I've been watching the response of Dems on McCains choice all night and I had to laugh. None of them, not a single one looked very happy about Gov. Palin. They blew her off in their typical “no-ones right except them” way but none of them were very pleased. They know it was the perfect choice to combat Barry, but they are trying to act like it was a mistake. Good Times!

Ok, after just reading through the “Flaws in Present Broadband Wireless Access Deployment and Business Models” white paper, below are some comments, confirmations and opinions of my own.

To set the scene, I am no more an authority on the “Communication Industry” other than what my experience dictates and I do not profess to have the correct answers or solutions to the connectivity issues of the world more than the next man (or woman). My background is in internet, telecoms, wireless and VoIP (with a smothering of IPTV on top). I have held senior level management positions with companies like Cable & Wireless and Hughes Network Systems, I have been through the boom, doom and new boom of the internet industry since 1997 and for my sins I own a WISP in southern Spain.

A step back in history:

During the mid to late 1990’s, internet became the must have commodity and any business that was not connected was a business not worth mentioning. I had the opportunity to grow with one of the UK’s (and Europe’s) largest ISP – Internet Network Services Ltd (INS) – and saw the real impact that internet had on companies and their business growth and/or evolution. I have personally experienced two other similar scenarios in my career that I can liken to the internet explosion:

1) What terminal emulation did to the mainframe to LAN convergence
2) What personal computers did to everyone.

INS was successful and one of 30+ other successful ISPs that Cable & Wireless purchased during 1999/2000 across Europe. Telcos know about telephones, they know nothing or very little about internet. A contentious point … maybe, but do you know any Telco that is delivering an adequate internet service, anywhere in the world? I interject my own experience from working at C&W.
In the space of 24 months I witnessed the demise of a 127 year old, pedigree Telco, I saw 31+ ISPs disappear, 1,000 of hard working and innocent people lose their jobs and livelihoods, $billions being wiped of share values and a black hole void in technical advancements for years. All because some Telco manager thought he understood the internet. C&W were not alone in this mistake and the ripple effect flooded into the infrastructure world as well (CISCO, Nortel, just to name a couple).

Then came the ADSL “broadband” phenomenon; suddenly the Telcos had a solution, a way to use their existing copper networks to offer internet (why did they bother spending $billions on unnecessary IP networks?). The marketing departments came back to life, promises were made and the consumer said “I want some of that please”. Just like with PCs the user took it, broke it and demanded more. The ADSL ISPs (who are of course the Telco’s that broke the internet in the first place) have only themselves to blame … they promised the world and the world said “ok, you deliver the bandwidth and we will deliver the content” … unfortunately, nobody bothered to discuss that little arrangement with the network planning department.

We live in a technical state of mind (since the inception of the PC) of “bigger, better, faster, more!” where we must have the latest gadget, the latest version, be part of the latest social networking group and we want it all “NOW!” … suddenly our lives depend on speed and connectivity.

Content is doing to the internet what software did to PCs … every new version needs and upgrade. Unfortunately, internet or telecommunication networks cannot be upgraded like a simple memory upgrade on a PC … what is in the ground, in the exchanges (locally and nationally) has a limitation and a big price tag to change (not good for the ROI in the submitted business plans).

ADSL was designed and delivered as a faster way to surf, a quicker way to download your emails and the occasional document. No ADSL provider foresaw Skype, Facebook, IPTV or any number of other “user” driven and bandwidth demanding applications. No ADSL business plan could have forecasted the changes in the user’s usage profiles … all believing that they could maintain high contention ratios and bursty traffic and that the customer would be forever satisfied. Oooops!
Whilst working for Hughes Network Systems Europe in 2003/2004, I suddenly realized that over 50% of the world is unconnected and unconnectable – I don’t just mean the far regions of outer Mongolia, I mean western and eastern Europe, UK, USA, Asia, Africa and everywhere in between – and that cable based networks have their limitations. I spent a year travelling around EMEA, meeting and negotiating with the senior management of incumbent Telcos and tier 2 alternatives, formulating alliances and channel relationships with them to add VSAT technology to their broadband product portfolios. It was obvious to me (and my colleagues at HNSE) that ADSL was not going to connect everyone and that a “wireless” solution was needed. Unfortunately, the majority of Telco management that I met did not share the same opinion … they (at the time) believed 3G was the second coming of the internet. Many $billion license purchases later, 3G is still not here and many millions of VSATs are still connecting people where ADSL falls short.

Now, I am not suggesting for a minute that VSAT is the future of the internet … cost and bandwidth restrictions aside eliminates its claim … though it got my mind thinking that cable was not the only medium to connect someone. Unbeknown to me there was a wireless revolution happening outside.

A small jump forward in time (2004/5) and I embarked on one of the most challenging, and personally expensive, endeavours of my career. I setup a wireless internet and VoIP company in southern Spain. First off, I have personal reasons for being in Spain (called Silvia) and secondly, anyone that knows Spain will know that:

1) it has one of the worst telecommunications infrastructures in Europe
2) it has a geographically dispersed population and a terrain that limits the commercial ROI of a wired based network.
3) It is the country of choice for second home purchases – southern Spain leading by a mile.
4) 55,000,000 tourists visit Spain during holiday season – mainly visiting southern Spain
5) The local network infrastructure plan did not (and never will) allow for the millions of new homes that were built to cater for the millions of second home purchasers.
6) VSATs were the only option (at the time) to connect these wealthy, successful, educated foreigners.

To be clear, over half of the premises that have been built in the last 7 years cannot have a telephone line (meaning no ADSL). Telefonica, the incumbent operator, considers the foreign customer “outside of its national remit and obligation” to deliver a telephone service and considers the transient nature of the foreign customer a “bad-debt” risk. Also, and to be fair, why would Telefonica spend €millions to upgrade the infrastructure when the customer is only resident for short periods of the year?

A WISP paradise you might say, a success story you might think? — humm, ok, for those that think you know about wireless – though have never actually gone out and setup a real “in the field”, commercial wireless operation, you really need to re-read the white-paper.

I am sure, with the right amount of investment (we are going through second round now) we could dominate our territory and become a notable name locally … but more than that? Just like cable, wireless networks have their obstacles and drawbacks to overcome.

When we started the company, we could only use 2.4GHz as a technology – 5GHz was still a licensed frequency and the 5GHz manufacturers were still in their infancy. LMDS was an option if you could afford the license fee and equipment costs (e.g. Avarion CPE was about €1000 each, we didn’t bother looking at the network equipment costs).

Like most, if not all WISPs, we started on a shoe-string budget and expected to grow organically until a white knight found us. Unlike most WISPs, we actually came from an internet and networking background – as opposed to the local village idiot who thinks that a WRT on the back of an ADSL network is a “network” (I’ll come on to FON later) – and so we went about securing our backhaul connections and securing tower location relationships. We started with MESH (LocustWorld) but soon gave that up due to the limitations of 2.4GHz, we spent a lot of money buying and testing the “branded” 5GHz technology that was available and then proceeded to build our own equipment, which is still operational now across our network (and outperformed the branded labels).

We decided on a 3 tier network infrastructure deployment and chose bridging instead of routing the core network points. The 3 tier model did two things, it controlled the flow of traffic around each point of our network segments and it brought us closer to our customer (reducing the last mile). All our subscribers connect via 5GHz antennas to a local relay point (last mile to second layer) and then our relay points connect to our backbone wireless network (third layer) situated on tall buildings or mountain tops. We set a minimum wireless connection rate of 6Mbps and a maximum last mile distance of 6km (both under very strict circumstances) – most customers fall within 4Km and have a wireless connection of 18Mbps or more.

Most LMDS network operators, and even some WISPS, try to connect the customer directly to their main tower points and forget the street level. It doesn’t work for various reasons:

1) Wireless signal strength reduces over distance and becomes susceptible to interference (i.e. buildings, trees, waters, other RF signals, etc). You can boast the power (within legal limits) but this also increases the noise level, so you don’t always achieve improvements.

2) The longest link is the weakest link. Meaning the speed of your network will be governed by the slowest of your users. A customer who is a mere few hundred meters from your network will be slowed down by the customer that is many miles away (I have seen LMDS installations that were covering nearly 30Km).

3) Imagine a shouting game, where you have a group of people standing at different distances all shouting to a central point. Each person has to wait its turn to speak and wait for a response. The person furthest away will have the most difficulty in being understood and will have to wait the longest for a response. All other people in the group will also have to wait until the furthest has been heard and responded to before they can talk again. That is basically a quick way to think how a wireless network operates – hence the design of our network and the accurate point made in the white paper about installing multiple points to reduce the distance between the user and the network.

4) Sector antennas, and access points, have a limitation on simultaneous users, which directly affects the connection experience of the customer. Having multiple customers connecting to the same antenna will not only slow the customers down but will slow down the overall network. A densely population of local access points, connecting to a second level, which connects to the core network, will improve overall network performance.

5) The least amount of wireless network between your customer and your internet backhaul will produce the best results and experience (it is one less piece of bandwidth to manage).

We have our own VoIP partition, working with a UK VoIP network operator, and we offer our customers a range of broadband internet tariffs, VoIP telephone services, IP-PBX, international numbers and very low cost call rates.

There are 2 other WISPs in the area – different parts of the coast – and I doubt that our combined customer base would exceed 2000 subscribers (and that is after we have all thrown a couple of million euros and 4 years of effort). Why?

Just some of the basic factors (not excuses) to give you a flavour of the complications of running a WISP business:

1) it is fair to say that we are all under funded and off the radar of serious VC companies

2) The cost to deploy a communications network, densely hitting street level and maintaining line of sight and the marketing costs to capture the audience is beyond the scope of a privately funded, start-up business.

3) Limited locations to mount our network points. Whilst we have commercial relationships with TV and Radio tower operators, these points are often out range of the customer’s premises (due to our minimum conditions and wireless policy). In addition, the electricity infrastructure on mountain tops is poor to say the least.

4) Lack of skilled staff and operational costs in Spain. The local skill levels are poor in Spain (technical and language ability) – so recruiting locally is not often a good thing. The salaries that we can afford to offer (and the % that we have to pay to the Spanish government) mean that we cannot afford to bring people in from the outside.

5) The politics of deploying a wireless network is often hampered by mis-information regarding the health risks of WIFI and the image issue of putting an antenna on the side of a multi-million euro mansion. We have urbanisations without hope of ever getting traditional connectivity that will not take our service because their community rules stipulate a “no-antenna” policy. To change such a policy requires a vote at the yearly AGM.

6) Education regarding alternative telephone and internet options. People just don’t believe that you can connect them without a wire.

7) Installation costs, service costs and other mis-marketing issues. We are not some multi-billion euro Telco that can wave away the installation costs, nor can we reduce our monthly subscription fees to match the adverts on the TV. You would be amazed how many times I have heard a customer complaining that “its much cheaper at home” … they just don’t seem to understand that they are in Spain.

8) We have probably been (unfairly) tarred with the same brush as the previous cowboys and let-downs that these customer opportunities have come to expect whilst living in Spain.

9) Finally, we all started too early. This obviously is a good thing from an understanding point of view but a fatally bad thing from a start-up business point of view. Our learning curves were steep, expensive and necessary to get where we are today.

If you are not funded, if you have not climbed a tower, actually installed/configured and tried/failed a 100 ways to connect a customer; if you have not immersed yourself in your local territory, understood your customer connectivity demands or understood how to acquire the customer (and believe me, your previous experience means naught) … then don’t try to be a WISP or try to write a WISP business plan.

Other points in the White Paper, that I would like to comment upon.

FON … I disagree that FON has a “good idea”. I have blog’ed Martin Varsavsky and I have challenged him on the same points as you have. FON do not own a network; they believe that the Telcos, WISPs and ISPs are going to let them use their infrastructures for free. Let’s think about this, bandwidth is a commodity that our own subscribers are demanding more of (for less money). The infrastructure costs we have suffered will not be repaid for years to come and the maintenance/upgrade costs are still being budgeted … why would we let a “potential” competitor abuse and use our network for free?

Ah, FON say they will contribute towards the cost by sharing a small % of the fees collected by paying Foneros … when I met them in Madrid last year they could not tell me how much.

From a customer point of view, why would someone what to share their, already contended, bandwidth?

From my company’s perspective and location, I have a large portion of the 55 million tourists entering my network space during the holiday periods. Just image what sort of bandwidth considerations we would have to budget if we allowed FON or any other roaming HotSpot user to access our network. It makes no sense commercially to upset our paying subscribers or jeopardise our own HotSpot revenues by allowing tourists to roam, nor does it make technical sense to “budget” for non-subscribing customers (how would you calculate it?).

Clearwire … I know one of the people that arranged the $900 million investment and start-up fund and I was impressed when he told me their world domination plans (this was 2 years ago). Since then, I have read the USA deployment failures, their financial problems and recently learned that a nearby city (Malaga) is going to be their first European deployment. It is a good point to mention that a year ago I read a similar FON press release saying that they were going to make Malaga the first WIFI city and were going to donate 2,000 FON enabled routers to be deployed as part of the “Club Málaga Valley E-27” initiative – article in Spanish.

I view Wimax with scepticism (at best) … things just do not ring true and it seems to be just another marketing hype. We know that today we can technically deliver what Wimax proclaims – I can give you a 54Mbps wireless link to my network now using standard, off the shelf 5GHz equipment and an equal amount of internet bandwidth if you want to pay for it. We can overcome the non-line of sight benefit, and build a better network, by deploying multiple access points using either 2.4GHz or 5GHz technology and we can do so at a fraction of the cost. So what is the purpose?

I read a Gartner report 3 years ago that showed the evolution of wireless, it suggested 2.4GHz would be phased out by 5GHz and then by Wimax all within 5 years. I am no longer the gadget addict that I used to be but I know that the Laptop manufacturers are still deploying 2.4GHz enable devices today, some even offering 5GHz, and I have yet to see a Wimax (or even Pre-Wimax (that term makes me laugh)) notebook in the shops. Other device manufacturers equally are still developing and manufacturing their devices with WIFI technology. Now, doesn’t that suggest that Wimax is not on the point of launch in the foreseeable future?

I am sure that somewhere there is a company developing Wimax USB dongles and I suspect that they will expect customers to purchase them at the rate of Bluetooth … but then, why would a user with a perfectly functioning WIFI device, located in a WIFI coverage area, want to purchase (yet) another device to connect to the internet?

Going back to a basic point, something that I mentioned earlier and which was mentioned in the white-paper, bandwidth has a price. The reason why ADSL networks are contended (and normally over contended) is to make money/profit. Sell the same 2Mbps a 100 times and hope that only a small % will want to connect at the same time and suddenly your ROI looks good. I think this mentality comes from the old dial-up days when people didn’t really have much to do on the internet and had to pay per minute. Today’s users are very different. Therefore, for WIFI, ADSL, Satellite, Wimax, 3G or even 4G operators to offer huge amounts of bandwidth they will have to either increase their contention ratios further or increase their network infrastructures dramatically.

Increasing contention ratios (or not increasing them) will have a damaging effect on applications such as VoIP or IPTV (e.g. latency, packet loss, performance degradation, etc.). The user experience will suffer and customers will churn. Increasing network infrastructure costs will effect profits, extend ROI and will probably result in higher costs to the customer.

How many people have signed up to 20Mb+ ADSL2 and been disappointed with a “best efforts” 1.5Mbps (or less) at the times when they really want or need to use the internet?

Whatever the last-mile connection format is used, the customer experience goes way beyond this leg. The network design needs proper planning, financing and foresight. R&D budgets and plans need to be quadrupled and fluid. Off-net networks and relationships (i.e. backhaul, VoIP, content providers, etc) have to marry with your business plan and customer promises. Product and Marketing departments really need to understand the limitations of the technologies and the customer desires and stop delivering messages to be “one up on their competitor”. Sales people need to become commercially and technically adept both internally and externally to ensure that not only are they setting the right expectation but also so that they can consult with the customer to ensure the product meets the requirement (it saves on support calls and complaints later). It goes all the way through the business and into the Alliances … so yes, based on my knowledge and experience there are “Flaws in Present Broadband Wireless Access Deployment and Business Models” and it goes much deeper than what we currently believe.

So let’s forget the hype, world domination press releases and let’s get back to basic business sense. The internet is big and its opportunity is growing by the second. One network or technology does not fit all – the very basic foundation of the internet is a collaboration of networks – and one supplier cannot deliver the solution to the problem.

HP shutters its Upline online data backup service – Why commoditized online backup service is not a sustainable business…

I was contacted by ChannelWeb to comment on HP’s decision to shutter their Upline online data backup business. The gist of what I commented was carried in the article “HP To Shutter Upline Online Storage Backup Service” by ChannelWeb…