Paul Tingey

From W3 to the World Wide Web

Fifteen years ago this week, on the 30th of April 1993, two directors from the CERN Particle Physics Laboratory signed and published a document which relinquished "all intellectual property rights to" and permitted "anyone to use, duplicate, modify and redistribute" a technology they referred to as W3.

Today W3 is better know as the World Wide Web, but the concept is the same; a scalable, platform independent information medium where documents are connected through hypertext links. This ground breaking idea was the brainchild of Sir Tim Berners-Lee, now the Director of the World Wide Web Consortium and a respected technology visionary.

It would be easy to say that, fifteen years ago, no-one could possibly have anticipated how the release of W3 into the public domain would change the way that information is created, recorded, transfered, used and (unfortunately) abused. However, it is clear that Berners-Lee and CERN developed W3 with a specific set of activities and applications in mind and, broadly speaking, those same applications still drive how the World Wide Web is used today.

In one paragraph of CERN's original two page press release the organisation outlines the application areas for which W3 was well suited. The list includes :

• technical design notes
• documentation
• news
• discussion
• educational material
• personal notes
• publicity
• bulletin boards
• live status information
• numerical data

Other than some obvious scientific and research overtones, these key application areas still hold true today although the terminology has evolved. So, for example, discussion now takes place in forums, publicity is sought through corporate websites and personal notes appear in blogs and on the pages of social networking sites.

In fact only in the last couple of years have we really begun to see the web grow out of it's infancy and push CERN's original vision further than may have been anticipated at it's birth. YouTube, deviantART and Google Maps don't easily fit into any of CERN's original application types and, as such, mark the beginning of what may be the second age of the web. Some refer to this adolescent web as Web 2.0 but personally I'm not keen on that label since it hints towards a step change in the infrastructure whereas in reality todays web is more about evolved applications than a revolution in the underlying technology. The term Web 2.0 only really exists because we don't have a snappier way to describe the trend towards web applications based on collaboration, software services and multimedia data that we are now witnessing.

And where does the evolution of the World Wide Web take us next ?  Well, those eager to get some pointers may be interested to read about Semantic Web, an evolution of the World Wide Web proposed by Berners-Lee which provides a data sharing framework and enables a highly integrated "web of data". Lets not give it a version number yet - let's just look forward to the more mature years of a "middle aged" future web.

Mobile Broadband dongles find success in UK

There seems to have many news stories about the imminent rise of 3G technology in the UK over the last few years. Looking back these stories would seem to have heralded repeated false dawns with 3G being relegated to little more than another mobile voice calling technology. However we now seem to have positive proof that 3G based mobile broadband is becoming a viable alternative to (or addition to) more traditional broadband technologies for UK users.

In a blog entry titled Mobile net takes off, Rory Cellan-Jones of the BBC includes what he describes as "an extraordinary graph" showing how the amount of data crossing the 3 mobile operator's 3G network has increased 14 fold in just six months.

According to 3 this massive increase in the amount of data throughput on their 3G network is due to what they refer to as dongles - plug-and-play USB modems that use High-Speed Downlink Packet Access or HSDPA technology (sometimes referred to as 3.5G) to enable laptops to use mobile broadband. 3 launched their dongles last October and immediately saw a huge increase in the traffic crossing it's network.

The dongles are being offered with a fixed price plan (for example 3GB of data for £15 a month) which has triggered a price war in the UK market with other operators such as T-Mobile and Vodafone competing with similarly aggressive deals.

The responses to Rory's post are generally positive about 3's mobile broadband service and it seems clear that possibly as many as half a million new users have signed up to 3's dongle and now use mobile broadband on their laptop (or desktop) machines.

Of course many of these users are likely to have made a choice between competing technologies which means that mobile broadband dongles may now be taking market share from, for example :

• traditional wireless LAN technology which, whilst fairly commonly found, is less pervasive and less mobile than 3G and is rarely provided as a free service in the UK
• fixed broadband services which often incur significant installation fees for new users and often require users to pay an additional monthly fee for a fixed phone service (a service which is itself under threat from low cost mobile payment plans)
• WiMax (or 802.16e) which may not have a significant market share currently but it's interesting to consider how the success of 3.5G/HSDPA based services may impact it's roll-out over the next few years

How this tangled marketplace of competing technologies evolves will become apparent over the next months and years but one thing seems clear - after several years of promise, 3G technology seems to have gained a significant foothold in the UK market.

"Making Multicore Work" Seminars

February has been a particularly busy month. So busy, in fact, that I was not able to make time to be at the Mobile World Congress in Barcelona. Instead I have spent most of the month completing a tour of seven European cities presenting at a joint Wind River/Freescale/Virtutech seminar series.

The theme of the seminar was "Making Multicore Work". The presentations included details of Freescale's newest and upcoming multicore devices, Virtutech's virtualised software development technology and Wind River's latest VxWorks SMP and multicore debug tools releases.

The subject of the seminar seemed particularly popular with the number and variety of attendees at each venue exceeding our expectations. Attendees included engineers and development managers from the Mobile Handset, Automotive, Defense and Networking industries.

The focus for my particular presentation was Symmetric Multi-Processing (more specifically VxWorks SMP) during which I outlined the challenges involved in programming for a multicore environment. I also presented a live demonstration of VxWorks SMP (running on the new MPC8572E based Freescale reference platform) in which I attempted to highlight those programming practices which may lead to problems in a multicore environment.

During my presentation I was keen to outline that problems may not only be caused by "bad" programming practices but may also be the result of apparently acceptable program structure which may fail due to assumptions made by the programmer. For example, programmers often expect memory operations to be performed in the order in which they appear in source code. However, modern processors use out-of-order execution [Wikipedia reference] techniques which may cause load/store operations to be re-ordered. In a single core system this is rarely (if ever) a problem due to the fact that the memory operation "pipeline" will be emptied/cleaned before or during the next context switch so the software knows little or nothing about the hardware optimisations happening beneath it. However, in a multicore system, where two or more tasks or ISRs are running concurrently, the programmer needs to be aware of the effect that out-of-order load/store operations may have and use appropriate memory barrier [Wikipedia reference] instructions where necessary. Without these memory barriers, out-of-order load/store operations on a multicore system could give rise to the worst kind of timing related bugs and race conditions.

I'm hoping that the audience at each of our venues took away something positive from the event. For myself, the event has reinforced my view that multicore devices will become increasingly commonplace over the next months and years. The attendance numbers and the reactions of attendees make it clear that engineers are being forced to think in terms of multicore (either Symmetric or Asymmetric multi-processing). The challenge for those engineers is to adapt the way they write code for a multicore environment whilst at the same time, fully utilising the hardware resources they have available.

The emergence of the Femtocell

Amongst all the news (and some would say hype) about Fixed Mobile Convergence, there is another market dynamic being played out which may lead to an alternative form of change. Fixed-to-Mobile Substitution (or FMS) aims to increase the proportion of calls made from traditional 2G/3G mobile handsets at the expense of those made from other devices such as fixed lines or dual-mode (or converged) handsets.

In reality this is a trend started by consumers themselves who, attracted by falling mobile call rates, have begun to use their mobile phones for all their voice calls whether they are outside of or within their homes. This substitution effect is most noticeable (for fixed line operators) amongst new users (i.e. in new households) who are often opting not to have a fixed line phone installed at all.

This is seen by many as being a potentially huge threat to the revenues of fixed line operators. For example whilst about 6% of homes in the UK are "mobile only", in Finland, where the effects of FMS have been particularly marked, the figure is closer to 30% leading to a sizable reduction in fixed line revenues. Obviously mobile operators, with an interest in underpinning their own revenues, are motivated to foster the success of FMS and have shown interest in any technology which enables this. The most significant of these is referred to as the Femtocell.

Femtocells are small "in building" 2G/3G access devices (similar in size to a wireless access point). They connect, on the radio side, to a standard mobile phone and, on the network side, to a users broadband connection and have the advantage of providing excellent mobile coverage within the home as well as low cost calling (via IP backhaul).

Earlier this week some of the big players in the Femtocell market announced the formation of a consortium of their own to "advance the development and adoption of femtocell products and services". Founding members of the Femto Forum include Airvana, ip.access, Netgear, Ubiquisys and picoChip (among others) with other members likely to be announced over the coming months.

The creation of the Femto Forum comes at a time when Femtocell products are just beginning to hit the market. Nokia Siemens Networks have, only this week, announced a 3G Femto Home Access solution which will begin trials at the beginning of 2008 and will be deployed later the same year. Ericsson announced a similar Femtocell product early this year which may become available later in 2007.

As our love affair with mobile telephony continues to move from strength to strength, it's just possible that we could see Femtocell technology becoming as commonplace as wireless access points are today. And, if the Femto Forum works its magic, it could happen sooner than we think !

Fixed Mobile Convergence : Evolution or Revolution ?

The last couple of months have seen a flurry of reports on the subject of the future prospects for Fixed Mobile Convergence (from, for example, Pyramid Research and IDC) which has, in turn, spawned a flurry of web articles and blog comment on the same subject. The consensus among most of these reports is that Fixed Mobile Convergence (or FMC) adoption is set to grow over the next few years and that those companies who are investing in it are, in the main, investing wisely.

However, how users (or should that be prospective users) view the move towards FMC is more difficult to judge. In an article on Computing.co.uk, a research Director from IDC describes how their recent report outlines their view that operators (at least in Western Europe) are finding it difficult to sign up customers to FMC services.

Apparently, the report goes on to suggest that FMC should not be sold as a new or differentiated service but, instead, should be marketed as a complement or extension to existing broadband and home network products. This seems sensible although it is a strategy which assumes that target customers have a good understanding of home networking, WiFi and associated technologies (some aspects of which are, well, less than intuitive in their current form).

Another interesting question is whether consumers (many of whom know little or nothing about FMC) are actually asking for the kind of services it can provide. This particular point is answered in a most interesting way in a recent blog entry by Carl Weinschenk on ITBusinessEdge.com which, for me, provided one of those, "I wish I had written that" moments !

Carl's main point is encapsulated in his title - "They don't know it, but users are driving FMC". He goes on to describe how it is users who are driving the trajectory of communications technology towards FMC by increasingly seeing "little or no distinction between what they can do with either [wired or mobile devices]". Essentially, the use of features such as syncing between devices and agile switching between networks has created an increasingly seamless feel to the way in which differing device types are used.

Of course there is a definite step between this "seamless feel" and true FMC but the point is that users are indicating, through their use of current technology, how they want technology to evolve into the future. So, for example, a line drawn forward from those technologies which are in widespread use today such as Web browsing, Email, VoIP, IM, Multimedia content and enhanced mobility (in general) could lead us to a point where truly unified FMC devices could be in widespread use tomorrow.

Maybe the lesson from the IDC report is that these new devices will be more readily accepted if marketed and sold as advanced but evolutionary fixed network devices rather than revolutionary mobile devices.

Climber makes "worlds highest mobile phone call"

Mobile telephony reached another frontier yesterday when the British climber Rod Baber set a new record by making the world's highest land-based mobile phone call from the summit of Everest. According to his sponsor (Motorola) Rod Baber made a call from 29,035 feet (8848 metres) to a voicemail box. You can listen to his message via a link on the BBC website.

It would seem that the phone call was well outside of the normal operating envelope for both the caller (who dialed prepared speed dial numbers using a small pointer because it was too cold to remove his gloves) and the phone itself (Mr. Baber had to tape the batteries to his arm to ensure that they remained warm enough to operate the phone). Mr. Baber also called home to his family as well as sending the world's highest land-based SMS message reading "One small text for man, one giant leap for mobilekind – thanks Motorola".

The new record was made possible when China aligned a mobile base station just 12 miles from the summit and in line of sight with the North Ridge of Everest. This should make it possible for mountaineers to stay in closer touch with the outside world and for rescue missions to use mobile communications on the mountain.

This is one of a number of "stunts" happening during the current Everest climbing season of which the most dangerous is likely to be an attempt by the Dutch climber Wim "Iceman" Hof to climb to the summit wearing only shorts, boots, gloves and a cap. By the end of this years climbing season it is expected that approximately 550 people will have reached the summit of the worlds highest peak.

Evolving Architecture Standards race on

On Wednesday of last week I had the chance to present at the Evolving Architecture Standards Conference in the UK. The event was focused on the ATCA, AMC and MicroTCA standards and gave delegates the chance to learn more about the evolution, capabilities and potential uses of these technologies.

Apart from presentations from representatives of Intel, Motorola, PICMG, Schroff and myself from Wind River, delegates also had the chance to take part in hands-on and tutorials lessons from Intel, Schroff and Wind River. There was also a day long exhibition featuring about 15 companies offering ATCA and MicroTCA related products.

Having spent some time talking to both delegates and exhibitors, there seems to be general agreement that ATCA based products are now emerging from the early adopter stage and gaining widespread support in the marketplace. The transition from innovation to adopted technology has been somewhat longer than some would have expected; maybe two years or more.

However, the consensus of opinion seems to be that MicroTCA will gain wide acceptance within a significantly shorter period of time. It is helped by the fact that the underlying technology (AMC) is well tested and readily available. MicroTCA will also benefit from a wider customer base across differing market sectors as evidenced by the wide range of registrants for last weeks conference who included representatives from industries such as Defense, Scientific research, Networking, Test Equipment and Industrial control.

The event was held at the excellent Williams F1 Conference Centre in Grove near Oxford and, at the end of the day, delegates had a chance to tour the Williams F1 Collection of more than 40 Formula One cars from past and present. If you are a Formula One fan like myself then this is one conference centre where it's definitely worth carrying a camera !

Carrier Grade Linux version 4.0 specification is released

Last week saw Carrier Grade Linux take a further step forward with the release of the latest version of the specification. The CGL 4.0 document, released on 26th February 2007, is the result of work by individuals in many organizations and heralds a new level of collaboration between all those involved in the Carrier Grade equipment business.

CGL 4.0 is the first version of the Carrier Grade Linux specification to be released by the Linux Foundation, a recently created group formed by the merger of the Open Source Development Labs (OSDL) and the Free Standards Group.

Improved alignment between the requirements of the telecoms equipment providers and the contents of the CGL Specification itself has been made possible through the work of the SCOPE Alliance. This consortium of equipment providers (such as Motorola, Alcatel/Lucent, Nokia, Huawei and others) has created a profile to outline those features of a carrier grade system which are most important for them and their customers to act as a kind of guidance system to keep the CGL specification in line with end user requirements.

The CGL profile created by SCOPE defines a total of approximately 250 individual requirements or features divided into seven groups; Availability, Clustering, Serviceability, Performance, Standards, Hardware and Security. Each of the individual requirements is assigned one of three priority levels; Mandatory, Desired and Roadmap. Obviously those requirements defined as Mandatory are seen as necessary to a successful Carrier Grade Linux system and, as such, must be present in any CGL 4.0 registered distribution.

Prior to CGL 4.0 there was no formalized registration process for Linux distributions claiming compliance. However, CGL 4.0 introduces a formal registration process and requires that any compliant distribution must meet all of the 135 Mandatory requirements.

The next step in the evolution of CGL will see the specification aligned with the Linux Standards Base (LSB) with the aim of increasing compatibility between CGL compliant and mainline distributions such that applications will run on either type of Linux in exactly the same way.

The CGL specification has been through a number of iterations (the first version was released in 2002) but the collaboration between standards groups, Linux vendors and equipment providers which is such a feature of the development of the CGL 4.0 specification marks a particularly important milestone in the story of Carrier Grade Linux. Cooperation between interested parties of this kind is likely to reinforce the position of CGL as the operating system of choice for telecom systems in the future.

Hello from 3GSM in Barcelona

I am writing this during the third day of the of the 3GSM Conference and Show in Barcelona. Having walked the halls (or at least some of them) over the last couple of days it is apparent to me now just how huge this show is; in the region of 1,300 stands spread over six huge halls and approximately 60,000 visitors make for a very lively atmosphere.

If I had to choose the most "talked about" single technology then it would probably be WiMax. However, maybe that is because it's the most recently maturing of the "convergence enabling" technologies. And in truth, convergence is the strongest theme of the show with converged networks, converged handsets and even converged silicon featuring widely in the show. 

Aside from the Conference, the show itself is organized into broad themes within each hall. The large network infrastructure companies find their place in Hall 8 with, for example, Nortel, Motorola, Alcatel-Lucent and Ericsson having huge stands around a central square. For sheer size however, Nokia-Siemens (still listed in the show literature as "Nokia and Siemens representing future Nokia Siemens Networks") wins the day. Not only do they have a huge stand on the main floor, they also have a separate two level stand best described as a business center incorporating meetings rooms, offices and a restaurant.

In Hall 7 the convergence theme is brought together by providers of mobile content and entertainment products. There is a particularly strong emphasis on mobile TV and this hall also includes a content theater and the 3GSM TV Studio.

Being 3GSM there is a strong focus on convergence at the physical handset. Many companies are showing their latest products supporting technologies such as 3G, HSDPA, EDGE and WiFi/VoIP. If visitor traffic is anything to go by the most popular handset manufacturer appears to be LG who are showing their beautifully designed Chocolate and Shine mobile handsets and who also, by the way, win my vote for the best quality carpet I've ever seen on an exhibition stand. The Blackberry stand also sees significant traffic by offering visitors the chance to try out the features of the newest Blackberry models. Other mobile device vendors such as Sony-Ericsson, Samsung, Nokia and Huawei are also showing their latest products.

As well as the halls, the central boulevard running through the venue is packed with trucks exhibiting hot new technology products as well as other items of interest such as stunt bikes and skateboarding. Luckily, as you can see in the picture above, the Barcelona weather has been very kind and makes it particularly pleasant to walk the outside spaces.

Day two of the show saw what was, for me, a particularly interesting headline in the 3GSM Daily (yes, this show is so big that it has it's own daily newspaper). Tim Berners-Lee, the inventor of the World Wide Web, called in his keynote speech for technologies which formed foundations (i.e. the basis of long term returns for everyone) rather than ceilings (i.e. the end of the road for innovation). This was seen by some as a call to end the kind of restrictive technologies and methodologies which have hampered the development of a truly open web (e.g. DRM and walled gardens).

Of course, there has to be a downside and at this show it has to be the sheer size which offers some special problems. Aside from a good pair of shoes, visitors and exhibitors need lots of patience to deal with the crowds leaving the show at 7:00pm each evening. Taxis and the Barcelona Metro struggle to cope with the tens of thousands of visitors all leaving at a similar time.

And the upside?  Well, good food and excellent weather in a beautiful city is a good start. Isn't it?

BT encouraged to be more open

I have often written in this blog about open source and its use in the kinds of equipment which sits at the heart of telecommunications networks. Carrier Grade Linux and other open source technology, whilst not quite pervasive yet, are becoming more and more commonplace as underlying component technologies in mobile and Internet core elements.

On the other hand, I haven't written much about user devices (or Customer Premise Equipment). However you can be sure that open source software is making an equally big mark in that area although, generally speaking, its use stays "under the radar" with most users remaining unaware of the nature of the software which operates the networked devices in their homes.

Occasionally however, events force open source software above the radar and into plain sight. The last few weeks have seen an example of this in the UK centering around a particularly widely used ADSL broadband Wireless router; the BT Home Hub. The device was launched about six months ago and supports an integrated VoIP phone as well as BT's Fusion FMC offering and their Vision IPTV offering. BT supply telephone and DSL services to a large proportion of UK household and, as a result, the Home Hub has gained many users in a short time.

A few weeks ago, after questions raised by users, BT admitted that the Home Hub was based on open source software (including Linux) and is therefore subject to the terms of the GPL. Since then the story has been closely followed by The Register, The Free Software Foundation Europe and various blogs (most notably Home Hub Blog).

Eventually, after a short delay, BT posted the source code for the Linux kernel used on the Home Hub on a web site (found here). However, the situation seems to be rumbling on with BT claiming they have published everything relevant and the Free Software Foundation Europe claiming that some elements (such as the top level Makefile) are still missing.

This situation and the continuing pressure on BT to open up their source code highlights, for me, how far we have come down the route to open source. It feels starkly different to the situation in the late 1980's and early 1990's when I was involved in the design of a test strategy for a telephony device which was approved under the BABT certification process. The aim of the testing process was to ensure that users were safe while using the product and also that the network itself would not be adversely affected by connection of the product. The process we put in place included safety tests as well as device functionality tests and, after the tests were successfully passed, we were required to physically seal the plastics so that the device could not be tampered with. These procedures were enforced by stringent inspections and failure of any part of the testing process caused a stop of the production line.

So, in some ways, it seems somewhat strange to me that we now have a telephone network within which open-ness (based on open source and open standards) is an integral part. Of course the true nature of the telephone network is more of a regulated network infrastructure incorporating isolated islands of open-ness but even that feels a world away from the closed and heavily controlled nature of the BT network in the 80's and 90's when even manufacturers (let alone home users) could not easily change the code in their devices without considering the impact on their certification status.

Of course open source and open standards are here to stay since they offer the opportunity to accelerate product development and should therefore be welcomed. I wonder if the product managers in charge of the BT Home Hub are thinking the same way?