Does that mean that the first three characters ever successfully sent were ‘lol’?

Seems fitting.

Wow! That’s amazing if true. The question is was “login” entirely resent again or if the last three characters of “login” were sent.

If it’s the former, it’s unbelievable.

The first message „lo“ was sent at 10:30 p.m, on October 29, 1969. The transmitting SDS Sigma 7 host computer crashed before transmitting the final „g“ needed for the receiving computer at SRI to add the final „in“ for „login“. It took Leonard Kleinrock and UCLA student programmer Charley Kline about an hour to recover from that before the computer effected a full "login" sending the three letters „log“ again.

The 50th anniversary of this day passed barely a month back. Pretty cool.

Here is a more detailed NPR Segment on this story with pictures: https://www.npr.org/templates/story/story.php?storyId=114280...

Interesting.

If message accumulation on sending side waits till EOL is pressed - or some other indication of end of message is entered - then network wouldn't actually see "lo" - they wouldn't be sent before the crash.

If input would be sent character-by-character, then first "l" would be sent, then "o", then system would crash. In this case the first message would be "l".

This wouldn't be TCP connection - that would require some handshakes beforehand, so more messages back and forth even before "l" is sent.

There were protocols before TCP; simplest way is to assume it's direct IP. So, first point-to-point, routed internetworked message - first IP packed reached destination across networks - seemed to contain payload "l".

What's wrong here?

Werner Herzog made an entire movie focused on exploring the connected world, and the earlier part of the movie speaks pretty extensively about the first message sent across the internet. Incidentally, the movie is titled, "Lo and behold", as a play on the first message being, "Lo".

https://www.imdb.com/title/tt5275828/