Social Networking Sites Opening Their APIs

prostoalex writes "Business Week magazine is looking at social networking sites opening their APIs to third-party developers to enable social applications not supported by the network itself. Facebook is setting an example by releasing their API from beta into 1.0, and many others are expected to follow the suit. Quoting from the article: 'Since Facebook, a network of 17 million college students, started a pilot program last summer, third-party developers have created some 100 new applications. Now a Facebook user name and password can be used to log in to content-sharing and chat site Mosoto, and to automatically import Facebook friends into Mosoto's buddy list for chat. Facebook itself does not offer a chat function.'"

Comments

Great, when do we get a Slashdot API?

[psykocrime]

Great, when do we get a Slashdot API????

Re:Great, when do we get a Slashdot API?

[Anonymous Coward]

Probably the minute Slashdot becomes social. /ducks

Re:Great, when do we get a Slashdot API?

[alx5000]

The specs are to be released tomorrow around 3:00 pm, 4:30 pm and maybe we'll have another PR during the evening.

Re:Great, when do we get a Slashdot API?

[TeamSPAM]

Why do you need one? Are you planning on making a dup detector or a dup submitter?

Re:Great, when do we get a Slashdot API?

[Anonymous Coward]

Is that a serious question? Why is a machine interface to data content useful?

Every site with any kind of data should have an API. Every one eventually will. The questions are 1. how long is it going to take and 2. are the sites going to provide it themselves (doubtful) or are we going to do things the hard way.

Re:Great, when do we get a Slashdot API?

[AusIV]

Anything you want to know about how to interface with slashdot you can learn from slashcode. The wonders of open source.

Re:Great, when do we get a Slashdot API?

[Anonymous Coward]

Correct me if I am wrong, but that is just the source code under slashdot not an API for external access to the data stored on a running instance (such as slashdot.com).

Re:Great, when do we get a Slashdot API?

[PopeRatzo]

Is there a benefit to having a public API to social sites that does NOT involve advertisement, selling or other forms of commercial exploitation?

I'm thinking about the mosoto thing. OK, it would be nice to be able to share files with online "friends" if it didn't open me up to lawsuits or jail. I wouldn't mind the idea of having streaming content that could be uploaded and shared among contacts like YouTube if it was something other than the beta test for an ad-supported replacement to commercial television.

I'm not doubting that there are useful things that a public API to a social site could create, I just can't think of any. Please give me some examples, and let's start with a few that don't involve somebody making a buck off the "society" that's created by these sites.

Re:Great, when do we get a Slashdot API?

[Anonymous Coward]

yeah its call RSS. check the link at the bottom of the page.

Re:Great, when do we get a Slashdot API?

[Frosty Piss]

Probably the minute Slashdot becomes social.

Listen, you insensitive clod...

Re:Great, when do we get a Slashdot API?

[fbartho]

I assume you're saying that you're unhappy with the http interface then?

Re:Great, when do we get a Slashdot API?

[Vellmont]

Why do you need one? Are you planning on making a dup detector or a dup submitter?

How about a real kill file rather than modding stuff down? I can't be the only one that doesn't even want to see AC replies under a certain level.

myspace has already done this

[President_Camacho]

Business Week magazine is looking at social networking sites opening their APIs to third-party developers to enable social applications not supported by the network itself.

You mean Myspace doesn't have enough third-party "applications"?

Re:myspace has already done this

[jbloggs]

Don't be foolish. That's what happens when you don't directly and security support services that should and will exist. We are entering into a new era of computing by building applications on top of networks that reflect the real-world. Its about enchanced communication, and it's not about if but when. I applaud facebook for being proactive on this, even if they view it as a business move to increase their popularity. Ultimately its good for the consumer, especially not to unnecessarily fragment social networks by requiring everyone to rebuild a network for just 1 feature.

Re:myspace has already done this

[Anonymous Coward]

even if they view it as a business move to increase their popularity

The good side of capitalism in action.

Advaned Programming Interface

[Anonymous Coward]

RE: Social Networking Sites Opening Their APIs

Great, now if I could just find a woman on one to open her API

Re:Advaned Programming Interface

[Anonymous Coward]

Pitty you who has difficulty finding the manual.

Re:Advaned Programming Interface

[Misch]

From thinkgeek:

Please note: Adding love.h to your partner object requires a few additional objects be streamed in before some functions are made available:

#include <love.h>

Partner significantOther;
Dinner dinner;
Flowers flowers;

significantOther << dinner << flowers;


Otherwise the call to significantOther.putOut() will throw an UninitializedMember() exception.

Re:Advaned Programming Interface

[Misch]

>man woman
No manual entry for woman

Re:Advaned Programming Interface

[alx5000]

I tried MSDN, but their implementation seemed buggy...

Re:Advaned Programming Interface

[Anubis350]

$ whatis woman
woman: nothing appropriate

Re:Advaned Programming Interface

[Anonymous Coward]

No manual entry for woman

I enter them manually all the time. Bug marked WORKSFORME.

Re:Advaned Programming Interface

[MooUK]

Wouldn't an "uninitialised member" be more likely to be caused by insufficient attractiveness of the other party?

Re:Advaned Programming Interface

[Nahor]

significantOther << dinner << flowers;

Their idea of a romantic dinner is to eat flowers!? Or is it that they think their wives look like cows?

Re:Advaned Programming Interface

[Kjella]

>man woman
No manual entry for woman

Personally, I think they should throw a false error message like: "Preparing man entry... ERROR: Index too large" just to see how many who'd try to fix it :D

Re:Advaned Programming Interface

[Eivind]

C++ allows operator-overloading.

Unfortunately they then fell in love with it, and completely overabused it.

Allowing "+" to add not just two numbers, but also say concatenate two strings is reasonable.

Using left-shift to mean in effect "print" is not.

That's almost as ugly as my favourite C++-wart:

How do you separate object++ from ++object when overloading functions ? Answer: You add a completely bogus "int" argument to the prefix-operator, so the compiler recognizes the two as having different signatures. Yeah ! Why "int" you ask ? No reason -- just because.

It's Web 3.0 !!!!!1!!eleven1!!

[EatingSteak]

Uh oh, Here comes "Web 3.0". I wonder... who is going to take credit for this one?

Facebook

[halltk1983]

Are they still only for college kids? Or am I getting me social stal... networking sites mixed up again?

Re:Facebook

[ben4242]

Facebook is used primarily by the college crowd, but if you are an alumni of a particular school, you can receive a Facebook account.

Re:Facebook

[MattPat]

Facebook is used primarily by the college crowd, but if you are an alumni of a particular school, you can receive a Facebook account.

High school students can get a Facebook account, too, and it's becoming increasingly popular. (They are, however, kept to their own subdomain ;)).

Re:Facebook

[cashman73]

Or, you can just sign up without having any particular affiliation. Facebook opened its doors to the general public sometime in September or October, 2006.

Re:Facebook

[TheNinjaroach]

No Facebook has now opened its doors and created "Communities" for different districts across your state. I don't know about using it from outside the US though.

Social Networking Protocol

[Carthag]

For a while now I've been hoping that a general protocol would come out and replace the centralized networking sites. It would be fairly trivial to create a handshaking system over a simple p2p network that allows you to set the friend-status of other nodes. These nodes would then be able to access a local profile based on their status. The profile could contain pretty much anything that the user wishes to include in it. It'd have to be user-friendly though. Of course the hand-shaking needs to be secure so people can't crawl the network for personal information, but that could possibly be done with public/private keys or a similar scheme.

I don't have time to code any of this, though, but it would be a million times more efficient than the current system where you have some friends on some sites and some friends on others.

Re:Social Networking Protocol

[Bogtha]

There's a decentralised RDF-based "Semantic Web"-type version in the form of FOAF. You can already browse it with software like FOAFnaut etc, and generate your own FOAF file with FOAF-a-matic. There was a crawler called Plink, but that seems to be dead now.

Re:Social Networking Protocol

[drinkypoo]

It would be fairly trivial to create a handshaking system over a simple p2p network that allows you to set the friend-status of other nodes.

I can't tell if this makes sense or not. Please define "node" in this context.

If what you mean is another webserver, then I don't think this is at all necessary or in fact even desirable. I think what is necessary is simply exposing congruent properties. For example you need to provide a means for account validation from a hashed password without providing the password back to the caller, and you need to make various values available. This is best done with a standard format for information interchange, and I don't mean the american one.

In other words it would be best if in addition to any custom APIs provided, the sites also provided a standard one. It should be simple XML for back-and-forth compatibility. This is pretty much all that is needed for collaboration between them, provided you implement the authentication system. That way you can have an affiliation of any type of sites and share member between them.

Keep in mind, however, that most of these social networking sites will fight you with everything they have. They depend on attracting as many visitors as possible and convincing them to eschew all others, which is easy because most people would prefer not to flop between sites.

I still think the actual answer is just to run your own blog, and let google (and others) handle the social aspect. Why associate myself with myspace? Of course this is still hard for a lot of people, but it's getting easier all the time. For example I could go with a hosting provider with fantastico, install drupal (or wordpress or whatever) via that, and then use the appropriate functionality to tie myself in with a network of other like-engined sites. There are also modules for some of these to participate with others...

Re:Social Networking Protocol

[Carthag]

I mean that each user/profile is a node in a p2p network. A seperate application, like an IM client, based on an open networking protocol. It could actually be integrated into your IM client across networks.

Re:Social Networking Protocol

[jaydonnell]

It would be fairly trivial to create a handshaking system over a simple p2p network ... It'd have to be user-friendly though. Of course the hand-shaking needs to be secure

HAHAHAHAHAHAHAHAHAHA! Surely you jest.

Re:Social Networking Protocol

[stickystyle]

That would be...

http://appleseed.sourceforge.net/

Re:Social Networking Protocol

[shadwstalkr]

What's your business plan for this decentralized protocol?

Nothing New

[Anonymous Coward]

This is nothing new. The API's to many social networking sites are already open.

Considering the fact that Flickr is one of the first ones, see this particular use of the its APIs:

http://blogs.sun.com/MortazaviBlog/entry/persopoli s_the_takht_e_jamshid.

And in...

[Adambomb]

other news, Marketing data miners wring their hands and cackle with glee. Coming up at 11.

Above and beyond

[CLorox]

These open APIs are allowing developers to go so far beyond what Facebook was originally designed for, that it makes you wonder if these addins will spin off with their own system, eclipsing the original site. Mosoto http://www.mosoto.com/ doesn't just add IM style chat from your Facebook friends, but file sharing and streaming audio to everyone in your list as well.

How often will MySpace API Change?

[queenb**ch]

Consider the fact that Tom or what ever his name really is changes the software on Myspace.com about as often as I change my socks, I'm wondering how well that's going to work. Ask anyone who's ever tried writing something for the ebay API.

Furthermore the whole "online friends" site is sooooo jr high. I always get a mental image of a bunch of bow head cheerleaders standing around in the girls bathroom "Nuh-uh, I do so have more friends than you."

2 cents,

QueenB

Re:How often will MySpace API Change?

[FinestLittleSpace]

The eBay API is relatively robust and /does/ use compatibility levels to deal with backwards compatibility, however you're right in that if you try and program something too specific and that feature is cut from eBay, you're on your own.

bah

[DuroSoft]

yea well they wont even let us get xml versions of our news feeds from facebook so MEH.

Re:bah

[zantolak]

And how do you intend to syndicate them anywhere else when you have to be logged into Facebook to see anything?

Security?

[amplusquem]

They have had a facebook plugin for the latest version of AOL Instant Messenger for a while now and there's already been concerns about phishing issues. What's to stop a program from using the APIs to connect to facebook but then also keeping the facebook username and password that have been entered? Once you let 3rd parties connect to your system, security goes wayy down.

Re:Security?

[Anonymous Coward]

You don't use a username/password to log the user in. Users log in through the facebook website which passes an timed expiry identification key back to the 3rd party application.

The CIA would like to take a moment and say...

[unsigned integer]

Thanks. I mean seriously. I'm sure that they have tons of facial recognition programs / data collection scripts just WAITING to poll the stuff you people are throwing out into the public domain and link it to your name, address, age, anything else that they can slurp up automatically!

I hope no one on myspace and their ilk have political aspirations .... ever. You'll find your "blog" turned right back in your face when you're running for an important office.

Re:The CIA would like to take a moment and say...

[Frequently_Asked_Ans]

well, i'd just like to say, "I stand by all blog entry's, comments (on /. and other sites) and other such crap i've posted on the interweb"

Admin-less wiki with a secure session image

[Anonymous Coward]

Did some people try a millenium ago to have a passwordless moderation system with a secure randomly generated image? (that changes every time you load it like on profilemix.com's world visitors map) I think this would solve many of the basic problems of wiki administration... unless you hate democracy. (if it worked)

everyone with dirt has already been blackmailed..

[Anonymous Coward]

I think everyone with any dirt at all that was worthy of being blackmailed has already been blackmailed. so you're safe!

Standardised APIs?

[aharth]

Some social networking sites (e.g. livejournal.com and d.hatena.ne.jp) already provide basic data export in FOAF (Friend-of-a-Friend) vocabulary. Search engines such as Swoogle and SWSE aggregate some of the content published in RDF. The problem is that crawling large database-driven sites with millions of files takes years when adhering to the Robots Exclusion Protocol. On the other hand, an API can provide on-demand integration, but with every site building their own API, a lot of schema wrapping (e.g. via XSLT's) is needed to aggregate data. Vocabularies such as SIOC could provide a standardised API and data format for all sorts of community sites, which would facilitate the integration of data from multiple places.

flickr's API

[kisrael]

On a semi-related note, I just made a tool that scrapes flickr to extract flickr notes as an HTML imagemap. I started it as an HTML screenscraper but then saw there didn't seem to be support for this in the flickr api.

Why REST?

[neelm]

It there a tenant of Web 2.0 "Thou shall not use WSDL/SOAP"?