Hello all,
I'm a user of the MyFaces implementation of JSF and would like to give my
opinion on this.
It appair that there are two opinion on this issue based upon two motivations:
1) Lock the versionnumber of MyFaces to the specifications.
2) Use the common major.minor.patch-versioning-schema independant of the
spec-versioning.
The upside of the first option is that users immediately understand which
version of JSF they can use a MyFaces release for.
The downside of the first is that the MyFaces communicity 'looses' the option
to use the common versioning schema.
Although this might be pervented if the MyFaces-community should decide to use
a combination of both by starting to use the schema:
specMajor.specMinor.implMajor.implMinor.implPatch.
This would give us:
- MyFaces 1.1.1.1.5 for the current release.
- MyFaces 1.1.2.0.0 for some major rework of the current implementation.
- MyFaces 1.2.1.0.0 for the initial release of a MyFaces Implementation of the
JSF 1.2 ( as currently being developed in the trunk )
- MyFaces 2.0.1.0.0 for the first implementation of the JSF 2.0 spec.
Although this could become a problem with the next release which would be
1.1.2.0.0, which is 'lower' then the current 1.1.5-release. Besides using 5
digits for version might be a bit much.
So I don't like the locking of the versioning numbers of MyFaces to the
versioning of the JSF-spec.
Although I agree with everybody whom have stated that it would be convenient if
the version of MyFaces would directy give information about the version of the
jsf-spec, I don't think that MUST be applied.
I think that most people, who use MyFaces, are smart people ( they propably are
software developers ;-) ), so they will understand that release x.y.z of
myfaces does not neccessary mean that it implements exactly version x.y.z. of
some specification.
Of course the website/documentation should specify the implemented version of
the spec in such a way, that it can easily be found. ( Like a compatibility
matrix or like Tomcat, which has a table on their homepage, which version of
tomcat implements each version of the spec. )
My suggestion would be to use the normal major.minor.patch-schema, resulting in:
- use MyFaces 1.x.x for implementation(s) of any JSF 1.1-spec.
- use MyFaces 2.x.x for implementation of the JSF 1.2-spec.
- Deal with the other problems when they emerge.
Possible solution if there ever comes a JSF 2.0-specification:
* Rename MyFaces to YourFaces ( Change the name from 1st person to 2nd-person,
just like the specification )
An consequently on release of JSF 3.0:
* Rename the product to HerFaces, when it wins the vote over HisFaces. ;-)
With kind regards,
Marco Beelen
-----Original Message-----
From: Bruno Aranda [mailto:[EMAIL PROTECTED]
Sent: dinsdag 22 mei 2007 15:13
To: MyFaces Development
Subject: Re: MyFaces 2.0.0 (was Re: Tomahawk 1.1.5 release plans?)
Ok, I see your points of having a more flexible versioning of
myfaces-impl (as Martin says, myfaces-api is not going to change ever,
a part from bug fixing). The only thing is that I think is more
natural to the standard user to know which jsf is implemented by
looking just at the version of the myfaces-impl, instead of having to
go through documentation, and the confusion can be greater when
myfaces-impl 2.0 and jsf-ri 2.0 are out there, both artifacts
implementing different versions of the spec. Of course, I know that
they are completely different things, but not everyone does.
Development-wise I am with you that myfaces-2.0 would be more
meaningful and flexible and I like it, but I think it is a matter of
compromise to avoid future confusion.
This is one of the issues with more controversy since a while!
Bruno
On 22/05/07, Paul Spencer <[EMAIL PROTECTED]> wrote:
> Bruno,
> Regardless if the version number, I would expect the community
> and PMC would prevent this from occurring.
>
> Paul Spencer
>
> Bruno Aranda wrote:
> > Hi, I can imagine a free evolution of myfaces-impl, but this would
> > come at a cost of incompatibility with the RI. If we add new
> > signatures and other artifacts depend on those signatures, that
> > artifact is depending in the implementation and cannot be used with
> > other implementations (e.g. RI). Is this really what we want? This is
> > why I think that the impl should not grow and should be restricted to
> > be *just* an implementation of the api.
> >
> > My 2 pences,
> >
> > Bruno
> >
> > On 22/05/07, Martin Marinschek <[EMAIL PROTECTED]> wrote:
> >
> >> I've always been of Manfred's opinion - it would be better to decouple
> >> spec version numbers from implementation version numbers, so I'm...
> >>
> >> +1 for MyFaces-Impl 2.0
> >>
> >> if we don't do that, we force ourselves into an artifical corset in
> >> which we cannot move - we can only increment minor version numbers,
> >> and that means that almost no changes have been committed (users would
> >> expect only bug-fixes), whereas the implementation could grow in
> >> functionality significantly independent from the spec.
> >>
> >> MyFaces API can stay with a version number of 1.2, though
> >>
> >> regards,
> >>
> >> Martin
> >>
> >> On 5/21/07, Zubin Wadia <[EMAIL PROTECTED]> wrote:
> >> > It is a discussion about the core - I am only trying to establish
> >> WHY there
> >> > are two schools of thought on this - refer to Manfred's post to this
> >> thread
> >> > on May 18th.
> >> >
> >> > Cheers,
> >> >
> >> > Z.
> >> >
> >> >
> >> > On 5/21/07, Mike Kienenberger <[EMAIL PROTECTED]> wrote:
> >> > > I thought we were simply discussing MyFaces Core.
> >> > >
> >> > > Let me clarify my vote:
> >> > >
> >> > > +1 1.2 MyFaces Core for JSF 1.2.
> >> > > -1 2.0 MyFaces Core for JSF 1.2.
> >> > >
> >> > > Don't care for Tomahawk/Trinidad/Tobago. These are no longer
> >> > > tightly-coupled to a specific MyFaces core release, and should use
> >> > > whatever versions make the most sense. This is already true for
> >> > > "shared", Trinidad, and Tobago. It's going to happen anyway for
> >> > > Tomahawk once Myfaces 1.2 becomes trunk since Myfaces 1.1 releases
> >> are
> >> > > going to be few and far between once the majority of committers have
> >> > > switched to 1.2.
> >> > >
> >> > > While there have been matching releases for Tomahawk and Core up to
> >> > > this point, this has been due to the elimination of the previous
> >> > > coupling between Core and Tomahawk (a process that was more involved
> >> > > and took longer than anyone expected).
> >> > >
> >> > > For tomahawk, my "don't care" suggestion for versioning would be to
> >> > > use the same version as "shared" as long as that makes sense.
> >> > >
> >> > >
> >> > > On 5/21/07, Zubin Wadia <[EMAIL PROTECTED]> wrote:
> >> > > > There will always be an impedence mismatch here because MyFaces no
> >> > longer
> >> > > > represents the "Spec" but also various component projects. So I see
> >> > > > Manfred/Matze's point.
> >> > > >
> >> > > > This is why I have always advocated letting the Component
> >> initiatives
> >> > reign
> >> > > > alone in terms of their version order, release frequency and
> >> alignment
> >> > with
> >> > > > MyFaces and/or the Sun RI.
> >> > > >
> >> > > > And to think that we have the same exposure as the Tomcat
> >> community is
> >> > > > pushing it. We are nowhere near as big as them - yet.
> >> > > >
> >> > > > So while they can start naming their releases after varieties of
> >> > Hibiscus
> >> > > > flowers in the future - we can't.
> >> > > >
> >> > > > I'm still +1 on 1.2.
> >> > > >
> >> > > > Cheers,
> >> > > >
> >> > > > Zubin.
> >> > > >
> >> > > >
> >> > > > On 5/21/07, Bruno Aranda < [EMAIL PROTECTED] > wrote:
> >> > > > > +1 for 1.2
> >> > > > > -1 for 2.0
> >> > > > >
> >> > > > > I do agree that using 2.0 may cause confusion, as unlike what
> >> happens
> >> > > > > with tomcat, there will be a future version 2.0 of the spec when
> >> > > > > myfaces 2.0 is there already. People, unaware of the versioning
> >> > > > > procedure of the myfaces project, will go and fetch this version
> >> > > > > thinking that it is the implementation of jsf 2.0.
> >> > > > >
> >> > > > > Cheers,
> >> > > > >
> >> > > > > Bruno
> >> > > > >
> >> > > > > On 21/05/07, Mike Kienenberger <[EMAIL PROTECTED]> wrote:
> >> > > > > > +1 for 1.2.
> >> > > > > > -1 for 2.0.
> >> > > > > >
> >> > > > > > I see no advantage to using major version numbers which
> >> differ from
> >> > > > > > the spec. I see the disadvantage of confusion.
> >> > > > > >
> >> > > > > > Also, Manfred, you can have a -1 vote on this issue, but not
> >> a veto.
> >> > > > > >
> >> > > > > > "Vetos only apply to code changes; they do not apply to
> >> procedural
> >> > > > > > issues such as software releases."
> >> > > > > > http://www.apache.org/foundation/glossary.html
> >> > > > > >
> >> > > > > > See also
> >> > > > > >
> >> > > >
> >> >
> >> http://mail-archives.apache.org/mod_mbox/incubator-general/200606.mbox/[EMAIL
> >> PROTECTED]
> >>
> >> > > > > >
> >> > > > > >
> >> > > > > >
> >> > > > > > On 5/18/07, Manfred Geiler < [EMAIL PROTECTED] > wrote:
> >> > > > > > > Hi folks,
> >> > > > > > >
> >> > > > > > > Like Paul Spencer I'm also still
> >> > > > > > > +1
> >> > > > > > > for
> >> > > > > > > MyFaces 1.x.y --> JSF 1.1
> >> > > > > > > MyFaces 2.x.y --> JSF 1.2
> >> > > > > > > MyFaces 3.x.y --> JSF 2.0
> >> > > > > > > MyFaces 4.x.y --> JSF whatever comes next
> >> > > > > > >
> >> > > > > > > Here is my explanation for the "why":
> >> > > > > > > This one is similar to Tomcat version numbering and I do not
> >> > remember
> >> > > > > > > anyone complaining about having a Tomcat 5.x that is an
> >> > implementaion
> >> > > > > > > of Servlet 2.4 and Tomcat 6.x being a Servlet 2.5 container.
> >> > > > > > > If there will be a "release vs. spec table" on the MyFaces
> >> > Homepage
> >> > > > > > > (like the one on http://tomcat.apache.org/) nobody will
> >> ever be
> >> > > > > > > confused.
> >> > > > > > > The big advantage of having (only) the major number
> >> aligned to the
> >> > > > > > > spec is the degree of freedom with minor (x) and fix (y)
> >> number.
> >> > It is
> >> > > > > > > a well known and successful pattern to have this
> >> major.minor.fix
> >> > > > > > > version numbering scheme. With the 1.2.x versioning on the
> >> other
> >> > hand,
> >> > > > > > > how could we ever differentiate between a minor release
> >> (with new
> >> > > > > > > features and maybe slightly changed API for non-spec
> >> stuff) and a
> >> > bug
> >> > > > > > > fix only release, if we may only count the last number up?!
> >> > > > > > > Remember the Tomcat jump from 5.0.x to 5.5.x when they did a
> >> > complete
> >> > > > > > > rewriting of the core stuff? How could they ever have
> >> expressed
> >> > that
> >> > > > > > > in version numbering if they had stolidly aligned their
> >> tomcat
> >> > version
> >> > > > > > > to the servlet spec 2.4?
> >> > > > > > >
> >> > > > > > > And do not forget:
> >> > > > > > > There is not only the implementation. There are 3
> >> component libs
> >> > under
> >> > > > > > > the MyFaces umbrella. And IMHO it is much more important
> >> to align
> >> > all
> >> > > > > > > the myfaces stuff (compatible to each other) within one major
> >> > number
> >> > > > > > > (2.x) than aligning all the stuff to the spec version. For
> >> the
> >> > > > > > > component libs it is even more important to have that
> >> degree of
> >> > > > > > > freedom for counting up a minor number whenever there is
> >> an API
> >> > change
> >> > > > > > > and let the minor number unchanged for a bug fix release.
> >> > > > > > > MyFaces is getting more and more important. Also for tool
> >> vendors.
> >> > So
> >> > > > > > > there will be more and more people and stuff out there
> >> who/that
> >> > relies
> >> > > > > > > on our APIs. We should be oblivious to this responsibility.
> >> > > > > > >
> >> > > > > > > Sorry, but this is my binding
> >> > > > > > > -1 veto
> >> > > > > > > on having 1.2.x for our next spec 1.2 implementation as
> >> long as
> >> > the
> >> > > > > > > only reason for having 1.2.x is a "cosmetic" reason only
> >> to help
> >> > > > > > > people not being confused.
> >> > > > > > > Perhaps I missed something. If so, please explain to me
> >> what is a
> >> > > > > > > proper technical or organizational or consequential reason
> >> for
> >> > having
> >> > > > > > > 1.2.x as version for our next major (sic!) release.
> >> > > > > > >
> >> > > > > > >
> >> > > > > > > Thanks,
> >> > > > > > > Manfred
> >> > > > > > >
> >> > > > > > >
> >> > > > > > >
> >> > > > > > >
> >> > > > > > > On 5/18/07, Kito D. Mann <[EMAIL PROTECTED]> wrote:
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > +1 for 1.2
> >> > > > > > > >
> >> > > > > > > > -1 for 2.0
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > Using a " 2.0" version is going to confuse people.
> >> > > > > > > >
> >> > > > > > > >
> >> > > >
> >> > ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
> >> > > > > > > > Kito D. Mann - Author, JavaServer Faces in Action
> >> > > > > > > > http://www.JSFCentral.com - JavaServer Faces FAQ, news,
> >> and info
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > * Sign up for the JSF Central newsletter!
> >> > > > > > > > http://oi.vresp.com/?fid=ac048d0e17 *
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > From: Grant Smith [mailto:[EMAIL PROTECTED]
> >> > > > > > > > Sent: Friday, May 18, 2007 1:16 PM
> >> > > > > > > > To: MyFaces Development
> >> > > > > > > > Subject: Re: MyFaces 2.0.0 (was Re: Tomahawk 1.1.5 release
> >> > plans?)
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > +1 for 1.2
> >> > > > > > > > -1 for 2.0
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > On 5/18/07, Mathias Brökelmann <
> >> [EMAIL PROTECTED]>
> >> > wrote:
> >> > > > > > > >
> >> > > > > > > > +1 for 1.2
> >> > > > > > > >
> >> > > > > > > > 2007/5/18, Matthias Wessendorf < [EMAIL PROTECTED] >:
> >> > > > > > > > > So,
> >> > > > > > > > >
> >> > > > > > > > > any interest in making this to 2.0.0 ?
> >> > > > > > > > >
> >> > > > > > > > > -Matthias
> >> > > > > > > > >
> >> > > > > > > > > On 2/23/07, Manfred Geiler <[EMAIL PROTECTED]>
> >> wrote:
> >> > > > > > > > > ...
> >> > > > > > > > > > I am
> >> > > > > > > > > > +1 for Paul's suggestion:
> >> > > > > > > > > > JSF 1.1 -> MyFaces 1.x
> >> > > > > > > > > > JSF 1.2 -> MyFaces 2.x
> >> > > > > > > > > >
> >> > > > > > > > > > and I am
> >> > > > > > > > > > +1 for JSF 2.0 (or JSF6 or whatever) -> MyFaces 3.x
> >> > > > > > > > >
> >> > > > > > > > > > --Manfred
> >> > > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > --
> >> > > > > > > > Mathias
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > >
> >> > > > > > > > --
> >> > > > > > > > Grant Smith
> >> > > > > > >
> >> > > > > > >
> >> > > > > > > --
> >> > > > > > > http://www.irian.at
> >> > > > > > > Your JSF powerhouse - JSF Consulting,
> >> > > > > > > Development and Courses in English and
> >> > > > > > > German
> >> > > > > > >
> >> > > > > > > Professional Support for Apache MyFaces
> >> > > > > > >
> >> > > > > >
> >> > > > >
> >> > > >
> >> > > >
> >> > >
> >> >
> >> >
> >>
> >>
> >> --
> >>
> >> http://www.irian.at
> >>
> >> Your JSF powerhouse -
> >> JSF Consulting, Development and
> >> Courses in English and German
> >>
> >> Professional Support for Apache MyFaces
> >>
> >
> >
>
>
------------------------------------------------------------------------------
Notice: This e-mail message, together with any attachments, contains
information of Merck & Co., Inc. (One Merck Drive, Whitehouse Station,
New Jersey, USA 08889), and/or its affiliates (which may be known
outside the United States as Merck Frosst, Merck Sharp & Dohme or MSD
and in Japan, as Banyu - direct contact information for affiliates is
available at http://www.merck.com/contact/contacts.html) that may be
confidential, proprietary copyrighted and/or legally privileged. It is
intended solely for the use of the individual or entity named on this
message. If you are not the intended recipient, and have received this
message in error, please notify us immediately by reply e-mail and then
delete it from your system.
------------------------------------------------------------------------------
