Skip to main content

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index] [List Home]
Re: [rt-pmc] EclipseRT white paper draft 2

Douglas Clarke wrote:
Ian,
I do not believe this paper in its current form does much to clarify what RT is or how to get started with it. The scope touches too many technologies trying to tie them together without any real explanation in many cases. From an EclipseLink perspective there are no concrete usage examples illustrating to readers not familiar with the project what it does and how they might use it.
Sorry you feel this way. It would have been nice to have some of this feedback a bit earlier in the process. I think you are expecting too much out of this white paper. We are just trying to provide a high level of what is possible not a detailed explanation.
Some suggestions: 1. Reduce the quantity of Eclipse projects/technologies in this paper so that more specific usage can be explained.
Do you have suggestions of what projects can be dropped? Given #3 it would seem Swordfish but anything else?
2. Avoid vague statements that do little to explain what the technologies are or how they are used together:

    * SOA: - "The platform nature of EclipseRT means that SOAs built
      on Swordfish can also use other EclipseRT components such as
      EclipseLink and BIRT for the usual enterprise application needs.

I will try to clarify.
3. The SOA Swordfish section is interesting but something I would consider leaving out. If the SOA angle is to remain we should highlight how some other projects can be used in this space independently of Swordfish as well as with Swordfish. As one example EclipseLink SDO offers the reference implementation of Service Data Objects allowing services to easily pass structured data between them crossing service and programming language boundaries. This infrastructure is leveraged with the Swordfish project but could be used in any SOA development efforts as I am sure other common technologies in the covered projects could be.
I beg to disagree. SOA is very topical and very relevant to the potential audience.
4. Some statements that I am not sure I understand:

    * At the end of the Platforms section there is the statement -
      "The management services available in EclipseRT, such as p2,
      EclipseLink, and Swordfish gives IT the ability to deploy and
      then maintain applications built on the platform." I am unsure
      what is being implied here about available management services
      of these projects
    * In the web Applications sections - "And, of course, Jetty is
      tightly integrated with the rest of EclipseRT, such as the
      Eclipse Rich Ajax Platform, EclipseLink and Equinox.".  Is this
      true? I believe Jetty can run as bundles on top of Equinox but I
      am unsure of the other tight integrations and the statement
      could be misinterpreted to indicate required dependencies.


I will try to clarify.
5. Adding EclipseLink specific usage context. I believe the common infrastructure projects that span Desktop, Web, SOA, Enterprise, ... usage may need to be given more contextual usage and explanation so that they do not stick out as just name droppings in the paper. BIRT, ECF, EclipseLink, can be used in all of these architectures which makes it tough to explain what it is and how a customer can use it. I would recommend calling out a common infrastructure section where some of these can be highlighted.
I think it would be great if someone wanted to created a summary of all the EclipseRT projects. Unfortunately that is not the intention of this white paper.


Back to the top