Friday, March 14, 2008

Target Communication Framework (TCF)

I'm speaking at EclipseCon 2008A year ago at EclipseCon, I've been asked a lot whether there wasn't a lightweight Open Source agent for resource-constrained communications with remote systems planned or available.

And now it's here - and much more! The Target Communication Framework (TCF) is not only an extendable agent, it's a whole protocol framework that has the potential to make target communications a lot easier. TCF is a new incubating component of the Eclipse Target Management Project, and its unique benefits include
  • Transport-independent multiplexing of multiple services over a single protocol
  • Ability to transparently add 3rd party value-adding services in the communication chain
  • Auto-discovery and single setup of all target services.

As you can see, all communication links can share the same protocol, simplifying connection setup and allowing transparent tunnelling without unnecessary protocol conversions. In fact, any 3rd party vendor can contribute a value-add server to do transport conversion from a standard TCP/IP channel into custom channels such as JTAG or even proprietary hardware connections: all services can immediately route through the new transport and take immediate benefit of the value-add.

Besides Wind River and Eclipse, the Power.org association is also actively working on TCF as a potential emerging standard. If you work in the embedded space, consider joining the effort now to discuss your unique needs!

If you want to know more, join me at EclipseCon for the TM Tutorial on Monday at 1:30, the TM Short Talk on Thursday at 11:10 or the DSDP BOF on tuesday at 7:30pm - or simply drop us an E-mail or browse the TCF online documentation.

Friday, March 07, 2008

Just one more word about E4

There's been some hot discussions around the recent announcement of "E4", a new component as part of the Eclipse Platform Project to host code and infrastructure for prototyping ideas towards planning Eclipse 4.0.

I've been both excited and surprised - like others - about this announcement. Excited because some great people finally get together doing some real work for the future of the best Open Source Platform I know. Surprised because I had expected more visibility of those efforts in earlier stages. And despite an excellent background post by Ed Merks and a plead by zx to just have things coming and contribute, there's three "Why"s that I just can't push aside:
  • Why all this secrecy? Why wasn't there more ideas and discussions on the corresponding bugzilla plan item? A lot of people made suggestions there or were listening to what ideas others might have.
  • Why wasn't there a call for a workshop like the provisioning workshop last year? The P2 effort seems to run excellent, why not make a similar announcement again?
  • Why is it mostly Wind River people challenging the process? Don't the other strategic Eclipse members care about innovation and the future of Eclipse as we do, or are we just too young and passionate?
I don't want to over-analyze or over-discuss this. I'm truly thankful for the effort and energy that some of the most talented people are putting into some demos I definitely don't want to miss at EclipseCon. I can't wait to learn more about what's baking, and I'm burning to get involved into an effort that's destined to become the most innovative and powerful platform to come.

If you think that the Wind River guys are the bad guys just nagging, forgive our passion and probably misformed words. Give us a chance to get together - there's a lot of ideas and patches just waiting to get applied, and a strategy that just wants to drive innovation.