Web 2.x Blog

I have recently arise from my stalemate of working in the office back into the Web 2.0 blogosphere again. I gave mashup a serious look after a manager in my firm helped me discover Dapper and Kapowtech. The diagram next to it is taken from the premium programmableweb.com, where thousands of open mashable API exists. It shows how mashable APIs co-exists in the mashup ecosystem.

Web 2.0 has come a long way since the old days where web applications merely falls into the “geeks create, consumer use” prism. Today, user-generated content dominate the internet, with the likes of Facebook, blogs and wikis. The IT geeks merely need to provide a platform for which these applications can be sprung to life (i.e. Facebook platform, Google’s new OpenSocial for the matter…). Such emerging trend of use of web applications has prompted the need for fast (with zero deployment time) lightweight applications that are composite/compact. In another words: The User must be given the flexibility to build/mashup what they want, and when they want. (note the capital “u” for user..)

So where does all this fits within the enterprise?

The traditional Enterprise IT application falls the following category :-

1. Pre-built applications that are tailored as a one-size fits all (think of the typical enterprise portal).
2. X number of applications built specifically for X number of users (think of maintenance and upgrade cost when the needs of these x number of uses changes rapidly).

The downfall to the above approach:-

1. The concept of Web 2.0 applications are normally emergent. This means the applications has to evolve and adapt to changing needs in the business world, which is not uncommon. Pre-built application do not “emerge” and incur huge course for IT to customize and maintain.
2. As pointed out by Kapowtech, there is always the long tail of IT applications or information that are locked away in some repositories that are undiscoverable, but when used collectively, produces great business value. The idea of WIKI is based around collective intelligence, and the idea of a Enterprise is not any different.
3. Access and leverage of enterprise information is not always straightforward. As every knowledge worker in the company has specific needs, the ability to mashup specific information to suit specific users will provide great strategic value to the firm as a whole. Information silos, as implemented in the past, do not work in this context.

The pre-requisite technology to enabling mashup would be to be able to “break” the system apart using distributed technologies. This is where SOA plays a bit part as a the enabling technology behind mashup. The broken up specific functions of a IT system are typically exposed as various forms of API that are able to be consumed and aggregated as a information focal point. The discussion on SOA is tremendous, and it is beyond this discussion to explain it.

I am currently writing a whitepaper regarding Enterprise Mashups for my firm, and a link to it will be posted if it gets published. The whitepaper will discuss all of the above in greater detail.

Tags: Mashup, SOA, Web 2.0