Tag Archives: 2007

Graham Helsby and his *BIG* sandwich

Here’s my mate Graham with a very large sarny…


Now that’s a *Big* sandwich…

Last years photos from Sun’s Customer Engineering Conference 2007 (CEC 2007)

Found these the other day, posted up to Flickr…


Enterprise Architecture @ Sun Microsystems

As I mentioned in this post Enterprise Architecture Practitioners Conference 2007 here is my presentation in full from that event – along with the speakers notes.Not yet attached is the overview of differing architectural skill / capability ‘domains’ which I ‘white-boarded’ at the event – but I’ll do a post later which includes a web friendly version of that and link it in.Don’t forget that you can use the ‘Body Image Size’ function over on the right hand side to resize the images to something you find more acceptable.

Enjoy !

Enterprise Architecture @ Sun Microsystems

  1. Welcome to the Conference !
  2. Introductions
  3. Contents
  4. The 3 major uses of EA @ Sun Microsystems
  5. EA used internally at Sun
  6. Sun internal EA has delivered…
  7. EA used externally for Sun Customers : Why do customers involve Sun in there EA Programmes
  8. EA used externally for Sun Customers : Supporting customers EA teams helps Sun to…
  9. EA used externally for Sun Customers : Three EA case studies…
  10. EA case study no. 1 : A large Utility company : When EA goes ‘Bad’
  11. EA case study no. 2 : A large Government department : EA & SOA – the perfect marriage ?
  12. EA case study no. 2 : A large Government department : EA + SOA = SOA Adoption Roadmaps
  13. EA case study no. 3 : A large Government organisation : “It’s about the people, stupid !”
  14. Enterprise Architecture and Innovation…
  15.    …using EA to perform customer analysis
  16. In Summary – Enterprise Architecture ‘Best Practises’
  17. Where can you get EA help at Sun Microsystems ? Try over here…
  18. Q&A;
  19. Thank you very much & enjoy the rest of the Conference !

Welcome to the Conference !

Hello everyone,

I’d just like to start off by thanking the Open Group for organising this event, specifically
John Spencer, who kindly asked me to present to you all today, and Allen Brown, the
Chairman of the Open Group, who’s given Sun, and I, a very warm welcome today.

And I’d also like to thank you all for coming along to see me, I hope you enjoy the


So who am I ?

My name is Wayne Horkan, and I’m the Chief Technologist (CT) for the UK and Ireland at Sun

I’ve been the UK & Ireland Chief Technologist for the last eighteen months, and I’ve been
at Sun for the last seven years, in total.

As CT I provide Technical Leadership to the entire UK and Ireland Sun organisation, with specific responsibilities for the senior customer facing Business and Technology Consultants in that organisation. A significant part of my role is spent engaging with our customers, and I sit on a number of our customers Technical Governance structures and boards – all of whom are responsible for generating and managing EAs.

Furthermore historically Sun and the Open Group have had an excellent relationship, with
Sun having been a founder member of the Open Group.


I’m going to cover the three major uses of EA at Sun Microsystems, including three case
studies from our customer engagements.

After that I’m going to discuss our “Seven Rules for Successful EA”, which were drawn
out from lessons learned from these and other engagements.

That’s going to be followed by a short Q&A; before I finish.

The 3 major uses of EA @ Sun Microsystems

If we look at the three major applications of EA at Sun Microsystems, they are:

  1. Internally – EA used by Sun internally to manage it’s own IT estate
  2. Externally – where Sun consultants have been involved with customer EA teams and
    EA initiatives
  3. Both – Or rather EA practiced internally to better understand our external customers’ IT

EA used internally at Sun

EA was introduced into Sun Micro by our previous CIO – Bill Vass.

Before Bill came to Sun he worked in the Office of the Secretary of Defense, Office of the
CIO. In this capacity, Bill was director of three large sectors of the U.S. Department of
Defense’s IT infrastructure. and was very familiar with EA from the introduction of

His roles at the DOD included:

  1. Director DOD where he was software manager for over 30 projects
  2. CIO at the Pentagon in infrastructure
  3. Networks operation Manager at the Security’s Division

Sun has an EA office which co-ordinates EA across Sun’s IT Operations (‘ITops’) – Bill
set this operation up about four years ago.

Our current CIO – Bob Worrall, having taken over from Bill about a year ago has the EA
office reporting directly to him.

Sun internal EA has delivered…

It has delivered 3 important pieces of work:

  1. An EA of Sun’s IT Ecosystem – but an EA is not just a means in itself. By developing
    an EA we have been able to start work at EA Planning, and thus able to deliver even more
    definite business benefit.
  2. Project Helios, which has been delivered into production, an online Runtime,
    Service/Asset Management System. This for me is “Asset Management evolved” – as it
    allows you to view at Runtime Business Services to Business Units, to Application
    Services, to Application, to Operations Team, to Infrastructure & Hardware, to Data
    Centre & Location and to Facilities Management. It is a cornerstone of our SOA adoption
    roadmap and will become an online management information system.
  3. Project IBIS, stands for “Integrated Business Information Solution” and is a
    concentrated effort to reduce complexity across the Sun IT Estate. The IBIS project will be
    looking at consolidating circa 1,000 business applications, world-wide, into a single
    Oracle Financials runtime. We are achieving this strategy by replacing the atypical “80%”
    of all functionality with a single implementation of Oracle financials, combined with an
    SOA approach of encapsulation & aggregation for the remaining “20%” functionality.

In the UK we have a saying about Cobblers Shoes – the saying goes that cobblers shoes
are always in poor condition because the cobbler could at any point repair them – that’s
what he does of course. And I think a little of that happened at Sun.

Because were a technology firm, full to the brim with smart innovative technologists, it
was always supremely easy to construct applications, and even worse – easy to support
them well into production.

Thankfully Project IBIS is going to change that. We have spent the last 2 years auditing
all related and impacted business processes and applications across Sun and have
developed a plan & route into the future. Furthermore we have implemented a governance
structure which will minimize this happening again.

Obviously Project IBIS is a massive concern – affecting circa 40,00 staff worldwide, not
only producing efficiencies in the Supply chain, simplification of the Estate, but also
reduction of support required and the costs attributable to such a difficult to manage IT

So you can see we have a very live, well placed and well sponsored EA function within

EA used externally for Sun Customers : Why do customers involve Sun in there EA Programmes

Next I’m going to talk about the use of EA externally with our Customers.

Sun gets involved with customers EA initiatives, and we are often asked to supply Senior
Technologists into customers EA governance structures to help with EA definition.

This happens for a number of reasons, including:

  1. Vendor representations – customers want tier one vendors to act as IT / Technology
    industry representatives.
  2. Decision ‘Backup’ – validation of opinions and techniques based upon experience of
    seeing it being done elsewhere.
  3. Trust – the customer trusts Sun for impartial advice.
  4. Reliance on core technical stock – if the customer has a large amount of Sun technology
    they might want representation from us in moving that forward.
  5. Consultant specialist knowledge, experience or relationships – an ex-employee, for
    example, or someone involved in a previous implementation.

EA used externally for Sun Customers : Supporting customers EA teams helps Sun to…

As the Technology Leader for the UK and Ireland, I actively encourage these relationships
– as they:

  1. Give Sun a chance to demonstrate value – outside of pure product.
  2. Allow us a better understanding of the aims and challenge of our customer
  3. De-risk implementations by giving us opportunity to contribute to the design and
    implementation of new systems as well as a long term and strategic view of the customer
    IT ecosystem.

Additionally it allows us to “up skill” our own staff in EA delivery.

EA used externally for Sun Customers : Three EA case studies…

I’d just like to talk briefly about some of the customers who we’ve worked with their EA teams.

I’m going to cover three case studies:

  • firstly a large utility company – “When EA goes ‘Bad’”
  • secondly a very large government department – an example of “EA and SOA”
  • thirdly, another extremely large, government organisation – “EA Technical Governance” or “it’s about the people, stupid !”.

EA case study no. 1 : A large Utility company : When EA goes ‘Bad’

We got involved with the EA work at this customer around 3 years ago. It was extremely
interesting to see how EA was being used, how it had evolved and the issues that they had.

I call this “when EA goes bad” – because in my mind it was the best (or worst, given your
viewpoint,) example of “Ivory Tower Syndrome” – a problem when EA teams become
isolated, withdrawn and stop communicating.

Issues included:

  1. EA was over “deliverable focused” and “data driven” – with a lack of a “big picture”
  2. Not Inclusive – leading to severe “Ivory Tower Syndrome”.
  3. Lack of clear objectives for the EA team to deliver against.
  4. £80m spend on a single project cancelled – due to the EA Team.
  5. No / lack of Senior Sponsorship / Visibility – Accountability

I personally conducted two reviews for them. Firstly, a review of the vendors to their 70m
pound outsourcing deal – where we subsequently had to redesign their proposed operating
structure so that it was based upon a cross cut of technology skills and vendor and
project inclusion.

Secondly a review of an ongoing project which had incorrectly been identified by the EA
team as a strategic component. Initially this was a combined workflow and customer
management system – including remote, wireless, handheld data collectors (for ‘pipe’
asset location).

The programme had been due to deliver £1 m in a year savings, initially costing £8M This
had very soon ballooned to £11M, then £40M, and finally £80M – including requiring an
estimated further £40M, so a total of £120M, to be delivered.
You may ask why a programme with a 120 year return on investment (ROI) would still be
being delivered.

Unfortunately the EA council, which had become very parochial due to a lack of new
members and lack of integration with the project members & architects had deemed parts
of the program to be fit to repurpose as an Enterprise Service Bus (ESB) – effectively they
said that the programme was delivering a Strategic corporate component.

This wasn’t the case as it was built as an “end-point system to end-point” system model
(commonly called Point to Point). A legacy approach to integration, one that, for apart
from very small systems, has been discounted. It’s extremely costly in terms of connecting
new systems and would not scale as an Enterprise corporate component – neither in terms
of ease of re-provisioning nor development nor Runtime reusability.

If the EA team had listened to the development team they would have known this – but
because of “Ivory Tower Syndrome” and the organisational hierarchy involved they were
not open to learning this particularly bad news. The EA team had become elitist, out of
step with the business, disconnected to the project teams, and had stopped communicating.

After reviewing the Programme it was shut down by the incumbent IT Director, with the
resultant loss of over 400 jobs. Obviously we were congratulated for saving the other
£40M – but I would have preferred for the original EA team to have not become so
inwardly focused, and to not have mistakenly championed a system which would have
never have been able to grow into the strategic corporate component that they had
imagined. We subsequently helped the customer to rebuild its EA function with a new,
inclusive Governance structure.

EA case study no. 2 : A large Government department : EA & SOA – the perfect marriage ?

We were asked to get involved with this customers EA board when they were reviewing
their SOA implementation and had asked Sun to develop their SOA Adoption Roadmap.

We found a legacy enterprise model which had not been updated for a number of years –
unfortunately it had been delivered by Business Analysts so it was very much a catalogue
of business processes. Very little was related to real world application / logical or
infrastructure / physical technical domains.

We found that to deliver a realistic SOA Adoption Roadmap – in terms of ability to be
implemented – we needed the beginnings of an Enterprise Architecture to provide a
contextual backdrop upon which to base our SOA Adoption plan.

So we started by rapidly developing a light weight yet inclusive EA Model which included
the major corporate components that were due to be involved in the SOA implementation.

EA case study no. 2 : A large Government department : EA + SOA = SOA Adoption Roadmaps

We were able to define the strategic roadmap (and roadmaps) for implementation of an
SOA across the organisation, once this was done. Furthermore, we could understand the
implications of what we were planning to do by their impact on the other corporate
components. This allowed us to model differing roadmaps, and judge them on a variety of

The SOA Adoption Roadmaps could be compared and judged against:

  1. Functionality delivered – i.e. what capabilities would be introduced at which point in
    the overall implementation plan.
  2. Costs to implement – for each stage, or alternate stage, what the cost would be, as well
    as it’s relationship to previous and future stages.
  3. Complexity of implementation – how difficult it would be for each stage, and some of
    the risks involved (large discreet stages, versus, small “phased” stages).
  4. Non functional, but strategic and re-useable – the “cross the board” estate capabilities,
    such as the messaging sub-system (Enterprise Service Bus, or ESB, as it’s also known),
    non-functional items such as aggregated audit, which could be re-used outside of the SOA

EA case study no. 3 : A large Government organisation : “It’s about the people, stupid !”

This case study is about EA & Technical Governance – and I’ve called it “it’s about the
people, stupid!”. EA is really about the people – those helping to deliver it, those
influenced by it, and those who it delivers for. I feel that it is imperative to get the right
mix when building an EA function.

The organisation at the centre of this case study is the single largest employer in the UK,
with around 1M staff. It is a distributed organisation supporting the needs of all subjects
in the UK.

It is currently going through a massive change program – an ambitious plan to provide an
integrated stack of functionality across the organisation and it’s subsidiary organisations.
Comprised of a central message subsystem and five major Service Provider Regions. Sun
is involved in delivering the messaging subsystem, or rather G2G backbone (very similar
to a B2B system) which supports all data traffic.

I was asked to intervene in an issue with the prime vendor / supplier to the end customer –
one of the largest UK telecoms companies.

When I got on site I realised that the issue was one of technical governance model and
that although we had teams of application architects and teams of infrastructure / system
architects we did not have any Enterprise Architects. People who would get a contextual
view of what was being attempted, nor, as they weren’t any, our ability to map to the
customers EA team !

By insisting we had a team of EA’s led by one of Sun’s most experienced Enterprise
Architects we were able to move the project out of escalation and avoid severe financial
penalties, which would have been levied if communication had continued to fail between
our teams.

Enterprise Architecture and Innovation…

At Sun we see ourselves as a force for innovation – and the vanguard of disruptive

So it shouldn’t be a surprise that we take EA principles and methods and shake them
around a bit … quite a bit.

At Sun in the UK and Ireland we are using EA in the pre-sales process, obviously where
possible and legally acceptable, to map customer accounts.

   …using EA to perform customer analysis

We also use EA techniques to help understand our customers better. Often our customers
are more than happy that we are so interested in them, and the successful evolution of their
IT / IS Estates.

We call these rapid, light weight, EA methods collectively as Sun LEAF (Light-weight
Enterprise Architecture Framework)

By mapping out our customers IT / IS Estates, and by relating Business Units, to
functional and non-function corporate components and the application architectures,
software stacks and physical architectures that under pin them we are able to learn an
unprecedented amount about our customers use of technology.

We can use the information to identify a number of items – such as which corporate
components consist of Sun Technology – and more importantly which ones don’t – and
thus are applicable for targeted sales. Which business units have systems we have no
visibility at all – and so we need to investigate more. What technologies connect together
and how easy it might be for us to displace them with the least amount of impact on the
customer estate.

The Sun LEAF programme is currently in its second pilot prior to roll-out. Some significant
success have already been made, which I can’t currently disclose at the moment.

In Summary – Enterprise Architecture ‘Best Practises’

That wraps up the overview of EA use at Sun Microsystems. However, I thought it would
be useful to look over the best practices we have identified and developed at Sun for EA.

I think it’s obvious that EA needs to rapidly speed up it’s time to delivery – without a large
drop in the quality of EA being delivered. Basically we, as Enterprise Architects, all need
to be smarter at our game. That’s why I welcome the focus on EA methods, tools and
standardisation that this conference, and the Open Group bring.

I’ve seen too many EA programs either fail to deliver against there original goals, get lost
in the detail without delivering the “big picture” / contextual view promised, and,
increasingly in the UK and Ireland, where the average time in role for a CIO in a top 200
company has slipped to 2 to 3 years maximum, fail to deliver in the timescales of the
original sponsor – only to lose sponsorship & support and be closed down without
delivering at all.

  1. EA must relate to a definite deliverable goal or aim – otherwise it can become an in
    means in itself. Obvious examples include:

    • Reducing complexity – and the cost reductions and savings that obviously result
    • SOA Adoption Roadmaps – or EAI / Integration
    • Plan for the future of the technical Estate – fashionably called EA Planning.
  2. Don’t labour it – keep “fast and light weight”. Unless you really need to – get quick
    demonstrable ‘wins’ onside to encourage buy in and acceptance. Think about the concept
    of “gearing” – delivering discrete pieces of the EA which can relate to actual business
    value which can then be used to justify the next stages proposed.
  3. Inclusion – Involve people from the wider technology eco-system, both internally and
    externally – both vendor and project staff
  4. Rotation – avoid “Ivory Tower Syndrome” by rotating staff in and out of the EA team
  5. Get the right Technical Governance skills base mix (NB Use the whiteboard to
    demonstrate – EA Technical skill domains spider diagrams developed by Sun)
  6. Must get the right level of sponsorship – EA functions straight into the CIO office.
  7. Keep aligned to the Business Strategy Board – EA is about the Technical Strategy for a
    company, and it must relate to the Business Strategy of that company – or will end up
    delivering something that the business can not adopt.

Where can you get EA help at Sun Microsystems ? Try over here…

So where can you get help from Sun re: Enterprise Architecture.

If you have a Sun account team, and I know some of you do, then use that account team to
identify local experts – reference this event, my presentation and me too.

Alternately you are very welcome to get in touch with me, and I’ll either help identify the
best person to help you in the circumstance, or even work with you directly.


That concludes my presentation on Enterprise Architecture @ Sun Microsystems – I very
much hope that it’s been enjoyable and educative.

My contact details are on the slide behind me – if you’d like to get in touch I’d be very
happy to hear from you.

We just have time for a short question and answer session – so any questions ?

Thank you very much & enjoy the rest of the Conference !

Thanks a lot for all your questions, and thank you very much again for coming along to
see me today.

Just like to say “all the best” and I hope you all enjoy the rest of the conference.


Related Links:

Enterprise Architecture Practitioners Conference, 2007

I’ve finally prepared my slides from when I presented at the Enterprise Architecture Practitioners Conference, 2007, so just thought I’d finalise by covering the day and getting the slides online…

I have to say it was a great day – Enterprise Architecture (EA) is one of my favourite disciplines and getting to present on the subject for the conference was something I’d looked forward to since the previous years event.

The conference was hosted in Paris, at the Paris Hilton, just off the Sienne – it was hosted by the Open Group, and I was kindly asked to present by John Spencer, the recently retired Director of The Open Group’s Architecture Forum, and once I was there I was warmly welcomed by Allen Brown, the President and Chief Executive Officer of The Open Group.

As you can probably guess I’m very passionate about EA, in fact any (almost all) of the methods and techniques for getting a contextual, “Big Picture” view of an IT / IS estate ecosystem (what I like to call ‘Macro IT’) – and I’ll be writing a little more about some of these methods in latter posts.

As an industry we’ve been pretty successful to date about understanding and evolving the knowledge and experience around both Infrastructure and Applications. What we’ve been much less successful at, and I believe that it’s due to demand (and compromised by time to deliver), is understanding how these technologies, are put together as systems in a holistic manner, and how those systems themselves co-exist in Enterprise IT eco-systems (and to an extent how they compete with each other as ‘functional’ entities).

Last year when I presented at the Enterprise Architecture Practitioners Conference, 2006 (hosted in London), my slides were much more visual, and focused on how EA was being used at Sun, they were well received, but the comments afterwards were that they were too ‘high level’. So this year I decided to focus much more on Case Studies where we had been involved with Customers EA programmes. The change in direction must have worked because I got a half room standing ovation – the first I’ve received of that size (circa 200 there – I was still on a high about it for at least a couple of weeks).

Because of this there are two version, one with my speakers notes attached and one without – I’ve also created a ‘blog post from the slides (including the speakers notes) which I’ll put online tomorrow.

PDF with speakers notes

PDF without speakers notes

Last years slides are also available (“Enterprise Architecture Practitioners Conference 2006”), and are also on the ‘Architecting the Enterprise’ web site.

Related Links:

And finally…

…after much comment – from workmates, colleagues, friends and acquaintances (but not, I might add, from family members) – as to the lack of a blog of any sort, and as the peak of the blogging phenomenon apparently arrives on the collective horizon, I’ve decided to go and start one.

According to Gartner (here), and reported by a number of my favourite news sites (here, at the Register, and here, at the Inquirer, for example) the growth of blogging will finally peak at around 100 million blogs by 2007. This is part of “the Gartner top ten predictions for 2007 and beyond” – and so where better to start than at a short review of these predictions. I have to admit that I’ve been asked to report on technology trends, make market predictions and even, occasionally, present on futurology – and find the subject of trying to second guess the future as fascinating as it is frustrating.

When it comes to predictions such as this it’s important to bring your own views and opinions to bear, and in regards to objectivity, I like to recall this quote from Warren Buffett: “Forecasts usually tell us more of the forecaster than of the forecast”.

  1. Through 2009, market share for the top 10 IT outsourcers will decline to 40.0% (from 43.5% now), equalling a revenue shift of $5.4 billion: Gartner believes that as the market share for outsourcers declines (based upon current trends) there will be consolidation of outsource businesses and some closures. This is an extremely common prediction, in these circumstances, and is almost always true of maturing markets – frankly I can’t see how you could get this wrong – although it’s good to be reminded about the situation of the outsourcing market, even for those of us who are living with it’s effects everyday.
  2. Only one Asia/Pacific-based service provider will make the global top 20 through 2010: Because of the relatively small number of “global players” in consulting coming from Asia it will limit the ability of the Asian economies to grow revenue streams quickly and become global leaders. I find it really difficult to see how the Asian economies will be limited in growth – I can see them simply buying into the areas they need to, when they need to – some of them already make significant revenues from IT offshoring, and building consultancies out of these organisations is very likely, for instance, Tata & Wipro both have consultancy departments which are growing and maturing fast.
  3. Blogging and community contributors will peak in the first half of 2007: Trends in the average life span of a blogger and the current growth rate of blogs means that there are already more than 200 million ex-bloggers, therefore, the number of bloggers will peak around 100 million mark circa the first half of 2007.Although significant growth in bloggers may have slowed down, I believe it will settle down to more natural growth, plus I don’t see how this has factored in the increasing Internet population, nor the increasing world population. Growth may bottom out for a time but I would then expect there to be slow and progressive growth, much more based on the increasing number of people getting adequate access to Internet resources, fuelled by the overall growth of the world population. And that is without considering the next MySpace phenomenon, where massive expansion unexpectedly grows the overall blogging population.
  4. By 2009, corporate social responsibility (CSR) will be a higher board – and executive-level priority than regulatory compliance: regulation has become a key issue for government and the corporate world, with the aim of ensuring more-responsible behaviour. However, the need for companies to be socially responsible to their employees, customers and shareholders is growing as well. The future will see corporate boards and executives make this social dynamic a more-critical priority. Collective awareness of corporate responsibility is growing, and not purely around green issues. How corporates treat people and conduct themselves in the wider world has needed focusing on for a long time. Plus I believe that CSR will be good for businesses, related to game theory, specifically the Nash equilibrium (developed by John Nash), where choice of optimal behaviour when costs and benefits of each option depend upon the choices of other individuals. Working with a community finance charity, I know that there is already research into this area, which suggests that this is the case – this is another topic I hope to approach in the future of this blog.
  5. By the end of 2007, 75% of enterprises will be infected with undetected, financially motivated, targeted malware that evaded their traditional perimeter and host defences: the threat environment is changing — financially motivated, targeted attacks are increasing, and automated malware-generation kits allow simple creation of thousands of variants quickly — but our security processes and technologies haven’t kept up. Security in IT is an absolute arms race – and this is not going to change – but whilst money can still be made from the sector, legally in defence, and illegally in offence, we will continue to see a strong and healthy IT Security market. This is not to say anyone should be complacent, as the problems at TK-Max have just shown, we have to be rational about these issues, and approach them in a sensible manner.
  6. Vista will be the last major release of Microsoft Windows: the next generation of operating environments will be more modular and will be updated incrementally. The era of monolithic deployments of software releases is nearing an end. Microsoft will be a visible player in this movement, and the result will be more-flexible updates to Windows and a new focus on quality overall. Microsoft have always been able to move there business empire juggernaut into new areas – however they are very often not the first, nor the second , but have often manipulated the market by having the largest installed base volume – look at how slowly they embraced the Internet for one (for instance, Microsoft didn’t even make the first 100 Internet Domain Name Registrations, more here, Sun are joint eleventh, b.t.w.), but when they did it – they did it. Although there have been reports of ramp down of Windows Operating Systems staff, I very much doubt Microsoft are ready to let go of exploiting that installed base to the maximum – there are still a lot of PC users not yet connected to the Internet and they will need a distribution mechanism to support those users.
  7. By 2010, the average total cost of ownership (TCO) of new PCs will fall by 50%: the growing importance and focus on manageability, automation and reliability will provide a welcome means of differentiating PCs in a market that is increasingly commoditised. Many of the manageability and support tools will be broadly available across multiple vendors. However, vendors that can leverage these tools further and can graduate from claims of “goodness” to concrete examples of cost savings will have a market advantage. I’m finding this hard to believe: three years to see new PC TCO fall by 50%. Most people agree that the majority of cost is in systems management, and even with the expected growth in systems management tools (and the ilk), it doesn’t approach issues such as increasing functionality forcing TCO up. Nor the logistics of having staff support a distributed PC estate. Having helped manage the maintenance of a large, distributed PC estate, I know that one of my largest costs were the staff required to physically go around the buildings in question, often the PC required direct support, or for some reason would not be connected to the network (either off, broken, network / network card failure, not loading drivers, etc.). I believe that Thin Client and Ultra Thin Client (UTC) technologies like Sun’s Sun Ray, which have a centralised server model, most readily approach these issues, and that there will be an increase in the acceptance of what used to be referred to as Network Computing (NC).
  8. By 2010, 60% of the worldwide cellular population will be “trackable” via an emerging “follow-me Internet”: local regulations have arisen to protect users’ privacy, but growing demands for national safety and civil protection are relaxing some of the initial privacy limitations. Marketing incentives will also push users to forgo privacy concerns, and many other scenarios will enable outsiders to track their users. Another believable prediction, although given the fact that this is “60% of the world wide cellular population”, I feel it might imply that this will be in what used to be called the “first world”. The issue of evolved versus planned systems play into this, and we all need to think more about what the implications are, so that we can put in place the best governance model available.
  9. Through 2011, enterprises will waste $100 billion buying the wrong networking technologies and services: enterprises are missing out on opportunities to build a network that would put them at a competitive advantage. Instead, they follow outdated design practices and collectively will waste at least $100 billion in the next five years. I can well believe this prediction: having reviewed, audited, and helped close down a number of very large IT programmes, I am sadly becoming de-sensitised & acclimatised to the amounts of money which are needlessly lost, however this does not mean any of us should be dismissive about the massive amounts wasted. In fact I’m surprised it’s estimated as low as $100 billion.
  10. By 2008, nearly 50% of data centers worldwide will lack the necessary power and cooling capacity to support high-density equipment: due to higher densities of processors proliferating, problems in this area continue to grow, and although the power and cooling challenge of high-density computer equipment will persist in the short term, a convergence of innovative technologies will begin to mitigate the problem by 2010. This is a trend which I see emerging every day amongst our customers, especially those who still host out of large cities, such as London. We have field research into this area, which easily backs this claim up too. However this prediction, doesn’t appear to take innovation, such as processor multi-threading, into account: technologies like Sun’s Niagara, and Niagara 2, will have a significant impact on this area, due to the massive decrease in compute footprint. This is an area which I’ll be exploring in more detail over the next few months.

At first I’d thought my title for this post was to do with my finally getting a blog online – but I think it’s as applicable to some of the predictions above. In the next few posts I hope to be looking at the UK Government IT ecosystem, macro IT estate issues & definitions, some more about me and what I do at Sun, and maybe just a little bit about motorbikes.