The state of things for XAF 8.2

As some of you have noticed already, we have recently had to make a few changes to our plans for XAF 8.2. One important reason for this is that we are busy working on important architectural modifications/features, mainly to support our Reusable Domain Model plans that Roman has blogged about a few times. Other reasons are related to the amount of support work we're doing, as we're seeing large numbers of suggestions and questions for XAF, and of course also a fair number of bug reports. Our aim is to incorporate many "small" suggestions and requests for minor improvements, since we want to make the basis of XAF as flexible and stable as possible.

There's no way we can provide a feature out of the box for every use case out there, but it is our goal to be able to provide a solution for the vast majority of problems, even if it involves some manual work. That's why your feedback is important to us and we're spending a fair amount of time evaluating it and acting on it. Thank you very much for being involved!

Of course a lot of support work is also being done to answer customer questions, about implementation details of XAF, approaches to certain solutions, and ever so often about basic information around the parts of your applications that we can't write for you, such as how to create working domain class hierarchies, where and how to implement business logic and so on. We are aware of this need for information, which we find to be much greater for XAF than for any other Developer Express product. Our new technical evangelist Gary and I have been busy working on an outline of content we want to publish on a variety of topics. Gary will soon be blogging about his experiences with XAF, and you will see technical articles appear as well (within a few weeks is my best guess at this time). Slightly longer term plans (talking a few months here) also involve publishing more video content, which is something that many have requested. As usual we're having to prioritize, but we'll pick up on additional ideas as soon as time allows.

The architectural work we're doing is a problem for our 8.2 plans in two ways: first, it consumes time, second, it means that features or changes which depend on base class implementation had to be put on hold for 8.2. The second reason is why our workflow plans are being delayed, for example, and a few other things as well. On the other hand we're building a solid basis for the implementation of these features, and I don't think I'm promising too much if I say that 8.3 will contain more than the usual amount of new things, partly due to this "preparative" work we're doing.

Finally let me say that in spite of all this the 8.2 release is not going to be boring by any means - I wouldn't want this post to be misunderstood that way. Of course the usual disclaimer applies - these are our current plans, and since we're working towards a specific deadline for the 8.2 release, it is possible that they will have to be changed. But if things go the way we think they will, 8.2 is going to have these new features (not an exhaustive list, just a few highlights):

• Ribbon support in Windows Forms
• A data analysis tool based on the XtraPivotGrid and XtraCharts
• IIS Medium Trust support
• ASPxScheduler support
• An abstracted data layer - use XAF without XPO - important update, see below

Keep the feedback coming!

Update to the "abstracted data layer" entry - There are two things that must be added to this. First, my language wasn't very good when I wrote the entry - it does actually sound like we're going to create a new data layer, which is not what we're doing. What I meant is that we're working on abstracting the existing data layer, in other words decoupling it from the rest of our architecture. Second, while the purpose of this work is of course that other data access technology than XPO can be used with XAF, the team tells me that we're not quite at the point where they would be comfortable with a recommendation to go ahead and start creating additional data layer implementations. Our current work will take away direct dependencies between XAF and XPO, but there are still functional areas like criteria support as well as certain assumptions about the way the data layer works, which might make it very hard to implement a compatible data layer that is not based on XPO. So up until the point where we create a new data layer ourselves and see that it is possible, we can't really call this a fully supported extensibility point. My apologies for any confusion in this regard.

Free DevExpress Products - Get Your Copy Today