After just a few days of no or unreliable service, the Skype ecosystem is under attack with commentators and competitors alike both taking a look at the leading VoIP service and seeing more than on chink in the previously polished armour.
[UPDATE: See Skype Blames Massive Reboot For Failure]
As you would expect, the competitors came out on the attack. In Australia, phone card distributor and more recently ISP and VoIP service provider, gotalk, released a statement saying it figures the outage created an opportunity for other players to make a raid on Skype’s loyal subscriber base.
CEO Steve Picton reckons the outage made Skype’s domination of the Australian and New Zealand VoIP markets “even more vulnerable to attack” from full service local competition. Picton already had Skype in his sights, however. Earlier this month his Tier 2 telco, gotalk, announced plans to “challenge the local Skype domination” by launching Australia’s firstPrePaid VoIP service attached to all their CardCall phone cards.
The junior telco provides a range of wholesale and retail telecommunication services through its own switching centre in Sydney’s Castlereagh Street which handles over a billion international minutes per year which in turn equates to 20% of all Australian outbound international call traffic.
“A recent research project undertaken by PhoneChoice, “Home Phone and Internet Provider Comparison (July 2007)” specifically compared gotalk prepaid VoIP with Skype Pro. Based on typical usage, the PhoneChoice survey clearly showed gotalk’s prepaid VoIP was cheaper overall than Skype’s Pro,” Picton said. Of course this doesn’t account for any number of free calls users might be able to access to the 200 million plus Skype subscribers.
“When you add in the fact that gotalk is an Australian owned company, a full service telco business offering 24/7 locally available customer service – (as opposed to Skype who can only be contacted by email) – I believe gotalk can be seen to be a more than competitive alternative to Skype,” the statement said.
Whether the outage, which Skype blamed on a rogue algorithm that prevented users from logging on to the system, will have any long term affects on user loyalty is doubtful. But Mark Main, senior analyst at Ovum sees the outage, which remained intermittent for several days, as just one issue in an ongoing decline at the peer-to-peer telecommunications company.
All Is Not Well
“There have been a few signs in recent months that possibly all is not well with Skype. Numerous people have remarked to me that ‘Skype is getting worse’. The difficulty analysing such problems is that all the evidence is anecdotal and certainly not based on a good statistical sample. That the user base continues to grow suggests that the overall user experience has been good enough, so far,” says Main.
“There is still a danger that services designed to be highly disruptive to traditional telecoms business models have been developed without sufficient regard for resilience, something we have been saying since consumer VoIP came to the fore during 2003. Telecoms engineering is no different to any other product development – there is always a commercial penalty to pay by compromising reliability or quality. You still broadly get what you pay for in telecoms.”
“Outages and quality problems with VoIP service providers are sadly still too common – most service providers have glitches that can last from a few minutes to a few hours. That Skype has gone so long without a prolonged problem is quite an achievement,” pints out Main. “However, this event is probably Skype’s biggest hiccup to-date, as it approaches its fourth birthday.
“Rather more worrying is the admission by Skype that it is an algorithmic problem – Skype always claimed scalability due to it peer-to-peer mode of operation. A problem with the peer-to-peer algorithm also makes it more likely that very large numbers of users have been affected at some point, although Skype hasn’t yet been specific about the scale of the outage,” he continued.
“So the bottom line is that while Skype is now a well-established service it has largely flourished through its simplicity, good-enough quality/reliability and user-endorsement. However, there are no contracts to tie-in users and the approximately five million people that we estimate pay Skype for value-added services are those with a more compelling reason to stay. Skype will need to work hard to make this outage event a one-off or its loyal user base could be enticed away by other, better VoIP offerings. There is plenty of choice,” concluded Main.
What Went Wrong
Over the weekend Skype posted a notice to its site that the problems were resolved and that the service had stabilised and returned to normal. The company committed to providing its users with a more detailed explanation of what went wrong on Monday (US time).
But already the pundits have been out in force attacking Skype even before they have all the facts at hand. Skype was forced to deny it was the victim of a Denial of Service attack. A rumour which started when a number of Russian websites claimed responsibility for the outage and published code for the supposed attack.
The company has also denied that a scheduled maintenance window the day before had anything to do with the outage, or that it was the result of the Microsoft Windows monthly patch cycle which occurred just before. [Update in fact Skype now blames the regular Windows Patch Cycle for simutaneously rebooting a large number of machines worldwide]
In fact somebody somewhere has claimed the glitch was in the Skype system since it was originally written. (sorry I still can’t find the link for this one.)
Whatever the reason for the failure, many pundits, bloggers and commentators are now arguing that the reliability of VoIP services remain a critical factor in widespread take-up.
Celebrity VoIP blogger Om Malik (who let’s face it can be a bit of a goose sometimes) wrote a provocative post titled: “Does Skype Outage Expose P2P’s Limitations?” which argued that: “Peer-to-peer by its very nature is supposed to work without a problem…” but with users unable to log in it “makes you wonder about how resilient are P2P services.”
Then later had to come back with: “On second thoughts, I want to be clear that if you are going to build a mass market consumer service on P2P and use authentication servers or add layers on top of the basic architecture, then you are on shaky ground and need to build in some sort of redundancy,” he wrote after another blogger pointed out he was confused. But the supposed failure of P2P architectures was out there. With Om telling the likes of “Joost, Babelgum and other P2P companies should be concerned about their business prospects going forward”.
This opened up a chance for Sightspeed, another P2P VoIP start-up to use the topic as a bit of leverage to promote its own service which users company servers instead of the Skype Supernodes, which Om at least initially seemed to think was the cause of the failure.
Quite apart from the looming security issues that have dogged the technology, it could be this very public failure that is ultimately blamed for any delay in widespread adoption of the technology.
Perhaps it won’t have much impact at all, as Garrett Smith, the Anchor blogger at Smith on VoIP wrote in a posting “Did Anyone Care About The Skype Outage?” that there was “not much coverage of this event outside of the industry and even for those of us within the industry, it was not life-altering.”
“It would seem,” he continues, “that the lack of doom and gloom caused by the Skype outage is a sign that Skype is not being used as a primary, mission-critical means of communication. Businesses and individuals are most certainly using Skype, but they are doing it in a manner that simply augments their normal voice service.”