Re: [Standards] BOSH and legacy auth - do we have to be backwards compatible?

2013-03-04 Thread Stefan Strigler
This again leads me back to the question whether we should change SHOULD to MUST within this paragraph: If no stream:features/ element is included in the connection manager's session creation response, then the client SHOULD send empty request elements until it receives a response

Re: [Standards] BOSH and legacy auth - do we have to be backwards compatible?

2013-02-15 Thread Winfried Tilanus
-BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 02/08/2013 06:28 PM, Peter Saint-Andre wrote: Hi, I think we can remove the XEP-0078 references. It has been obsolete since 2008. Steve, there seem to be no objections. Do you write a patch for it? Winfried -BEGIN PGP SIGNATURE-

Re: [Standards] BOSH and legacy auth - do we have to be backwards compatible?

2013-02-15 Thread Stefan Strigler
OK! Am 15.02.2013 um 13:14 schrieb Winfried Tilanus winfr...@tilanus.com: -BEGIN PGP SIGNED MESSAGE- Hash: SHA512 On 02/08/2013 06:28 PM, Peter Saint-Andre wrote: Hi, I think we can remove the XEP-0078 references. It has been obsolete since 2008. Steve, there seem to be no

Re: [Standards] BOSH and legacy auth - do we have to be backwards compatible?

2013-02-08 Thread Winfried Tilanus
On 02/07/2013 01:50 PM, Stefan Strigler wrote: Hi, Stream features are only provided by non legacy servers which might accept legacy auth for backwards compatibility with legacy clients. Legacy servers don't know about stream features. I have been rereading XEP-0078 on this, and you are

Re: [Standards] BOSH and legacy auth - do we have to be backwards compatible?

2013-02-08 Thread Peter Saint-Andre
-BEGIN PGP SIGNED MESSAGE- Hash: SHA1 On 2/8/13 2:37 AM, Winfried Tilanus wrote: On 02/07/2013 01:50 PM, Stefan Strigler wrote: Hi, Stream features are only provided by non legacy servers which might accept legacy auth for backwards compatibility with legacy clients. Legacy