This week I’ve been thinking about how to generalize on what Linxter is. Currently we use the term platform-as-a-service to describe our Internet Service Bus. However, as we follow the whole cloud of conversations about these new categories of computing, we begin to question the accuracy of this classification. Leading us to ask the question, what kind of an as-a-Service (aaS) are we?
There are so many *aaS acronyms out there, how do we compare and contrast Linxter with other services? The namespace is becoming increasingly convoluted, but the concept is ever more crucial to businesses and governments. This is being validated in many places, and not just in tech publications. We recently read in the Economist that "... everything and anything computer-related will end up being offered as a service.", and John Garing, the CIO of the U.S. Defense Information Systems Agency, on the need for cloud-based IT services was quoted "We have to get to this standard environment that is provisionable and scalable" in the New York Times.
As organizations choose not to “own” an increasing amount of functions outside of their expertise, there will be even more what-have-yous-as-a-service. This leads us to an impasse of having non-scalable terminology. Thinking of the more prominent *aaS offerings (Software, Platform, and Infrastructure) we do not exactly fit any of these as defined by others. For anyone who is not really clear on the differences of Infrastructure-as-a-Service or Platform-as-a-Service, Markus Klems does a bang-up job getting you up to pace. Should we coin yet another *aaS term or craft one of these existing terms to more closely fit our ideology?
If pressed to devise a new term, I would say a hyperconnective MOM-as-a-Service is the most accurate, but is this really a step forward in enabling this conversation? Is this helpful? The MOM acronym goes back a ways, and stands for Message-Oriented Middleware. Our MOM is hyperconnective, as per Nortel’s notion of hyperconnectivity, in which anything that can be connected to the Internet will be. And we are an *aaS because our ISB is offered as an infrastructure/platform in the cloud (http://linxter.com/implementation).
During my search, I came across Boomi (http://boomi.com), which is an IaaS. They are not an Infrastructure-as-a-Service, but an Integration-as-a-Service provider! We are running out of letters! So now, there are two fundamentally different offerings that share the same acronym. That said, Boomi provides a really cool service, which is database-to-database field level mapping between systems. Their IaaS seems especially well-suited for SMBs.
For Linxter, Infrastructure-as-a-Service does not concisely fit what we do, as it can be generalized as simply leasing hardware in the cloud. Our offering brings much more than this to the table. And Software-as-a-Service seems much too simple to describe our solution, as it implies a niche end use. With Linxter, one could build their own SaaS offering that uses Linxter to take care of their communication plumbing needs. Platform-as-a-Service seems to suggest a more complex solution than what we offer, but it seems like the best acronym for us right now.
Returning to defining Linxter, let’s begin by dissecting what it does. Linxter provides software developers a secure, reliable and auditable messaging solution for the applications they write – applications which can run on servers, workstations, laptops or devices (whatever that evolves to mean). They are able to communicate regardless of their platform, network, or location. Furthermore, this is done while maintaining independence from each others’ internal processes. The communication is reliable in that message delivery is guaranteed even in the presence of software and network failures. With Linxter, applications can be dynamically provisioned, their messaging settings dynamically updated, and their communication channels dynamically changed as needed.
The services provided by the Linxter Internet Service Bus are autonomous, interoperable, and scalable. Location transparency was designed-in from the beginning, allowing multiple ISBs to be set up around the globe, with various kinds of redundancy. The specific location of the services is not important, and can change over time based on the loads and security needs of the back-end systems, which can be either shared, dedicated, or even walled-off depending upon the needs of customers.
Aside from all the acronyms, communication plumbing is a serious business as we move into this age of hyperconnectivity (http://hyperconnectivity.com) and the Global Information Grid (http://www.nsa.gov/ia/industry/gig.cfm).
As I stated in an earlier posting, “… even more exciting than the explosion in the number of IP (Internet Protocol) enabled devices, is the way developers will be able to innovate through the interconnection of all of these things.” This is why we created Linxter.
Back to my original question, does platform-as-a-service suffice or do we add hyperconnective MOM-aaS to the mix? I think we all know the answer to that.