|
 RSS Feed
 In the previous blog, I talked about program to program design versus program to people design. Google does this well. For each type, they create platforms for third parties to develop applications (serving the long tail in the process) and allow for customization. Take mashups (see the network figure). This is clearly applications to applications design.First of all, notice that many companies unbundle their functionality and provide them as separate components (googleMaps, GoogleSearch, Blogger etc.). They provide platforms to make it easy for developers to use these components via APIs and toolkits. On the program to people front, Google allows users to customize their homepage using pre-fabricated components that are availabe on Google. This allows third parties to develop widgets and make them available on Google. To encourage this, Google provides toolkits and promotes these widgets on thier website. Finally, through the magic of AJAX, Google allows endusers to customize their homepage by moving and adding widgets to suit thier needs. Couple of highlights. First, encourage third party development of funtionality. Support this through APIs and toolkits. Second, allow for customization and user control.
s
As I was preparing to teach a design class, it occurred to me that we over emphasize the objective of an end user and neglect the role of other programs as consumers. The shift from Web 1.0 to 2.0, and the consequent emergence of the Web as a platform, means that we should look at design from additional perspectives.
In the era of Web 2.0 applications will be aggregated (portals), remixed (mashups) and discovered. Given this, how should you describe API needs and how do you tag applications or reports generated by applications? I hope to emphasize some of these concepts in my class.
Comments (0 )
s
This has been going on for a while. MSFT was the front runner followed by Yahoo and then Google entered into the fray and won. There was a time when MSFT never lost such battles. AOL benefits by getting more ad revenues, Google made a defensive move by preventing MSFT from gaining any traction. MSFT, however, lost in many ways. Their search engine and Ad Center program continues to lag. Even worse, AOL engineers analyzed their engines and certified it to be inferior to others out there.
Comments (0 )
s
In today's WSJ there is an article about this. It is an attempt by MSFT to attract new users and create stickiness. This attempt is tricky because they may attract the wrong kind of users. In the race to become the next business platform, Google, MSFT, Yahoo and eBay are attempting to create direct and indirect network effects. Yahoo's recent acquisition of del.icio.us will promote user to user interaction and contribute to direct network effects. Would this revenue sharing model from MSFT impact either type of network effects?
Comments (0 )
s
This makes sense. First of all, social tagging is a good alternative to brute force search. Second, although I had earlier predicted that Google would buy del.icio.us, Yahoo is more into social tagging (with Flickr) and would be able to internalize del.icio.us with their company culture.
Comments (0 )
s
As I was mulling over the issues behind opensourcing, outsourcing and standards, I realize that customers want interoperability and product companies care about architectural control. On the surface, these are seemingly conflicting demands.
Customers purchase software products hoping that this product will allow them to exchange information with other products/applications that they currently have and also with any future products that might consider. In fact, they would love it if product companies gave them platforms that would make it easier to develop interoperable products. Furthermore, if these platforms conform to open standards and are extendable with out incurring any penalty, this is nirvana. These open standards would prevent any holdups, since the product company could switch easily to a competing platform. This is, however, bad news for the product/platform provider, since they could get commoditized.
From a product provider perspective, they would like to provide platforms that are proprietary and adopted by all. Their objective is to control the platform and create toll booths to collect rents from every customer using the platform. These end solutions are reached for narrow product categories. By and large, most markets will have platforms that are somewhat open and somewhat private/proprietary.
In a purely open platform, the customer is very happy and the platform provider has to use their knowledge to create complementary products and compete with their customers. The other option is for them to provide vertically integrated solutions on top of their open platforms. An example of this is IBM using the Linux OS and providing integrated solutions on top of that. Another strategy could be to open the platform “kernel” but provide extensions to it that are proprietary. For example, Apple took BSD and forked it to create their OS. When a platform provider allows API-based access to their proprietary platforms, customers adapt by using these platforms as a black-box and writing their code in a manner that makes dependencies explicit. This is the situation with Google and complementors who write mash-ups that work on the Google platform. The reason Google will be allowed to control the architecture is because they make it easy for these complementors to innovate. Google benefits because its business model depends on the ad-generated traffic from these complementors.
Comments (0 )
s
Error sending email
Email sent successfully
|