Talk:Basic Encoding Rules

This page should redirect from "BER". Efadae (talk) 20:52, 1 April 2009 (UTC)

The paragraph stating the reader can view BER as a kind of "binary" XML seems odd to me. I think it should be striken. Kdz 01:49, 6 April 2006 (UTC)


 * Agreed. Although the format is parseable without knowing the exact schema, the essence of BER is to strictly conform to a particular ASN.1 schema, and is not really comparable to XML in that respect. Looks like this article needs work, I'm putting a cleanup tag on it. intgr 18:43, 6 April 2006 (UTC)


 * Agreed. If "BER is a basic self-describing serialization of ASN.1 datatypes that include atomic and non-atomic types such as structures and lists". Although this is basically stated in the article, I use the terms datatype and structure differently. A decoder can reveal the datatypes and values, but it cannot reveal the names of the elements without the schema. In most BinaryXML implementations a decoder could probably determine the element contents, but not their name, or datatype as defined by the XMLSchema, so less information would be in the stream. --208.198.163.221 19:02, 20 April 2006 (UTC)

So who is this Huw Rogers guy exactly? ...and is his comment helpful at all?


 * The same question popped into my mind. -- intgr 22:26, 5 April 2006 (UTC)


 * The comment seems pointless (unless maybe applied to ASN.1 {joke}), but the alignment issue with the serialization was a concern for some hence PER as an aligned option which can make decoding easier, if not faster. --208.198.163.221 19:02, 20 April 2006 (UTC)


 * Huw Rogers has a page at http://www.bbc.co.uk/derby/bbc_bus/meet_the_team/huw_rogers.shtml - still not sure why we should reckon him as relevant to the topic. --lpoulsen 18:48, 20 April 2006 (PST)

I doubt that this is the same Huw Rogers of the link above. I know another Huw Rogers who led a team recently that was investigating the BER protocol for internal use at his former place of emplyment (which is where I work now). And I don't know of any reason that this comment is relevant, just because he said it.

Article cleanup
I have cleaned up this article a good bit, based on readings of the actual standards and a couple of ACM studies or articles about it. I could not find a good reference for Huw Roger's quote (every single hit on Google was clearly a direct copy of the FOLDOC and/or Wikipedia reference), so I left it out.

I'm not entirely sure how to deal with the "its inefficient" idea in general, though. Based on the number of comments on various ASN.1-related mailing lists, BER syntax clearly has a broad perception of being inefficient. But none of those comments was very authoritative, or made reference to any published or documented study; they were merely repeating "common knowledge". I did find one article that referenced the perceived inefficiency by way of refuting it, so I put that in.

I think the cleanup tag can probably be removed now, but I'd like other people to read over the article first.

Kutulu 15:34, 26 September 2006 (UTC)

Thanks, I think it's pretty good and comprehensive now - cleanup removed. I think the lead section could use some expansion, and the article should mention that "encoding formats" are officially called "transfer syntax". I'm also not convinced that PKCS uses BER and not DER, however I can't be bothered to check right now. I fixed some minor grammar/punctation/capitalization errors to fit Manual of Style while at it. Note you would normally use  for bold text instead of   - as the latter can be misleading and will magically turn into a link when the article is moved/renamed. -- intgr 16:24, 26 September 2006 (UTC)


 * According to the PKCS #7 document, it uses BER in some cases and DER in others: The standard is designed such that the enhanced content types can be prepared in a single pass using indefinite-length BER encoding, and processed in a single pass in any BER encoding." ... "Since DER encoding is required by the signed-data, signed-and-enveloped data, and digested-data content types, an extra pass may be necessary when a content type other than data is the inner content of one of those content types.".  I'll try to expand the lead section a bit, but I'm trying not to get overly technical. Kutulu 19:21, 26 September 2006 (UTC)