Microsoft Global Case Study
Another little thing we’ve been working on over the past couple of months is completing this Microsoft Case Study, with a Redmond based team.
Web-based Accounting Solution Increases Business Growth, Customer Flexibility
I do believe in the Software + Services model. Obviously there is benefit to Microsoft in involving client technology but there are important user benefits as well.
At a simple level, Office is a great example of introducing a rich client. We make sure we can get most data into Excel.
I think in the life cycle of SaaS we are in the early stages where we are pulling everything back to the server into the multi-tenanted SaaS model and implementing new applications over base SaaS frameworks.
Framework services include things like rendering, security, authentication, service points, reporting, provisioning, monitoring and a bunch more things. Communication with the core business logic is implemented in a Service Oriented Architecture (SOA) fashion so it becomes conceptually straight forward to access applications in other ways.
Next year we’ll see the introduction of off-line functionality inside the standard browsers, so new scenarios start to open up. Also I think we’ll see obvious scenarios where having the power of a client makes a lot of sense as well. Say for intensive computational work or for offline useage. These clients will access the server based logic we are all building now. In the browser cross platform client technologies like SilverLight and Flex make it even more compelling.
I think we’ll see more parts of applications delivered for the mobile web or even as mobile applications.
I’m certainly not precious on doing everything on the web but before we go to far in introducing rich clients we need get the applications written for the server first.
So I believe we’re still very much in the early phases of SaaS. The re-architecting and framework development that is going on now build the foundations for all sorts of delivery scenario’s over the next few years.
blog
I really believe that client technologies such as Flex and maybe even Silverlight ;-) will become more and more prominent over the next few years and as such I wonder what effect this will have on back-end implementations and even our approach to architecture.
Adobe’s AIR and the forthcoming off-line accessible web applications soon to be afforded through SQL Lite (supported by the likes of Firefox) will help catalyze this development.
Personally, I have spent the better part of this year building an application platform for my business based on .Net and an HTML/AJAX front end. Through my recent contract work for the MetService I have spent a lot of time getting back into FLEX and made the decision to re-build my businesses’ client software in FLEX.
For me there was a solid business case to make this rather dramatic decision. The question of what to do with the back-end was not so straight forward - leave it in .Net but modify it to respond directly to FLEX via a presentation server layer such as WebOrb, change it into a Web Service / some kind of XML service, etc…. Strangely, I realized that I had a lot more options than I had when the UI was HTML and possibly even more astonishingly I didn’t care that much. Quite simply, I had moved so much of the business logic and functionality into the client (FLEX) that I didn’t really care how the server did it’s thing.
What I realized through my deliberations is that the more power we entrust in the client the less important the server becomes. In our application, for example, the server became nothing more than a gateway to permanent persistence, simple data storage and a utility for the orchestration of other web services.
In many ways I welcome this. Technologies such as FLEX bring a state-full, user-oriented desktop-software-style approach to software design. I.e we build applications based on what we want to offer the user, not based on what we can do. Flex offers the added advantage that it looks and feels good too. Ultimately the server becomes a simple (exchangable) SOA element, nothing more.
The true value of the application is what it does and services it renders, not the technology we’ve invested in to make it possible.
With the direction Adobe and now Microsoft are taking in giving more power to the client, I think we’ll see a rapid increase in the capabilities of web applications and a haziness between what is web and what is desktop. This is good for companies and users alike. HTML was a means to an end. Now that we have new means, we have new ends.
So let’s hope to see a corressponding focus on infrastructure, without which NZ will lag even further behind the world.
Hi Rod
I could use some clarification concerning the software + services model, which you champion so ardently. I downloaded and printed the white paper you posted a while back - and it confused me.
I would like to ( very, very briefly ) outline the architecture of a couple of projects I have completed for Virgin in recent months, and ask you to explain how they would be done using the SAS model, and what benefits that would bring. Would that be ok ? And if so, should I post it here or go offline and use email instead ?
Also, you said “Next year we’ll see the introduction of off-line functionality inside the standard browsers, so new scenarios start to open up….”
Examples ?
Thanks
John
If you happy for it to be in the public domain I’d be very interested to have a look at how what you’ve done would work under a S+S/SaaS framework.
email me at benkepesATgmailDOTcom if you’d like to discuss…..