We need a tag evolution

May 8, 2008

Nova Spivack, the founder of Twine, held an interesting presentation about the future of the Web on the Next Web conference in Amsterdam. He thinks that we are currently in the process of Internet evolution in which tags are having an increasing significance. He predicts that in the next 10 to 15 years tags will have an increasingly important part while keywords will gradually disappear.

An interesting discussion about the subject took place on Techcrunch when Eric Schonfeld posted this thread asking the question “Is Keyword Search About To Hit Its Breaking Point?“. 97 comments have been posted so far and one of them especially caught my attention:

John Clarke Mills

Tags are nothing new, that is for sure. But what if you could tag an object, or entity, with another object. So instead of tagging objects with strings, which falls back on a simple full-text search, you could tag something with an actual representation?

I think that John has really nailed the point.

The problem with both keywords and tags is that they are just words. But what would happen if, instead of words, we used objects? What if we used unique concepts that would always and everywhere have the same name and would refer to the specific object?

Wikipedia & DBpedia

How can we reach an agreement on the names of such a large number of concepts? Well, it’s already been done and can be found in the largest collection of concepts in the world – Wikipedia. Wikipedia, besides having a standardized way of displaying articles, also has a standardized way of naming titles, which have been created and are constantly perfected by social consensus.

Currently there are over 2.36 million articles in English language on Wikipedia. The titles of Wikipedia articles are unique and cover almost all the concepts we can imagine.

However, the “problem” with Wikipedia is that it is not made for machines, but for humans. Its search capabilities are limited to full-text search, which only allows very limited access to this valuable knowledge-base.

Fortunately, there is DBpedia, which represents community effort to extract structured information from Wikipedia and to make this information available on the Web. The DBpedia.org project uses the Resource Description Framework (RDF) as a flexible data model for representing extracted information and for publishing it on the Web.

For example, the web page about Semantic Web on Wikipedia looks like this, while on DBpedia it looks like this (there is also an alternative that is easier to read by humans).

This practically means that based on the name of the tag we can learn more useful information about that tag, its properties and connections to other tags. That is why I believe that DBpedia web pages are good representatives of the “objects”, the references of which will be tags.

Characteristics of new tags

Unique name

Unlike classic tags, which are just words, new tags represent references to unique concepts that have their own URL. For example, the tag “Coca-Cola” has a reference to URL http://dbpedia.org/data/Coca-Cola http://dbpedia.org/resource/Coca-Cola (actually, the name of the tag is just the last part of the URL).

So, instead of having different tags for the same concept, which is the case with classic tags (cocacola, coca-cola, coca+cola, CocaCola) there will be just the one unique “Coca-Cola” tag.

Disambiguation

But what if we wanted to add a tag that has more than one meaning? Let us look at the example of “library”. What are we referring to – “a collection of books”, “collection of subroutines used to develop software” or “the Seinfeld episode called ‘The Library'”?

It is simple – we’ll just use different tags: Library,
Library (computing)
and The Library (Seinfeld episode).

Tag properties and its connections to other tags

New tags are references to objects, and objects, as we know, have certain properties. In DBpedia there are some properties that are common to all tags, such as: an abstract, a picture (if existing), labels in multiple languages, type and subject to whom the tag belongs.

For example, if we look at DBpedia page for Keith Richards we can learn some additional properties about him (year of birth, type of voice, genre of music he plays…) as well as his connections to other tags (born in Dartford, current member of The Rolling Stones, plays Fender Telecaster and Gibson Les Paul, occupation: Music producer, Musician and Songwriter…).

Classification of tags

As I mentioned earlier, tags belong to different groups and form a structure. A system that supports such tags has an advantage over other systems because it automatically classifies tags and so “knows” what Microformat, RDFa, Web Ontology Language and Thesauris have in common. They all belong to the subject Knowledge representation. That’s why with Faviki it is possible to follow the content by subject and not only using one tag (see Knowledge representation page).

Conclusion

I think that tags will truly dominate in the near future. But those will not be the tags that we are used to, but their “smarter” offspring. I believe that the results of this evolution will make the foundation for the future Internet which will handle objects and their properties instead of just web pages. Present situation is not ideal but it makes a good foundation for the development of the universal language that could connect people and the Internet in new and exciting ways.

21 Responses to “We need a tag evolution”

  1. John Bäckstrand Says:

    Great post. Dbpedia is an interesting project in and of itself, but I really, really think wikipedia needs to be “more semantic”, too. It seems like its getting, but a bit too slow for my taste.

  2. Matt Foster Says:

    This idea may already be well on its way, the idea is that instead of tagging objects with words you tag them with URLs. This reminds me of using Google’s “related” operator.


  3. Glad you liked my comment although I should mention I am a little biased due the fact that I work for Radar Networks🙂. The tags as objects project has been the apple of my eye for a while now. Our system does currently implement tags this way, however; they are not exposed in this fashion yet. More to come later.

    I liked your post though, you definitely grasp many of the ideas and goals of the Sematnic Web. Cheers.


  4. Thank you all for comments.

    @John Bäckstrand
    I agree Wikipedia needs to be more semantic. However, I don’t think Wikipedia will solve this problem. I believe that semantic web will be brought to ordinary users by new startups.

    @Matt Foster
    I’m not sure I understand you completely. Using ‘related’ operator shows similar websites, but that doesn’t solve the problem of universal concepts.

    @John Clarke Mills
    I’m glad you’ve read the post! It’s good to hear that Twine implements tags this way. However, it must be a hard job to convert page content into the tags.


  5. […] we are currently in the process of Internet evolution in which tags are having an increasing signifihttps://faviki.wordpress.com/2008/05/08/9/References – Subroutines – PEOPedit Reference to subroutines. Slightly different from the three […]


  6. Hi, and thanks a lot for this much needed tool ! Here is my question:

    Some concepts do not exist in DBpedia because they have no Wikipedia item. What are my options if I want to use such concepts ? For instance, how to express that an item is about “Köln Easterhegg 2008”, an event Wikipedia has no clue about ?

    I imagine a mashup between Faviki and del.ico.us that would ally the reliability of an ontology with the expressivity of a folksonomy. An item would be the combination of:
    – A Faviki item
    – A del.icio.us item if more expressivity is needed, with a tag identifying the Faviki item (for instance a faviki_d_818 tag)

    The important thing is to really make the effort to find the concept in DBpedia. Only when the concept does not exist in DBpedia can a weaker concept definition be used.

    Any thought about that ? Thanks again !


  7. Having users insert correct tags is essential for Faviki. Thus, I feel the need to suggest various things I think could improve the tag insertion user interface:

    I want to tag a page as being about “Representational State Transfer” (aka REST). So I begin typing REST and I am given a list of possible matches. Unfortunately, this list does not include “Representational State Transfer”, even though it is on the Wikipedia disambiguation article.

    1) When I begin typing a tag name, the suggestion list is to short. In case an ambiguous keyword has been typed, the suggestion list should at least include the disambiguation article’s concepts.

    2) The suggestion list is too short, it could be split in navigable pages just like Google results.

    3) Disambiguation pages are not concepts themselves, they should not appear directly, but their content is particularly interesting. They could appear in a special way in the suggestion list.

    When I start typing “tupl”, I see among others tuple and tuplet, and I ask myself: am I sure “tuple” is the concept I want to express ? What is this “tuplet” concept that maybe fits better ? So I click on “tuplet” to learn more, but nothing happens. A while after, I am about to add a “SOAP” tag, but I hesitate: even though SOAP is a protocol for me and my colleagues in my little environment, it is probably not a universal concept, maybe for most people (or just in another industry) it means something else ?

    4) When the mouse is over a suggested concept, it would be nice to have a small popup appear, with the DBpedia description of this concept (which is usually the first paragraph of the Wikipedia article).

    5) I was about to ask for a “recent tags” section in the suggestion list, but it seems to exist already, mixed with what I can imagine are “popular” tags🙂

    Thanks ! Nicolas.


  8. Two minor issues so far:

    I think there is a problem with the bookmarklet when trying to tag https websites. I tried with Firefox and Opera on different https websites: the bookmarklet loads forever, the usual GUI is not shown.

    I can not add the “Triplestore” tag, even though this concept is defined in Wikipedia. That is because DBpedia did not parse it, probably because the article has no infobox. I guess such issues are to be dealt in DBpedia.

    Do you intend to set up a collaboration space (mailing list, or bug tracker, or wiki) in order to enable a Faviki community ?
    For those interested, there is an enthusiastic discussion about Faviki on the DBpedia mailing list.


  9. There is a text overlap glitch (at least with Firefox).
    How to reproduce:

    1) Open the main page.
    2) Click on your name in the upper right side of the page, it opens the list of your links.
    3) Put the mouse over a link, click on the orange “edit” button that appears.
    4) Click “edit title and notes”.
    5) Click “close”.
    6) Now you see that the link’s entry is unreadable because various labels overlap.

    Thanks !
    Raoul Nicolas.


  10. Hi, and thanks a lot for this much needed tool !

    Thank you for your time and effort🙂. I’m glad you use Faviki!

    Here is my question:
    Some concepts do not exist in DBpedia because they have no Wikipedia item. What are my options if I want to use such concepts ? For instance, how to express that an item is about “Köln Easterhegg 2008″, an event Wikipedia has no clue about ?

    You can describe it with tags that exist. I agree that Faviki limits expressivness but that’s the whole point. I believe that this lack of expressivness actually pays off in the long run – it will be much easier to find bookmarks, yours or anyone else’s.

    Having users insert correct tags is essential for Faviki. Thus, I feel the need to suggest various things I think could improve the tag insertion user interface:
    I want to tag a page as being about “Representational State Transfer” (aka REST). So I begin typing REST and I am given a list of possible matches. Unfortunately, this list does not include “Representational State Transfer”, even though it is on the Wikipedia disambiguation article.

    Agree sometimes is hard to find a tag with autocomplete list, especially when it comes to abbreviations. Google search it’s great in that cases. Type ‘rest’, click on the G button and see what’s first on the list (you can add it by clicking the ‘copy’ link).

    1) When I begin typing a tag name, the suggestion list is to short. In case an ambiguous keyword has been typed, the suggestion list should at least include the disambiguation article’s concepts.
    2) The suggestion list is too short, it could be split in navigable pages just like Google results.

    You’re probabbly right. This is not the definite height of the list (many things on Faviki are not definite), it is limited by height iframe in a bookmarklet and for the sake of simplicity. Again, Google search can be very helpful in these (ambiguos) situations.

    2) Disambiguation pages are not concepts themselves, they should not appear directly, but their content is particularly interesting. They could appear in a special way in the suggestion list.

    Yes, letting people tag with disambiguation tags is obiviously bad idea. I hope I fix it asap.

    When I start typing “tupl”, I see among others tuple and tuplet, and I ask myself: am I sure “tuple” is the concept I want to express ? What is this “tuplet” concept that maybe fits better ? So I click on “tuplet” to learn more, but nothing happens. A while after, I am about to add a “SOAP” tag, but I hesitate: even though SOAP is a protocol for me and my colleagues in my little environment, it is probably not a universal concept, maybe for most people (or just in another industry) it means something else ?

    Totally agreed. I have the exact same problem when use Faviki. In such cases I usually search with Google – Google returns tags which are often the most representative ones for the search term.

    4) When the mouse is over a suggested concept, it would be nice to have a small popup appear, with the DBpedia description of this concept (which is usually the first paragraph of the Wikipedia article).

    I’m thinking how to solve this problem without reducing simplicity. Hope it will be implemented soon.

    5) I was about to ask for a “recent tags” section in the suggestion list, but it seems to exist already, mixed with what I can imagine are “popular” tags

    First are listed already used tags by community (above the line). It works for now.. in the future, maybe you’ll get used tags only by you and your friends.

    I think there is a problem with the bookmarklet when trying to tag https websites. I tried with Firefox and Opera on different https websites: the bookmarklet loads forever, the usual GUI is not shown.

    Yes, this is a problem. Do you have an idea how to solve it?

    I can not add the “Triplestore” tag, even though this concept is defined in Wikipedia. That is because DBpedia did not parse it, probably because the article has no infobox. I guess such issues are to be dealt in DBpedia.

    .. and there are new Wikipedia pages not included in DBpedia.. Working on that problem now.

    Do you intend to set up a collaboration space (mailing list, or bug tracker, or wiki) in order to enable a Faviki community ?

    Deffinitely. Expect that soon.

    There is a text overlap glitch (at least with Firefox).

    Just squeezed the bug !

    Thanks,
    Vuk


  11. Thanks for your answers Vuk !
    Some comments on your answers:

    A> DBpedia expressiveness:
    It is fine for popular things, and indeed people will more easily find popular URLs. I have a sense of limited expressiveness when I am working on very specialized topics, most of them having no Wikipedia coverage, or maybe just a paragraph or sentence in Wikipedia. I think bookmark sharing is especially valuable in all of the very specialized fields, with small communities. Hopefully, Wikipedia grows very fast and will cover more and more fields with more and more details🙂

    B> Difficulty to find a resource:
    Indeed, the Google button is great🙂

    C> Suggestion list too short:
    Have you considered an AJAX scrolling list ?

    D> Hesitating between two resources:
    D> Mouse-over short description of resource:
    Maybe the short description could appear somewhere on the right of the “Save” and “Close” buttons ?

    E> Recent tags:
    I was thinking about my own recent tags, because most of the time I explore many websites about a particular topic, and then move to another topic.

    F> Problem with HTTPS:
    I don’t have the source code of ajax/edit_entry.php unfortunately… is it open source ?

    G> Concepts in Wikipedia but not in DBpedia
    It’s a great opportunity to get more people involved in DBpedia🙂

    H> Collaboration space
    Wikia offers free wiki hosting, that could be an easy and flexible start.

    I> Overlap
    Is the fixed version in production right now ? The bug does not seem fixed to me. Please send me your email (mine is nicolas.raoul at gmail) and I will send you a screenshot.

    Have a nice day in Beograd (love this city!)🙂


  12. Hi there, congrats for your site.
    Two little comments:
    first, the link on the “About” page (http://www.faviki.com/pages/about/) to the contact page is broken (the last link under “send it here”).
    Second, how can I invite friends?


  13. @Carsten

    Thanks, the link is fixed.
    to invite friends: Click on your name in the upper right side of the page, it opens your page – the list of your links. Then click on the friends tab and you will see the button ‘invite friends’. Type in emails and click on ‘send’ button.


  14. Vuk i think sindice could help you find notable URIs for notable and less notable thing.

    We index DBPedia, as a start, but then add many other datasets e.g. DBLP (which has the names of most computer scientists in the world) Geonames, which has names of locations, linkedin microformats have a lot of professionals etc etc.

    The query language has quite some options to refine the outcome and/or we’ll be happy to cook an api for you with just selectedly “clean” datasets.🙂 just let me know.. we’re all exited to see the (s/S)emantic web getting to life.


  15. @Giovanni

    Thank you very much for comment! I think Sindice is a great project and I’m glad we share the excitement about the Semantic web revolution🙂

    I have sent you email with more details.

    Cheers

  16. Bergholt Says:

    First of all, congrats on launching a solid product with a great intuitive feel to it. I really like the integration of the Zemanta engine.

    Are there any plans on integrating with, say, Technorati or similar services?

    Technorati seem interested in collaborating:

    “If you are a developer who is working with tags, we want to work with you! Contact us to discuss how your site, community, or tool can become better integrated with Technorati.”


  17. @Bergholt

    I am glad you like Faviki and find it intuitive! Thanks for pointing out Technorati page, the qoute about collaboration looks very interesting.


  18. […] uses the information about tags from DBpedia datasets. DBpedia extracts structured data from Wikipedia, which is constantly growing.  Last release – […]


  19. […] already wrote on this blog about the need for a tag evolution back in May, so I was happy to present the idea, that has matured in the meantime, to a wider […]


  20. […] 11, 2009 As strong believers in the semantic tagging (we wrote about it here and here), we are happy to announce that today one big  step toward realization of the idea is […]


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: