DataDear was a project funded by FUSION, the national funding programme administered by the Malta Council for Science and Technology. The project developed by the DataDear Consortium enhanced innovative tools which help SMEs transition their business data processing to the cloud. Scope Solutions led this project and the University of Malta was a partner.
The University’s academics contributed
with research into cloud architectures and systems, offered technical advice,
and also undertook outreach activities targeted at SMEs.
The DataDear project created a series of innovative tools based on state of the art technologies to help SME end users access and process data stored on the cloud without reinventing their internal processes and reporting, and retaining the ability to operate with familiar software such as existing accounting and spreadsheet software. The product enables the connection from a spreadsheet to multiple Software as a Service (SaaS) cloud applications and facilitates the exchange of data between them in a robust and controlled environment.
One of the primary target markets is professional accounting practitioners, who invariably have practiced with spreadsheets for a significant part of their career; accountants are also an ideal target as they constantly need to integrate data from various applications to deliver effective financial reports.
The University team consisted of Prof. Kevin Vella and Dr Joseph G. Vella from the Faculty of ICT. The project also included Mr Thomas Mercieca, who concurrently undertook a master's at the Department of Computer Science.
The team worked cohesively on the main work packages and one can identify the following principal contributions: movement of data between, and the coupling of, SaaS applications in a seamless method with acceptable computational performance and data consistency; understanding the business activities workflow and how the interaction can best be improved for benefitting the quality of end users' computing work when using an application like DataDear with SaaS software. Furthermore, consistent and frequent interaction with Scope Solutions, a local leader in accounting solutions related to SaaS environments, helped to improve the focus and context for the project’s objective to be met.
An analysis of the workflow process, for example how an accountant can produce a costing of a new service, can greatly improve the efficacy and efficiency of an activity. Furthermore when encoding of the processes required to orchestrate the steps for data retrieval and the computation of results is undertaken one can then ensure that the workflow produces proper results and halts if any of its dependencies are lost or some required data is inaccessible. The research here was supplemented with a careful analysis of spreadsheet models that typically show a workflow output, based on a representation grounded mathematical graphs, so that the spreadsheet model can be verified against some of the many known issues associated with spreadsheet’s use. It is worth reminding the readers of a number of cases where using a spreadsheet has led to embarrassing and expensive blunders.
Another technical challenge is ensuring that the interaction with cloud services is transparent to the end-users and that it is computationally efficient in terms of response time and also on the resources required. In this work, we have demonstrated how a number of architectures, based on microservices, are available to build the processing pipeline, and how multithreading can improve one's interaction with the spreadsheet client. Mr Mercieca's research specifically dealt with ensuring that complex data interaction, for example, those involving OLAP queries, can utilise data servers’ architectures to deliver better response times, and also supporting a number of reporting capabilities.
Yet another strand was the investigation and proposal of having a spreadsheet posting back data to the SaaS data servers that had initially provided the input data. The real issue here is respecting the freshness of data, i.e. data that was retrieved from the cloud may have since changed, and ensuring that posted data adheres to the business rule currently in place at the SaaS database. A few mechanisms have been identified and a possible choice should be based on actual data interactions.
More information can be requested by the project coordinators.