Friday, July 18, 2014

Stitching Matters

photo credit to Deseret News Education Travel

Stitching matters.  Stitching holds disparate components and materials together in a form something close to what we had in mind as a solution to a problem. Could be clothes.  Could be wicker baskets.  In the tech world, stitching things together typically involves integrating components or services together into a unified solution.  And it matters. A lot.

I have a few applications on my mobile platforms that rely on integrated cloud storage services for saving my work.  Now, I've tried many cloud storage services. iCloud seems to be continually on the fritz for one reason or another.  Google Drive (or whatever they're calling it today) consumes CPU resources to the point that my devices start to smoke.  Dropbox has issues with losing data during sync. Box...well, they don't seem to really care about individual users anymore as they're not targeting enterprise users. Sky Drive, Live Drive, Hummingjay - you name it, I've probably tried it.  I finally settled on Copy: reliable, significant space for free, always working.

But, there's an issue with Copy.  Yup, you guessed it, no stitching.  I have yet to find spreadsheet, word processing, or presentation applications for mobile that integrate with Copy.  So I'll probably be moving back to Dropbox and leaving Copy...and backing up everything much more often.  Because I'd rather have a solution that works well some of the time than a solution that doesn't work at all.  Stitching matters.

Stitching is a differentiating factor in enterprise applications as well...especially if we're talking about SaaS.  From my worm's eye view of the enterprise applications market, I don't see many customers buying everything from a single SaaS vendor.  It's usually a mix of on-premise and SaaS, often involving multiple vendors. But those customers want unification across the enterprise:  single sign-on, single data source, consistent look and feel, a single business process, and so on.  So one of the bigger challenges in SaaS becomes stitching.  Not just doing it, but doing it well enough to meet expectations.  Stitching matters.

So when someone asks me for advice on choosing a SaaS provider, I always suggest that (among other things) potential buyers consider the stitching.  What integration comes pre-built out of the box?  Do the APIs and integration points comply with industry standards for SOAP and/or REST and/or JMS?  Are there cloud integration services available from the SaaS provider or a partner?  What stitching is available to handle larger data loads?  Can you do those data loads yourself or does the SaaS provider insist on doing them for you?  How about services integration with mobile platforms?  If the stitching doesn't meet your specific needs, will the SaaS provider assist?  If so, at what cost?  Stitching matters.

Now down to what Oracle has in regards to stitching ('cause Oracle is what I write about here).  It seems like Oracle is doing really well with service-based integration and application co-existence as their Cloud Application Services (aka Fusion Applications) continues to evolve.  

We're seeing an expansion of APIs and pre-built integrations, especially utilizing REST.  Both inbound and outbound integration of very light and simple data loads looks good as well.  The stitching in these areas looks pretty good.

Inbound and outbound integration for large and complex data loads, on the other hand, has some room for growth.  The tools are there, but the processes involved in using those tools in Oracle's Cloud still have some evolving of their own to accomplish.  But it'll get there.  Because Oracle's leadership knows: stitching matters.

Sunday, June 29, 2014

Changing My Lab

As we continue to see more inroads made by cloud and SaaS technology, I have to admit to a downside...what can I work with in my local lab.

My local lab consists of two iMacs, an Apple Airbook, several arrays of 1-TB external hard drives from OWC, and a serious effort in leveraging virtual machines (I'm personally a VMware fan).  It wasn't too long ago that I could run most Oracle technology on my little lab platform.  And I like it that way...I prefer to have total control over all my prototyping and development environments.

But as Oracle transitions to the cloud, the software footprint for the various Oracle technologies I work with has just become too big for my lab platform. Fusion Applications, the Business Process Management Suite, OBIEE, the SOA Suite...heck, even E-Business Suite 12.2.x is pushing the limit.  Some of the footprint growth is due to new features, some to code bloat or technical debt, and some is just the nature of the beast with enterprise applications built for the cloud.

I'll also admit to being cheap here.  There's no subscription fees for AWS or Oracle Cloud coming out of my personal wallet.  Just not gonna happen.  And investing in new hardware seems to be akin to stocking up on buggy whips or vacuum tubes.

So, what's my plan?  My intent is to go back to basics for my home lab.  I can still run an Oracle database quite nicely (anything ranging from Enterprise to APEX to Express).  I can also run Oracle ADF (but, due to performance issues, I choose Glassfish over WebLogic).  OBIEE sample apps also fit and run well.  But I'll have to look to my company, my customer, or some other kind-hearted soul for bigger sandboxes - Fusion Apps or Middleware, OBIEE, and anything having to do with SOA integration or business process builds.  I'm not crazy about this solution, but it is what it is...

What about the rest of you out there?  How do you folks (especially the non-Oracle employees) handle your personal sandboxes these days?  Any ideas or suggestions?  Comments welcome.

Tuesday, June 24, 2014

About User Groups

I'm hanging out in the middle of nowhere this week...Fort Riley, Kansas.  Here to visit my granddaughters.  Which means I'm missing ODTUG's KScope14 conference.  Missed the OAUG/Quest/IOUG Collaborate14 this year as well.  Will also be absent at OAUG's ConnectionPoint14 in Pittsburgh.  Will be missing a few others that are usually on my calendar as well (But I made it to UTOUG Training Days, Alliance14, and the MidAtlantic HEUG conference - will also make it to the ECOAUG later this year).

With all the user conferences missed in 2014, I've had some folks asking if I still believe it Oracle user groups.  The short answer is yes.  The longer answer is yes, but I do believe the user group model needs to change a bit.

Attend a user group conference this year (sorry, Oracle OpenWorld does not count - it is NOT a user group conference).  Look around at the faces.  Other than those working the partner sales booths, the vast majority of those faces will be middle-aged and older.  See, when user groups were first formed, the model was built to appeal to Baby Boomers and Echo Boomers.  And the big thrill was face-to-face networking.  Now that the Baby Boomers and Echo Boomers are riding off into the enterprise technology sunset, the user group model can only flourish by changing the model for those who take our places.

Face-to-face networking is still important, but just doesn't seem to hold the same level of importance for these younger workers.  Easily accessed on-demand education sessions on the web (for free), virtual gatherings on GoogleTalk, facilitating group chats on focused subjects, information in short snippets...simple, quick and virtual channels of information delivery seem to gain more traction with the rising generation than annual, huge national or international conferences when it comes to enterprise apps.

So, yeah, I still believe in user groups.  But, as long as you're asking, I think the model will need changing in order to flourish into the future.

I'm going back to the grandkids now...

Friday, June 20, 2014

Learning From The Earnings Call

So now that we've heard the Oracle FY14 Q4 financial results, what did we learn?  Most noticeably, Oracle continues to be a company in transition...and that transition is beginning to take hold.  Changing a company and an ecosystem as large as Oracle's is like turning a takes time.  But we have some pretty clear evidence that the battleship is turning.

Let's look at the numbers for a minute.  Oracle's SaaS and PaaS for FY14 came in at $1.12B (US).  That's up 23% from last year.  Heck, they're going that part of the business like a weed.  Oracle is already well past Workday in terms of annual revenue (although the comparison is a bit unfair because Workday isn't really in the PaaS business, so let's not count them out by any means), and is approaching about 50% of the latest SaleForce revenue numbers.  Comparatively speaking, Oracle is an up-and-comer in the space.

But the transition is not without some pain.  New software licenses were flat for the quarter and earnings-per-shared missed expectations.  That tells me two things:  1) some of that cloud growth is at the sacrifice of software licensing deals; 2) Oracle has yet to figure out how to make cloud services as profitable as license deals.  We're likely to see the second point work itself out as the pressure for margins ramps up through existing Oracle customers take advantage of the Customer 2 Cloud program.

So what we're seeing is the continuing transition of Oracle into a SaaS and PaaS provider, but with some speed bumps along the be expected in any transition, especially one of this magnitude.  So, you may ask, what's driving the growth?

IMHO, Oracle offers three factors that differentiate their SaaS and PaaS offerings from other market competitors:

1) A database that offers the benefits of multi-tenancy without commingling data.  That's a huge advantage in overcoming security fears of many potential cloud customers.  And, from what I can tell, security fears are the biggest objection for most potential cloud customers.

2) A very well-design User Experience with Simplified UI.  While there's still work to be done (let's build responsiveness into ADF, the tool used to build the UX, so that we can build once for all platforms), Oracle's UX has built a big differentiator in the marketplace.  And offering up the templates and design patterns so that developers can build their own apps with the same UX is a great approach.

3) Deep pockets.  Check out that Oracle balance sheet.  With all that cash, Oracle can afford to invest in growing their cloud business.  That includes investing in strategic customer accounts.

So while Oracle's Q4 report could have been better, I personally saw what I was looking for: tangible signs of progress in Oracle's transition.  Learned from the earnings call.  Ya'all go ahead and sell that stock.  Poo-poo all over the outlook.  Whatever.  The battleship is turning.

DISCLAIMER:  I hold 10 shares of Oracle stock that I acquired while working for Oracle in the 90's.  I keep it just for sentimental value.  I'm admittedly a big fan of Oracle technology.  And while I'm not always a fan of Oracle's business tactics, I think they're a very smart company from a strategic perspective.  So, it's true...I'm biased.  Now you know.

Wednesday, May 21, 2014

Flipping The Fit Gap

I've been in the software business a long time now...I can remember from first-hand experience how things were done 15 and 20 years ago.  It seems like one of the biggest changes from then to now comes in our approach to "Fit Gap".

Fit Gap sessions are part of most packaged application implementations approaches.  The idea is to compare user requirements with software functionality in order to see how well the software "fits" with those requirements and where the software fails to meet those requirements (each of the latter is a "gap").

In the past, the focus was on fulfilling identified gaps by manipulating the software:  personalizations, extensions, customizations...what ever was needed to support changing the software to fit end user business practices.  But times have changed.

Enterprise Application systems now come with industry best practices baked into the software.  In fact, when Enterprise Software vendors talk today about "transforming your enterprise", they're really focused on three things:

1.  Leveraging the best business practices baked into the software to improve a customer's business processes.

2. Providing valuable information about your enterprise from all that data collected by the enterprise applications system (transactional reports, business intelligence and analytics, etc.).

3. Improve your operational agility and provide an opportunity to redirect your IT focus by moving your IT operations to the cloud.

The changes in Enterprise Applications have changed the nature of the Fit Gap process.  The emphasis has changed.  The objective is no longer to simply move your business processes to a new technical platform.  Instead, the target is to determine how well the best practices baked into the applications will work for your enterprise.  In other words, will this software package make my business better?

Over the past few years, I've seen a change as Enterprise Applications customers have picked up on the change.  There's less of an effort on building big requirements up front.  Instead, those customers are running the software "plain vanilla" out of the box in a test run (or Conference Room Pilot 1 or CRP 1).  Afterwards, the Fit Gap is conducted...but with a focus on how well the current enterprise fits into the business processes baked into the software.  Changing the software mostly occurs when the customer enterprise is, for one critical business reason or another, unable to fit into a best practice.

Big change when you think about it.  Fit Gap has been flipped from an emphasis on "make the software fit us" to "how can the software help transform our business"?  Fit Gap has been flipped!

Tuesday, April 22, 2014


Convergence (\kən-ˈvər-jən(t)s\)

1  :  the act of converging and especially moving toward union or uniformity; especially :  coordinated movement of the two eyes so that the image of a single point is formed on corresponding retinal areas

2  :  the state or property of being convergent

3  :  independent development of similar characters (as of bodily structure of unrelated organisms or cultural traits) often associated with similarity of habits or environment

4  :  the merging of distinct technologies, industries, or devices into a unified whole

"Convergence." Merriam-Webster, n.d. Web. 22 Apr. 2014.

We're seeing convergence in process, right now, in Oracle's enterprise applications.  Some examples:

Simplified UI:   Take a look at Simplified UI in Fusion Applications Release 8.  Then look again at the Simplified UI recently announced for the E-Business Suite.  And then take another look at the new UI for PeopleSoft 9.2.  They're all looking and working in a similar fashion.  Built on different technologies, to be sure, but providing a similar user experience.  Functional design patterns are driving a convergence in the user experience for Oracle's packaged applications.

Continuous Delivery:  PeopleSoft uses a continuous delivery model - feature packs developed incrementally, then automatically distributed to end users for review and application in an incremental fashion.  Not upgrades, but incremental enhancements of the existing version...which may eventually lead to the end of major upgrade events for PeopleSoft altogether.  With online patching as of release 12.2,  the E-Business Suite seems headed in the same direction.  With Fusion on the cloud as SaaS, moving to a continuous delivery model is not a difficult matter (in fact, short of the branding of new releases and the perception that branding creates, Fusion Applications is already built on a continuous delivery model).

Cloud:  The convergence here seems to be one of industry strategy.  Commercial industries on cloud, public sector industries on cloud, and the recently announced clouds directed at the higher education space.  Oracle is utilizing their basic cloud applications services, then supplementing with additional services built for specific industries.  Again, the higher education market is the best example of this:  Student Cloud, Higher Education Cloud, and new higher education specific functionality for Fusion Applications.

These are simply the most visible examples of the convergence trend taking place within Oracle's enterprise applications.  There are more taking place, and I'm sure the future holds even more.  We're barely at the beginning of this trend.

The real question here:  what does this mean for the customers?  IMHO, it's significant...but each customer will probably come up with different answers.  Just some ideas to get the gears turning:  co-existence between different Oracle applications with a single, consistent user experience; hybrid platforms running transactions on-premise, but with big data processed on the cloud to provide information through a combination of transactional reporting and embedded analytics; moving from teams supporting major upgrade efforts to small teams engaged in continuous delivery, thus freeing up more resources for important, strategically-oriented work.

That's all I can jam into this post without turning it into an epistle.  Hopefully I've left you with your brain cells jiggling about what convergence within Oracle's enterprise applications means to you.  Thoughts?  Comments?  You know what to do.

Monday, April 21, 2014

Simple Is The New Cool

We've all heard it: "simple is the new cool".  It's true.

We certainly see it in the UX world.  Simple designs for human interaction.  Tell me what I need to do, how to do it, and make it easy.  If I only had a dollar for every time I heard that mantra... And when we consider the market moving to the cloud, UX is the major differentiation - and simple is carrying the day every day.

That last paragraph?  Double-down on it for mobile.

And we see it in system architecture as well.  Big emphasis on unraveling institutional spaghetti bowls of interfaces, services and business processes in favor of simple, straight-forward enterprise architectures.

Packaged enterprise applications?  Sure.  How many times over the past five years have we heard and seen examples of "reconsider your customizations"?

And let's not forget about embedded analytics that blur the line between business intelligence and transactional reporting.  We're making it simpler to identify and understand the information we need, rather than getting wrapped around the axle about how to classify the source of that information.

Everywhere you look in the enterprise applications world these days, it's all about simplification.  Simple is the new cool.

So how do you start with simple?  You start with a strong focus on the problem you're trying to solve.  What are you making better?  And, by the way, problems like "our company needs to sell more software/services to end users"?  Not the type of problem I have in mind here.  Think about a warehouse working taking inventory.  A college freshman attempting to register.  An AP specialist attempting to identify and analyze payables trends.  Those are the types of problems you want to solve.  Those are the types of situations you want to make better.

Get focused.  Simple is the new cool.