For many years now you haven’t been able to look at web services without hearing the term “The Cloud”. Cloud computing (or SaaS, Software as a Service as it’s also known in the business world) is described by Wikipedia as “the use of computing resources (hardware and software) that are delivered as a service over a network (typically the Internet)”. It’s a broad term that describes most Internet-based services but some “household name” consumer cloud services are Spotify and Netflix.


Over the past few years I’ve become known as something of an expert in the field of cloud-based telephony services. It wasn’t something I intended to happen but simply a side effect of my use of the cloud in my telephony projects. At the last count there have been over 100,000,000 voice minutes routed through my various cloud telephony services.

But the users of those services didn’t know they were using a cloud service, because I used the cloud for my own advantage to enable me to build affordable infrastructure. What I want to write about today is what the concept of The Cloud can mean for the end-user and why services that utilise cloud computing for their infrastructure could be so much better if they thought about the relationship between their infrastructure cloud and their users: “user-centric cloud computing” if you will.

The cloud and their users! Too many services concentrate on the relationship between the service and the end-user’s device, be it a PC, Mac, phone, tablet or TV. While it’s important to distinguish between the device it’s also important to remember that the ultimate endpoint is the user themselves. Netflix gets this; there are four different types of device from which I regularly watch Netflix and the experience is customised for each device type (large screen interface for the TV, small screen interface for my phone and somewhere in between for tablets and PCs) but no matter what device I’m using the content is the same for ME the end user. I can start watching a film using my phone on the train, pause it when I arrive at my station and continue watching on my living room TV from my PS3. This is a user-centric cloud service. If I had a criticism of this side of Netflix it’s that they are aiming at the living room but don’t take into account that living rooms have multiple users many of whom are families who don’t want a separate Netflix account for each household member and even if they did it’s not simple to switch between accounts on most devices.


Now what about Spotify? In my opinion Spotify completely drops the ball when it comes to being a user-centric cloud service - in fact it’s the perfect example of a device-centric cloud service that neglects the relationship between the service and the end-user. For example, yesterday I drove to London for dinner, on the way home I played the album “She Wants Revenge” by the band of the same name. When I arrived home I had a bit of work to do so I sat down at my desk and loaded up the Spotify desktop client which welcomes me with the “What’s new” page. I wanted to continue listening to the album from where I left off in the car but this isn’t an option, even my History page only shows me music I’ve played through the desktop client and has no knowledge of what I’ve been doing outside of this client except for syncing playlists I’ve created. The two clients are completely separate experiences, Spotify uses the cloud to deliver music to me but they are by no means a user-centric cloud service.

Now think about this from a telecommunications point of view and you can see what I was thinking when I showed off my “one phone number, multiple devices” demo at Mobile Monday last year. Impossible Telecom was designed from the outset to be a user-centric cloud mobile operator; your phone was simply a conduit to the service. You can use the Impossible website to have an SMS conversation with a friend, when you leave your desk you can continue that conversation on your phone (yes, iMessage does this but it’s not tied to your operator although some would say this is a good thing) the same with phone calls (you can achieve some amazing things with WebRTC these days!).

The telephone is arguably the oldest cloud service in existence but it’s never been a user-centric cloud service, the world is lacking a user-centric mobile operator like Impossible Telecom and it makes me sad that it’s been put on hold for the time-being… but at least I know it’s there, waiting for when the world is ready. 

I noticed that Spotify was sending 3.2MB/s of data while I wasn’t even listening to any music. This is because Spotify uses clients as nodes for distributing music between users so they can reduce the load on their server.

This is a good idea but I pay for Spotify Premium so I feel I shouldn’t have to use my bandwidth to reduce the load on their infrastructure - this is especially bad if you don’t have an unlimited bandwidth account or even worse you are using a mobile dongle where data overage is expensive or are sharing bandwidth with other people such as in an office.

Thankfully I use Little Snitch on my Mac (if you have a Mac then you should use Little Snitch too!) so it was trivial to see what was using the bandwidth and block it by removing the existing Spotify rules and adding the following three rules:

Restarted Spotify to kill and open connections and my upload dropped to 9KB/s… Problem solved :)

I can’t believe it’s been nearly a whole year since I last blogged anything meaningful about my plans to start a mobile network!

I had planned to be transparent about the process and have written several blog posts since but haven’t been able to publish them due to NDAs and not wanting to “rock the boat” with existing and potential suppliers, many of whom were being very cooperative in helping us get off the ground.

So having not published much it’s quite disheartening that what I can publish is the post talking about how it’s all gone terribly terribly wrong!

So where was I a year ago? Looking at how little I have now it’s hard to believe that I had almost everything in place to start a mobile network! All the suppliers had lined up and delivered much of what was needed and pending a few minor but mission critical tweaks we were looking good for a late summer launch.

But these minor issues we needed resolving became major issues for other people and their lack of progress left me looking at ditching our existing partner and seeking alternative routes to market (which is to say another host network to launch on). I spoke to several networks as well as MVNEs who work with networks but reading between the lines the message was this: innovation isn’t welcome here.

If I wanted to I could launch a new MVNO in less than a month but it would be nothing more than an “airtime reseller” (one network actually uses this term to describe MVNOs).

My apologies for being so vague but I would still like to launch this one day - it will always be something I chase after but for the time being the industry isn’t ready to support it and I no longer have the funds required to make it happen.

My closest friend told me over Christmas that she thinks I’ve stopped caring about the things I used to care about and I think she might be right. I’ve been working for my own companies for the better part of the last decade and while there have been some successes as well as failures the past six months have left me emotionally and financially drained and, if I’m being totally honest, I’m tired of hearing bad news* - I think it’s time I took a job somewhere innovating for other people.

In short, I think it’s time to look through my book of contacts and find some work. I’ve built a ton of technology for Impossible Telecom so I may spin some of that off as stand-alone products but I think it would be as more of a small lifestyle business than trying to build the next Vodafone.

* - I wrote down a quote from a video interview with Marc Andreessen about doing startups which ends with: “…you’re gonna feel alone and isolated and you’re gonna get lots of bad news”; he’s right, but remember the saying: “if your dreams don’t scare you, they aren’t big enough”.

Back in January I found myself working on my tax return having put all my documents in storage for the summer while I was in SF. This meant things weren’t as neat as they usually were and I found myself revamping my document organisation & storage methods.

I was originally against scanning everything and wanted some fancy boxes with files to store paperwork but a few different people convinced me to look at Fujitsu’s ScanSnap document scanners. I settled on one of their cheaper units, the ScanSnap S1300 which was around £250 from Amazon.

It took about 15 minutes to get everything up and running including setting up the software after unpacking the scanner and plugging it into my iMac (it rather handily includes a USB power cable as well as the included mains adaptor so I’m powering it entirely from my computer, although this does use up 2 USB ports). The included ScanSnap software will scan pages, perform OCR and then spits out a fully searchable PDF file. You can also set an application to scan into as well as saving files so I have my scanner set up for one-button scanning to both Evernote (as the Scan To application) and Dropbox (by telling the ScanSnap software to also save it to a Dropbox folder).

This means when I receive a letter (or bill, statement etc, I tend to scan everything except junk mail these days - just in case!) I just pop it into the scanner, press the big blue button that lights up and in a matter of seconds I have a copy on my desktop, pushed to the cloud on both Evernote and Dropbox. The next time my iMac performs a Time Machine backup a copy of the document will be stored on my Time Capsule and of course Dropbox will sync the file with my laptop as well. The original hard copy can now be filed or shredded depending on it’s importance. When I need to find a document I can search for it in Evernote or on my Mac using Spotlight (remember, the ScanSnap spits out fully searchable PDFs and I’ve found the OCR to be very reliable). With the mobile clients for Evernote I have access to all my documents while I’m on the move!

I appreciate that some people may not be comfortable uploading their documents to services such as Dropbox or Evernote but I love using them both so for me this is absolute document storage bliss!

You can pick up a Fujitsu ScanSnap S1300 using my affiliate link here or check out the rest of the ScanSnap range.

More details on scanning to Evernote using the ScanSnap software can be found on the Evernote Blog here.

I get very annoyed at unhelpful customer service or products that don’t work as advertised so I thought it would be nice to write about some positive consumer experiences. Hence “Friday Favourite” in which I will attempt to say positive things about a product I own or a service I use.

First up is the Jawbone Jambox, a bluetooth speaker from the people that make the highly regarded Jawbone bluetooth headset.

I picked this up when I was in San Francisco because I didn’t have any speakers in my little studio apartment. I originally intended to get the “classic” B&W Zeppelin but after trying it in the Apple Store it was clear that it would have been overkill for my studio. Jawbone had just released the Jambox so they had a demo display in the store and while the audio quality didn’t come close to the B&W Zeppelin it was definitely good enough to fill my apartment with sound, it’s wireless, portable and at $199 it’s a fraction of the cost of the Zeppelin (you can pick up the Jambox for around £130 in the UK).

The device is a small black (other colours available!) rectangle with a rubberised top and bottom, the middle part is a metal grille which wraps around the unit. There are stereo speakers on the front (not that you’re going to enjoy much stereo separation here) and bass is ported quite effectively to the rear meaning that if you want a bit of extra bass you can place it facing away from a wall (or even better, a corner but this can make the sound a bit boomy).

Something I hadn’t anticipated is that this device has almost exclusively lived in my bathroom where it pumps out tunes while I’m in the shower. With the volume settings on my iPhone set to full and the Jambox to half it’s still loud enough to be heard comfortably over the sound of the extractor fan and shower.

Another good use is to drag it down to the kitchen to provide background music while cooking (or more likely in my case unpacking takeout). The device also features a microphone and serves as a very competent speakerphone / conference unit - obviously calling on Jawbone’s experience in building bluetooth headsets.

The marketing material shows hipsters using it as the sound system for a party which might be a bit ambitious but for every task I’ve thrown at the Jambox it’s performed wonderfully.

If I had one criticism it’s that the audible battery low notification doesn’t always give you much time before the battery dies but with an indicator on the phone’s screen it’s hardly the end of the world.

I highly recommend the Jawbone Jambox for anyone looking for portable speakers, even though it’s at the top end of the price range for this kind of device.

You can buy your very own Jambox from Amazon using my referral link here

At last week’s Mobile Monday London event I gave a brief into and demonstrated one of the features offered by Impossible Mobile. Thanks to BennyCrime who filmed it and heckled via the medium of paper.

I’m no Steve Jobs but I think I made my point. 

(The sounds is a bit low as it’s filmed on an iPhone from the audience)

The guys over at have an amazing set of photos from a trip to the top of The Shard in London Bridge. At the bottom is an epic ultra-high resolution “gigapixel” panoramic shot of the city that you can scroll around inside a flash applet.

On my desk I have 2x 27” Apple Thunderbolt displays plugged in to a 27” iMac and a 30” Apple Cinema display plugged in to a MacBook Air. That’s a total resolution of 11680 pixels wide by 1600 tall (at it’s highest - on the 30” cinema display). You might have seen me rant in the past that it’s hard to find photos with a high enough resolution to have a single desktop across all screens.

This photo of London (my city… even though I live in Winchester) was exactly what I was looking for. Unfortunately it was trapped inside this silly flash applet.

After some investigation it turns out the applet was loading tiles and stitching the visible images together (a bit like Google Maps). At the most detailed resolution the total image was made of of 1,776 individual 685 × 685 JPG tiles (the edge tiles were a bit smaller) to make a massive 50690 × 15949 image.

I spent about an hour looking at the traffic generated by the flash app and writing a bit of code to drag down all the tiles from the silentuk server and merge them into a single image. I left it to merge last night and found my (quite high powered) iMac unresponsive this afternoon when I returned to it.

After about 10 minutes of optimising the merging code I re-ran it and after about an hour it spat out what is quite possibly the most incredible 126MB photograph I’ve ever seen. Which then crashed Photoshop when I tried to open it (Preview displayed it fine) - a few more lines of code to shrink it down to a photoshop friendly (but still larger than my native desktop) size and I was in business!

Sure I could have asked the guys at SilentUK for a copy of the photo but I’m impatient and I like a challenge (I’d still LOVE to get my hands on the original RAW files that made this).

I’m not going to release my stitched copy or the code required to stitch it because it’s not my photo to share and I believe they plan to monetize the photo by making prints available. It’s not exactly hard to do if you’re determined enough (plus it’s fun!).

Here is what the finished wallpaper looks like on my desk:

And here is what the final photo looks like with the areas I used for wallpaper cut-out - as you can see it’s still only a small part of the original photo:

Update: The guys at SilentUK have been in touch because they didn’t want people stealing the image and have subsequently removed the panoramic view while they figure out how to protect it. Sorry… that’s me, I come along and ruin everyone’s fun :(

Just a quick note to say that I’ll be giving a brief overview and demo of the Impossible Mobile service at the Mobile Monday London Demo Night this coming Monday (2nd of April). This will be my first public demo so it promises to be all kinds of awkward :D

Registration is free but spaces are limited, find out more and register here.

I will of course be around for the post-event drinks, come and say hello :)

People who know me from my “career” as an aggressively vocal blogger and mobile industry commentator may remember that a few years ago I heavily criticised the UK MVNO GiffGaff regarding the way they handled a short outage and some customer service incidents (read it here if you haven’t seen it).

GiffGaff experienced another outage on Friday that knocked out voice and SMS traffic for the better part of an afternoon. I’m very pleased to say that the way they handled themselves both during and after the incident has been absolutely impeccable. At the time of the incident I saw tweets explaining that they were looking into the problem and today I received an SMS and email apologising for the outage and directing me to a series of blog posts explaining what went wrong. As well as a plain English explanation and apology from MD Mike Fairman (here) there was a technical explanation from CTO Nigel Sudell which went far beyond any reasonable expectation of disclosure as you can see here.

One of the issues with being an both an outspoken critic and an entrepreneur in the same industry is that I must hold myself to the same standards I hold others to - and expect my own critics to hold me to those same standards. When I announced that I would be launching my own MVNO my comments regarding GiffGaff began to occupy a part of my mind; through the service I’m building with Impossible Telecom I would be able to put into practice the sort of customer-centric disaster handling I, as a consumer, wanted see and I believed I would show GiffGaff and other service providers how it should be done.

With the way the GiffGaff team handled this recent outage I can honestly say that the bar has been significantly raised and my goal is now to match the level of transparency and openness displayed by Mike and Nigel. I believe this is the biggest compliment I’m able to give them.

After my original blog post I met up with Mike for some beers and a chat about GiffGaff and the mobile industry in general. I’d like to extend an invitation to Mike and Nigel for another few rounds of beers… This time the tab is on me :)

There, I said it… it’s out in the open. I’m launching a mobile network.

The 31st of January was my last day at Bababoo Inc where I worked since 2010 when they acquired Pibbix, the last startup I founded. It was a great journey during which I built a global cloud-telephony platform that I was very proud of and I ended up moving from London to San Francisco.

Since returning to the UK late last year I’ve been thinking about what to do next. Like many “entrepreneurs” I have lots of ideas for new companies and I keep a list of them with notes in case I ever find myself with the time and capital to launch them.

I’m not even sure when I added “Launch a mobile network” to the list, it’s been on my mind for as long as I can remember and I played around with the idea as a lab project back when I was working on Pibbix. I think it’s a desire partly born from wanting to use the kind of mobile network I knew I could build and wanting to make the jump from b2b services (where all my previous startups have concentrated) to consumer.

At first I dismissed it as an impossible task and concentrated on building a mobile app studio… but with the mobile network idea lingering in the back of my mind I started to prototype a few things, I made a few calls to some contacts and before I knew it I had a couple of SIMs in my hand and calls flowing between them through my own telephony switches. A few more jigsaw pieces came together and it started to look feasible that I could actually launch my very own mobile network.

Obviously this will be an MVNO model (a Mobile Virtual Network Operator - ie: we’ll be using a host network’s towers and spectrum). This is the bit where most industry insiders start to shake their heads; most MVNOs don’t fare well unless they are attached to a big existing brand such as a supermarket or have a large marketing budget aimed at a niche or ethnic markets. Almost every MVNO competes on price - their USP is that they are cheaper than the other guys… there are a lot of other guys and it’s getting very cheap!

I believe there is room for another kind of MVNO though, one that doesn’t try to offer the absolute rock-bottom pricing and instead offers reasonable pricing with a USP of features & quality; useful features that no other provider offers (which I will go into in more detail and even demo in future blog posts). An MVNO where the person on the end of the customer service line understands the requirements of the customer and has the power to do what’s necessary to resolve any issues (I don’t know about you but I’m fed up with the awful “customer service” offered by mobile networks).

I’m still working on this essentially by myself so it’s going to be a lot of hard work and there are a number of technical, commercial, bureaucratic and regulatory hurdles to overcome but given the progress so far I’m very confident that brilliant things are going to come out of this. 

So that’s the story of how I ended up as the Founder, CEO and Tea Boy of Impossible Telecom Ltd - the company behind the UK’s newest & most kick-ass mobile network.

I plan to be quite transparent about the whole process so over the coming months I’ll be using this blog to demo features, rant about issues and document the process of getting from now to being launched and generating revenue as transparently as I can be while still respecting various NDAs and commercially sensitive information.