Edgar Landivar, creator and current project Manager of Elastix received several questions from the community this week. In this brief interview Edgar answers questions related with the creation, visions and expectations for the Elastix project.

We'll continue doing more cycles of this kind with other actors of this project in the future.*

Rafael Bonifaz: How does it feel to know that the work you started is being used by thousands of people around the world?

I feel really thankful to know that thousands of people choose our product over many other alternatives available, and it is very comforting to see how we could build a really active community around the project. This also makes me feel committed to maintaining the level of innovation and quality that has been a goal of this project since the beginning.

Did you ever imagine that success would be so big?

To tell you the truth, when we started the project I was very optimistic and we knew this project would be successful, but I will be lying if I say that at that point we knew what was going to happen. Now, knowing that at anytime soon we would reach the million downloads in a relatively short time exceeds all my initial thoughts.

Josiah Spackman: How Elastix begun and what did motivated you to build it when you could have used an alternative distro?

Well, this is a large story ;)

We started PaloSanto as a company dedicated to developed open source businesses, and we've been working in this sector since 1999. At one point our clients started calling us asking for Asterisk based PBXs, we begun installing the operating system and then the components one by one. Soon we realized that there were too many software packages involved in every installation and that each one could become a point of failure.

In addition, customers were also asking us to develop reports for their calls (CDRs) this conduct us to write a LAMP based software to show those reports. We called that software package Elastix and that was the real beginning of the name, in this point it wasn’t a distribution. We started to install it along with other software packages, scripts and other minor things to complement and automate our installations.

When the Elastix reports interface was mature enough we decided to release it under the GPL license and publish it online. We had some downloads but nothing too exciting.

What motivated us to make a distro?

Well, after a few months we decided to analyze some of the implementations we made and realized that virtually every one of them had different combinations of packages and versions which, in a logic sense, would be difficult to maintain in the future. All of these without counting the installation time, which was quite large.

Seeing the potential problem in the future we decided to try some distributions available to see which one to use. The idea was that a PBX distro would solve the problem of maintenance and would reduce the time of implementation. But we would still have to install our reports interface on top, even when this wouldn’t take too much time.

One Friday we had a meeting to decide what distribution to use in our future deployments and all of them had some weaknesses and would involve extra work, so after debating a while someone jumped out with the idea of building our own distribution and make it public. The thing seemed to make sense because we could include the tools we built earlier as part of the distro and recycle the name Elastix.

Anyway the decision was not easy and we decided to do a pilot project for 12 months ... and you know the rest of the story...

To put things in context I must say that in 2003 we had launched a desktop Linux distribution targeted to the local market, but it wasn't very successful. This experience gave us know how tools to create and maintain our own distribution. Certainly, it made things easier for the birth of Elastix and gave us a point to start. It is most likely that without the earlier "failed project" Elastix had not seen the light.

Franck Danard: It is possible to have an "elastix LiveCD" ISO image made by the Elastix community? Something similar to other distros like Ubuntu, etc...

There are people interested in this, but honestly it is not in the roadmap, at least for now. Anyway, if the community is willing to contribute in any way with this work, we would have to re-evaluate this alternative. As always, we are open to ideas.

Frédéric BAGNOL: Do you think it will be nice to integrate BigBlueButton on Elastix?

Of course, we are currently planning future versions of Elastix and the Web-based conference functionality has been a point of discussion. We included an Add-on called Elastix WebConference that has been written in-house, but without video support.

BigBlueButton is a software package that we are evaluating; maybe it is what we're looking for.

Dr. Sichendra Bista:Why Elastix have not switched to FreeSwitch when FreePBX 3.0 has already defaulted to FreeSwitch instead of Asterisk? What prevents you from doing so? I feel like Elastix could add much more feature with FreeSwitch than with Asterisk, don't you think so?

We have been evaluating FreeSwitch for a while and it is certainly an incredible software package. At least in the short term we are just experimenting with FreeSwitch but we don’t have very specific plans about it.

I must say that we are relatively comfortable with Asterisk, so a possible alternative will be to maintain two distros, instead of migrating everything to FreeSwitch, but that would demand a lot of work and that is a factor to analyze.

Rodrigo Perez: I like Elastix, has everything you need, but there is something I do not like, how to develop modules, I find it too complicated. Have you ever thought of using a framework like CakePHP for example? This would allow greater fluidity in the development and has a really powerful and simple plug-in system.

Personally I've made some experimentation with CakePHP and other frameworks but we have not taken the decision to change.

Currently we use a framework called Neo. Neo is a specialized framework for Elastix and the real problem could be that there is not enough information about it. This has been a problem in the past and we are trying to fix it working on more documentation oriented to developers.

Ivan Romero: Does the project have any plans to add security modules and security policies?

We worked hard on security in Elastix 2.0. On one hand our own development process has been modified to check for possible security holes in our stage of testing, and security-related bugs are now solved as quick as possible.

In addition, in Elastix 2.0 we have included a mandatory routine for changing passwords after installation, this will avoid that default passwords used in Elastix can be in the public domain.

Regarding the modules you ask, we are planning to include a pair related to what you mention: a basic module for “telephony-oriented” firewall and a security audit module.

Gotten Gore (ATCOM): We noted Elastix's training program focus on Spanish countries/areas. How do you think the support for non-Spanish users will be and does Elastix have any training plan for non-Spanish areas?

At the moment we are developing the training material in English and we are very happy with the expectation generated with this initiative.

In addition, we will launch the certification training courses online. So we will serve non-Spanish users with our certification program very soon, either physically or online.

Are you planning to build Elastix on hardware?

If you mean to port Elastix into an embedded board, we’ve made some experimentation with Alix and other embedded boards in the past and now we are trying to port Elastix to a Blackfin based embedded board. But it is only experimental, nothing serious at the moment because this project is not on our roadmap.

What is the main advantage of Elastix from your opinion?

There is not only one advantage in my opinion. But some relevant ones are:

- We are focused in the development of a UNIFIED COMMUNICATIONS SERVER, not only a PBX. This allows us to bring more integrated functionality to the end user and of course a better communication experience.

- We have a very active and stable project, growing every day. We are developing new functionality and fixing bugs at a good speed. This also guaranties that the project will be well maintained in the future and the current users can download updates, documentation and new versions for a long time.

- Our business model has been always built around the open source. So the Elastix project will be always free. We have not hidden plans to close the code or things like that.

There are some limitations about concurrent calls with Asterisk. Do you have any good idea for Elastix solving this problem in future versions?

You have to keep in mind that Asterisk is not specifically designed to handle thousands of concurrent calls. It is focused to bring a very rich set of functionality to the companies that would cost a lot of money with other alternatives.

For those large scenarios you'll have to combine Asterisk/Elastix with other components to balance the call load. We are able to handle thousands of calls with this approach and bring a rich set of functionality at the same time. Everything tying to keep the costs low, this is a good deal.

Of course, I am not saying that there is nothing to improve in Asterisk but most of the time the people tend to compare Asterisk with other kind of beasts like large switches and it is not a fair comparison.

Sergio Alecha: When would Elastix start using the Real Time methodology in its configurations?

Good question, it is something we have discussed a lot here, actually we have an experimental library. However, in terms of configuration we use FreePBX and it manages the configuration based in plain text files.

*Note: Questions were answered in Spanish and English and translated to both languages.