www2009 in depth


The World Wide Web Congress has reached its eighteenth edition and took place between April 20 and 24 at the Palacio Municipal de Congresos de Madrid, with madri+d as official partner in the dissemination of the event. This event is the world's largest Internet congress and is held annually, bringing together researchers, developers from large and small companies. The Internet has a particular characteristic compared to other fields of research, and that is the great speed with which the results of research can reach the market, so that both researchers and technology-based entrepreneurs, as well as large companies such as Google or Yahoo have presented and discussed numerous topics. In the debates, academic topics were mixed with Internet monetization without any problems.


As the profile of the participants was heterogeneous, the organization has adapted to their particularities: it is worth mentioning that hardly any paper has been used, not even the agenda that was delivered with the accreditation to the event has been used massively, instead the congress website (www2009.org) has been constantly updated, as well as a wiki that has served to compile the conclusions, and has been open to contributions from people inside and outside the Congress, the use of this collaborative tool has significantly conditioned the working procedures in the rooms where it has been used, so that in the morning session an agenda of topics of interest was developed, which later, in the afternoon sessions was developed, discussed and documented; the rules of the working sessions that were recorded in the wiki were two: the pages generated may or may not have a prefixed format (depending on the session) but there must be wiki consensus. All this required a powerful and stable wireless connection that worked well, although during the work sessions there was a drop in the connection speed, this did not cause any setback; another requirement was to have plugs to power the laptops at any point in any of the rooms.
As for the attendees, they have been mostly foreigners, and specifically, the percentage of Asians is very high, mainly Chinese, Indians and Japanese, especially from the side of the Academy. As organizational curiosities: each morning, before starting the sessions, a room was set up for yoga, or that in each room there was a minimum of ten electrical plugs.

Starting points

The topics discussed have been based on a series of principles or trends on which there is consensus among researchers, firstly the necessary simplification and homogenization of the web so as to allow access to advanced services to people with little knowledge of computers, so that a large number of lines of research are aimed at weaving an Internet without the end user perceiving changes, or at least without having to learn new computer techniques, instead seeking to simplify navigation. Another basis is the growing importance of the mobile Internet. The decentralization of the Internet has an influence on interpersonal relationships. It is worth mentioning that very representative Internet people such as Tim Berners Lee, Mary Ellen Zurko, or Ricaro Baeza-Yates were discussing on an equal footing with congress attendees at numerous working tables; This decentralization also reaches the hardware, the devices that are interconnected through the Internet, in what is called Cloud computing or cloud computing, with respect to the physical support, the hardware, there have been no presentations focused on it, as long as the neutrality of the Internet is maintained, the hardware does not need a space in the Congress; although threats that loom over the neutrality of the Net have been listed.
The generation of content occupies, in itself, a secondary place in the topics of the congress, although there is consensus on the need to systematize and qualify it to facilitate its delivery to the end user; however, information is the real value of the Internet, but it can be anywhere, so the work focused on making it more accessible to the end user.
The socialization of the web deserves its own section, while in WWW 2007 in Banff (Canada), there was no special section for social media, social networks, and in WWW 2008, in Beijing, there was already an interest of the community for the social, with a space on social networks, however, WWW 2009 shows the current trend towards the social web, has had its own space and is the 12% of the articles accepted for the Congress.
The Internet is multidisciplinary, and the term Computer Science has been discussed as a meeting place for various disciplines with computer science: sociology, ontology, statistics, linguistics, mathematics, etc., from an integrating and forward-looking point of view, while there have been few commercial presentations of products oriented to the consumer market. From this point of view, computing is presented as just another tool, as a means and not as an end.
Semantics (linguistics) and intelligent systems (software that learns from what the user does on the Internet) open ways to achieve an orderly web (properly linked) and easy to use, so that the user finds what he is looking for, without having to be aware of the data processes that his search originates. Navigation on images, sound and video opens new possibilities to extract information from both objects and users, thus indexing and extracting information from images, sound or video by signal processing is a field with great possibilities.
The social impact of the Internet is much greater than the economic impact of the web, and its neutrality must be preserved in the face of economic interests. Likewise, the media impact of this congress is considered much less than the social impact that its conclusions will have on the population, given that the WWW2009 debated the future of the Internet and the main players in the sector were present, both from the technological demand side - the companies - and from the supply side - the Academy.


Social computing: work is being done to obtain social patterns and comparable data on the behavior of Internet users based on browsing patterns, and especially on the valuation that each user gives to an object, for example, a photograph or a text document, on the basis of which extrapolations are obtained of how the user will value another different object that he does not yet know or has not yet valued, so that there are direct applications in the field of content recommendation to the user. To carry it out, multidimensional tables of collaborative filtering based on the user are used, and allow progress in image recognition, because the identification of the object present in an image is something that, today, a computer can not do, for the experiment the so-called human computation is used, in this case people are shown an image and are asked to assign some labels (tags) both Yahoo and Google are currently focused on this line of research. So-called human computation through games currently offers a great field for experimentation. Social computing is a very recent discipline, but one that is destined to remain in existence due to the great possibilities that Web 2.0 is opening up, and is based on the development and launch of new video games aimed at gathering information; for example, in 2003 it is estimated that users played Windows Solitaire for 3,000 million hours.
Trust in a user or organization through its records is a current issue for companies like ebay, which works on data extraction and processing in order to provide trust indexes, for ebay it is necessary to have two trust indexes, one transparent, such as the one they use now, and which is based on the opinion that the buyer offers the seller after a transaction, and another opaque where other circumstances such as the value of the object, its type, the answers that the seller offers, etc ... are weighted. Because trust is a relative index, while reputation is global.
Another current issue is the fight against unwanted mail (spam) by means of social tools, for which it is necessary to define a series of tags dynamically, work is currently underway to automate this tagging within the documents that circulate through the social networks, with these methods Yahoo has achieved efficiency quotas in the social filtering of 97% spam. Search engines are in themselves a source of unwanted information, this fact is corroborated by the low number of such information present in del.icio.us, a news aggregator that does not have a search engine; the counterpart of aggregators so based on the information provided by the user and so little based on search engines is the weight that a small oligarchy of users has in it, for example, in Digg, 25% of the news have been submitted by 20 users.
Two current web trends are simplicity, as shown by the success of Twitter or Google, and mobility, so that developers of small applications for cell phones have had their own space at WWW2009, according to Tim Berners Lee, cell phone access to the Internet will facilitate the incorporation of developing countries to the Internet. With a view to enabling universal access to the Internet, he also proposed emphasizing automatic translation of the web, which ties in with one of the specific areas of work of the Congress, web accessibility for the disabled, which also had its own space. Other challenges for the future are device-person interfaces and database management, while for Verizon the greatest weight was, as for ebay, on trust and security; and identity on the web in the case of Nigel Shadbolt.
Several workshops have been held to propose ideas, which have been reflected in the wiki As an official conference speaker, some of the topics of the brainstorming sessions were: what mobile devices mean for the Web, social computing, information mining,...
The extraction of information from Web 2.0 occupied a fundamental place throughout the working sessions, since in no way does it seek to create more obstacles for the user but to homogenize the current platforms, create communication interfaces between them, and above all to correctly channel the flow of information between the user and the documents that the user wants; in the words of Ricardo Baeza-Yates, it does not matter so much how one searches but why one searches for something. Premise adopted by the Linked Data user community (http://linkeddata.org/).
The extraction of patterns from audiovisual material occupied a good number of presentations, specifically from Yahoo, Google or Carnegie Mellon University among others, this field opens new possibilities to the work of search engines, it is about opening sound, image or video files and being able to assign them a series of attributes, in the case of Yahoo this is materialized in the automated organization of videos in webs 2.0.


Leave a Reply

Your email address will not be published. Required fields are marked *