I was referring to voice instant messaging (using VoIP) as a revenue opportunity for operators, not VoIP as a replacement to circuit switched voice.

In voice instant messaging we agree that there is no real-time QoS requirement; moreover my argument is that with voice instant messaging there are a lot more use case catered for, where voice is used as the evolution of IM for mobile consumers.

Andreas

A very thought provoking piece, thanks.

While I agree that the carriers need to consider how to embrace VOIP, you somewhat glossed over the fundamental problem with VOIP. Quality of Service.

The PSTN is tuned for end to end voice communication. Some network operators are beginning to offer end to end digital voice services (we have one from BT and its still not quite there yet in terms of end to end voice quality).

But the issue with internet based VOIP (and I use it a lot, Skype, ventrillo and others) is that it does not match the end to end quality of service offered by the traditional PSTN because its routed over a network that is not tuned for voice packet transmission, and no amount of clevers by Skype at the end point can totally hide this fact. In fact at a fundamental level TCP/IP cannot deliver end to end real-time QoS (it was never designed to), so the network has to try and hide this fact by using bandwidth to hide the issue, but at a basic technology level bandwidth just does not solve the problem.

That said. For VOIP based services like voice messaging, there is no end to end real-time QoS requirement. So these are indeed revenue opportunities. But the challenge for carriers is how to build hybrid systems delivering an end to end realtime voice comms capability mixed with a VOIP app service, while not killing their revenues or losing quality of their service in their current baseline revenue earner. Either that, or

1. somehow get VOIP QoS to match PSTN QoS – not sure how unless every network infrastructure provider does what BT his doing with their 21st Century program for IP upgrades. It will happen, but the current economic downturn will reduce the investment needed to make this happen and so this will take a while.

2. Get the whole world to accept a reduced QoS for realtime voice calls. Not likely!

So, in short, good stuff, but lets not get rose tinted spectacles just yet.

Geoff

I 've updated the post to reference the use cases showcased by Solaimes – very innovative indeed, yet rather simple to implement.

Andreas

I agree with your points but I understand how difficult is evangelizing about them

a) PTT has a lot of use cases not cannibalizing circuit switch voice calls, and could be offered even being conservative only as a service enabler (mobile server interaction for UGC and CRM purposes), not cannibalizing mobile to mobile cs calls.

http://www.youtube.com/watch?v=wI_nPT0xAcs

b) Voice, CS, PTT and VoIP have still potential helping to increase multimedia services as improving browsing with a multimodal approach not needing a on device application and smartphone as "entry level handset" for rich user experience.

http://www.youtube.com/watch?v=O3ipH3gDG0o

and c) NaaS and also the Communication Solutions as a Service will be the trend very soon, we tried to evangelize about that

http://blog.solaiemes.com/2009/01/communications-…

Of course, Google will not deny (or wants being seen as denying) external contributions. It's also a benefit of Android vs S60 and LiMo who have higher barriers to contribution (especially LiMo).

But just that you 'submit' a piece of code does not mean that that code will be successfully reviewed into the tree. It then becomes a commercial and technical process of reviewing contributions.

In short, Android isn't open, neither closed. The point behind 'open is the new closed' is that open visibility to source code does not mean influence over the product itself.

Makes sense?

Andreas

Hmm, according to http://source.android.com/project *anyone* can be a contributor. In theory. And I find it totally acceptable that in order for a prospective member to join OHA they have to contribute in advance. But imho nothing you have written justifies that OHA is closed.

What do you think?

Tote

Thanks – I 've updated the post to reference Hanashi.nu. Good to see software innovation coming out of OEM labs.

Tote,

Google accepts contributions (see their contributor agreement on source.android.com) but they do not allow public reviewers or commiters AFAIK. To become an OHA process there's certain commitments you have to make; for example many software vendors (e.g. PacketVideo) have released (part of) their source code under the Android repository. I assume there is a financial transaction that's also involved.

Re: the app store comparison, yes, I 'd love to update that with Ovi Store, etc once we find the time (or otherwise a project to justify the time investment

Cheers,

Andreas

Very nice article, but it's really the slideshow that I feel myself in heaven. That is fantastic!

However, some things to add/ask:

- What did you mean on Google OHA being closed? I keep hearing in the news that 'this and that company has joined OHA', then how can it be closed?

- You could update MAS-part with Ovi Store and other prospective (emphasis here) MASes, too. Having mentioned GetJar and Handango are getting less relevant over time. Otoh, currently they're still relevant.

Thanks