Re: [Standards] Rename XEP status identifiers

[Kevin Smith]

On 27 Sep 2017, at 15:08, Sam Whited  wrote:
> 
> On Wed, Sep 27, 2017, at 02:08, Kevin Smith wrote:
>> Are they not then going to be upset if there are backwards-incompatible
>> changes in a new namespace? We try not to do that in Draft XEPs, but
>> that’s the reason they’re Draft rather than Final, to give us that
>> option.
> 
> "Stableish”?

+1

/K
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Sam Whited]

On Wed, Sep 27, 2017, at 02:08, Kevin Smith wrote:
> Are they not then going to be upset if there are backwards-incompatible
> changes in a new namespace? We try not to do that in Draft XEPs, but
> that’s the reason they’re Draft rather than Final, to give us that
> option.

"Stableish"?

—Sam
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Kevin Smith]

On 26 Sep 2017, at 20:35, Sam Whited  wrote:
> 
> On Tue, Sep 26, 2017, at 14:19, Ivan Vučica wrote:
>> And now, to bikeshed a bit on the proposed naming: To a casual reader,
>> stable has similar implications as final. Especially if said reader is
>> used to Debian's use of the word.
> 
> That seems like exactly what we want.

Are they not then going to be upset if there are backwards-incompatible changes 
in a new namespace? We try not to do that in Draft XEPs, but that’s the reason 
they’re Draft rather than Final, to give us that option.

/K
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Dave Cridland]

On 26 September 2017 at 20:19, Ivan Vučica  wrote:
> And now, to bikeshed a bit on the proposed naming: To a casual reader,
> stable has similar implications as final. Especially if said reader is used
> to Debian's use of the word.

I think to developers, the two *do* have similar implications.

Dave.
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Ivan Vučica]

On Tue, Sep 26, 2017 at 6:57 PM Sam Whited  wrote:

> On Tue, Sep 26, 2017, at 12:37, Ivan Vučica wrote:
> >
> > On 26 September 2017 at 14:47:27, Sam Whited (s...@samwhited.com) wrote:
> >
> > As others have said, the real naming problem is "draft". We can't
> > actively advance draft as much (since final really is final and can't be
> > touched ever again)
> > Is that a bad thing?
>
> To be clear, the fact that the name "draft" confuses people into
> thinking it's something that's not yet ready is a bad thing, not the
> fact that draft takes a long time to advance (which is a good thing).
>
> > Conversely, is it a good thing that certain XEPs have changed a lot since
> > the first draft?
>
> Your use of the phrase "the first draft" illustrates my point. I am not
> sure what you meant by that, if you meant "the first revision in the
> Draft status" then XEPs should not have changed a lot. If you mean "the
> first published revision of an XEP in Experimental", then it's not a bad
> thing that they've changed a lot, that's just the development process.
>

Don't get me wrong, I would be in favor of replacing draft with something
else. I am just not sure "we don't want to mark XEPs final as we can't
change them later" is a valid reason.

Maybe there should be more effort to actually make XEPs final and backwards
compatible, especially since many clients and servers claim compatibility
with XEPs, not with revisions of XEPs.

If I search today for software that implements XEP-0313, I often cannot
know if it'll actually be compatible with my server without digging into
the code and seeing that, ah, no, it's not compatible -- it has no support
for :2 namespace.

For searchability, I would have been happier if incompatible namespace
changes were made in a new XEP.


And now, to bikeshed a bit on the proposed naming: To a casual reader,
stable has similar implications as final. Especially if said reader is used
to Debian's use of the word.

Perhaps something like "ready" or, to borrow Debian conventions, "testing",
would be better?
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Sam Whited]

On Tue, Sep 26, 2017, at 12:37, Ivan Vučica wrote:
> 
> On 26 September 2017 at 14:47:27, Sam Whited (s...@samwhited.com) wrote:
> 
> As others have said, the real naming problem is "draft". We can't
> actively advance draft as much (since final really is final and can't be
> touched ever again)
> Is that a bad thing?

To be clear, the fact that the name "draft" confuses people into
thinking it's something that's not yet ready is a bad thing, not the
fact that draft takes a long time to advance (which is a good thing).

> Conversely, is it a good thing that certain XEPs have changed a lot since
> the first draft?

Your use of the phrase "the first draft" illustrates my point. I am not
sure what you meant by that, if you meant "the first revision in the
Draft status" then XEPs should not have changed a lot. If you mean "the
first published revision of an XEP in Experimental", then it's not a bad
thing that they've changed a lot, that's just the development process.

—Sam
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Ivan Vučica]

On 26 September 2017 at 14:47:27, Sam Whited (s...@samwhited.com) wrote:

As others have said, the real naming problem is "draft". We can't
actively advance draft as much (since final really is final and can't be
touched ever again)
Is that a bad thing?

Conversely, is it a good thing that certain XEPs have changed a lot since the 
first draft?





signature.asc
Description: Message signed with OpenPGP using AMPGpg
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Evgeny Khramtsov]

Tue, 26 Sep 2017 14:22:17 +0200
Goffi  wrote:

> I've seen that there was a need 
> to get disco items in XEP-0355. I've tried to update my Prosody
> implementation and Pubsub component to test it, and now that I see
> it's working, I want to update the XEP.

I actually found the disco part the most irritable. There is a state
where a component is connected, a server sent a disco request and no
disco response is received yet. In this state it's hard to say anything
about correctness of work of a component: this can be use case
dependent.
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Jonas Wielicki]

On Dienstag, 26. September 2017 09:47:10 CEST Sam Whited wrote:
> As others have said, the real naming problem is "draft". We can't
> actively advance draft as much (since final really is final and can't be
> touched ever again), so renaming it to something else ("Stable" sounds
> good to me) seems sensible.

+1.

signature.asc
Description: This is a digitally signed message part.
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Sam Whited]

On Tue, Sep 26, 2017, at 06:15, Dave Cridland wrote:
> > Should we rename the status names that we use in XEPs? One of the recurring
> > criticisms about XMPP that I read is "Pretty-standard-feature XYZ has a XEP
> > that is only "experimental"! By doing some window dressing, we will improve
> > the perceived maturity and stability of the protocol.

I agree, "experimental" seems accurately named to me (most people
probably shouldn't implement them, a few early adopters might be
interested, it might change, etc.). If that isn't the case (eg. an XEP
is ready for wider implementation) then it should not be experimental
anymore, it should have been advanced. That is, the problem is that we
advance things too slowly, not the name.

As others have said, the real naming problem is "draft". We can't
actively advance draft as much (since final really is final and can't be
touched ever again), so renaming it to something else ("Stable" sounds
good to me) seems sensible.

—Sam
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Florian Schmaus]

On 26.09.2017 15:38, Peter Saint-Andre wrote:
> On 9/26/17 5:15 AM, Dave Cridland wrote:
>> On 26 September 2017 at 10:03, Guus der Kinderen
>>  wrote:
>>> Hello all,
>>>
>>> Should we rename the status names that we use in XEPs? One of the recurring
>>> criticisms about XMPP that I read is "Pretty-standard-feature XYZ has a XEP
>>> that is only "experimental"! By doing some window dressing, we will improve
>>> the perceived maturity and stability of the protocol.
>>
>> Left shifts are bad.
>>
>> I'd rather we chose to aggressively advance XEPs, and where we find
>> that impossible, try to fix the problems preventing it. In other
>> words, I would rather improve the actual maturity and stability of the
>> protocol, rather than merely play with the perception of it.
>>
>> We need a playground, and we need that playground to be in the open -
>> Experimental XEPs should be providing this playground.
>>
>> That said, "Draft" is an IETF-ism that could easily enough be replaced
>> by "Stable" - it's been dropped at the IETF (incorrectly, I feel, but
>> still) so the reference is largely gone.
>>
>> Dave.
> 
> I'd be fine with "Stable".

+1 for s/Draft/Stable/

- Florian



signature.asc
Description: OpenPGP digital signature
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Peter Saint-Andre]

On 9/26/17 5:15 AM, Dave Cridland wrote:
> On 26 September 2017 at 10:03, Guus der Kinderen
>  wrote:
>> Hello all,
>>
>> Should we rename the status names that we use in XEPs? One of the recurring
>> criticisms about XMPP that I read is "Pretty-standard-feature XYZ has a XEP
>> that is only "experimental"! By doing some window dressing, we will improve
>> the perceived maturity and stability of the protocol.
> 
> Left shifts are bad.
> 
> I'd rather we chose to aggressively advance XEPs, and where we find
> that impossible, try to fix the problems preventing it. In other
> words, I would rather improve the actual maturity and stability of the
> protocol, rather than merely play with the perception of it.
> 
> We need a playground, and we need that playground to be in the open -
> Experimental XEPs should be providing this playground.
> 
> That said, "Draft" is an IETF-ism that could easily enough be replaced
> by "Stable" - it's been dropped at the IETF (incorrectly, I feel, but
> still) so the reference is largely gone.
> 
> Dave.

I'd be fine with "Stable".

Peter




signature.asc
Description: OpenPGP digital signature
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Goffi]

Le mardi 26 septembre 2017, 13:15:57 CEST Dave Cridland a écrit :
> On 26 September 2017 at 10:03, Guus der Kinderen
> 
>  wrote:
> > Hello all,
> > 
> > Should we rename the status names that we use in XEPs? One of the
> > recurring
> > criticisms about XMPP that I read is "Pretty-standard-feature XYZ has a
> > XEP
> > that is only "experimental"! By doing some window dressing, we will
> > improve
> > the perceived maturity and stability of the protocol.
> 
> Left shifts are bad.
> 
> I'd rather we chose to aggressively advance XEPs, and where we find
> that impossible, try to fix the problems preventing it. In other
> words, I would rather improve the actual maturity and stability of the
> protocol, rather than merely play with the perception of it.
> 
> We need a playground, and we need that playground to be in the open -
> Experimental XEPs should be providing this playground.
> 
> That said, "Draft" is an IETF-ism that could easily enough be replaced
> by "Stable" - it's been dropped at the IETF (incorrectly, I feel, but
> still) so the reference is largely gone.
> 
> Dave.


After a short discussion on the xsf@ MUC room, I'm just adding here my 
experience: I'm author of XEP-0355 and XEP-0356, and both have been written 
and modified lastly more that one year ago.
These XEPs have at least 2 implementations (one in Prosody that I've made, and 
one in Ejabberd), and it's used by at least one component (SàT Pubsub that I'm 
developping) which can potentially be of interest for the whole XMPP community 
(being a generic PEP/Pubsub component).

I've had nothing to add to these XEP during the last year, but I wanted to 
wait for maturity of implementations, and maybe feedback before requesting a 
move to draft. Meawhile these 2 XEPs have been deferred.

Recently, by testing compatibility with Movim, I've seen that there was a need 
to get disco items in XEP-0355. I've tried to update my Prosody implementation 
and Pubsub component to test it, and now that I see it's working, I want to 
update the XEP.

My point here is that it's not because a XEP is not touched at all that it is 
abandonned, it may just mean that the author is waiting for implementation/
maturity before requesting an update.

"Deferred" status is OK IMHO (I roughly consider it as equivalent to 
experimental), but it may refrain people to do an implementation, as they may 
thing the author is not taking care of the XEPs.

It was just to add my experience to the discussion, I'm not in favor of any 
workflow change, and not opposed to it either. But I think the time  is more to 
consolidate implementations and fix standard issues than to change workflow 
(which is working more or less at the moment).


Goffi
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Dave Cridland]

On 26 September 2017 at 10:03, Guus der Kinderen
 wrote:
> Hello all,
>
> Should we rename the status names that we use in XEPs? One of the recurring
> criticisms about XMPP that I read is "Pretty-standard-feature XYZ has a XEP
> that is only "experimental"! By doing some window dressing, we will improve
> the perceived maturity and stability of the protocol.

Left shifts are bad.

I'd rather we chose to aggressively advance XEPs, and where we find
that impossible, try to fix the problems preventing it. In other
words, I would rather improve the actual maturity and stability of the
protocol, rather than merely play with the perception of it.

We need a playground, and we need that playground to be in the open -
Experimental XEPs should be providing this playground.

That said, "Draft" is an IETF-ism that could easily enough be replaced
by "Stable" - it's been dropped at the IETF (incorrectly, I feel, but
still) so the reference is largely gone.

Dave.
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

Re: [Standards] Rename XEP status identifiers

[Georg Lukas]

* Guus der Kinderen  [2017-09-26 11:06]:
> By doing some window dressing, we will improve the perceived maturity
> and stability of the protocol.

Absolutely +1.

MUC is 15 years old, and it's still in "Draft". We really need better
names (though we probably need to discuss those at length, first).

To quote a quite pragmatic (but maybe not quite serious) proposal from
the xsf@ MUC:

> "Work in progress", "Almost done", "Done"


Kind regards

Georg
-- 
|| http://op-co.de ++  GCS d--(++) s: a C+++ UL+++ !P L+++ !E W+++ N  ++
|| gpg: 0x962FD2DE ||  o? K- w---() O M V? PS+ PE-- Y++ PGP+ t+ 5 R+  ||
|| Ge0rG: euIRCnet ||  X(+++) tv+ b+(++) DI+++ D- G e h- r++ y?   ||
++ IRCnet OFTC OPN ||_||


signature.asc
Description: PGP signature
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___

[Standards] Rename XEP status identifiers

[Guus der Kinderen]

Hello all,

Should we rename the status names that we use in XEPs? One of the recurring
criticisms about XMPP that I read is "Pretty-standard-feature XYZ has a XEP
that is only "experimental"! By doing some window dressing, we will improve
the perceived maturity and stability of the protocol.

Regards,

  Guus
___
Standards mailing list
Info: https://mail.jabber.org/mailman/listinfo/standards
Unsubscribe: standards-unsubscr...@xmpp.org
___