Uno de Waal

Uno de Waal’s online space covering everything about web2.0, social networks and internet related developments in South Africa and how it fits in with the rest of the world.

Google you bastards


Dammit, why does their Toilet Seat look better than my Toilet Seat?


24.com launches Answer.it, gets deeper into the social web


As part of our recent foray in social services (not the welfare kind, the web kind) at 24.com, we’ve launched Answerit - a site where you can ask and questions and get human answers.

It’s a very well thought out product - you can thumbs-up answers, follow commentators and sections, view rankings and a whole bunch of other nice user features.

A question you might ask is “Why? Isn’t there something like metafilter or Ask.Yahoo?” Well, yes there is. But there’s also relevancy in being local (how do you like this one), as well as fostering the community that has been built up by 24 (we have one of the largest blogging audiences/platforms in South Africa).

Slowly but surely we’re builing up a good suite of social products - Laaik.it, Answer.it, Play as well as a number of communities in our other sections like Entertainment, Wheels and Health. We’re going to be doing some even more useful and exciting things in the short space coming up :)

Other people mentioning Answer.it:

Pre-release Vincent Maher.
Stii.
Matthew Buckland.


Questions around dataportability


I really liked the post on RWW today about dataportablity by Josh Catone. It’s something that I’m struggling to get my head around – who owns that data?

I’m a believer that while you own the data, the activity, i.e. what you did on the site and what you put on there, is partially owned by the site. Just because you uploaded a photo to Flickr, and you used all the cool nifty uploading, tagging, rotating, sorting, organizing tools that Flickr offers, you should now be able to port all of that to Facebook? Same, Facebook has friend tagging as a feature, and it’s a piece of technology that they pioneered, and now you want to take that over to Flickr?

That part doesn’t sit well with me. I feel that the services/sites have invested quite a lot of R&D time into building these features, and for you to now be able to up that data and take it somewhere else…

The example Josh used is quite a good analogy – can I take my data from Store 1 to Store 2?
Store 1 has brilliant data capturing, sorting and customer research. Store 2 is a mom and pop shop, no technology, but they can tailor a lot of niche services to you, if they can get that data. Yes, I would like that data portable, but Store 1 has heavily invested in that data, and it would not have been possible to get to that data without the infrastructure that they built.

Now, with some dataportabilty chucked in, we can quickly take the attention data from Store 1 and plug it into Store 2?

Is that attention data owned by Store 1? Yes, I believe it is. Unless they agree to making that Attention Data public, it would be a breach of service to take that data out. Unless there are laws governing it. Principle Six in New Zealand does exactly that embodies what I want to happen.
(thanks for the link Pete! )

Why do I say “Yes, I would want that”? It means that I don’t need to input that data again, and the new service would just start working automatically.

This is something I’m struggling with so far – I know it’s a good thing. But how do you keep the technologies safe that companies have invested in, and promote more research and innovation? No site would like their users to be able up and off to another site.

Should I be able to select what data I want, send that over to the new service? I believe so. oAuth is going to help me.


Social news sites are more valuable than Facebook.


If you take a look at the data that Facebook has on you, and the quality of that data you’d be able to see that they really don’t have that much qualitative data and can only target pretty rudimentary ads.

For making it easier, let’s split it in 2 groups: Profile data and activity data.

Profile data is the data that you fill in - Favourite books, movies, interests, music. Honestly I’ve never filled in this data and I don’t know of someone who has maintained that, it also doesn’t reflect your true preferences (typical case of “you don’t really know what you like”).

Your work details are in there as well - what you are busy doing and more or less in what field of business you are in. People don’t maintain that too well either. Then there’s also your network data - which shows in which area you are, but that you could probably get more accurate data on with IP targeting.

Last but not least - your relationship status, age and sex. Well golly gosh, ain’t that fantastic.

Up to this point, it’s really difficult to target an ad to a person. With this data, for e.g., would you be able to target an Audi ad to me, knowing that I prefer Audi over BMW? I don’t think you can target something specific to person with those details. Would you be able to know that I am looking for a vintage Mercedes Benz SL500 convertible? And be able to serve me ads for that?

I don’t think so.

But luckily, there is the opportunity to expand! We have a whole set of other data: Groups, photos, events, Friends, Pages and applications - which we can lump into activity profile data.

However, none of this data is structured enough to be able to give real value. Does Facebook know that I have an intense dislike for 7de laan if I have joined the following groups:

I struggle to think that they would.
(as a side note - there isn’t a Fan Page for 7de Laan…)

Similarly, the events that I attend are only tagged on a very rudimentary level, Facebook does not know that some of the music I listen to is nu-electro-afro-coochie-pop, but that I also like DJ Krush and really want him to come to Cape Town. Does it take into the account the nature of post-modern, ironic party organisers who label their Electro dress-ups a “Party - Barbecue”?

The photo’s we have of all our events and so are also not really semantic in any way - does Facebook know that I am wearing Diesel (and not Nike) sunglasses, or drinking Amstel (and not Windhoek) beer? Nope. Fan pages are a cool addition and it does allow me to specify the relationship between a product/person/music (i.e. “I like this”), but it’s still a very rudimentary system. Does it know that I like a genre of music (which I haven’t specified on my profile)?

My friends list isn’t too great either - I can’t get any more data out of there than I can from my own profile.

Applications has the best repository of info, but it’s in a data silo. Your last.fm application can’t talk too well with you Facebook profile and make you stop seeing “Beat Bladder Infection” ads along the side, and rather “Watch the new Justic video”. This data still sits in a silo and I’m thinking it sits more with the 3rd party app than it does with Facebook.

Still, at the end of the day I don’t believe Facebook could target me an ad for a vintage SL500 Merc.

Enter social news

Now, on the other hand, DIGG, or any social news site, has much more data on me. While they don’t have my profile data (like age and sex) they can find out what I like and don’t like much better. I click on links and submit articles, all these are in semantic form. If I’ve submitted articles on vintage cars, rated anything up or down, Digg would be able to better target an ad tailored to me. It’s because of that that I feel Digg (or any social news site) should have a higher per-user valuation than Facebook. That data is more valuable and deeper than Facebook could ever have.


Linkedin RSS


I’m quite liking the LinkedIn RSS feed for network updates, except for one thing, it doesn’t do what you expect it to do.

I’d like it if there was a link to both people’s profiles, especially the new person whom I don’t know.


What’s this?


20 points to whoever knows what this picture means:


Microsoft invites: Pull your address data out of FB


The Facebook/MS Messenger product allows you to pull your contacts email address’s out of Facebook and displays the addresses in plain text, ready for scraping.

Now this might not seem like too big a deal - but remember that of all the things that you can’t get out of Facebook (their API is quite extensive) the email address is the one that’s obscured. Essentially this means that you can’t map your friends in Facebook to friends on another service - which means Facebook is giving lip-service to dataportability.

However, if you were in a good position with Facebook, say if you were Microsoft, and say, if you were recently involved in quite a big business deal, you might be able to say “Hey, give us that data”.

Which would then give you a screen like the one below:

blur.jpg

So why is this important?

No other company is allowed to do this except Microsoft, and it’s not possible to do this without some kind of screen scraper. Obviously, if you are in bed with them you are allowed to.

Services like Outsync would have a field day with this data, but similarly, data miners could too.

But still it doesn’t do what I want it to do - I want it to show me who of my friends are already on MSN!
The last step is still a pretty idiotic mailer:

blur-4.jpg

Ideally, this would have mapped to your service already, added the contacts and there you go. But I’m just idealistic.


Socialthing: winning


 I’m trying out Socialthing and I’m really impressed with the “adding your services” task. It’s pretty quick and easy and uses a really cool interface.

One of the most interesting pieces is that they grab your Facebook news Feed, or a part of it. You’ll see that you Socialthing stream is different from your Facebook stream (the FB part). I’m also quite interested in how they do this - there is not API hook for your newsfeed (remember, it’s the juice that keeps you on FB) and it’s against the Facebook ToC’s. So Socialthing, how are you doing it?

All this got me thinking: The best Authentication pattern for me was the Flickr one - Their OAuth pattern is the way all services should work. Why does Flickr support OAuth, as well as FireEagle, but not Delicious? Hrm…

Regarding FriendFeed - I was expecting SocialThing to work in a similar way. While I might not be friends with people like Chris Messina, Brian Oberkirch or Chris Saad, I still want to follow what they are doing - which is why I use FriendFeed. I thought Socialthing would do this, but at the moment there isn’t the ability to add “people who aren’t really my friends”. I like being able to do that. While I think there are similarities between Socialthing and Friendfeed as both are Lifestreaming apps, they serve different purposes. So there might be a little bit of a gap in there, somewhere that both these apps are missing?

Socialthing also has grouping (for e.g. Twitter Messages) - something that FriendFeed guys put out as a major thang - I think that’s like… Basics. Otherwise you can use RSS, no?

Either way, a cool service. Check out some of the design patterns here below.

FriendFeed_1205153380605    socialthing!_1205152685494   socialthing!_1205139805703


Gmail OAuth


This entire fiasco with Gmail is going to do wonders for OAuth. There’s no publicity like bad publicity. I claimed it a while back and I still firmly believe that if the services make your data portable through proper authenication standards, you wouldn’t need to screenscrape it with a user/pass.

W00t oAuth.


Wasa and Google Analytics


Now that Google Analytics is looking to open up their analytics packages, I wonder how long it would be before you could grab that data and show it all in nice little user segments.

Taken the whole shebang lately with Wasa, you’d think that this would be a great tool to showcase your stats in some ranked way - very similar to the OPA.