Accredited Standards Committee Doc. No.: L2/98-414

NCITS* Date: 02 December, 1998

L2, Codes and Character Sets Projects: 12-M, 13-M

Reference: L2/98-280

Reply to: Edwin Hart

To: L2 Applied Physics Laboratory

11100 Johns Hopkins Rd.

Laurel, MD 20723-6099

Voice: +1 443 778-6926

FAX: +1 443 778-1093


Subject: Question Vote to Withdraw X3.4 and X3.83

Action Requested: Reconsider

I believe that L2 made a serious mistake in voting to recommend that NCITC withdraw X3.4 and X3.83. Based on the information included here, I request that the L2 members reconsider the withdrawal.

X3.4-1986, 7-bit ASCII

The best reason for not withdrawing X3.4 is because it is a politically charged issue.

Other standards important to the information technology community reference X3.4 and use X3.4 as the base character-set.
The ANSI/ISO standard for the C programming language references X3.4 as the base character-set. The Internet standard for electronic mail (SMTP) references X3.4 as the base character-set. Moreover, the MIME standard (extension to SMTP) retains the 7-bit restriction of the SMTP standard.

7-bit ASCII has become ubiquitous and is referenced everywhere.
ASCII has immediate name-recognition by vendors and users of information technology. ANSI will destroy the trust of the public were it to withdraw the 7-bit ASCII standard.

X3.4 is not the same as the IRV of ISO/IEC 646-1991.
Although they share the same repertoire of graphic characters, X3.4 defines the control characters encoded in the C0 area but ISO/IEC 646 does not. This has implications on implementing 7-bit ASCII and conformance.

X3.83, US procedure for registering code table according to ISO 2375

X3.83 is not the same as ISO 2375.
ISO 2375 specifies a registry for coded character sets. X3.83 defines specific requirements and the procedure for the US to submit registration requests to the ISO registration authority for ISO 2375.

Although the US has not submitted any registrations for several years, I would still recommend not withdrawing this standard. If the US would need to register a coded character set, it would be easier to have the procedure in place rather than reinventing one.