Had a briefing with Michael Vizard a few weeks back and just picked up this piece he wrote. I like the way he has articulated our discussions around the problems as he sees them. Here's the link. Let me know what you think.

RETHINKING APPLICATION INTEGRATION

Well, for one thing, this economy has put clarity on the need to help enterprise users become a lot more productive in their environment, with applications they already have. I have pitched this message over the last 3 months, so let me try it here on you and see what you think!

"I propose that we have turned our enterprise users into manual workers". Sure, they use computers but most of their business processes are manual. What I mean is, look at an automobile factory of 30 years ago. We would have well trained technicians, manually building cars down the production line but it was all very manual, each step requiring the mostly manual activity before the next step and so forth. Today, just go and look at an enterprise user and watch them, using their desktop applications to put someone on a new insurance or mobile phone plan. It's mostly manual. just because users have computers, it does not make them productive if the many applications they use, don't work together in any form. For example; user logs into System A, B, C, D and E. Lookup customer verification info in system A. Enter phone number of SSN in System B, use account number from B to go to system C to verify address. Change address first in system C and then D. Log into System F now because we need to check a credit score or address verification. System E has the details of the current Insurance plan or Mobile Phone plan and user needs to work out if customer qualifies. Customer doesn't qualify but another new plan might be available so we go to system F........ and so on and so on, on virtually every users desktop, every enterprise, every where in the world! How manual is that?

If only every desktop application was "open" to be enable these workflows to be automated for our users. The users can get on and do what they know best and focus on the customer. We have have left our users behind. Spending trillions of dollars of integration at the back end but then still delivering new disconnected UI's to the user on the front end (desktop), and the problem gets worse every year!

Products from companies like OpenSpan, enable Desktop applications to become Open. Almost instantly. it's game changing stuff. Just imagine what you could do, if only ALL of your applications were enabled to allow automation. Desktop productivity will go through the roof.. More later....

Just posted a blog on how wrapping legacy applications is no longer limited to green screens. Technologies now exist to robustly "Virtualize" the presentation layer and this enables any kind of business application to be integrated or extend through SOA. Client side or server side.

Wrapping Legacy Applications for SOA or anything else..

The key takeaway in 2009 is, do more with less. Use what you already have. The transactions work, they are just needed to be used in different ways. We always talk about re-use. Well lets start with RE-USING what we ALREADY have. Enjoy. Feel free to comment.

 2008 seemed to all be about...

Just got back from CES. I reckon numbers were down 50% from my visit 2 years ago. How could I tell without using sophisticated integration and measurement tools? Well, I got a great rate on a top hotel and there were zero lines for cabs and buffets !

What did I see? A great thin OLED (less than a 1/4"). A 150" plasma from Panasonic - not sure you'll be picking one up at Best Buy's anytime soon! Also, big price reduction on the 103" plasma. Down from $99,000 last year to about $60,000 this year. Not quite a red tag deal though eh?  :)

Saw plenty of 3D tech but overall, I was pretty unimpressed with most things I saw. Lots of solutions, looking for problems in my view.

Saw a good demo of Windows 7 and got to talk to one of the designers for  a while so will report back on the new UI in my next post.

I might bore some people with my common phrase.. "A legacy application is any application written more than 5 minutes ago" but it quite simply is true. And with that truth comes the need to integrate a 5 minute old application with some other application - older OR newer than that one.

Even if you write a bunch of SOA services and wrap those to create a new application, it's a legacy application. No developer EVER thinks of their application at the time as legacy - it's just an app!

However, it never ceases to amaze me though, how little effort goes into an applications UI to make it "open" for integration. Do developers really think users like to use copy and paste? Do you know, when you click a hyperlink, all that really is going on is the equivalent of "copy this URL", "alt-tab to browser", "paste into browser text box" and "hit enter". Yet if we had to that that manual process for every web site we wanted to visit, we'd get very mad.

Having API's on every UI (application) for automating any workflow is my goal. Not to solve the copy and paste problem but to make every enterprise users workflows (lots of legacy apps), as automated as possible. The ROI is astronomical. Desktop Productivity is the end game. The tools are here...

The New Enterprise Desktop

I have always felt that desktop virtualization is a fantastic technology but the millions of end users will see only frustration on their own desktops - forcing users to rely even more on copy and paste to do integration won't get you the business buy-in!

Technologies to add API's and automations to the apps in the VM already exist, enabling desktop apps to communicate across the bubbles. It's time.

Also Check out :  Is Virtualization an Integration Solution or Nightmare?

If there ever was a time when desktop productivity will be as important as it's ever been, it will be in 2009. The cliché of 2008, "do more with less, and do more with what you have" will dominate from now on.

CIO's, answerable to their boards and shareholders will see that innovation on the business side will be the key drivers for their own success. For some, it's about survival and for others it's about bringing back profitability in the business. Frequently overlooked until recently, looking at enterprise users and how efficient they and their desktop applications are will be crucial in delivering massive savings to the bottom line.

Just because your users use a computer and the software running on it, does not mean you do not have MANUAL processes. Just the opposite. Many users use their keyboard and eyes to run 2, 5, 10 or more applications on their desktops. Just watch your users and tell me, the way they use desktop software is mostly manual, right? Automate these processes and allow your users do more in less time. Run the numbers of $$ savings? It's huge and this bit, business already knows and why they want it NOW. Spend a $1m to save $5m or even $10m or more in the first year. The math is easy. The technology is here and it can be done quickly. What are you waiting for?

David Linthicum writes on "Data Integration at the User Interface" - 

Very little, if anything! That's why virtualization has been harder to sell to the the business users.

Unless you put your integrated applications in the same VM, even existing desktop integrations will break. Excel in one VM and your integrated Trading System in another VM now won't be integrated (except for copy and paste )! 

There are technologies now that enable you to integrate applications running in different VM's but strangely, for most desktop applications, there are none from the VM vendors themselves.

I remember, way back when (yawn), when I had to integrate Lotus 123 with data from a mainframe through a terminal emulator - both running on a DOS PC. It wasn't easy to get these two products to integrate but thanks to both DOS products having at least a nominal API, it was possible and added significant productivity gains over the manual re-keying of the data.

27 years later, I find most enterprise users still running desktop applications with no API's or integration capabilities what so ever at the desktop. Even new applications fall foul of this. Amongst the many other pains this causes our users, dramatic productivity improvements are missed because user workflows across multiple applications are mostly manual.

There are tools today that can "instantly" create API's into any existing (new or old) desktop application. With that, productivity gains are seen immediately. These API's are client side and can even be exposed as Web Services.

I have some thoughts about why the problem still exists today. I'll share that later. What are your thoughts?

 There are 3 Priorities for IT and Business in 2009 for immediate and dramtic cost savings !

 

1. Optimize what your users already have and do
2. Optimize what your users already have and do
3. Optimize what your users already have and do

 

OK, a little tongue in cheek but it's TRUE.... Have you ever looked at a real user lately in your company? Really looked at their desktop and what they do with it? Please, stop what you are doing and go and look. This is really the only place you can see with your own eyes the results of your organizations labors for SOA, Integration, BPM, BAM, Mash-ups, Web 2.0 and what ever other technology you've been working on for the last 5 years! Your users are your consumers!

 

What do you think you your users will be looking at or doing on their desktops? I am not a mind reader but I'd be surprised if it isn't at least 4 of the 5 below;

 

1. Viewing more than 5 applications daily, possibly 10 or more. This will include completely different UI models such as Windows GUI apps, Web Apps, Java Apps, Mainframe apps and maybe even some DOS apps.
2. Entering common keystrokes 1000's of times a day to integrate the above applications which will include ALT-C, ALT-TAB, ALT-V and mouse clicks, Focus, double-click, right click, copy, Focus, right click, paste. Also typing in the same text twice (SSN, Phone etc.,)
3. Executing laborious repetitive workflows, like looking up in two or more systems to see if the customer exists in two or more of the systems, copying data out of spreadsheets to do a trade and moving the data back to another system
4. Logging in with user name and passwords repetitively to at least one of the applications, maybe all of them, even SAAS applications or ones you wrote a few years ago
5. Writing text on a notepad or post-it-note to be used later (to enter data back into a spreadsheet or notes log).

 

I could go on, but you get the point. Enterprise has spent a fortune on the server integration technologies but rarely do they focus on what business is demanding. That is, the optimization of what our users already have and do. If you deliver your users yet another application, make sure it doesn't just sit there and add to the desktop conundrum of too many un-integrated applications. It kills desktop productivity and adds to your enterprise costs. Fact.

 

Optimizing the desktop allows enterprises to save hundreds of millions of dollars by automating even the simplest of workflows on their user's desktops. Think of the time wasted daily by your users and what you would save through automation. I have seen desktop automation be implemented in less than 4 weeks that saves over $500k a year by automating the tasks of less than 20 users. I've also seen companies save over $20m a year by pushing out just one small automation across 20,000 users.

 

The time has come. There are products out there today that can be rapidly implemented to integrate, automate and ultimately optimize all of the applications your users have on their desktops. IT can be trained in days and massive savings to the bottom line can be seen in weeks!

 

With automation and desktop productivity improvements comes better monitoring, compliance, less errors and elimination of duplication. 2009 - will be the year of desktop integration and dramatic productivity increases. Remember, I said it here first.

 

My company, OpenSpan has one such product and I encourage you to look at a demonstration that you can view on line. Do your own math? What would just 2% increase in user productivity mean to your CFO in 2009?