Google open-sources WebRTC

[Posted June 1, 2011 by corbet]

Google has announced (in early May) the release of the WebRTC code which, it hopes, will help to jump-start the creation of a new generation of communication applications. "Until now, real time communications required the use of proprietary signal processing technology that was mostly delivered through plug-ins and client downloads. With WebRTC, we are open sourcing the voice and video engine technologies from our acquisition of GIPS, giving developers access to state of the art signal processing technology, under a royalty free BSD style license. This will allow developers to create voice and video chat applications via simple HTML and JavaScript APIs."

Google open-sources WebRTC

Posted Jun 1, 2011 8:23 UTC (Wed) by ajb (subscriber, #9694) [Link] (15 responses)

And Microsoft just paid $8.5 billion for Skype. Oops.

Counting chickens before they've been conceived

Posted Jun 1, 2011 9:51 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (14 responses)

The only thing that would make that investment irrelevant would be the existence (not announcement of a desire to create) of a far larger open network. If Skype in fact has tens of millions of ordinary users who routinely make telephone calls through it, a "far larger" network just isn't likely.

Basically it's the walled garden thing again. Having the Internet, nobody cared that they missed out on MSN because MSN was small and content poor and the Internet was already huge and had everything you could want. But suppose MSN had included exclusive movie streaming back in the mid-1990s. Suddenly nobody wants the Internet, except as a freebie thrown in when you pay for MSN from Microsoft.

For telephony the "content" is people (or to a small and reducing extent, businesses) you want to call. Skype users are walled off, you can't call them unless you're a Skype user. If 1% of the people you want to call are Skype users, they're the annoying holdouts and you tell them they should switch. But if 50% of people you want to call use Skype, you switch to Skype too so that you can call them.

Counting chickens before they've been conceived

Posted Jun 1, 2011 10:51 UTC (Wed) by epa (subscriber, #39769) [Link] (6 responses)

If you can just tell them to go to a web page to make a video call, then the established base of Skype or any other network becomes largely irrelevant. So we need the microphone, camera, and digital signal processing accessible from Javascript.

Counting chickens before they've been conceived

Posted Jun 1, 2011 11:29 UTC (Wed) by tialaramex (subscriber, #21167) [Link] (5 responses)

"I need to call Mom"
"Here's a web page"
"OK, how do I call Mom?"
"She needs to go to the web page too"
"Um, and how will she know to do that?"
"Call her?"
"That's what I'm trying to do!"

Counting chickens before they've been conceived

Posted Jun 1, 2011 11:57 UTC (Wed) by rvfh (guest, #31018) [Link]

> "Call her?"

Send her an e-mail, a SMS, a facebook message...

Counting chickens before they've been conceived

Posted Jun 1, 2011 22:36 UTC (Wed) by jmalcolm (subscriber, #8876) [Link] (2 responses)

< Call her?

I think that what people forget is the the unifying network is the plain old telephone system. There are far more people on that than on any VOIP network. I can call out to a phone number (DID) from any network (my SIP provider say) and it can route back to wherever callee likes (like Skype perhaps).

As long as the telephone system stays substantially larger than these competing networks, it is still anybody's game.

Counting chickens before they've been conceived

Posted Jun 6, 2011 18:48 UTC (Mon) by eean (subscriber, #50420) [Link] (1 responses)

Which is actually what Skype excels at. For $30/year I have a Skype-in US number and can make unlimited US calls. There are SIP providers that do the same thing, but Skype is #1 for consumers.

Not that I think Microsoft spent its money well. Since it's a platform developer its a conflict-of-interest for Microsoft to own Skype. This immediately devalues Skype since other companies won't want to play ball. And Skype is hardly a monopoly so other companies don't have to.

Counting chickens before they've been conceived

Posted Jun 6, 2011 18:57 UTC (Mon) by dlang (guest, #313) [Link]

skype didn't work with anyone else anyway, so why do you think it makes a difference?

Counting chickens before they've been conceived

Posted Jun 3, 2011 20:09 UTC (Fri) by elanthis (guest, #6227) [Link]

The vast majority of Skype calls I've ever been on were pre-scheduled meetings. Just saying.

Counting chickens before they've been conceived

Posted Jun 1, 2011 12:12 UTC (Wed) by job (guest, #670) [Link] (6 responses)

I think Google Talk qualifies as a both larger and open network for voice communication over IP. With the number of people on the Android platform, all of which are potential users, their market share will get even bigger.

(Skype isn't really that big compared to the other players in that particular market. Even Microsoft's own MSN chat is larger, making their investment in Skype even more hard to understand.)

Counting chickens before they've been conceived

Posted Jun 1, 2011 14:38 UTC (Wed) by erwbgy (subscriber, #4104) [Link] (5 responses)

Skype now makes money. According to their SEC filing, in 2010 they:

had 663 million registered users
billed 2.8 billion minutes
had an average revenue per customer of $97
made an operating profit of $20.6 million on revenues of $859.8 million

Counting chickens before they've been conceived

Posted Jun 1, 2011 22:39 UTC (Wed) by jmalcolm (subscriber, #8876) [Link] (4 responses)

At $20 million per year it would take 400+ years for Microsoft to pay off their investment in Skype.

I am not sure "they are making money" really justifies this purchase.

Counting chickens before they've been conceived

Posted Jun 2, 2011 5:21 UTC (Thu) by ghane (guest, #1805) [Link] (2 responses)

There are two possible options to waiting 400+ years.

1. You expect income to increase sharply, and continue. Your payback period may then drop (next year to 100 years, a couple of years later to 30).

2. You don't care for the $20m anyway, you hope to sell it for $12b next week to Symantec. (Hey, it worked for Skype already :-) )

I am not saying either is likely, just that P/E is not what drives (at least my view of) investment, but (current P)/(future E). The Price is a done deal. But I am not really interested in their last years income, from which I do not get a cent, but in what I expect/hope will be their *next* year's income. I am buying a share in their future, not their past anyway (I hope no unsettled lawsuits turn up, but I can hedge against that).

Slightly off-the-mark, but what use is a new-born baby?

--
Sanjeev, who is still waiting for his $8.5b deal, but has lots of free advice for everyone.

Counting chickens before they've been conceived

Posted Jun 3, 2011 11:32 UTC (Fri) by erwbgy (subscriber, #4104) [Link]

Most people sleep too much and new-born babies are very good at ensuring that this doesn't happen. Everyone should have one :-)

Counting chickens before they've been conceived

Posted Jun 3, 2011 17:02 UTC (Fri) by jjs (guest, #10315) [Link]

Yea, but last year's income can be an indicator of future income (we don't know the future). The last several years can provide pointers to the direction for future income.

"You pilot always into an unknown future, facts are your single clue. Get the FACTS." - Lazarus Long

Counting chickens before they've been conceived

Posted Jun 3, 2011 2:18 UTC (Fri) by csigler (subscriber, #1224) [Link]

> I am not sure "they are making money" really
> justifies this purchase.

Since when has "they make money" (profit) ever justified any Internet company purchase? It seems to me that, over time, the opposite has proven true; "they lose money" is the far better justification.

Doesn't it go something like this?:

1. We should buy Company_X!
2. But they post a large net loss every quarter, and their losses are increasing?!
3. But they have a userbase of Y million customers and this number is increasing at Z%/month (along with other metrics like unique page views/month, percentage of returning purchase customers/month, or whatever statistics can be squeezed out of the customer data to make them look uber-attractive to someone)!!! Just think what will happen when all Y million of those users become _our_ customers!!!
4. ???
5. Profit!

Clemmitt (who knows there's money to be made on this new-fangled Internet but is too stupid to figure out how ;^)

Google open-sources WebRTC

Posted Jun 1, 2011 8:54 UTC (Wed) by merge (subscriber, #65339) [Link] (1 responses)

I'd love to have a javascript jabber client that somehow works anywhere :)
... not sure if that's what the outcome will be

Google open-sources WebRTC

Posted Jun 6, 2011 10:30 UTC (Mon) by amonnet (guest, #54852) [Link]

have you tried jwchat?

Google open-sources WebRTC

Posted Jun 1, 2011 10:31 UTC (Wed) by robert_s (subscriber, #42402) [Link] (1 responses)

Reading the site, it all seems very video-chat specific and not very... repurposable. Just yet another web API for browsers to implement.

Google open-sources WebRTC

Posted Jun 3, 2011 17:06 UTC (Fri) by n8willis (subscriber, #43041) [Link]

Just because it's a "web API" doesn't mean that only the full-blown "browsers" will implement it.

Nate

Google open-sources WebRTC

Posted Jun 2, 2011 20:37 UTC (Thu) by cmccabe (guest, #60281) [Link]

It would be so awesome to have a free web-based videoconferencing system accessible to people on every platform. Preferrably one that hooked into an open source server backend like asterisk. Then setting up videoconferencing for your company would be as easy as "service asterisk start"...

I have been using the Linux skype client for almost a year now. The user interface is ugly and clunky, it's 32-bit, lacks features found in the Windows client, and updated on a schedule of roughly never. I guess I have to give them credit for at least supporting PulseAudio, but it's a pain.