The notation of OID strings as 1.3.6.1.4.1 started appearing in the ISODE
ASN.1 compiler, in the late 80's. It was not part of the ASN.1 standard; in
fact, ASN.1 defines its own set of format, that can mix numbers and
litterals. In ASN.1, this was called a "value notation." A standard ASN.1
textual representation would have been, for example, {1 3 6 1 4 1}, or {1 3
6 foo(1) bar(4) 1}, as in
        fooBar OBJECT IDENTIFIER ::= {1 3 6 foo(1) bar(4) 1}
Why the ASN.1 representation never really caught up in user interfaces is
left as an exercize for the reader...

> -----Original Message-----
> From: Peter Furniss [mailto:[EMAIL PROTECTED]]
> Sent: Thursday, June 22, 2000 4:13 PM
> To: '[EMAIL PROTECTED]'; 'Michael Mealling'
> Subject: RE: Where is the OID "dot convention" spelled out?
> 
> 
> Michael Mealling sent :
> 
> > For all the ASN.1 folks out there:
> > 
> > I'm in the midst of writing up the OID URN namespace document 
> > (see 
> http://www.ietf.org/internet-drafts/draft-mealling-oid-urn-00.txt)
> > and it has come to my attention that none of the ASN.1 standards 
> > define the dot-notation that we use in all sorts of RFCs. 
> I'm specifically
> > referring to the practice of inserting dots in between each 
> arc as in:
> > 1.3.6.1.4.1
> > 
> > Is anyone aware if this is actually spelled out somewhere? I don't
> > have the newest ASN.1 docs in front of me so if the're in there
> > a page reference would be great.
> 
> It is an IETF convention, not one defined in the ASN.1 
> standards. It is used at least as early as RFC 1157 (snmp), 
> which refers to "the familiar dot notation" - familiar, I'd 
> assumed, in the dot notation of IP addresses. (which seems to 
> be first mentioned in RFC 790, though it says it was already 
> in use then (and had leading zeros)
>  
> The ASN.1 standards use spaces between the fields and 
> {braces} round the whole lot in text representations. They 
> also use name(number) to identify what the fields are, and 
> the first part of a text representation can also be an 
> assigned asn.1 value (so a whole subtree can be specified 
> without repeating the value of the root all over the place)
> 
> 
> Peter Furniss
> ISO/IEC JTC1 OSI Maintenance Rapporteur (on behalf of BSI)
> OSI Maintenance website: http://www.furniss.co.uk/maint
> 
> 

Reply via email to