Digium announces closed handsets

I did a serious double take when I read the PR this week about the plans that Digium, the original developer of Asterisk, has to enter the handset market with 3 devices in Q2 this year.

Yet another SIP handset with a new logo on it isn’t the surprising bit and until I read the detail I assumed that this would be some sort of soft Digium OEM branding of hardware from an established player. It is the sort of thing that we have thought about a few times and would allow them to exploit the Asterisk or Switchvox brand to give them more handset revenue whilst keeping their open credentials intact by continuing to play nicely with the rest of the handset market. That isn’t what these are. They are Digium developed and manufactured and will apparently have proprietary features specific to closed source Digium applications. If that is true then it really is a bold departure.

As most folks will know Digium has a range of products which, up to now, have been mostly focussed around the open SIP VoIP comms market based on its early leadership in this field from developing the Open Source Asterisk code base a decade ago. Their original business model centered on monetising the ground breaking Asterisk code by selling telco interface cards to folks that used it, and latterly from a commercial PABX based on Asterisk called Switchvox which they purchased in 2007.

I guess that this isn’t a great place to be commercially any more. Digium have a number of strong competitors in the interface card space with no real differentiation other than the Asterisk brand association. The importance of these cards is in any case declining at their CPE end of the market which is moving at great speed away from local legacy telco interfaces towards SIP trunks. So they really need to make Switchvox work for them which is what I assume this is all about.

When we developed our Asterisk based PBX product in 2004, we could see the prospect of a huge market shift away from expensive and inflexible PBX systems which were at that time all using proprietary lock-in and exploiting the lack of standards between the PBX and the handset to keep the customer captive for all parts of the solution. Asterisk was a glimpse of the future, it’s adoption of cloud friendly open SIP standards broke this lock-in and enabled us to develop products which were flexible and worked with the many emerging SIP handset vendors like snom, Polycom & Sipura to deliver very competitive end to end solutions. Yes, there was an engineering problem to solve to make all these open handsets behave seamlessly but this was not that hard although it was resource consuming. Today we automatically discover and provision over 70 different handsets from half a dozen vendors but maintain compatibility with many more due to our use of exclusively open interfaces.

In the meantime purchasers have recognised the distortions caused by proprietary lock-in and even the old stalwart vendors now boast open SIP handset interoperability because it is now a “must have” on many procurement specifications.

From early information on these handsets, it looks like Digium may have launched a bid to rewind time and take Asterisk back into the 80s and 90s world of closed proprietary phone systems to make a few extra bucks on handsets in Switchvox sales. If they have then it will be interesting to see how that works for them.

This early information is very sketchy and more detail is promised soon on how the handsets will integrate. It could be that Digium intend to use their own handset designs to pioneer open functionality that gives smoother operation of features between handset and switch. If so then that will be great move for entire industry and we look forward to being able to add support for three more handsets that provide a further proof point to our customers that open solutions are the way to go! Who knows, Digium may be able to build a great business out of providing well thought out open handsets that create as much of a buzz as Asterisk first did a decade ago.

 

About Rob Pickering

Rob is a reformed software engineer who has spent much of his life developing computer networking applications and infrastructure. His career started in the 1980s with hands on development of the Internet TCP/IP protocol stack through a spell as a development manager at 3Com in the 1990s before founding ipcortex where he is currently our CEO.
This entry was posted in commercial, industry view and tagged , , , . Bookmark the permalink.

2 Responses to Digium announces closed handsets

  1. Steven Sokol says:

    Rob,

    Thank you for covering our recent announcement. The launch of our phone product line is big news for Digium and the Asterisk community, and we’re happy for all the press coverage we can get. I could not help but notice both the use of the (rather inflammatory) word “closed” in your headline and a number of inaccuracies in the text. We have not abandoned our roots and we have no intention of pursing a “lock-in” strategy. Our phones are far from “closed” and, in fact, represent a significant opportunity for Asterisk users, integrators and solution providers, including IP Cortex.

    First, lets get one thing straight: all of our phones use the open SIP standard for communications. They are not proprietary. We have concentrated our testing efforts on Asterisk and Asterisk-based systems like Switchvox, but they should work with any SIP compliant call control platform. Digium remains as dedicated as ever to the power of open standards, including SIP.

    The primary driver in our decision to build our own phones was the lack of control over the user experience provided by third party manufacturers. The current crop of third party SIP phones are inflexible — they lack the necessary level of granularity in configuration to create a truly awesome user experience. Menus are not context sensitive. Button actions are fixed. Worse yet, there is no way to add any form of application-level functionality beyond what the manufacturer includes in the phone’s firmware.

    We’re changing the paradigm by including an open API that lets end users and integrators (like you) create apps that run on the phone. And no, this is not just another lame XML-based “micro-browser”. Our app engine runs real JavaScript code that can interact directly with core phone features and respond to phone events. Apps can access data on the PBX or anywhere else using HTTP requests. They can run in the background or foreground. They can respond to SIP messages, button actions or state changes. You will even be able replace the standard phone apps with customized versions that tie into your PBX if you wish. See if you can get an Aastra or Polycom phone to do that.

    Another major reason for building our own phones was the ridiculous complexity of deploying and configuring third party gear. Yes, it can be done, but all of the standard approaches are utter hacks. You can use the old “NMAP” trick where you use a hacker’s tool to search your network for phones, but that’s haphazard at best. You can use the “Option66” DHCP setting — if your customer’s router or DHCP server supports it. You can also snipe DHCP requests but that depends on intercepting the requests and sometimes fails if the real DHCP server beats you to the response. Regardless of the approach you take, you have to set up a configuration server, craft configuration files and keep them in sync with the call control system. Yuck.

    There are no broadly accepted industry standards for discovery, provisioning or firmware updates. Digium’s answer to this challenge is the DPMA or Digium Phone Module for Asterisk. DPMA is an add-on software module that simplifies the aforementioned mess. Our phones use the Bonjour service discovery protocol (aka Zeroconf / mDNS) to find servers. They use SIP to communicate with the server. The DPMA handles a special class of SIP messages related to provisioning and application data transfer. The payload of these messages is encrypted, both to prevent the capture of passwords by nefarious intermediaries and because Digium needs to prevent the wholesale cloning of our phones. (Open source is not a suicide pact.) The DPMA, much like our G.729 and Fax For Asterisk modules, is commercially licensed. However, we’re not charging for the DPMA. Licenses are freely available to end users. We’re also more than happy to provide free redistribution licenses to manufacturers of Asterisk-based systems. Even yours.

    Digium is a successful business that has grown significantly in spite of a sluggish economy and a changing dynamic in the communications industry. And, yes, phones represent a large opportunity for Digium as well as our ecosystem. We estimate that there are over 1.3 million new IP phones hooked up to Asterisk-based systems each year. If you assume that the average price for an IP phone is $120, that works out to over $150 million each year. And that number is growing. I challenge you to consider the opportunity that these phones can bring to your business. With the features included and the price of these SIP-based IP phones, we have received an overwhelming response of partners desiring to integrate these phones into their solutions. I think you’ll find that you too will get better phones, better margins and happier customers.

    If you’re interested, please do give us a shout. We would be happy to work with you.

    Cheers,

    Steven Sokol
    Marketing Director
    Digium, Inc.

    • Rob Pickering says:

      Steve, sorry I missed this comment at the time that your made it.

      Thanks for the clarifications on the nature of the API support. Smoother integration between PBX and handset is certainly something where there is plenty of room to improve the end user experience.

      I’m sorry if the headline caused offence, I certainly didn’t intend to be inflammatory, merely reflect my surprise at the apparent proprietary nature of the handset provisioning interface. Thank you for your clarification on the way that you intend to licence this, I can kind of understand and certainly sympathise with your commercial reasoning!

      As to our integrating support for these devices, I’d be more than happy to discuss this with you. The situation isn’t conceptually any different to other vendors who make provisioning details available only under NDA etc.

      Thanks again for the comments.

      Rob.

Leave a Reply