Questions? Feedback? powered by Olark live chat software
Print Page  |  Contact Us  |  Your Cart  |  Sign In
The Boomer Bulletin - 2010
Blog Home All Blogs
Search all posts for:   


View all (46) posts »

Understanding the Shift in IT from Servers to Services

Posted By Eric Benson, Tuesday, May 11, 2010
Eric Benson

I don’t think it comes as a surprise that hosted applications (also known as cloud computing, software as a service (SaaS) and many other terms) are being touted as the future of IT.  In fact, if I listen to the sales people I talk to, oftentimes their cloud services are faster, cheaper, more reliable and respond faster to change than conventional platforms.  Sounds enticing until you think of the things they are missing in their statements.
What about…

  • Security (so I have passwords, but how is my data safe on your servers?)
  • Data portability (what happens if I decide to leave?)
  • Feature equality (so, where do my print jobs go to your hosted server?  How does work, exactly?)
  • And so on.

For some technology professionals, this is enough to say "enough” – and not look any further.  I’ll get to this in a bit, but this reaction is a definition problem.

Cloud Computing, Automobiles and Postmodern Thinking

I think that regardless of how smart someone is, or how well they try to define the whole hosted applications model, it’s a term to define something that isn’t well defined.  Two other examples make my point.  The first cars were called horseless carriages.   No one quite knew what they were, so they were defined by what they weren’t.   The same thing is true of the term postmodern.  Implicit in the term is "after modern”, which is definition by omission. 
Software as a service is very similar.  Take a term you already know (software) and define it by being a service instead of being an application.  Same thing with cloud computing – it’s being computed out there (in the clouds, very clear, right) and not in here.  How much "vaguer” can you get?
My hunch is that this really hits a bone with a lot of IT people, but they might not be sure why.  Let me describe a personality trait that I don’t think most technology people would oppose. 

  • If a task is unknown or undefined, it will take enormous amounts of time/energy/resources to complete.  Also, the list of objections will contain many entries and is very detailed.
  • If a task seems easy or exciting, it will take almost no time/energy/resources to complete.   The list of to-dos is exceedingly short and undetailed.

Now, take the largest shift in application provision in current memory and define it with a fuzzy term. Which end of the reaction spectrum above will come out ahead?

Don’t read this as me saying hosted applications are not ready for primetime.  The model has been around since the mainframe, the only thing that has changed is the scope of the network.  If you have multiple offices, you are already doing cloud computing.  Think about that for a moment.
Hosted applications and services are built on mature platforms using the same backbone you use to do almost every communication in your office.  The key is not to attack the term or the ambiguity that ensues from hosting in the cloud, but to address the places where it works for you.  You can’t do that by treating hosted applications with a double standard.  Treat them with the same standards as your internal servers and applications.  You’ll start to see where the benefits are if you do so, but you have to address them with a common foundation.

Getting away from the overwhelming 5% reaction

We’ve gone through a transition over the past few years towards being fully hosted, and although the transition has not been easy at times, it’s working.  I would recommend Jim Boomer’s article The Changing Role of the IT Professional for a snapshot of the whole picture.   One of Jim’s comments that I wholeheartedly agree with is that you should focus on the 95% of the features that hosted applications do answer instead of the 5% that they don’t.
An example from the Boomer Technology Circle™ meetings can help here.  This question came up -- How do we manage our printing in the office when we moved to a virtualized environment (translation – most printing is handled through a print server; with servers hosted, how is this function handled?).  Although this is a valid question, it focuses on the 5%, not the 95%.

Now seems like a perfect time to address the key objections to hosted applications.  Let me take the three I mentioned and quickly discuss who I moved beyond them.

So, what about…

  • Security?
    How many of you can say that your data is completely secure in your office?  Security will always be a concern, and I can’t think of a better location for my data than redundant SAS-70 data centers. 
  • Data portability
    Look at your office – why haven’t you moved away from CCH/Thomson/UltraTax or some other program?  Data lock-in is part of being in business.
  • Feature equality
    List the features you really use in your xxxx - whatever you are moving.  Then compare that list to the vendor you are looking at.  Not the 9000 features that came installed, but the 90 you use.  The ones that are missing can either be addressed or you have the wrong solution (or timing).

Moving to a hosted infrastructure is very similar to other transitions that have a 5% problem – in other words, every other problem your firm faces.  When you hire someone, you know what you are hiring for, you know where they will work, and you know who they’ll work with.  But, the interaction that new hire has with all of those variables is unknown.  The same is true of the economy – you can provide quality services to a filtered client list, but there are many variables that have nothing to do with you.  Be consistent in treating internal and external technology as part of a business solution.

In technology, the unknown quotient gets a LOT of play, and I think some of it is deserved.  If a disaster plan is not in place, you may not survive as a business if a disaster does happen.  I used to think that this passive-reactive job pressure was the reason for objections to hosting, but the more I talk with people, the more I realize this isn’t true.  Planning for the worst case scenario is important.  However, when you look into something like hosted applications, I don’t think the typical negative reaction is worst case scenario planning – it’s the unknown task taking immense amounts of time problem.
The opposite is true as well – I recently talked a firm IT person out of moving email hosting because his reason for moving was it "looked easy, cool and had a huge amount of storage”.  This is the other end of the scenario – easy or exciting tasks take no time.

Define the whole model, then decide which services you can move

So, how do you avoid these pitfalls?  In my experience, the way towards success is to define the whole picture before you choose where you’re going.  I am not talking about a complicated network diagram here – I mean something that defines the core services that technology provides, and what resources are needed to do so.  In our office this is a matrix that we developed through a process improvement team.  You can see the main process steps that our company uses to do work, what tools are used at each stage, and what services those tools provide.  You can view by tool, by service, or by process step.   Whole firm workflow has a place in any technology plan – or strategic plan, and it helps tremendously when you try to refine a large topic like hosted applications down into a manageable project.

Regardless of how you do this model, it is crucial to have one.  Technology professionals can discuss moving services with the teams that use them because everyone is on the same page.  Planning for the future becomes easier; you can see where improvements need to be made more quickly.  In this way, everyone can come to an understanding of where you are at.

It’s not a question of if, but when

Hosted applications (and infrastructures) are the future.  They should be treated seriously whenever an upgrade or change need to be made in your infrastructure.  Start with the options that make sense using a reasoned, team approach.  Make sure you are starting from the right direction – what do you need versus what’s "out there”.  Cloud computing can be a storm if you let it be, but that’s true of any technology decision made without the right process and team. 

Define what you are doing first, let that determine where you are going, and then use that knowledge to see where hosted applications fit.  Your firm will benefit from solutions that truly fit the hosted application promise of cheaper, more reliable and accessible from anywhere.  There are quite a few robust solutions that are very cost effective, flexible and can change the way you do business for the better.  Be proactive and move when it makes sense.  It’s not a question of if any more, but when.

Tags:  Eric Benson  Information Technology  IT planning  new technologies 

Share |
Permalink | Comments (1)

Comments on this post...

Byron Patrick, CPA says...
Posted Tuesday, May 11, 2010
Hi Eric, nice post!

You are absolutely right... it’s not IF but WHEN. Fortunately we have a number of clients that for them, WHEN is NOW.

To answer your questions:

"Security (so I have passwords, but how is my data safe on your servers?)" Redundant Cisco firewalls, SSL 128bit encryption, MS Threat Management Gateway, strict NTFS permissions.

"Data portability (what happens if I decide to leave?)" It’s your data, you get it back.

"Feature equality (so, where do my print jobs go to your hosted server? How does work, exactly?)" Your local printers at home, in the office, at a client site, etc are autocreated when you login to your virtual desktop, therefore, you print directly to them.

Eric, I agree with you, there is significant ambiguity and vagueness related to our types of services. However I will argue, if your vendor of choice doesn't have the answers you should look elsewhere. In addition, there are enough companies using these services now that references are a great opportunity to set realistic expectations.

Thanks for your post Eric, there is so much nonsense and misinformation on the web about these services it is nice to see someone helping to create some clarity.
Permalink to this Comment }

Community Search
Sign In
Sign In securely
Recent Blog Posts