bothered to notice I've been MIA... > > >>> Yeah, posters here are entirely missing the point on where the > >>> software cycles are! I've been watching satellite TV and working up > >>> my home theatre system. Anybody else spent $3000-$5000 on video > >>> elecronics lately? Just what _processors_ are in those components! > >>> Sheesh! Surely enough to rival the auto electronics. > > >>> Seems peer-to-peer to me - oh wait! Maybe the satellite is the server, > >>> and the DirecTV receiver is the client! Or maybe the BluRay DVD > >>> player is the server and the remote clicker is the client? Or is it X- > >>> Windows type of client/server "inversion" whereby the TV is the > >>> Display Server and the VHS player is the application client??? > > >>> I'm soooooooo confused! > > >> <grin> All depends upon your point of view of course. > > >> The video server on your digital TV is the client for the content > >> server from your blue ray player, and a client for your digital video > >> recorder, which is also the client for the cable broadcast network, > >> except when it is acting as a server to provide reports back to the > >> cable company, or an audio server to drive your amplifier, which is > >> definitely serving your speakers. > > >> Except when it is a client being controlled by your iPhone. > > >> Simple eh? > > >> -Paul > >
From: Paul on 25 Oct 2009 00:55
On 2009-10-23 14:00:37 -0500, Richard <riplin(a)Azonic.co.nz> said: > On Oct 23, 5:44�pm, Paul <paul-nospamatall.rauler...(a)mac.com> wrote: >> On 2009-10-22 13:23:05 -0500, Richard <rip...(a)Azonic.co.nz> said: [snip snip snip...] >> >> Are you talking about e-mail? �You know, the stuff defined in the dicti > onary >> as: >> >> a system for sending messages from one individual to another via >> telecommunications links between computers or terminals. > > http://en.wikipedia.org/wiki/E-mail > > """E-mail started in 1965 as a way for multiple users of a time- > sharing mainframe computer to communicate.""" > > Yes, they do indeed seem to be talking about e-mail and "sending > messages from one individual to another via telecommunications links > between computers or terminals". That isn't a bad article, but I am not sure when IUM was made a part of CTTS. CTTS led to MULTICS which led to UNIX, of course, so those concepts would have been passed along that route. ITS had some kind of inter user messaging as well, though I have never seen it. -Paul > > >> ARPANET was not open to general public, no, but I certainly had an >> arpanet account. So did a heck of a lot of college kids, and a lot of >> people in defense and aerospace. It got a heck of a lot of use. Bitnet >> was about the only other comparable system, and it was far more >> restricted. >> >> The ARPA plan for ARPANet was submitted and approved in mid 1968 - >> about one year before we walked on the moon. >> >> UUCP protocols usually required a map to find where they were going, as >> in "bang addresses" (which is why most of us from that time still call >> the "!" character a bang...) Note that in those days, the "internet" >> was small enough most people kept entire maps of the net on their host >> computers. �Bill Joy and DNS made all that irrelevant, and I think >> Tomlinson is credited with coming up with the at address style. (I may >> be wrong there.) >> >> UUCP is a protocol, layered over a communications network. It was >> adapted for use with modems, which spread the popularity. But that >> didn't happen until the late 1970's. The foldest UUCP map I can >> remember would show wolfvax and ucbvax connected by at least 10 or 12 >> connections. Most of them would have been UUCP over network, though the >> long hauls would have been over dialups. There were certainly >> connections to ARPANet, Berknet, and others on that map as well. >> >> As late as 1984 and 1985, I was pulling e-mail and newsgroups to a >> small AT&T 3B2 computer in my home office, and allowing folks to dial >> in and use it from the local area. I believe my primary connection was >> to Rutgers at that time. >> >> All this to point out that e-mail was in pretty good use before UUCP >> became the de-facto standard. It was not the the first widespread >> system, though it did become the most used, at least for a period. >> [ snip snip snip...] |