Understood, no problem. As I see it, a little bit of both is needed. If
no one points out what really is wrong with Spring, no one seems to
realize it and is quite complacent with it's shortcomings. That being
said, I agree negative campaigning just for the sake of stirring trouble
Piotr Steininger wrote:
I am still a bit on the fence about campaigning against Spring. I
think there is far more to be gained from promoting CDI, Weld
Extension, Seam 3 and the EE 6 profile (with positive messages), than
jumping into a debate with gung-ho Spring believers. From what I saw
in my consulting engagements people who used Spring for a while are
stuck in a specific mindset of application development. For them
anything beyond singleton bean wiring on app startup does not exist or
is too mysterious or scary. Yes, it does sound funny, but I once had
the same mindset, but I saw the light ;). I converted to Seam thanks
to so many examples and great books like Dan's Seam in Action.
So if our goal is to entice the "not-yet-lost" souls, then we have to
send more messages that are pro-Weld, and not necessarily against
Spring. How to do it? In a few ways. First, excellent and exhaustive
documentation. I think Gavin, Pete, Dan and the community has done an
awesome job in that arena, and have set a golden standard. Second -
example projects. I think nothing shows the prowess of a technology
stack than a real proof that it works, works well, integrates with
other key technologies (JSF2, JPA, BPM, etc), which are part of the
Enterprise application ecosystem. These examples though need to go
above and beyond the simple ones already out there (we need the simple
ones too) - something along the lines of the Photo Album for
RichFaces. Third, tutorials, articles and blog posts detailing the
aforementioned examples - in another words - creating the hype and
excitement about the CDI's programming model, and the ease of creating
powerful apps in an elegant and concise way.
But at this point, there are still a number of important pieces
missing, like the Weld extensions and Seam 3 modules, which provide
integration with key technologies for business apps (i.e. BPM or
Security) or complement these technologies. These are needed to create
the example projects, which are essential as the meat of the article
to be written about Weld and Seam.
My key goal is to build an enthusiastic community around CDI, rather
than go after Spring. I think Gavin has a better handle on disproving
any allegation against CDI or anything bogus about Spring. I simply
don't have the technical know-how, just an implementer experience.
On Thu, Dec 17, 2009 at 3:39 PM, Reza Rahman <reza_rahman(a)lycos.com
Please consider posting your comments on InfoQ...it would help a
greater number of developers see past the Spring FUD...
Piotr Steininger wrote:
One thing to add to your comment is that they have done
nothing to Spring Webflow, which on the cover, promises the
scopes/context that Seam 2 delivered (primarily the
conversation scope). I had the misfortune of being forced to
use the Spring stack on one of my engagements, after having
great success with Seam on another project. I can't explain
how harrowing of an effort it was to get Webflow working even
remotely like Seam (in Webflow all conversation scoped beans
had to be manually defined in the xml). And the docs - yes
there are some, but extremely minuscule for JSF integration
and absolutely NO mention of necessary JSF 1.2 configuration,
which I had to figure out on my own.
I also had to make it work with Spring Security - this one is
clunky and difficult to set up, no to mention buggy. Yes, the
security was bumped to 3.0.0 and aligned with JSR-250, but
where in the world are the docs?! There ar en't any for 3.0,
and the ones for 2.x aren't any better than the ones for Webflow.
As a software architect who often gets to decide the
technology stack (but sometimes gets overruled :( ), I look
for a complete /ecosystem/ for the application I'm designing,
not just individual pieces with big promises. I need stuff
that works, works well, and works together. Not only that, I
look for something simple, maintainable and maintained by a
strong community. And my measure of strong community is the
quality of documentation and ac cess to knowledgeable
resources, without support contracts.
This is where Spring falls entire short of its promises, yet
again. So my time and money are with CDI and Seam.
Thanks for your advocacy and hard work!
On Thu, Dec 17, 2009 at 2:35 PM, Gavin King
weld-dev mailing list
weld-dev mailing list