Skip to main content
[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [rt-pmc] Re: Swordfish review slides
Title: Re: [rt-pmc] Re: Swordfish review slides
Bjorn and fellow PMC members,

Now I have reviewed the document and had a number of conversations with Oliver who is the project lead for Swordfish.

The project team was not aware that a 1.0 release inevitably implies graduation as well — the development process documentation doesn’t seem to be 100% clear about this. If the PMC had had a chance to review the draft docuware before Bjorn did, they would surely have noticed that and advised the project accordingly.

There’s no doubt that the project is definitely not ready for a graduation review at his point. My question is more whether the project should have a 0.x release or not. On one hand, there is no user community and no diverse developer community yet, which is certainly to some extent due to the poor communication job the project did in the past. But on the other hand, they’ve learned their lessons and the current status of the framework itself should be good enough for a 0.x release.

Essentially, we have a chicken/egg problem here. There can’t be users from other projects as the whole RT area is new to Eclipse (I consider Riena and RAP not as pure RT projects, but Smila and EclipseLink). A release 0.x would provide the necessary momentum for adoption in other projects. So I’m in favor of approving a 0.x release.

Ricco


Am 03.03.09 17:42 schrieb "Ricco Deutscher" unter <ricco.deutscher@xxxxxxxxx>:

Bjorn,

thank you very much for your comments regarding the Swordfish release review docuware. The slides you’ve reviewed, however, were not meant to be final — Oliver sent them to EMO as a draft on Anne’s request, and we apologize if they haven’t been marked clearly as such.

Nevertheless, we understand your points and we will come back to you within the next 1-2 days.

Best regards,
Ricco


Am 03.03.09 01:16 schrieb "Bjorn Freeman-Benson" unter <bjorn.freeman-benson@xxxxxxxxxxx>:

RT PMC members,
The Swordfish project has sent in some slides for a 1.0 release review. In reviewing these slides, I note the following items which make me wonder how involved the PMC has been in preparing this project. Specifically, two issues:

  • The Larger Eclipse Community: the purpose of these reviews is to explain the project and it's status to the larger Eclipse community. The review is not just a checklist of hoops to jump through. For example:

    • Schedule: the purpose of listing the schedule is to demonstrate to the larger community the project's ability to set a schedule and to meet the schedule. Thus the schedule page in the docuware needs to list the actual planned milestones and planned dates and actual dates. Not just first and last milestones: all of them, with details.
    • Standards: the purpose of listing the standards is to explain to the community what standards the project is implementing and tracking. Thus the statement "strives to be standards compliant" is not good enough.
    • ...etc...

  • Maturity for a 1.0 Release: it's hard for me to see the project as ready for a 1.0 release given that:
    • All the API is provisional. That could be ok except that there's no plan for hardening the API and thus one wonders whether proper care has been put into the API design. The purpose of "provisional API" is not to allow projects to avoid spending the effort to create good APIs - rather, the purpose is to allow an "almost completed" API to be road tested with actual clients before be finalized.
    • The project only supports one messaging engine. It's hard to claim that something is a framework if it doesn't have at least two, preferably three uses/users.
  • I also note that the PMC has not yet approved this release. The PMC is supposed to approve releases before they are sent to the EMO.

I'm inclined to say "no" to this release review unless you all on the PMC can explain why I'm wrong here. Thanks.

- Bjorn

Back to the top