Category Archive

Last week I was asked for Qhuba’s opinion on Green IT.

We hadn’t approached the subject for a while, the first time we did we simply thought carbon compensation was the best way to go, planted a few hundred trees (well, 1800 to be more exact) and just moved on.

Now we’re being asked more and more what we advise companies to do if they want to do green IT.

Here are three less conventional things you MUST do.

1 - Put in your policies that you will virtualise EVERYTHING you can, but only when opportunity truly arises

Why? The math is a little complex. If you have 3 small machines that you virtualise into 1 bigger machine - you save power right? Yes, true, but if you factor in the environmental cost of creating a new machine (and throwing away the 3 older ones before they are truly done with), it doesn’t make sense to pro-actively virtualise.

2 - Move to nuclear power. (Oops, we said the ‘N’ word).

Why? At the moment, whilst we are waiting for greenhouse gas problems to be resolved, the fastest thing you can do to reduce your carbon production is make sure that you are not taking power from coal or gas-fueled plants. Go Hydro or Nuclear if you can.

3 - Move to SaaS.

Why? Putting your software in the hands of a (very) large data centre makes sense from all kinds of angles. Particularly where they are building hydro plants right next to the datacentres.

You’ll note there are lots of other things we didn’t mention. As a company we believe people should live close to their work or work from home as much as possible. This doesn’t only cut down on the travel costs (both environmental and cash versions), but increases their own personal productivity and enjoyment. Fewer traffic jams = fewer frustrated people, and of course less bad emmisions.

When we planted our trees we made sure they were on places where we are sure they will live at least 60 years long, and not be cut for timber after 20 or so years. You know of course that the first years of a trees life they actually produce more greenhouse gases than they absorbe…

So do Green IT in a variety of ways - but please don’t go and buy new equipment to do it with - that doesn’t make sense.

For a long time CIO’s and other IT folk have been forced to start pushing the business benefits of changes. This is a good thing. We need to focus more on the benefits of doing IT than on the costs.

BUT… There is always a BUT.

Yesterday a colleague asked advice for an organisation that’s moving into open-source and is looking to the next step. (We at Qhuba believe in a very slow - but sure - move to open source).

They are looking to justify the next step. Of course there are the obvious items such as reduced license costs, but to truly sell it to the business they are looking for the ‘business case’. So the question arises: What is in IT for the business?

Well, painfully, not very much. At best you hope - like with any migration - that the business will just keep on running after you’re done. After all, new IT tools are just that - tools. They do not trigger the step-changes that can be made if someone truly improves a business process.
These tools are no doubt essential to many of those changes, but they are not the catalyst.

The best you can hope for is landing up with much of the same. Any benefits such as end-user productivity and reduced TCO have been achieved many years ago. All we can focus on when doing projects around the bread-and-butter IT services is reduction in cost or effort from the IT department.

An that kind of brings back the world of “what’s in it for me, the CIO”.

We’ve been longing to stay away from that world, and now the only way we can justify the benefits are by again turning the focus on our own (IT) cost and reducing them.

This in itself though should be the wake-up call for CIO’s. If there is no further gain to be had on the business side - we should call it a day with IT improvements in those areas, and just stop doing IT there altogether.

Yes - stop doing IT in those areas. Just like most large companies have stopped producing the majority of their energy themselves, so we need to stop producing our own ‘bread-and-butter’ IT services ourselves.

If you’ve landed up asking youreself: “What’s in IT for me?”, by all means go ahead and take the next step - just make sure it’s the last one you force yourself to take - give it up to some SAAS provider, or find another solution altogether.

I’m on the move again. This time it’s not a lifhgt to Rome or the US, but I’m relocating my weekday home to another city in holland.

Why? Traffic.

Incredible really to have this as a reason in todays’ world where we have incredible possibilities for telecommunication. This got me to thinking. We work increasingly globally, and are more electronic-interrupt driven than process driven. So much so in fact that when we find there are interrupt not included in a process, we formalise the process so far that we can be sure the next time it’s required, the correct interrupt will be generated causing us to do some work.

Let’s up the abstraction level for a moment to create a little understanding around the point.

Think about your yearly planning cycle. In large companies nowadays it’s an incredibly scripted process, with the dates of starting, drafting & delivering known well in advance, frequently even the year before. It’s become such a massive process we’ve automated as much of it as possible. In many cases an e-mail reminder goes out automatically to remind people that it’s that time of year again, and they’d better get going with the plans.

Works right? Well…. yes, BUT, is it the best way, or just the way we’ve gotten used to doing it?

Go back to the reasons WHY you want to review your plans. It’s all to do with maximising business benefits whilst minimising costs and associated effort. Off course I’m glossing over a number of other reasons to do it here. But basically it’s just good business (sensible) practice to do so.

So why wait until the year is over until you do it? There is no good reason.

If we can identify things now that would benefit us, then why not weigh up the pro’s & con’s now and cut to the chase? Why wait until just before review time to review how it is you’re doing and what it is you’re doing?

Just as any auditor knows, waiting until the end of a major project to do an audit will only lead to a pat on the back or a flaming of the project, but NEVER to an improvement of the outcome itself. Doing an audit at the start and during however, reduces the risk of the flame, and increases the value of the outcome.

For those who disagree, I challenge that you don’t have enough understanding of your audit departments’ capabilities, and are not getting value for your money in doing so.

So here’s the advice: Let’s move away from scripted business, and move to a much more fluid way of working in which continuous improvement is the driver, not our pre-programmed interrupts. I don’t mean we don’t set goals anymore; goals are essential, but I DO mean we check more frequently (but don’t script this part either) how we are achieving them, if we are achieving them, and if we should still strive to achieve them.

Maybe it’s a process politicians can take when half-way through a war as well. If your objectives have changed, or the outcomes needs to be something different, just stop doing what you’re doing and go and do the other.

Easily said. Easily done?

I’m off to the US soon, to Portland to  the Ubuntu Live event.

Now I have family living in the US, so I thought I’d book a flight from Paris to LA, then do the Portland leg as an internal US flight, stopping in LA a few days on one side of my intercontinental trip.

Easier said then done. It took me quite a while to get the right-priced tickets going at the right dates/times to make changing flights OK. Conceptually, it’s simple. This is what I wanted:

1. Fly from Paris to LA
2. Connect and fly through to Portland
3. Present at Ubuntu Live 
4. Fly back from Portland to LA
5. Stop over a few days, surf a bit, do the Californian thing for a while, then get a flight back to Paris

Now I can’t be the only one wanting to do something like this, but do you think it’s possible in any of the systems like Expedia, Lastminute.com etc? Not a chance. Not only does it want to offer me a full set of nights’stay in hotels in LA (I might want the night before the flight back, but not the rest), but I also have to manually sync up my departure & arrival times.

Of course the fact that I booked the flight for 1 person only, but made 2 seperate bookings means I need to select which itenerary I want, although, to me, it’s just one single trip.

 So I started thinking about all the other things that we have as services that are not quite there yet.

One of the other on-line services I use a lot are the bookstores. I most frequently use Amazon and Barnes&Noble, although I’ll shop anywhere if they can get me the latest Harry potter a day earlier.

Now what really impresses me about those sites is that in most cases (as long as you’re buying from them and not a 2nd hand seller via them), you can chop and change your order, add books or CD’s to it up until the day it ships. I’m pushed to think what more I’d want from them. (I know, out-of-print books - but that’s the hobby thing)

Now how can I have such a good experience with cheap articles (books), and a much worse experience with more expensive ones (the flights/hotels)? Surely there’s something going wrong here. I can’t believe that flight planning is more complex than shipping books, particularly as it’s the airlines themselves doing the difficult part of airplane maintenance. All Expedia needs to do is check prices, availability, and go. It’s not like they need to schedule the flights in themselves (although, wouldn’t that be nice?).

I suppose this is a plea to anyone thinking about providing any online service to, once you’ve built it, actually go and look properly how people are using it. And by that I don’t mean checking out the usage statistics, and seeing how browsing habbits change if you move the banner from the top to the left. I mean actually physically going out there and looking at what people do with what you give them.

We all have different ways of doing things. That makes us more interesting as a human race. Don’t force us into an unnatural pattern because the IT systems work better with it. Fix the problem.

Make machines work more human for us.

Pauwl

Let’s face it, the Germans determine pretty much which car you drive if you have lots of money.

After all, Bugatti and Lamborghini have gone north to Germany from an ownership perspective, to join BMW, Mercedes-Benz and Audi.

And now they’re determining your Browser, next your OS?

It’s true, it’s not Microsoft anymore that determines whats on your company desktop, it’s SAP.

If you run Firefox, try to read the SAP annual report here.
If you’re thinking of running Linux, first ask yourself if you want to be bound to a browser-client or not.

Now I know the browser client has gotten a lot better (fully functional?) in the last years, but are companies OK with leaving the client fully behind?

So what’s the deal? Is SAP in bed with Microsoft now? Why isn’t there a SAP client for Linux, or a roadmap for it? Or have I overlooked some information somewhere?

With the impending release of Ubuntu Linux for other devices (see here) it’s clear a fully functional, viable alternative for tablets, and mobile devices is coming.
But then back to SAP. It seems we’re being locked-in to Windows platforms by SAP.

Time to get on the barricades and see if we can free our corporate ERP’s from the platform restrictions, and really set our users free!

ACTION: Go and find out from your SAP representative what they’re doing about getting SAP going on Ubuntu Linux.

Go now.

Well,

according to Microsoft, there are 230 patent infringements in Linux. But they’re not going to sue about it…. yet (see ZDNet)

Happy now?

No. nowhere near it. Some comments are about Microsoft ‘playing nice’ for now.

Playing nice would be to be honest and say which infringements you think they are, and let the community respond by refuting it, or by simply changing the code and making the problem go away.

But then… You might wonder how many open source license terms the Windows products violate.

Or is it just me?

Come on Steve! Spill the goods! Give us a chance.