Back in December I posted a short piece about whether there really is a future in writing software any more. The piece was driven by some comments made about some research initiatives in new software tools going on in Microsoft & Sun and how they would affect us going forward. While we have a lot to discuss technically here, for the wider audience typified by the readers of this Blog it’s really a case of “So What? When this stuff emerges from the labs, will it be useful to my business?”. The question we are really trying to answer is not about writing software per se, more about whether it makes sense to write software for my own business.

If you try to be logical about it, I’d say that most businesses want to deliver value to their customers that is better than the competition can deliver. One way to deliver that is by having computers support what you do to add value, and that in turn means translating your own specific business needs into something that computing can deliver for you.

The trouble is that it is easy to be logical about this, but we have to start adding in reality before the accountants and bank managers do it for us. Writing software is not cheap, and for the majority of businesses the computing economics comes down to a balance of “Will my business work if I change the operational processes to work with an off-the-shelf piece of software at lower costs?” against “What added value can I get to boost my bottom line by creating something myself that fits my processes?”.

Given that software development can be difficult to run in business terms if you are not able to dedicate the correct resources to it, the pragmatic solution tends towards the use of off-the-shelf packages, but that means that some of your value add can be eroded.

Whilst the words coming from the academics & technologists may not say it directly, there are some exciting innovations coming in the world of software development that will ultimately benefit our businesses, If we can make it easy to create that added value, and make it robust so that we can trust software to run our business processes then we make it cheaper to customise the support for own business models and thus be more competitive.

In December I mentioned the research work going on around building software robustness through better testability & documentation, don’t you hate it when you go to a web site & it is down? The work behind some of these sites is very substantial, and it’s not surprising that some do have problems. Any improvements in testability makes for better customer experiences in the long run.

However that doesn’t solve the fact that we need to make it easy to build that added-value, and one of the most promising areas is the new developments on Domain Specific Languages or DSLs. Putting it crudely, a Domain Specific Language is one where any software is created using language that is familiar to that area (or domain). By doing this, the job of creating the software is made easier as the person doing the work is working in their own context not that of the computer and so development of business solutions and value is made easier; faster & more relevant to the business need.

That is the good news. The not so good news is that the people creating these new DSLs haven’t really got to grips yet with the business context, and most of the DSLs emerging are aimed at the technical world for now. If you’re a mathematician then it’s great news, but if you’re running a plumbing supplies business you’ll have to wait for a little while longer.

Had you asked me back in December how long the plumbing supplies business would have to wait then I would have rolled my eyes, shrugged my shoulders and muttered something under my breath about length and pieces of string. However news of a real live application for a veterinary practice being rolled out using a language called Clojure has given me hope that we are starting to see, dare I say “green shoots of”, emerging business uses of these new languages. It is early days yet, and to be fair Clojure is still quite a technical language, but it’s being applied in a business context and that is what makes it so encouraging.

So will we see the academics turning out something really oriented to businesses which combines ease of use with robustness? If they could do it using Cloud Computing so that you could rent it as you need it then that would provide smaller businesses with cost-effective ways to offer value-add on-line. Could we get that this year? I really would like to think so.

As 2009 starts we are all predicting what we think the coming year holds in store for us, this Blog being no exception. However, there is a trend in the predictions that has grown to full scale articles concerning The Demise of SOA (Service Oriented Architecture). It feels like it was only a few months ago that we were being told how SOA was the next big thing in computing & that we would all be doing it soon.

For those of you are interested, Service Oriented Architecture is about decoupling the software components that we use from each other so that in future we can more quickly mix & match the applications we use to create useful tools that help our businesses.
Quite frankly though, it doesn’t matter what SOA is, it’s the turnaround in the commentary that I’m more concerned about. Most of the commentary I’ve seen focuses on two aspects:

Firstly that take up of SOA principals has been very slow and thus the inertia isn’t building up as expected. Apparently we expected most organisations to throw out all of their systems and completely re-implement their systems using SOA…hmmm? What has actually happened is that most businesses who are using SOA have implemented smaller projects so haven’t realised the full benefits.

Secondly the economic downturn means we can’t afford to re engineer everything to SOA, and cloud computing will take over anyway.

It may be me just being grumpy, but from a business perspective I would find it extremely difficult to justify spending a lot of money on throwing away things that work for my business; and then replacing them with other things that work in the same way SOA or not. Surely that applies to any size of organisation? What makes more sense is that when a business is looking at making a change, then at that point the principals of SOA may be applied by the IT providers, and that most of the business won’t really care as long as what it gets is better for the business as a whole than what was there before.

To me, I’d say that the whole SOA piece is very much alive & well, but evolving. In this way we see that businesses are taking steps to implement tools on a project by project basis which may well introduce SOA over a long period of time, and when the next round of business iteration comes along then the long term benefits of SOA are realised – which funnily enough is where the commentators are saying that SOA has failed.

Our much vaunted Cloud Computing and Software as a Service (“SaaS”) actually relies on SOA to deliver much of what it does, so “Cloud” will replace “SOA” as the visible artifacts that business see.

It’s quite valid for us to want to know about “The next big thing”, but we also need to keep in mind that change actually takes a lot longer on the whole that the commentators would have us believe. In real world timescales we’re still building up a head of steam.

Just because something hasn’t changed lock, stock & barrel in the blink of an eye doesn’t mean it is lost, just that pragmatic business sense is actually winning.

Cloud Computing, Software as a Service, Application Service Providers, Virtual Desktops and Live Workspaces. Just some of the terms which have been bandied about over the past few years. With Windows Live Mesh, now available as a Public Beta as part of the ‘Wave 3′ release of Windows Live I sense these technologies coming of age and are ready to hit the mainstream. Big time.

So what’s it all about and why does it matter?

Live Mesh allows users to access and share their files seamlessly across the web and across multiple devices. So nothing really new there. It also allows users to share applications – again cool perhaps, but not altogether groundbreaking. Live Desktop looks and works in much the same way as your computer desktop (if you use Vista, that is) making it instantly familiar, intuitive and a breeze to use. Nice, fast and funky; but again we’ve seen most of this before.

Scratch a little deeper and some of the real potential of this new platform reveals itself. Live Mesh allows synchronisation between multiple devices (including mobile and macs), and people, providing an open, shared workspace in which diverse and distributed teams can collaborate. A ‘feed’ provides up to the minute information on users activity in relation to the workspace.

The world is changing fast and with the increasing pressures on business to improve efficiencies, ‘green-up’ and adopt flexible mobile working, the reality is that todays combination of email, shared folders and local applications is creaking at the seams. To date, the Achilles heel of the cloud computing paradigm has been its dependence upon ‘always on’ connectivity. So when the WiFi on the train decides to have a rest or my spangly new 3G data card hits its download limit I am, to put it politely, stuffed. However, when you are offline (intentionally or otherwise), Mesh provides a client which allows you to continue work without the connection. Then, when you are back online everything you have been working on can be neatly synchronised.

As it is closely integrated with the rest of Windows, Live Mesh is a logical step towards the ‘hyperconnected’ environment which has been looming for some time. It will allow us to step meekly into this world from our familiar PC-centric environment without the quantum leap of betting our entire working lives on being always connected. I think we will like what we see.

Mesh can be previewed at www.mesh.com

Andrew will shortly be living and working for a week using only his mobile broadband connection and sharing his experience in a diary. Watch this space and wish him luck.

I thought that company directors spent some of their time playing golf with their peers as part of “networking”. However it turns out that the modern IT director, such as ours, seems to think that a good time spent networking with his peers involves discussions around relevant topics like “Strategic Resourcing”. It was no surprise when he returned from a recent event and started asking me about using free open source software.

They had been discussing the effect of collaboration tools, especially instant messaging, on company culture. The CIO of a local utility company had mentioned that they were rolling out messaging using open source products which had saved capital expenditure compared to the commercial offerings. Without morning coffee I couldn’t really think of any cogent arguments either way and by the time I had he was away discussing something else of importance like the role of IT in business transformation.

The question of whether to use Open Source software or pay for commercial products keeps coming up and my answer is still the same as it has always been: It depends!

I’m not an automotive engineer, and when I sit in my car I’m not overly concerned about how it does what it does. I just want it to work. Frankly for most businesses I expect IT to be treated in the same way; concentrate on the business itself and how IT helps to make it work, not on how IT works. If that fits with how you want to do business, then you’ll want IT products to work without deviating your time away from the core activities.  Despite the occasional report to the contrary, software businesses put a lot of effort into testing their products to work well.

Larger companies like Microsoft or Symantec develop & test the processes around installation as well as ongoing use. They also publish specifications for how their software looks when you use it so that if you are familiar with one product, then another should be easier to find your way around as well. Of course this costs money & you pay for that.

That’s not to say that there aren’t some really good software packages in the open source world. OpenOffice comes to mind as a good alternative to Microsoft Office and it is free to use. It looks different to Microsoft Office and some of the features work in a different way. My family experiences with OpenOffice are probably typical of using open source & other free software:
My eldest son got to grips with OpenOffice almost immediately. He found his way around it quickly and used it to produce a lot of documents including some high quality camera-ready brochures ready for professional printing. But then he did study Computer Science for a year at university as part of his course so I’d expect him to be pretty savvy at this stuff.
His younger brother could use OpenOffice but found that the differences to Microsoft office were enough to put him off. He just wanted to get on with working, not with trying to find out how to do something when he’d already learnt once elsewhere. He gave up and bought Microsoft Office as he felt he was faster working with it.
My father-in-law took one look at OpenOffice and declared it so different to Microsoft Office that he gave up in confusion.

I try to think of who is using the software and what they want to spend their time doing. Microsoft Office skills are widely taught now so getting staff going with them is easy. A major organisation, like the one that the CIO above represents, may well have in-house training and support teams that are able to get staff up to speed with other products cost-effectively. For them open source software could offer significant savings, that’s why we see some local authorities taking this route. However for a smaller business without access to that sort of infrastructure then the cost of mainstream products may well be balanced against the ease of using them.

That for me sums up the whole open software product portfolio. If you are prepared to spend the time (and indirectly your money) then you can get some very good products cost-effectively. Most come with no guarantee, and you don’t get a telephone number to call if you have a problem. If that is for you then go for it, just be prepared and recognise that the up-front saving may be eroded by the in-life costs.

As I said, “It depends”

PS. I use open source software at home

My blogging colleague Neil Strenge & I tend, as each week progresses, to watch the clouds. We each have our reasons, but what we share is a desire to foretell what the weather will be during the coming weekend. Unfortunately at that point our wishes diverge, Neil wants to see signs of wind coming whereas what I’m looking for is an indication of quiet weather. In truth it’s not the clouds themselves that we try to read, but the sequence that we see the clouds change that tells us what is coming.

The hot topic in technology circles is “Cloud Computing”; and at the moment we don’t have to look too far to get news. Search most media sites for “Cloud Computing” and the list shows many recent entries on the subject. What is proving far more difficult is reading the forecast for when we’ll be using the cloud, but last weeks announcements by Microsoft are quite important.

One problem here is that the term “Cloud Computing” has different meanings to different people, so when you read another news article on the “cloud” then you need to bear the context in mind. Mention “cloud” and you could be talking about:

·         Hosted computer facilities running somewhere out in the world
·         Development tools allowing programmers to write applications that can be hosted out somewhere in the world
·         Applications software running somewhere out in the world

I’ve made some pretty gross simplifications above, but the key words are somewhere out in the world. In cloud computing you don’t need to know where you are getting the service from, only that you are getting the service.

For a small business, I’d say that it’s the third category of business applications that will be the most useful. Buying software to run my business usually comes with a few added (some might say “diverting”) challenges such as acquiring the compute power needed to run that software, or training staff to use the software and keep it running day after day.

How much easier it is to contract for the service being run for you (often labelled “Software as a Service”), our own hosted email service for example. That way, as you grow the problem of expanding that service is with the provider not with you, you just ask for more seats or more capabilities and it is the providers job to deliver as agreed. That is the promise of cloud computing, the ability to easily scale your business software according to your needs. The problem is that there are not a lot of business software applications available in the cloud right now; first we need the hosted compute facilities and then, more importantly, the software developers to use the development tools to get on and produce the business services for us.

It comes back to reading the sequence of the clouds: we can see the compute facilities for example in the BBC news item broadcast last Monday and summarised on the BBC website. We can see providers such as Google making development tools available as described by Stephen Fry in the Guardian newspaper last month. Now we have to wait for real business applications to come to the market to add to small number of offerings currently around dominated by SalesForce.com. The forecast is looking good, but we don’t know how long we’ll be waiting. That makes Microsoft’s announcement of new facilities under the “Azure” codename last week suddenly look very interesting.

It’s always been a computing mantra to reuse software wherever you can rather than start all over again. The Microsoft Azure announcement offers the software developers a chance to quickly change existing applications to run on a cloud model, and being Microsoft there is a lot of software out there ready to change. Hopefully this means that we will see an acceleration in the number of business applications coming into this market place Watch this space!

Your business ISP ain’t what it used to be. The days when the ISPs job was simply to connect you to the internet and provide some webspace are long behind us. Most business ISPs are now able to offer expert IT management, hosted applications and Software as a Service (SaaS) in addition to the more traditional ‘connectivity’. So what is this all about and what opportunities does it provide for businesses in these difficult times?

SaaS can be attractive to a wide range of businesses from the SoHo (Small office, Home office) startups and SMEs through to larger corporates and public sector organisations. In fact any organisation that seeks to outsource some of its hardware/server requirements, software management and maintenance can now benefit. This can serve to make the bean-counters happy; one of the key advantages of SaaS is that it reduces the need to recruit and retain expensive in-house skills. SaaS allows businesses to outsource the management of key IT infrastructure, concentrating on its core competencies and keeping costs manageable and predictable in times when cost reduction is the norm and restricted borrowing facilities are constraining capital expenditure.

Most of these services are made available on a pay-as-you-go basis. So for instance an employees Microsoft Exchange mail facility can be ‘rented’ for as little as £5 per month per user. What is more as the business grows and reorganises the additional requirements can be accommodated with ease. As most SaaS services are delivered securely over the internet they are ideal for the ever-growing army of road-warriors and home workers who have no permanent office base.

The range of SaaS services is mushrooming. Hosted email services are growing in popularity, following the trajectory of the US where over 40% of email and messaging services are now delivered using an outsourced model. Add to this hosted services for CRM (Customer Relationship Management), Document Management, ERP (Enterprise Resource Planning), and IP telephony and it is clear why many IT departments are redefining themselves with a focus on the delivery of strategic projects whilst the ‘nuts and bolts’ software is handled by specialist service providers. Certain sectors are seeing rapid uptake, salesforce.com for instance is building on its early success in CRM by redefining a new range of ‘Platform-as-a-service’ solutions. No wonder then that this burgeoning sectors is expected to be worth over $10bn by 2009.

The great news for customers is that many of these services are available on a trial basis, many free. So with little or no commitment it is possible to try SaaS for size and see how it fits your business needs.

Things that might be of interest to you: Web Hosting and Email, Microsoft Exchange, Home Office, Large Networks, Business Broadband