Aug 29, 2014

The Role of Open Source in IT/Telco

By David Krhut

There is a strange discrepancy in the IT/Telco word. In certain areas, open source products are joyfully embraced as technology cornerstones, with no one questioning their viability. In others, they are treated as something not to be touched with even a very long pole.

Open source examples that do not raise alarm from IT executive are easy to highlight. Linux, which was first developed in 1991, has evolved tremendously into a foundation platform for many other new areas and technologies (e.g. Android OS). However, from an IT/Telco perspective it is still mostly relegated to server farms, where it serves as a powerful and reliable operation system.

Right now, open source is also winning against proprietary software in the current IT craze of “Big Data.” Analytical solutions can be composed from solely open source components that in some cases provide much higher performance and functionality than multi-million-dollar, “serious” products.

The biggest issue with open source remains perception.

Open source systems are not (yet!) suitable for business critical areas (e.g. CRM, rating, billing) of tier 1-2 telco operators. However, if open source were to be carefully deployed in back-office and/or non-customer facing areas, then these alternatives could definitely deliver levels of productivity and quality that are very competitive with proprietary solutions.

The biggest issue with open source remains perception. Even after so many years and plethora of successful products, it is still seen as something unsecure, created with some obscure technology in a dorm room or garage by a bunch of unreliable students/enthusiasts.

Paid software offers the impression of higher security and stability. “If we paid so much money for this software, it must be worth it!” And if something goes wrong it might be more convenient to shout at someone who took our money than at those lazy students in their dorms filled with pizza boxes …

But maybe there is other factor at play. Given the never-ending migrations, consolidations, upgrades and other system-changing initiatives that are constantly happening in any big IT organization, the proprietary software is often considered a more expensive but less risky investment. The cost difference (when all these possible risk considerations are incorporated) might be not so appealing for the open source movement.

Four key factors go into the decision between proprietary systems versus open source alternatives:

1. Upfront Cost

Open source is not completely free, but it can be cheaper than proprietary solutions. Thanks to our extensive implementation experiences and analysis we have ascertained that the general cost of an open source implementation can be one-half or even one-third the cost to implement a proprietary product.

The main difference is obviously in the license cost. At the beginning the organization does not have to pay the license fee. However, the IT budget should still include expenses for support, installation, training and also eventual customizations of the system, along with other, hidden costs described below.

2. Development/Future Roadmap

Development of certain open source systems can be “spontaneous” and unpredictable. Not all development teams maintain a publicly available long-term roadmap outside parties can use to evaluate whether a system would be still around in three to five years. In some extreme cases, the development team can suddenly discontinue its endeavors—a situation no serious organization depending on the solution can really afford.

Development of certain open source systems can be “spontaneous” and unpredictable.

Proponents of open source will argue that “if you have source code available, you can do anything,” but telecom IT departments usually do not keep extensive (and expensive) internal development teams that could overtake on short notice the development of such an abandoned product.

In the end it mostly depends on the complexity and popularity of the solution:

  • If it is simple, it can probably be maintained/overtaken at no or minimal additional cost.
  • If it is more complex and successful, there are usually sizeable communities and other companies that are already invested in such product (and are indirectly influencing the roadmap).
  • Alternatively, someone may completely overtake the development of a successful product and now only a limited part of the solution is still offered as open source.

To all these points, friendly/hostile takeovers—or even discontinuations—of software products are also an inseparable part of the world of proprietary software.

3. Support

When dealing with support of open source product, the company can rely on four “pillars” – the core product development team, surrounding community/forums, its own internal development team and a system integrator.

Even in a “perfect storm” failure scenario—such that none of these groups can immediately solve the problem for you—it’s likely that 99 percent of issues that might befall you were already encountered, discussed and fixed (or at least workarounds were offered) by someone else from the community.

Having said that, we understand that the statement “we will solely rely on community support for this application” might not look good in IT strategy presentations, so most open source vendors or system integrators come to the rescue with their offer of support licenses that ensure brisk reaction times from their bug-cracking teams. (This comes at a cost, but this expense can be planned in advance.)

Additionally there is a very positive “side” effect of full access to the open source in relation to support: significantly reduced vendor lock-in. The code is available to “anyone,” greatly reducing the vendor’s monopoly.

4. Security

A few years ago we met a customer who during our discussion regarding open source alternatives stated categorically that “open source is not secure, because it is open source.” The statement meant that everybody can see what is inside the code and possibly plant inside some nasty back-door functionality that would then hack/steal everything.

It is true that the availability of open source does not assure that all relevant IT experts will efficiently comb its innards for any signs of malfeasance. On the other hand, the interested clients can themselves see (and compile!) what is inside and whether it complies with established internal security standards. External consultants can also perform security audits.

If necessary, the source code can be changed to eliminate the unwanted parts (although such branching from the main source code can bring other challenges).

Additionally there is a very positive “side” effect of full access to the open source in relation to support: significantly reduced vendor lock-in. The code is available to “anyone,” greatly reducing the vendor’s monopoly.


So in the end, how can we best advise enterprising organizations that are willing to taste the open-source freedom?

When and where possible, try to include at least one open source alternative in the mix of requested systems/applications—if only just to see what the open source community has to offer and at which conditions. Do not consider the open source just because it might be “trendy” for given area. Always ask whether the application really fits into your portfolio.

Additionally look beyond license cost savings (although those are always nice). Consider the planned roadmap, size of the support community, “reputation,” number of integration partners and the development skills of your internal team.

Proprietary software is also not without its risks and open source might give you the much needed flexibility that is missing in your current mix of IT capabilities.

There really are high-quality open source alternatives in areas like (desktop) virtualization, cloud computing, data warehousing, big data, messaging, workflow and rule engines, application servers, CMSs, development tools, trouble ticketing systems, revision control, configuration management, monitoring and, backup and restore. If you are interested in getting more information about any of these areas or receiving general advice regarding where, what and how, please do not hesitate to contact us. 

comments powered by Disqus

Innovation meets performance.