Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

The origin of “log in” (2011)

designcult.org

Post reply on HN
Print view View original

11–20 of 64 posts

Re: The origin of “log in” (2011)

#11

Post by m12k »

jeffbr13 wrote: post #5

Like the author, I frequently get hung up on which of "log in", "sign in", or any of their variants is 'more correct' every time I put together software with users. "Authenticate" might be the most correct verb, although it's not very user friendly?

Personally I prefer "log in" to "sign in" because it makes it stand out more from "sign up", thus hopefully leading to fewer users accidentally selecting one when they wanted the other. And regarding the variations, I've always been under the impression that "log in" is the act of logging in, while "login" is either the situation/event ("a login happened") or an alternative term for username.

Re: The origin of “log in” (2011)

#13

Post by cynwoody »

It's interesting that the 1964 CTSS manual† linked from the article (OCR'ed by Google††) uses "login", one word, no blank. See page 6 (page 14 of the PDF).

I knew about ships' logs, but I didn't know that the word "log" originated from the use of a wooden log tossed overboard attached to a knotted rope, so that the observed speed could be entered into a book that eventually came to be known as a ship's log.

Also interesting in that manual is that each "segment" of a command (what we now call a token) was six characters, blank-padded. The 7094 had 36-bit words and 6-bit bytes.

†http://bitsavers.trailing-edge.com/pdf/mit/ctss/CTSS_Program...

††http://webcache.googleusercontent.com/search?q=cache:SapGOxk...

Re: The origin of “log in” (2011)

#14

Post by cynwoody »

jgrahamc wrote: post #6

My father used to use a system (IBM?) that required him to 'track on' and 'track off'. Anyone come across this?

In the sixties, the IBM time-sharing system CP-67† would physically punch an "accounting card" on one of its card punches whenever a user logged off or did some other billable event, such as printing a listing.

CP-67 ran on the 360/67, which was a 360/65 with special DAT hardware that allowed it to provide each user with their own virtual machine with its own address space.

†http://www.multicians.org/thvv/360-67.html

Re: The origin of “log in” (2011)

#17

Post by bananas »

jgrahamc wrote: post #6

My father used to use a system (IBM?) that required him to 'track on' and 'track off'. Anyone come across this?

Yes! I had to extract some data (in 1997!) off a DOS payroll package written in about 1985 that ran on an IBM 286. It was churning away quite happily. It seemed a bit like a 3270 terminal style application so I assume it was inspired by something distant and IBM. That had track on and track off!

On a side note, it was pretty easy to get the data off and reverse engineer the file format using visual basic at the time.

Was quite a fun project :)

Re: The origin of “log in” (2011)

#18

Post by protomyth »

jgrahamc wrote: post #6

My father used to use a system (IBM?) that required him to 'track on' and 'track off'. Anyone come across this?

Around here, it was "wand in" and "wand out" because they used a light pen to read a barcode on their badge to access the AS/400. My Dad still says "wand in" when he logs into his Mac.

Re: The origin of “log in” (2011)

#19

Post by mseebach »

cynwoody wrote: post #13

It's interesting that the 1964 CTSS manual† linked from the article (OCR'ed by Google††) uses "login", one word, no blank. See page 6 (page 14 of the PDF). I knew about ships' logs, but I didn't know that the word "log" originated from the use of a wooden log tossed overboard attached to a knotted rope, so that the observed speed could be entered into a book that eventually came to be known as a ship's log. Also inte…

Isn't that in reference to a command literally named "login", presumably due to the lack of a good convention for separating words without the space character?

Like "useradd" on contemporary Linux.

Re: The origin of “log in” (2011)

#20

Post by jeffbr13 »

m12k wrote: post #11
jeffbr13 wrote: post #5

Like the author, I frequently get hung up on which of "log in", "sign in", or any of their variants is 'more correct' every time I put together software with users. "Authenticate" might be the most correct verb, although it's not very user friendly?

Personally I prefer "log in" to "sign in" because it makes it stand out more from "sign up", thus hopefully leading to fewer users accidentally selecting one when they wanted the other. And regarding the variations, I've always been under the impression that "log in" is the act of logging in, while "login" is either the situation/event ("a login happened") or an alternative term for username.

You're right that "log in" is more distinct from "sign up" (and works because "logging up" doesn't make much sense), and I definitely agree with you that 'login' is a noun, if anything!

Both "log in" and "sign in" are anachronisms though - they're justified when you're actually signing-in-and-out of a computer system or building to log your visit, but the metaphor doesn't technically make as much sense when you're talking about authenticating the user on a website.

The evolution of the terms does let you nicely abstract away the actual mechanism of authentication for a user, which I suppose is the point.

Post reply on HN