Watermelon: la web de datos enlazados en un eterno grafo de conocimiento > ricardo+alonso+maturana

    2 resultados



    Publicado el 22.6.2012 por Ricardo Alonso Maturana

    Emprendimiento y tecnología: la aventura de crear una compañia de Social Knowledge Business

    El pasado 21 de junio ofrecí una ponencia titulada 'Emprendimiento y Tecnología: la aventura de crear una compañía de Social Knowledge Business', en el marco del Tecnalia’s KBE Day. En este post, escrito en inglés para la propia conferencia, hago un resumen tanto de mi trayectoria como de la GNOSS.

    Entrepreneurship and Technology: The adventure of Creating a Social Knowledge Business Company (Ricardo A. Maturana's lecture in Tecnalia’s KBE Day)

    I offered a lecture during the Tecnalia’s KBE Day (Zamudio, Bizkaia; 21th June 2012), a seminar with Knowledge Based Entrepreneurship companies and professionals.  This is a transcript of the lecture ‘Entrepreneurship and Technology: The adventure of Creating a Social Knowledge Business Company’:


    I am the founder and the current CEO of www.gnoss.com. In October 2007, Luis Cacho, President and cofounder of arsys.es, joined the project. 

    I don’t know what expectations you have about this presentation. I’m not an academic person, so I don’t do research on this topic. I don’t have any general ideas about this and of course I have no recipes. I usually avoid giving some advice. So, what could I say with sense about undertaking new paths in technology or about creating new technological companies? I only have my own experience widened with some reading, some good conversations and, at last, I have walked this path with some friends and my family who wanted to share this adventure. To sum up, I feel capable to tell you my own story and think about it carefully, but no more.

    First of all, let me to tell you something about the knowledge company I’ve created: GNOSS.

    GNOSS works with semantic technology

    GNOSS is a social software platform which includes linkable and customizable social networks, which can be managed and in which it is possible to take part with a federate identity – a user can work simultaneously in several networks with personal or collective identity; organization, class or group, based on a structured web of data not on a web of documents. To put it briefly, www.gnoss.com is a space for social linkable networks that run on semantic web technology; a web 3.0 project offering solutions for knowledge management, collaborative work and informal learning.

    GNOSS can be used as:

    • a corporative tool
    • a learning tool

    GNOSS & Didactalia in Linking Open Data Cloud (September 2011)

    ​This story begins at the beginning, that is, just before I decided to launch GNOSS and, as a consequence, to create a technological start-up.


    When I was sixteen, I was studying at High School in La Salle, in the Deusto neighbourhood. Until that moment, I really enjoyed reading comics and books. Of course, I had read the appropriate books for my age, like Salgari’s and Stevenson’s, or those of a very popular author named Martin Vigil. Related to comics, I liked almost anyone, from Mortadelo and Filemón to Flash Gordon. But when I was sixteen, something happened: as the reading material at the end of the literature textbook, I found the beginning of a Borges’ short story: “Las minas circulares”.  When I read that singular page, I noticed that I had never read anything like that before. That day after school, I went to a bookseller, now closed, named “Paradiso” (like the celebrated poem by Milton) and I bought “Ficciones”. That day, with those few words from the beginning of the Borges’ short story, I went into a new dimension made of metaphors, concepts, literature, formal languages… The bricks to build a vision of the world.

    Borges not only was an inspiration, but he also represented the possibilities of the mind and thought in the good stories, especially when they are mixed with the most important of all the high human faculties: the imagination. The following years, I became a persistent, interested and passionate reader. I found out that the world has other worlds inside and that words contain all things and all possible worlds. I was definitely fascinated by the power of words and language, and by their dominance over the things.

    After Borges, other writers came, such as Cortázar and Stendhal, Tolstoi and Flaubert, Carpentier, Conrad and Greene and many others.

    First big learning: From then to now, I’m deeply convinced that it’s impossible to have a solid vision of the future without a solid imagination full of good metaphors. And the vision of the future has inside the whole of opportunities. Imagination and metaphors represent the form of the future in our present and in our mind. The killer faculty to create anything (and it is the same if we talk about painting or tech enterprises) is the imagination and its contents, not the intelligence.

    2. ENJOY

    Enjoy every moment, specially the worst ones, and difficult ones.

    Finally, I had to choose what I wanted to be in the future. Between us, that means what kind of university studies I preferred. I chose Philosophy in the University of Deusto. There, I discovered the pleasures and suffering of thinking and I had the opportunity of reading some a few but fundamental books: Plato, Descartes, Wittgestein, Russel, The philosophy of language… I discussed, studied and had the opportunity, so to speak, of seeing and thinking about the lower surface of things, the other side of the moon, the reverse side of the world. Those days were full of happiness in the strange way that happiness becomes alive in the extreme youth, a germinal period in which everything is still possible. I lived with passion those warm days and I enjoyed doing what I wanted to do without any utilitarian restriction. At last, those golden days came to end. But something, something important, remains up to now: you have to follow your real interest, listen to your heart. We all have the duty to deploy our talents, all of them, without any other consideration. That means living with passion and connect your deepest being with your actions and with what you do at every moment.

    After this, I taught Philosophy, I wrote my thesis on models of innovation dissemination and social knowledge engineering. I also promoted an institute for adult education. I was consultant in advanced training and knowledge systems. I’ve always felt that all these activities were deeply connected. GNOSS, our technological project, contains elements of all these experiences, like a soft perfume.

    Second learning: your life is unique and you only live once. Everybody has the duty to try to be happy and useful. Nobody has the right to waste the time and life that has been given. Connect the deepest interests in your life; living with passion, that’s the formula of good life and good business too. Creating and undertaking are passions. 


    When I was 38, I realized that it was the time of ‘now or never’.

    I wanted to work by myself and create and foster my own project. So, any time is always good.

    I remember worst times than the current ones. My father founded our enterprise group in 1977. They were bad times here, in the Basque Country, and also in Spain. While our economy was collapsed, the new political frame was under construction and we lived in the middle of a major political violence. We are not worst now. So, the present time is always the best one for doing what you want to do. If you feel inside the impulse for creating, for undertaking a project, you have to follow it up without any doubt. Tomorrow could be too late. “Don’t leave for tomorrow what you can do today”, we usually say.

    Third learning: Those who feel the urge to create and undertake, but against these deep impulse, put off or avoid this commitment, will end feeling melancholy and sad passions. 


    Once, when I was very young, my father told me that he didn’t want to work for anybody but himself. This is a matter of principles. We are educated in the belief that the most important aspiration in our life is to reach security. All our educational and cultural programme, all the social energy of our mothers, fathers and teachers, collaborate in setting this kind of value up. Tones of creativity, talent and personal possibilities of development are sacrificed in the security alter. But the consequences are cruel: we have built up a culture of bitterness and doubt; most people have got unreliable secret dreams. “There is a time for everything, and a season for every activity under heavens: a time to be born and a time to die; a time to plant and a time to uproot” (the Bible says).

    But the time goes by and, what happens when we haven’t done what we’d like to do? The answer is quite simple: social bottom of bitterness and envy, which punishes those who try to undertake projects. In our society, the social penalties for failing are too hard and rigorous, frequently are the first cause to drop out in the race for creating new enterprises and economic value.

    Fourth learning: be yourself, don’t work for anybody but yourself and don’t listen to the discouraging background noise of our fearful society.


    Our education is focused on contents, but less in applications, personal development and character formation. And success, as Woody Allen says, is not a question of intelligence but courage, audacity and bravery. The character formation is the key cultural competence to make effective a knowledge and open creative society. A society based on the creativity that fosters everybody’s talent. But now, in spite of the superficial rhetoric about the importance of innovation and talent, to be an entrepreneur, to found and promote and enterprise or business, to have the ambition of developing technology, is something that implies a hard fight against the social elements. But the school doesn’t work for the formation of character. Undertaking demands a lot of energy and a solid character even in those societies that don’t punish the failure; imagine how hard and difficult it would be when the society turns its back on whom has express the preference of living like an entrepreneur. The society that wastes talents is doomed to repeat its history as a farce.  Don’t forget, at least in the case of Spain, that our modern history, with the exception of the last thirty years (and this with shadows) is a story of decline and fall. I think that we have the duty to do something like the Basque dream (or Spanish dream if you want), a path that allows everybody to rise and aspire to a better life got over their efforts, will, merits and talents. Even in this case, character will be the key.

    Fifth learning: undertaking needs a self-confidence and a solid character, even more if we consider our educational, cultural and social values and environment.


    To do anything in the long term, you need first a solid vision. Without it, it is impossible to aggregate resources and talented people for your project. It’s necessary to visualize the point of arrival because any entrepreneurs has to communicate the project clearly to many players, pointing out the process and benefits of making the journey together. But visions need dreams. Martin Luther King said: “I have a dream”. In that moment, the dominance of the WASP people in the USA began to decline. After the Luther’s dream, there is the end of the apartheid that the black people suffered for centuries, and the rise of a society based on civil rights for everybody. Dreams are not enough. After having a dream, you have to be sure that you are not reinventing the wheel. And if you are sure, then you have to consider carefully all the things and circumstances from going from here to there, the arrival point that lightens your dream. Be careful! The entrepreneur who is not in love with its idea is an odd species. There is no place for crazy ideas, so consider carefully your project almost twice. And after all, JUST DO IT.

    Action is always beneficial. To connect thought and action is a magical formula that leads to sure success because, in this case, success is not only the final result, but also the process itself, which expresses the will to live the way you want.

    Sixth learning: Just do it.


    The will to do things is a necessary condition, like hard work and character, but again it’s not enough. All the projects need the adequate resources and… money, especially technology projects. In fact, money is used to get excellent resources; without them, will, character, vision and good ideas are almost nothing. To develop technology is expensive and requires to add several resources. Regrettable, this country doesn’t see itself as a technology creator, but as technological services designer. And here, when I say technology, I mean IT technologies, more precisely Internet technologies. This underlying belief makes it very difficult to get financial resources for undertaking in the field of technology. The people who could invest in this sort of projects don’t have faith in them. Here it is important to distinguish between a technological project and a business based on Internet technology. These last one are easy to finance because the investors have the expectation of a quick failure or a quick return on investment. Probably, we don’t have the financial culture and institutions to produce high-tech with global ambitious and, as a consequence, our bets are not enough aggressive, so we lose once and again generating in this way a vicious loop. Perhaps we don’t have nor the cultural conditions, nor the financial structures to aspire to have critical influence in the global Internet, but I think we should be brave for being a major player in the Spanish speaking Internet. This is the second agglutination and market, just after the English speaking Internet of the whole West Internet. Our project, GNOSS, has requested several financial resources. We have sixteen people working in our company; four of them are PhD, many engineers, etc. It’s not cheap. We have been fortunate, because we’ve been able of supporting this project with the solid commitment of our partners (Maturana Group and Luis Cacho, President of arsys.es) and with the collaboration of the public sector (CDTI).

    Of course, nobody reasonable should expect to get a loan from a bank, not now, but neither in the past when our financial system was extremely occupied laying and laying bricks. There are also public funds for innovation, but… it’s not the best way to do the things quickly enough. Bureaucracy is always too slow and demands too administrative task for an entrepreneur.

    Seventh learning: Be careful with the money. It’s never enough.


    Make a team is a critical subject for a technological entrepreneur. The reason tells us that we have bet for talent. Nevertheless, it’s convenient to take care when you want to hire someone. Some rules:

    • Avoid sad people; that is, people who see the bottle half empty, people who say: “I already said…”, and similar sentences. This kind of people resolves bored and discourages the rest of the team. There is no vaccine against this kind of people: only one of them can break down the moral of the best team.
    • Keep away from proud people and those who think that know everything. These people are always a brake. They usually aren’t able to ask for help and therefore, the way of solving problems is inefficient and slow.
    • Prevent against people who have communication problems and are cold from an emotional point of view.
    • Choose warm, smiling and optimistic people.
    • And last, spend money from the beginning in a very good human resources consultant, specialized in recruitment: it will be the best-spent money, for sure.

    Eighth learning: Always talent, avoid sad and proud people, choose the optimistic once and spend money in a very good human resources consultant, specialized in recruitment.

    9. PARTNERS?

    Ask yourself: Actually, do I need a partner?

    Frequently the conflicts between partners come from the different point of view about who of them are creating more value. This kind of discussion is not easy to solve. There are usually two points of view: the one of the investor and the one of the entrepreneur. Money trends to believe that it is the key factor, in spite of the fact that many brilliant projects fail because the entrepreneur and promoter loses the power of making decision. Be careful with partners and try to maintain always the power of decision. Technological start-up projects depend critically on the mood and moral of the promoter. Money can do almost nothing without an enthusiastic entrepreneur.

    Ninth learning: Partners only if needed and remember: money is always coward.


    Success is a very antiheroic task. Creating technology is not brilliant at every moment. Most days we spend time breaking stones. Talent and money are not enough without work, work and hard work.


    • ​IMAGINATION is the key human faculty for visualize future, opportunities, projects and business.
    • ENJOY. Your life is unique and you only live once. Everybody has the duty to try to be happy and useful. Nobody has the right to waste the time and life that has been given. Connect the deepest interests in your life; living with passion, that’s the formula of good life and good business too. Creating and undertaking are passions.
    • NOW OR NEVER. Those who feel the urge to create and undertake, but against these deep impulse, put off or avoid this commitment, will end feeling melancholy and sad passions.
    • BE YOURSELF AND BE FREE, don’t work for anybody but yourself and don’t listen to the discouraging background noise of our fearful society.
    • BE SELF-CONFIDENT AND BUILD A SOLID CHARACTER. Undertaking needs a self-confidence and a solid character, even more if we consider our educational, cultural and social values and environment.
    • HAVE DREAMS AND JUST DO IT. (Have dreams, build a clear vision, consider carefully all the circumstances, and just do it.)
    • MONEY!! Be careful with the money. It’s never enough.
    • MAKE A TALENTED TEAM. Always talent, avoid sad and proud people, choose the optimistic once and spend money in a very good human resources consultant, specialized in recruitment.
    • PARTNERS? Partners only if needed and remember: money is always coward.
    • WORK A LOT!! Talent and money are not enough without everyday work, work and hard work.

    Now, you can compare this with Fernando Trias de Bes says about the key reasons for failure when you start a new company ¿Are the more or less the same? I think basically yes. However, Trias de Bes considers some interesting causes in addition.


    Fernando Trias de Bes, MBA professor at ESADE and writer of books as “The good luck” or “Lateral Marketing”, surprised everyone with his new book: “The black book for the entrepreneur” (El libro negro del emprendedor) were he explains his theory about the main factors for failure when starting a new company. This is based in his own experience, his work at the MBA school (ESADE) and many “chats” with different experienced entrepreneurs. He got to compile 14 key reasons to fail. Those factors are the following:

    • About the person:

    1 entrepreneur with a reason but without motivation
    2 not to have an entrepreneur mentality
    3 not to have a strong character to fight when is needed

    • About the business partners:

    4 count with them when in reality you don’t need partners at all
    5 choose partners without an election criteria
    6 share the same percent of capital when everyone is not giving the same value
    7 lack of communication and understanding between partners

    • About the business idea:

    to think that success only depends on that "great idea"
    9 start within fields you don’t like or lack knowledge
    10 start within non attractive fields (saturated, little growth)

    • About the impact in your family live:

    11 make the business dependant on your family economy and material needs
    12 not knowing the impact that starting a business will have in the quality of your personal live (time for your family, hobbies, …)

    • About management of the new company:

    13 create business models in which you don’t get benefits quickly and neither in the long term
    14 to be an entrepreneur (want to create something all the time) instead of a business man (like to manage growth) and not know the right moment to retire




    Publicado el 14.6.2010 por Ricardo Alonso Maturana

    Social data; web semántica; y datos abiertos y enlazados: ¡knowledge internet! / Social data, semantic web, opend and linked data: knowledge internet!


    El proceso general de digitalización de la realidad está generando una capa de representación de la totalidad de las cosas y personas, pero también, y esto resulta especialmente relevante, de sus relaciones. A, por ejemplo, ‘Juan Rulfo’ y, pongamos, ‘El llano en llamas’ les une el hecho de que el primero es el AUTOR del segundo, lo que queda expresado mediante la proposición: ‘Juan Rulfo es el autor de ‘El llano en llamas’ Como veremos, la web semántica posibilita computar (deducir, razonar, buscar...) con proposiciones de ese tipo, lo que ofrece grandes posibilidades. Cuando disponemos de los datos y de sus relaciones organizados de acuerdo con los lenguajes de la web semántica (de los que he hablado en ¿Mis datos en manos de terceros? Ventajas de expresar contenidos con estándares de la web semántica), decimos que “los datos están estructurados". Por desgracia, la mayor parte de los datos de la web no están expresados de ese modo. Sin embargo también están estructurados, aunque no del modo en el que lo entienden las máquinas. Buena parte del trabajo consiste, precisamente, en convertir nuestros documentos HTML, .doc, etc...en datos estructurados. En cualquier caso y en buena medida, todos nuestros datos los estamos organizando de forma útil para las personas gracias a las redes sociales. De hecho, construir una web más inteligente parece la condición necesaria para manejar y utilizar de manera eficiente esa enorme y creciente cantidad de datos que está produciendo la actividad humana en las redes. A esta clase de datos categorizados y organizados gracias al trabajo social de las redes les llamaremos Social Data.


    Social Data

    Habitualmente los Social Data incorporan el punto de vista de las personas que los editan, o bien, dependiendo de su naturaleza, el de un grupo. Esta información que añade información a una información original suele conocerse con el nombre de metadatos. Los Social Data son, por tanto, metadatos construidos gracias al trabajo social en la red de las personas. Las colecciones de metadatos más habituales en las redes sociales son conocidas por el nombre de folksonomías, que expresan el punto de vista personal o la perspectiva personal con relación a una información o documento determinado. En ocasiones, las redes sociales, en especial cuando se trata de redes profesionales, incluyen puntos de vista más normativos para calificar la información: las taxonomías o los tesauros. La acción de asignar un determinado metadato a una información o documento se conoce con el nombre de etiquetar. El etiquetado social es la fuente primaria de producción de Social Data.


    Google es demasiado idiota para entender lo que la gente necesita

    Dado que la racionalidad humana tiene un carácter intencional o finalista, los Social Data aportan una información muy relevante a la hora de recuperar la información con una determinada finalidad o intención. Este es el modo general en el que los humanos desean recuperar información, pero no el modo en el que resuelven este problema los sistemas y buscadores. Pensemos por ejemplo en nuestra experiencia de búsqueda y recuperación de información con Google, el buscador más extendido. Google nos ofrece como resultado de una búsqueda una, habitualmente larga, lista de resultados o posibilidades ordenados en función de la relevancia que atribuye a cada uno de ellos. Eso no estaría mal en primera instancia, pero lamentablemente es todo lo que podemos hacer con Google: no podemos hacer una segunda pregunta relacionada con la primera. Imaginemos que me gustaría conocer qué hay sobre ‘buscadores semánticos que utilicen procesamiento del lenguaje natural y, además, lógica borrosa, escrito por mujeres en 2008, en alguna universidad de California o en San Francisco y alrededores’. Podemos intentarlo introduciendo en la barra del buscador "buscador+semántico+procesamiento+del+lenguaje+natural+lógica+borrosa+san+francisco+2008", pero como puede imaginar casi cualquiera el resultado puede resultar sorprendente. En todo caso, a nadie se nos ocurriría utilizar ‘mujer’ o ‘alrededores’ o ‘alguna universidad de california’ como criterio de recuperación de información, porque conociendo a Google no resulta una expectativa razonable.


    Google padece el síndrome del savant o del sabio. Es un idiota con algunas capacidades muy desarrolladas, un idiota inteligente capaz de comparar todos los caracteres de un texto con la secuencia que le proponemos, pero incapaz de entender lo que le solicitamos. En nuestro ejemplo, las categorías ‘documentos escritos por mujeres’ o ‘producidos en una universidad de california’ son difícilmente traducibles a una secuencia de caracteres que ofrezca la perspectiva de un conjunto de resultados aceptable. ¿Por qué no podemos interrogar a nuestro buscador sucesivamente, tal y como hacemos en nuestra vida ordinaria, para resolver problemas que requieren manejar grandes cantidades de información? Nuestro cerebro no está diseñado para manejar enormes cantidades de información, sino relativamente poca pero significativa; por eso nuestra razón produce como resultado de nuestras indagaciones largas listas ordenadas por relevancia, sino que opera estableciendo sucesivas condiciones o restricciones crecientes que nos conducen a un resultado o a unos pocos cuyo valor relativo pueda establecerse de un sólo golpe de vista. En una partida de ajedrez, nadie en su sano juicio consideraría una estrategia razonable el ordenar en una lista en función de sus posibilidades de conducir a la victoria, todas las posibles jugadas vinculadas con una posición dada de las fichas. Obviamente lo puede hacer Deep Blue, pero los humanos carecen de esa capacidad para computar. A cambio ‘saben lo que quieren’, lo que les permite considerar sólo la información útil para ese fin. 


    Contextos de interpretación de la información que pueden ser entendidos por las máquinas

    Los Social Data añaden un contexto explícito de interpretación a cualquier información o documento, por lo que constituyen el elemento social de la dimensión semántica de la web. La semántica de la web se construye según vemos social o colectivamente y esa información es útil porque las personas no son demasiado diferentes. 

    Una ontología concreta expresa un modo de categorizar, modelar o representar nuestro conocimiento con relación a un campo, entidad u objeto determinado. Lo normal es que las ontologías representen las entidades, que en nuestro lenguaje natural son denotadas mediante los nombres propios y comunes, y sus relaciones. Como lo hacemos en un lenguaje que puede ser ‘comprendido’ por las máquinas, éstas pueden entenderse entre sí (interoperar o intercambiar datos con independencia de los formatos de almacenamiento y de las aplicaciones de gestión), pero también ‘entendernos’ e interactuar con nosotros de manera inteligente; también podemos verlo al revés, gracias a las convenciones de la web semántica nosotros podemos conversar con ellas utilizando nuestras capacidades de un modo natural, esto es, podemos razonar con ellas. ¡Esto representa una gran oportunidad para todos, que se añade a las que ofrecía el etiquetado social!

    En la práctica, hay muchas cosas que necesitan conocer los sistemas, especialmente en el contexto de una red social, para poder comunicarse con sentido con las personas y para que interoperen con otros sistemas. Para que esto sea realmente posible,  precisamos que todos ellos hablen con las mismas palabras, esto es, que utilicen las mismas ontologías. A estas ontologías sobre las que existe un acuerdo (que puede ser universal, muy amplio o...menos amplio) las denominamos vocabularios. Algunos vocabularios de carácter muy general resultan especialmente importantes. Dado que los sistemas funcionan sobre la base de documentos digitalizados y descripciones de personas, las ontologías que representan nuestra idea general de lo que es un recurso o documento digital, las que modelan la descripción de una persona y aquellas que describen un sistema de categorías o tesauro resultan especialmente importantes porque permiten conectar a la mayor parte de las entidades que existen en la webEllas representan del modo más inclusivo a casi cualquier contenido que puede encontrarse en internet y por ello hacen que las máquinas y los sistemas puedan interoperar entre sí con pocas restricciones.

    La web semántica puede definirse como el conjunto de convenciones que hace posible estructurar los datos contenidos en los distintos formatos de documentos (que generalmente están desestructurados, lo que en realidad significa que no pueden interpretarlos las máquinas), con el fin de que tanto las máquinas como las personas puedan interactuar (interoperar) entre sí de un modo más humanizado, intuitivo, eficiente y satisfactorio que lo que sucedía con la web basada en la computación. Es la base para el desarrollo de una web más inteligente y...autoconsciente.


    Open Data y Linked Data

    Open Data designa una filosofía y práctica que persigue que determinados datos estén disponibles de forma libre a todo el mundo, sin restricciones de copyright, patentes u otros mecanismos de control. Los datos pueden estar abiertos y, cuando no se utilizan los estándares de la web semántica, no resultar aprovechables por terceros.

    Aún cuando los datos de una determinada aplicación web se expresen de acuerdo con los estándares de la web semántica caben diversas posibilidades:

    • Los datos pueden ser abiertos, pero no estar enlazados
    • Los datos pueden ser enlazados, pero no estar abiertos

    La posibilidad de datos que sean tanto abiertos como enlazados es cada vez más viable, tanto desde el punto de vista tecnológico, como de negocio. La Web Semántica sólo puede funcionar con datos que sean tanto abiertos como enlazados. Nos referimos a esto en un post anterior. En esta entrada, la figura representa el grafo del conjunto de iniciativas y aplicaciones enlazadas que constituyen la Linked Data Web o de la Web de los datos abiertos y enlazados

    Linked Data Web implica una manera de publicar contenidos en la Web que:

    •  favorece la reutilización
    •  reduce la redundancia
    •  maximiza la conectividad (real y potencial)
    •  hace posible el “efecto red” a la hora de añadir valor a los datos


    En definitiva:                                                    Linked Data = Open Data + Open Standars


    La web semántica tiene ya un tamaño considerable, que irá aumentando a medida en la que se vayan estructurando los datos de más espacios de la web (a la par que se crean espacios con los datos ya estructurados). Los datos estructurados permiten estrategias de búsqueda que en lugar de ordenar una lista de posibles soluciones en función de la relevancia (en lugar de obligar a los humanos a entender la lógica de las máquinas), permite ir razonando hasta localizar el resultado o pequeño número de resultados que responde a las restricciones o condiciones del razonamiento. En definitiva, permiten las búsquedas basadas en el razonamiento o búsquedas facetadas.

    Por otro lado, la web semántica posibilita ofrecer como resultado de una determinada búsqueda el conjunto de contextos relacionados con ella, como por ejemplo personas relacionadas, documentos relacionados, imágenes relacionas, etiquetas o metadatos relacionados, etc…Esto posibilita el poder desarrollar y evolucionar las búsquedas desde la perspectiva humana de la exploración.

    En resumen, Linked Data Web sería:

    •          Base de datos global
    •          Diseñada para que las máquinas ‘hablen’ y ‘piensen’ al modo humano·        
    •          Los objetos que maneja y conecta representan cosas (como personas, películas, imágenes, libros, plantas, etc…, esto es, cualquier cosa que podamos representar mediante una ontología) y no, como ocurre en la web HTML, documentos (páginas web)
    •          Los enlaces representan relaciones entre entidades o 'cosas'
    •          Para ello se precisa de un alto grado de estructuración en las descripciones de esas entidades
    •          Es preciso, por tanto, que la semántica de las cosas sea explícita

    Las tecnologías o estándares asociados con su desarrollo serían: URIs, HTTP, RDF, RDFS/OWL


    El siguiente grafo muestra el conjunto de iniciativas que forman parte de la web semántica y sus diferentes grados de interacción




    Imagen: linkeddata.org


    www.gnoss.com es un sistema de redes sociales enlazadas cuya ontología se expresa de acuerdo con los estándares de la web semántica. gnoss.com, además de ser un espacio Open Data, es un espacio de Linked Data, esto es, sus datos son enlazables, interpretables y expresables desde cualquier web que trabaje dentro de los estándares de la web semántica. Pero la web semántica son, en el fondo, social data pues la semántica expresa siempre un acuerdo formal o informal entre personas: no hay posibilidades de entendernos sin una idea común acerca del significado de las palabras. Esta semántica se va construyendo, es un ‘work in progress’ que se puede expresar de diferentes modo, pero que finalmente, dentro de la lógica evolutiva de la web semántica, tiende a concretarse en vocabularios estándar. Estos estándares son los que en el corto y medio plazo se irán imponiendo para resolver los profundos problemas de aislamiento a los que nos somete el no hacerlo así. Los sistemas de salud, las administraciones públicas y las grandes corporaciones están asumiendo la necesidad de trabajar con ellos si quieren aprovechar el potencial de sus sistemas y de la relación entre ellos y las personas. Poco a poco lo irán haciendo el resto de las empresas y personas. 




    The general process of reality digitalization of is creating a layer of representation of all things and people, but also of their relationships, and this is particularly important. For example, ‘Juan Rulfo’ and, let’s say, ‘The Burning Plain and Other Stories’ are linked by the fact that the first one is the author of the second one. This is expressed by the proposition: “Juan Rulfo is the author of The Burning Plain”. As we shall see, the semantic Web enables compute (deduce, reason, search...) with proposals of this kind, which offers great possibilities. When you have the data and their relationships organized according to semantic web languages (I have spoken about them in My data in the hands of others? Advantages of expressing content with semantic web standards), we say that “data are structured”. Unfortunately, most web data are not expressed in that way. However, they are structured too, but not in the way that the machines understand. Much of the work involves specifically converting our HTML, doc, etc. documents into structured data. In any case and to a large extend, we are organizing all our data in a useful way for people thanks to social networks. In fact, building a smarter web seems the necessary condition to efficiently manage and use the huge and growing amount of data that human activity is generating on the networks. This kind of data which are categorized and organized through the social work on social networks will be called Social Data.


    Social Data

    Social Data usually incorporate the perspective of people who edit them, or, depending on their nature, that of a group. This information that adds information to original information is generally known as metadata. Therefore, Social Data are metadata built thanks to social work in the people network. The most common metadata collections on social networks are known by the name of folksonomies, which express a personal view or personal perspective in relation to a particular information or document. Social networks, especially when it comes to professional networks, sometimes include regulatory viewpoints to describe the information: taxonomies or thesauri. The action of assigning a particular metadata to some information or to a document is known as tagging. Social tagging is the primary source of Social Data production.

    Google is too stupid to understand what people need

    Given that human rationality has an intentional or finalist nature, the Social Data provide very important information when retrieving information with a specific purpose or intent. This is the general way in which humans want to retrieve information, but not the way in which systems and search engines solve this problem. For instance, let’s consider our experience of information search and retrieval with Google, the most widespread search engine. As a result of a search, Google shows us a typically long list of potential results sorted according to the relevance it gives to each of them. At first, that would not be a bad option, but unfortunately that’s all we can do with Google: we cannot ask a second question related to the first one. Imagine that  you would like to know what’s on ‘semantic search engine using natural language processing, and also fuzzy logic, written by women in 2008 in any university in California or in San Francisco and surroundings’. We can try entering into the search bar: "search+engine+semantic+processing+of+natural+language+fuzzy+logic+san+Francisco+2008". But almost anyone can imagine that the result might be rather surprising. In any case, no one would use ‘woman’ or ‘surroundings’ or ‘any university of california’ as a criterion for information retrieval, because it is not a reasonable expectation when you know Google.

    Google has the savant syndrome. It’s an idiot with some highly developed skills, an intelligent fool that can compare all the characters in a text with the sequence that we are proposing, but unable to understand what we are asking. In our example, the categories ‘documents written by women’ or ‘produced in any university of california’ are difficult to translate into a sequence of characters that offers the prospect of an acceptable result set. Why can’t we query our search engine successively, as we do in our ordinary life, to solve problems that require handling large amounts of information?

    Our brain is not designed to handle huge amounts of information, but relatively small, but significant. That’s why our reason doesn’t produce long lists sorted by relevance as a result from our inquiries, but it operates establishing successive conditions or restrictions that lead to one result or to a few ones of which relative value can be established at a glance. In a game of chess, nobody in his right mind would consider as a reasonable strategy ordering in a list, according to their ability to lead to victory, all the possible moves associated with a given position of the chess pieces. Obviously, Deep Blue can do it, but humans lack the ability to compute. In turn, they ‘know what they want’, what allows them to consider only the relevant information for their purpose.


    Contexts of information interpretation that can be understood by machines

    Social Data add an explicit context of interpretation to any information or document, so they are the social element of the semantic dimension of the web. The Semantic Web is built according to social or collectively perspective and that information is useful because people are not too different.

    A specific ontology expresses a way of categorizing, modeling or representing our knowledge in relation to a determined field, entity or object. Ontologies normally represent the entities, which in our natural language are denoted by proper and common names, and their relationships. As we do it in language which can be ‘understood’ by the machines, these ones can understand each other (interoperate or interchange data regardless of storage formats and management applications), but also ‘understand’ us and interact with us intelligently. We can also see the opposite: thanks to the conventions of the Semantic Web, we can talk with them using our skills in a natural way, that is, we can reason with them. This represents a great opportunity for all, in addition to those offered by the social tagging!

    In practice, there are many things systems need to know, especially in the context of a social network, to communicate meaningfully with people and to interoperate with other systems. For this to be really possible, they all must speak the same words, that is, using the same ontologies. These ontologies for which there is an agreement (which may be universal, broad or… narrower) are called vocabularies. Some very general vocabularies are particularly important. As the systems operate on the basis of digital documents and descriptions of people, the following ontologies related to them are specially important because they allow you to connect most of the entities that exist on the web: a) ontologies that represent our general idea about a resource or a digital document, b) the ones that shape the description of a person and c) those that describe a system of categories or thesaurus. They represent the most inclusive way to almost any content that can be found on the Internet. Thus, they make the machines and systems to interoperate with each other with few restrictions.

    The Semantic Web can be defined as the set of conventions that makes it possible to structure the data in different document formats (which are usually unstructured, what actually means that machines can’t interpret them), so that both machines and people can interact (interoperate) with each other in a more humane, intuitive, efficient and satisfying way than what happened with the computer-based web. This is the basis for the development of a more intelligent and… self-aware web.


    Open Data and Linked Data

    Open Data means a philosophy and practice that pursues that certain data are freely available to everyone without restrictions of copyright, patents or other control mechanisms. Data can be open but not usable by others, when semantic web standards aren’t used.

    Even when data from a particular web application are expressed in accordance with the standards of the Semantic Web, there are several possibilities:

    • Data can be open, but not be linked
    • Data can be linked, but not be open

     The possibility of data being both open and linked is increasingly viable, both from a technological and business standpoint. The Semantic Web can only work with data that are both open and linked. I referred to this in a previous post. The below figure represents the graph of all the linked initiatives and applications which form the Linked Data Web.

    Linked Data Web involves a way to publish content on the Web that:

    • promotes reuse
    • reduces redundancy
    • maximizes connectivity (real and potential)
    • makes possible the ‘network effect’ when it comes to adding value to data

    In short:  Linked Data = Open Data + Open Standars

    The Semantic Web has already a considerable size, which will increase as data from more web sites get structured (at the same time that spaces with already structured data are being created). Structured data allow search strategies to reason until they reach the result or small number of results according to restrictions or conditions of reasoning, instead of ordering a list of possible solutions depending on the relevance (rather than forcing humans to understand the logic of the machines). In short, they allow searches based on the reasoning or faceted search.

    On the other hand, the semantic web enables to offer as a result of a particular search a context set associated with it, such as related people, documents, images, tags or metadata, etc... This makes it possible to develop and evolve search from the human exploration perspective.

    To sum up, Linked Data Web would be:

    • Global database
    • Disigned for machines to ‘talk’ and ‘think’ the human way.
    • Objects that it manages and connects represent things (like people, movies, images, books, plants, etc…, that is, anything that can be represented by an ontology), but not documents (web pages) as it occurred in the HTML web
    • Links represent relations between entities or ‘things’.
    • This requires a high degree of structure in the descriptions of these entities.
    • Therefore, it is necessary that the semantics of things is explicit.

    Technologies or standards associated with its development would be: URIs, HTTP, RDF, RDFS/OWL

    The following graph shows the set of initiatives that are part of the semantic web and their different degrees of interaction


    Image: linkeddata.org

     www.gnoss.com is a system of linked social networks whose ontology is expressed in accordance with the standards of the semantic web. gnoss.com, besides being an Open Data space, is a Linked Data space, that is, its data can be linked, interpreted and expressed by any site that works within the standards of the semantic web. But the semantic web is in the background, social data, since the semantics expresses always a formal or informal agreement between people: there is no chance of understanding without a common idea of the meaning of words. This semantics is being built, is a work in progress which can be expressed in different ways, but finally tends to be translated into standard vocabularies, within the evolutionary logic of the Semantic Web. Those standards are the ones that will go imposing in the short and medium term to solve the deep problems of isolation generated when not doing it so. Health systems, public administrations and large corporations are taking the need to work with them if they want to exploit the potential of their systems and the relationship between them and people. The rest of the companies and individuals will be doing it little by little.