Origin Of The Abbreviation I18n

I18nGuy Home Page

What Is "I18n"?

"I18n" is an abbreviation for the word "internationalization". The term "i18n" is derived from its spelling as the letter "i" plus 18 letters plus the letter "n".

I+n1t2e3r4n5a6t7i8o9n10a11l12i13z14a15t16i17o18+n

Technically, the term is not an acronym, as acronyms represent expressions that are derived from the first letters of words. (For example, "DBCS" is an acronym for "Double Byte Character Set".) I believe the correct name for this type of abbreviation is numeronym- a number based word. Examples include "K9" for canine, and the French "K7", pronounced "K-sept" for the word cassette. Another example is "411" for the concept of "information".

What Is The Origin Of "I18n"?

I asked this question on the Unicode Consortium public mail list and the NELOCSIG mail list on Thursday 2002-10-10. Here is my compilation of the discussion.
I would be glad to update this page if readers send me additional information, preferably with some evidence (pointers to books, mails, or other documentation).
I have noted author's names in the thread excerpts. I would be glad to either remove them or add other personal identifiers (e-mail address or web site) if people would like to be anonymous or well-identified. (I would prefer stronger identification to better establish credibility and in case someone else wants to do further research on the subject.)
Many of the links are now dead. I removed the links but kept the addresses as parenthetical remarks in this style

Usage Of The Terms "Internationalization" And "Localization"

The term "internationalization" was used as far back as 1985 at Apple and DEC. Unix seems to have picked it up at least in the late 80s.

Not all companies used the term. Based on documentation I have from before 1990, IBM and LOTUS seemed to prefer the term "international-enabling". (They were separate and distinct companies at the time, Lotus being DOS-based may have inherited "enabling" from IBM.) They didn't use the term "internationalization" until the '90s.

The term "Localization" was also in use in the mid-80s, by all of the vendors. (Or at least by IBM, DEC, Apple, Lotus, HP, Unixes).

Summary Of The Origin Of "I18n"

The story goes as follows:
A DEC employee named Jan Scherpenhuizen was given an email account of S12n by a system administrator, since his name was too long to be an account name. This approach to abbreviating long names was intended to be humorous and became generalized at DEC. The convention was applied to "internationalization" at DEC which was using the numeronym by 1985.

Use of the term spread. Searching the net, we found uses on-line as early as 1989. It was being used on /usr/group, which evolved into UniForum. The X Window standards community was also using the abbreviation by 1989. Looking in printed texts, the earliest reference I could find was in the book Soft Landing in Japan, published by American Electronic Association, 1992.

The extension of this naming convention to the terms Localization (l10n), Europeanization (e13n), Japanization (j10n), Globalization (g11n), seemed to come somewhat after the invention of "i18n". The terms Canonicalization and Normalization, defined more recently, also have numeronym forms (c14n and n11n).

Thanks to the many contributors to this thread.

Excerpts Of The Email Discussion

Author: Arnold Winkler

Here is my recollection:
Sometime around 1991 in a IEEE P1003.1 (POSIX) meeting, Gary Miller (IBM) was writing on the blackboard. After having spelled out Internationalization a few times, he first abbreviated it to I--n and a bit later (obviously after counting the letters in between) used I18N.
L10N did not show up until quite some time later. I have no idea who used it first.

Author: Radovan Garabík, searching with Google:

"i18n" is first mentioned in USENET on 30 nov 1989, "l10n" is first mentioned on 30 august 1990. The posting also mentions that the abbreviation is not widely used.

Author: Hideki Hiura

The acronym "I18N" appeared before 1991, since I recall I have already used I18N in '89 ;-).
The beginning of this kind of acronym was S12N(Scherpenhuizen) at DEC, as far as on the record, as an email address for him on DEC VMS.
By 1985, I18N became an acronym for Internationalization in the I18N team at DEC, by following this Scherpenhuizen's S12N convention.
Among the standard organizations, the /usr/group (It became UniForum later) was the first one using I18N as an acronym for Internationalization, in '88.

Author: Jim Melton

...I18n was in use long before 1991. I first started using it myself in perhaps 1987, having picked it up from colleagues at Digital Equipment Corporation (remember *them*?); I have no idea where they got the term, though. However, I first encountered L10n quite some years later, possibly as recently as 1994 or 1995, but I don't recall it as clearly.

Author: Rick McGowan

The earliest reference I can find to "i18n" in my old e-mail trail is the following e-mail to the "sun!unicode" mail list by Glenn Wright. This was Oct 5, 1989. By that time, the term was definitely current, as Mr. Hiura suggests.


> From upheisei!attunix!sun!glennw Thu Oct  5 15:59:05 EDT 1989
> Date: Thu, 5 Oct 89 12:46:22 PDT
> From: sun!glennw (Glenn P. Wright)
> To: sun!unicode
> Subject: Next Unicode meeting
>
> Time:
>
>       Monday Oct 9th. 4pm
>
> Place:
>       Sun Microsystems, Building 5, 2550 Garcia Avenue
>
> Agenda:
>
>       Discussion of Current X3L2 proposal status
>       Discussion on future Unicode committee organisation
>       Joe will have updates on the Unicode chart
>                       (depending on the As pitching rate)
>       Liason reports -
>               X/Open,
>               Joint i18n meeting,
>               Ad Hoc meeting in Peking (oops).
Author: Tex Texin

I went thru my i18n books to scan for mentions. The earliest mention I could find for i18n (the abbreviation) was 1992, in "Soft Landing in Japan".
It seems like 1993 some books mention it, and 1994 and thereafter it is consistently mentioned.
The term "internationalization" seems vendor oriented. IBM preferred "enabling" (and NLS). I have some early DEC books (1987 or so), and the abbreviation wasn't used.
The lack of use in books before 1992 could be a choice by the authors that the term was jargon or slang and wasn't relevant to explaining the concepts.
It seems to have passed from DEC to Unix usage, if we believe the reports I am getting.

Author: Barry Caplan

Barry's web site i18n.com has a story on the origin, The Strange Birth Of The Term 'I18n' Friday March 01, @11:39PM:
(http://www.i18n.com/article.pl?sid=02/03/02/0345215&mode=thread&threshold=)

Author: John McConnell

I can confirm Hiura-san's version. I heard it from Jurgen Bettels, who've I've known since '84 and worked with Scherpenhuizen in the Geneva office at the time. Scherpenhuizen managed the ISO work.

In the days when bytes were precious, VMS had a character username limit. Some anonymous system administrator shortened Scherpenhuizen's username to S12N to fit. It may have been especially funny because Scherpenhuizen was an unusually large man. Whatever, it became an office joke for long words. Internationalization had just become a hot topic in ISO then, so it was applied, and stuck.

I suspect Eamon MacDermott may have been the person who spread the term from DEC to the rest of the world.

Author: Tor Lillqvist

Well, the first occurence of "i18n" in Google's USENET archive seems to be Rob Robason, Nov 30 1989, 11:38 am, Subject: cp gives poor error message (so what else is new?) (http://groups.google.com/groups?selm=5570339%40hpfcdc.HP.COM) from Nov 30, 1989.

"l10n" occurs first in Dominic Dunlop's Article on Internationalisation Dated Aug 30 1990, 7:56 am (http://groups.google.com/groups?selm=1990Aug30.115608.3729%40tsa.co.uk) from Aug 30, 1990.

Author: Hideki Hiura

The statement I posted earlier,

hiura> The acronym "I18N" appeared before 1991, since I recall I have
hiura> already used I18N in '89 ;-).
was exactly refering to the use of "I18N" in the committee the article on i18n.com referring. As some of you may remember ;-), I was a part of this committee, so I know this committee was not the origin of the term I18N.

We've created two mailing lists at MIT for the X Window System Internationalization activity, called mltalk(multilingual talk) and i18n-si (Internationalization sample implementation) back on those days.

DEC was heavily involved in the development of X Window System since very early phase, I can imagine there were some idea exchanges on the naming, so it is possible that the rep. of DEC passed the hint to this committee.

Author: Barry Caplan

Here is a mention from 1989-12-02 11:24:11 PST only 3 days later:
Jeffrey S. Haemer, Dec 2, 1989 19:24:11 GMT, Subject: Standards Update, IEEE 1003.1: System services interface
(http://groups.google.com/groups?q=i18n+1988&hl=en&lr=&ie=UTF-8&selm=454%40longway.TIC.COM&rnum=7) that says:

> 5.  Messaging
>
>      The UniForum internationalization (I18N) folks brought forward a
>      proposal for a messaging facility to be included in P1003.1b.
>      The working group decided that it needs some more work but will
>      go into the next draft.
>
>      [Editor's note -- The problem being solved here is that
>      internationalized applications store all user-visible strings in
>      external files, so that vendors and users can change the
>
>December 1989 Standards Update  IEEE 1003.1: System services interface
>
>
>                                - 5 -
>
>      language of an application without recompiling it.  The UniForum
>      I18N group is proposing a standard format for those files.]
This indicates to me that UniForum might be a place to look for earlier references....
This is a very interesting thread from 1990: Dominic Dunlop's Article on Internationalisation Dated Aug 30 1990, 7:56 am
(http://groups.google.com/groups?hl=en&lr=&ie=UTF-8&threadm=1990Aug30.115608.3729%40tsa.co.uk&rnum=20&prev=/groups%3Fq%3Di18n%2B1988%26start%3D10%26hl%3Den%26lr%3D%26ie%3DUTF-8%26selm%3D1990Aug30.115608.3729%2540tsa.co.uk%26rnum%3D20)
Author: Mark Davis

We used the term "internationalization" in Apple in late 85. We might have also used it earlier than that, I don't remember.

Author: Timothy Greenwood

I concur with the stories from the other DEC folks and certainly remember Jan Scherpenhuizen and S12N.

Some idea of a lower date for common use of I18N are books that talk about internationalization but do not use the abbreviation. It is not used in the July 1993 X/Open Internationalisation Guide nor the summer 93 Digital Technical Journal on Product Internationalization. Nor do I see it in the 1991 'Digital Guide to Developing International Software'. This was based on an internally distributed DEC manual - I believe that I have a copy at home. These dates tie in with the findings from Tex. I suspect that the term was in internal use, but not considered as fit for publication. The term internationalization itself is not used in my earliest reference, proceedings from an internal DEC conference on International Opportunities and Differences' April 1985.

Author: Bob Jung     2002-10-25

...My memory may be fuzzy...

I started attending many of the /usr/group internationalization SIG meetings from early 1986. (I had just returned from 5 months in Japan where I worked on the original AT&T MNLS and JAE development.) As noted on your page /usr/group evolved into Uniforum and the SIG continued working as a Uniforum SIG.

At one point in one of the early meetings, someone was writing the long "internationalization" word yet again on the white board. Since we didn't like writing this long word all the time, someone suggested we find an abbreviation. After some discussion (I don't know if the idea was borrowed from somewhere else), the suggestion of "i18n" was agreed upon and since then that SIG "standardized" upon it.

BTW, it was at this SIG that we conceived (or at least proposed) the locale model. (Aside from Tex- perhaps another origin page beginning...)

Maybe if we can find some of the early SIG members, we can try to confirm my memory. I think these folks were involved at that time:
Brian Boyle, a consultant based in Berkeley
Karen Barnes of HP
Greger (long last name which I cannot remember) of DEC