[Standards] BOSH patches, hopefully the last before final
christian.schudt at gmx.de
Sun Feb 2 12:29:46 UTC 2014
sure, I can understand that.
I am of course not in the position to decide about such things (may vs should vs must…) and I can hardly weigh up the real impact of such a change, but I thought status "Draft" is liberal enough to allow such changes.
(If not, I'd be interested in what kind of changes are allowed)
XEP-0047: In-Band Bytestreams changed something, too, when it became Final:
"Per a vote of the XMPP Council, advanced to Final; in addition, specified that the 'block-size' attribute defines the size of the data chunk before instead of after base64-encoding, to accord with existing implementations."
During Draft status the "block-size" was the base64 encoded size. In Final status it is now the byte array size _prior_ to base64 encoding.
This change surely broke some implementations too, which can be seen here as an example: http://community.igniterealtime.org/thread/45667
Am 02.02.2014 um 13:01 schrieb Winfried Tilanus:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA512
> On 02/01/2014 06:17 PM, Christian Schudt wrote:
>> The whole issue with the 'from' attribute is no big problem of
>> course. It was just something that caught my attention.
> OK, clear.
>> On the other hand, I am not considering it a "major thing" to
>> change a MAY to a MUST in the BOSH specification, especially
>> because the draft status clearly says "but some changes to the
>> protocol are possible before it becomes a Final Standard". But I am
>> no expert in XEP definitions. Maybe a SHOULD is a also possible!?
> Well, the 'MAY' says it is optional to do. "SHOULD' says: "do this,
> but if you know very well what you are doing and have good reasons to
> do something else, then you can do something else, but don't complain
> with us if it breaks anything." "MUST" says: do this or it breaks.
> In your implementation it breaks things, so I understand your request
> to change this to a SHOULD. But changing it to a SHOULD would mean
> that quite a lot working implementations have to be changed. So we
> differ in opinion here: I consider that as a major thing.
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1
> -----END PGP SIGNATURE-----
More information about the Standards