February 16, 2007

New blog!

Posted in Uncategorized at 11:01 am by jayfresh

I’ve decided to continue this blog at a new address:

http://jayfresh.wordpress.com

I thought it was about time I went with a snappier pseudonym… 😉

January 5, 2007

Software Development as a Service (SDaaS)

Posted in Uncategorized at 12:46 pm by jayfresh

For me, Software as a Service (SaaS) has two facets:

  • A full exploitation of multi-tenancy in software allowing an emphasis on collaboration, synchronisation and community
  • An abstraction from the labours of software administration and deployment

This change in the consumption of software must be echoed in the production of software, so that we can think of an ideal of “Software Development as a Service” (SDaaS). This mirrors the driving forces behind SaaS:

  • A diverse marketplace for IT skills is unified in a community of support, re-use and competition
  • The infrastructure necessary for software development is managed and abstracted away, leaving developers with more freedom and the flexibility to apply the most appropriate solution to a situation

As with SaaS, which has seen its greatest take-up in the SME sector, SDaaS will be initially adopted by smaller groups of developers, with Enterprise adoption limited to relatively autonomous in-house development teams. In step with SaaS, as trust grows in the security and capability of hosted-software services, SDaaS will become the natural choice for developing these type of applications. At the extreme limit, large IT contracts will be awarded not to giant integrators running bespoke solutions, but to companies who can deliver at a fraction of the cost by exploiting the skills and infrastructure of the global marketplace.

November 29, 2006

Service Delivery in a Web 2.0 world

Posted in SaaS, service delivery at 6:06 pm by jayfresh

Wow, this is a big problem for business right now… So let’s say that you have a whole bunch of applications, web services and other consumables that you want to sell to your customers, or re-sell through your channels. Doesn’t it make sense to have a single, consistent interface for the users to manage their accounts and purchase items; for managers to re-package products, set price points, create special offers and go so far as to create entirely new products from existing sub-products?

What about metering, fault reporting, usage metrics, linkage to billing systems (not mention the countless other Enterprise systems you might want to link to)?

This single, consistent interface is what I tend to think of as a “service delivery layer” in a software stack. Sure companies have been creating these things for years, particularly since the days of ASP in ’99, but the landscape has really changed and new solutions are required:

There is a need to consume third-party services, to offer them back as hosted applications, to allow developers to upload their own applications into the marketplace (Salesforce-style).

There is a requirement for simplicity – it’s no good if it is a technical, manual job to onboard new products, to configure, to release new products or offers based on existing products. This is a world where competitors with the right systems will be able to react instantly to changes in the market and their competitors product releases.

There is a need to offer some sort of service-level agreements, but the delivery of this service should come from the service delivery platform, not from the application itself. The same goes for all the commercial wrap around the products. The developers need to focus on their applications and should not have to worry about pre-integrating with billing systems and supporting different levels of service natively. In the world where third-party applications and services are to be “mashed up”, this conformance from the developer cannot be expected.

Companies like JamCracker, HP, Cerillion and Mircosoft are creating platforms to enable all of this, but they are not going about it in the same way by any stretch of the imagination. The companies who will win here are those that build modular, extensible, flexible environments, where the hassle of onboarding a new application into the service layer is as minimal as possible, even nill. The company that wins will be the one with the best understanding of the business problem they are trying to solve and where this market is going.

In many ways, this is a step towards future-proofing product development, as we cannot know where the Internet is going to take us over the next months and years. What we do know is that consumers and businesses are going to expect more flexibility, personalisation and ease-of-use. We have to bring these qualities to the touchpoints with these customers or we will be beaten down by faster-moving competitors who take the best of global innovation and sell it on under their brand.

October 31, 2006

JotSpot & Google – business unusual

Posted in development, google, jotspot at 7:09 pm by jayfresh

JotSpot has been bought by Google for an undisclosed amount, so what’s this all about? I guess it’s not obvious – Om Malik is even running a poll on his blog asking whether this makes sense…

I have been a Jot customer for nearly a year now, and I have some very particular views on where their strengths (and weaknesses) lie. I have very little doubt that Google hasn’t bought Jot for their userbase (60k+ users, circa 1k corporate) or for the applications that they have released (let’s face it, online spreadsheets and family trackers are hardly revolutionary). The one thing that Jot is doing very well is advancing the field of DIY applications.

Blogging brought DIY publishing to the masses, along with wikis – they created an explosion in content creation and a huge incentive for software developers to innovate around the mould, either off their own back or in response to users’ feedback. If JotSpot (and their ilk – check CogHead for example) do the same thing for applications, what does this mean for consumers and businesses worldwide?

1. We could cut bespoke, stove-piped development – JotSpot has developed a toolset that lets you rapidly roll out applications without requiring heavy technical skill. Whilst the scope of the applications is currently around information management (which is a massive field anyway), fast forward six months to a year and we will be looking much more at complex Enterprise applications created piecemeal by non-technical business users.

2. Much more customization and personalization – part of the reason why business software often sits on the shelf is that one piece of software will not satisfy the individual requirements of lots of different departments. Salesforce.com know this… A core facet of the JotSpot platform is that the applications you create from their toolkit tend to be very focussed on collaboration, personalization and the user’s individual experience. You can build applications that share a common set of data and layout, but can be customized for the individual, very easily.

3. Lastly, the whole DIY element means that the volume of potential developers goes up by an order of magnitude, from those with deep technical understanding to those with an understanding of the business logic, which are often the people with the requirements in the first place. This frees up developers to create innovative capabilities and respond more generically to their customers, adding to the whole platform rather than one application.

I’m sure it will be a little while before we see anything filtering down from Google, although it has to be said that their decision to make JotSpot free from this point onwards will make a lot of people happy (including me). I’m sure that as a combined force they will be raising the stakes in SaaS, application development and the Online Office, and probably other areas as well. Keep it coming!

October 30, 2006

Virtual Realty

Posted in SecondLife, Uncategorized, virtual worlds at 1:34 pm by jayfresh

There was an article in the FT today about real-world rights to assets in virtual worlds, such as Second Life.

I was just talking about this last night and earlier this week. It makes a lot of sense that in our increasingly virtual world, abstract entities on the web are equally valuable and important as those things we can actually touch. In fact, we’ve been used to this for years and years – as soon as you put money in a bank and monitor it, move it and spend it online or over the telephone, you’re dealing with virtual entities. The stockmarket is another example, and investments such as FTSE trackers take their worth from observing the index – they never even own any shares!

Clearly, we are completely used to dealing with abstract representations of worth, so what is the big problem with putting value on, say, property in virtual worlds? There are people making their living buying and selling on Second Life, and GigaOM’s Om Malik even thinks that eBay should set up retail outlets on Second Life. As hinted in the FT article linked above, the big problem is probably the fact that worlds like Second Life were set up as games, and the mainstream is not going to be comfortable with putting their cash into investments that could arbitrarily disappear at the whim of games publishers or other users – what good is a virtual villa if it can be destroyed tomorrow by a wizard’s whirlwind?

My feeling is that there is a lot of potential here – the world is going online, and there will be increasing numbers of investors who will see the appeal. Two things strike me as significant: First, if a games publisher codifies the rights of individuals in a virtual world as equivalent to those without, although there are obivous legal hurdles to get over. Until this happens, the risks of investing in these worlds are likely going to be inflated. Second, when virtual worlds, which are essentially closed, start integrating as Om suggests, with the rest of the Internet. There is already a massive eCommerce intrastructure on the web, and if these worlds can leverage what already exists in terms of user-base, brand, and popularity, there could be whole new trading channels appearing. Less commerce-focussed, the concept behind virtual worlds seems like a natural extension of social networks, with each user’s place in a community taking on a more geographic meaning.

If you’re thinking of getting into this whole virtual eBusiness thing, you might want to give these guys a call – MillionsOfUs – the world’s first virtual management consultants?

[EDIT]

There is a hell of a lot of stuff in today’s paidcontent bulletin about online gaming, including a mention of gaming company OGPlanet launching an Asian game, Albatross18, in the US, where it is free to play but you pay real money for virtual items such as clothes and accessories.

Don’t miss the large report on Second Life…

“We think the reason [for all the press interest] is that Second Life is perceived as having moved from being a game to a virtual development platform. This is what we thought the internet was going to be 10-15 years ago – a collaborative space that’s both social and creative.” – Justin Bovington (Founder, Second Life)

October 23, 2006

IdeaBrew – a storm in a Hogshead

Posted in IdeaBrew, Web2.0 at 5:41 pm by jayfresh

First meeting of IdeaBrew last Wednesday, gathered in a local over nachos and platters of snacks.

James Shi is running this initiative to construct a Web 2.0 ideas factory – now that he works in Al-noor’s business office we have a direct line of sight to the big guy.

Around ten or fifteen people showed up – new OneIT graduates, guys from the business office, a couple of other interested parties. We’ve been throwing ideas around on an online whiteboard for a while, and we had a couple of presentations about these and some general brainstorming afterwards.

We’ve now got three major threads to develop: extending the BT employee directory into a social network; creating an internal collaboration site (not unlike BaseCamp…); creating a closed-loop-control system for the CEO’s business office.

These sound pretty cool, especially if we can build common functionality to underpin them all. I think that’s an important theme in all this lightweight Web 2.0 stuff – using the websites as views on a common pool of data.

October 18, 2006

DIY-web going solo

Posted in salesforce programming apex at 10:48 am by jayfresh

I’ve been interested in the DIY-web for a while now, by which I mean sites where you can create your own applications rather than just content e.g. JotSpot (http://www.jot.com), CogHead (http://www.coghead.com) and others. When you look at what DIY-publishing did for the web, in the form of blogs and wikis, it is very exciting to think about what will happen when DIY-app creation becomes as easy and mainstream as publishing is now.

I am writing this post after reading about Salesforce.com’s announcement of their Apex programming language, which in theory will give developers unlimited access to the resources and capabilities of the AppEx platform, letting them create native AppEx apps.

http://www.computerwire.com/industries/research/?pid=F1F115DF-6942-498E-93DC-48809F66475A

This is interesting not because it will make things simpler, but because it shows that Salesforce is serious about positioning the web and its AppEx platform as a development platform of choice. Instead of people saying that they write Javascript or .Net or Java applications on the AppEx platform, they will just talk about writing Apex applications.

[EDIT]

I’ve just watched a couple of demos of some of the things you can do with Apex. The language looks very complex (it looks a lot like Java), so this is definitely not a step in the direction of simplicity. However, because the language is well integrated with the platform, any package of functionality can be exposed as a web service or via the API. This means that developers can build in their own widgets or tools that can become part of AppEx applications.

They are keeping a very close eye on your Apex usage metrics , which would put them in a good position to charge for usage if they wanted to.

[FURTHER EDIT]

Note the under construction message at http://adnsandbox.com/, which is where the Apex presentation movie is held.  Is Salesforce constructing a web-based development environment, a la Laszlo Explorer (http://www.laszlosystems.com/lps-3.0/laszlo-explorer/)?