I couldn’t attend nor even follow the stream at Sapphirenow, but I picked up a few tweets on Integration. Well actually, Seb pointed one out to me.
As much as I detest it, I’ll have to base this post on the limited info I retrieved – although I did browse the usual placeholders for SAP news of course.
If you read my latest post on SAP and Integration, you might presume I was a little overexcited about what was going to be announced
Well, the excitement wore off. Really off
I would have expected for SAP to take Integration seriously. In the light of the new developments, i.e. them buying a few billion dollar of SaaS HR, the ambition to seriously go mobile and the current fuss about all their disparate Integration solutions, I would have like them to come up with a single solution
In stead, they talk about Cloud Integration. I understand they want to build / compose their own Integration platform. SAP has been trying to do that for years, and it has brought them some kind of integration but certainly not the kind they need, let alone what they need ofr the future – that is, the future as I see it, so disagreement will abound there.
From the SAP event newsroom:
SAP recognizes that for many of its customers, heterogeneous IT landscapes and deployments across on demand and on premise will continue to be the norm and it is crucial to provide integration to make such a hybrid solution landscape work. To address this need, SAP intends to deliver a cloud-based integration technology, comprised of on-demand solutions for process integration and data services, with out-of-the-box content to connect the loosely coupled line-of-business on-demand solutions to other SAP solutions whether on premise or on demand. For integration to third-party solutions, SAP plans to offer its own cloud-based integration technology and also plans to enable its vast ecosystem of partners, including solutions from Dell Boomi, IBM Cast Iron and Mulesoft
I think the latter three are legacy (customer expectation) that came with Successfactors, or at least I hope they are. Boomi promises code-free integration with o.a. Atomshpere but is a very basic point-to-point integration solution that only works with known applications on either side – not surprisingly.
IBM Cast Iron? Again, the “Configuration, not coding” meme. By the looks of it, it connects to SAP business-in-one and ByD.
In Enterprise integration, about 99% of what you deal with is unexpected or exceptional – and you can’t configure the unknown.
Even though handcoding has been admittedly absent from the Integration field for decades, when it comes to making a data mapping, apparently this is a differentiator for these two.
Cast Iron also seems to advocate the direct hard-coupled approach of linking two applications together. It’s an approach that nobody will use at that level. For POS systems you might want to ram your connector in there and retrieve or even update some data yourself, but on an application level I’d strictly exchange service requests and replies.
On a whole, Cast Iron vaguely reminds me of Pervasive’s Data Integration – yet on a much, much smaller scale.
Mule? That one has been around for a while, and slowly gaining enterprise track. Mule iON makes sense to me, and I wanted to have a peek at their System Integration Edition but that only led me to a form after which I got “Thank you, you’ll be contacted”
Back to SAP’s Integration roadmap – or is it merely a Cloud Integration platform as it states?
Yet another solution to the Integration challenge, or rather, as I said earlier, it seems yet another disparate workaround to an issue, rather than a solid solution to an Enterprise challenge.
Where SAP makes their entry usually based upon Buy Before Make, I don’t understand why they choose to make their own Integration platform where there are so many proven technologies out there.
TIBCO, Axway and Pervasive are what come to mind – yet the real solution doesn’t lie in offering a single vendor (although that is not within the mindset of most SAP people, I reckon).
Furthermore, the question remains: what will happen to all the different Integration opportunities already present within SAP? From what I understood from Vishal Sikka, these will remain, at least for now
What SAP could have done, is shed some light on what the future is for all their Integration solution. If this is it, I might have misinterpreted the information and need lots, and lots more. Will this be their single integration platform regarding of vendor, location, and device? I didn’t read that memo anywhere
If SAP were to go my way, they would have made their own stack Integrationable – instead of presenting a technology roadmap for their Integration platform. What do I mean? I’m talking about creating a humanly legible service repository across the board: what services does SAP offer where?
I am talking about business services of course, amply describing the business rules and exceptions for each – the very start for any Integration. If these are logically and functionally known, the syntactical variations in which these can be subscriped to can be disclosed. Add to that the necessary mechanisms of acknowledgement and error handling, slap on versioning, and it would all become a matter of Business decisions rather than Technological ones – and change the Integration approach from bottom-up to top-down
With such, and of course a price tag to each variant, SAP Integration would become a whole lot more insightful and intelligible. Integration vendor, tool or platform would be free of choice, and even mid- and long term ROI would be able to be calculated. The necessary tech dynamics would remain, but the current volatility in SAP Integration would disappear; a very important step if you ask me
Maybe that will come tomorrow, as we still have one day left of Sapphirenow. Mind you, I would have given a limb or two to attend but it simply wasn’t doable. I may have misinterpreted some facts as a result, and am willing to stand corrected as usual.