RE: [xml-dev] Is it time for the binary XML permathread to start up aga

[Alessandro Triglia <sandro@mclink.it>]

I won't discuss ASN.X or RXER here, but since you mention ASN.1 BER and DER, I would like to point out that the XML Encoding Rules of ASN.1 ("XER", ITU-T Rec. X.693 Amd. 1 | ISO/IEC 8825-4 Amd. 1) have been available for years, can be downloaded for free from the ITU-T website, are completely IP-free, and are implemented in a few commercial toolkits.  I gave a presentation on this subject at XML Europe 2002.

ASN.1, when used together with its XML Encoding Rules, is a viable schema language for XML documents.  Yet an ASN.1 schema is somewhat more abstract than an XSD schema, and any ASN.1 schema can be used to generate messages encoded in any set of encoding rules--from XML (XER) to BER, DER, or PER.  The choice of a set of encoding rules to be used with a given schema is usually fixed (in a standard, binding, or profile) to ensure interoperability.  However, things don't always need to be so rigid.  For example, if someone creates an ASN.1 schema that is intended to be encoded primarily in XER (thus generating XML documents), a closed set of endpoints could decide to use PER instead of XER when talking to each other in order to improve performance (there are big differences in both size and processing time).  This is quite a natural thing to do with ASN.1 because in ASN.1 the focus is on the so-called "abstract value" rather than on the actual document being transferred (the enc!
 oded message, or the "bits-on-the-wire").  From ASN.1's point of view, XML, PER, and BER have the same status--they are considered as alternative ways of representing an abstract value (of a specified type) on the wire.

Alessandro Triglia
OSS Nokalva
www.oss.com 


> -----Original Message-----
> From: Amelia A Lewis [mailto:amyzing@talsever.com] 
> Sent: Thursday, July 19, 2007 20:55
> To: xml-dev@lists.xml.org
> Subject: Re: [xml-dev] Is it time for the binary XML 
> permathread to start up again?
> 
> Heh.
> 
> For that matter, this past week also saw the publication (at 
> IETF) of a variety of RFCs documenting an intersection of XML 
> and ASN.1.  ASN.X is an XML schema language "semantically 
> equivalent" to ASN.1 (RFC 4912).  It would be interesting to 
> see if they've got a type system (if they do, it would be a 
> first for the XML Schema world ... :-).  Even a type 
> collection would potentially improve upon the WXS disaster.  
> Then there's Robust XML Encoding Rules (RXER), for XML 
> instances that represent ASN.1 (or semantically equivalent 
> ASN.X?) schemata.
> 
> Disclaimer: I haven't had a chance to read and review them.  
> Second disclaimer: ASN.1 is one of those things that I think 
> of fondly, but never quite find a use for.
> 
> Still ... there are already libraries out there that can 
> decode various ASN.1 encodings (BER, DER).  So this 
> potentially overlaps (strongly? partially? not at all?) with 
> EXI, except that ... well, there's already code available.  I 
> doubt that there are schema validators available for RXER 
> against ASN.X, but I doubt that it would be *terrifically* 
> hard to extend existing validators.  Depends upon the 
> formalism (if any) underlying ASN.X, I suppose.
> 
> See rfc-editor.org, RFCs in the range 4910 to 4914.
> 
> Amy!