▲ 169 points
back
93 comments
That March 1977 map always brings back a flood of memories to this old-timer.
Happy nights spent hacking in the Harvard graduate computer center next to the PDP-1/PDP-10 (Harv-1, Harv-10), getting calls on the IMP phone in the middle of the night from the BBN network operations asking me to reboot it manually as it had gotten wedged...
And, next to me, Bill Gates writing his first assembler/linker/simulator for the Altair 8080... (I tried talking him out of this microcomputer distraction -- we have the whole world of mainframes at our fingertips! -- without success.)
(Edit:) We also would play the game of telnet-till-you-die, going from machine to machine around the world (no passwords on guest accounts in the early days), until the connection died somewhere along the way.
Plus, once the hackers came along, Geoff Steckel (systems guy on the PDP-10) wrote a little logger to record all incoming guests keystrokes on an old teletype, so we could watch them attempting to hack the system.
On the subject of that last item, it is to my amusement that modern internet scanners are completely confused by a 1970s operating system. They record a "hit" when they find an open telnet port, but then get stuck because there is no recognizable prompt after the system banner message prints. They find a running FTP server but get confused that it does not use recognizable filesystem semantics. They get even more confused when it ignores passwords because the system has none. By all rules and tenets of security doctrine this system should be the internet equivalent to a smoking crater, instantly and utterly destroyed by advanced security threats beyond the imaginations of its creators.
PS: It is also amusing that an unmodified 1970s SMTP server can still deliver messages to gmail and receive responses back, given only the provision of a SPF record. Sadly, the coming mandatory requirement for DKIM will finally make this no longer a possibility.
PPS: It is much less amusing to attempt to read the gmail user's responses on a terminal.
Surprisingly, as I discovered earlier today, Gmail (still? newly?) supports sending plaintext-only messages!
Dear Sir, could you just, you know, continue writing? I just love these stories, would love to hear more!
Nah, it'd come out too much as "almost famous".
I did manage to avoid being Microsoft employee #12 or so (my buddy Bob Greenberg was #8, I think?, and encouraged me to come join them), and Adobe employee #8 (I knew Chuck Geschke from some earlier work done as an undergrad extending his PhD thesis to Harvard's extensible language ECL), due to various life circumstances. I guess God didn't want be to be a spoiled rotten billionaire.
Another near miss was co-consulting with Len Bosack at HP setting up Lisp Machine networking, and wondering how the heck the then-nascent Cisco was ever going to sell more than a few hundred routers (based on the same Sun-1 boards developed by Andy Bechtolstein at Standford that we used at Imagen, the first typeset-quality laser printers, a spinoff from Don Knuth's research at Stanford) to universities and government labs.
As Gates said, those of us who grew up with the ARPAnet and came to take it as a simple fact of life like electricity didn't see the Internet juggernaut coming.
Yes, please! I love these comments on HN, this is blog material.
If this man wrote a book I’d read it.
I worked for Peter Kirstein for many years - he always had wonderful stories to tell.
In the article Peter talks about the temporary import license for the original ARPAnet equipment. The delayed VAT and duty bill for this gear prevented anyone else taking over the UK internet in the early days because the bill would have then become due. But he didn't mention that eventually if the original ARPAnet equipment was ever scrapped, the bill would also become due.
When I was first at UCL in the mid 1980s until well into the 90s, all that equipment was stored disused in the mens toilets in the basement. Eventually Peter decided someone had to do something about it, but he couldn't afford the budget to ship all this gear back to the US. Peter always seemed to delight in finding loopholes, so he pulled some strings. Peter was always very well connected - UCL even ran the .int and nato.int domains for a long time. So, at some point someone from UCL drove a truck full of obsolete ARPAnet gear to some American Air Force base in East Anglia that was technically US territory. Someone from the US air force gave them a receipt, and the gear was officially exported. And there it was left, in the US Air Force garbage. Shame it didn't end up in a museum, but that would have required paying the VAT bill.
If only it could have disappeared into a vault until now... the bill could have been inflated away!
The most hilari-depressing part of the story was the funding politics and grantwriting headaches that have never changed:
– the NPL couldn't set up a British inter-network because of pressure from GPO;
– they couldn't connect to ARPA via Norway because of the Foreign Office;
– then, UCL couldn't get funding from SERC;
– then, UCL couldn't get funding from DTI because it didn't have industrial interest (although, to be fair, it was the department of "industry")...
...and then nearly a decade later government bodies were trying to take it over.
(It looks like the IMP/TIP was literally funded by petty-ish £££ that the NPL superintendent could get his hands on without further approval. To be fair, GPO did fund the link to Oslo.)
Peter once told me that in 1973, the only two organizations permitted to do telecommunications were the Post Office and the Ministry of Defense. So to legally connect UCL to the ARPAnet, he needed an exception clause. Somehow he got both the Post Office and the Ministry of Defense to sign off that they were not interested in computer-to-computer communications, in perpetuity, so that UCL could do so instead. He said he never tried to hold them to it later.
Spot on. Would be interesting to see what sort of hoops you'd have to jump through in an academic lab today to secure similar money.
That 5k GPB in 1973 is 77k in today's pound, or about 95k USD at current exchange rates.
Every time I have a meeting with government, especially regulatory agencies about getting something done, it isn't easy because these words have very different meanings, which can lead you into bad places quickly:
support - engineer means "compatible, works with" govt means "aiding a cause"
business rules - engineer means logic, govt means literal rules that have force of law
If you want to get results, you have to be really careful - if you say you are supporting something, the govt people may think you are aiding a cause they or whoever appointed them oppose. If you talk about rules, govt people assume a 2 year fight, expensive process, and lots of hearings - so it gets weird.
The trouble was that it was quite unclear to a researcher, even in one of the research council networking hubs, how to get access to the gateway, and it may have cost. I gave up trying before going to work in Oak Ridge for the summer (where I was taken aback by the primitive computing, at least "outside the fence"). For some time (mid-80s to early 90s? I don't remember) we were generally dependent on the infamous BITNET email gateway to communicate with the rest of the world from the well-developed UK network. It was "interesting" to deal with code in a Swedish 6-bit character set sent through the EBCDIC gateway to ISO 646-GB. (The Fortran Hollerith formats were added interest...)
Peter Kirstein died in January 2020, likely around the time when the internet finally reached Wales.
So I was just reading through a 1988 Swedish popular book on "data communications". Not a single word on Arpanet/etc. Many other network technologies and attempts at global networks described.
My point: "Internet" wasn't very well-known "even" in 1988 outside of well-connected places.
Book: Scandinavian PC Systems, Valentino Berti: "Introduktion till datakommunikation"
The inter-networking part of Internet was specialist knowledge restricted to those researchers actively working in the space. But countries had rich national networks back then, e.g. UK universities had a thing called JANET (joint academic network) that allowed, say, someone at the University of Kent to send files to someone at the University of Durham. The hosts were heterogeneous but the protocols were kinda sorta in place (there was a lot of X.25 leased lines and UUCP dialup, if I recall). Kent sticks in my memory because they could do commercial email in the old path!to!destination style if you knew the right guy to call. And Durham because they had this incredibly wacky mainframe OS, Michigan Terminal System, which I have never seen anywhere except there and at Newcastle (a town 40km up the road from Durham).
Any idea when JAnet connected to the Internet? When I first used it ~1994 I remember they had a single 2Mbps connection to the USA for the whole of JAnet.
What connection did the housing in the dorms at Kent have? I seem to remember serial ports in the rooms, but could have been ether?
It looks like Archive.org has it.
https://archive.org/details/michigan-terminal-system-distrib...
I was in the basement of UCL computer science (in the Pearson
building) in 1988. Our lab had a very special yellow (Don't ever touch
that!!) cable that ran across the ceiling between joists, then off
under UCH toward Telecom tower. Of course we hung bits of origami on
it with cotton. Apparently that was JANET. I never heard anyone say
"The Internet" back then, but we did have a coms lecture where
"inter-networking" was a thing. Nice to read some old names in that
piece.
Until relatively recently I worked with JANET (or Janet - lower case - as it is now) as part of Jisc, the UK's NREN. I also worked with the wider European org, GEANT, that runs the academic networks across Europe. We were (and still are ) very proud of Janet.
I was also in one of those basement labs in the Pearson building in 1988. Not exactly the nicest place to work, but some great equipment. I particularly remember that year coding a graphical application in NeWS (Sun's original network window system, long before it became X/NeWS) on a Sun Workstation there, which was an amazing piece of kit for its time. Also remember the DecStation 3100s we had down there that would periodically catch fire.
Memories. I had an unfiltered 100Mbit ethernet port, with public static IP, in my dorm room at Manchester Uni in 2000 thanks to JANET. It was amazing compared to the ISDN line at home (...which until then I had thought was the bees knees). It took almost another 15 years before I could get something faster at home than what I'd had at uni (!)
I think knowledge was more localised to certain universities or researchers.
Norway was the first country outside of USA to be connected to Arpanet in 1973 to share seismic data, but the universities did not gain general access until 1983.
I posted submitted a link to a story here: https://news.ycombinator.com/item?id=42655568
IIRC Finland was connected by then.
By '73? My understanding is the only other country outside the US to have a connection at that point was Norway, and only NORSAR and maybe one other location. NORSAR only because it was a critically important seismic array used by NATO to monitor Soviet missile tests.
As far as I can tell, Finland first got ARPANET connection via NORDunet in 1988 [1][2][3], though possibly indirectly a few years before as there was a connection to Sweden a few years earlier.
[1] https://siy.fi/history-of-the-finnish-internet/
[2] https://csc.fi/en/news/funets-anniversary-40-years-of-action...
[3] From [2]: "On Thursday December 1st, 1988, the first routing test was carried out, allowing IP packets to pass from Finland via Nordunet to the USA, in effect the National Science Foundation’s NSFNET and Arpanet networks. This connected Finland to the international Internet via the Funet network. The following message can be seen as the beginning of the Finnish Internet."
> The little black book of the internet
The article doesn't mention the Coloured Book protocols, but I'm pretty sure this phrasing isn't accidental: https://en.wikipedia.org/wiki/Coloured_Book_protocols
While it could be, "little black book" commonly refers to an address book, and has since long before the internet. While it's hard to established when it became a commonly understood term for that, -as it was used in a more general sense before that, and often simply referred to small notebooks -, it has seen use to refer to address books since at least the 19th century[1]. (EDIT: It's also mentioned in Aretha Franklin's "Runnin' Out of Fools[2] from 1964)
Also, now I feel very old. "Little black book" was an ubiquitous term when I grew up, and so using the term would just be a synonym to "address book" that nobody would think twice about.
[1] E.g. The House and Home: A Practical Book - Volume 1 (1896): "And there is a little black book with red lettering seen on every writing-table and carriage-cushion wherein puzzled mater-familias finds her bearings annually among her cherished acquaintances, many of whom the little black book alone keeps in her recollection!"
[2] Guess you got back (Guess you got) To my name (To my name) In your little black book
Red book was particularly interesting (modulo lack of security) long before "the Grid"; it worked between various computer centres.
> (2018)
I'm pretty sure the article is more recent than that... After some searching I found the Conversation's RSS feed for technology[1] which says it was 2025 after searching the page for "internet" and looking through the results
<id>tag:theconversation.com,2011:article/45404</id>
<published>2025-01-08T16:44:10Z</published>
<updated>2025-01-08T16:44:10Z</updated>
<link rel="alternate" type="text/html" href="https://theconversation.com/how-britain-got-its-first-internet-connection-by-the-late-pioneer-who-created-the-first-password-on-the-internet-45404"/>
<title>How Britain got its first internet connection – by the late pioneer who created the first password on the internet</title>
The preface says it was originally written "a few years before [the author] died", but never published (until now - having been (re)edited), so the (2018) makes sense.
> I'm pretty sure the article is more recent than that
You seem to be confusing the date it was written with the date it was published.
I believe there are multiple works where that interval is over 50 years. I would not be surprised if a gap of a century has been surpassed.
Not relevant to the article, but for the history of UK academic networking mentioned in various comments, Wikipedia's account looks about right, though I'm not sure it's up-to-date concerning the regions: https://en.wikipedia.org/wiki/JANET
I was using SRCnet in 1981, when Liverpool Physics had a dedicated link to Daresbury (national lab) whose speed I don't remember at that stage. Unfortunately the infamous PDP11 "terminal concentrators" for interactive use then were horribly unreliable. RJE to the cloud, where analyses ran, worked well.
The story has it all. Government stupidity and shortsightedness. Nonsensical self-harming bureaucracy. A plucky, persistent, woefully underfunded Brit who eventually succeeded in spite of the state.
Here's a debunking and a history of the myth the ARPANET/Internet was designed to survive nuclear war:
http://9ol.es/nuclear-myth.html
A bit more:
https://history.stackexchange.com/questions/15035/did-the-co...
> he was prevented from extending his project outside the lab by pressure from the British Post Office, which then held a monopoly on telecommunications.
The UK and telecoms don't make a happy pairing - we always seem to do the wrong thing:
https://www.techradar.com/news/world-of-tech/how-the-uk-lost...
Dr Cochrane knew that Britain's tired copper network was insufficient: "In 1974 it was patently obvious that copper wire was unsuitable for digital communication in any form, and it could not afford the capacity we needed for the future."
He was asked to do a report on the UK's future of digital communication and what was needed to move forward.
"In 1979 I presented my results," he tells us, "and the conclusion was to forget about copper and get into fibre. So BT started a massive effort - that spanned in six years - involving thousands of people to both digitise the network and to put fibre everywhere. The country had more fibre per capita than any other nation.
"In 1986, I managed to get fibre to the home cheaper than copper and we started a programme where we built factories for manufacturing the system. By 1990, we had two factories, one in Ipswich and one in Birmingham, where were manufacturing components for systems to roll out to the local loop".
BT let me off a £2,000 bill 'cus my dialer.exe got hijacked and I was calling a Seychelles premium line.
I'd turned off the sound on my 56k? modem and didn't spot the extremely long beeps and boops.
That's what got me started on basic security...zone alarm (pre-shitty-ness), proxomitron, hosts files (and other blocking lists/ banner blockers), autoexec.bat to delete histories, being caught out by dark patterns etc.
Fun times....'cus the internet was fun at the time, too. Nowadays, despite a trove of bookmarks, I visit very, very, very few new-to-me sites.
I really hope cable management has improved at LLNL since the 70s when that picture was taken.
Time-sharing is interesting. The same kind of thing is happening now with AI.
let's thank him for making some absolutely god-mode architectural calls (protocol layering, manufacturer agnosticism, etc.)
I had the great fortune to work at UCL in the early 1980s, just as SATNET came to an end as a project. I worked on what became ISODE, and related stuff as well as doing operations bits and pieces. I was far too junior to figure in Peter's planning but I will say, he was a very interesting HoD: he didn't suffer fools gladly (and I am one) but at the same time could be quite forgiving, if you were at least entertaining about your foolishness.
His most preferred model of funding work was to have 9/10ths of it done or a sure-fure thing, and then use the funding to work on the next idea, so he was guaranteed to have goods to deliver at the end of the project. He didn't always carry it off but when it worked it was superlative. I was on at least one OSI (protocol) project with 6 partners across industry and research in Europe, and the work was very unequal. That said, very fine dinners. I have fond memories of INRIA canteen food having wine and fresh fruit. UCL had baked beans and pies, the staff club had the same baked beans and pies but you could eat them under a superb Stanley Spenser oil painting of the resurrection.
The first Cisco Router turned up one day, it had annoyingly noisy fans which blew air a useless direction compared to the rest of the racks. As Mark Handley has pointed out below the basement was full of trash: a fantastic Prism-wedge shaped digital copying stand used with the British Library to photograph rare works, and an unbelievably expensive CCD digital camera attached gathering dust, a BBN Butterfly (it was a heap of crap frankly) running pre-BGP routing, a BLIT terminal and depraz mouse, the first Dec and Sun workstations.
We ran a project with the slade school of art doing digital arts design with them, lovely people. I still have some of the reject works.
Peter liked inviting people to come and be at UCL. During my time Bob Braden (SATNET) was there, and Mike Lesk (UUCP) -and they were also very nice and approachable. Tiny tea-room, I caused a kurfuffle washing out the most disgustingly stained coffee mug there, the owner of which was bulding the patina both to see how thick it got, and to discourage others from using his cup.
Peter had one standing rule you did NOT break: If there was an inter-departmental meeting with the people from ULCC (at that time down at Lambs Conduit St) he expected you to "vote against" any proposal they brought to the table without question: Departmental politics ran deep.
Peter was a committed skier, he was hardly young when I got there and he was still avidly visiting the alps as often as possible.
UCL ran the gateway from JANET (X.25 based non internet) to the ARPANet which required you to login to a unix jump host and use kermit to connect across, and lodge FTP requests using JANET "grey book" FTP protocol to talk to the FTP client through a conversion script. The gateway got hacked every now and then. Some of us were a bit unkind and said it used Kermit because Peter, who was somewhat small, frog-like and had thick glasses he peered through, appreciated the joke that he was Kermit personified.
I've put 2018 as an approximation above because the intro says "a few years before he died" in 2020. If anybody can figure out the actual year, we can change it.
Edit: in HN titles, if you see a year at the end in parens, that indicates the year that the article originated. If you see a year that's not at the end in parens, that's part of the article title, meaning it's probably about something that happened that year. That's the convention anyhow.
"The impact of the internet on our way of life in its first 40 years has been immeasurable. It has expanded and developed in a way none of us envisaged in 1975", suggesting 2015?
It says it was newer published (before now), so would not 2025 be correct?
[stub for offtopicness]
Sorry if I'm unclear on what 2018 is supposed to mean (understand it's an approximation). But internet service is much older than that. According to Wikipedia,
https://en.wikipedia.org/wiki/Internet_in_the_United_Kingdom
Pipex was established in 1990 and began providing dial-up Internet access in March 1992, the UK's first commercial Internet service provider (ISP).
That's about the same time it became available in the U.S. I got home internet service (dial-up) in the mid-1990's.
I’m happy that you Brits finally got the Internet in 2018. It took a while, but I hope it was worth the wait.