about processes and engines


OpenWFE 1.7.1pre7 has just been made available for download.

It contains lots of bug fixes and performance enhancements.

We’re very close to a 1.7.1 release, probably next week.

The new webclient, based on the GWT (Google Web Toolkit) will be made available in the 1.7.2 release (watch for 1.7.2preX releases). The server side stuff is here, I’m now building the user interface. I’d love to implement a ‘droflo 2′ based on that toolkit as well, but, we’ll see.

A small recap on the new features (1.7.1) : decision tables, easier distribution of process segments across engines, a <forget> expression (still have to blog about it / document it), a <skip> expression (used within cursors and loops) and many other small things.

A bit of roadmap (1.7.2) : a new [g]webclient, a channel system for inter-process correlation/communication, failure strategies (rescuing flows upon break)…

Lots of things to blog about, but I first have to keep the documentation in sync with the features as well (the gap is very limited though). Blog posts are documentation too.

(I also have to release the latest version of Lukas’ OpenWFE IDE, which gets better every day)

Written by John Mettraux

June 22, 2006 at 9:36 pm

Posted in bpms, openwfe, workflow

3 Responses

Subscribe to comments with RSS.

  1. Hi John,

    what about the generic user? Will this feature be implemented in 1.7.1?

    Any something else: You mentioned the failure strategies. Yesterday, I’ve built a never ending loop because I forgot to increase my counter-variable. To solve this, I stopped the engine, removed the workitem manually and started the engine again. Is there a better solution?

    Best regards,


    June 23, 2006 at 6:09 am

  2. Hi Sebastian,

    maybe this discussion had better take place on the developers mailing list. But let’s continue it here.

    The generic user is not high on my priority list, I’m working on other things that are more relevant to me (and my customers). Let’s say it could make it into 1.7.2. My ideas about it are still a bit fuzzy, once I get convinced about the right way to do it and implement it, it could become a reality in a matter of hours.

    A never ending loop : nice, the failure strategy would ‘catch’ it (the runtime exception) and save the ‘dead’ workitem into some ‘dump’ file.
    The dead workitem could then be used to restart a flow with the values it contains, but manually.
    Of course, other failure strategy implementation may be used.

    Best regards,



    June 23, 2006 at 6:19 am

  3. Ok John, thanks. I will wait for 1.7.2 :-)


    June 23, 2006 at 9:11 am

Comments are closed.

%d bloggers like this: