Monday, December 29, 2008


I used to read a book while watching TV. I sometimes try to read a magazine while watching TV but I end up not being able to concentrate well on either. I thought this was just an age thing. Well, it seems it isn't. According to Barbara Shelly's recent post:

Multitaskers think they can focus on more than one thing at the same time. What's actually happening, researchers say, is that parts of the brain are switching from task to task, losing time and focus with each pivot.

With certain exceptions - practicing Buddhists come to mind - people have always valued the ability to do more than one thing at once. Knit and watch TV. Cook while listening to music. Drive and talk. But in the last two decades, devices that were supposed to make us more efficient and connected have actually made us more attention-deprived and anti-social.

I have to say I agree with this wholeheartedly. I hate it when somebody is texting while in a meeting, or checking their Blackberry. It's actually rude. Now, I have research to back up the fact that it is not efficient either!

Monday, December 22, 2008

Owning Clouds

The new tech war is on for the Cloud operating system. Current leaders are Amazon and Google. But Microsoft are getting serious. Will they be too late? Planned release for their Azure platform is end of 2009. They weren't first in most of the other areas they have come to dominate so this battle is going to get really interesting.

Who else will enter the war? IBM, Oracle, SAP, Cisco...

Is actually a contender for a leading platform or really just a good application solution?

Sunday, December 21, 2008

Startups - some hope

Another follow up from my recent posts on startups here and here, I decided to open my Dec '08 issue of Wired magazine which had been lying around on my desk for a while. Saw an interesting article by Daniel Roth -- "Back to the Garage - when the economy is in turmoil, the time is ripe for ambitious innovation".

He certainly makes a compelling argument. Here's some food for thought. Some of the major tech companies and when they were founded. Not exactly in major boom times.

Google - 1998
Yahoo - 1994
Siebel Systems - 1993
Cisco - 1984
EMC - 1979
Oracle - 1977
Microsoft - 1975
SAP - 1972

Sequoia on startups and the economic downturn. Interesting view from the investor perspective.

Friday, December 19, 2008

Interesting approach to big bank bonus problem

Just saw over at an interesting post on how Credit Suisse is handling the problem of giving their senior executives bonuses or not. Give them bonuses in the form of the toxic mortgage-backed assets that lie at the heart of the credit crisis! I expect this approach to be cloned by other financial services institutions in the not so distant future...

Thursday, December 18, 2008

Hire the best

Following on from my recent post on whether now is a good time for startups or not, I was reading this blog from Seth Godin about hiring the best people.

He is so right. I've seen so many examples recently of companies going for ease of hiring in these tough economic times rather than looking for the best people. They post a job, wait for loads of CV's to show up, and then try to match a CV exactly to the job spec they posted rather than looking for that unique person who will take their organisation to new heights.

During my career I have always found the best hiring approach is not to hire somebody making a sideways move (i.e. doing the same job in a different company) but rather to hire somebody looking for a new challenge. These people will want to prove they can do the job, they will work harder, they will have more loyalty to you and your company, and will be better employees all around.

Don't hire for mediocrity or average, hire the best.

Wednesday, December 17, 2008

Roofless data centers!

Check out Microsoft's new roofless data center designs or as Nick Carr says "trailer park computing".

This is certainly an interesting twist on the whole mega data center build-out currently going on - particularly by Microsoft itself and Google. It also turns the green IT challenges on its head by building it into the design.

Tuesday, December 16, 2008

Good time or bad time for startups?

I was reading some of Michael Arrington's posts from the Le Web conference and this post triggered a question in my mind - is now a good time or bad time for startups?

Arrington makes this statement:

But Europe’s persistent background pessimism was out in full force, even at an event full of entrepreneurs. Americans dominated the stage and spoke mostly about the tremendous opportunities that arise in down markets. Engineers are much easier to hire. The press have fewer startups and stories to divide their attention. The pond certainly gets smaller, but there are far fewer people fishing, too. For most startups, this is a time to blossom.
Hmmm, as a European (but one based in Dublin, Ireland which tends to mean I have a split US/Euro perspective on things) I can see where he is coming from but can also see some of the downside.

Yes, it is true that there are tremendous opportunities in a down market and engineers are easier to hire (but I'll come back to this point in a minute), and press coverage is probably easier to get. However, if you are well funded and can ride out this down market it's a great time. If you are in the position of raising new money either to get started or for follow-on funding then you won't have an easy time. Doom and gloom persists in the world of finance. Forget trying to get a loan.

On the question of Engineers. Any talented engineers already employed somewhere are unlikely to make a move to a startup unless it has serious funding behind it. Granted, there are many seriously talented engineers on the market because of this downturn so they should be easier to hire and probably for reasonable packages (but don't try to underpay because as soon as the market picks up they will jump ship).

So, the opportunities for startups in this market are most likely related to their funding situation. Well funded = good times; Poorly funded = bad times.

Sunday, December 14, 2008

Something to think about this Christmas

Something to think about this Christmas. Check out this YouTube video. It's also a fantastic example of how to give a great presentation.

Emerging Cloud utilities

One of the noticeable aspects of today's SaaS (Software as a Service) world is that most companies going this route are taking a solo approach i.e. hosting a version of their software on the web and selling direct to customers as an alternate channel.

This is a good starting place but it still creates problems for the customer, particularly if they utilise multiple SaaS solutions, around seamlessly integrating these services for use in their business. As many IT professionals already know, the integration challenge is already tough enough for applications hosted inside organisations. This new SaaS model just adds to the complexity. Granted, SaaS or Cloud solutions also create huge advantages but the industry needs to seriously consider this integration challenge.

Integration is not just about integrating web based applications using something like OpenID but also considering hybrid models where some solutions will be provided using SaaS and others will still be hosted internally.

If you agree with Nicholas Carr's viewpoint in his latest book "The Big Switch: Rewiring the World, from Edison to Google" that managing IT environments will migrate from being managed directly by companies themselves to a utility model similar to the electric grid, then these challenges around managing hybrid environments (SaaS and in-house) will only be temporary. I agree with Carr but I don't believe the switch will occur quickly. How long will it take? My guess is probably 10-20 years for a complete transformation across all industries to pick up all the laggards. Some companies will make the switch much earlier than that - in fact, there may be a competitive advantage for SME's during this period where they can gain efficiency over the big guys.

So, who will manage these new utilities? Will it be the current utilities (most likely the telecoms related ones) just expanding into this new area? Will it be the Amazon's / Google's / Microsoft's of the world? Or will new utilities emerge? The most likely scenario is probably all of the above - "horses for courses".

Thursday, December 4, 2008

All about Google

In case you missed it on TechCrunch.

Everything you always wanted to know about Google but were afraid to ask

Discusses some of Google's key success factors including Scalability, Network Effects, Data Mining, Openness, Cocreation, Business Model.

Wednesday, December 3, 2008

ITIL v3 and the Cloud

First off, for those not familiar with it. What is ITIL? Here's the Wikipedia definition:

The Information Technology Infrastructure Library (ITIL) is a set of concepts and policies for managing information technology (IT) infrastructure, development and operations.

ITIL is published in a series of books, each of which covers an IT management topic. The names ITIL and IT Infrastructure Library are registered trademarks of the United Kingdom's Office of Government Commerce (OGC). ITIL gives a detailed description of a number of important IT practices with comprehensive checklists, tasks and procedures that can be tailored to any IT organization.

I didn't think much of the early versions of ITIL. I thought they were too techie focused. However, the latest version (v3) is a different ballgame. It starts to focus on the whole area of Service Management. This includes Service Strategy, Design, Transition, Operation, and Continuous Service Improvement. All critical components for running a SaaS or Cloud Computing environment.

If you're thinking of using services from the Cloud or building your own, or even for services running within an enterprise, you could do worse than reading the ITIL v3 books. From experience, I've seen how the methods, guidelines and processes can make a huge difference. I believe a sensible approach should be taken to its use rather than blindly following it word for word. One great thing about it is that it introduces a common vocabulary which is half the battle.

If you only get one thing from ITIL it should be the Utility and Warranty concepts.

  • Utility is perceived by the customer from the functionality (or attributes) of the service that has a positive effect on the performance of tasks associated with desired outcomes.
  • Warranty is derived from the service being available when needed, in sufficient capacity or magnitude, and dependably in terms of continuity and security.

In many environments, too much emphasis is put on the Utility side (i.e. the functionality) of a software service or application and not enough on the Warranty aspects. In the world of SaaS and Cloud computing, the warranty pieces are just as important, if not more important.

Tuesday, December 2, 2008

Some apps not suited to the Cloud

A lot of thinking is going into how all IT services are going to move into the cloud. I tend to agree that the economics of Cloud computing are too great for organisations to ignore. However, I'm starting to think some of the commentary and services being provided are similar to those we had around the .com boom. At that time, everything was being turned into a .com, even when it made absolutely no sense. I fear the same is happening with Cloud computing. Some services don't make sense in the Cloud, or at least the way they are currently being considered. For example, Office apps may not make sense out in the Cloud. Even Microsoft are jumping onto the bandwagon.

I think it makes total sense for documents, spreadsheets, presentations etc (i.e. the data) to be stored out in the Cloud securely. I'm not as convinced that running all the apps themselves in the Cloud is such a good idea. A hybrid model will be best in some cases - even though this will be a more complicated service to create.

At this stage I think we all have to face the fact that the MS Office suite is pretty much a de-facto standard - even with Microsoft messing with the user interface in Office 2007 to make the learning curve interesting again! I don't have the purest view that some others have around the Cloud where everything runs inside the browser and must be hosted remotely. I don't mind thick-client apps when it makes sense. In certain cases they make total sense i.e. photo, video, sound editing and I would argue Office-type apps (particularly for the power user types).

Where I think the Cloud could add a huge amount of value is helping make this type of a hybrid environment (Cloud and thick-client apps) better is around support services for upgrades, patches, helpdesk, training, pro-active maintenance (i.e. fix it before you knew it was broken) etc. I know some of these services already exist but they could be so much better, particularly if they were provided as an integrated service. Think of the potential around SME services and corporate services in this area. In fact, in some cases the software could even be given away and revenue driven from services.

Most organisations rely on IT support services companies of some sort to help them with their IT environment. All of these companies need to start thinking about moving away from "man in van" onsite support and start looking at providing real Cloud based managed services (including infrastructure and application support) - not the services they currently call managed services.

Monday, December 1, 2008

SaaS and the network risk

One of the more under talked about areas in the whole SaaS and Cloud Computing discussion is the transfer of risk to the network. The services provided from the Cloud are typically more robust and scalable than those hosted in-house. Medium to Large organisations will have their own LAN's and WAN's designed to allow all their employees access to corporate services. Essentially, the network is under their own control.

With SaaS and the Cloud, the networks are all publicly provided. This certainly simplifies the connectivity and management side of things. However, the risk to the provided SaaS service is now in the network. If connectivity to the network is lost (from any node), those at that node have no service. I know this is stating the obvious but the point here is the types of services being provided - they're starting to become more mission critical. This means the accessibility requirements go up. It's not good enough for an organisation to have a five 9's up time from the SaaS provider if they can't access it.

For critical services the configuration of the network at each node now needs to be considered. Backup links, automatic failover, connections from DR sites etc start to become critical. Organisations using SaaS and Cloud services now need to seriously look at their own networks and telco providers and in particular their connectivity to the Internet. This includes access for teleworkers and other remote workers.

I've seen some innovative offerings from mobile network providers recently for SME's (Small, Medium Enterprises) who are starting to offer broadband packages and routers with a backup 3G network connection. If the DSL line fails, the router automatically switches over to the 3G link. It might be a slower speed but at least you still have access. When WiMax starts to get more pervasive the service options will increase even further. These types of services could also be ideal for large corporates to give to remote / tele workers.

If you are seriously considering utilising Cloud (or SaaS) based services for critical corporate applications, you need to have a good long look at your current network and your Internet service providers. Do a risk analysis and consider all mitigation options. You may need to spend a bit more throughout your network but it should be thought of like insurance.