Showing posts with label telco mashups. Show all posts
Showing posts with label telco mashups. Show all posts

Monday, July 30, 2007

WITA : An Architecture Standard for Telco Mashups

One big difference between the Web world and the telecom world are standards. I don't mean that they use different standards (although they do), they have a different approach the standards entirely.  Other than the two worlds of standards represented by the Telco and Web world, there's a third sort of standard - an implementation standard.   A "this is how we tend to do this sort of stuff" standard.  For instance, there's really no one PSTN standard - it's just how we have put together many other standards to make something that works.  We need to establish a standard for Telco Mashups; a set of conventions as to how we tend to write these applications.  Since I've been at this, I've noticed a standard pattern I'll call the Web Integrated Telco Architecture - WITA.   From where I sit, the an application written to the WITA standard is more scalable, more reliable and easier to write than anything else I have seen put forward for telephony enabled services. As I put together my telco mashup applications, and see companies like Gaboogie and Twitter do the same, I see that we all use approximately the same approach. If we can standardize it, give it a name, we can build the community quicker. 
 
The telco world has organizations such as the ITU and ANSI; indispensable tools in creating standards that telephony vendors need to design their equipment.  An ITU standard takes several years to  ratify and formalize, and a typical vendor cycle of development lasts around eighteen months. The Internet community runs off of the IETF, and creates nearly all the web standards used in networking, such as TCP, HTTP and FTP.  The IETF is a bit different than the ITU, in that the IETF documents stuff that's already working, so that the standards lag the availability of a product.  IETF standards take much less time to appear as drafts, and Internet application development lasts three to six months. 

Another sort of standard is an architecture standard.  Strictly speaking, IMS is an architecture standard, not an implementation standard.  In the Web world, AJAX is an architecture (or web development technique) that uses Asynchronous Java and XML to make rich, interactive web sites.  Ajax is a standard because that's how people normally them together - there is no committee and there are no plenary sessions.  What is the telco mashup standard? What is an application that uses WITA? I'll define it today, and we can delve into it over the next week. I use WITA for nearly all my telephony mashups, and I've seen many others do the same.

Web Integrated Telco Architectures use the following standard components:
  1. Voice XML: WITA uses VXML to handle inbound interactions with human beings, and most outbound voice messaging applications use VoiceXML to make those messages rich an interesting.    The VXML scripts are delivered by a web services application, and post the inputs the collect to a web services application.
  2. Ruby on Rails : WITA uses Ruby on Rails to implement the web services application. It's responsible for putting the logic on top of the database, and for rendering views to the user in the form of VxML scripts, Web Services APIs and graphically intensive web pages. 
  3. Telco Web Services : WITA uses telco web services to deliver telephony features such as outbound messaging, conference calling, click to dial and SMS messaging.   These web services are called from the Ruby on Rails frame work, and provide the scaling and reliability components of the architecture.
VxML + Rails + Telco Web Services == WITA.

Wednesday, July 25, 2007

Paprika

Recently, the man in the Purple Shirt challenged the community to finally come up with compelling real time communication services. I suppose it was spoken during a moment of frustration for Jeff, as this industry has spent a lot of time and energy making these wonderful IP based technologies, but we are still doing pretty much the same old things with them. For those that do make something that turns him on, he might provide some early-early seed capital, and more importantly, he would provide some visibility and friendship. I applaud Jeff's continuous efforts to move voice technology forward, and as I've said before, I'm here to add my efforts to his. To help developers, Aswath pitched in by providing some pointers to how he'd do it. Well, here's what I have to add to the discussion, and it's a single word.

Paprika

To use voice in a compelling way, recognize that voice is a spice, not a main ingredient. Voice and other real time communications brings out the flavor in some other application, but it isn't the star. A compelling application starts with solving a real customer problem, and unless your customer happens to be a telco, carrying voice probably isn't the issue. The issue is something else. Take any vertical and check to see if I'm right. Here's a classic example: entertainment. American Idol had an innovative idea, which was to make a TV show where the people emotionally enroll in the outcome. If you think about it, it's just like sports: people watch because they care about their team. But how did they deeply involve the audience? They made them vote. How did they make it compelling and unique? They made them vote using text messages, which made the process unique and pandered to their core, young audience in one fell swoop. Brilliant. Compelling.

Why do we have such a voice services focus? My bet is that telephony has been so hard, for so long, that the people involved only know the telecom industry. We don't know the problems faced in other verticals because we've been so focused on the problems in ours. I bet you that if you spent just the smallest amount of time looking at transportation, or financial services or education, you'd find all sorts of places you could sprinkle basic communications into the mix to make it more delicious. Because we haven't done that, Jeff's food is bland and boring. He deserves better, you deserve better, and most importantly, customers deserve better.

Monday, July 09, 2007

Time to brag yet?

No, I don't think so, but boy do I want to.

I caught a link from mashup meister John Musser from Light Reading called Telco Web 2.0 Mashups. In it, Caroline Chappell speaks about the impact Web 2.0 technologies are having on IMS deployments in her new report. I think Caroline what has is mostly right from a technology standpoint, but misses on some critical cultural points. For instance, the mashup phenomenon has, at it's core, this idea of the perpetual beta. Without regard to how mashable the carriers make their offerings, it won't make them any more fleeter of foot. That said, I am in complete and violent agreement with Caroline, and I hope my industry has the good sense to listen to her advice.

Now, if I were to brag, I'd point my readers back to something I wrote nine months ago....