Email inventor Ray Tomlinson RIP

[ppppenguin]

Amongst other things tomlinson instroduced the now ubiquitous @ symbol into common usage

http://www.bbc.co.uk/news/technology-35742210

It's worth looking at comment 117 in the Have Your Say. Written by none other than Vint Cerf, one of the true fathers of the internet. Or at least claiming to be Vint Cerf and making a comment that makes sense if he is.

117. Posted by vint cerf

on 12 minutes ago
Ray Tomlinson was a good friend and deserves credit as one of the first three software engineers to implement TCP (later TCP/IP). He was an Internet Pioneer in the best sense of the term. His first implementation forced us to revise the protocol to include a three-way handshake for duplicate detection. He was a wonderful and warm-hearted person and a joy to work with. I will miss him.

[Catkins]

Amongst other things tomlinson instroduced the now ubiquitous @ symbol into common usage

http://www.bbc.co.uk/news/technology-35742210

It's worth looking at comment 117 in the Have Your Say. Written by none other than Vint Cerf, one of the true fathers of the internet. Or at least claiming to be Vint Cerf and making a comment that makes sense if he is.

The comment definitely has the air of authenticity.  For instance I wonder how many people on HYS would know TCP/IP was originally just specified as TCP.  This is probably fairly arcane knowledge.  In case anyone doesn't know what this is all about, suffice to say originally TCP was specified as a compete protocol, it later became fashionable to think about network protocols as layers, each one doing just one thing, each additional layer adding functionality.  So the packet routing aspect became split into a separate layer (IP), with the end-to-end channel semantics being layered on top now called TCP.   The two layer combination being called TCP/IP.   

There is another lighter-weight protocol specified called UDP also layered on IP (or UDP/IP), which because it doesn't do packet loss detection and re-transmission was often used to do video/audio transmission over the internet.  At least I used it to do that back in ~1992.  I don't actually know if most video streaming now uses TCP or UDP (*).  I suspect TCP as many household routers now routinely block UDP.

(*) My last networking research grant finished in 1996.

Getting back on topic, one thing which is often forgotten about nowadays in the UK is that for a long time the internet (or more specifically the internet protocol) only existed in the US.  Not that we didn't have national networks in the UK/Europe, but we used a different incompatible system called the X.25 packet switched network (from 1980 - 1992).

We did have email, but it wasn't quite the email of today.  For a start it used a completely different protocol and it was a completely separate implementation.  The protocol used was called the grey book protocol (one of the so called coloured book protocols https://en.wikipedia.org/wiki/Coloured_Book_protocols, so named due to the colours of the book covers).  It took the concepts of the internet email system and the "@" symbol but it was subtly different.  The most conspicuous difference was addresses were the other way around.

Internet email addresses were of the form catkins@comp.xxx.ac.uk,

the equivalent grey book email address was catkins@UK.AC.XXX.COMP.

Sending email to the US was more of an art than a science, as you had to wrap the destination email address within the address of a gateway that would forward it onto the internet.  There was one machine in the UK connected to the internet at Imperial college London if I recall correctly.



 

[Nick]

PSI - that takes me back !

I wrote a PSI mail transport for VAX Mail back in about 1988 - at that time there were multiple networks - the main one that was a precursor to the full internet in the UK was JANET - the Joint Academic NETwork.

My addresses at the time were:

• PSImail: +234213300154::NICK  
  
• NICK@NCDLAB.ULCC.AC.UK
  
• Janet:  NICK@UK.AC.ULCC.NCDLAB
  
• NICK%NCDLAB.ULCC.AC.UK@UKACRL
  

ULCC (University of London Computer Center) were the forwarder, not Imperial.

Here's a Google-archived post by me on this from December 1989 : https://groups.google.com/forum/#!topic/...3EmoqJv_mo - the earliest post of mine they have is from the 27th September 1989....

All a long time ago !