Slashdot Mirror
Greatest Task of Web 2.x: Meta-Validation
CexpTretical writes "This Technology Review article about Web 2.x problems fails to mention the 800 pound gorilla in the room when it comes to fulfilling the dreams of the Semantic Web — i.e., assumptions about the validity of metadata or tagging schemes. We can add all of the metadata and/or tags we want to web resources but that does not mean that the 'data about the data' honestly or accurately describe the resource or are 'about the data' at all. This is why Google does not place much importance on the metadata already contained in HTML document headers for search ranking, because it cannot be trusted. And to validate it would require more effort than to search and index that data from scratch. Ensuring or verifying the validity of metadata would be a task equal to that of initially creating it, but would have to be repeated on an ongoing basis. Hence all of the talk about 'trusted networks,' which then require trusting the gatekeepers of those networks. Talk about 'semantics.'" Slashdot's moderation and meta-moderation offer one example of getting useful metadata in a non-trusted environment.
The tagging system might be a better example, or at least an example of mostly useless meta information.
What about the removal of accurate metadata, such as Slashdot's disabling of the "dupe" tag?
Especially here at Slashdot where a certain type of groupthink is very prevalent, it's not so much a matter of whether a comment is insightful or interesting so much as it adheres to the consensus view of the moderators. A non-conforming view is labeled 'Troll'. So in one sense, the metadata provided by the moderation system is useful in that you can tell at a glance how well a comment conforms to the Slashdot zeitgeist just by looking at its moderation score.
However since posts lower than zero do not get displayed automatically, views that are unappealing to the Slashdot community are relegated to obscurity regardless of their validity and correctness.
Linux sucks.
I thought it was "elephant in the room"? Googlefight!. We're talking orders of magnitude here... Please tell me that lame TV commercial that botched the idiom isn't starting a trend? I think 800 lb gorilla should remain as the Urban Dictionary's "an overbearing entity in a specific industry or sphere of activity" and not expand to the more abstract, from Wikipedia, "an obvious truth that is being ignored"
You're completely correct. Most people don't moderate based on how a post or a story meets a certain set of criteria. They only ever get to the level where they "agree" with a story, or "disagree" with it.
When it was first becoming popular, I used Digg for a few weeks. Various people would post comments in stories I had submitted, saying how they had just "buried" the story as "OK, This is Lame" because they disagreed with it. Of course, that's now how Digg is intended to work. It's about a story's merit as a story, not about how it may conflict or agree with one's opinion.
I've seen other people suffer a similar fate. John C. Randolph, who many of us know for his past work at Apple, is often the victim of that sort of stupidity. Unlike 99.99999% of the Digg users, he has some clue as to how Apple works, and what sort of projects they're working on. Yet time and time again he's the victim of morons who outright claim his stories are "inaccurate". Unfortunately, there are so many morons that they completely outnumber those of us who know who jcr is, who know of his great work, and who know how perfectly accurate his information is.
At least here at Slashdot, there's some limit as to who gets moderating privileges. It tends to be only the most intelligent individuals. Contrast this to Digg, where any 12-year-old cocktard is given the ability to moderate stories that tower above their intellect and understanding.
Working with metadata from a non-trusted community is a few orders of difficulty harder than working with trusted metadata. All the examples from non-trusted user groups that I've seen are either 1) only able to track fairly simple data or 2) ambitious but disappointing. I'd put Slashdot's moderation and metamoderation in the first category. Relevance, quality, and a few kinds of description are possible, but these are fairly simple things to track. Most internet resources would require metadata that is much harder to validate to be useful.
A primary example of this that comes to my mind is the current crop of music recommendation services. The idea behind these sites is that they can, through one of various methods, recommend music to you based on what you like. I've experimented somewhat extensively with Pandora and Last.fm, and the difference in the quality of their suggestions is amazing.
Last.fm uses community data for recommendations. It tracks tags that users attach to songs and the collection of artists that each user listens to. Based on what artists you have listened to or which tags you select, it attempts to point out other artists you might like.
Pandora makes recommendations based on musical qualities. The data the service uses comes from the Music Genome Project, which paid people who have studied music to catalogue the musical qualities of songs in their database. Employees listen to songs and select which attributes are applicable to the song from a list of hundreds of attributes. To use the service, you enter some songs and artists that you like, and based on the musical attributes of those songs and artists, it recommends other songs you might like.
The results that the services provide, at least in my case, are like night and day. Last.fm's recommendations are heavily influenced by what's popular and how a common user would categorize an artist or song. They sort-of hit the right areas, but it doesn't get much better than Amazon's recommendations. Pandora's recommendations always seem to be more on target, even though it uses only a few artists or songs that you enter at the start, in contract to Last.fm, which can use my entire play history.
I guess a lot of this can be chalked up to the difference between association and relation - without some type of new innovation, it seems that community-based metadata can only be based on association, which is a far cry short of relation. Yes, it is a type of relation, but a set of data has qualities that a few simple tags from users are not going to be able to touch. It seems to me the next generation of metadata will only be possible when we can figure out a way to get the sort of data that Pandora uses from a community group. It's a daunting challenge that tagging and simple user activities like the Google Image Labeller have just started to slightly touch.
For metadata to be useful at all, there has to be some way to come to a consensus, and the most logical way to come to a consensus is by what the majority thinks. However, there are too many examples where the majority is wrong for metadata to be truly useful in my opinion.
Here on Slashdot, there is a selection process and a reputation system that determines who has the ability to moderate. How does this "Web 2.0" address the fact that anyone can attach and moderate tags?
We all know what to do, but we don't know how to get re-elected once we have done it
You can't search on them, you don't have any incentive to tag them for yourself (since everyone is limited to the same 5 tags or so), and you can't get "More articles like this". Is it any shocker that they've turned into a veritable festival of in-jokes which provide no information you couldn't get from reading the summary? Heck, after you've read the headline you can provide all the tags:
"Is Linux ready for desktop?"
yes, no, fud, notfud -- and it would be marked omgponies, dupe, and thistagisfreakinguseless if any of those options weren't automatically stripped.
Its almost like tags are designed to be useless here, in a way that they're not with delicious (put the periods in wherever you want them -- I use www.delicious.com and I am so very glad it works). I can use delicious as a "Hmm, I want to read this later" bookmark-shared-across-machines, to categorize Java samples for my own use later, and to do things which are of use to *me*. The social aspect grows naturally from the personal uses, because when you mark Sun's whitepaper as being about Java or this photo on flickr as being of sakura everyone else gets to piggyback on your diligence. But if there isn't any personal use possible then tagging is just textual autoeroticism.
You can mark me fud and omgponies if you want.
Help poke pirates in the eyepatch, arr.
So, I'm really dubious about one of the myths about Google and metadata: that Google doesn't use metadata because it's unreliable.
Google does, in fact, use metadata -- tons of it. Google uses explicit metadata built into headers (like the description, robot control); it uses the rel-license microformat; and it uses titles and h1 headers. It also uses some crucial metadata that's not self-reported by the Web site -- namely, the number and text of links inbound towards a page. It also uses metadata in HTTP headers.
Google also uses lots of data that is unreliable or could be dishonest. After all, there's a huge dark business of blackhat SEO that has its sole intention to trick Google's bots into thinking pages are more important (or are on a different subject) than they actually are. There is no particular part of an HTML page or any other Web resource that cannot be a lie. Web spiders have to deal with this all the time, and they have to balance the information they get from different data sources to determine what's true and what's not.
It's true that Google's search results don't depend as heavily on the specific meta keywords the way many first-generation search engines did. But I think that's more a consideration of the remarkable naivete of early search engines than anything else.
Evan Prodromou | evan@prodromou.name | http://evan.prodromou.name/
An example of unaccountable, gameable metadata that generates untrustworthy info that is almost as useless, through abuse, as it is useful.
Slashdot's moderation could:
Those few improvements could introduce some accountability and feedback into the now mostly abused meta/moderation system. Until then, Slashdot has little to teach the world about the right way to accumulate useful metadata in an untrustworthy environment.
--
make install -not war
I think the tags are great; they let me get my whole article's worth of Slashdot groupthink in just a few seconds of skimming.
... "
For instance: "IT: Vista Designed to Make Malware Easy" is tagged "troll, fud, vista, notfud, microsoft". I mean -- that's it! That's the whole discussion right there. Point, spastic head-nodding, counterpoint, rehash of the original article. Thank you sir, may I have another.
I'm hopeful that on some future "Slashdot Mobile," they'll remove everything but the titles and tags, and display it as a feed. Maybe after that, they'll even get rid of the titles, so you can just see a constant stream of tags.
Forget a boot stamping on the face of humanity; that's the future for you: "microsoft fud notfud troll itsatrap google dupe evil internet hardware nvidia slashvertisement pigpile dupe sun esr fud ubuntu dupe microsoft dupe
"Ladies and gentlemen, my killbot features Lotus Notes and a machine gun. It is the finest available."