Showing posts with label hcard. Show all posts
Showing posts with label hcard. Show all posts

Wednesday, May 6, 2009

Yet Another Post on Spock.com

Techcrunch recently did an article on Spock.com stating that Spock is running low on cash and may be bought by Intelius. This is no surprise to me. Spock really didnt offer anything new. Spock basically did exactly what Wink.com was doing except that the user could edit other peoples pages, they even had a feature called Spock Power that controlled the editing ability of users. The idea at this stage was pretty cool, but then something happened. Slowly everything novel about Spock started to dissapear. First to go was the Spock Power. They did this in an attempt to stop spam. It may have worked, but along with the spam went most of the users that used to visit the site on a regular basis. Secondly, you could no longer edit other peoples pages. This made the site effectively exactly the same as other people search engines.

I had an idea for Spock to make a name for themselves but they never bit. Shortly after I wrote that article, Jay Bhatti contacted me and wanted to talk data portability. I pled my case and told him what they should do and why they should do it. He told me that Spock was treading a thin line because their data - which is mostly harvested from other big name social networks - is only allowed to be harvest from these sites because it basically provides links back to, and advertisement for the originating site. Spock became a less social site because the big name social networks threatened to shut off their flow of data. He cited this as a reason why they couldnt make Spock's data more portable. I disagreed and cited 4 different areas that Spock could improve upon that wouldn't piss anyone off and I put concrete suggestions for improvement into an email that was later passed to their engineering department and most likely forgotten. Here is the entirety of that email:

Spock Data Portability Initiatives:

1. Implement XHTML Friends Network (XFN) microformat

a. For every link in the “Web” section of a user’s search result add a rel=”me” to the link. This will allow for identity consolidation which will make Spock THE destination for identity consolidation because this information is already being aggregated on your site. I would suspect that this would draw a lot of traffic to Spock because anyone who wanted to consolidate their online identity could easily go to Spock and add any links in their web section that aren’t already there. I have already done this, but without the rel=”me”, services like Google SocialGraph won’t be able to pick up on it.

b. For every related person in the “Related People” section add a rel=”x” where is x is equal to one of these attributes “friend, acquaintance, contact, met, co-worker, colleague, co-resident, neighbor, child, parent, sibling, spouse, kin, muse, crush, date, sweetheart”. I would recommend using these attributes when the user themselves have used one of these attributes. If the user has used a term outside of these attributes, just use “contact”. You can also use multiple values like rel=”friend met colleague”. You could even take it further and map some terms to these values i.e. map “girlfriend” and “boyfriend” to “sweetheart”.

c. I see no reason for social networks to not want you to do this. All you’re doing is promoting connectedness between people on the web. This will actually help social networks recommend friends to their users because they can go to your site and see what users are friends on Spock that aren’t already friends on their site.

2. Implement OpenID

a. Follow Plaxo’s guide for OpenID-Enabling your site

b. OpenID eliminates the need for multiple usernames across different websites, simplifying your online experience. This is basically doing your users a favor. Rather than making them have to remember what email they used to register with your site and what password they used, they just use their OpenID login and password. The login and pass are authenticated at the OpenID provider and the session is returned to your site authenticated. It doesn’t get any easier. As OpenID gains momentum, I believe this will draw many more users to your site. I can’t tell you how many times I’ve been to a website that wants me to register and I just leave because I don’t want feel like going through the hassle.

3. Implement the rel-tag microformat

a. Just add a rel=”tag” to the tags link. For example “software engineer” would then be “software engineer”.

b. This isn’t as important to me as numbers 1 and 2. I t should drive more traffic to your site as it better let’s search engines know what the page is about. It also makes browsing tags easier. Well, you guys already have your own implementation for viewing the tags, but rather than showing tag searches as http://www.spock.com/q/”software-engineer”, you would show it as http://www.spock.com/tag/software-engineer.


Thursday, January 15, 2009

Open Source Contribution

So, I recently contributed to an open source project. The project is called org.microformats.hCard. It's a Java hCard parser that I was using in my masters project. I needed the parser to also have support for XFN so that I could parse and store hCard + XFN friends lists, so I added it. Thanks to Reinier Zwitserloot (the creator of this project) for working with me to get these modifications into the baseline.

I now have no excuse to not work on my masters project. Oh wait, I am taking two graduate level programming courses this semester along with full time work. Nevermind.

Monday, March 10, 2008

Spock and Microformats - It Just Makes Sense

For those of you who haven't been fortunate enough to visit spock.com, I suggest you do so. Spock is the newest (and most likely the best) people search engine on the web. If you have any web presence at all, you probably already have a page on the site.

I've been a faithful user of Spock since its launch, but I've always felt there was something missing. At first, I thought it was the lack of the typical social-networking whiz bangs like page comments and polls, but after having a chat with Jay Bhatti, co-founder of Spock, I realized that the real intent of Spock is people search. Their not trying to create another social network disguised as a people search engine like some (I'm not going to name anyone in particular). I'm still not sure I agree with this stance given that users tend to get bored without these features once the initial building of their profile and network has completed. This could prove detrimental to Spock given that their users provide much of their search content. I've since realized that this is not the missing link. Actually, what's missing are the links themselves, the semantic links.

Spock is a perfect use-case for using microformats given that most of their data is suitable to be formatted in preexisting popular microformats. Let's take for example the current categories of a Spock search result: web, pictures, tags, news, related people, contact. XFN could be used to markup the web, news, and related people sections, rel-tag for the tags section, and hCard for the pictures and contact sections.

The first benefit is that users of Spock wouldn't need to build their own XFN relationships by hand, Spock would automate the creation of XFN HTML through their user interface. Google has very recently released their Social Graph API which harvests relationship information present on the web in the form of XFN and FOAF, making it publicly available for use. If Spock implemented XFN, it would vastly improve the functionality of the this API, allowing for easy creation and use of relationships between people on the web. This may seem like a deflection of use from Spock to Google, but ultimately it would increase the use of Spock in order to build and maintain XFN relationships.

Another benefit is that contact and picture information could be automatically exported to a vCard using Firefox add-ons like Operator and Tails Export. Current Spock operations allow for the import of contact information but the only way to get it out is through their API. While this is good for developers, it doesn't do much for users who want to quickly get and use the information they want.

I hope Spock realizes the benefit of microformats and chooses to implement them sooner than later. I look at this as a time to shine for Spock, and stand out as a truly open, web 2.0 website.