looksoftware Looks Ahead
Published: May 16, 2011
by Dan Burger
You may get into some rather heated debates about whether the IBM i platform deserves to be branded as legacy or modern, but what that really comes down to is how it is being used. There's no question a lot can be learned about the platform, and that can be said for a lot of folks who are its staunchest supporters, where it sometimes can be said that love is blind. Brendan Kay is not blind. He sees a lot of potential and, more importantly, he sees ways to put the IBM i platform to good use.
Kay is the interim CEO of looksoftware, a company in the midst of much IBM i turmoil over the modernization of applications and even the relevancy of RPG as a programming language with a future. Some organizations have put RPG out to pasture and others have sat on the fence unable to make a decision on their specific programming future. The reason Kay is a temporary CEO, in case you missed it, is because he moved into that role since the recent death of Marcus Dee, a sad occasion not only for the looksoftware family but for many throughout the IBM midrange community. Last week, I had an opportunity to talk with Kay during the annual looksoftware user conference, called lookahead 2011, in Las Vegas. We discussed the technological aspects of the company that were championed by Dee and will shape the company for the years to come.
I started my conversation with Kay by complaining that IT is supposed to be conquering complexity, but it seems that in many ways--at least at the enterprise level--just the opposite is true. Although I'm not aware of any software company marketing itself with a can of worms logo, I still thought it was a fair question to ask whether looksoftware was part of the problem or part of the solution.
Brendan Kay: I think in general IT is getting more complex all the time. My personal opinion is that a lot of people approach simplicity thinking of it as a starting point and that complexity is something added in. So if you are careful not to add it in, then you'll end up with a simple solution. I have a different view. I think you start in the middle of the simplicity-complexity continuum. If you want things to be simple, it is not enough to say we are going to be careful not to make them complex. You have to actually devote your resources and focus to making them simple.
As an example, one of our customers started with an interface modernization project for a call center. The first step we took was to define the underlying services that were required to satisfy call center inquiries and develop an interface with the call center operators. It had an immediate benefit of reducing their average call time by 40 percent. That was only phase one. The next thing was to implement an integrated voice response unit to those same services. That increased the time during which calls could be made to the call center. That was phase two. Phase three was about allowing those services to be available to specific business partners who needed information. Each of those phases leverages the services that were provided in the original phase.
It's not an easy thing to strip complexity from a solution. Most companies talk about it, but don't do it.
Technical people generally don't mind complexity. In fact, they are often quite proud of demonstrating how they understand something that is complex. When presented with a complex problem they come up with complex solutions and are proud of themselves for doing it.
One of the core focuses of looksoftware is simplicity. Do more with less, as in the case of this call center.
RPG Open Access is another really good example. All the features and capabilities are built into it. The customer experience is very simple: run one command to add the handler, recompile, and execute the program.
We put effort into making our handler dynamic and into creating an intelligent handler that can determine at execution time how it should be behaving. There is complexity in there, but what we've done is taken the complexity and dealt with it in our handler. The customer doesn't have to deal with it.
Dan Burger: RPG Open Access is very complex. What made it worth the effort to take on this methodology?
Kay: We had a number of questions about RPG OA. We had a lot of ideas about what we could do. For us the complexity was about whether it would be necessary to have multiple handlers and whether that would become unwieldy to produce and manage them. And what about programs I can't open-access enable because they were written in COBOL or because they are system APIs. That was another area we thought might undermine Open Access. We set out to resolve these things. That's probably an indication of the way we respond to complexity. We actively seek out the complexity and we solve it.
Our initial focus was the integration of OA programs with non-OA programs (5250 programs). That allows existing applications to be brought across without re-architecting the integration points. We proved to ourselves that this could be done. Then we made sure all functionality could be assessed.
A key architectural decision was to write one handler to do everything.
Burger: Mobile access to applications is as hot as any topic in IT right now. What's your take on that with regard to the demand in IBM i shops and how do you see it affecting RPG developers?
Kay: Mobile access to applications is starting to come to the fore. It is even more so in the South American market where access through mobile devices is much greater than what we see in North America, for instance.
The mobile technologies that looksoftware is pursuing are HTML5 and CCS3. They are the keys to providing a rich and optimized interface for mobile devices . . . even though HTML5 is not a standard and probably has a couple of years to go before it is formalized as a standard.
Using looksoftware, RPG programmers can write their applications like they always have. They can look at the way things are presented in our designer and if they want to move things around or change properties, or insert new controls and tie them to functionality, they can do all that through a standard WYSIWYG design. They just point it to what they want it to look like. We leverage HTML5 and CCS3 to create the optimized experience. For those with CCS skills, they can utilize them to customize if they want to, but the typical user gets packaged styles, which means they don't have to learn those methodologies.
It also allows those who are familiar with Visual Studio and Eclipse development environments to work with RPG applications to make modifications and leverage those applications without going down to the RPG level.
Our designer allows someone with a .NET background to make changes to the way an application is presented to an end user without making changes to the DDS. [Almost always unfamiliar territory for those not schooled in RPG.]
We are not about encouraging anyone off the platform. We are providing the functionality that some people say you have to leave the platform to get. Instead, RPG can be the central point of all this.
There have been circumstances where companies have decided to move off the IBM i platform because they didn't think it was possible to do these things in RPG, only to find it was possible after the RPG group pointed out what could be done.
Burger: Do you see an increasing number of IBM i shops migrating from the platform?
Kay: These days more shops are willing to access the real benefits of migrating as opposed to just going along with the belief that they should migrate. The move has to be justified. Is there a return on investment? We get excited when we get into a process where that is occurring. We can show that companies don't have to get off the platform. We are showing RPG is not a dead end technology.
We are focused on not just writing new code and reusing old code, but creating an integrated environment that can leverage assets in existing applications and expand on them with new features. We are very excited about Open Access and what it allows us to do. One of the traditional problems with RPG was the bottleneck between the program and the ultimate user interface. That bottleneck was DDS and 5250. It lost a lot of the richness of the application. With RPG OA, the interface is controlled within RPG, not an external interface definition that is bolted on.
Our goal with RPG Open Access is not to modernize the end user's interface, but to modernize the RPG language.
Burger: Let's look at some other industry trends and how they apply to looksoftware beginning with software as a service.
Kay: SaaS is a significant trend in the IT industry, and in some IBM i shops it represents the present. For others, perhaps, it will be part of the future. For those where it is the present, it is very much about an ISV who is selling software without an upfront fee. For customers, there was a pretty large hurdle to purchase in this manner. SaaS avoids that.
It has two significant implications: It is easier to get customers and it is easier to lose customers. Vendors will need to make sure their software is very good.
Software as a service will affect our customers, and everything that affects our customers, affects looksoftware. We have architected our products to support many of the things that software as a service requires. One is simple deployment. On top of that is support for all the major communications that run across HTTP and through firewalls, for instance.
Burger: There continues to be confusion when talking about software as a service and cloud computing. How do you distinguish them?
We see them as overlapping. Software as a service is not synonymous with cloud, which does mean a lot of things to different people. To some people SaaS is cloud. Others think of underlying services as coming from the cloud. We always have to explore what people think the cloud is before having a discussion.
We have capabilities built into our products that are based around the directions we see cloud moving. Web services enablement, for instance. Being able to take transaction-specific segments of functionality from applications and expose them as a Web service for people to consume over the cloud is another thing.
The work we are doing in mobile computing space is important because some people see using their Android or iPad tablet on a 3G network as cloud computing. There are sophisticated HTML capabilities provided by those devices.
Burger: In terms of looksoftware investments, what are the biggest areas?
Kay: The first one is IBM i. Our technology is about leveraging IBM i and allowing people to do more with those systems.
OAR Gives looksoftware Customer Technological Freedom
Marcus Dee, CEO of looksoftware: 1960-2011
.NET App Modernization Tool Unveiled by looksoftware
RPG Open Access Spurs Development for looksoftware
Post this story to del.icio.us
Post this story to Digg
Post this story to Slashdot