463 West Street

Voice is an Application, but Telephony is a Service

2007-05-30T14:38:00.000-04:00

I realize that by using the word "telephony" in the title, I've immediately categorized myself as a bellhead. A curmudgeon. An anachronism.

But it struck me the other day, when seeing an ad for one of the many cable voice services for only $30/month, or $35/month, or whatever - why are people willing to pay $30/month, or $35/month, or whatever for telephone service?

Because it works. Because it's easy to set up and use. Because it does what you need, and very little that you don't need. Because you don't have to configure it, administer it, upgrade it, manage it, or figure out how to make it work with anything else.

Because all that stuff is taken care of behind the scenes. The service providers provide a service - which includes configuration, management, upgrades, interworking, reliability, quality, and all the other "non-functional" requirements that you need for something to just work.

The "functional requirements" - codec, transmission, addressing, directories, and the like - can be met by an application. But ask any systems engineer - the functional requirements are the easy part.

Martin has "gone cold" on VoIP - because, basically, it's inconsistent and it's a pain. I'd posit, though, that it's not VoIP per se he's gone cold on - it's that he's trying to use an application where he's accustomed to using a service. And where he needs a service. And, therefore, where he's willing to pay for a service.

There is certainly a place for "VoIP applications". Just as there's a place for accounting applications, architectural design applications, and financial planning applications. But just as accounting applications, architectural design applications, and financial planning applications haven't resulted in massive unemployment for accountants, architects, or financial planners, the existence of VoIP applications doesn't mean there's no longer a need - or a willingness to pay - for voice services.

The key is to focus not on the "VoIP" part - but on the "service" part.

Blogger Dates and Times

2006-11-07T17:40:00.000-05:00

A couple of weeks ago, I created a draft post in Blogger on fixed-mobile convergence - really, only a title and a couple of sentence fragments, to remind me what I wanted to write about when I had time.

Then today I saw Andrew Schmitt's post on cableco fixed line service, and cannibalized this draft post to write a reply.

Alas, Blogger uses the original create time and date of the post to time- and date-stamp it, not the publish time and date.

So, while it looks like I wrote the reply to Andrew's post about three weeks before he actually wrote the post itself, that's just Blogger's quirkiness at work...

Rumors of Fixed Line's Death are Somewhat Exaggerated

2006-10-19T11:02:00.000-04:00

This was originally a comment at Andrew Schmitt's excellent Nyquist Capital blog, but it got kind of long and unwieldy.

In a post titled "Someone Tell the Cablecos Fixed Line is Dead", Andrew writes:

"Everyone agrees fixed line is a dying, low margin business. Yet Cablecos like Comcast (CMCSA), Cablevision (CVC), Shaw (SJR), and Time Warner (TWX) are feverishly trying to capture market share in this business. Why?"

I have a few reasons. Actually, about 70 billion reasons.

There are 175M fixed lines in the US. At $35/month (the going rate for voice over cable), that's over $70B of annual revenue. The number of lines is declining, but it's only down 9.1% over the peak (192.5M lines in 2000) - even at the highest reported annual rate of decrease (-3.5% 2003-2004), there would still be over 120M fixed lines ten years from now. For the MSOs to not go after this market, and instead to rely on a fixed-mobile convergence play (which is likely to be several years away) would be leaving a ton of money on the table (worse: leaving a ton of money in the hands of their primary competitors).

(By the way, the second major driver of POTS line decreases - aside from cellular substitution - is substitution of high-speed internet service for second phone lines used for dialup access.)

Sure, the PacketCable EMTA is useless when the MSO transitions to a SIP-based fixed-mobile convergence offer, but it's so cheap that Cablevision (for one) is shipping all their new Optimum Online customers cable modems with EMTAs whether or not they subscribe to Optimum Voice. So for maybe $10 per HSI sub in incremental capex (I don't know the actual cost difference between a cable modem without an EMTA and with an EMTA - $10 is a guess), you capture about $18/month in incremental revenue (over 50% of Cablevision's HSI customers purchase OV, at $35/month - the take rate is lower for other MSOs, but Cablevision shows that 50% is an achievable target) - and you deprive your main competitor of about $50/month in revenue for each customer you capture (since cable voice customers tend to be disproportionately high-paying voice customers).

When a SIP-based FMC solution is available, the MSOs will start shipping their cable modems with built-in 802.11n hubs, and the handsets for their "mobile" service will be dual-mode 11n/cellular. By that time, the cost of the embedded 11n hub will be low enough that they can throw it in just like they throw in the EMTA now; it'll give the customers the home networking capabilities and the platform for the dual-mode handset at home. In the meantime, they'll have captured a thousand dollars or so of revenue away from the ILECs per customer. And there will still be probably 150M-160M fixed lines sitting out there to be plucked...

Just as importantly, the MSOs will have established credibility as voice providers, established a base of voice customers, and built IP-based voice networks which (if they play their cards right and make smart equipment choices) will be largely re-usable in a converged fixed/mobile network.

Having a fixed component to the service - a "landline" - is important not because of the whiz-bang features like single number and integrated address book and all that - it's important because spectrum is a scarce and expensive resource. So when a customer uses their dual-mode handset at home, the traffic avoids the scarce, expensive licensed spectrum and hops on the abundant, cheap home WiFi network with its good in-home coverage and its DOCSIS connection back to the MSO servers.

Internet Telephone

2006-05-04T09:51:00.000-04:00

No, not that kind of internet telephone. The internet version of the party game where everyone sits in a circle, one person whispers a sentence to the person next to him, he whispers it to the next person in the circle, and so on around the circle until it comes back, radically different, to the original person.

Let's start at the end. Techdirt has an article about proposed US data retention laws, which also includes a statement that

the FCC has (surprise, surprise) authorized new taxes on ISPs to pay for the mandatory wiretap access on broadband and VoIP services.

Wow. The FCC has authorized new taxes on ISPs. Let's follow the link and see exactly what CNET news.com said.

The Federal Communications Commission voted unanimously to levy what likely will amount to wiretapping taxes on companies, municipalities and universities, saying it would create an incentive for them to keep costs down and that it was necessary to fight the war on terror.

Hmm. "... to levy what likely will amount to wiretapping taxes on companies." Perhaps we should see exactly what the FCC said.

The R&O isn't available yet, but the release (PDF) says:

...the Order concludes that carriers are responsible for CALEA development and implementation costs for post-January 1, 1995 equipment and facilities, and declines to adopt a national surcharge to recover CALEA costs.

Wait a minute. The FCC said it's not going to authorize, levy, or impose new taxes to recover the cost of "facilities-based broadband Internet access providers and interconnected voice over Internet protocol (VOIP) providers" implementing CALEA. In only two hops and about 12 hours, that changed 180 degrees into "the FCC has authorized new taxes on ISPs."

If you want to be outraged about the civil liberty implications of requiring ISPs and VoIP providers to support CALEA, fine. If you want to castigate the FCC for appearing to extend CALEA obligations that were always intended for public networks (the term in the statute is "telecommunicatons carrier", which includes in its definition "common carrier for hire") to private network providers such as universities, go right ahead. If you want to bemoan the FCC's apparent intent to define ISPs as "information services" when it suits one policy goal but as "telecommunications providers" when it suits another, be my guest. You can even rail against the unfairness of the FCC providing $500M for ILECs to implement CALEA but providing nothing for the ISP or VoIP industry if you'd like.

But if you don't even get the facts right, your arguments lose a lot of their credibility.

Everything Old is New Again

2006-03-10T23:22:00.000-05:00

Mr. Blog writes on "Net Neutrality, 1996", with his thesis being that 10 years ago, the Internet took off due to the ubiquity of dialup access; the telco raked in significant money by providing dumb transport (second phone lines) with no knowledge of or control over the content (V.34, anyone?). "Neutrality" at its finest.

I'd extend this thesis to say that the Internet took off due to the ubiquity of flat-rate dialing for dialup internet access. In the rest of the world, where metered usage was far more prevalent than in the US, the widespread use of the Internet grew a lot more slowly.

Now if anyone thinks back to those bucolic days of yesteryear, they may remember constant rumors called the "modem tax". The rumor was that telcos would monitor the phone conversations for modem traffic, and charge modem calls differently than voice calls -- charging per-minute or capping the minutes that could be used on the flat rate line.

Substitute "perform deep packet inspection" for "monitor the phone conversations for modem traffic", "guaranteed QOS" for "modem calls", "charging based on QOS" for "charging per-minute", "capping the downloads" for "capping the minutes", and "broadband access" for "flat rate line", and that paragraph could have been written about the current situation. Except then, the telcos universally denied any intention to take such action. Now, they're advocating it.

Ripoff or Reasonable?

2006-03-10T16:30:00.000-05:00

Tom Evslin is irate that AT&T is charging soldiers 21 cents a minute for calling card calls from Iraq to the US, and doesn't permit access to other phone card providers' access numbers. Jack Decker is even irater.

According to the article in the Prepaid Press that Tom cites, AT&T charges 21 cents per minute for calls from Iraq. This price, the meme goes, is outrageous.

You can buy prepaid cards almost anywhere in the world to call the US for less than two cents a minute.

Also outrageous, the meme goes, is that the soldiers are denied the chance to use a less-expensive calling card.

But when a company appears to be abusing their monopoly position to pick the pockets of people who have no other choices, I get upset.

Disclaimer: I, like Tom, used to work for AT&T. I wasn't anywhere near Carpetland. I did go there for a meeting. Once. They didn't invite me back. But one boss of mine was in the same part of the building as some of the Labs folks who were responsible for establishing the calling centers in Iraq and Afghanistan, and they were rather proud of what they had done - there aren't many places in the Labs where people hang pictures of their projects on their doors, but that was one of them. So I got curious. Is AT&T taking unfair advantage of their position in Iraq to gouge the soldiers?

I went looking for the international prepaid calling cards that allow you to call the US from Iraq for less than two cents a minute. Perhaps AT&T's biggest LD rival, MCI (now Verizon Business), would have a better rate. Doing some searching, I found three MCI international prepaid cards that show rates from Iraq to the US (Not sure all the deeplinks will work):

Branded as MCI ConnectHome: 45 cents per minute
Branded as AbsoluteGlobal, sold through ZapTel: 45 cents per minute
Branded as MCI World Traveler Phone Card: 56.3 cents per minute

Hmm. Maybe the other of the Big Three, Sprint: Nope. Not available.

OK, obviously the Big Three are all thinking the same way. But the Prepaid industry is ultra-competitive, and there are a lot of small companies that have extremely low prepaid card rates. So let's look on Google for "international calling card" and see what we can find:

WorldTraveler Card (PhoneCardSavers): 56.25 cents per minute - good to see that it's the same rate that MCI /VZB uses. Although Calling Card Plus has the same card with a rate of 45 cents per minute.
iDPhoneCard.com: Not available
PhoneShark.com: No cards available for service from Iraq.
OneSuite.com: Not available

There were a number of sites that had rates for PC to phone or callback cards, but I was unable to find any calling cards with rates from Iraq to the US that were less than 45 cents per minute - more than twice what AT&T is charging.

Other than anecdotal evidence, is anyone able to document phone-to-phone calling cards with a rate for calls from Iraq to the Continental US less than what AT&T is charging?

Would it be preferable for AT&T to allow callers to use other companies' calling cards? Sure. Would it help the soldiers? Doesn't look like it.

WiMath

2006-03-09T14:35:00.000-05:00

Intel demos a WiMAX-enabled laptop on a scooter that is delivering speeds of 2 Mb/s.

"Support of data rates up to 75 Mb/s" is a widely-bandied-about figure.

I've even seen some people claim 150 Mb/s.

Alas, Shannon-Hartley's not just a good idea, it's the law.

WiMAX supports a number of modulation types and coding rates; the best is 64QAM 3/4, which provides a raw data rate of 4.5 bits per baud, or 4.5 bps/Hz. Let's be generous and assume that a hypothetical WiMAX network has such a dense forest of base stations that everywhere Sean Maloney rides his scooter in San Francisco, his WiMAX card is operating at 64QAM 3/4.

The 2.3 GHz WCS band licenses in the US are for 5 MHz channels; 2.5 GHz BRS/EBS licenses are for 6 MHz channels. So with that 6 MHz channel, the maximum theoretical data rate for WiMAX is 27 Mb/s. Hang Sean's scooter from a crane right in front of the base station antenna, and he won't get any better than 27 Mb/s raw channel capacity.

That's raw bits, though. You have to then factor in the WiMAX PHY overhead (about 15%), MAC overhead (about 20%), and MAC convergence sublayer overhead (about 10%). Now you're down to about 16.2 Mb/s user payload throughput. Remember, we're still talking best case - 6 MHz BRS/EBS channels, and enough base stations that you've got 100% coverage with 64QAM 3/4.

Well, OK, 16 Mb/s is still pretty fast, right? But remember, this is a shared medium, and that 16 Mb/s is combined upstream/downstream. How many WiMAX cards does Intel hope to sell? Because even with only a 256 kb/s upstream, that's only about 7 simultaneous users at 2 Mb/s.

150 Mb/s? Not hardly.

75 Mb/s? Not even close.

2 Mb/s? Sure, if a carrier builds a dense forest of base stations, because you need small cells both to get as much 64QAM coverage as possible and you need a lot of them to provide the capacity needed.

AT&T-BLS Merger

2006-03-06T09:27:00.000-05:00

Over/Under on the number of Star Wars Episode 6: Return of the Jedi/"rebuilt Deathstar" references among the blogniscenti: 12.

Verizon VCast Music Service to Launch in January

2005-11-29T09:08:00.000-05:00

While getting a new phone at the local Verizon Wireless store yesterday, I saw on the back of the counter a piece of paper with sales and customer service information for a Verizon VCast Music Service, showing a launch date of January, 2006. Details are sketchy in my mind (I'm willing to look at something sitting in plain sight; snatching it up would have been a bit tacky), and there was no information on pricing or content sources, but I do recall that four phones would be supported - two LGs, "The V" and the VX8100, and a Samsung, all of which would require an "upgrade package", and one yet-to-be-released phone the maker of which I can't recall.

Google Betas and Stock Price

2005-11-19T00:31:00.000-05:00

Mark Evans compares Google's stock price with its launch of new Betas and official products, and Om Malik notes it.

Now, I suspect that Mark was half kidding, but his "analysis" also "inspired" me to look a little deeper.

Of the ten days that Mark notes new product or Beta introductions, Google went up 6 and down 4. On the six up days it gained $17.57; on the four down days, it lost $15.24.

So in other words, on the 10 days Google introduced new products or betas, it had a net gain of $2.33, or $0.23/day. On the 307 days Google didn't introduce new products or betas, it had a net gain of $300.21, or $0.97/day.

Which just goes to show that given five minutes with a spreadsheet and daily stock prices, you can prove just about anything you want.

Towerstream

2005-11-08T17:00:00.000-05:00

This hardly merits a post, but Andy Abramson doesn't have comments turned on in VoIPWatch.

He writes, "I keep hearing that Comcast has an interest in acquiring (Towerstream) to compliment their existing wireless spectrum portfolio."

Towerstream uses unlicensed spectrum for their standard offering; they use licensed spectrum strictly for point-to-point backhaul and 100Mb/s+ enterprise service, and that's 18 GHz and 23 GHz licenses.

So if Comcast bought Towerstream, it could be for their networks, could be for their enterprise customers, but is unlikely to be for their spectrum.

How Mainstream Can Blogs Get, Anyway?

2005-11-07T23:33:00.000-05:00

Nick Carr writes that "...the blogosphere is going to end up looking a lot like the old 'mainstream media'... A relatively small number of high-traffic blogs will dominate the market, and then there'll be a whole lot of more specialized blogs with fewer readers."

I agree that a relatively small number of blogs could dominate blog traffic, and there could be a whole lot of specialized blogs with fewer readers. But I think the orders of magnitude of "relatively small", "whole lot", and "fewer" are radically different than they are for, say, magazines, because of the radically different cost of entry for a blog as opposed to a magazine.

The Magazine Publishers of America represents more than 240 publishers with approximately 1400 titles. Five titles account for 20% of the magazines sold. I'm guessing (hard to find statistics) that (a) a lot more than 5 blogs are needed to account for 20% of the total blog page hits; (b) there are a lot more than 1400 blogs published by bloggers in the US; and (c) the minimum readership for a blog to continue to be published is a whole lot lower - i.e., 0 - than the minimum readership for a magazine to continue to be published. (Like this one, for example...)

I believe in The Long Tail.

UTOPIA Take Rate

2005-09-21T22:54:00.000-04:00

According to Light Reading, the UTOPIA project has signed up only 500 customers out of the 5,000 to 8,000 currently reached by the network. Light Reading seems to think that's a bad thing.

According to the UTOPIA project's website, they began construction in 2004. And according to my math, 500 out of 5,000 to 8,000 is somewhere between 6.25% and 10%.

So for an access network that's less than two years after start of construction, they've captured between 6% and 10% of what the cablecos call "homes passed".

In terms of comparison, to pick a cableco at random, Comcast was selling Comcast @Home in 1999 (and maybe earlier - that's just the earliest Comcast press release I can find that refers to internet service), and as of 2Q05 - six or more years later - was at 18.9% of homes passed.

So 6%-10% penetration really doesn't seem that bad.

Their feasibility study indicates that they've got three years to hit 10% penetration for breakeven, and only need to get up to 25%-30% in years 4+ to stay at breakeven. Personally, I wouldn't be that negative on 6%-10% in year two.

Sprint WiMax, Bro

2005-09-15T11:51:00.000-04:00

Broadband Reports picks up on a Yonhap News blurb that Samsung will provide WiBro equipment to Sprint-Nextel for a trial. (I haven't seen an announcement from either Samsung or Sprint-Nextel.) Update: Sprint-Nextel has an announcement out.

Sprint-Nextel has previously trialed Flarion (now Qualcomm) FLASH-OFDM equipment and IPWireless TD-CDMA equipment, and has announced trials of Motorola WiMAX equipment.

There would not seem to be a great match between the 2.3 GHz spectrum that WiBro typically operates in and the 2.5 GHz BRS/EBS spectrum that Sprint-Nextel has so much of, but the same could probably be said for FLASH-OFDM and TD-CDMA.

If I had to guess, I'd guess that Sprint-Nextel will use the trial to look more at the mobility aspects of broadband wireless technologies than the WiBro radio technology per se.

Internet Bubble 2.0

2005-09-08T09:31:00.000-04:00

What's the difference between Internet Bubble 2.0 and Internet Bubble 1.0?

In Internet Bubble 2.0, eBay is rumored to be in talks to buy Skype, a company which some have estimated to have revenues in the neighborhood of $70M, for between $2B and $3B.

In Internet Bubble 1.0, Skype would have bought SBC.

I Could Do Better

2005-08-27T01:37:00.000-04:00

Business Week Online has an article from a Standard & Poor analyst which does a good job of perpetuating several popular WiMAX myths, confusing various technologies, and generally confusing the hell out of anyone who's trying to understand broadband wireless access.

Let's start with the simple fact-checking that you would think a professional analyst writing in a respected publication would have been able to get right: WiMAX compatibility testing is being performed by CETECOM, el Centro de Technologia de las Comunicaciones, S.A., not "Cenecom, the WiMax Forum Certification Laboratory, which opened in early August". CETECOM has been around since 1991; it's the WiMAX testing that started in August.

Then there's the ever-popular, "has a service range of up to 50 kilometers and provides data rates of up to 280 megabits per second per base station." This is even more outlandish than the usual hype, which limits its irrational exuberance to 70 Mb/s and 50 kilometers.

That "up to" is a great two words. My car can go up to 200 miles per hour and get up to 100 MPG. If I pushed it off a high enough cliff; otherwise, its performance is somewhat less.

I suppose with 80 MHz of spectrum and subscribers within half a mile of the base station (with line of sight to the base station antenna) that would enable the base station to run all the subscribers on 64QAM you could get 280 Mb/s out of a single base station. And I suppose with a really tall tower, really flat terrain, and remote units with their antennas sitting in a window facing the base station antenna, you could push a signal 50 km. But in anything approaching the real world, you're not going to get either 280 Mb/s or 50 km.

Realistically, with a 5/5.5 MHz channel (US WCS or BRS spectrum), you'll get around 13 Mb/s user payload, and in something approaching normal suburban/rural terrain, you'll be able to push a reasonable signal out 3 to 5 miles. Useful? Sure, especially for DSL-equivalent services where you can run oversubscription of 25:1 or higher. But let's not get carried away.

And by the way, that WCS or BRS spectrum is licensed and usable for WiMAX (though the WiMAX Forum profiles for 2.3/2.5 GHz haven't been completed yet). So saying that "there's no authorized radio frequency spectrum from the Federal Communications Commission for the issuance of WiMax licenses" is, well, wrong.

The article also says, "AirBand, Clearwire, and TowerStream... believe bandwidth is adequate within 50 megahertz to 500 megahertz bands for Wi-Fi and WiMax without seeking radio frequency licenses from the FCC." Actually, AirBand and Towerstream use frequencies in the 5 GHz unlicensed bands (that's 5000 MHz); Clearwire is the second-largest owner of BRS frequency licenses and has leased additional licenses (some from Sprint/Nextel).

And I'm not even going to get into the butchering of OFDM, OFDMA, and FLASH-OFDM.

I do not pretend to be a professional analyst; I write about things that interest me. I have some knowledge from my day job, and I have half a dozen tabs open in Firefox as I write this to track down the rest. If I can get this stuff right, you'd think a professional analyst who gets paid to follow this stuff could.

Maybe I'm in the wrong line of work.

CALEA and VoIP

2005-08-06T23:19:00.000-04:00

I'm probably going to be unpopular for saying this (the blogger community tends to be rather anti-establishment, I find), but: I feel the FCC did the right thing in extending CALEA obligations to "interconnected VoIP service providers."

CALEA came about largely because of, believe it or not, ISDN. Unlike a traditional phone line, you can't wiretap an ISDN line. At one time, it was actually believed that ISDN would be widely adopted in the US (please, hold your laughter), and this would have a detrimental effect on the ability of law enforcement to administer authorized wiretaps.

Fast forward a few years, and there's a new problem for wiretaps; namely, cellular phones. Can't tap one of them, either. In my opinion, cellphones became the major impetus for CALEA continuing to move forward once it was apparent that ISDN was pretty much a non-starter in the US.

The intent (I feel like a Constitutional scholar, talking of "original intent") was to provide capabilities to law enforcement that would enable them to execute lawfully authorized wiretaps (and pen register trap and traces, but let's just stick with the term "wiretap" for now -- it's so semantically loaded) substantially equivalent to what they could execute on a traditional phone line, on "non-traditional" telephones or the equivalent. No more, no less.

So, along comes VoIP. Can't tap VoIP, either. Just as I dislike asymmetrical regulation, I dislike asymmetrical statutes - and requiring landline and wireless carriers to implement CALEA capabilities while not requiring VoIP service providers to implement equivalent capabilities is asymmetrical. Does it serve the public interest if a law enforcement agency gets a lawfully-authorized intercept order against a person, but is only able to execute that intercept if the person's phone is landline or wireless, not VoIP? I don't see it.

That said, I do have some problems with the way the FCC is going about it. For one thing, the FCC established a deadline of 18 months from issuance of the order. The original CALEA statute (passed in October, 1994) had a deadline of four years; the FCC eventually established a deadline of June, 2000 (five years and 8 months), and granted extensions to June, 2002 (over seven and a half years) to any carrier that asked. Seven and a half years for the industry to figure out how to make three-way calls and pen register traps transparent to the calling parties and easier to manage, versus one and a half years to establish entirely new technology for intercepting packetized voice transported over a broadband connection that may be owned and operated by a third party, strikes me as a bit (here we go again) asymmetrical.

It will be interesting to see how this plays out. The telephone companies enlisted the support of the telephone equipment manufacturers, defined a standard, played for time, and got the government to pay a significant amount of money to get compliance ($450M paid or owed as of 9/30/03). I don't know if the VoIP service provider community is as adept at the regulatory dance.

Clearwire Growth

2005-08-06T13:52:00.000-04:00

Steve Stroh's BWIA/WiMAX site points to an article on Clearwire's ramping up for growth. The article uses, among other things, a perusal of internet job posting sites to project locations where Clearwire may begin offering service.

One thing that I've not seen anyone pick up on, other than discussing rumors, is that Clearwire has entered into a lease agreement with Sprint to lease 54 BRS licenses in 21 BTAs. It requires some digging through the FCC ULS database to get all the details -- and the work done to uncover them was sufficiently part of my day job that I'm not inclined to list them all here -- except to say that one of them is in fact call sign B050, covering BTA 050, Boise-Nampa ID (one of the cities mentioned in the Puget Sound Business Journal article), and if you look at the Admin data associated with that license it'll clue you in how to find the rest.

It seems to me that the FCC's recent actions are the best thing that could happen for broadband wireless access. Eliminating the requirement that cable companies and telcos make their local access transmission facilities open to other service providers virtually forces aspiring competitors to a wireless access approach (unless you think broadband over power line is viable, which I'm not too sure of).

FCC Approves Sprint/Nextel Merger

2005-08-03T17:17:00.000-04:00

While officially the FCC only approved the transfer of control of spectrum licenses and leases from Nextel to Sprint, they're basically approving the merger, with very few conditions.

One condition, which will be of interest to the WiMAX watchers out there, is that Sprint committed to "offer service using BRS/EBS spectrum to at least 15 million Americans" (hmm, do they have to check passports of everyone in range of their towers?) within 4 years and to another 15 million within 6 years.

Of course, they can get out of this commitment if "circumstances beyond (their) control" prevent them from meeting it. Sort of like the circumstances beyond the control of SBC that prevented it from meeting a commitment it made to the FCC when it purchased Ameritech, to offer competitive local service in 25 markets outside its regions.

It is interesting to see the difference between the public safety obligations being placed on carriers in a mature industry such as cellular telephony and those being placed on service providers in an immature industry such as VoIP. As noted in Commissioner Copps's statement, Nextel will not meet the FCC requirements on cellular E911 until the end of 2007, two years after the FCC deadline. Not only is the FCC not requiring them to cut off service to their customers (unlike the guidance from the Enforcement Bureau given to VoIP service providers), it is not even imposing merger conditions related to E911 compliance.

Some would call this a double standard.

VoIP E911 Notification

2005-07-26T23:57:00.000-04:00

Jeff Pulver points to an FCC public notice regarding the notification requirement for interconnected VoIP providers to tell their subscribers that 911 might not work on their VoIP service.

I have to say I find disturbing the requirement that service providers disconnect all customers who have not affirmatively acknowledged the notification. As Jeff points out (I haven't independently checked his numbers, but see no reason to dispute them), 1.5M landlines in the US do not have E911, and significant numbers of wireless phones do not have E911, but there has been no requirement on ILECs or Cellular carriers to get affirmative acknowledgdment from these customers, and especially none to shut off service to customers who don't affirmatively acknowledge.

Is the public interest better served by customers not having VoIP service at all than by customers having VoIP service without E911? How does this square with the logic that the public interest is better served by ILEC and Cellular customers having service without E911?

(Let's ignore for the moment that the ILECs that do not provide E911 are typically rural independent telcos, which have an inordinate amount of lobbying clout.)

If you're an ILEC customer without E911, it's OK for the ILEC to give you a phone line without E911, because they're often the provider of last resort - if they don't give you a phone line, you don't have a phone line. But if you're a VoIP subscriber, I guess the logic is that you can always get phone service from the ILEC if the VoIP service provider cuts you off.

But isn't that a regulatory impediment to competition? If the FCC considers intermodal competition (e.g., cable companies competing with telcos for telephone service) a better method of ensuring consumer protection than, for example, regulations like local service unbundling, shouldn't they at the very least be avoiding asymmetrical regulation (such as requiring VoIP service providers to notify - and collect affirmative acknowledgement - of E911 limitations, but not requiring the same of telcos)?

This only applies out in the boondocks, where independent telcos don't provide E911, you say? Well, let's look at some of Packet8's caveats about when E911 may not work as expected, for example:

When there is an electrical power outage, service outage or suspension/disconnection of Packet8 service due to billing or other issues.
When a change of address has been reported, but not yet been updated on the Packet8 account.
When the local PSAP receiving Packet8 E911 emergency service calls does not have a system configured for E911 services that enables the operator to capture and/or retain automatic number or location information.
When due to technical factors in network design and/or in the event of network congestion on the Packet8 network, a Packet8 E911 call may produce a busy signal or experience unexpected answering wait times and/or take longer to answer.

Those could apply to my landline just as easily. If the ILEC has a service outage, my E911 service won't work. If I move and my ILEC drops the ball getting my new address into ALI, my E911 service won't deliver the correct address. If the PSAP can't receive or capture ANI/ALI, they won't get it. If there's a major emergency that results in mass calling, my E911 calls may not get through right away. But my ILEC isn't required to notify me of these limitations (I wonder how many people realize that telco switches have line concentration, and particularly in newer developments served by digital loop carriers, if everyone in their neighborhood picks up the phone and tries to call 911 at the same time, many of them won't even get a dial tone, let alone get through to the PSAP), and clearly aren't required to cut off my phone service if I don't affirmatively acknowledge these limitations.

Again, I fail to see how, in a deregulatory model that is supposed to encourage competition over regulation, imposing asymmetrical regulation like this is in the public interest.

I'm also trying to figure out exactly how you put a warning sticker on a softphone, but that's another topic...

Sprint/Nextel, WiMax, and Moto

2005-07-26T23:33:00.000-04:00

It says here that Sprint and Nextel, after the merger, will transition their iDEN customers to CDMA. That means a lot of smart Motorola wireless techies will need something else to do.

It says here that Sprint expects to someday build a network using its 2.5 GHz spectrum, which is reported to cover 85% of the US population and is covered by one of the first WiMax profiles to be available.

It says here that Motorola has "expanded its strategic focus to bring comprehensive WiMax (802.16e) solutions quickly to market," including "increased R&D (and) resources."

Seems to me that Motorola has decided to take the lemons from the Sprint/Nextel merger and make itself some lemonade. And it seems that they're the team to beat to grab Sprint's mobile broadband wireless business, whenever Sprint does decide to build something. Consider:

Sprint is going to be looking for a mobile broadband wireless technology, not a DSL replacement. When Moto talks WiMax in the US, they mean 802.16e mobile.
Both Sprint and Nextel have bought Motorola, and existing purchaser relationships can mean a lot.
Motorola can show Sprint a solid corporate commitment to WiMax; Sprint's other two major vendors (Lucent and Nortel), not so much. Lucent is reselling Alvarion's WiMax line; Nortel is partnering with LG Electronics.

You've got to like the idea of a network supporting tri-mode devices that'll switch between WiFi (spotty coverage, high bandwidth), WiMax (wide coverage, moderate bandwidth), and EV-DO (ubiquity, lower bandwidth), depending on what's available.

NASA's Idea of Multimedia

2005-07-26T15:12:00.000-04:00

I watched the Discovery launch today, with my heart in my throat until the solid rocket motors were jettisoned. (I'm in the generation for whom "where were you when the Challenger exploded" is a touchpoint question, much as the Kennedy assassination was for my parents' generation, and 9/11 will be for my kids'.) The difference from Challenger being that instead of watching it in my basement apartment at college on my 13" black and white TV, I watched the streaming video from NASA on my laptop.

Cool, I can watch the Shuttle launch here at the office without having to find a TV. One thing irks me, though, and that's the use of the term "multimedia" to describe streaming video to a computer. (Sorry to single out NASA; they've got better things to do than drive meaningful usage of technology-related terms in the English language, but it was their usage that made me think of it.)

No. Throwing the NASA TV feed out a webcast isn't multimedia. Show me thumbnails of four camera feeds and let me select which one I want to watch (so I can cut away from the shot of the Governer and his sister-in-law to, gee, maybe the launch pad). Superimpose the countdown clock over the video feed so you don't have to keep cutting back to the camera shot from the viewing stage where the clock is visible. Show a map with the transatlantic abort sites. After the launch superimpose the vehicle telemetry with position, altitude, and speed. Even better, have a live map showing vehicle position with altitude and speed superimposed. That's multimedia.

Microsoft, GIPS, Skype, and Money

2005-07-14T22:53:00.000-04:00

So Microsoft has licensed the Global IP Sound VoiceEngine product, which provides the broadband codecs that Skype uses. (Several other bloggers have already noted this; I picked it up from Andy and James.)

Previously, when I conjectured on how much Skype was paying for its GIPS license, a commenter claimed that GIPS doesn't always charge on a per-port basis; rather, they "use the business model of each of their customers," and that since Skype doesn't charge for downloads, the license fee will be a percentage of revenues.

Hmm. MSN Messenger is a free download, just like Skype. Think GIPS is getting a percentage of Microsoft's revenues?

Facetiousness aside, this item and some recent commentary about the tension between Skype the platform and Skype the company developing applications got me thinking on several levels.

First, if in fact GIPS is getting licensing fees from Skype based on Skype's revenue, I would think they could be getting nervous about an explosion of third-party applications that contribute nothing to Skype's top line. Particularly if those applications are ones that don't drive Skype interworking with the PSTN, which is the only place Skype makes money currently. If I'm using Skype as a platform for video communications using one of the third-party video apps, there's nothing particularly driving me to buy SkypeIn, SkypeOut, or Skype VM.

By the same token, Skype itself has got a fundamental conflict, which makes it very different from, say, Microsoft. It makes no money off its basic platform. The more ability it gives developers to invent creative new applications, the less likely it is to be able to sell similar applications itself. Mindshare is great, mass adoption is great, but it doesn't pay the bills.

Companies in a situation like this have several options:

Limit the ability for third parties to develop applications that compete with in-house applications. Keep some capabilities out of the API, for example. Doesn't exactly endear you to third-party developers.
Charge developers in some way for access to the API. Also doesn't endear you to third-party developers, and inherently reduces the number of people who will want to be third-party developers.
Charge for the platform. Probably a rather unpopular move, and dangerous if user barriers to switching are as low as people think.
Charge for parts of the platform. Segment the product into two, for example: a "Skype Lite" and a "Skype Pro", with more capabilities in "Skype Pro" -- but only "Skype Lite" available for free.
Build applications that are so much better than what the third parties will build that customers will buy yours instead of someone else's. Hard, given the ability of third parties to focus more narrowly.

Personally, I'm betting on Number 4.

How Free is Free?

2005-06-15T22:54:00.000-04:00

The great thing about selling software is that the incremental cost is very small - typically the cost of production and packaging.

If the software is delivered via download, even that cost goes away, and the incremental cost per unit shipped can be arbitrarily close to zero.

The exception is if the software includes any licensed technologies - then there can be a per-unit cost for licensing.

"So what's this got to do with telecom," you ask? Well, Skype has claimed that its marginal cost per user is less than US$0.01. But Skype licenses the Global IP Sound VoiceEngine product, and it's not free: Global IP Sound's 3Q04 results include in their highlights "First royalties from Skype Technologies S.A."

Also in their 3Q04 results, Global IP Sound notes that "Deployed software reached more than 26 million endpoints (measured by downloads and shipments)." Doing a pencil-and-paper interpolation from a chart in a Skype Journal post, it looks to me like Skype had about 21M downloads as of the end of 3Q04. Since this is the first quarter in which Global IP Sound realized revenue from Skype, let's assume every Skype download up to the end of 3Q04 generated revenue for GIPS in 3Q04.

Let's also assume that the Skype marginal cost per user is as much as US$0.01, and 100% of that marginal cost is the license for GIPS VoiceEngine. The royalties from Skype to GIPS would therefore have been around US$210k, or about KSEK 1,610 at current exchange rates (I don't know what the exchange rates were in 3Q04, but that's a level of precision way beyond this analysis anyway).

Now, Global IP Sound's total revenues reported for 3Q04 were KSEK 7,639. Which means that since Skype didn't contribute more than KSEK 1,610, other GIPS customers contributed about KSEK 6,000.

But if total deployed software reached 26M endpoints, and 21M of those endpoints were Skype, then the other 5M shipments generated KSEK 6000 for GIPS, or an average of SEK 1.20 per endpoint - over 15 times the upper limit on the revenue per endpoint that GIPS could be getting from Skype.

So either:

Skype has a tremendously favorable licensing contract with GIPS (possibly getting a very good price in exchange for the publicity Global IPSound is getting out of being associated with Skype, and the potential business they can get licensing their products to other customers who want to make products that interwork with Skype) - in which case other GIPS customers could be getting pretty upset; or,
Skype's marginal cost per customer isn't really less than $0.01.

Perhaps that's another reason why Skype is moving into retail - not only is the viral marketing growth potentially slowing, but giving away a non-free product can cut into one's margins...

Off-Topic: Pictures and Stories

2005-06-14T10:31:00.000-04:00

Why, in an article headlined "US Regional Jets Under Scrutiny", over a caption of "During the first half of 2005, 33 percent of all domestic U.S. flights were on regional jets", does CNN run a picture of a Saab 340 turboprop? Couldn't they find any pictures of a CRJ? (I found 461 just on airliners.net; I'm sure with all of CNN's resources they could have found one picture they could license...)

Skype Codecs

2005-06-14T00:04:00.000-04:00

Aswath believes, based on some rather cloudy statements in a CommsDesign interview, that Skype has built a proprietary codec that gives them better sound quality than POTS or other VoIP.

Personally, I'm skeptical. For one thing, codec design isn't easy. For another thing, there are a bunch of perfectly good wideband codecs out there (GIPS iSAC, which Skype is known to use; G.722, which goes back to the eighties; G.722.1 and G.722.2, more modern standard wideband codecs (G.722.2 is also known as GSM AMR-WB), though there may be licensing issues), so why reinvent the wheel?

Furthermore, the Global IP Sound VoiceEngine package comes with iSAC (known to be used by Skype), iLBC (ditto), G.711 (mentioned in the CommsDesign article), and with G.729 as an optional component (known to be used for SkypeIn and SkypeOut). If you're licensing a package with that complete set of codecs, why take the time and energy to build a brand-new wideband codec?

Then again, maybe there are a bunch of really bright Estonian speech processing gurus working for Skype. (Speech processing gurus who are Estonian, not gurus at processing Estonian speech...)

Maybe it's the Language Barrier

2005-06-13T23:47:00.000-04:00

Aswath takes a shot at some claims in the CommsDesign interview of Niklas Zennstrom. Reading the interview, one has to wonder: is Mr. Zennstrom intentionally being misleading about how Skype compares to other VoIP architectures, or is he just uninformed?

In answer to a question about how Skype claims better quality of service than other VoIP providers and regular phone lines, Mr. Zennstrom replies that Skype's peer-to-peer technology means that "the connections between end users are set up directly over the Internet to end users, rather than having to go through a central server somewhere."

But in every VoIP architecture, the (media) connections between end users are set up directly over an IP network (the Internet or a private IP network). The only case where the media connections do not go directly between the end users is when a Session Border Controller is in the media path - but this an exception in the same way that Skype uses supernodes for NAT/firewall traversal, in which case the Skype media connection does not go directly between the end users either.

The call/session control information will "go through a central server" in an H.323 or MGCP network (though it may not in a SIP network); however, how call/session control information is handled has little to do with sound quality.

So is Mr. Zennstrom intentionally mixing call/session control information with media connections, or does he not understand the difference?

Clearwire Spectrum Acquisition

2005-06-10T13:35:00.000-04:00

Several blogs have made note of a purchase by Clearwire of 2.5 GHz BRS spectrum from CT Communications, nee Wavetel, and questions have been raised about what areas the licenses cover.

You could search the FCC Universal Licensing System, but then, why would you be reading this when you could be having fun searching ULS?

So here goes. The licenses cover 6 BTAs (Basic Trading Areas):

BTA062, Burlington NC
BTA141, Fayetteville-Lumberton NC
BTA229, Kingsport-Johnson City TN
BTA295, Middlesboro-Harlan KY
BTA368, Raleigh-Durham NC
BTA382, Rocky Mount-Wilson NC

There are also nine specific licenses within those BTAs, and one license in BTA074, Charlotte-Gastonia NC.

Note that the NC BTAs are all in the Raleigh-Durham area. Clearwire did not purchase CTC's licenses for six other BTAs in NC.

Bring Out Your Lawyers

2005-06-07T23:11:00.000-04:00

One of the most interesting (where by "interesting" I mean "capable of generating megabytes of informed, partially-informed, and wholly-uninformed debate") aspects of the FCC VoIP E911 Order is the determination of exactly to whom it applies.

For the one or two of you out there who might not have read it yet, the Order identifies something called "interconnected VoIP services," and defines it as having the following characteristics:

(1) the service enables real-time, two-way voice communications;
(2) the service requires a broadband connection from the user’s location;
(3) the service requires IP-compatible CPE; and
(4) the service offering permits users generally to receive calls that originate on the PSTN and to terminate calls to the PSTN.

VoIP provided by a cable MSO? Check.
Bring-Your-Own-Access residential VoIP such as Vonage or AT&T CallVantage? Check.
VoIP as part of Xbox Live? Nope.
VoIP as part of a chat application with no PSTN dial-in or dial-out capability? Nope.
VoIP tie-lines between PBXs? Nope. (No inherent dial-in/dial-out.)
Skype? Hmmm.

Well, it enables real-time, two-way voice communications. It requires IP-compatible CPE (since footnote 77 explicitly states that "IP-compatible CPE includes... a personal computer with a microphone and speakers, and software to perform the conversion (softphone)."

The other two traits are more interesting.

The Skype application makes use of adaptive, low-bit-rate codecs, and Skype could certainly claim that a broadband connection is not inherently required.

And Skype, in and of itself, does not "permit users generally to receive calls that originate on the PSTN and to terminate calls to the PSTN." You need to buy SkypeIn and buy SkypeOut minutes for that.

It'll be interesting to see Skype (or, more precisely, Skype's lawyers) respond to the Order. As has been noted several places, they've already said it doesn't apply to them at all - though I haven't seen them play the "no broadband connection required" card yet. I don't know if that argument will fly; keep your eyes open for the next Ex Parte Skype has with the FCC to see what arguments they make. They may have to fall back to an argument that Skype per se isn't an interconnected VoIP service, but Skype plus In plus Out is. (Which Richard points out the FCC pretty much concludes in Para 58.)

Next up: Jurisdiction.

VoIP 911 - Location Updates

2005-06-06T14:52:00.000-04:00

This is the first on what I hope will be a number of posts deconstructing the FCC VoIP E911 order.

The Order requires that Interconnected VoIP Service Providers must "provide their end users one or more methods of updating information regarding the user's physical location." Fine; simple enough to provide a web page for users to enter their location when they take their service on the road.

Alas, not so fast. The Order also requires that the provider include "at least one option that requires use only of the CPE necessary to access the interconnected VoIP service."

I suspect the simplest way to implement this - and the only way it can be done in 120 days - is to go ahead and build the web application, encourage your users to use it (as I suspect that the overwhelming majority of VoIP users who take their ATA on the road also have a laptop with them), and give them the option of calling the customer service number to update their location.

Of course, given some of the reports of customer service wait times of up to an hour at Vonage, it's not clear a "call customer service" option would satisfy the FCC requirement that "any method utilized allow an end user to update his or her Registered Location at will and in a timely manner..."

Server-Side Skype

2005-06-01T23:27:00.000-04:00

Aswath points to a white paper by Connectotel that describes server-side Skype applications, and notes that one could use the techniques described to build competitors to SkypeIn and SkypeOut (and, I'd add, Skype Voicemail).

Alas, playing in this sandbox has some risks.

For one thing, there are limitations. To quote one section of the Connectotel white paper:

It is important to be aware that granting authorization within the Skype client is currently a manual process for which no APIs exist. This implies a potentially large amount of work for personnel who will have to grant or deny authorization one-by-one to the userswho request it.

In other words, authorization can't be easily automated, which makes it rather hard to scale. This limitation of the API points out the second risk: If you do something with the API that starts taking money out of Skype's pockets, Skype can always change the API. After all:

Skype, in its sole discretion, reserves the right to add additional features or functions, or to provide programming fixes, updates and upgrades, to the Skype Software. You acknowledge and agree that Skype has no obligation to make available to You any subsequent versions of the Skype Software. (Skype EULA)

And

Furthermore, you acknowledge and agree that Skype, in its sole discretion, may modify or discontinue or suspend Your ability to use any version of the Skype Software, or terminate any license hereunder, at any time. Skype also may suspend or terminate any license hereunder and disable any Skype Software You may already have accessed or installed without prior notice at any time. (Skype EULA)

Not something I'd build a business case on.

Broadband Price Wars

2005-06-01T10:49:00.000-04:00

SBC has escalated the broadband pricing wars, reducing the introductory price of their entry-level DSL offer to $14.95/month. (One-year term, and the price goes to the then-current price at the end of the term.) This is only $2/month more than their dialup internet access (bundled with local/LD service).

Next step: Incent dialup internet subscribers to convert to DSL and shut down/sell off the dial access network, reducing operational expenses. (Telcos love opex reduction business cases.)

Cablevision VoIP-Ready

2005-05-23T22:48:00.000-04:00

I have Cablevision's Optimum Online service at home, and recently I've been having intermittent problems. After some adventures with their automated customer technical support application, I got through to a real live person, who said it was probably my old cable modem. Cablevision would replace it at no charge; all I had to do was go to their walk-in center and I could get a new one.

Interestingly, the new cable modem is a Motorola Surfboard SBV5120 with built-in ATA. [Correction: with eMTA, embedded Media Terminal Adapter; thanks to "VoIP Dude" for the clarification.] I don't subscribe to Optimum Voice (Cablevision's VoIP service). But when talking with another CTS rep later in the day, he mentioned that they're giving out cable modems with eMTAs (though he didn't call them that) to all their customers, so if the customer later decides to buy Optimum Voice, they don't need a cable modem swap.

This tells me a couple things. First, the price differential between the SBV5120 and the SB5100 has to be pretty small. Second, Cablevision has to be expecting a pretty reasonable take rate for Optimum Voice among current Optimum Online customers.

VoIP and 911

2005-05-19T23:08:00.000-04:00

I think I lose my Telecom Blogger Card if I don't write something about the FCC's VoIP E911 order. So some observations.

First, basing any conclusions on the FCC's press release, instead of the Report and Order that isn't available yet, is risky. The language is imprecise, the requirements are fuzzy, and the obligations are unclear.

That said...

The FCC's determination of who it's placing obligations on seems broad. "Interconnected VoIP service providers," that have obligations placed on them, are defined as those who "enable customers to receive calls from and terminate calls to the public switched telephone network." By that definition, Skype (for example) could be considered an "Interconnected VoIP service provider", as Skype (the company that offers the SkypeIn and SkypeOut services) enables customers to receive calls from and terminate calls to the public switched telephone network. The definition of "Interconnected VoIP service providers" says nothing about how the provider markets the service, or whether there's an ATA supporting 2500 sets - all it says is "enable customers to receive calls from and terminate calls to the public switched telephone network."

The consistent use of the term "interconnected VoIP provider" to describe who the FCC is placing obligations on, but a change to the term "telecommunications carrier" to describe who the ILECs are obligated to provide E911 access to, is worrisome. While I suspect that the overwhelming majority of VoIP providers make use of a CLEC or a third-party provider such as Intrado to provide E911 interconnection, this apparent limitation of the obligations of the ILECs to their current "statutory obligation to provide requesting telecommunications carriers access to their 911 network" makes Chairman Martin's statement that "VoIP providers may interconnect directly with the incumbent LECs’ 911 network or purchase access to this network from competitive carriers and other third-party providers" sound somewhat disingenuous.

Finally, the Commission doesn't seem to have thought through the implications of "nomadic VoIP" - taking one's ATA (or using one's softphone or IP phone) somewhere other than the "subscribed address".

The first conclusion the Commission reached is that "Interconnected VoIP providers must deliver all 911 calls to the customer's local emergency operator."

The second conclusion the Commission reached is that "Interconnected VoIP providers must provide emergency call operators with the ... location information of their customers... Although the customer most provide the location information, the VoIP provider must provide the customer a means of updating this information, whether he or she is at home or away from home." (Emphasis mine.)

For a VoIP provider to deliver 911 calls to a customer's local emergency operator, they have to have E911 trunks to the selective router (tandem switch) serving the PSAP responsible for the local area in which the customer is located. Some carriers (AT&T, SunRocket, AOL) have indicated they may have to limit the availability of their VoIP services to areas in which they are currently able to offer E911 (i.e., areas where they have E911 trunks to the ILEC tandem).

Fine. But what about when someone takes their Vonage WiFi phone to their vacation house in Wisconsin, where Vonage doesn't have E911? Even if Vonage gives them the ability to update their location information - even if Vonage had the ability to figure out their location information autonomously, the holy grail of VoIP E911 - Vonage can't deliver that customer's 911 calls to their local emergency operator, because they don't have the trunks into the local 911 tandem.

Basically, a requirement to deliver all 911 calls to the correct PSAP for nomadic customers is a requirement for every VoIP operator to establish trunking (either directly or through a third party) into the 911 tandem serving every one of the 6200 PSAPs in the country.

Personally, I think it would be amusing to see each of the 400 VoIP providers in North America send a few thousand Access Service Requests to the ILECs, for E911 trunking into every single 911 tandem in the US. We'd see how well the ILECs could service a million ASRs in the next 120 days.

On the other hand, I'd note that Intrado's stock went up 13.44% on Thursday.

Holes in the Lawn

2005-05-17T00:14:00.000-04:00

OK, my last eight posts have been about Skype, so I've got to find something else to talk about.

As I mowed the lawn over the weekend, I realized I have a lot of holes in my lawn. There are the patches that get squashed where number one son and I stand to play catch. There's the spot under the swingset where number two son drags the toes of his sneakers when he's on the swings. There are the patches where the flowerpots go as bases for the kickball games.

I'm glad I have holes in my lawn.

Skype Bits and Bytes

2005-05-16T23:55:00.000-04:00

Lars writes rather dismissively about the bandwidth requirements for Skype over a wireless data connection, quoting and attempting to correct an article titled "Why Skype for Mobile isn't a Big Deal" in an online publication called The Feature, which is owned by Nokia.

First, the quote from The Feature:

Skype says its software uses 0-0.5 KBps when idle, or 3-16 KBps when on a call. That's 30 KB per minute when idle or beween 180 and 960 KB per minute on a call -- which on many mobile networks would run up a huge bill quite quickly.

Then, Lars's attempt at countering the math:

To use the author's own words, there is a bit of a disconnect in his calculations between bandwidth (measured in kbps) and storage (measured in KB) and consequently between his deductions and reality. Multiplying the kbps by 60 does NOT lead to a KB value per minute (as the author has done). There are 1.000 bits in a kilobit (10^3 bits) and 1.024 bytes in a kilobyte (2^10 bytes), while one byte consists of 8 bits. So to convert from a bit-value to a byte-value you need to divide the bit-value by 8. Hence, 1.000 bits are 125 bytes. So by using the author's figures, a bandwidth rate of 16 kbps for Skype would actually mean transfering 2.000 bytes per second or 120.000 bytes per minute or exactly 117,1875 KB per minute - a far cry from the 960 KB per minute claimed.

The only problem is, the author of the article in The Feature was directly citing Skype's figures, which are stated in kilobytes per second:

How much bandwidth does Skype use when there are no active calls?

On average Skype uses 0-0.5 kilobytes/sec while idle. This is used mainly for contact presence updates. The exact bandwidth depends on many factors.

How much bandwidth does Skype use while I'm in a call?

Skype automatically selects the best codec depending on the connection between yourself and the person you are calling. On average, Skype uses between 3-16 kilobytes/sec depending on bandwidth available for other party, network conditions in between, callers CPU performance, etc.

So doing the math myself - Skype uses up to 4 kilobits per second (kb/s) while idle, and anywhere between 24 and 128kb/s for an active call. And the original author's math is exactly correct: an idle Skype client will use as much as 30 kilobytes per minute, and a Skype call will consume anywhere from 180 to 960 kilobytes per minute.

Also note that generally-accepted practice is to use the ISO abbreviation "k" for kilo = 10^3, and the character "K", which does not stand for anything, for 2^10. Thus, a kilobit is 1000 bits, a kilobyte is 1000 bytes, and a Kbyte - usually shortened to simply a KB or K - is 1024 bytes.

So a one-minute Skype call will result in data transfer of between 175K and 937K. Using the Vodafone D2 rates Lars quotes, 0.0963 cents (Euro) per KB, that'll cost you between 17 and 90 cents (Euro) per minute. (For those of you in the US, the cheapest Verizon Wireless plan seems to be 60MB for $59.99, or 0.0977 cents US per KB, so that Skype call would cost between 17 and 92 cents US per minute.)

Why Skype chooses to cite bandwidth figures in kilobytes per second instead of the more common kilobits per second is left as an exercise to the paranoid.

SkypeOut and SkypeIn Gateways, Yet Again

2005-05-13T23:25:00.000-04:00

I think I can milk this topic for one more post.

Looking again at James Enck's item on the Skype architecture, I realize I inferred "media gateways" where I read "gateways". That's not necessarily the correct interpretation. "Gateways" could mean IP-IP gateways, performing a Session Border Controller-type role. This would be the point at which the internal Skype protocol, whatever it is, is decrypted and translated to the version of SIP required by the particular carrier partner, and the media stream is decrypted (as well as little details like counting minutes are taken care of).

This makes more sense to me than Skype converting IP to TDM to handoff to their carrier partners. Still not as disruptive as simply building these capabilities into the client software and letting it run on an arbitrary supernode, but I guess sometimes architectural purity can take a backseat to pragmatism.

Skype and SIP

2005-05-13T22:55:00.000-04:00

In a comment on a previous post, muppetmaster pointed to an ongoing discussion in the Skype Forum. In it, a Skype staffer says directly that "skype client does not implement/use SIP". Apparently a field in the shared.xml Skype config file is named , leading some to believe (as did my interpretation of Skype error codes) that SIP was implemented in the client.

Of course, it's possible that the Skype developers simply borrowed SIP terminology in developing their own protocol. (Developers like to borrow things.) It's also possible that the Skype proprietary protocol is 99.999% SIP. (As I said, developers like to borrow things.) And except to the protocol zealots, it doesn't really matter either way. The Skype protocol is unpublished and its transmission is encrypted, so unless Skype decides they want to allow arbitrary SIP clients to interwork with Skype, it's not going to happen.

And as long as Skype has a larger user base than the SIP world, they're not going to want it to happen. (For a detailed examination of why this is the case, see this paper, quoted below:)

...if network value scales like n log(n), as we argue ... then relative gains from interconnection depend on the sizes of the networks. In this case the smaller network gains considerably more than the larger one. This produces an incentive for larger networks to refuse to interconnect without payment...

Skype Supernodes

2005-05-11T10:48:00.000-04:00

James Enck, commenting on the ongoing banter between myself and Aswath on SkypeOut and SkypeIn gateways, notes that he has "long assumed that Skype has 'seeded' a number of its own supernodes globally from day one." The Baset and Schulzrinne Skype Protocol Analysis observed what they called "bootstrap supernodes" - seven IP addresses and port pairs that always showed up in the Skype Host Cache on the initial installation of the Skype Client. It's very reasonable to think that these bootstrap supernodes are run by Skype, supporting James's assumption.

SkypeOut and SkypeIn Gateways, Part 3

2005-05-11T09:58:00.000-04:00

James Enck reports that Skype "has built and deployed its own-spec gateways," and Aswath is puzzled. As am I, but then, I'm a natural-born skeptic. Aswath's puzzlement is centered on (1) why use G.729a for SkypeOut calls instead of the GIPS iSAC codec if they're your own gateways, and (2) do the calls get handed off to the carrier partner as TDM and then converted back to VoIP?

More conjectures: Media gateways need hardware. I don't see Skype buying DSPs and building boards from scratch; seems more likely they'd be buying the hardware from someone like AudioCodes or Brooktrout. This hardware doesn't support the GIPS iSAC codec. GIPS offers a version of its VoiceEngine(TM) product for embedded devices, which could conceivably be embedded in a third-party media gateway card - but this version doesn't support the iSAC codec. So I can see Skype's inability to use iSAC for SkypeOut/SkypeIn calls. And if you're unable to use the GIPS codecs, then G.729a isn't an unreasonable choice if you want Skype to work reasonably well over low-bandwidth connections (since G.711 is a bandwidth hog, as I noted previously).

If Skype is indeed converting G.729a VoIP to G.711 TDM to handoff to its carrier partners, and the carrier partners are carrying the traffic as VoIP across their networks (pick a codec) and converting back to G.711 TDM to handoff to the PSTN, that would do more to explain the recurring complaints about SkypeOut sound quality. The more codecs you string together, the worse the sound gets.

If James's info is correct, Skype starts looking less and less disruptive and more and more like a telco. (And not only a telco, but a vertically-integrated telco.) And it's not clear to me that being a telco is something Skype is very good at - or even something they should want to be good at.

SkypeOut and SkypeIn Gateways, Part 2

2005-05-10T10:20:00.000-04:00

Aswath provides an alternative conjecture on how Skype clients signal to Skype's carrier partners' media gateways. The two main differences between our theories (if I'm interpreting him correctly) are (1) he thinks that Skype is running dedicated servers as supernodes for SkypeIn and SkypeOut calls, where my theory is that they're using "any old supernode"; (2) he thinks the clients are talking the internal Skype protocol to supernodes, which translate to SIP, where my theory is that the clients are talking SIP with a supernode acting as a SIP proxy.

Given the dearth of actual data this is all built upon, it's hard to argue either conjecture. But there are some differing implications depending on which way it may work.

If Skype is in fact running dedicated servers as supernodes for SkypeIn and SkypeOut, growth in these services will require investment on Skype's part to meet the capacity demand. Given that Skype is collecting revenue for these services - and the revenue is prepaid - that's not a killer; however, it makes the business model less compelling. If they're using "any old supernode", growth is cheaper since they're making use of arbitrary users' computing resources.
Aswath's point about dedicated servers providing a few well-known interconnection points is a good one, though.

The other distinction, whether Skype clients are running "native SIP" with a supernode SIP proxy server, or whether a supernode translates for "Skype-ese" to SIP, I think is more of academic interest.

SkypeOut and SkypeIn Gateways

2005-05-08T01:23:00.000-04:00

I've been thinking about SkypeOut and SkypeIn.

For Skype-to-Skype calls, Skype, as has been documented, uses a proprietary call signaling protocol and the GIPS iSAC and iLBC codecs. It is also reported that Skype uses G.729a for SkypeOut calls. I'm assuming the same holds for SkypeIn calls

Skype shows Colt, Level3, iBasis, Teleglobe, Cable & Wireless, and B3G as "Carrier Partners", implying that these are the companies that provide the interfaces to the PSTN.

Somewhere, SkypeOut traffic has to be converted from G.729a packets to G.711 A-law or mu-law TDM (and the reverse for SkypeIn). And somewhere, call control signaling has to be converted to some standard PSTN signaling for network interconnection.

A couple of questions come to mind:

1. What are the media gateways that convert G.729a VoIP to G.711 TDM, and who owns and runs them?

2. How do the Skype clients communicate with the media gateways?

Now, I have no real information on this topic and have done no experimentation to try to figure this out (it'd be interesting to see Salman Baset and Henning Schulzrinne revisit their Skype Analysis cited above for SkypeIn/SkypeOut). However, some detective work and reasoning can lead to some interesting conjectures. Bear in mind that these are just that - conjectures. Differing views are welcomed...

First conjecture: Skype uses direct IP handoff to their carrier partners; the carrier partners do the TDM conversion.

Reasoning: first, several of Skype's Carrier Partners offer VoIP services with direct IP handoff to the carrier network (e.g., iBasis DirectVoIP, Teleglobe VoIPLink, Level3 (3)Voice Termination). Second, Skype's architecture has an almost congenital avoidance of centralized servers under Skype's control. "Apart from the login server, there is no central server inthe Skype network." [Baset and Schulzrinne] And given Skype's model, I'd find it surprising if they invested in any hardware when they could avoid it.

Second conjecture: SkypeOut call control signaling is SIP from the Skype Client through a supernode serving as a SIP proxy and then to the carrier partner's media gateway.

Reasoning: The carriers cited above provide IP handoff using either H.323 or SIP. SIP is a text-based protocol that would be relatively easy to work with for developers familiar with internet methods; H.323 is a binary protocol defined using ASN.1 that wouldn't. SkypeOut error codes (10403, 10404, 10408, 10484, 10487, 10500, 10503) map directly to and have the same meaning as corresponding SIP Response Codes. Skype uses supernodes to assist with functions like firewall traversal and dealing with NAT, so it seems logical they'd use it for SIP proxying as well - and it seems to me counter to the peer-to-peer model to have supernodes translating from Skype's proprietary protocol to SIP. Though the SIP proxy has to be smart enough to know not to encrypt the messages going to the carrier partner's media gateway.

Alternately, Skype could be using dedicated SIP proxy servers instead of supernodes. That would, however, cause them to incur more capital cost for SkypeOut. Which also seems unlikely.

Not conjectured upon at this time: How does the Skype supernode "find" the appropriate partner's media gateway for a given PSTN number?

Serial Codecs

2005-05-02T23:15:00.000-04:00

In a comment on Tom Evslin's post on SkypeOut, Edward Vielmetti notes "some really bad artifacts calling out from SkypeOut through the PSTN into someone else's VOIP phone, enough to make the call too poor quality to carry on a business conversation."

Skype uses the Global IP Sound iSAC codec for Skype-to-Skype calls, but uses G.729a for SkypeOut calls. Most VoIP providers use G.711.

Various reports (sorry, no links I can find) indicate G.729a gives a Mean Opinion Score (MOS) in the range of 3.75 to 4.0, which is perfectly acceptable. G.711 MOS is generally between 4.1 and 4.5. However, serial coding (G.729a to G.711) generally drops the MOS between 0.5 and 0.75 points.

Not knowing details about Skype's implementation of the G.729a codec or the circuit gateway they're using, it's hard to say more.

VoIP Driving Broadband??

2005-04-21T18:38:00.000-04:00

One of Niklas Zennstrom's statements at VON Canada (Page 10 of the presentation) is that "VoIP is Driving Broadband." Interesting statement, and rather provocative. How many of the 100M Skype downloads were people who were just about to download Skype over their V.92 dial-up internet connection, and then said, "Gosh, I better go out and get that broadband connection before I use this"? How many of the 32.5M broadband subscribers in the US went out and got their broadband connection so they could get Vonage, AT&T CallVantage, or one of the cable VoIP services?

Somewhat less facetiously, is there any data related to broadband purchase decisions that shows that "ability to use VoIP" was a factor in the decision for any significant number of purchasers? I haven't found anything regarding decision factors in broadband purchases on the Broadband or VoIP categories of itfacts.biz.

Widespread adoption of broadband access is clearly an enabler for VoIP, but a claim that VoIP is "driving broadband" seems a bit of a stretch to me.

Synchronicity

2005-04-19T20:00:00.000-04:00

So today, Om Malik cites a Yahoo Finance story that Verizon is offering naked DSL, and Aswath Rao cites a Russell Shaw blog entry about how Verizon's iobi service could be a VoIP killer.

How could Verizon really kill VoIP? How about an incentive to Verizon DSL customers: Keep your telephone service, and we'll give you unlimited local and LD, plus iobi Home, for $19.95/month. E911, line power, five-nines reliability, and use of your existing home wiring, for five bucks a month cheaper than Vonage - with all the nifty features you get from Vonage except portability and a non-local DN.

The only advantage that would leave Vonage is the regulatory arbitrage of avoiding all the add-on fees (911 surcharges, USF, etc.). And as has been shown in the past (see ISP-CLECs and reciprocal compensation), regulatory arbitrage is not a sustainable long-term business model.

Skype: Windows for VoIP?

2005-04-15T21:36:00.000-04:00

One reason (I'm sure there are many others) that Windows came to dominate the PC business is that they had a published API. This led to an explosion of diverse applications - which is what most people buy PCs for, after all.

Of course, Microsoft also builds its own applications, and has integrated some of those applications into the OS platform, undercutting application developers that had contributed significantly to its success and leading to all sorts of nasty antitrust claims.

Skype has a published API. Third-party developers are able to build all sorts of diverse applications making use of a basic peer-to-peer voice over IP infrastructure provided by Skype. And while free peer-to-peer voice communications is pretty cool in and of itself - just like a windows-based operating system is pretty cool in and of itself - it's this potential explosion of applications that is the real "disruptive technology".

Questions remain. How will Skype manage the inherent conflict between providing an open platform for applications developers and providing applications on that platform - especially when they give the platform away for free and make money on the applications? Can a third-party developer bundle the Skype software with an application they sell, built on top of Skype, without violating the limitation in the Skype Distribution Terms to "non-commercial gain"? Do enough applications lend themselves to a peer-to-peer model to be worthwhile? (Interestingly, the three applications that Skype is selling - SkypeIn, SkypeOut, and Skype Voicemail - all seem to hybridize the peer-to-peer model with some level of specialized nodes, either gateways or servers.)

But the potential that is unlocked by an open API is tremendous. Tremendous enough to make Skype the Windows of the VoIP world? Time will tell.

Software Management Humor

2005-04-14T22:53:00.000-04:00

Seeing Ronald Gruia's pointer to Tom Evslin's translations of programmer-speak reminded me of something I got almost ten years ago. I received it from Jim Tomayko, who was at the time on the staff of the Software Engineering Institute at Carnegie Mellon University. I don't know if he was the original author, but I give him all the credit. (There's no copyright notice on it, so if you read this, Jim, I hope you weren't planning on publishing it in a Big Book of Software Humor and making a ton o' money...)

"Essentially complete"
Translation: It's half done

"Schedule exposure"
Translation: It slipped three weeks ago

"We predict"
Translation: We hope to God

"Screen design is lagging"
Translation: Not a single screen exists

"Risk is high but acceptable"
Translation: 100 to 1 odds. Or, with ten times the budget and ten times the people, we stand a 50/50 chance

"Potential show stopper"
Translation: The team has updated their resumes

"Serious but not insurmountable problems"
Translation: It'll take a miracle

"Basic agreement has been reached"
Translation: The &%$#@s won't even talk to us

"Results are being quantified"
Translation: We're massaging the numbers so that they'll agree with our conclusions

"Task force to review"
Translation: Seven people who are incompetent at their regular jobs have been loaned to the project

"Not well defined"
Translation: Nobody's even thought about it

"Still scoping the requirements"
Translation: See "Not well defined"

"Not well understood"
Translation: Now that we've thought about it, we don't want to think about it any more

"Requires further analysis and management attention"
Translation: Totally out of control

"Results are encouraging"
Translation: Power-on produced no smoke

Eliminating the Service Provider?

2005-04-07T20:56:00.000-04:00

In a comment on the previous Disappearing VoIP entry, Aswath writes, "the only promise of VoIP is end-to-end direct communication, unaided by any provider."

Any provider is a bit of an overstatement - someone's providing the broadband network and the software - but I get the drift: eliminate the voice service provider. Yes, VoIP enables this in a way that is impossible in the POTS network. By the same token, I could download a free copy of Movable Type and host this blog on my home PC - but I find it easier to use Blogger. I could run a mail server, but gmail works fine and, again, is a lot easier for me. Eliminating the service provider is possible for a lot of applications, and there will be those who want to do so, for flexibility, control, features, dislike of the service provider, and a variety of other reasons. But there will plenty of people who are perfectly content to let someone else do the heavy lifting.

If the only promise of VoIP is end-to-end direct communication without any voice service provider, there will be a very large category of people who will not particulary care about VoIP.

PaleoISPs, Walled Gardens, and Adoption Rates

2005-04-06T19:41:00.000-04:00

I came across an interesting historical nugget today. In September of 1988 the General Manager of GEnie (does anyone out there remember GEnie?), in a speech to the Electronic Networking Association, cited a prediction that "...by the year 2000, it is estimated that 50 percent of the U.S. population will be using videotex services on at least an occasional basis."

What I found interesting was that the timing was pretty close - Harris Interactive's surveys of internet usage, as cited by clickz.com, show the 50% threshold being crossed some time in 1999, with 63% online as of 2000 - but the services were so wrong. So wrong that the spelling checker in Microsoft Word doesn't even recognize the word "videotex".

Back in the pre-WWW days of the late 80's and early 90's, companies like GEnie, CompuServe, Prodigy, and The Source provided online information services in closed communities - what today are referred to as walled gardens. Then came the Web and the ability for anyone, anywhere, to provide information, applications, and services. Instead of a service provider deciding what content, information, applications, services would be offered to the market, anything that someone thought would be useful could be offered to the market - and the market could decide what would succeed and what wouldn't.

And a small group of people will never be as good at identifying what ideas will be useful to the market as will a very large group of people with the ability to try anything.

The PaleoISPs are largely gone - CompuServe survives as a part of AOL, and Prodigy's name lingers in the SBC Yahoo domain names; The Source was bought and dismembered by CompuServe, and GEnie was bought and later shut down by IDT. Meanwhile, somewhere on the order of 75% of Americans have home internet access.

Lessons for VoIP providers are left (for now) as an excercise for the reader.

Access Charges

2005-03-25T22:07:00.000-05:00

In Light Reading's story on Level 3 withdrawing its forbearance petition, they make the statement that carrier access charges "make up the subsidies used for programs such as Universal Lifeline and E911."

I've seen statements to this effect before, but they're wrong.

Universal Service is funded by carriers paying a percentage of their interstate and international revenue directly into the Universal Service Fund. (Before the CALLS plan was adopted in June, 2000, access charges did contribute to the USF.)

Many (most?) carriers levy a E911 surcharge that ostensibly funds the E911 system, though it's not clear all the money goes to E911. But the E911 surcharge is a user charge, and has nothing to do with carrier access charges.

Access charges are local exchange carrier revenue, plain and simple.

Disappearing VoIP

2005-03-23T22:44:00.000-05:00

Irwin Lazar writes, among other things, that VoIP will have succeeded when "customers will be in a position to buy low-cost phone service that looks, acts, and performs exactly like the phone they have today, perhaps with some added features such as web-based call control."

The ILECs could do that tomorrow. (OK, the added features such as web-based call control might take a bit longer. Though Verizon is offering them now with their iobi service.) All they have to do is slash prices, and customers will be in a position to buy, yes, low-cost phone service that looks, acts, and performs exactly like the phone they have today. And VoIP as a technology will have disappeared from the customer perspective, though not perhaps in the way that Irwin meant.

To date Vonage, for example, has raised $205M in total financing as of 8/25/04. Verizon, for example, has $2.3B in cash on hand and a free cash flow of $4.3B.

In a race to the bottom, which would you bet on?

VoIP succeeded for a little while as an international calling arbitrage play. It can succeed for a little while as a low-cost alternative to POTS. Long-term, providers need to take advantage of the capabilities of VoIP and the architectures that support it to do things that POTS can't do. Like improved sound quality via better codecs, like fixed-mobile convergence, like presence and universal messaging and applications that people haven't even thought of yet.

Otherwise, my money's on the guy with the deepest pockets.

Level 3's VoIP Forbearance Petition Withdrawn

2005-03-22T09:21:00.000-05:00

Level 3 has withdrawn its forbearance petition on VoIP access charges the day before the FCC was required to rule. Light Reading had reported last week that sources close to the Commission were saying that the FCC was going to deny the forbearance request rather than letting the deadline pass, though they may have issued a subsequent ruling further exempting VoIP from access charges.

Currently, VoIP calls are not subject to any access charges due to an Enhanced Service Provider exemption. Had the forbearance request been denied, carriers would have been required to pay switched access charges for VoIP. Switched access charges are paid to the local exchange carrier for calls sent to, and calls received from, the LEC; the average price of switched access is around 1.2 cents per minute for interstate calls and about 1.8 cents per minutes for intrastate calls.

Level 3 maintains that VoIP calls should be subject to reciprocal compensation charges, under which the interconnecting carrier pays for calls sent to a LEC, and receives payment for calls received from a LEC. The average price per minute for reciprocal compensation is about 0.1 cents.

It seems that with the clock running, Level 3 felt that the FCC would potentially reach a decision unfavorable to them and the VoIP industry. Withdrawing the petition allows things to continue in their current state of uncertainty, which for VoIP carriers is certainly preferable to an unfavorable FCC ruling with a possible future rule change.

The carrier switched access charge structure was originally developed at the time of the AT&T divestiture to compensate local exchange carriers for the cost of "exchange access", connecting users' telephone lines to interexchange carriers' networks. It was an implicit subsidy mechanism; rates were set more than an order of magnitude higher than rates for interconnection of geographically-adjacent local exchange carriers (e.g., a Bell company and an independent telco). Switched access charges have come down significantly since divestiture; however, they are still more than an order of magnitude higher than reciprocal compensation charges, which generally tend to be more reflective of the actual cost of interconnection.

Broadband Voice and the PSTN

2005-03-14T21:48:00.000-05:00

Several blogs (Om Malik, Aswath Rao, Andy Abramson) write about how the voice quality of SkypeOut is not as good as the voice quality of Skype because Skype uses a wideband codec (the GIPS iSAC codec) for Skype-to-Skype calls but uses G.729a for SkypeOut calls. Aswath comments that it "looks like the interconnect providers still do not see the need to upgrade their gateways to support the wideband codec."

The problem is, even if someone built a VoIP gateway that supported the iSAC codec, the circuit-switched side of the gateway would still be talking G.711 to the PSTN. You could get voice quality approaching that of a PSTN call, but you're not going to get the 8kHz of voice bandwidth that you can get on a Skype-to-Skype call (with sufficient network capacity and performance).

Of course, PSTN-quality is better than G.729a (by about 0.5 MOS points), so by using G.729a Skype is accepting a voice quality lower than PSTN for SkypeOut calls. I would conjecture that they choose to do this because G.711 over IP is a bandwidth hog (about 95 kb/s with 20ms frame size, compared to about 40 kb/s for G.729a). But given that a carrier can achieve PSTN-quality (at the expense of bandwidth) simply by using G.711 for the sessions that interwork with the PSTN, what's the motivation for them to upgrade their gateways to support wideband codecs?

Funny, though - with wideband codecs, sound quality could be a differentiator for VoIP services, rather than a detriment. (I'm by no means the first person to think of this.)

Network Reliability

2005-03-07T20:20:00.000-05:00

There's been a flurry of news about VoIP network outages the first week of March: Light Reading and Om Malik's blog (and a commentary on Om's blog) seem to have the most information, unless you want to dive into the Vonage Forum or Broadband Reports VoIP forum.

If you read product datasheets for VoIP equipment (softswitches, media gateways, application servers, take your pick), you will be flooded with terms like "carrier grade", "99.999% availability", "fully redundant", "live software upgrades", and "no single points of failure". I've even seen a vendor claim of "99.99994% availability", which means that over a five-year deployed life, a system will be unavailable for about a minute and a half.

So how do you reconcile all this extremely reliable equipment with networks that go down for hours at a time?

One possibility is simple: Vendors lie. Or, to put it less judgementally, vendor claims of equipment reliability are theoretical calculations that are not borne out in real-world deployment.

Given the historical perception that VoIP is less reliable than traditional phone service, there is clearly an element of marketing hype in vendors' reliability claims. But any vendor that tries to sell into a telco knows that their claims are going to be held up to some level of scrutiny, and the methodology they use to forecast reliability and availability had better be "generally accepted in the industry". For the most part, that means forecasts in accordance with methodologies and models in Telcordia Reliability and Quality Generic Requirements. While 90 seconds of downtime over a five-year lifespan (which realistically means that of 20 boxes deployed for five years, 19 never go down at all and one goes down for half an hour) may tax one's credulity, I don't think that there's a strong reason to believe that VoIP network equipment is intrinsically less reliable than traditional phone network equipment on a box-by-box level.

I think the actual reason is much more subtle and deep-seated, and is pervasive across equipment vendors and carriers. It's not one of equipment, or of engineering, but of culture, and of a mismatch between corporate culture and customer expectations.

"Traditional" telcos, and "traditional" vendors, have a culture of reliability above all. If this is your mindset, you do things like rigorously testing new software before deploying it - and you deploy it at 1:00 AM on a Saturday, not midday on a weekday. You have processes and procedures in place to deploy with well-defined checkpoints, safe stop points, and capabilities to backout. Vendors with this mindset do their own rigorous regression testing before releasing new software to their customers.

This has its own downside - testing adds time and cost to software releases. Processes and procedures slow things down and require a different mindset. It's hard to "get an idea on Tuesday and deploy the service on Wednesday" if you have to integration test and regression test the software with everything in your network - let alone if you have interoperability with other networks to worry about.

Neither approach is wrong. The problem comes about when the company's culture clashes with the customer's expectations (which, of course, mostly come about from the company's positioning of the product). If Skype stops working for a couple of hours, there's some grumbling but not a lot of repercussions - after all, it's free, the users are self-identified early adopters who tend to be tolerant of glitches, and they buy into the value trade-off that Skype presents them. If SkypeOut has problems, the grumbling escalates, because SkypeOut is a paid service, and the customers' expectations are higher. And a mass-marketed "phone service" that you can buy at Circuit City generates expectations of reliability and availability consistent with that of the "phone service" that people have been accustomed to for the last 75 years. If the culture of the company providing that service is more focused on providing low-cost service with new features than on providing reliable service, problems are inevitable.

Companies - both vendors and carriers - have to decide what they want to be, they have to communicate that message and those expectations to their customers, and they have to live them. If your marketing message is that you're a Phone Company, you've just bought yourself 130 years of history that people are expecting you to live up to. Those expectations can help you win a lot of customers. But if you're really trying to be a fast-moving, low-cost provider of disruptive technology for voice communications, and your corporate culture is based on that model, perhaps "Phone Company" isn't the message you want your customers to walk away with.

463 West Street

2005-03-07T09:17:00.000-05:00

In 1907, Theodore Vail concentrated all of the research and development activities of the Bell System at the Western Electric plant at 463 West Street, New York. Eighteen years later, this building became the headquarters of a new company, jointly owned by AT&T and Western Electric, called Bell Telephone Laboratories, Inc.

There are now apartments at 463 West Street; what used to be Western Electric is now Lucent Technologies, Avaya, and Agere Systems; and AT&T is being bought by SBC in a sort of reverse-Saturn-devouring-his-offspring move.

Times in telecommunications continue to be interesting, and I hope this blog can contribute some small amount of insight, with some historical perspective.