Home page logo

interesting-people logo Interesting People mailing list archives

Re: NSF and the Birth of the Internet
From: David Farber <dave () farber net>
Date: Tue, 19 Aug 2008 14:41:07 -0400

Begin forwarded message:

From: "Michael O'Dell" <mo () ccr org>
Date: August 19, 2008 1:27:36 PM EDT
To: dave () farber net
Subject: Re: [IP] Re:      NSF and the Birth of the Internet

crucial to the explosion of the Web was the
revolution in PC graphics cards.

when Gopher first hit, I thought it was gonna be the
Internet's answer to MINITEL and be explosively successful.
I could play with HTTP but only because I had a Sun workstation
on my desk and some expensive color displays in the terminal
room at Bellcore.  The only thing that worked over dialup
modems on my Toshiba T1200 was Gopher because all it needed
was an ascii terminal emulator running on a Black&White
character-based display.

that was the technology which was already pervasive out there -
the installed base - and that was in the best position to exploit Gopher.

However, not long after that, this VGA graphics thing appeared
and while for the first year buying one caused serious pain,
they started dropping in price, the VESA bus slot appeared on
cheap motherboards, and the world changed, or it was about to.

packaged TCP/IP products with graphics first appeared for the
PC under DOS and enjoyed significant popularity for a while.

but then the seismic shift of Windows 95 hit. Bit-mapped graphics
became standard equipment, not just an option on Cadillacs.

at that point, who cares about Gopher?

well, to be honest, people at the end of waxed string phone
lines still cared, but GIF progressive rendering and lots
of other creative hacks made it all tolerable and the
obvious functional superiority of HTTP put the evolution
and expansion of The Web into warp drive.

Innovation on this scale seldom hinges on a single issue,
or even a few issues.  It's usually the fortuitous interaction
of a bunch of complex curves, each wiggled by its own economics,
producing little pools and regions of viability. Things live in
those pools, and things die when those pools dry up (remember
Diskless Workstations?). some species, though, manage to
crawl into the next pool that forms and get to play
The Game of Evolution all over again.

I've described The Internet (big I) as a huge petry dish.
the ISPs keep it pleasantly warm, pump in nutrients, and
take money from all comers for the ability to sit in the
petry dish and see what happens.

the ISPs don't particularly care which ones
live and which ones die - or at least they shouldn't.
Nature sorts that out with well-known techniques.

The ISPs get paid by each customer for his or her seat in the dish,
and as long as most of the customers don't die, they'll do just fine.
If the dish gets crowded, they make more dish (or should),
and if they aren't charging enough to make more dish,
*they have screwed up their economics!*

That's my definition of "network neutrality" -
"Life in the Internet Petry Dish"


David Farber wrote:
Begin forwarded message:
From: Dave Crocker <dhc2 () dcrocker net>
Date: August 19, 2008 10:43:12 AM EDT
To: dave () farber net
Cc: ip <ip () v2 listbox com>
Subject: Re: [IP] Re:     NSF and the Birth of the Internet
Reply-To: dcrocker () bbiw net
> From: Miles Fidelman <mfidelman () meetinghouse net>
????: First LAN connected to the ARPANET
Probably PARC.  Probably 1974-ish.
????: CSnet linked to the ARPANET
Since we already had email gatewaying running for the Army Materiel Command, at UDel, it was a small matter to start adding CSNet sites, as soon as NSF (and Arpa) authorized it. Also: My article about Rand's email work cites the role I believe CSNet played in the creation of NSFNet, and the role NSFNet played in the technical and operational style of the modern Internet.
> From: "David P. Reed" <dpreed () reed com>
>     But the idea
> that *Ray Tomlinson* gets the credit for electronic messaging, when
> Englebart did what he did is insane.
My understanding is that single-machine email was invented with the first time-sharing system at MIT in the 60s. I lived on Engelbart's NLS system, and indeed it was fabulous. Mouse, structured text, powerful editing, dynamic linking. But it was single-machine.
Ray invented "networked" email.
Nothing provided an early demonstration of the massive reality in the concept of network effects as did scaling email from one machine to a wide area network service. That's why Ray received the IEEE Internet award in 2004.
> From: Jim Thompson <jim () netgate com>
>> Also the word "gopher" appears nowhere in the timeline, although for a
>> couple of years it WAS the Internet...
> No, it wasn't.
While the question of how users perceive(d) the Internet can always get bogged down in the definition of users, metrics, or period of time, the role of gopher was fundamental. The Arpanet/Internet lived with "anonymous" FTP as its public publishing method for 15 years. As the Internet approached mass- market scale, a better mechanism was needed and various folk experimented with different approaches. For nearly 5 years, gopher dominated. In 1990, I was giving a half-day Internet introduction to some phone company folk, as part of a week-long seminar at CMU. I included a gopher demonstration, letting the class make selection choices, as we navigated from a page that divided the world into regions, until we finally got down to a page for the Wellington New Zealand Town Council and found a pointer to the Town Council minutes. At that moment, I finally understood how big an impact the Internet would have. A non-technical group was publishing mundane, non- technical information for non-technical use. If they were doing that then, everyone else would too. And by "everyone" I realized it would be literially everyone in the world. As the Web started to gain traction later, there was some question which of the two technologies would win. Gopher was far easier to publish, since it took any ol' text file, whereas the Web required specially-created files. On the other hand, the Web was multi-media and it let you give useful information to users with every key- click. With gopher, useful information was only at the leaves, and not the intermediate navigation nodes. No embedded links.

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

  By Date           By Thread  

Current thread:
[ Nmap | Sec Tools | Mailing Lists | Site News | About/Contact | Advertising | Privacy ]