There was a time when Jeff Pulver will periodically lament about lack of new and exciting features and will talk about Purple minutes that became Purple Applications. A few days back, David Beckemeyer raised the same concern. He points out that even though he has deployed many “Class 5 switches” via PhoneGnome with easy to use API that can be used to offer new and exciting services, only a handful have taken advantage of them. He admonishes the rest of us to exploit the PhoneGnome platform and develop features. In response to that, PhoneBoy wonders that it could be because there is not much demand for these features. Alec Saunders is a bit more optimistic and sees a new day dawning. My take on this is a bit different.
David quotes Simon Torrence as saying that with PhoneGnome one can “[deploy] far more features than could ever be deployed in IMS, and deploy them within weeks, not years.” The first part of that claim may be true, but I am skeptical about the ability to deploy them quickly. The reason is that PhoneGnome as do almost all standalone VoIP clients project the same limited interface to the users as the PSTN does. If it is difficult to develop feature in PSTN, then there are two main reasons – feature interaction and educating the user to invoke the features. PhoneGnome has not addressed either of them. This may not explain why the developers are missing in the first place; but PhoneGnome will face this issue quickly if David’s wish is granted.
Consider the service David quotes in his entry – Tellme DialTone 2.0. I do not have dialtone in my cordless phone; neither does the cell phone. Why do I need it for VoIP? It is nice that they use speech recognition to dial the number for me. But how do I say “not that John Smith, but the other one”? Wouldn’t having a screen based phone and the ability to select from my address book be simpler and cheaper? I may not have explained why the developers are missing, but I am pointing out that had they come, the situation will be same as it is in PSTN. So I say that we need to pay attention to user interface and simplify feature invocation.
Posted by aswath at February 2, 2007 08:04 AM
I disagree Aswath. if your app needs a diferent UI, it is easy with the Gnome platform. Even today, with a Softphone, one seamlessly uses their PC in place of the analog handset and can still place calls using the analog POTS line transparently. I'd argue there are MANY MANY cool things you can do without a radical UI replacvement, but if your awesome idea needs it, then no problem.
This is an exact example of one of the platform benefits - it unbundles the handset from the POTS line service (telephone service). There's a VAST ARRAY of things that can be done on this platform. You are presenting just yet one more excuse for developers to "wait for some magical future" that may well never come.
I also disagree that our platform "does nothing for feature integartion." Look how Iotum interacts with the ITSP selection feature automatically, how it integrates with the Vociemail options, and many other platform features, with no special effort required by iotum, etc.
I'm disappointed in your superficial analysis - it is very atypical for you.
Posted by: David Beckemeyer at February 2, 2007 11:51 PMDavid:
Probably there is a disconnect between what I am trying to say and what you are reading into it. Let me try once more, but I am not sure that I will succeed.
My point is that if the user can invoke the features only via a regular phone (with 12+1 buttons), then the features that can be developed will be limited for the same reasons that they are limited in the PSTN world – feature interactions and customer education of the UI. So we should be focused on changing the CPE whereas the focus so far has been in network infrastructure. I have expressed this many times before.
You may be right that many cool things can be done with ATAs themselves; if so I do not know why there have not been very many takers.
When I referred to PhoneGnome, I meant the ATA only and I suspect that you mean more than that – that explains why use the phrase “platform”.
I did not say that your platform does nothing for feature integration; I said PhoneGnome does not solve the feature interaction problem. How can it, when the problem arises from the limited UI at the phone. Take for example the feature interaction between call waiting/retrieve and three way conference when the second call is an incoming call. In PSTN (I suspect in PhoneGnomeas well), it is resolved by allowing only call retrieve. Contrast this with what iPhone is able to do (as can be seen by the demo at their site) – they are able to display two buttons – one for retrieving and one for bridging.
You may still think the analysis is superficial, but I strongly believe that changing the CPE is the one that is going to bring in the benefits of VoIP. One has to take a look at Verizon’s iobi service to see that.
Aswath:
I think you have this exactly right. Innovation on and in devices makes lots of sense. Devices which run lots of communications apps in addition to voice make sense.
Using VoIP both to drastically reduce the cost of calls and enhance connectivity of mobile devices (WiFi in buildings, for example) is innovation which extends the usefulness of vocie without adding complexity.
But there won't be a Voice 2.0. More about that at http://blog.tomevslin.com/2007/02/voice_20_fugget.html
I have nothing against innovation at the CPE, but feature development shouldn't be put on hold in the mean time.
One point I tried to make is that users are not limited to the 12+1 buttons of their phone as the only interface. This is one of the really cool things of the platform, how it ties web apps, or apps on any platform, to the phone experience. We have apps in Java on mobile phones that use the UI of the phone but are coontrolling PhoneGnome activities. Web-based apps, or apps using any GUI can interact with and manage telephony via the PhoneGnome platform, certainly to include the iPhone feature you describe for call accept/reject/bridge (or whatever) as fairly a trivial example.
Posted by: David Beckemeyer at February 4, 2007 05:31 PMThere is really no such thing as Voice 2.0. Saying so assumes that one we are building on Voice 1.0 and two, voice communications differ from other kinds of communications such as video or text. We actually should use the term CoIP (Communications or Collaborations over IP) which expresses our future communications better. I elaborate on this in my blog at http://flatplanetphone.com/wordpress/?p=38
Posted by: Moshe Maeir at February 5, 2007 03:19 PM