Why is Voicemail still in the Stone Age?

The other day I was thinking a little bit about some of the legacy products that I help to run here and it got me thinking about some of the features I use and how I'm going to replace them when I'm not able to have a free work account. This train of thought lead to thinking about how stagnant the Telephony market has been for the last 20 years. There are pockets of innovation, but no real mass-market adoption of new features.

I expect the following things from my voicemail:

1. Reliable
2. Voicemails retrievable via phone, online and through email.
3. Able to customize how calls are answered, if at all.
4. Able to listen in on the recording of a voice message in real-time while it is being left, and able to take the call if I so choose (message screening).
5. Get online notification via Instant Messenger of incoming calls and be able to dispose of them as I choose.
   

I get all these through AOL Voicemail, (granted #5 was a beta program we stopped), but if I wanted to get this otherwise, I think my only option is an Asterix homebrew solution. Is there seriously no commercial demand for these features? AOL Voicemail is $7.95 a month and get's me all of those things. PhoneValet looks interesting, but I don't have a MAC. CallWave used to have a Voicemail offering, but I cannot find it anymore. eVoice comes close, but even the paid version is only half of the features I'd expect. It'll be interesting to see how this space evolves as VoIP becomes more common place. In the meantime, I suppose I'll be installing Asterix.

Voice Services Help Pages!

After a couple months of sporadic effort, the Voice Services help pages have finally been moved over to the new Drupal voice platform. This has been a significant goal for me and hopefully will facilitate timely updates to these pages based on user need. The old platform made updates to content a little more work than it should have.

I have several goals that I want to push forward now that all the help content is on the Drupal platform.

1. Facilitate user's answering their own questions
2. Educate user's on more advanced usage
3. Provide a cost effective means of customer support

Customer support is expensive, I didn't realize how expensive until I started learning about it. Providing easy to find and comprehensive answers online is, by far, not only the most cost effective solution from a companies perspective but also the quickest for the user (if done right). The cost per user of a web site is minimal whereas e-mail or phone support is quite costly. Online help has a number of other upsides at the same time. To name a few:

• Related articles to further educate users
• Raise consumer awareness of complementary products
• Build search engine rankings

However, I'm only a fan of Online Help done correctly. I find online help sites without an obvious way of escalating your query very frustrating and disappointing. I think we've all, at one time, had to deal with a company that seemed to go above and beyond in trying to avoid being contacted by a user. I've structured our Voice Services help pages so that any time you are looking at a product's help online, there is a "Customer Support" link that will let you communicate with a live person. As we get common questions that are not answered within the online help, we will add them to help our users get the answers they want even faster.

Integrating AIM Call Out into KDE

Today I used Twinkle (an Open Source SIP Softphone) to add the ability to call phone numbers out of my KDE addressbook with AIM Call Out. The versions of software that I'm using:

• Debian Lenny
  
• KDE v3.5.9
• Kontact v1.2.9
• Twinkle v1.2

I followed our instructions to setup Twinkle to get the SIP calls working. Once I had that going, I did the following:

1. Start up Kontact.
2. Under the "Settings" menu, choose "Configure Kontact..."
3. In the left hand component menu, choose "Contacts -> General" settings.
   
4. There should be a section called "Script-Hooks". In the "Phone:" field type "twinkle --call %N".
5. Click "Ok" to save your settings.

Note: The AIM Call Out service expects E.164 compliant phone numbers. We do this because it is an international service and we'd be insane to try to guess what your dialing plan should be. A lot of the phone numbers in my address book are NOT E.164 compliant and I didn't really feel like going through and converting them all so instead I setup a conversion filter in Twinkle:

The expression here will work for 10 digit North American phone numbers. For a complete list of country calling codes, see the Wikipedia entry on the subject.

Mashing VoIP with Flash

I really hope the mashup of VoIP technology with Flash will become a killer application. The awareness among the general public may not grow to something like E-Mail or Text Messages via cell phone, but I predict flash and VoIP will be the required catalyst to solving the biggest technical problem with telephony today: Telephone Numbers.

Telephone Numbers were designed when locality mattered. Area codes, prefixes, etc. all revolve around physical location. In some sense, they've been virtualized in that you can port your number a short distance away and you can route calls for a number over a different number somewhere else. Regardless, at the end of the day a telephone number is generally tied to a physical location. Here's the crux of the problem though; people don't communicate with physical locations. People communicate with people. A cell phone is a partial step in the right direction, but why have a number at all? We have e-mail addresses and Domain Name Resolution (DNS), why in the world do I have to remember or lookup a phone number when I know who it is I want to communicate with? I think the first stage in moving from phone numbers to calling an "Identity" is VoIP and Flash. Bear with me here as I walk through my reasoning:

1. Flash allows the mass market to make phone calls with a click of the mouse. All the other requirements are already met (computer w/ Internet and flash enabled browser).
2. The method of contacting an identity is irrelevant once the interface for that communication is a mouse click. All the end-user knows (or cares) is that you click the person's name and then talk. You might be calling a POTS line, you might be talking over a cell phone or even a computer. To the end-user, the Medium is irrelevant (but not the UI), the content is all that matters.
3. When the Medium is abstracted to irrelevance, it becomes fluid depending on the situation.
   

I've been working in the telephony space for almost 10 years now and I've seen several attempts at or ideas for a ubiquitous communications revolution. The problem is always shoehorning it into the existing telephone network with the inherent concept of telephone numbers tied to locations. A new user interface for phone calls can make a phone number obsolete. Doing away with phone numbers and switching to some form of DNS addressbook will enable a whole new universe of communication possibilities. I look forward to it and I really hope that Flash will be the catalyst for the revolution.

Quality vs Cost

Nothing is free, it's an old saying and I fully believe in it. In the VoIP world, it is more true than ever. VoIP is a revolutionary technology; it has taken the high cost of entry of the POTS world and made it dirt cheap. Unfortunately, the dirt cheap idea seems to be taking over and the quality and availability we used to expect is falling by the wayside.

I have very high expectations of computers and technology. I switched to Debian GNU/Linux in 1995 and since then I've learned how stable things can (and should) be. A reboot should happen to upgrade the very core of the operating system, a hardware upgrade or a power failure. Anytime other than that and something or someone has failed to do their job properly. When I am on the phone, I expect high quality audio and a connection that stays connected. Why the world is starting to accept crap from telco's is beyond me. What happened to the 5 9's? (For those of you keeping track, that is about 5 minutes a year that your phone can be unavailable, whether you are on it or not).

Granted, anyone who has really worked with SIP will acknowledge it has some fatal shortcomings, so perhaps that is contributing to the degrading service quality we're seeing in the telephoney world. I don't think that is the sole cause. Honestly, I put a lot of the blame on the consumer. We don't vote with our dollars enough. For some providers, the cost of monitoring their trunks appears to outweigh the loss of revenue from when links go down. I find it incomprehensible, but with the focus so much on cost, quality gets the shaft.

I hope the pendulum starts to swing the other way, and customers actually get up and leave a telco that is failing to provide a quality connection. The nice thing about capitalism is that, at least in theory, the consumer market can punish those companies that fail to meet expectations. Hopefully the market will weed out the providers that charge $0.001 less per minute for about half the quality and availability. If that happens, those of us trying to provide the service we'd like to use ourselves can continue to do so.

Fun with Web 2.0

I've added an RSS feed I just created to my blog here; Top SIP Devices for AIM Call. This is the 5 most viewed device pages in the last 4 weeks on voice.aol.com. Go "Web 2.0" I suppose.

I clearly remember using Mosaic to browse HTML pages and thinking "Wow, this is so much easier than Gopher". Today, in about 10 minutes, I was able to create a live RSS feed of the the most popular articles in a subset of a web site and import that as a handy list on my blog. Seems like a great improvement over the early days of the World Wide Web doesn't it.

But when you think about it, the technology to do it really doesn't seem all that advanced. It's just a database query, a sort, a truncate and then an import. UNIX has been doing piped commands like that for, what, almost 40 years now? Congratulations Web 2.0, on being able to do things the Unix command line could do 30 years ago. I guess you're much more colourful than Unix was. That's something.

AIM® Phoneline Help Updated

I added some new questions to the AIM® Phoneline FAQ today. We get a surprising amount of requests about AIM Phoneline, which I guess is good. I've also added a "Customer Support" page for the APL product, which I hope will make contacting us about APL easier.

Hopefully these new questions help speed up the support for some folks. Don't forget to call yourself once a month to keep your account active if you don't have anyone else calling you. That seems to be the biggest complaint we get.

Getting Started

Recently, we posted an article on my voice portal that mentions the other bloggers in the office here. It's motivated me into starting a blog of my own, which is something I've been thinking about doing for a few years now...

So I guess this makes for the inaugural post on my inaugural blog. This reminds me a lot of the BBS days, how nostalgic.