Hi all,
Long story short, I think the Attic is appropriate.
I do like Tiles. It does work well. But the technologies it relies on
are unlikely to evolve in the coming years, requiring little maintenance.
There is more to a project than its code though. For open source
projects, a healthy, active community is important. And open source or
not, a collection of real world use cases proves it is useful.
Personally I've not had the opportunity to use tiles for a few years
now. I'm doing the templating in the browser, and that excludes Java.
Should we port tiles to another language - TypeScript perhaps?
Tiles could do more. I never managed to complete and merge my stuff
about decoupling Tiles from servlet-centric stuff, and using DI for
configuring it. But nowadays, these seem hardly relevant anymore.
Further improvements on tiles should, IMHO, focus on getting the
benefits of future technologies: Java 9 (lambdas and modules), reactive
streams/RxJava/Reactor, and keep it agnostic of the template engine.
Also it should move away from the HTML templating, unless we port it to
the javascript world. It would be most useful for code generation
though. Perhaps Vert.x is a good target. Or the "Infrastucture As Code"
world (Kubernetes - CloudFormation - Terraform...)
I can imagine future developments to Tiles, but it implies a complete
reengineering. I have plenty of ideas, I'd love contribute and provide
direction, but at the end of the day, my availability is too limited.
For sure the tiles community is healthy, but hardly active. The spring
community has chosen to focus on thymeleaf and deprecated tiles for all
practical purposes. Struts2 has never really achieved traction. Users
are only waking up for reporting bugs - most of those are servlet
configuration issues unrelated to Tiles.
We've been talking about the attic for some time now. I think the time
has come.
Hope this helps,
Nick
On 2018-03-20 08:31, Brett Ryan wrote:
Tiles is still my goto for templating, and AFAIK it’s the only view composition
framework that allows a standard component based method of composing views from
smaller parts.
That said, tiles has not really required imho more development as it works
really well the way it presently stands.
Unfortunately pivotal have in my eyes been a contributor to the demise of tiles
take-up by removing all examples of tiles use from the spring/mvc documentation.
Could an effort to revive tiles awareness be some investment in example usage
that is published within and contributed back to http://spring.io? Other
awareness strategies?
On 20 Mar 2018, at 23:24, mck <m...@apache.org> wrote:
All,
I do believe that it's time that the project rotates the Chair/VP position.
It is normal that the Chair position rotates every year or so, and is sign of a
healthy project.
Nicolas, would you be interested?
Otherwise, it may be a good time again to question whether Tiles should be
moved to the Attic.
Given that there exists a git mirror on GitHub I see even less reason today to
avoid the Attic. Code on github is so easily forked and extended. And with any
revival in activity the project can be brought back out of the Attic.
regards,
Mick
