Is legacy application transformation the next big thing in our space?

These days you can't go long without running into another product intent on making it easier to convert legacy applications into some new mobile-enabled form. These are interesting products that no doubt will have a place in our future, but today there are a lot of questions.

These days you can’t go long without running into another product intent on making it easier to convert legacy applications into some new mobile-enabled form. These are interesting products that no doubt will have a place in our future, but today there are a lot of questions. Will they be the next “big thing” in our space? Is this industry ripe for acquisitions and consolidation? Or will the complexities of application transformation slow everything down?

Many options

We’ve been certainly spent a lot of time looking at application transformation already. We can trace this all the way back to Citrix Project Vertigo, but more recently we’ve been looking at things like Citrix HDX Mobile (formerly XenApp Mobility Pack, formerly project Golden Gate) and Capriza; plus we saw Powwow exhibit at BriForum, and Gabe recently wrote about Reddo Mobility.

We’re actually talking about many different technologies. There’s remote desktop-based Windows application refactoring as well as web app refactoring. But besides that, the various mobile app development platform technologies are crossing into this space, too. These include backend resource connectors, “API-ification” for existing resources, multi-platform app creation tools, code-less app creation tools, and more. The bottom line is that no matter what type of existing applications you’re dealing with or what you want to get done, there’s a product out there that will help you transform it. Seriously—there are too many options to keep up with them all.

Complexity, too

Application transformation seems poised to take off, but how long will it really take? There are several difficulties:

First off, it’s just plain early. Most companies aren’t thinking about mobilizing apps yet, or are just deciding what the next step is.

Second, application transformation is complex. This isn’t just a plumbing issue. Mobilizing a legacy application is about way more than just transforming the user interface so it’s touch-friendly—it’s a whole new state of mind.

Third, dealing with all the different application transformation options could become unwieldy. It’s good that there products for all the different scenarios that you have in an environment, but then you end up with a lot of different point solutions for different applications. Managing and securing all those client apps and back-end silos could get really complex.

Acquisition targets?

Because of all the different options this space could be ripe for consolidation and acquisitions. It’d be great to have multiple techniques available but with a common management framework and from a single vendor.

Taking this thought further, Citrix, VMware, and other EUC vendors could be a natural aggregation point. There’s a whole spectrum of technologies involved here, ranging from remote desktop-oriented products on one end to mobile developer-oriented products on the other, and naturally Citrix and VMware will stay closer to the former, but the line is blurring and it could make sense for them to keep expanding. They’ve already going down that route anyway: Citrix bought Framehawk and has HDX Mobile. VMware has worked a bit on Unity Touch features in the View mobile clients and announced the AirWatch Appshield program to work with MADP vendors.

So when will this happen? As I said before, it’s early and this is complex stuff. I’m thinking out loud here, but I wouldn't be surprised to see the major EUC and EMM vendors paying even more attention to this going forward.

Join the conversation

5 comments

Send me notifications when other members comment.

Please create a username to comment.

Great article! I agree 100%, and I also hope that this becomes big, especially if we go in with the assumptions (1) users want to use mobile devices, (2) legacy enterprise desktop apps aren't getting replaced anytime soon.


Cancel

Thing is, I've seen this same thing 12 years ago, only that time it was refactoring legacy mainframe apps for the web. Companies like Jacada had semi-automated tools that let you webify and modernize mainframe apps for the modern web, without requiring any changes to original apps.


For a while, it was hot stuff, and then it totally puttered out because:


1. It was never automated enough, and ended up requiring very significant effort to achieve worthwhile results.


2. Even after all the effort, the resulting web apps tended to be clunky, and less usable than the original green screens.


3. Even small changes in the original app often required the process to be redone.


Maybe this iteration will be better - I certainly understand the need. But if anything, properly mobilizing a legacy Windows app is more difficult than webifying a mainframe app.


Cancel

In my opinion, transformation of User Interface paradigm from WIMP (mouse controlled windows based GUI) to NUI (natural touch based interface) is more fundamental then a lot these solution providers present to us.


Just as Dan mentioned, UI transformation was also not so successful 12 years ago, when character based terminal applications were transformed with same type of technologies to WIMP interfaces.


Hopefully most of the application developers have learned to use three-tier development, where UI is a layer which easily can be replaced when new UI technologies become available,


Hopefully most of the application developers have learned to use three-tier developement, where UI is a layer which easaly can be replaced when new UI technologies become available,


Cancel

This is a great discussion and great article from Jack.  As the CTO of Reddo Mobility (and CTO/founder of Gizmox prior to that), I can personally attest to the complexity of transforming legacy applications.  I really like the image of "API-ification" that Jack mentions.  To build on this concept, I would contend that you need to think carefully about how to define the legacy app as a "service" that can be efficiently interfaced with in order transform into something usable


I think there are at least three points to consider:


1. The application at the back-end has its own embedded application logic, flow and different states. Unless you want to completely rewrite and redesign everything, the guiding strategy for UI re-factoring should be to distill the legacy UI to its relevant components and NOT attempt to re-create it and all of its embedded application logic.


2. It is not practical to attempt to distill a UI based on the absolute locations and values of its controls.  You need to recognize and track the underlying Windows elements in order to quickly adjust to many of the potential changes in the original app. This was a key limitation to mainframe screen scraping technologies.


3. A complete re-factoring solution needs to also include an enterprise-class application delivery protocol and a set of mobile-optimized controls for rendering the re-factored application.  


I completely agree it is complex.  But doable.


Cancel

I love how this was framed!  I also love the comments so far and the perceived players in the space.  I love Dan's comparing this to having to connect the mainframe data to the web back in the day for you old folks ;) It seems to me based on that comparison, did that not happen, just not with the green screen notion originally thought of?


If we can do as Jack proposes, is this really that far away?  Seems to me one could combine the technologies in  play now to get pretty darn close?  The value seen from making applications which were not written to be three tiered as LJ points out being able to become so and connected to future focused applications would ultimately free us from being vendor locked based on old platforms and instead enjoy the new battlefield of new cool devices.


How many devices where you carrying 5 years ago?  Three?  Now?  This would be needed for true ubiquitous access correct?


Cancel

-ADS BY GOOGLE

SearchVirtualDesktop

SearchEnterpriseDesktop

SearchServerVirtualization

SearchVMware

Close