Are you New Here ? You might want to:
1. Take a look at some of the most popular posts
2. Subscribe to the RSS Feed
3.
about StartupDunia
4. Leave me a tip by email
Rediff API - the good, the bad and the ugly
Its been about two weeks now since the news about Rediff API, Sociali, was leaked. The formal announcement about the API was made by Rediff during Proto. As part of the effort, Rediff will provide funding upto Rs. 2.5 lakhs for interesting applications. While I read about the API, I wanted to check it out for myself before making any comments about it.
The Good:
Opening up the data via API is definitely a strong and a progressive move by Rediff. As simple as it sounds, taking down the walled garden is not an easy job to accomplish, both conceptually and technically. I’m just surmising here, but I definitely think that the success of Orkut and Facebook applications had at least a small role to play in Rediff’s decision to open up their API.
Let’s take a closer look at Rediff’s API from a technical perspective. Aping Facebook’s approach, Rediff has provided three ways of accessing their API:
1) API exposed by a REST interface
2) Sociali Markup language (RBML)
3) Sociali Query Language (RQL)
If you take a closer look, the API currently has exposed methods that mainly revolve around user profiles, friends, news feed and ‘wall’. Besides the profile data, I was able to locate access to only two types of application data via the API - iShare media related data and classifieds related data. As per Rediff execs, iShare is the first application they are starting with, and will slowly cover the rest of Rediff applications as well.
The Bad:
To begin with, I’m baffled as to why bulk of the Rediff API revolves around profiles, friends, wall and such features ? Not all the Rediff applications support these ’social’ features. Of the top of my head, only iShare, iLand, Blogs & Connexions (probably a couple more that I might be missing here) support user profiles, friends, wall and similar features. Part of these API methods will be useless from the perspective of other Rediff applications.
A large part of the disconnect seems to arise from the fact that several sections on Rediff are user generated content and hence revolve around user profiles etc., while several other sections like News, movies etc are editorially generated content. Unless a different set of APIs are released for the non UGC sections, the current set of API’s just wont work.
This also brings up another aspect related to the non UGC sections. It’s mainly the structured data that is generally most helpful when exposed via API. Say stock quotes and financial news related a particular stock. Or reviews, ratings and photos related to a movie etc. Exposing unstructured data via API yields limited benefits and inherently limits the scope of applications built around it.
The Ugly:
While I think that releasing the API is definitely a bold move on Rediff’s part, I think they should first consider cleaning up their messed up websites first. The biggest priority, in my opinion, should be implementing single sign on across the various Rediff services. Currently, when I login into Rediff Question and Answers, and then visit their iShare website, I need to enter my login credentials again. Take a point or two from Google and seamlessly integrate all your varied services first. Exposing data via API’s can afford to take a back seat.
In conclusion, Rediff’s API appears to be a great move on the surface. However, on digging deeper, it appears that the whole API thing hasnt been very well thought out and is a disappointment. It will definitely need much more tuning and tweaking before developers can leverage all the different Rediff sections into their applications.
Similar Posts
- Barrons crashes Rediff party
- Rediff Q2 revenue up by 18%, search in Indian languages coming
- BigAdda and Rediff iShare statistics
Comments
5 Responses to “Rediff API - the good, the bad and the ugly”


Why does rediff have to come up with a new data structure and RBML(whatever that means) in the first place. We have opensocial and before there is FBML(facebook mark up langauge - a very neat thing). Why does one more company spending $ and time to recreate something similar and expecting developers to work on RBML that is useless outside rediff network ?
I think Social media apps should be portable beyond networks. Any way, I hope few Indian companies will make it big on rediff . It will be interesting to watch in the coming days
We have developed application on rediff platform so I know little bit about it. RBML is FBML only just the name is changed. That is ok with me as they dont have to completely reinvent the wheel but I think the launch of the platform is immature. There are lots of bugs around and I am really not fan of ishare site itself. How many users can rediff get to this site with this UI and feature?
From vBharat.com » Rediff API - the good, the bad and the ugly…
Its been about two weeks now since the news about Rediff API, Sociali, was leaked. The formal announcement about the API was made by Rediff during Proto. As part of the effort, Rediff will provide funding upto Rs. 2.5 lakhs for interesting application…
Abhishek,
thanks for sharing your first hand experiences with the api.
Great review. Thanks for the insight on the API, after facebook anyone opening their platform makes for interesting read. I agree rediff needs to enhance their user experience first. A lot of Indian sites though popular suck at their user inexperience, that is one area that Indians need to get serious about.
If there is anything that iPod teaches is that external form is as important as the guts.
Cheers