Doug Kerr
Well-known member
The coded character set ASCII, the basis for almost all of today's coded character representation as itself, through its "extensions" (such as "code pages"), and thorough its super-successor, Unicode, first formally emerged on June 17, 1963 - fifty years ago - via the approval by the American Standards Association of its standard X3.4-1963, "American Standard Code for Information Interchange".
That version of ASCII (and that was not yet the official designation of the coded character set, just a widely-used acronym from the then-title of the standards document) only had upper-case alphabetic characters, the area of the code space where the lower-case letters were eventually mapped being shown as "for future assignment". This was part of some political silliness, of which there was an abundance in the matter of ASCII!
I had not been involved in the development to that point, but I joined the work shortly thereafter, and was heavily involved in the "finishing" of ASCII, culminating in the release of the 1967 and 1968 issues of the standard. (The name "ASCII" was formally given to the coded character set by the 1967 standard, a good thing, since by then, owing to a change in the name of the issuing organization, the document title already no longer spelled "ASCII".)
The lower case alphabetic characters were added, "where expected", in the 1967 version of the standard.
The 1968 version differed from the 1967 version only in that it included the provision that the character LF (Line Feed) could optionally have the meaning NL (New Line), a compromise between the traditional "teletypewriter" protocol of CR-LF at the end of the line and the then-customary "computer" protocol of a single "new line" character at the end of the line. The choice of LF as the character to receive this optional definition was based on a study I did regarding forward and backward compatibility implications.
I was a major contributor to the development of, and an author and principal editor of, the 1967 and 1968 ASCII standards.
The prior art
Prior to the emergence of ASCII, teletypewriter communication generally used one of several variants of a five-bit code properly known as the Murray Code. It was almost always (wholly incorrectly) spoken of as the "Baudot" code, which is a comparable, but wholly different code, used only for synchronous international telegraph operation, an obsolescent technology.
Transmission of "ready to typeset" text for newspapers, magazines, and even books was via a six-bit superset of one flavor of this five-bit code.
Teletypewriter-like data terminals used in some military systems used a six-bit code developed for the purpose.
Character representation in computers was typically by one of many wholly-different six-bit codes. One prominent one, used by IBM, was BCDIC, essentially a binary mapping of the "Hollerith" code used on punched data cards.
Big support from inside the Beltway
A strong supporter of the emerging effort to develop a standard coded character set was the United States Government. At the time, computers (and by that we meant what would today be called "mainframe computers") were leased (and in some cases sold) as complete "bundles" of all ingredients (software and all).
The US Government was keen on moving to a different model, in which one could solicit bids separately for the lease or sale of data systems by individual major units: central processor, control consoles, tape storage drives, printers, and remote terminals.
Of course for this to be possible would require a gigantic movement in the direction of standardization of interfaces and protocols. One small piece of this, but seen by the Government as the vital starting point, was the development of a standard coded character set in which all character data would be passed across interfaces, stored, and so forth.
Thus what became the ASCII movement, already underway by the information industry, was given great support and impetus by the Government, in terms of permitting government experts to serve on the development committees (very capable people, by the way), letting Government facilities host the committee meetings when it was "their turn", and so forth.
In fact, the first all-agency US Government standard in the information processing area, Federal Information Processing Standard No. 1 (FIPS 1), was in fact exactly the 1968 issue of the ASCII standard, by proclamation of President Lyndon B. Johnson.
For conspiracy theorists
A cynic might think as follows:
• Several large computer manufacturers at the time had prospered well under the business model of "completely bundled" systems.
• They might not be enthusiastic about the emergence, stimulated by the U. S. Government, of a "separate procurement of major subsystems" business model.
• Thus they might not be fully supportive of standardization efforts seen as technical keys to the practical emergence of such a model ASCII being the "poster boy" for this.
• Might then their substantial efforts in the committee work to develop ASCII and related standards not be done fully "in good faith"; might some technical positions strongly presented and supported actually be intended to "scramble" the work?
This is not the place for me to further reflect on this (although I could tell hundreds of wonderful stories). I can only say that I had great technical and personal respect for the committee members from those organizations, and shared much good technical and personal fellowship with them.
And I enjoyed the many "intellectual forensic debates" I had with them in the course of the work.
And I almost always won.
An early adoption
Finally let me recognize with pride the bold decision of the Bell Telephone System (my employer at the time) to, in 1961, in connection with the "mechanization" of its TWX (Teletypewriter Exchange) service, introduce new teletypewriters using what would soon become ASCII (prior to formal approval of the standard). This decision certainly helped encourage the adoption of ASCII, after its formal approval, by many other entities.
Still here
Many thought that ASCII would be a "flash in the pan".
But it's still here.
And so am I.
So, happy birthday, ASCII.
And while I'm up, happy birthday as well to my younger daughter, Linda, who was born two days after the first approval of ASCII (and her mother always claimed that this was the only way I could remember when her birthday was!). (Her older sister, Nancy, was born the day before my birthday, two years earlier.)
Best regards,
Doug
That version of ASCII (and that was not yet the official designation of the coded character set, just a widely-used acronym from the then-title of the standards document) only had upper-case alphabetic characters, the area of the code space where the lower-case letters were eventually mapped being shown as "for future assignment". This was part of some political silliness, of which there was an abundance in the matter of ASCII!
I had not been involved in the development to that point, but I joined the work shortly thereafter, and was heavily involved in the "finishing" of ASCII, culminating in the release of the 1967 and 1968 issues of the standard. (The name "ASCII" was formally given to the coded character set by the 1967 standard, a good thing, since by then, owing to a change in the name of the issuing organization, the document title already no longer spelled "ASCII".)
The lower case alphabetic characters were added, "where expected", in the 1967 version of the standard.
The 1968 version differed from the 1967 version only in that it included the provision that the character LF (Line Feed) could optionally have the meaning NL (New Line), a compromise between the traditional "teletypewriter" protocol of CR-LF at the end of the line and the then-customary "computer" protocol of a single "new line" character at the end of the line. The choice of LF as the character to receive this optional definition was based on a study I did regarding forward and backward compatibility implications.
I was a major contributor to the development of, and an author and principal editor of, the 1967 and 1968 ASCII standards.
The prior art
Prior to the emergence of ASCII, teletypewriter communication generally used one of several variants of a five-bit code properly known as the Murray Code. It was almost always (wholly incorrectly) spoken of as the "Baudot" code, which is a comparable, but wholly different code, used only for synchronous international telegraph operation, an obsolescent technology.
Transmission of "ready to typeset" text for newspapers, magazines, and even books was via a six-bit superset of one flavor of this five-bit code.
Teletypewriter-like data terminals used in some military systems used a six-bit code developed for the purpose.
Character representation in computers was typically by one of many wholly-different six-bit codes. One prominent one, used by IBM, was BCDIC, essentially a binary mapping of the "Hollerith" code used on punched data cards.
Big support from inside the Beltway
A strong supporter of the emerging effort to develop a standard coded character set was the United States Government. At the time, computers (and by that we meant what would today be called "mainframe computers") were leased (and in some cases sold) as complete "bundles" of all ingredients (software and all).
The US Government was keen on moving to a different model, in which one could solicit bids separately for the lease or sale of data systems by individual major units: central processor, control consoles, tape storage drives, printers, and remote terminals.
Of course for this to be possible would require a gigantic movement in the direction of standardization of interfaces and protocols. One small piece of this, but seen by the Government as the vital starting point, was the development of a standard coded character set in which all character data would be passed across interfaces, stored, and so forth.
Thus what became the ASCII movement, already underway by the information industry, was given great support and impetus by the Government, in terms of permitting government experts to serve on the development committees (very capable people, by the way), letting Government facilities host the committee meetings when it was "their turn", and so forth.
In fact, the first all-agency US Government standard in the information processing area, Federal Information Processing Standard No. 1 (FIPS 1), was in fact exactly the 1968 issue of the ASCII standard, by proclamation of President Lyndon B. Johnson.
For conspiracy theorists
A cynic might think as follows:
• Several large computer manufacturers at the time had prospered well under the business model of "completely bundled" systems.
• They might not be enthusiastic about the emergence, stimulated by the U. S. Government, of a "separate procurement of major subsystems" business model.
• Thus they might not be fully supportive of standardization efforts seen as technical keys to the practical emergence of such a model ASCII being the "poster boy" for this.
• Might then their substantial efforts in the committee work to develop ASCII and related standards not be done fully "in good faith"; might some technical positions strongly presented and supported actually be intended to "scramble" the work?
This is not the place for me to further reflect on this (although I could tell hundreds of wonderful stories). I can only say that I had great technical and personal respect for the committee members from those organizations, and shared much good technical and personal fellowship with them.
And I enjoyed the many "intellectual forensic debates" I had with them in the course of the work.
And I almost always won.
An early adoption
Finally let me recognize with pride the bold decision of the Bell Telephone System (my employer at the time) to, in 1961, in connection with the "mechanization" of its TWX (Teletypewriter Exchange) service, introduce new teletypewriters using what would soon become ASCII (prior to formal approval of the standard). This decision certainly helped encourage the adoption of ASCII, after its formal approval, by many other entities.
Still here
Many thought that ASCII would be a "flash in the pan".
But it's still here.
And so am I.
So, happy birthday, ASCII.
And while I'm up, happy birthday as well to my younger daughter, Linda, who was born two days after the first approval of ASCII (and her mother always claimed that this was the only way I could remember when her birthday was!). (Her older sister, Nancy, was born the day before my birthday, two years earlier.)
Best regards,
Doug
