DevExpress Newsletter 38: Message from the CTO

19 November 2010

This particular Message follows on some thoughts I’ve been having this past month or so:

Hindsight, easy; foresight, hard

There have been several instances recently of the truism that it's only hindsight that is 20-20. I'll kick off with our contribution: back last December, when we were laying out the plans for 2010, it seemed entirely feasible that we would get out the second major release of the year in October, maybe November at a pinch. And yet here we are in the latter half of November, having only just released the beta. Yes, certainly we were optimistic that the biggest release of DXperience we've ever done would arrive on time, but who could fault us for having ambitious plans and just going for it?

For this release, we decided to approach development slightly differently: a feature had to be complete before moving on to the next. Yes, we'd done this before, but this time "complete" meant including documentation, demos, designers, and all the other ancillary things that wrap a feature. So we could talk about completed features earlier than we've ever done in a release cycle. All benefits, no downsides, right? Except that it is human nature to make plans for one's own apps assuming that new feature X would be available in October: it was in the roadmap, remember?.

Another example from the opposite direction: rumors about Silverlight have abounded since PDC. I won't rehash them here, but in essence you can assume there are 3 camps: Silverlight will continue just fine the way it has for the past couple of years; Silverlight has been repositioned to just line-of-business apps and Windows Phone 7; or Silverlight's death warrant has been all but signed. (Of course, this is not a Holy Trinity; the truth, if anyone except Microsoft knows for sure, is somewhere on a continuum between them all.) It is human nature, then, to worry about starting to write a Silverlight app: whence Silverlight? is niggling at your mind.

These two scenarios are an example of why foresight is so hard or impossible: we are just terrible at predicting the future. Instead we should discount rumors and instead concentrate on the facts when deciding on a course of action. Facts that we know right now. Yes, we may be wrong, in which case, we fix the issues at a later stage; but we may be all right, in which case we move on to the next decision to make, the next project to do. (Sounds almost Agile, no?)

Yes, it would be nice to use feature X that DevExpress has announced in my app, but, you know, I'll get out this release and add it into the next one, once they really have provided said feature. Until then I shall view it as vaporware. Yes, I may spend 6 months writing a Silverlight app, worried about its future, but then I'll be able to laugh at all those who dithered on the sidelines once my app's released and successful.

Unless you're into gambling on stocks, shade the brightness of your foresight. Don’t let it illuminate everything you do.

Having said all that, we’re certainly going to be burnishing our crystal ball and peering into its foggy depths at our summit from 29th November to 3rd December. Remember to register and tune in: we have two webinars during that time, during which we’ll be talking about what we’re planning for 2011. Of course, the famous DevExpress RoadMap will also be published later as a result.

7 comment(s)
AutoTRAX

Does this mean you will return to do some work with

Winforms such as right to left*, improvements to the

Ribbon etc. Things used in real world rich clients?

* Yes not all the world uses left to right and its a big

market.

19 November, 2010
Alex Batchinski

Looking forward for the new Silverlight and MVC controls!

19 November, 2010
Felipe R Machado

I find it strange that Silverlight had never that much of an appeal with me until the recent announcements of its "death" by Microsoft analysts... With jQuery and other javascript frameworks and HTML5 on the rise, suddenly there was much less need for "web-rule-breaking" frameworks like Flash and Silverlight to create rich-client web applications. But with the "relegation" of Silverlight to Windows Phone 7 I really saw a great business opportunity and the biggest justification to devote many ours and resources to it (Ever tried to write Objective-C applications for the iPhone?) There is a huge market for Windows Phone 7 mobile applications, and Silverlight will be the Windows Phone 7 framework! I think DevExpress must urgently focus on providing components and utilities aimed at Windows Phone 7 (the interface paradigm is very different, re-engineering may be needed, etc), but this will certainly capture a lot of developers out there. Maybe a Windows Phone 7 suite (Silverlight controls aimed at that platform) with special pricing and great attention to the beauty of the interface and touch-screen interaction, I can't really know better than you guys, but I suggest that you do something to grab the attention of the many firms and people wishing to develop for Windows Phone 7 phones.

19 November, 2010
Michael Thuma

That's why is in beta, no problem at all. We are all very bad in guessing the future ... - What a poor kind of brain - it only works backward.  What counts is what works now ... this is the only thing to decide upon.

19 November, 2010
ayodahunsi

I Agree 100% with Felipe!

I just got my Windows Phone7, and of course the first thing that came to mind is how to develop apps for it.. and scouting the web, I saw one of DevExp competitors (will not mention names here), already has RAD controls specific for Windows Phone 7. This competitor even has a draw for a windows 7 phone+ their controls.

I just really hope DevExpress see the benefit commit resources to Windows 7 development.

19 November, 2010
David Brennan

Winforms aren't alone in not having right to left support from DevExpress, VCL doesn't either. There are many things I would like to see but RTL for VCL is right up there!

21 November, 2010
Jaime Alvarez [VOLUNDAT]

I think that right would then pay subscriptions for two volumes and their updates.

For now we would be paying one-year subscriptions for a single volume.

all plans can go wrong .... it's also true that effective teams provide results, the other simple explanations

DevExpress is one of the best components on the market, I have about 10 years with you, but sad to see that their competitors have more commitment to the dates offered.

Jaime Alvarez

22 November, 2010

Please login or register to post comments.