Why isn't this effort doomed to the same fate as Ozone 8?
Primarily because development is being done in a true open source fashion. In the past, previous systems were largely the work of a single contractor team that developed internally and then made the code public. The customer, contract, and contractor interactions made this a huge bureaucratic obstacle to transparency and accountability.
In addition, previous efforts could be characterized as "distributed requirements gathering, centralized development". The model is now closer to "distributed development, submit your requirements as code".
Explain "development is being done in a true open source fashion"
All development is being done in the open. The master repositories for code, design, and tasks are in the Ozone Developers organization on Github. We're using open source tools and standards whenever possible.
We are already prototyping a distributed development effort. There are 4 core development teams across multiple government organizations contributing to the development. This forces more discipline on standards and interfaces as well as decoupled design of the software, which was the source of a lot of technical debt in previous incarnations.
How do I run the prototype?
As of July 2nd, 2014, there's no single prototype for the Ozone Platform. The various projects on Github can be run individually, and should have notes on how to do so within them.
When will there be a prototype?
Keep in mind that while OZP had seeds that go back to December 2013, the big push really only kicked off in May 2014.
We're aiming for a rough prototype to come together by September. The goal of this prototype is to get initial feedback from developers on the API changes and from users from the subtle but vast differences in user experience that comes from having multiple windows (e.g. where do you put the launcher? is a launcher even necessary?).
What pieces can I poke today?
What about project X in the ozone-developers organization?
There's dead code in the project list right now. Anything prefixed with "owf8" is dead. The "owf7-refactor" repo is a community contribution of performance enhancements for OWF 7 that may be released as part of 7.Final. Otherwise, look at the Readme and harass folks if one doesn't exist.
But what about...?
Everything is still in flux, and the core team is still getting it's rhythm going. More info will follow
How can I contribute?
It's open source. Pick a task, fork the repo, implement it, and send us a pull request.
For more formal arrangements, contact the Applications Mall PMO.
Can you explain a few things about the new intended architecture?
1. What is the purpose of the controller projects and how do they relate to the webtops?
2. Is the hub-ui project intended to be the launcher from which users launch their application/webtops?
3. How will authentication work in OZP? I ran the the sample webtops and there was no login or authentication. In OWF you log into the one webtop to access your widgets.
4. Do you have any other documentation or presentations that you can share that would explain the vision for OZP and the intended architecture?
5. You mentioned contacting the "Applications Mall PMO" for more formal contribution arrangements. Do you have contact information?
On Wednesday, July 2, 2014 8:16:44 AM UTC-4, Jason Wagner wrote:
--
You received this message because you are subscribed to the Google Groups "ozoneplatform-users" group.
To unsubscribe from this group and stop receiving emails from it, send an email to ozoneplatform-u...@googlegroups.com.
For more options, visit https://groups.google.com/d/optout.
How far away is OWF from leasing an inter-widget communication (IWC) that works across multiple browser windows and tabs?
To unsubscribe from this group and stop receiving emails from it, send an email to ozoneplatform-users+unsub...@googlegroups.com.