more on Internet still reshaping history ( with comment by the editor)

[Dave Farber <dave () farber net>]

Begin forwarded message:

> From: "David P. Reed" <dpreed () reed com>
> Date: December 7, 2009 4:33:42 PM EST
> To: dave () farber net
> Cc: ip <ip () v2 listbox com>
> Subject: Re: [IP] Internet still reshaping history ( with comment by  
> the editor)

> One more bit of history: the Farber DCS ring was used as inspiration  
> for the design of the MIT token ring that my (Saltzer was my  
> adviser) research group was working on.  The design of our token  
> ring was adopted by several vendors: IBM and by Proteon, and with a  
> little variation, by Apollo Computer Corp. Saltzer himself conceived  
> of the "star shaped" ring configuration that had "hubs" which  
> facilitated physical reconfiguration by allowing splicing out of any  
> individual portion of the ring with only microseconds of  
> interruption, and IBM adopted that concept.  Proteon was the first  
> company to develop a multiprotocol router (Noel Chiappa was  
> responsible for that).
>
> The hub concept was later adapted to Ethernet, creating "twisted  
> pair" Ethernet at PARC, and spun out to Synaptics, and though I  
> don't know for sure, I'm pretty sure that Saltzer's invention of the  
> hub idea was the primary inspiration.
>
> So there is a huge tree of innovations that sprung out of just the  
> DCS ring.
>
> On 12/07/2009 09:38 AM, Dave Farber wrote:
> > Just to clarify history a bit,the IBM Token Ring was directly  
> > derived from the ring used in the DCS distributed system 1972 ish  
> > time frame --the first operational "cloud" like system. Djf
> >
> >
> > Begin forwarded message:
> >
> > > From: Bob Metcalfe <bmetcalfe () polarisventures com>
> > > Date: December 6, 2009 8:59:48 PM EST
> > > To: anthony.citrano () gmail com, dave () farber net, ip () v2 listbox com, gep2 () terabites com
> > > Cc: Bob Metcalfe <bmetcalfe () polarisventures com>
> > > Subject: RE: [IP] Re: Internet still reshaping history ( with  
> > > comment by the editor)
> >
> > > Reshaping history?
> > >
> > > Gordon Peterson below tries "to clarify the records" on the  
> > > history of the PC, the spreadsheet, and the LAN.
> > >
> > > Of course his clarification depends on the meaning of words like  
> > > invent and first, and on the fact that he was at Datapoint back in  
> > > the day.
> > >
> > > Hey, Dan and Bob didn't "invent" the spreadsheet; they merely took  
> > > what accountants had been doing for centuries and wrote some  
> > > computer programs to speed it up, no?  No.
> > >
> > > I don't think many people dispute that Gordon's ARCnet was the  
> > > first commercially successful LAN (maybe HyperChannel from NSC?),  
> > > ahead of Ethernet, which was invented on May 22, 1973 (before  
> > > ARCnet) but didn't get standardized (IEEE 802.3) and  
> > > commercialized until 1981.  I've always wondered whether Gordon  
> > > read our Ethernet paper in the July 1976 issue of CACM before  
> > > developing ARCnet?
> > >
> > > After all that, it took Ethernet a good long time to kill the non- 
> > > standard ARCnet and the imperfectly standard IBM Token Ring (IEEE  
> > > 802.5 RIP).
> > >
> > > I half joke that by 1982 there were people BUYING Ethernet whom I  
> > > had not met personally; by 1985 there were people INVENTING  
> > > Ethernet whom I had not met personally.  Today's "Ethernet" is a  
> > > far cry from the Ethernet that Dave Boggs and I invented (or not),  
> > > and it ships 350+ million switch ports per year, not counting WiFi.
> > >
> > > Fine.  Plenty of credit to go around.  Datapoint and ARCnet were  
> > > early and cool in their day, surely.  Thank you, Gordon.
> > >
> > > But then Gordon writes, below, that I didn't invent Ethernet.  I  
> > > find this not clarifying but annoying, for obvious reasons.
> > >
> > > So, I went and dug up this ditty, to annoy Gordon back:
> > >
> > > -------------------------------------------------
> > >
> > > How Ethernet is Exactly like Alohanet, not
> > >
> > > By Bob Metcalfe
> > >
> > > A paragraph from "Retrospective" in my 1996 book, PACKET  
> > > COMMUNICATION, which is a reprint of my 1973 Harvard PhD  
> > > dissertation, still available on Amazon.com:
> > >
> > > ...
> > >
> > > Anyway, despite numerous mentions of Abramson's Alohanet work in  
> > > this dissertation and in almost all Ethernet materials since, I've  
> > > heard repeatedly over the years that, because it uses randomized  
> > > retransmissions, Ethernet is merely an Alohanet rip-off.  Well,  
> > > these critics may have a point.  Ethernet is just exactly like  
> > > Alohanet.  Except Ethernet does not have a central controller like  
> > > Alohanet's Menehune.  Except Ethernet has a single transmission  
> > > channel with two-address packets instead of Alohanet's two  
> > > channels with one-address packets.  Except Ethernet, unlike  
> > > Alohanet, has carrier sense to avoid most collisions and to carry  
> > > long packets efficiently.  Except Ethernet, unlike Alohanet, has  
> > > collision detection so as not to waste bandwidth.  Except Ethernet  
> > > has retransmission back-off control for channel stability, like  
> > > Alohanet didn't.  And except finally that Ethernet runs at  
> > > millions of bits per second over cable within buildings, while  
> > > Alohanet ran at thousands of bits per second over radio among the  
> > > Hawaii islands.  So, with these few exceptions, sure, Ethernet is  
> > > exactly like Alohanet.
> > >
> > > ...
> > >
> > > /Bob Metcalfe (http://twitter.com/BobMetcalfe)
> > >
> > > -----Original Message-----
> > > From: Anthony Citrano [mailto:anthony.citrano () gmail com]
> > > Sent: Sat 12/5/2009 10:29 PM
> > > To: Bob  Metcalfe
> > > Subject: Fwd: [IP] Re: Internet still reshaping history ( with  
> > > comment by the editor)
> > >
> > > You probably saw this already...
> > >
> > > ---------- Forwarded message ----------
> > > From: David Farber <dave () farber net>
> > > Date: Thu, Sep 10, 2009 at 9:36 AM
> > > Subject: [IP] Re:  Internet still reshaping history (  with  
> > > comment by
> > > the editor)
> > > To: ip <ip () v2 listbox com>
> > >
> > >
> > > Begin forwarded message:
> > >
> > > From: Gordon Peterson <gep2 () terabites com>
> > > Date: September 10, 2009 12:41:50 PM EDT
> > > To: dave () farber net
> > > Subject: Re: [IP] Re: Internet still reshaping history
> > >
> > > Perhaps this was your point, but just to clarify the record:
> > >
> > >  1.  The first spreadsheet program worthy of being called that  
> > > really
> > > was Visicalc, which was a revolutionary concept (although I have to
> > > confess I'm among those who really didn't realize its significance  
> > > at
> > > the time...!  My reaction was "it's easy to write the expressions  
> > > for
> > > each of those cells/variables.")  Even from the Microsoft stable,
> > > their Multiplan spreadsheet software preceded Excel.
> > >
> > >  2.  The first "personal computer" worthy of the term really was the
> > > Datapoint 2200.  That machine was the machine for which the Intel  
> > > 8008
> > > (the first one-chip general purpose 8-bit microprocessor) was  
> > > created,
> > > the architecture whose evolutionary descendants still power PCs to
> > > this day.  It was the first totally self-contained desktop computer
> > > system, looking like a piece of office equipment, designed for use  
> > > by
> > > one user, integrating a keyboard and CRT display (and magnetic  
> > > digital
> > > storage) in the package, with an operating system, utilities,
> > > higher-level business-oriented programming languages, general  
> > > editor,
> > > text formatting package, communications programs, and more.
> > >
> > > The Apple II was principally notable, at the time, by the fact  
> > > that it
> > > had a more finished packaging than most of the other
> > > hobbyist/home-oriented machines then available.  (Its packaging,
> > > though, was certainly no more finished than that of the Datapoint
> > > 2200... much less so, in fact).  It's undeniable that the Apple II  
> > > was
> > > less expensive than the Datapoint 2200, but then again the 2200  
> > > did a
> > > whole lot more than the Apple II did, too.
> > >
> > > But anyhow, Microsoft didn't invent the spreadsheet (even as the
> > > "architects") and Steve Jobs (and Apple) didn't invent the personal
> > > computer (even as "the architect").  Both of those companies were
> > > latecomers to the party, even as "architects".
> > >
> > > (By the same token, Ethernet wasn't the first LAN, and Bob Metcalfe
> > > didn't invent it either... Ethernet was not much more than a wired
> > > implementation of the University of Hawaii's "Project Aloha", which
> > > was an RF-wireless local area network... and the origins even made
> > > more clear by the bizarrely inappropriate name "Ethernet" for a  
> > > CABLED
> > > network, where Project Aloha, being wireless, transmitted through  
> > > "the
> > > ether"...!)
> > >  [editor's note:  its interesting historically and the DCS  token  
> > > ring
> > > was contemporary with the ethernet. The design of the token ring was
> > > read by IBM with minor deletions and became a commercial token ring.
> > > (For those who would debate that I have a note from a very senior
> > > IBMer  acknowledging that fact. I got rather annoyed with IBM for
> > > publishing their design as if it  was their idea and wrote to them
> > > asking them to be  "professional".
> > >
> > > Metcalfe and I outlined a proposed joint paper showing how in many
> > > ways ethernet and the token ring were very similar. For  number of
> > > reasons it never got completed.
> > >
> > > I never found that worthwhile to go around demanding priority on
> > > innovations I made. I was happy, maybe naïvely, to look at what it
> > > contributed to the field and more importantly what graduate students
> > > got their thesis based on the  working on the projects.]
> > >
> > >
> > > It is strange how some fictions about "who created what" get fixated
> > > in popular culture.
> > >
> > >
> > >
> > > -------------------------------------------
> > > Archives: https://www.listbox.com/member/archive/247/=now
> > > RSS Feed: https://www.listbox.com/member/archive/rss/247/
> > > Powered by Listbox: http://www.listbox.com
> > Archives        



-------------------------------------------
Archives: https://www.listbox.com/member/archive/247/=now
RSS Feed: https://www.listbox.com/member/archive/rss/247/
Powered by Listbox: http://www.listbox.com