Firefly.. softphone and network but not P2P

A reader anonymously wrote in to suggest that I give Freshtel's Firefly a test drive. While it looks like another nice option, competing with FWD and others, it doesn't appear to be P2P... am I missing something? I believe that there is an important distinction between the VoIP providers with centralized arhchitectures and a P2P architecture. The three providers I have found so far in the later category are Skype, ITALK2U, and Peerio -- anyone know of others? Of course each of these has drawbacks starting with the reasonable argument that Skype shouldn't be considered P2P at all... (see earlier post on Why Skype is No Different...)

Review of ITALK2U

Alec Saunders sent email with a link to his review of ITALK2U and a comparison to Skype. Thanks Alec! In part he writes:

iTalk doesn't yet support conference calling, or encrypted calling, and doesn't have the Skype missed calls list, but other than that appears to be a functional clone of Skype. It's also a bit more standards based, in that it supports H.323, rather than a proprietary signalling protocol, but it uses a proprietary codec, which makes that somewhat moot. Voice quality is similar to Skype, which is to say very good, but I don't know how well it will handle dial-up. The codec seems to be some kind of a learning codec, in that quality improved over the duration of the call. MSN messenger is the same way.

Click on over to Alec's blog for the rest of his rundown and screenshots.

Another Skype competitor

Here is another competitor to Skype in the peer to peer VoIP world -- called Peerio it appears to have a number of advantages over Skype... with a few caveats though. First of all it is open source, second it has a developer API which will easily allow others to build Peerio compatible applications, third it claims to be entirely serverless (as opposed to the directory architecture required by Skype) and fourth it is based on H.323 and SIP standards.

Now for a few caveats... their web site definitely leaves a lot to be desired with lots of holes to be filled (although the promise is there). More problematic is the half-hearted approach to open source. On the site they write:

Despite our efforts, there are two necessary exceptions to the open source principle:

1) The source codes for the Peerio444™ core will not be released.
2) Peerio444™ will not function in embedded systems and is limited to standard PC use (x86 and PowerPC).

Why is that?
The source codes for the Peerio444™ core will be kept proprietary so that we can avoid spawning hundreds of proprietary commercial systems (see Linux) that could jeopardize our ability to maintain a Forever Free platform. To be Forever Free we must protect the system from profiteers. Support and maintenance of the Peerio 444™ core will be limited to our research and development team.

This is like saying that you can do any work on the car except for modifying the engine -- we have welded the hood shut to prevent you from making modifications to anything inside that compartment.

This flies in the face of good open source policy for a whole bunch of reasons but I'll just mention a few here:

• At anytime Peerio can change the definition of what is inside the core and what is outside the core, endangering other projects that have been built on Peerio.
  
• The argument that this protects the system from "profiteers" is spurious -- by contrast placing assets in the open source community creates a level playing field
  
• Most importantly, Peerio will defeat its own strategy by holding back this key element of the product -- open source only works as a viral means of distribution when it is truly open.

Part of the point of open source is that by using it, a developer is no longer dependent upon the whims of a given company for the future of their development project -- The developer knows that the core building blocks will always (Forever?) be available. By holding back this key element as proprietary property Peerio eliminates that key advantage of open source and will significantly reduce the number of companies and individuals willing to invest in the future of the project.

On another front, I have been meaning to test the beta of ITALK2U which has been out for awhile from Endoreal/LitFiber. I downloaded one version and couldn't get it to work, but have been meaning to download a newer version. If anyone out there has had any success, please let me know.

How cheap can it get?

Already it is significantly cheaper to use VoIP providers like Vonage for your telephony needs, but now there is a price war between the low cost providers! According to this NetworkingPipeline story, Vonage Cuts Prices for Internet Calling

The company insisted it was not launching a price war in a market that is just starting, saying that the price cut reflected its decreasing costs in doing business.

Right. And as the article goes on to point out, AT&T announced their CallVantage plan just 6 weeks ago with an introductory price of $19.99 a month for the first six months. I am guessing that would have slowed Vonage's customer acquisition rate.

More time to blog...

After almost 7 months of work and a considerable financial investment, Fred Ball and I along with the founding partners have decided to turn out the lights at CallTrex.

The immediate cause of death was our inability to close follow on financing to help grow the company. A call center model only works at a significant scale of operations. Below a certain threshold, fixed costs represent too large a part of the financial model and it is impossible to make money. We were operating at about one-third of the scale that we needed to be to start generating top line profit. At least another $300,000 was needed to arrive at this first milestone. A number of angel investors moved in the direction of funding the business but pulled back at the last minute. Others simply moved too slowly. In the end, the Board of Directors expressed a lack of confidence that we would close this angel financing round in time and approved a plan to shut down operations.

As in so many cases, there were a set of underlying causes of death that led up to this end point. A large number of people had warned that selling to the small and medium business market was fraught with peril -- this certainly proved to be the case. Customer acquisition was slow and expensive and we never reached a scale where a larger distribution partnership made sense. In addition we suffered from technology problems which could have been solved with more resources, but we found ourselves spending money to fund the call center operations and starving the technical team. Technology problems sometimes meant the loss of customers which further slowed our growth rate.

Despite all of these challenges, CallTrex had a unique value to offer the marketplace, with an innovative system for connecting our customers to their remote staff and allowing that remote staff to provide a great service to our customer's callers. It is sad to see it go. As we have notified our customers that the service will be turned off, we have received some terrific praise on how great our service was.

I will be taking some time off and thinking about what I want to do next. Definitely spending more time with my daughter -- 7 months old and doing great!

Asterisk Use Accelerates...

The adoption of a technology outside the initial core group of visionaries is an important milestone on the path to greatness or abandonment. Recently a number of folks have pointed out that Pulver's "Free World Dialup" now supports Asterisk -- you can read about it here on -- Pulver's own blog . A comment over in North American Bandwidth news:

Soon, protocols will be irrelevant since all VoIP devices will support all of them.

An interesting point -- though it misses a key issue. Translation can replace standards as long as the standards are transparent and processor speed is cheap and plentiful. This is true for all open protocols -- think currencies, human language translation... But not so for closed, proprietary protocols. As an open-source VoIP offering, Asterisk makes it easy for anyone to support since their protocol is published. Translation for Skype, on the other hand, might prove much more difficult.

On another front, I spoke to someone who was told by Global Crossing that they are now doing tests to certify Asterisk. If they decide to formally support Asterisk, this will be an even stronger indication that we are on the road to wide adoption rather than oblivion...

COMMENT 5/8/2004 1:40:50 AM Roland Tanglao (www) said:

Hi Ted:

Yes, it can be difficult to translate and emulate a closed protocol like Skype, but it's not impossible.

If Skype doesn't interoperate and coexist with other VoIP protocols but in spite of this still becomes popular with the mass market, somebody will reverse engineer it!

Corporate Cockroaches

One should never underestimate the lowly cockroach. If any creature could survive a nuclear holocaust, it would likely be cockroaches. They have a capacity for survival that surely puts them in the Darwinian All-Stars. A few days ago a friend of mine (who works for a very large company) and I were talking about the people that somehow survive layoff after layoff inside his company. "They are cockroaches!" he exclaimed. A kind of corporate cockroach. Through a combination of hoarding information, avoiding accountability and spending their work time on building a network of relationships these individuals find cozy cracks and crevices to survive the reduction in force bombs that explode on a regular basis.

One can't entirely blame the cockroach, however. If corporations made their employees feel a little more secure in their positions, perhaps they wouldn't feel the need to spend their intellectual energy on how to protect themselves instead of on how to get their jobs done. Or is society to blame because we learn to attach our own sense of self-worth to the work we do, so losing a job can be a more terrible blow to self-confidence than virtually anything else?

Whoever is to blame, a manager's task is clear -- take out the RAID can and exterminate the beasts. Watch for those three signal events -- hoarding information, avoiding accountability, and inordinate amounts of time "networking." Businesses have to be able to have "roles" and not "faces" as another of my friends recently stated. We each play a role within or organization. Part of our job should be making sure that someone else can play that role if we move on. And we should look to our teams as people playing roles as well. The moment you believe that a particular person is essential, rather than a role being essential, you should recognize that you have a problem.

Will VoIP kill the telco star?

A nod to pop music group the Buggles* in the title of this Red Herring website post on VoIP -- Will VoIP kill the telco star the latest article pointing out how mainstream telephony is fumbling the opportunity to emerge as winners in the new IP based world that is rapidly emerging...

A particularly interesting point made in the article -- a report of one individual's experience in switching over to VoIP -- the phone company did nothing to try and keep this business:

Qwest did not handle the transaction well, failing to complete the hand-off of my old telephone number and cutting me off from the world for several days. Rather than competing for my business – there was no call asking if they could find a way to keep my business after I requested the number be moved – they made it difficult for me to stay connected.

I love the comment from the Qwest customer service person as well when the reasons were explained for the switch:

Wow, that's a lot of savings every month.

The author reports savings of 87% on his total phone bill.


*(Video killed the Radio Star on their Age of Plastic album)

Too late to close the barn door...

Great blogging of the VON show by industry pundit Paul Kapustka. I like his post today in particular about the strength in numbers approach to the regulatory questions being asked of the VoIP industry. When VoIP was a flash in the pan with a few startups playing, the ancient telecom companies might have been able to deal the industry a fatal blow with some heavy regulatory action. But now that there is real momentum and many large players getting together to create a political lobbying power, it is reasonable to believe that the cow is long gone from this barn...

Why Skype is No Different

Great link found on Lance Tracey's Blog on why Skype is no different from other VoIP services. A thoughtful analysis by Aswath Rao on what Skype really is and is not, and how its technical architecture embodies faults that undermine the P2P marketing message. Excerpt:

Some have expressed reservation because Skype is proprietary. There have been previous instances where proprietary consumer items have found wide adoption without incurring huge collective cost. VCR is one of the examples that come to mind. But in this case there are some differences:

• Alternatives, based on standards are available
  
  
• Skype uses mostly well-known and open technologies; only the protocol semantics is proprietary
  
  
• Even though Skype (for that matter VoIP) is naturally a “product” and not a “service”, Skype views it as service. For example, they do not allow an enterprise to use their own GIS, instead of the global one, even if communication will be restricted to internal use alone.
  
  
• As I am told, there is no way to directly address another client, even if the IP address is known. Windows Messenger from Microsoft has the same limitation, whereas NetMeeting allowed direct communication.

In this respect also, they are just like other VoIP providers. It is disheartening to see that even those whose middle name should be P2P, think like this.

Competition for Skype

It was bound to happen. The model is just too darn attractive. Other companies were going to enter the p2p VoIP business, it was just a matter of time.

I just finished talking with the founder of one such company about his company and product -- John Jarvis of LitFiber told me that they are just a few weeks away from a beta release of their new p2p VoIP product iTALK 2U.

John claims that the sound quality in their product is significantly better than what Skype can achieve due to a significantly better codec that his team has built from the ground up. Even more interesting is the way he is approaching the market though -- he has built his product to be H.323 compliant.

Compliance with VoIP standards means that users of his product should be able to connect to any other VoIP infrastructure that supports the standard. This is a terrific competitive advantage over Skype's current closed architecture approach and should really transform the game. With this product, any of the companies that have made VoIP investments will be able to leverage the p2p VoIP phenomenon using LitFiber's product.

I eagerly await the release of the download...

Anyone in their garage can now be a telephone company...

Jim Kohlenberger blogs on VoIP
Excerpt:

VoIP has decoupled voice from the physical infrastructure itself. Now it can go independent of who your physical provider is. Of course, this challenged incumbent interests in that regard. Because of this decoupling it has allowed many new things. Microsoft build SIP (the protocol) just like HTML (http) and FTP, SIP is a protocol. Anyone in their garage can now be a telephone company for the entire planet. In shopping e-commerce you can get a voice rep. You can do online voice conferences. Included in game boxes and game consoles where kids use VoIP to communicate. Used in instant messaging. Star-trek type communicators are now used in Hospitals (Wi-fi based). All sorts of other things. Skype. Free World Dialup. Metcalf’s law: value of Internet goes up as you can talk to more people. New companies are broadband VoIP providers. Vonage, 8x8, etc. Plug in a regular phone to IP network and make calls. All types of new services are possible. Right now they look like regular telephones because that is what we expect. Many things are possible in the future. Question is how do we get there? How do we allow them to flourish so these new services can be made possible?

COMMENT:
On : 3/25/2004 7:27:06 AM Joshua Ruihley (www) said:

The "Jim Kohlenberger" post on Smart Cog isn't actually a blog posting by Jim. The post is of my notes from his presentation at the Internet Commons Congress yesterday.

Conferences, Panels, Speakers... getting your time and money's worth

Putting on a conference is a huge undertaking. Having it be a good conference with inspiring speakers, fascinating roundtables, and great networking is even more difficult. So the organizers of this week's Open Source Business Conference should get credit for meeting most of their objectives and inspiring, fascinating, and networking at least some of the time.

Two major places where the conference felt weak to me:

1) Too many gratuitous keynote speeches by event sponsors. Its OK to have a keynote from someone at a sponsoring company, but they have to have something to say! A tired flogging of the sponsor's products and market strategy (especially if the speech has been given before at other conferences... you know who you are) doesn't do the sponsor any good and cheapens the conference experience for attendees.

2) The panel format is difficult at best to pull off. The two key ingredients are a tough moderator and some tightly defined issues that the panel (and audience) can dig their teeth into. Going deep on an issue and really uncovering some great ideas is much more valuable than a skin-deep flyover of a broad set of questions.

Criticism aside, the idea of a conference on Open Source Business is definitely one whose time has come and I am certain that this conference will only continue to improve with time.

My thoughts on the open source investor roundtable

This group of venture capitalists seem to be largely interested in investing in commercial software -- gee, if you can use open source as a leverage point against a larger commercial competitor to get distribution scale or infrastructure -- great. But there is no open source business model. Proprietary software advantage is what will succeed.

This is quite a contrast to Ray Lane's presentation of the watershed that the software industry is confronting, or the comments that Clay Christensen reportedly made yesterday (I wasn't here) about how open source transforms the way we think about the software product -- instead of a core piece of software wrapped in service, documentation, etc -- the wrapping becomes the product.

So are the VCs wrong and stuck in the past? Or are they the pragmatic ones and the idea that open source can change the software industry is polyannish at best?

More from OSBC -- Open Source Investor Roundtable

Panelists:

Opening comments:

Dave Power, Fidelity Ventures -- open source is a way of getting a headstart and a way to build a community, but it is not a business model in itself

Peter Fenton, Accel Partners -- open source is one of the top three disruptive forces in the software industry, we don't make a bet on organic uptake but it has to be there, a viral broad and growing community around an open source product -- there has to be run time, or production value -- the oxygen that allows you to charge for support and thirdly there needs to be demand side economies of scale -- a dominant provider vs. many small providers, and lastly there has to be product drift over time -- sendmail doesn't have much product drift for example, that is the key to harvesting revenue streams for open source vendors

Tim Guleri, Sierra Ventures -- As Peter said, there has to be an organic uptake of the product, it has to be a group of people that has control over the way the open source community contributes to that open source product

Steve (?) -- we are looking at companies that are building applications on top of an open source platform


Moderator Question -- as venture capitalists you are making significant early bets, what are the business models you are making on open source?

Dave -- three models that make sense -- 1) a dual licensing model, an open source version but also a commercial version that they can make money on; 2) enhancing open source software with proprietary software and 3) a service model

Peter -- those are the revenue models. But as a VC we want a scalable, high margin revenue model. Consulting companies shouldn't be looking for venture money.

Tim -- there is a commercial tipping point at which big companies have deployed a certain amount of an open source product and then someone asks, is there a commercial version of this product?

Steve -- we are in a period of transition in the way that corporate america thinks about open source -- they have started to embrace it with the caveat that they still want to have a commercial vendor that gives them comfort that there is a support function -- but will they continue to evolve? will they become completely comfortable with using open source without the need to have a commercial vendor associated with these technologies?


Moderator Q -- relationships with the communities, as the products become more commercial how does the community react?

Dave -- in each case in the beginning there is a community leader of the open source product, a person that reaches out and then there are hundreds of thousands of downloads, that creates a community. But over time as you get more successful the companies that form around these products are doing more and more commercial software development so there is a challenge about how these companies continue to nurture that community that got them started. We are in a window where there is an opportunity for startups to do something special here -- but large companies are starting to figure out how to manage these open source communities.

Peter -- the starting point is some mercurial founder that gets the ball rolling, and builds a viral contributor base but this shifts over time. Once a project is established the issue becomes maintaining and this requires a different kind of support from the company built around the project

Tim -- we make sure that we fund, maintain, and make sure the open source community thrives -- in our market segment we keep the commercial and open source in sync (security product)


Moderator Q -- Where is the value of open source over time?

Dave -- the value will be up and down the stack

Peter -- anywhere you can get huge distribution efficiencies and broad developer interest

Tim -- every segment of software will get affected in a positive vein by open source. I think the SMB space is where a lot of this innovation is going to occur -- maybe GM won't want to run their financials on open source, but the 10 person parts company will want to.


Audience Q -- comment on standardization for interconnectivity APIs

Tim -- standard XML protocols that are self describing

Peter -- standard product sets - JBoss, mySQL, running on Linux -- a standard open source software stack and that drives standards

...

David Ritter continues...

but I was wrong about having power... yes there is a plug, but no it isn't providing any electricity.. so I am signing off for now and will be back when I find power...

Dave Ritter Keynote

Making Connections: Lessons on the Power of Networked Communities

David Ritter is Vice President of Strategic Technology for the Boston Consulting Group

Opening comments:

Open Source is about people. And it is about networking

Breakthrough creativity solving complex challenges by participants who are geographically and organizationally dispersed, who don't do it for direct monetary reward but display unusual enthusiasm...

Ray's closing comments

We are just starting the "golden age" of our industry

Open source is the key factor (besides web services) that will start that

Software will be delivered as a service. Software IS a service. A service company is a different DNA than a product company. Software companies have to recognize this shift.

Software has always been a service business and we just haven't been honest about this.

Ray continues -- How has the formula changed?

90s formula -- ROI = I x M / C

invention times momentum divided by the cost

NOW -- ROI - P(I) x U / TCO

Proof of innovation times useability divided by total cost of ownership

Software industry at the start of this decade is a $200 billion industry, 70% in the US, representing 1% of the US GDP, and employing 325,000 people

Today, (2004), Open Source Software and Web Services changes the marketplace. Apache is 2x Microsoft. Microsoft is only 3x Linux. India and China are becoming enormous. And the enterprise is adopting OSS

Here is what enterprise CIOs are saying about Open Source:

Positives -- Lower TCO, Equivalent Functionality, Equivalent Quality, Bi-directional scalability, Faster bug response fixes

Some challenges -- Informal support, velocity of change, no roadmaps, functional gaps, licensing caveats, ISV endorsements

What does the software industry look like in 2010?

OSS destruction > Growth -- the industry will be the SAME SIZE at the end of the decade because Open Source will destroy as much revenue growth as occurs in other parts of the industry

US Share < 60% -- more and more is going to move off shore

<1% of US GDP -- GDP will grow faster than revenue in the software industry

< 325,000 people -- the US is not supplying the engineers for this industry

Ray continues -- what the software industry needs to do

More predictability -- not necessarily perfect software, but better predictability about when software will work

Ubiquitous Access -- this is a worldwide marketplace

Continuous Updates

Service-Oriendted Architecture (network/federated data)

Flawless Service