As some of the Twitter folken are aware, I’m in the process of finalizing details for my forthcoming transition back to the East Coast for my usual summer of Sox. At some point between May 1 and, say, May 15th, I’ll be pulling up stakes and migrating from Denver to the family place in Georgetown, ME.

Unlike last year, however, the plan this summer is to rent my loft here in town out rather than letting it lie fallow. This introduces the typical moving complications that I’m used to dealing with, but one that I’m not. One that involves telephony. One that I’d love suggestions on.

Herewith a description of the problem, past solutions, and so on.

The Toll-Free

A few years back, we acquired the 866-RED-MONK vanity toll-free number, which has proven reasonably popular. At least among the non-PR types that don’t default to my cell phone given any opportunity. The way it works currently is this: AT&T holds the toll-free, and redirects to a number of our choice. It used to be our number in Maine when I was based there, but these days it points to the landline here at the home office. This works fine when I’m here, and less fine when I’m not.

The Problem

When I’m not here, obviously I’d prefer to not have the toll free point to a line here. Initially, I thought Grand Central would prove to be the perfect solution, and tried to point our toll-free at that. AT&T disagreed, and declined to point our toll-free at a number that didn’t have a physical address associated with it. Ultimately, my “brilliant solution” was forwarding the landline to Grand Central, so someone calling the toll free would get bounced as follows:

Toll-Free==>Landline==>Grand Central==>Grand Central #s

Fun stuff. With renters poised to occupy the space here, however, I’m not going to be able to pull that off this summer, as I need to disconnect the line. Ergo, I need a more elegant, not to mention permanent, solution.

Here are the things I’m thinking of.

The Solutions?

1. Move the Landline to VOIP, Point the Toll Free at That:
   This has the same problem as Grand Central – the lack of a physical address – though I could probably fake this better simply by billing it to the office here in Denver. Still, it’s unclear if I could sneak this by AT&T.
2. Move the Toll Free to Someone Other Than AT&T:
   I’d be happy to do this, but know very little about toll-free providers. Particularly whether any of them are VOIP friendly or not.
3. Lose the Toll Free:
   It’s not essential to our business, but I’d prefer not to lose it if that can be avoided.
4. Get a Landline in Maine, Point the Toll Free at That:
   This is problematic, due to the fact that getting additional lines at the family place is complicated (we’re on an island), and that I don’t typically rent an office for the entire duration of my stay.
5. Point the Landline at Another RedMonk Employee:
   We could do this, of course, but given that a great many of the people that dial the toll-free are looking for me specifically, this would be less than convenient.

As you can see, none of my options are looking great. Those of you with more VOIP expertise than I possess are actively encouraged to suggest anything that might work.

The purpose of the GrandCentral service was summed up in succinctly by David Pogue’s piece, “One Number That Will Ring All Your Phones.” That is, more or less, what GrandCentral does. You sign up, receive a new phone number, and tell GrandCentral about your existing lines: cell, office, home, Gizmo, whatever.

From that point forward, whenever someone calls your GrandCentral number, all of the phones you’ve configured will ring simultaneously. In my case, as an example, instead of trying my cell phone which generally doesn’t ring when I’m at home (a slab concrete infrastructure will do that), you can point yourself at my GrandCentral number which will ring my cell phone and my office line simultaneously. Simple, but quite useful. If you have more than one phone line, anyway. If you only have one number, there are still some useful perks but the value is diminished.

Following Raven’s recommendation, I’ve been testing the service out for a couple of weeks and thought I’d do a quick Q&A in the event that any of you are interested – and several of you have already asked about it.

Q: Any disclaimers?
A: Not that I’m aware of. I’m unaware (but curious) of what GrandCentral’s underlying infrastructure is composed of, but I don’t see anything to disclose.

Q: Let’s start with the question on everyone’s mind – is the service free?
A: Currently, yes. And even in future, GrandCentral is promising to have a free version. This is what their FAQ says on the subject:

Yes, we’re excited to say that we will always offer a free version of GrandCentral, even after beta. Our free version will include unlimited inbound minutes, unlimited voicemail (up to 30 days old), and access to all of our core features.

During beta, we’re giving everyone unlimited access to our premium features. In exchange, all we ask is that you send us your feedback (good or bad) to beta@grandcentral.com. We’ll read every comment.

Whether or not that’s a good idea is open to question.

Q: What do you mean?
A: Well, telephony’s a technology burdened with vastly different expectations in terms of uptime. So much so that “dial-tone” levels of service are a common euphemism for serious uptime – whether you agree with it or not. Contrast this with, say, email, which most of us could afford to lose for a while (or in my case, permanently). Given this expectation, GrandCentral is likely to be judged in large part by how available the service is; thus far, I haven’t experienced any outages, but two folks I know have. One of them contacted customer service in a live chat session during one of the outages, and was told “it’s a beta.” While true, it’s probably not the best way of delivering that message. Fortunately, they eventually expect to roll out for pay, premium services. The question will become how continuing to make available a free service impacts – or doesn’t – GrandCentral’s ability to keep everything running smoothly. I don’t know the answer to that yet, but I’m willing to give it the benefit of the doubt in the meantime.

Q: Is the service at all related to the GrandCentral web service-ish firm?
A: Tim O’Reilly seems to indicate that the answer is no, saying “I was on the board of the company formerly known as Grand Central, which is connected to this one via its name and its primary investor, Halsey Minor.”

Q: Apart from the whole one number thing, what does GrandCentral do for you?
A: As with many next generation telephony services, it’s got a lot of bells and whistles. Take voicemail, as an example. Similar to CallWave, which I still employ on my cell phone, it does all of the basics. Records voicemails digitally, makes them available on the web, notifies you that you have them and so on. But it also lets you record custom greetings by entry in your address book – friends might hear an informal greeting, while business contacts get something more professional. Speaking of professionalism, all the work folks in the audience are missing out on one of the more gratuitous features: ringshare. This lets you change the usual ringing sound heard while dialing with music of your choice (note to Linux users, for some reason the Flash interface to upload tracks doesn’t work – try a Windows machine – and yes, they’re aware of the problem). So while my friends and family might hear M. Ward or Pearl Jam while they’re waiting for GrandCentral to try and find me, you folks miss out, because that wouldn’t be “professional”

Q: Apart from the voicemail and messaging, does GrandCentral have more sophisticated features, say rules based call direction?
A: Surprisingly, the calling rules portion of the application is fairly primitive. You can take your existing numbers and give them the ability to ring certain numbers only, but that’s about it. Apart from the option to not ring your Home number during business hours, there’s very little time based processing you can do, such as letting certain select callers ring through at any hour, but directing late night calls to voicemail. They do, however, offer some sort of blacklist/rules based telemarketing filtering, but I have little experience with this.

Q: How about call screening?
A: Yup, it does that. You can turn it off completely, screen only blocked callers, or screen everyone. Right now I’ve got it screening everyone, meaning that when you call me I get a message saying, “Call from YOUR NAME (often hilariously mispronounced – it especially has trouble with my friend Marguerite’s name), press 1 to accept, 2 to send to voicemail and so on.” I’ll probably just screen blocked callers after I’m finished testing the service.

Q: What does the service mean, realistically, for the average person?
A: Well, if the discussion at our last Denver Tech Meetup is any indication, it may mean nothing to many of you. But I already have a couple of friends who are in and out of their respective offices that are thinking about using it so that myself and their other friends don’t have to keep calling multiple numbers to track them down. It should be of significant value also to folks like myself that tend to be fairly mobile workers; consultants, analysts and the like. If I’d had this last summer, as an example, no one would have had to bother trying to remember my summer office number – I could just have GrandCentral patch you through seamlessly. The short answer is that many of you won’t care about this at all, but those of you that do will find it very interesting, I think.

Q: What about privacy issues?
A: Tim raised this issue when he talked about the service, and it’s a very legitimate issue. As more of our information gets sucked up into servers that we do not control directly, the risks for intentional or even unintentional mischief go up. At least in theory. I’m fairly sanguine about the risks, however, because in this case I’m not outsourcing anything that couldn’t be discovered if, say, I lost my cellphone. Your mileage may vary, however, as always.

To their credit, GrandCentral’s got an FAQ just around privacy and security, and their privacy policy looked decent after a quick once over.

Q: What’s the catch?
A: The catch is very simple: it’s (yet) another phone number. If you’re like me, you’ll have to drag a certain percentage of your contacts kicking and screaming to learn a new number (in a couple of cases, I just changed the number on their cellphones). And until everyone cuts over to the new number – assuming you can be successful in transitioning everyone – you’ll have voicemail in two or more places: the GrandCentral voicemail and the voicemail of your land or cell lines.

Oh, and they don’t do international for the folks in the audience who are reading this from abroad. I think there’s a yet in there, however, as it sounds like they have plans to launch that service soon.

Q: Once you’ve finished testing the service, do you intend to continue using it?
A: Barring something catastrophic, I do. I’ll watch the transition from free beta service to premium offering very carefully, but it would appear they’re doing what they need to do there. It’s been very convenient for me on multiple occasions, and even some of my luddite friends have had positive things to say about it. It has its occasional glitches – one call this morning rang the cellphone 30 seconds ahead of the office line, which went to voicemail after I picked it up mid-ring – but the reliability thus far has been within acceptable tolerances. It’s likely, in fact, that within the next few weeks I’ll be repointing our toll free line – 866.RED.MONK – over to the GrandCentral service rather than my office landline. This should help you all find me easily even after I’ve completed my traditional summer migration back to my pseudo-home state, Maine.

Q: What is your GrandCentral number? Can anyone use it?
A: Sure, knock yourself out. My number is 617.395.5685 (chose a Boston area code because that’s what my cell area code is, and that’s the number I’ve had longest). And if you’re bored, just try the form below. You plug your number in, click the button and it will ring your phone and connect you to me – automagically.

X-Lite Screenshot

Originally uploaded by sogrady.

For those of you that are interested, there’ve been a few new developments in my personal Asterisk installation project. Couple of setbacks as well, but at least it’s moving forward. A quick summary:

1. TrixBox Installation:
   Complete. Using the aforementioned and linked NerdVittles howto, I’ve not only completed an installation of TrixBox and FreePBX, I’ve set up an account (or “trunk”, in Asterisk-speak) for my new provider, TelaSIP.

2. New VOIP Provider:
   While I considered integrating my existing Packet8 VOIP service into my Asterisk installation via hardware, I’ve continued to experience the very frustrating call drops where I can hear the caller but they can’t hear me. As discussed before, I gave Packet8 a chance to fix this and they were very responsive, but given that the problem was not solved I’m not inclined to give them a second chance.

   With Packet8 on the way out, I needed a new provider, and for the time being at least I’m going w/ TelaSip. I’ve heard good and bad things about them – often in the same conversations – but ultimately I’m choosing them because they seem to be very responsive to support questions over email. The fact that NerdVittles included them as a default in their howto didn’t hurt their chances either. We’ll see how they perform – I’m already having an issue (described below), but for now they’re the choice. The account is also about $10 per month cheaper than Packet8′s plan.

3. Number Transfer:
   TelaSip issued me a new number at the time of account creation, but ultimately I’d like to preserve my old Packet8 number. The good news is that the number is indeed portable, so I should be bringing it with me to TelaSip. The bad news is that the porting process – according to TelaSip’s form – is expected to take a minimum of 30 days. Contrast that with the porting process for cell phones, which when I switched from Verizon to Cingular took about 45 minutes. Given how little I use this phone, however – most of my friends and family still use the cell phone despite the fact that it doesn’t work in my apartment – this isn’t a dealbreaker.

4. Softphone Choice:
   Ultimately, I plan on connecting my Asterisk info back into a standard phone, probably by connecting my existing Uniden cordless to the system using something like this rather than a new digital ready phone. But for now, I’m using what’s called a softphone to connect to Asterisk. Think Skype. Unfortunately, neither Skype nor Gizmo can apparently be used as softphones to connect to my Asterisk instance, although Gizmo can call it because it speaks SIP. That would have been ideal, because then I could just run that and have both business and personal connectivity. But they don’t, from what I’ve been able to determine.

   I then considered Ekiga, the softphone shipping in Ubuntu by default, but it’s agonizingly slow on my machine – possibly because of its Evolution Data Server dependency (my GAIM instance was affected – I think – by something similar). Then I considered Tapioca, but I can’t find Edgy builds for the package and the Dapper builds conflict with some of my installed libraries. So for the time being, I’ve default over to the free but not open source X-Lite. I can’t say that I love the interface – the menus in paticular need a lot of UI love – but it works more or less adequately so far. Definitely open to suggestions here if any of you have them.

5. What Works:
   TrixBox seems to be operating just fine. All of the typical Asterisk stuff works just fine – I can dial in and check my voicemail, I can create wakeup calls, I can dial 611 and get up-to-date airport weather information read to me in a slightly creepy robot voice, and so on. Pretty neat.

6. What Sort of Works:
   I was about to write this up as not working, but as it turns out send voicemail to an email address functionality works just fine – Gmail just dropped it in my spam folder presumably because it came from a non-standard domain (determined from an nslookup of my Comcast IP). The catch? VLC can’t decode the WAV file, so there’s something wrong with the creation part. That should be solvable.

7. What Doesn’t Work:
   Inbound and outbound calls. Yeah, the two things you expect a phone to be able to do. The outbound problem may be a service related one rather than something I’ve misconfigured, because I’m getting the standard telephone error message of “All circuits are busy, please try your call again” which indicates to me that I’m at least connecting to the regular phone network at some point. I have a request into TelaSip to see what they make of it.

   The inbound issue is more problematic, because I suspect the issue there is something that I’ve done wrong. Calls to the number TelaSip has issued me are picked off by their voicemail system, and seemingly never arrive at my Asterisk instance. Not quite sure where I start to fix this, but I’m sure Google will turn something up.

And there you have it – my current progress with TrixBox, Asterisk and VOIP. One of the things I plan on looking into as soon as I can get the basic inbound/outbound calling problems solved are the follow me features. Being able to have friends or family routed to specific numbers – say work during the day or cell while I’m travelling – would be quite nice. It’s easy to do with internal extensions, I’m not quite sure how to do it with external numbers yet. But I’m sure someone’s figured it out.

Any questions, suggestions, or corrections, drop a comment.

Trixbox Screenshot

Originally uploaded by sogrady.

After a friend of mine dropped off a used Dell Dimension 4500 that his employer had condemned to the recycling heap this morning, I finally had the dedicated VOIP box I was looking for and set to work getting Trixbox set up on it. Trixbox [1], for those of you who aren’t open source telephony geeks, is a project designed to put a nice face on Asterisk – one of the more important open source projects going, IMO – and make it easier to use. In between running a couple of errands today, I shepherded it through the installation process which failed at first when using the i586 option, failed a second time for an undetermined reason, but third time was the charm and everything finally laid down relatively neatly using the default 386 option.

After the simple, CentOS based installation process, I would not have been successful with even the most basic configuration were it not for Nerd Vittles truly superlative documentation. This was precisely what I needed to take the next step, and its documentation like this that allows the only moderately geeky – such as yours truly – to get things up and running quickly.

The immediate plan is to get this box up and running to cope with my home phone line – currently serviced by Packet8. Unfortunately, according to the Packet8 forums, my current provider does not place nicely with Asterisk. A decision, therefore, needs to be made as to a.) continue w/ Packet8 and secure a hardware device (FXO card) to integrate the VOIP line into my Asterisk infrastructure, or b.) go with a more SIP friendly provider like Axvoice or TelaSIP – both of which come recommended by the Nerd Vittles crew. I’ve got pending questions open in a few forums to help me figure that one out, but if any of you have experience and/or recommendations here they’d be much appreciated.

Once I connect a VOIP provider to my new installation, I’ll be able to route my voicemail to email (though Packet8 does do this now), schedule wakeup calls, get weather reports by phone, set up proximity calling (routing calls automatically based on whether or not my Bluetooth cellphone is detected), and more. There are also automated podcasting solutions, hold music functionality, custom inbound calling rules (which would allow me, in theory, to do away with one line), and so on. Boring, Asterisk is not.

Longer term I’d like to route the RedMonk toll free number (866.RED.MONK) in here, then setup custom forwarding rules to send you to whichever of the three of us you’re looking for. Little ways to go before I get there, however.

In the meantime, let me express my sincere appreciation for the work of the Trixbox team, their forum participants, Mark Spencer and the folk from Digium, the gang over at Nerd Vittles and everyone else who’s chimed in with advice. Will keep you posted on future progress.

[1] Formerly called Asterisk@Home, and recently purchased by Fonality.

The question to me is not whether or not Skype is worth $3 billion – yes, that’s billions with a B – dollars, but rather where they go from here. While some analyses have focused on the revenue generating potential of Skype – or lackthereof – from where I’m sitting, the real value of Skype to potential acquirees may not be what it can produce, but what it can take away from, say, the established landline incumbents. Or, on the other side of that coin, the quasi-telecom status it could immediately grant a non-telecom player with grand ambitions.

The underlying assumptions fueling both the valuation and the assessments of Skype’s potential are, as near as I can determine, a simple reflection of the wild growth and popularity the application has experienced. It’s clearly come a long way from the bleeding edge early adopter days, and the assumption generally seems to be that further growth, while not guaranteed, is at least highly likely. Such is the benefit of achieving critical mass; doubters are more easily cast aside.

But while I wouldn’t put myself in the doubter category just yet, I must admit that I’m growing more skeptical. I’ve heard complaints in the past, of course, about the fact that Skype’s gone the closed route with respect to its transport protocols. But in the absence of a credible alternative, it seemed to be something of a moot point. A little while back, however, I’d heard somebody mention a project called Gizmo, then promptly forgot about it. Reminded of the effort by the excellent Brian Capouch when we spoke last week, I’ve been paying more attention, and keeping a keen eye out for traction. And you know what? I’m seeing faint traces of it, here and there.

For those that are unfamiliar with Gizmo, as I was, it’s more or less a Skype alternative that’s decided to base itself on the open SIP standard rather than go the proprietary protocol route as has Skype. Now from what I’m led to believe, SIP does not enjoy universal popularity amongst the VOIP crowd, but given a choice between that and a proprietary protocol they’ll choose SIP every day of the week and twice on Sunday.

In the absence of credible SIP alternatives, as previously mentioned, this was essentially a non-issue for the runaway train that was/is Skype. The SIP-loving – or at least tolerating – VOIP crowd may have grumbled about it, but there was nothing really credible to offer up as an alternative. Enter Michael Robertson, of MP3.com and Lindows fame, who created SIPphone, which in turn created Gizmo, expressly to compete with Skype. Suddenly Skype has a legitimate competitor (I’d give you a comparison of the two, but Gizmo’s Linux offerings at the moment are Linspire and Debian packages only – c’mon guys, what about the rest of us?); besides, of course, the IM platforms that have studiously been adding voice chat capabilities.

Who cares, you might say? Why should it matter that Skype’s protocol is closed? The sound quality’s great, and it’s always worked for me. Well, maybe you shouldn’t. If you see a future for things like Asterisk, however, I believe that it does matter – it matters a great deal. Because if one believes as I increasingly do that we’re progressing towards a future where voice is just another stream of data, having that data passed around in formats open and agreed upon by many will open doors for you, while proprietary formats will close them. Intrinsically flawed as it might be, email here – not IM – is the system to emulate. The power of standards is demonstrated every time someone on Exchange successfully emails someone on Notes, while the limitations of proprietary approaches are evident whenever an AIM user wants to IM a Yahoo user. Should proprietary voice formats take hold, it could be like the instant messaging world today, where the network predetermines your audience. Would you want a phone that only allowed you to talk to other Verizon customers? I sure wouldn’t.

But it’s not just Asterisk that can leverage open standards, as the recent Google Talk announcement highlights. No, Google Talk doesn’t support Gizmo at the moment, nor does it support Skype. But because Gizmo is SIP based it might – and soon. Technorati’s Niall Kennedy (there’s a good Irish name if ever I’ve seen one says that such an effort is already underway. There are also those who see Google Talk itself as a competitor to Skype. On that front, however, I tend to side with Steve Gillmor who – in his trademark tones (which as an aside, I quite enjoy) – says:

“They [Google] aren’t even competing against Skype, because the reality of Skype’s ubiquity is an asset of the broader platform, not something to be targetted for extinction.”

While Asterisk proliferates amongst the technical elite – the community that put Linux on the map, Google Talk will be there to open doors to the masses. Whether or not Gizmo can capitalize on the opportunity – potentially at the expense of Skype – remains to be seen; I don’t think that call can be made yet. But just as Google opened doors for itself by relying on the preexisting Jabber protocol, so too can Gizmo steal a march on Skype by going the standards route.

Either way, it’s clear to me that VOIP is coming, in a big way (and, by extension, that traditional telecom as we know it is in trouble). It’s equally clear that Skype’s played a large role in getting us to where we are today, if only by popularizing the concept of VOIP amongst the people with the power to make it a reality. The question in my mind, however, is whether the closed over open stance taken by Skype will cost them a similar or greater role in the future to come.