[ih] origins of the term "hyperlink"

Sun Apr 12 11:20:04 PDT 2020

I got curious and started with a google books ngram search, then started
combing through individual records provided with their context. I made sure
to filter for noise -- there are a lot of books indexed that are 1990+
editions of earlier books that contain the term "hyperlink" in the
additional notes.

This DTIC report on on a hypertext project from 1987 does NOT use the term
"hyperlink", only "link":
https://archive.org/details/DTIC_ADA188179/page/n3/mode/2up/search/%22knowledge+garden%22?q=%22knowledge+garden%22

The earliest *published* mentions I can find are from 1988.

"PC AI" trade publication mentions hyperlinks and linked documents.
https://books.google.com/books?id=dKYdAQAAIAAJ&q=%22hyperlink%22&dq=%22hyperlink%22&hl=en&newbks=1&newbks_redir=0&sa=X&ved=2ahUKEwib6Omgt-PoAhXUFjQIHSPKDucQ6AEwAnoECAIQAg

PC Magazine in December 1988 mentions a linked document product suite
called PC-Hypertext from a company called MaxThink, which had an entire
program called HyperLink:
https://archive.org/details/PC-Mag-1988-12-13/page/n151/mode/2up/search/maxthink

THE FINAL LINK HyperLink is not so
> much a single product as it is a combina-
> tion of related utilities. This $89 module
> has programs that let you move MaxThink
> and Houdini networks into hypertext net-
> works. Since the PC-Hypertext system is
> designed to connect separate files, Hyper-
> Link has a utility that automatically divides
> a large text file into many smaller ones.
> Another program creates cross-reference
> lists for each file and then combines them
> into one long list.
>
> HyperLink also includes an ENCODE
> program, which compiles text files into a
> hypertext system that can be used with PC-
> Hypertext. All jumps ate established by
> placing a filename between the left and
> right angle brackets. The files used by EN-
> CODE can be created with any editor ca-
> pable of writing ASCII files. HyperLink
> even includes a utility designed to make a
> mini-expert system out of a hypertext net-
> work.
>
>
Possible asides and dead ends:

There is a concept in graph theory called a "hypergraph" and I found a 1977
paper that uses the term "hyperlink" in a discussion of a data structure
for a database system built around the hypergraph. I'm unsure if there is a
direct connection with the hypertext terminology but since it's a computer
science paper I thought I'd include it.
https://link.springer.com/chapter/10.1007%2F978-3-642-48908-2_3

HyperLink magazine was a publication dedicated to HyperCard development,
which is far as I can tell began publication in 1987 around the same time
as HyperCard was released. Whether this was an intentional play on an
already-established concept of a hyperlink or if it merely was for "linking
up" HyperCard developers, I can't say.

I'm also finding some commercial product naming, like "hyperlink" referring
to a special kind of networking interface here:
https://books.google.com/books?id=Cc4EAQAAIAAJ&q=%22hyperlink%22&dq=%22hyperlink%22&hl=en&newbks=1&newbks_redir=0&sa=X&ved=2ahUKEwib6Omgt-PoAhXUFjQIHSPKDucQ6AEwBnoECAcQAg

My conclusion:

It seems that in 1987-1988, "hyper" was a popular prefix to attach to
anything to make it seem more "techie", like "cyber-" was for some years in
the 1990s. There was a general flowering of things called hyper-[noun] in
those two years. This of course matches up roughly with the invention of
the WWW in 1989. The idea that "hyperlink" was coined by Ben Shneidermanin
1988 seems wholly plausible to me! At least I can find nothing to
contradict the claim. My guess is that multiple people independently coined
"hyperlink" in 1988, including whoever was working at MaxThink on their
HyperLink product in 1988.

-Darius

On Sun, Apr 12, 2020 at 9:54 AM Steve Crocker via Internet-history <
internet-history at elists.isoc.org> wrote:

> Jeff and Vint,
>
> Thanks for this very informative dialog.  One small question about a date:
>
> > Bush published the idea in 1945. Engelbart named the idea "links" about
> > 1945 but did not tell anyone.
>
> Was it really 1945 when Engelbart names the idea “links”?  I’m guessing
> this was an unintended typo.
>
> Steve
>
> Sent from my iPhone
>
> > On Apr 12, 2020, at 12:45 PM, vinton cerf via Internet-history <
> internet-history at elists.isoc.org> wrote:
> >
> > I asked Jeff Rulifson if he could help us figure out when "hyperlink"
> > entered into usage. Jeff was the principal programmer for Douglas
> > Engelbart's NLS editing system in which such links were introduced. Ted
> > Nelson coined terms like "hypertext" in his Xanadu concept - the two were
> > contemporary in the 1960s.
> >
> > Here is Jeff's response:
> >
> > Vannevar Bush, (Atlantic Monthly, July 1945, p 107) introduces the idea
> of
> > "associative indexing, the basic idea of which is a provision whereby any
> > item may be caused at will to select immediately and automatically
> > another." Bush goes on to discuss the idea of a named trail of such
> > associations. Engelbart's contemporaneous handwritten marginal notes in
> his
> > copy of the magazine call such a trail "links". The Computer History
> Museum
> > has a photocopy of Doug's personal copy with his handwritten marginal
> > notes. (I have a TIFF of their copy.)
> >
> > By the time of the 1968 demo, I had fully implemented links. I used a
> very
> > general implementation. A link had two parts: the name of a document and
> a
> > search command. The named document could be the current document or any
> > other document available for searching. The search command was a regular
> > expression. Users could make their links very simple or extremely
> > sophisticated. As far as I know, this was the first general
> implementation.
> >
> > In 1965, Ted Nelson had coined the term hypertext but never had an
> > implementation. I believe he used the term links but I cannot find the
> > paper for verification.
> >
> > By 1967, not knowing about Nelson or Engelbart, Andy van Dam had built a
> > system called HES. Andy's system had the notion of links and they were
> > called "links". (See
> > http://www.digitalhumanities.org/dhq/vol/4/1/000081/000081.html)
> >
> > Bush published the idea in 1945. Engelbart named the idea "links" about
> > 1945 but did not tell anyone. Andy used links in a line editor in 1967
> and
> > I implemented a general version in 1968. Up till then, they were called
> > links.
> >
> > When was "hyper" added as a prefix?
> >
> > The article at
> >
> https://www.psychologytoday.com/us/blog/seeing-what-others-dont/201801/the-invention-hyperlinks
> > claims Ben Shneiderman invented them in 1988 and implies Shneiderman
> coined
> > the term. You can see more of Shneiderman's claims at
> > http://www.cs.umd.edu/hcil/hyperties/. Shneiderman added the prefix
> "hyper"
> > 43 years after Doug named associative indexing "links" and 20 years after
> > my implementation.
> >
> > The article at https://en.wikipedia.org/wiki/Hyperlink#History
> attributes
> > the term hyperlinks to Nelson in 1965. But, to my knowledge, Nelson used
> > the term "links" at that time.
> >
> > My bet would be that hyper was added between 1980 and 1987. Maybe it was
> a
> > journalist and we will never know.
> >
> > Jeff
> >
> >
> > -------
> >
> > Also:
> > [image: Bush with Engelbart Annotation.png]
> > --
> > Internet-history mailing list
> > Internet-history at elists.isoc.org
> > https://elists.isoc.org/mailman/listinfo/internet-history
> --
> Internet-history mailing list
> Internet-history at elists.isoc.org
> https://elists.isoc.org/mailman/listinfo/internet-history
>