Skip navigation

The Object Oriented Web

November 2, 2007 – 2:35 am

Space InvadersI don’t know about you but personally speaking, I have the feeling that a little something is missing in the Semantic Web as it is today. Both RDF and microformats aims to normalize data so that websites can communicate between each others. It would be a big step forward, similar to the invention of writing if you’d like, the Semantic Web becoming a kind of writing for computers. Notice the impact of writing for humanity and you’ll understand better the helplessness of computers which haven’t got it yet. If an isolated individual is not up to much, it is the same for the computer unable to communicate with its peers. Imagine the upcoming possibilities when computers would be able to “talk” to each others thanks to the semantic writing.

But in the end, it comes down to change the “World Wide Web” into a “World Wide Database“. It would be already a big step forward if we could get there, but I think we can go even further, as of right now, beyond the Semantic Web we could go directly to the “Object Oriented Web”. There is no need to search for this term on a search engine, because it doesn’t exist, at least not yet. This absurd idea occurred to me while I was reflecting about how to make decentralized social networks. I thought about data going from one person to another thanks to the “spreading” feature (see my previous post) and noticed that it could be interesting to join a certain number of functionalities to those data. Look at the blogs for instance, they usually have a “comment” feature and it would be interesting if this feature was available wherever the posts appear, whether it is on the original blog, a RSS aggregator like Google Reader, or a social network. Just consider the blog articles as objects capable of ubiquity composed of data (properties) as well as actions (methods) and you’ll start to envision what I mean by the “Object Oriented Web”.

In short, if the Semantic Web aims to establish standards for different type of data, the “Object Oriented Web” adds the possibility to normalize a certain number of actions. Beyond the “comment” feature applicable to almost any type of object, we can imagine many other possible actions. The first ones to come to my mind are: collaborative tags, vote feature “à la” Digg, “human translation” feature, system to flag malicious objects (illegal contents, spams, etc.), “buy” or “donate” features to remunerate the object’s authors, etc. All these elementary actions could be normalized in order to allow the portability of the objects and the interoperability of the services. When we purchase something in a store, it is convenient to be able to use different credit cards, Visa or American Express for instance, because the machine recording payments should always work. Likewise, it seems interesting to allow the interoperability of the new online payment services, because the customer should be able to choose between Paypal, Amazon FPS or Google Checkout. If this is so obvious for payment services, it should be the same for numerous other services more or less universal, such as those mentioned above: comments, tags, translations… For example, if the “comment” feature of WordPress doesn’t satisfy me, I’d like to use the one from CoComment, or any other service specialized in comments.

Clearly we should go towards the greatest interoperability as possible. Nevertheless it appears necessary to allow the authors of the posted objects to 1) define the possible actions on their objects, 2) control the services that will be in charge of these actions. The creator of a “music” object will authorize an action for receiving donations while refusing a specific donation service, because this last one would be against his deontology for instance. Conversely, an author could force a particular payment service because he gives a percentage of the amount to charities. Actually this freedom of choice should apply at different levels: 1) the objects authors, 2) intermediary services hosting the objects (social networks, blogs, online stores…) 3) people receiving the objects will themselves have access to a choice of actions and services, more or less various according to the choices made previously.

You should now have got a better idea of what I mean by “Object Oriented Web”, which is a concept that goes further than the Semantic Web, but that paradoxically seems to be easier to accomplish for leading to concrete applications relatively easy to apprehend. But how is this going to work? I cannot say exactly, all I have right now is the overall vision of a system allowing dematerialized things to spread on the web in a way that goes beyond the form, beyond semantic, like a powerful mix of data and actions giving objects their own existence. All that in the most open way possible while preserving freedom for authors, intermediates, and finally recipients like you and me, who receive and spread the different objects forming the web.

This is a translation of the article “Le Web Orienté Objet” originally written in French.

By Manuel Vila | Posted in Decentralization, Object Oriented Web, Semantic web | Tags: , , , , , , , , , , , , , , , , , , , , | Comments (10)

Social networks

October 26, 2007 – 8:59 am

Once again, the main idea behind the social networks comes from a reversal process. We’re dealing with an approach focused on the people (user-centric) and not on the applications allowing to produce various data (text with blogs, pictures on Flickr, videos on YouTube, etc.). Rather than indicate to our contacts the numerous RSS feeds representing our “digital life”, we are going to point at a unique address (our OpenID) whereby they will access to any shared data. Even better, they will be able to add us in their contact list in order to receive automatically our new data (our “lifestreams”). To draw a parallel between an existing tool, add a RSS feed to an aggregator like Google Reader comes down to add a contact in our social network. But there is a major difference because this new approach simplifies things a lot while introducing many new fascinating possibilities. Just take a look at Facebook or at any “lifestreaming” tool to convince yourself. And we haven’t seen nothing yet because it is possible to pick any application (homepage, blog, forum, mail, newsgroup, chat,…) to reinvent it, by imagining a completely new interface, following the logic of social networks that is to stay centered on people rather than on tools. And everything can change, it is not only a matter of interface but it can open up unexpected new horizons, with new paradigms to create.

So the basis of social networks is the individual. From now on, people can be represented by OpenID, there is no need to create as many accounts as websites anymore. We create an account somewhere and we use it everywhere. Then come relations between persons, in other words the social network. As scholarly as the term seems it actually covers a very simple idea. Just imagine an old fashioned address book where you keep all your contacts. There is family of course, close friends or acquaintances. But it can also includes the plumber, the mechanic, or colleagues, etc. At last it can lead to add people who don’t know us but that we wish to follow on the public lifestream: artists, scientists, journalists, bloggers, politicians, priests, philosophers,…

In the social networks, people share any kind of objects (texts, pictures, videos, etc.). Then those objects can be spread from one network to the other, a little bit like “word of mouth”. In one click any type of data can be propagated (whether you are the author or not) to part of or all our contacts, who can themselves rebroadcast it. This is how an information is going to multiply in order to affect a lot of people in a viral process. 1) I receive (or create) an information that I wish to communicate to my friends, therefore I click the “Spread” button. 2) My friends receive the information and they can spread it again if they want to. 3) Friends of my friends will then receive it in turn, while amongst them some don’t even know me. If they are interested by the information they may ask themselves “who’s that guy?”, take a look at my lifestream and eventually add me to their contact list. Boom (© Steve Jobs)! I’ve extended my social network, I made new friends! :)

The “Spread” feature seems essential because it is how the information often circulate in the web or in real life. That’s all I do when I send a link to a friend, blogs quote themselves, petitions via e-mail (or jokes) propagate from person to person thanks to the “Forward” feature, etc. This is what we call a decentralized communication system, where the information doesn’t only come from major medias but from anywhere. Every one of us, small or big, detaining the potential to create a worldwide buzz in one click and a few friends relaying the information.

Unfortunately, this “Spread” feature, as important as it seems, is either missing or poorly implemented in almost every existing social network. How often do you use the Facebook “Share” feature? It is as important to propagate information as to receive it. Spread is crucial, it’s a basic function, a natural mechanism.

The social network that I have in mind is based on two big ideas. 1) Allowing a maximal decentralization, we’ll get to a social graph common to all applications, a social network not trapped within a unique platform like Facebook. 2) Set up an infrastructure with both a user-centric and object-centric approach. I’ve previously mentioned the “Spread” feature applying to any kind of data but we can imagine many other features such as tags, comments, votes, translations, summaries, derived or related objets, bans, alerts (illegal contents, spams,…), modifications (Wiki), purchases, donations…

To sum up, we’ve got people exchanging objects, each one of them having a certain number of features more or less generic. The whole thing should entirely be decentralized, objects could be hosted anywhere (even at home) and multiple actors could be in charge of the functionalities. For instance, there could be several services to manage comments, several services to manage purchases, votes, translations, etc. But the whole thing would be completely compatible and consequently interchangeable. For exemple, if the “Comment” feature from Blogger.com is not satisfying to me, I’ll get the opportunity to use another service allowing me to pretty much do the same thing but with an approach that fits me better. In other words, we’ll be able to create or customize our applications by assembling components. In the context of the Semantic Web, it leads to normalize the function (API) on top of the data. We could call it “Object Oriented Web”. But this is another subject that I will treat in another article. A new world opens up to us, a unifying world where everything can converge in a wonderful way.

This is a translation of the article “Les réseaux sociaux” originally written in French.

By Manuel Vila | Posted in Decentralization, Object Oriented Web, Semantic web, Social network | Tags: , , , , , , , , , , , , , , , , , , , , , , , , , , , , | Comments (10)

Datahubs

October 24, 2007 – 5:19 pm

To begin with, there is a very simple idea: websites should indicate themselves their changes to the search engines. I’ve already touched on the subject in a previous post, right now search engines have a reversed approach which consists in crawling the web constantly looking for the slightest modification. Don’t you think it’s silly? Think about the number of web pages to visit, imagine the cost to get the lowest frequency between each visit. Consequently, it seems difficult to consider the development of new search engines today. Nevertheless, the advent of semantic web should lead to their multiplication, in a vertical way, while search engines are getting specialized more and more in specific fields.

Crawling seems to be the “boring” part for the search engines and if they want to be distinguishable from each others it won’t be with crawling, the innovation should be better in the matter of indexing, ranking, etc. But can we consider that some day search engines as Google or Yahoo agree to pool their crawling? Surprisingly, I think so.

But before we go further, we must know what we’re talking about. All in all, crawling consists in making a sort of “backup” of the World Wide Web. Somehow, generalist search engines need to own a full copy of the web. Therefore they need to scan and scan again the web to get the freshest copy. Worse, vertical search engines have got the same problem, even though their favourites domains are limited to a few topics, they still have to examine the whole web since the information they are interested in could be anywhere. At least it is the present approach.

How can we improve things? First and foremost I’m more and more convinced we need to reverse the process. Search engines should not question websites, websites should inform search engines of any changes. Today a web developer would have to set up a robot.txt file (or a Sitemaps) to insure the best indexation as possible. Tomorrow he will add a mechanism able to inform search engines about any changes that have occurred in his database. It shouldn’t be a problem for modern websites based on the MVC (Model-View-Controller) paradigm, they’d just have to add a “plugin” at the model level. To sum up, this plugin will be in charge of alerting search engines in order to report any “Create”, “Update” or “Delete” action.

At the end, this reversal process should enable the development of real time search engines. Imagine that as you enter your keywords, results will appear as the web changes! By the way, if websites have to contact search engines, which ones are they going to pick? Are they going to restrict it to certain ones? Of course not, websites should be able to spread their modifications towards a maximum of search engines, from the most important to the most specialized, without even having to know them before. How could this be? I’m thinking about some kind of relays diffusing the “modifications feed” as widely as possible. Let’s call it “datahubs” if you want. Datahubs will be linked to each other in a completely decentralized way and if a website sends an information to a certain datahub, every other datahubs will receive the exact same information, by a cascade process. In another way, if I want to create my own datahub, I will only need to connect somewhere, to another datahub, to receive all the changes happening everywhere on the web. For its part my datahub will be able to spread the incoming data to other datahubs. Surely, my datahub will have enough bandwidth to let all those informations transit.

It would be interesting to evaluate the total bandwidth necessary to transfer in real time all the web modifications, but I think we can estimate it shouldn’t be too high. In fact it actually should be pretty low and if I had to guess and give an approximate number, I’d say that 1Gbps would be enough if we stick to the textual data! Today we can find some hosting companies able to provide this bandwidth for less than $100 a month. Try to figure out the total cost of the constant crawling done by Google, Yahoo and MSN (only to quote the main ones) compared to the few dollars necessary to accomplish the same thing with the datahub idea.

Nevertheless, crawling isn’t everything. If we wish to create a true search engine we are going to need to accumulate a very important mass of informations in order to achieve some basic operations such as parsing, indexing, ranking, etc. Consequently, if we want to create a new generalist search engine we’ll still need to think about a huge infrastructure. On the other hand it would actually be very cheap to make vertical search engines specialized in specific fields, thanks to the datahub concept. Besides it would be one of the main feature, datahubs allowing to declare the datatypes that they wish to receive and spread.

At the end, it appears to me that the necessity of datahubs is obvious and the potential is so big that I can hardly imagine all the possible applications. But this idea is fairly new to me, I just started and I barely know the actual state of research on that matter. Aren’t they any people working on it already? Your feedback is welcome.

This is a translation of the article “Les datahubs” originally written in French.

By Manuel Vila | Posted in Decentralization, Object Oriented Web, Search engine, Semantic web | Tags: , , , , , , , , , , , , , | Comments (0)

Backlinks

October 19, 2007 – 4:03 pm

The more we use the internet, the more we realize the necessity of finding new solutions to better organize the growing mass of informations. Today we actually have a certain number of tools to add meaning to the informations that we drop all over the web. A comprehensible meaning to computers, allowing them to help us better organize things. That’s the big idea behind the semantic web, an idea which appears to us everyday more and more obvious. And in this field, we already have many advanced technologies, starting with those offered by the W3C itself: XML, RDF, OWL,…

But I have the feeling that something important is still missing to allow the semantic web to really take off. To understand what it is we have to take into account the fact that the web is in permanent change, data is continuously added, modified or deleted. The fact of having semantic informations doesn’t change anything, it just challenges more everyday the search engines to crawl the whole web constantly in order to detect the slightest change. It would be so much more efficient (and simple) if websites could alert the search engines themselves: “Hi google, I’ve changed, please visit me!”. Furthermore, it would strongly reduce the indexing time, until we come up to “real time search engines”. You might say that Google is in charge, that they earn enough to endlessly increase their infrastructure. You’d be wrong to think that because it is not only Google’s business. From the expansion of the semantic web should emerge various specialized websites in the need of aggregate an important mass of informations focused on one field but able to appear anywhere on the web. So what, everybody’s going to crawl separately?

But on top of the issue regarding the search engines, there is so much more. Let’s imagine that I want to make an online address book. The easiest way consists in storing the data myself; I will save my addresses in any database and that’s it. There is another approach, a little bit more complicated but so much more interesting: you can store some kind of link making it possible to get the information back at the source at any moment. In my address book, for instance, I will have a record directly connected to my favorite restaurant. Nevertheless this restaurant has got a website which includes exactly what I’m looking for: the restaurant details are showing in a way that my address book recognizes it, as the hCard format for example. Therefore I will subscribe my address book to this website.

Then, thanks to the subscription, the restaurant details would be available “on demand” and show in my address book. Of course, we can optimize the process by keeping a copy of the informations in a local database. But let’s make it clear, it is only a temporary copy, a kind of “cache” if you want. The real data stays at the source, in the restaurant’s website. And if a change occurs, like per say the restaurant has moved, changes will be automatically reported in my address book.

This “data subscription method” seems to be an interesting way to reach a kind of decentralized database able to work at a worldwide scale. But there is a much more essential aspect: the idea of “backlinks”. Actually, a subscription comes down to weave a bidirectional link from one information to the other. This smallest concept actually really have enormous consequences. The computer would now be able to determine how data are connected to each other and suddenly become a lot more intelligent.

Let’s take a look at another example to understand exactly what’s going on. What if our restaurant is willing to collect comments from the customers, to show them in its website in a sort of visitor’s book for instance. The restaurant would simply add a form on its website so the customers could save their comments. But the restaurant’s owner is a lot more ambitious, he also wishes to show the comments that have been stored somewhere else, on other websites. Either gastronomic critics or miscellaneous opinions given on the web, our restaurant’s owner would like to display all of them on his own website. Unfortunately, there is no existing easy solution to accomplish this according to the current web. There is no easy way to search all the information related to our restaurant.

This is when the concept of “backlinks” could be very useful. Actually, chances are that the miscellaneous comments spread on the web already include a link to the restaurant’s website. But unfortunately in our old web, those links are one-way. The restaurant’s website doesn’t even know about them, unless it makes a “link:” request with a search engine or if it considers the “HTTP Referers”, but it stays unsatisfactory (in the world of blogs, there is something about it called the Pingback protocol). No matter what, let’s say that the links are bidirectional: when a comment would be posted somewhere, the restaurant’s website would be alerted in order to store the corresponding “backlink”. Finally you would only have to go up to the different links and find all the comments linked to our famous restaurant.

The relational databases don’t operate differently, the “backlink” concept is prevailing and we couldn’t imagine it in a different way. But this isn’t how the web works, is it good or bad? I cannot say… However, if we want to someday achieve the World Wide Database dream, I think we should seriously consider the use of mechanisms bringing subscriptions and “backlinks”, therefore allowing the semantic informations to really “exist”.

This is a translation of the article “Les liens retour” originally written in French.

By Manuel Vila | Posted in Decentralization, Semantic web | Tags: , , , , , , , , , , , , , , , , , , , , , , , | Comments (1)