Alec Saunders SquawkBox

It appears that Vonage’s (VG) worst nightmare is about to come true.  South Carolina based Motley Rice (motto: “Litigating Today For a Better Tomorrow”) has filed a class action suit on behalf of Vonage Shareholders, alleging improper conduct in the sale of pre-IPO stock to Vonage customers.  There are a number of things said, but the most outrageous is the claim that Vonage, failing to sell the stock to institutional investors, then turned to its own customers.

The Complaint further alleges that, Defendants, realizing that institutional investors who normally buy in IPOs would be reluctant at best to purchase Vonage shares as-priced, pre-sold at least 13.5% of the Company’s IPO shares to Company customers in violation of NASD Rule 2310. NASD Rule 2310 requires that a company recommending the purchase or sale of its securities to a customer must have a reasonable basis for believing that the recommendation is suitable for the customer. The Complaint also alleges Defendants had no such reasonable basis in this case and improperly crammed investors into the Vonage IPO regardless of their suitability.

It appears that Vonage management may have been actively contemplating scenarios like this for some time.  Kathleen Day, writing in Saturday’s Washington Post, outlines a number of possible scenarios, including the uncomfortable spot that Vonage would find themselves in if they were to sue customers who refused to complete the purchase of IPO shares.

Other blogosphere commentary takes Vonage shareholders to task for wanting a risk-free ride on the IPO. As Cynthia Brumfeld commented, “it’s only the beginning of a very ugly ride for Vonage”.

Mike Arrington has written about GotVoice, a service that collects all of your voice mail into one inbox.  It does this by calling each of your voice mail systems periodically throughout the day, and collecting the voice mail from them.  Conceptually, this is just like have one email client poll multiple SMTP servers, and collect your email.  Except, of course, that this is far more primitive.

Returning to Carl Ford’s feature interaction post, this is a perfect candidate for passing a call off net to a third party.  In a SIP world, if I choose to use a third party voice mail provider, then the network proxy should simply route the call to my chosen voice mail provider.  Traditionally, UM systems have offered the opposite functionality; they grab an unanswered call back from the network, and force it to the UM system.  They do this because of inconsistent feature implementations on networks.  For instance, in the GSM world the network can be programmed (by end users, no less!) to offer different behaviour and different voice mail systems on call forward busy, call forward no answer, or call forward.  No such facility exists in the PSTN.  GotVoice takes a polling approach, which is workable but will incur network charges to poll and retrieve the voice mails.  The best approach would be to have the voice mail all deposited on the server of the customers choice from the beginning.

On the Pulver Blog, Carl Ford has written a short piece on feature interaction.  He notes that the web model for SIP doesn’t seem to be being applied as one would expect.  In effect, telecom companies are rolling out SIP services to their islands, because they don’t want to pass control of the call to a proxy.  Feature interaction is the boogeyman that gets raised.  Feature interaction is the idea that some features may interact differently from expectations because of interactions with other features.  For instance, try setting up Vonage’s simul-ring feature on your home phoneline so that it rings your cell phone as well as your landline, and then phone your house from your cellphone.  You can’t call home any more because your cell phone voice mail picks up every time. If you’re interested in more information on feature interaction, check out the ICFI ‘05 web site.  ICFI is a somewhat regular conference on feature interaction which has been running since the mid 1990’s. 

Feature interaction is the concern raised, but isn’t this really just a case of a poorly designed feature?  To use the Vonage example again, shouldn’t Vonage’s proxy have the brains to be able to figure out that it’s one of the simul-ring phones initiating the call?  I suspect the real issues are security and billing issues.  For instance, have billing systems been designed to handle the case of what happens when the call goes off network, and potentially returns later.