Today’s pattern is "Innovation in Assembly". This is the third pattern as defined by Tim
O'Reilly at a conference on Web 2.0. What this turn of phrase means
is allowing access to your data via an application programming
interface (API). An API allows controlled access to the data
underneath. You can lock down, or open up as much as you want. An
external developer can create an innovative application that is
assembled with one API to one platform, or alternatively multiple
API's to multiple platforms, creating a mash-up.
The purpose of creating an API to your
data is to stimulate ideas to extend your Web 2.0 platform. All
the main players of Web 2.0, Twitter, Facebook, Flickr and Amazon
have API's. This approach encourages enterprising users/developers to
think up new ways to utilize the data created by the Web 2.0
platform. When data is accessed via an API the platform can track
this access, and generate statistics that they can use for marketing
and research. Knowing where your traffic comes from is an important
statistic.
A good example of innovation in
assembly is gigpig.fm. Gigpig mashes together event information in
Brisbane and globally from various sources. Gigpig doesn't create
the original data, it uses API's to Web 2.0 platforms such as
Eventful, Bands in Town and Last.fm. They format the data in their
own way and allow searching for live gigs. They re-assemble the data
for a specific purpose, allowing those interested in just one type of
event to cut through the clutter. They match the Innovation in Assembly
pattern by:- Utilizing various API's to access multiple Web 2.0 platforms.
- Enhancing and improving pre-existing ideas
- Take single pieces of data from various sources and mash them together to form useful information.
- Enhance their users experience by providing extra information, by use of Google maps to provide venue locations.
The screen shot below shows the information mashed together, including a location map courtesy of Google maps. They also show where this information is sourced, so you can always go to the original source. This shows transparency, as they don't hide where they get their content. You can see they also show nearby gigs as recommendations of other gigs of interest.
This pattern is all about piecing together data from singular or various sources to make new information or to simply enhance what is already there. Web 2.0 sites want developers/users to extend their platform as this creates an ecosystem, for the platform to grow and improve. This in turn equates to financial growth. If people are heavily invested in an ecosystem, they will want to stay in that ecosystem.
References:
Mashable.com article: Why the Best Innovations are about Relevance, not Invention
What is an API?
Technorati: ZQK7UBHEJW7P
What is an API?
Technorati: ZQK7UBHEJW7P
I think I understand how this works from your post, but just to clarify; Gigpig requires certain other web developers API's to be made avaliable before any of this "mashing" can occur? So say if ticketek didn’t have API included by the developers, Gigpig would be unable to utilize that site information?
ReplyDeleteThat is correct, ticketek would have to provided a way for developers to access their event information. Gigpig gets their information from other websites that have gathered the original data, so in this case, live music events. Gigpig then hooks into all these other websites (that have provided a way to access their data) to get the event information to give to their users.
DeleteA truly interesting post. I think we will witness the emergence of more and more similar services in the near future that would present users with information combined from multiple web information sources in a useful way. Current trends show that this will be most likely accomplished through the increasing use of Web APIs. As a result of this, I can image a future scenario where web service providers would be "segmented" into two quite distinct roles: certain services would be specialising in providing the data part (the "database backend" role), while others would focus their efforts on presenting the information gathered from a variety sources in a unique and useful way (the "user interface" role).
ReplyDeleteI think combining services from many sources into new, or enhanced services is the way of the future. The wheel has already been invented, and now its up to developers to invent new ways of using it.
DeleteIt's amazing that users can get various information via this website even though this doesn't create the original data. I think this seems to need only a web designer to effectively present the data from other sources, neither a contents developer nor a database manager. This is a good example. Thank you for sharing a good information.
ReplyDeleteThank you for your comment. I think that a lot of web 2.0 sites that utilize other sites API's, don't create original content. Their success comes from extending an idea.
DeleteGood example of this week's pattern! I like how Gigpig allows users to comment using their Facebook account. I like this feature in many websites because it allow users to simply use their services without need to register and create an account.
ReplyDeleteI agree, there are so many sites out there that you may only use a couple of times. It saves having a million accounts if you can utilize one of your social network accounts to login.
DeleteHey Elizabeth really nice post, i am not commenting here because i feel obliged to, its because i genuinely liked your post. You have by far the most concise and succinct explanation of API's that i have read, you put it together so well and it made me read more!. Gigpig seams like a great example for Innovation in Assembly, it caters for a set audience and does it well by being a fantastic events information aggregator. Ashwag mentioned above me how its great to allow users to comment user their own facebook accounts. I Think this is a great feature for this site as its definitely puts a face to a name so to speak but it allows user to contribute their opinions without having to make some silly account which is fantastic. In my opinion this is a rather underutilized tool that far too many people dont use on their websites. This feature is something we will definitely see more of in the future, would you agree? sorry for posting off my google account, it would not let me link my wordpress :(
ReplyDeleteThanks for another great comment :) I agree the sign on with another social media account is a great feature. I have on occasion decided not to bother with a website because of having to create yet another account. So I hope we see more of this in the future.
DeleteHi Elizabeth, I like your style of writing. All of your posts are well laid out and informative. Just to jump in to your conversation with Nick, I agree that being able to use Facebook to access new sites like Pinterest is a good feature, actually Bitdefender keeps asking me to log in with FB now too. However I am still loathe to do this as my FB is quite private and I don't use FB games/apps for the reason of having had my account hacked. Am I just being paranoid as once your on FB you've pretty much sold soul anyways ?
ReplyDeleteI wouldn't say its paranoid. Letting external sites access your Facebook account involves similar amount of trust as for using apps/games. If you don't trust a site then its best not to let them near your personal information. I like to use an already established account for accessing a site I might not use much to avoid having accounts for every single site I have ever visited. What I usually do, is once I have finished doing what I needed on the site, I then revoke access via Facebook/Twitter so they cannot access anything further.
DeleteI have thought about, revoking access later that but as this is the semester for new learning re Web 2.0 maybe I should just embrace it all and regret later.... the first page I 'StumbledUpon' was Bike EXIF. So I liked them, then they had a FB page so I liked that, then Twitter and they have Pinterest as well, so really it is inescapable/inevitable...
ReplyDeleteBlogspots comment box really doesn't like Chrome.