<NOT speaking for IBM>

summary: McClanahan should clearly state *in some major publication*

* that JSF does/will not "replace Struts"

* how JSF and Struts will likely tend to specialize, in future

* how probable specializations will complement (and compete) in
  webapp development

I.e. pretty much what he has already said in this list, but much more
visibly.

details:

Craig R. McClanahan Sat, 20 Mar 2004 20:57:04 -0800 (rearranged)
> There is going to be tremendous support for JSF in the industry;
> fortunately, we can continue to maintain and enhance Struts without
> having to give that up (thanks to the integration library). Instead,
> we can embrace it

The problem, as I see it, is how to make the industry understand that
JSF will also embrace Struts? (and not in the sense of "embrace and
extend" :-) More below, esp re SFIL.

> My personal vision is that Struts developers will focus their energy
> on the controller and model tiers, leveraging the existence of
> standard (and not) technologies in the view tier.

Personally I agree strongly, and FWIW have advocated something very
similar (i.e. JSF both for view AND as a "M$-killer" for Model1 apps)
in local fora, e.g.

http://ns.cnconsulting.com/pipermail/juglist_trijug.org/2003q4/001106.html

ISTMT this also would also make a lotta sense for JSF, since (again,
it seems to _me_--my bosses may disagree) 

* no one framework is ever gonna do all of Java MVC web application
  development/execution the way that most IT folks want to do it
  (since "most folks" can never agree on much of anything 
  specific :-)

* no one framework is ever gonna have all the resources/mindshare to
  do all of Java MVC web application development/execution "right"
  (presuming "right" could be agreed on), therefore specialization
  makes sense

* MVC is a natural partition for such specialization

Unfortunately

* The JSF community seems to be putting out a competing message, not a
  complementary one: "JSF will replace Struts," or even "JSF is a
  better Struts."

� E.g. Geary said, flat out (not only is it in my notes, I believe
  it was verbatim in a slide), "Is JSF a replacement for Struts? Yes!"

  I challenged him, saying that while JSF 1.0 (with Tiles) can do
  pretty much everything Struts 1.1 can, Struts 2.0 seemed to be
  focused on doing things (e.g. struts-chain) that did not seem to be
  in the JSF plan. At which point he backed off, but continued to
  suggest that JSF should be favored for new-project development. (To
  his credit, Geary also made clear that "JSF and Tiles is a sweet
  combination.")

� One also hears that JCP is "more standard" than the Apache
  process, thus a more better target for development orgs. (Typically
  the Apache process is also deprecated by association with the
  unfortunate Struts 1.0--1.1 delays.) A popular variation asserts
  that JSF will eventually become part of the J2EE spec, while Struts
  never will.

* The "JSF replaces Struts" line has traction. I have heard it from
  consultants (and not just Geary), ISVs, and from ... highly-placed
  persons who I believe should know better :-(

* The "JSF replaces Struts" line has practical impact (which demands
  a substantial, visible response--more on that farther below)

� Development organizations have limited budgets. Managers of
  development orgs always want to "pick _the_ winner" (not just "a
  winner") if they can. There are of course a lotta webapps still to
  be written, and still a LOT of Model1 and "Model1.5" webapps out
  there, many of which folks wanna make more MVC. I suspect managers
  of their development groups will be most receptive to the "JSF (and
  not Struts) for new project development" line.

� Java tool developers face an esp crowded field of Java MVC web
  apps. We are gonna _hafta_ tool JSF, and we want to--it's nicely
  designed, and we wanna target the {Model1, "departmental developer",
  SMB, ASP.NET} space. But when managers of Java tool developers hear
  that "JSF will bury Struts," and hear about their budgets, they are
  gonna wanna say things like, "going forward we expect to actively
  tool JSF and to sunset Struts."

  Note that while I expect tool adoption/quality to be crucial for JSF
  (which very much seems "built to tool"), I do not consider it quite
  so important for Struts. That being said, good tools help, and I am
  very proud of my group's Struts tools, such as our web diagram
  editor. (FWIW I expect to be equally proud of our JSF tools in the
  very near future, and to continue to improve and extend our Struts
  support.)

So ... what to do about this? For starters, "we" can advocate that

* JSF is NOT gonna make Struts obsolete

* JSF AND Struts {is, will be} a sweet combination

but unfortunately I suspect that will not be enough: something's gotta
come "from the top," by which I mean (not entirely in jest)
McClanahan.

Steve Raeburn Sat, 20 Mar 2004 11:40:45 -0800 (rearranged)
>> As the creator of Struts and spec lead for JSF, I think Craig is in
>> a unique position to understand where all this is going. I take the
>> fact that he has accepted the role of Chair of the newly formed
>> Apache Struts PMC as a sign that he believes Struts has a strong
>> future and that he's willing to help its continuing evolution and
>> growth.

True that ...

>> Craig has previously stated that there is still a role for Struts
>> to play.

... and that--note however that he has done so (IIRC) "in the Struts
space," e.g. this list. Where I slightly disagree with Raeburn is his
claim that McClanahan has adequately spoken with, to paraphrase,
"deeds, not words":

>> He has demonstrated by developing a means to integrate the two and
>> making it available in the Struts distribution. See Struts-Faces in
>> the Struts contrib directory.

SFIL is unfortunately open to another interpretation, which I hear all
too often: that it is not so much an "integration library" as a
*migration* one. Unfortunately this interpretation resonates with an
allegation I have heard (I suspect I am not alone, but YMMV) far too
often from JSF advocates: that McClanahan is at best acceding to, and
at worst conniving in, the transition from Struts to JSF.

The former position seems to be Geary's. He said <from notes>"Craig
has a lot invested in Struts"</from notes>, implying <from memory>that
McClanahan is just unwilling to just pull the plug on his Struts
involvement</from memory>, which Geary clearly claims to be
obsolete (at least, Real Soon Now--he does give props to its
stability).

The latter position is more troubling, not because I know of any
evidence for it, but because I have been hearing it for over a year
now. Variations include:

* "Craig says one thing in the Struts space, and another thing in the
  JSF space." Whenever I hear this I ask for a pointer, and have never
  received one.

* "Craig works for Sun, and Sun wants JSF, not Struts." My rejoinder
  is, then why do they continue to allow/tolerate/support his
  continued work on/for Struts?

Note that such responses are usually met with nothing more than a smug
smile. I suppose this is due to, less charitably,

* the weakness of the position (their's, of course--not mine !-)

* the human desire to believe in the existence, and one's personal
  possession, of esoteric knowledge (cf the continuing popularity of
  the Middle Eastern monotheisms)

and more charitably

* the absence of the opposing view ("JSF + Struts, not JSF > Struts")
  outside the Struts space

esp

* the absence of the opposing view in "Faces space"

(and, umm, my position "in the food chain," or lack thereof :-)
Perhaps I'm missing something, but a (all-too-) quick google and
search of

http://forum.java.sun.com/forum.jsp?forum=427

(BTW another thing I prefer about Apache projects: the list/archive
style) found no prominent statements of the "JSF + Struts" position
outside the Struts space, certainly nothing to equal McClanahan's
statements on this list.

Therefore ISTMT

+ McClanahan could relatively easily state the "JSF + Struts" position
  in some major publication, either in the Java space (e.g. JDJ,
  JavaWorld) or, better yet, outside/above (e.g. eWeek,
  ComputerWorld). I say "easily" because

� IMHO McClanahan is widely respected, not only in both the JSF and
  Struts communities, but the wider Java community. When he talks (esp
  on this subject, which IIRC has already generated fair buzz), they
  will publish :-)

� the required verbiage could be easily generated. If an interview,
  all McClanahan need do is talk; if a more standard sort of article,
  the point could be relatively easily made by recycling his posts to
  this list. (To "spike" the "if it's so easy, why don't YOU do it"
  response, it's because I have too many bugs on my code right now :-)
  More importantly, who would listen? The whole point of this exercise
  is VISIBILITY.)

  and I specify "publication" because

� a talk typically doesn't provide something easily {quoted, linked
  to}, and I believe serialization is important in this matter. (I.e.
  the next time I hear one of the arguments rebutted above, I wanna be
  able to say, "go HERE and read THIS" :-)

� something that exists in hardcopy, as well as online, is more
  likely to reach managers (just my guess, YMMV)

  (though the individual pubs mentioned above are just OTTOMH).

+ McClanahan should state the "JSF + Struts" position in some major
  publication. The normative form is IMHO warranted by

� his unique and unquestioned position relative to both JSF and
  Struts

� the ease with which the action could be accomplished (as
  previously asserted)

� the practical effect such a statement could have (viz the effects
  of the "JSF replaces Struts" meme, farther above)

</NOT speaking for IBM>


---------------------------------------------------------------------
To unsubscribe, e-mail: [EMAIL PROTECTED]
For additional commands, e-mail: [EMAIL PROTECTED]

Reply via email to