As enterprises move more and more of their data into the cloud, the reality is they lose more control over the data they store on services from companies like Netsuite, Oracle (News 
- Alert) and Salesforce. In order to get programmatic access to your data you need to become familiar with the APIs of each vendor and then deal with the intricacies of what information they make available.
And let’s say you decide to switch from one cloud vendor to another – you have to first get the data out into a new proprietary cloud format and learn a new set of APIs to get at the data.
The situation isn’t unlike the database world before the advent of SQL – where a single, slightly altered database language worked with a variety of databases. To make life even easier for database developers, in the late 1980s and 1990s, a variety of fourth-generation application languages, or 4GL, programming tools emerged which were part pseudocode-like language and part SQL. It was like Pascal or even BASIC met SQL.
A typical program would use an SQL select statement, which would feed into a While Loop in an application allowing a programmer writing an invoicing program to select all customers with an account balance greater than zero, for example.
One of the common 4GL programming languages at the time was made by database company Informix. The co-founder of the company, Roger Sippl has a new venture that could bring the power of 4GL to the cloud. I had the chance to catch up with him recently and learn about how – for the last four years – his company Connection Cloud has effectively positioned itself as the database middleware of the cloud.
Rather than having to learn each language, a programmer uses an ODBC driver from Connection Cloud which deals with the intricacies of each cloud vendor, allowing a virtualized representation of the data to be seen by programmers and users. As you can imagine since it is a service, you can write queries which access data from multiple clouds at once and even tie the results into reporting apps, spreadsheets or any other software. You could even write queries to take data from some clouds, massage it and write it back to the same clouds or even other ones.
Of course you could use the cloud data for any reason – even to provide data to local apps. These decisions are totally up to you. Moreover, besides SQL, you can use web services and JDBC to access the data.
The company has been the report-generating back-end of cloud-based subscription management system Zuora (News - Alert) for over a year, but its future business model will likely rely more on end-users and programmers. Connection Cloud is still in beta but will be out in about a month. At that time, pricing will likely be around $99/seat – a free version and/or free trial may exist as well. There will also be workgroup and enterprise pricing for volume purchases.
There are currently connectors for Salesforce, Zendesk, Intacct, Google (News - Alert) Docs and Spreadsheets, Netsuite, Zuora and Facebook. And this data can be used in applications such as Excel, JasperSoft, Yellowfin, Tibco and Tableau. Expect more connections to be added over time.
In addition, programmers will be able to use languages such as Java, Ruby on Rail and PHP. According to Sippl, Connection Cloud allows these languages to act just like 4GL– as if the database was located in MySQL under the desk.
Other connectors are on the way soon – the company actually describes its offering as Data Liberation Services.
One of the major challenges the cloud presents to companies is that they feel a loss of control over their data. Connection Cloud should seriously alleviate these concerns. In fact, by abstracting cloud data, they make the move to the cloud much more attractive.
Moreover, such a tool is a natural enabler for big data and analytics data, which can now easily leverage cloud-based services. In other words: If getting to the data is now much easier, companies and third-parties can spend their time coding applications that sift through the data in these clouds to determine which customers may be most likely to purchase something soon or which may be contemplating cancelling a company’s service based on changing usage trends. I can envision cloud data becoming far more important in customer service, helping with upselling and retention as a result of companies having easier access to the virtual goldmine of information treasure locked up in their hosted applications.
Edited by Brooke Neuman
