And then, six years later…

15 March 2012

Way back on the Ides of March 2006, I started at DevExpress in my current position of CTO (“What? Stuck? No promotions since then?” Etc.). Six years to the day.

A lot of things have happened during that time but, rather than look back in a semi self-congratulatory way (which, after all, is boring to anyone but me), I want to look forward over the next few years – in essence the remit of what a CTO should do: identifying trends and technologies to watch and follow. To be honest, though, my predictions aren’t really going to impress anyone who already follows the tech news and blogs. And, yes, I dare say there’ll be some “well, if the CTO is saying *this*, why the heck are DevExpress doing *that*” moments.

Crystal ballFirst off, the big thing is the move away from desktops and laptops to tablets. We’re already seeing this trend in the retail space: Horace Dediu in Asymco, using some very conservative assumptions, recently estimated that sales of tablet units would overtake sales of traditional PCs in fall 2013, some eighteen months away. [When will tablets outsell traditional PCs?]. Already we are starting to see what’s become known as the Consumerization of IT, where workers, especially knowledge workers and upper echelons of management, want to use their own smartphones and tablets in their work. IT departments are putting into place the required infrastructure and applications (especially security-related) in order to cater to this trend. [Consumerization Of IT: Getting Beyond The Myths].

Second, we are starting to see a more pragmatic view to supporting all these new devices. No longer is it a 95% Windows world out there: the Mac is becoming more prominent, and of course these mobile devices all have different OSes (there’s iOS, Android, the forked Kindle Fire Android, and, soon, WinRT and WOA (Windows on ARM). Either your development team becomes proficient in many different OSes, IDEs, run-time environments, or you hire islands of expertise to cater for mobile, or you target the common denominator: the browser. And that means HTML5, CSS3, and JavaScript. [Will HTML5 replace native apps? It might: here's how to figure out when]. Yes, I know full HTML5 support is spotty at the moment, but the spec is due this time next year, JavaScript Harmony possibly at around the same time, and the browsers are being updated on a schedule faster than ours. And don’t forget the whole world of the hybrid app (HTML5/CSS3/JavaScript with some run-time libraries that access the features of the device).

Third, given that premise, I see a massive rise in the utilization of open-source software. Already, after maybe a year or so of meteoric rise, jQuery has become the de facto client-side library to use. It’s almost part of the JavaScript run-time, it’s so ubiquitous. The big thing that has still yet to consolidate though is the whole UI on the web. Yes, there are numerous open source (and even commercial) client-side UI libraries out there, all of them in the middle of a bonanza of development it seems, but no single one is dominant. Google have even started a very early draft of a specification for Web Components on the W3C site. [Web Components Explained]. I dare say given major funding from Microsoft, Google, Apple or any combination of the three, we’ll see one of the open-source UI libraries gain ascendancy, just as jQuery rose to its prominent position.

Fourth, since open-source libraries will start to dominate, I think we’ll be seeing a lot more work from vendors to enhance the development experience. In other words, sure, your controls are free and freely available, but you’re going to need some better tooling to write those web applications. Tools for designing and branding, for sophisticated data binding, for offline use, for debugging, for testing, for support, all those things that you forget need to be done when, in your enthusiasm for the free, you jump into the morass of client-side web development. Hand-crafting markup is tedious and pretty much equivalent to writing in assembly language. I’m not saying that vendors won’t be able to sell their own controls, as DevExpress continues to do, but that the development infrastructure for those controls will become more important.

Fifth, alongside the move to better web applications, the need for storing data in the cloud (and sharing it once it’s there) will become much more important. Not only the ability to store files in the cloud, but have web services providing data from cloud servers, to have and use a database in the cloud, etc, etc. Whether we’re talking about Amazon, Rackspace, Microsoft, or even Apple, the day of the cloud has yet to really reach us, but with the advent of more and more mobile always-on and always-connected devices, the cloud will become of strategic importance.

Sixth, I’ve said this before, and I’ll say it again: we shall continue to see the rise of the UX designer as being a primary part of any application. We consumers have been way spoiled with our devices and the UX on them and we shall continue to be. Our other applications: at work, or online, had better keep up. We want better ways of displaying and manipulating our data. We no longer want to be surprised by our interfaces, or have to read documentation on how to use them. The day of the grid is perhaps over, but there is no great contender interface to replace it just yet; it still has to be devised and invented.

Now, I’m not trying to say that any of this will come to pass in 2012. Of course not. Current platforms and development environments will be with us for a few more years yet. But I reckon that in, say, three or four years’ time (when I’ll be ten, as it were), perhaps with some of it in 2013, writing rich client-side applications targeting HTML5 will become as prevalent as writing native apps today.

Of course, DevExpress will continue to monitor these trends and others and make our plans accordingly, but rest assured we’re not going to suddenly abandon our current platforms. Indeed our philosophy is to strengthen them by providing tools, libraries, and controls that bridge what you know now and this brave new world.

Let’s see what develops.

11 comment(s)
richard morris

Congrats on the Milestone :)

15 March, 2012
José Enrique

Developers would like to say: "HTML5 is the way, one ring to rule them all"

Users and clients say: "I want a native ring that fits on my finger and looks pretty"

Gurus ingeniously argue on both trends.

And again developers don't admit "... I'm lost in a sea of doubts"

16 March, 2012

Nice article!  I can't wait to see DevExpress's tools in MonoDevelop :)

16 March, 2012
Martin Brekhof

>> I think we’ll be seeing a lot more work from vendors to

>> enhance the development experience

enter XAF, best argument ever to buy DevExpress components.

16 March, 2012
Julian Bucknall (DevExpress)

Richard: Thanks! The CTO shoes you left just about fit now.

Neal: You're talking native, I was talking markup :)

Martin: Exactly...

Cheers, Julian

16 March, 2012
Michael Thuma

we shall continue to see the rise of the UX designer as being a primary part of any application.

- Agreed. UX is about usability, especially.

I am following this whole bunch of technologies and evolution now for 8 years - 3 years in detail. Why.

Allow me to provide a very unspectacular example. OPR you can drill down, stylish in fashion of a 'newspaper'/'article'. This sounds weird but this what users want and what works mobile.

(Daily practice)

In 2003 I developed a Pack and Go variant of an OPR in Excel, merged data from several sources (40 cubes/queries) the report was interactive, allowed commenting and navigation. Never this beautiful, this was the possible using Excel only and heavy VS with Code generation for the interactive parts. One word on drill down -  There is not a lot required maybe one or two characteristics (dimensions) one added in the columns one added to the rows -  in practice this turned out to be enough..

(Lab experiment)

It was possible via pure web technologies too. The result was good but it required heavy java coding, excessive use of XSLT but the result has been an XML file with a view rendered defined by styles. Nothing big but it worked. The definition process was to heavy. (This was just in the labs). Information Builders provided something similar but I found it afterwards. In their world it was a solution for Logistics - Mobilized Controllers in  logistics.

Since these days I am observing the scenery and I think we have arrived.

Drivers are

a) Mobilization, phase 2. Eliminate the need to sit the whole day in front of the computer. The trend in Europe will be come mobilizing those who are controllers today. In modern trusts this is already daily practice. In the U.S. reality a lot more.

b) In Memory. This is lightning fast.

Finally the screen and the print-out seem to become married in the interactive document.

16 March, 2012
Peter Thorpe

Congrats on the Milestone.

The HTML 5 move is definately still rolling I moved more and more projects to web recently that were difficult in the past and find my self using your ASP.NET controls client side code more and more. Only really use postbacks/callbacks for data binding. Looking forward to more SVG use on the web to improve the UI's.

17 March, 2012
Peter Meier

Well done on the 6 year milestone.

Going by your post, it sounds like all your staff will soon be using tablets to compile software, write manuals, record videos etc through browser based applications. I wouldn't discount native applications and large screens just yet.

20 March, 2012
Bruno Cossi


Congrats on the milestone - at the risk of sounding melodramatic, my life (at least the professional one) is better thanks to DevExpress (perhaps a bit of a hyperbole but not a lot!).



21 March, 2012
Adam Russell

Julian, congrats on the milestone - has it been 6 years already!

on your point the sixth, bravo! it's still hard, but getting easier, to get good UX embedded into product development DNA.  And I guess it depends on different folks interpretation of what UX means.  On a current mobile app development project I'm managing we've been blessed with a wonderful UX guy who's taken us on a User Centred Design journey, starting with "what problem are we trying to solve?"  and "what value is some user going to place on solving that problem"

when the product mgt guys talk about features "X" and "Y" and the tech guys want to use technology stack "A" and "B" this process helps us resolve back to the core of what we are trying to achieve.

it's not perfect but it's getting better

22 March, 2012
Abdul Ahad Monty


Keep up the good work.

25 April, 2012

Please login or register to post comments.