Cloud Source
Showing results for 
Search instead for 
Do you mean 

Is SoMoClo the next buzzword? Maybe, but where is sensor data?

on ‎07-10-2013 12:43 AM

Somoclo.jpgEvery 5 years or so, the IT industry comes up with a new term to represent the industry/technology evolution. Do you remember mainframe, mini, client/server, DCE,  object orientation, SOA, web services, cloud …. And I could go on like this. SoMoClo, a term representing social, mobile and cloud, is, according to Aberdeen and a number of others the next term to use (or hype?).


Fine, it’s an opinion. But beyond the barbaric term, does the integration of social media, mobile devices and cloud represent the evolution of IT. In my mind the answer is no, and I will explain you why.


Living in a connected world

It is an understatement to pretend the world we are living in is increasingly interconnected. When I joined HP more than 30 years ago, our customers were flabbergasted that we could get a response from our US divisions in 3 working days. This happened using an e-mail environment. Today, if I do not have an answer within 10 minutes I’m getting nervous. Time has been shrinking.


But that is not all. Our motorways are increasingly intelligent, and automatically calculate the best throughput speed while advising GPS systems of traffic density. Airplanes transmit terabytes of data while flying to speed-up ground & maintenance procedures. You have the current weather conditions in any city around the globe at your fingertips. And there are many other such use cases.


Let me give you a simple example. Let’s assume for a moment you are responsible for procurement in a food company manufacturing apple based products. You are in the northern hemisphere and typically buy local apples at the end of the summer and New Zealand apples early in the year. Frost when the apple trees are blossoming is damaging for the harvest. Now, in your local area it is easy to know harvest might be endangered, just look out of the window. But in New Zealand, it is somewhat more difficult. So, you may want to monitor frost out there (using weather information) and the state of the apple trees (using for example pictures shared online or monitoring social media).  If you get to know of potential damage, you may want to secure supply, either in New Zealand itself or elsewhere in the world. Being the first to know gives you the best chance to ensure your availability of supply.


This is just one example. It is by combining information from multiple, unrelated sources, that you can draw conclusions addressing your concerns and make the appropriate decisions.


Three categories of data

Cloud is a mean through which you can access and analyze data. All in all, there are three types of data:

  • Enterprise data (often called Business Intelligence),
  • Social Media information,
  • Machine/sensor data.

Each data type has its own characteristics. Let’s take a moment to discuss them in more details.


Enterprise Data

As its name indicates, enterprise data is gathered by the company itself and often stored in databases and other structured environments. Most of the data is numeric, although we all know the character fields containing comments. The data is typically stored within the company and often associated with one or another of the key enterprise applications such as ERP, CRM or financials. We have analyzed that data for many years now and enterprise controllers or analysts know all about how to exploit it.


Environments such as SAP HANA provide you with powerful analysis tools to understand how your enterprise behaves, where your bottlenecks are, where you can improve etc. It allows you to identify the elements that influence your company and its eco-system. But you are depending on the data you can collect yourself. Most data items are structured (often numerical) although some comment fields allow the users to provide some hindsight about what happened at a given moment in time. It serves as a source helping you take decisions.


Recently, new sources of enterprise data have started to be exploited. These include e-mails, call center conversations etc. Being unstructured, they required new analysis methods. In most companies, such analysis is mainly done for compliance reasons, but these sources can add value elsewhere.


Social Media Information

Social media information is the exact opposite of the enterprise data. By definition it is stored in the cloud and it is unstructured. It’s a combination of Twitter, Facebook, LinkedIn and other social media platforms, combined with the content of forums, blogs, wikis, hangouts etc. Here you are not looking for a number, a value, but for the sentiment expressed in a sentence. It’s human data if you want and it needs to be interpreted to understand the tone of what is being stated.


Sales and marketing people are very interested in this type of data as it allows them to understand what their customers think about their products, their services and their brand. But it requires a totally different set of analysis tools.


In a holistic view of the information sources enterprises may want to have access to, I’d like to add TV, radio, press and other information sources to this category. I know they are typically not included here, but for the purpose of gaining a 360 degree understanding of your market, it is important not only to limit you to the traditional social media sources, but also to include other, equivalent sources of news and information. Very often news sources represent the views of journalists and organizations in the same way as tweets and blogs, except journalists have a certain deontology ensuring the quality of their information. But that is a whole different discussion.


Machine/sensor data

Most of this information is generated by external sources and is very often accessible for free or for a small fee from external sources. Being it the movement of mobile phones from one tower to the next one, allowing you to identify the speed at which traffic goes through a road segment, being it weather conditions, water temperatures, pollution information, container locations, you name it, once digested, this data can provide you with valuable information, even giving you competitive advantage. Most of this information is structured, often numerical.


Increasingly devices and infrastructure are equipped with sensors. They generate the vast majority of this information. But we are just scratching the surface on exploiting it. New businesses, new services will be created from using the data for specific purposes. An article in Discover Performance from last April, titled “From the Internet of Things, a business intelligence bounty” describes some use cases related to the analysis of sensor data. But I believe we can go one step further by integrating all three categories so enterprises and end-users can understand their environment and act accordingly.


Exploiting data for competitive advantage

A number of years ago, we started analyzing warranty records entered by our own engineers and third parties fixing some of our products, to gain early warning of potential issues with new products. Combining the results of such analysis with the final assembly test results allowed us to reduce the size of our warranty reserves drastically.


Monitoring social media information, allows us to spot potential issues even more quickly and act upon them. We no longer have to wait till a warranty record is created to see something is happening. Taking part in the social media dialog, understand what happens and doing something to correct it, not only allows us to spot issues quickly, but also demonstrates responsiveness.


Again, this is just an example. In a just in time environment, understanding when a truck needs to arrive (enterprise data), what the traffic situation is along the way (machine/sensor data) and whether there is a potential social unrest (social media data) that could affect my shipment, can keep my production lines going.


Over the last 15 years we have gone from a period where we did not have enough information to take decisions to a complete data overflow. It is only now that we are starting to understand what we can do with the available data. The question is now which tools and analysis methods to use. The holy grail consists in a combination of structured and unstructured data analysis tools. And that is where HP’s HAVEn technology stack delivers its value.


Is SoMoClo the right term?

I hope I showed you the importance of the combination of machine/sensor data in the mix, and I gave you a couple examples of where they really made a difference. I feel the term SoMoClo misses the point as it does not references that aspect. Yes, in some of the associated diagrams, there is a small M2M sign included which could be associated with what I talked about.


My question is twofold. Do we need acronyms to highlight where we are going? If we really do, something like SeDaMoCloS or MaDaMoCloS would probably be more accurate. But this starts to sound a lot like Damocles, isn’t it. Would have all that data at our fingertips and not exploit it be our 21st century “sword of Damocles”? Maybe…




If you want to understand more about HAVEn, read this whitepaper.


0 Kudos
About the Author


on ‎07-11-2013 10:22 AM

As a former SoMoClo co-conspirator, I'd point out that So stands for "social", not "social media."  Social and networked interactions are much larger than social media; they include voice, video, and log transactions as well. 


Also, in light of that larger social picture, you say that "there are three types of data

  • Enterprise data (often called Business Intelligence),
  • Social Media information,
  • Machine/sensor data"

But this leaves out the data making up the majority of bandwidth: voice and video.  This streaming data represents a fourth type of data that seems to fall off the radar.  All four types of data are more valuable when they are part of a network topology rather than static or broadcast, as well.


And do documents count as an additional static form of data or do they get pooled in with social media?  The interactions between people, applications, and documents form the enterprise social relationship topology and represents an important aspect of the challenges of new IT, but this ecosystem is still poorly understood.



on ‎07-11-2013 11:22 AM

Hyoun, I want to react to your message as I believe you may not have understood my blog entry completely. First, when I speak about sensor data, I do NOT speak about log transcripts etc. I do speak about data gathered from sensors, being it along roads, being it temperature, pressure etc. Any sensor information is welcome, and that goes way beyond what the word "social" means.


Second voice & video. I actually talk about call centers in the enterprise data (and that to my knowledge is voice), I do talk about extending social media to media communication, and that includes voice (radio) and video (TV). 


Yes, I admit, I did not specifically mention voice and video in the sensor data, as I did not use one of my favourite sensor data examples around surveillance cameras and how that information can be automatically analysed so the owner can be warned if something strange happens. Appologies for that omission. 


What I mean is that voice and video are formats of data, and these formats can be present in all three types of data. When I talked about data types, I did not specify their format, as in my mind there are hundreds of different formats these data items can take. Documents are another format of data, enterprise reports are enterprise data, poems, love letters are social data, and I could go on like that.


Hope this explains it a little better. Thanks for commenting.

on ‎07-12-2013 09:43 AM

To Hyoun's point, 'social' is defined by as bi-directional connectivity. A sensor, whether fixed or mobile, that is connected to the cloud and sends or receives data, whether wired or wirelessly, is thereby social. 

on ‎07-17-2013 12:16 AM



Frankly, good try, but if I go back to the definitions of social, it does not work. Indeed, all definitions of the ther "social" relate to society and its organization. To give you one example, look at the one of the free dictionary. That is why "social media" is using the term quite correctly.


Sensors have to do with our environment in the widest sense of the term, not with our society. Since I wrote this blog entry, I have been thinking about the subject quite a lot. In my mind, the spilling of social information through social media is a temporary thing. As people better understand the implications to their private life (yes, I am a European, so sensitive to that) they will change the wya they use the technology for it to become a vehicle of communication rather than a place to publicise ones social live.


OK, news and media will continue, and that is why I do not advocate the removal of the social element in the discussion, but I'm just pointing out there is a whole new world that is omitted. And that world is related to the "sensing" of our environment. Frankly, I don't care how we call that. At HP we speak about CeNSE, the central nervous system of the earth, as a vision, and I'm not in any way form or shape advocate that to be the name, but beyond a name, I want to point out we have a brand new revolution at our doorstep many of us seem to completely overlook. 

Nov 29 - Dec 1
Discover 2016 London
Learn how to thrive in a world of digital transformation at our biggest event of the year, Discover 2016 London, November 29 - December 1.
Read more
Each Month in 2016
Software Expert Days - 2016
Join us online to talk directly with our Software experts during online Expert Days. Find information here about past, current, and upcoming Expert Da...
Read more
View all