RawSugar
November 16, 2005 11:23 AM Subscribe
RawSugar - tastier than del.icio.us? Their hierarchical tagging scheme and the way they handle suggested items between users seem slick. But the social bookmarking phenomenon keeps growing, including specialty sites for academics and scientists. With Yahoo!'s My Web 2.0, it went mainstream. Google's tagging too. Heck, even Amazon's doing it. The usefulness of a social bookmarking site depends on its popularity, so is the concept getting hurt by all this balkanization? Have things gotten out of hand if there's a swiss army bookmarklet tool generator to speed things up? When will the tagging craze die down?
Tags are nothing more than keywords, and have been around for a very long time as a technique for classifying information. Tags (and keywords) are in general only as useful as the skill and consistency with which they are applied.
Give me information classified by a professional every time.
posted by seanmpuckett at 11:58 AM on November 16, 2005
Give me information classified by a professional every time.
posted by seanmpuckett at 11:58 AM on November 16, 2005
RawSugar is indeed slick, but I'm still somehow a sucker for the zen-simple interface at del.icio.us. Josh Schachter rocks, I've been a fan since memepool. He's even started pimping out his home page lately.
posted by mullingitover at 11:59 AM on November 16, 2005
posted by mullingitover at 11:59 AM on November 16, 2005
seanmpuckett writes 'Give me information classified by a professional every time.'
I think the idea behind tagging is that no matter how good that professional is at setting up the information hierarchy, there will inevitably be flaws. As the system grows, the flaws will become more cumbersome and eventually make the system suck. A folksonomy is more flexible, and is able to take advantage of the collective wisdom of the users in real time.
posted by mullingitover at 12:03 PM on November 16, 2005
I think the idea behind tagging is that no matter how good that professional is at setting up the information hierarchy, there will inevitably be flaws. As the system grows, the flaws will become more cumbersome and eventually make the system suck. A folksonomy is more flexible, and is able to take advantage of the collective wisdom of the users in real time.
posted by mullingitover at 12:03 PM on November 16, 2005
Tags are nothing more than keywords
Furthermore, if search engine AI had developed more quickly, we'd not be having a discussion about tags today because we wouldn't need people zipping around and basically "hinting" site indexes.
posted by VulcanMike at 12:04 PM on November 16, 2005
Furthermore, if search engine AI had developed more quickly, we'd not be having a discussion about tags today because we wouldn't need people zipping around and basically "hinting" site indexes.
posted by VulcanMike at 12:04 PM on November 16, 2005
Tags are good.
Tags will endure and grow until something better replaces them.
posted by nofundy at 12:06 PM on November 16, 2005
Tags will endure and grow until something better replaces them.
posted by nofundy at 12:06 PM on November 16, 2005
Tagging's been with us since the beginning, but the only early player to take advantage was Amazon. All this 2.0 stuff is about really using what we have.
posted by xammerboy at 12:07 PM on November 16, 2005
posted by xammerboy at 12:07 PM on November 16, 2005
I've been a fan since memepool
My god, Memepool! Remember how essential that site was? I haven't been there in close to two years, maybe. Do people still click on it as often as they used to? It was the original.
posted by Peter H at 12:18 PM on November 16, 2005
My god, Memepool! Remember how essential that site was? I haven't been there in close to two years, maybe. Do people still click on it as often as they used to? It was the original.
posted by Peter H at 12:18 PM on November 16, 2005
I clicked on the "Top 10" list expecting to see the top ten linked to sites and got another list of lists. Not a good way to start but hell, I'll bookmark it with delicious and check back later.
posted by KevinSkomsvold at 12:22 PM on November 16, 2005
posted by KevinSkomsvold at 12:22 PM on November 16, 2005
tagging is more than classification, it's also a vote, an indicator that something had enough value to someone for them to mark it up. Classification by a professional is unlikely to make these distinctions or at least make them so rapidly. Each method of classification has its advantages and it's a mistake to assume this is an either/or question.
Now as for the balkanization . . . why is it a bad thing? Balkanization is perhaps better referred to as self-selecting communities, which may make the value of tags greater. In fact ths is a beef I have against delicious, that I can assemble a subset of people (except via the inbox) or join a subset/group whose interests are closer to mine.
The real question is when does that bastard with some submarine patent pop up and lay claim to all this fun innovation.
posted by kingfisher, his musclebound cat at 12:26 PM on November 16, 2005
Now as for the balkanization . . . why is it a bad thing? Balkanization is perhaps better referred to as self-selecting communities, which may make the value of tags greater. In fact ths is a beef I have against delicious, that I can assemble a subset of people (except via the inbox) or join a subset/group whose interests are closer to mine.
The real question is when does that bastard with some submarine patent pop up and lay claim to all this fun innovation.
posted by kingfisher, his musclebound cat at 12:26 PM on November 16, 2005
kingfisher, I'm not sure I understand what you're arguing, so forgive me if I'm misinterpreting you, but: The fact that "balkanization" is nigh-universally read with negative connotations speaks to why it's a "bad thing."
"Balkanization", to me, implies that things are grouped together into an aggregate that isn't greater than the sum of its parts -- e.g., where metcalfe's law doesn't obtain. You could easily get that with a "user-voting" type of tagging.
I'm not saying that self-selection can't work; I am saying that it's not a holy grail. Evolutionary systems tend to produce optimal results for the system, but not necessarily for the users of hte system. If you design a taxonomy, you have the opportunity to describe how people ought to understand something. Sometimes you want to do that. For example, a user-specified tagging system could easily end up classifying 802.11g under "health risks".
Relying on keywords (i.e., tagging) for organization entails saying that it's OK to lose some things. I'm just not down with that. There are too many important things that aren't of obvious significance when they're first described.
posted by lodurr at 12:40 PM on November 16, 2005
"Balkanization", to me, implies that things are grouped together into an aggregate that isn't greater than the sum of its parts -- e.g., where metcalfe's law doesn't obtain. You could easily get that with a "user-voting" type of tagging.
I'm not saying that self-selection can't work; I am saying that it's not a holy grail. Evolutionary systems tend to produce optimal results for the system, but not necessarily for the users of hte system. If you design a taxonomy, you have the opportunity to describe how people ought to understand something. Sometimes you want to do that. For example, a user-specified tagging system could easily end up classifying 802.11g under "health risks".
Relying on keywords (i.e., tagging) for organization entails saying that it's OK to lose some things. I'm just not down with that. There are too many important things that aren't of obvious significance when they're first described.
posted by lodurr at 12:40 PM on November 16, 2005
It's not the groupware aspect of tagging that makes me love it, it is that I can maintain just a single bookmark to a site even if it concerns several different topics. No more trying to figure out if I put that book mark to the intercooler review in Projects-> Automobile-> Dodge -> Caravan or Reference-> Automobile -> Tests-> Compenents.
posted by Mitheral at 12:46 PM on November 16, 2005
posted by Mitheral at 12:46 PM on November 16, 2005
But mitheral, is that really anything inherent to do with tagging, or is it really a UI design feature? My Drupal site uses multiple-parent taxonomies that let me do exactly that.
What am I missing, here? How is tagging fundamentally different from taxonomy? It's just defined on the fly, without planning.
posted by lodurr at 12:53 PM on November 16, 2005
What am I missing, here? How is tagging fundamentally different from taxonomy? It's just defined on the fly, without planning.
posted by lodurr at 12:53 PM on November 16, 2005
lodurr writes 'How is tagging fundamentally different from taxonomy?'
They're emergent. Gestalt. Taxonomies are meant to be predictable. Tagging can go anywhere. That's the beauty of it, it doesn't assume it can foresee every possibility from the beginning.
posted by mullingitover at 1:15 PM on November 16, 2005
They're emergent. Gestalt. Taxonomies are meant to be predictable. Tagging can go anywhere. That's the beauty of it, it doesn't assume it can foresee every possibility from the beginning.
posted by mullingitover at 1:15 PM on November 16, 2005
mullingitover: They're emergent. Gestalt. Taxonomies are meant to be predictable. Tagging can go anywhere. That's the beauty of it, it doesn't assume it can foresee every possibility from the beginning.
Isn't this a bit of a straw-man concept of taxonomies? After all, the Linaean taxonomy over it's extended history has been quite comfortable with ad hoc reorganizations and additions when necessary.
Some of the links in Rawsugar were not working for me using Safari. Oh well, pretty colors at least.
posted by KirkJobSluder at 1:20 PM on November 16, 2005
Isn't this a bit of a straw-man concept of taxonomies? After all, the Linaean taxonomy over it's extended history has been quite comfortable with ad hoc reorganizations and additions when necessary.
Some of the links in Rawsugar were not working for me using Safari. Oh well, pretty colors at least.
posted by KirkJobSluder at 1:20 PM on November 16, 2005
Lodurr, I understand the connotations of "balkanization" and I am using the term a bit ironically because I felt the use of the term was not fair to the debate. There are other ways to describe what's happening, for example self-selecting communities, competing systems, etc.
In general, I am in favor of a microformat for tagging that can be passed among sites or aggregated as needed. I would also imagine that tagging systems need to allow for the formation of groups where greater specialization might lead to more accurate tags, or at least for groups where I might be more likely to find interesting or relevant results.
Now you answered your own question as to why tagging is . . . different from taxonomy (note I removed "fundamentally" because I am not sure what you mean by that). Taxonomy tells you what people "ought" to think about something; whereas tagging gives you an idea of what they do think about something. It's also very interesting for tracking interest over time, or it has value for casual terminology (e.g., "cool") that do not appear in professional taxonomies.
Once again, I don't see the two as opposing systems. I don't know why they must be. Each has advantages -- if I want to find something that a lot of people similar to me (based on tags and overlap) have found interesting, a tag system will do that. It helps in situations where there is so much information (e.g., the freaking internet) that professional classification is impossible. Think of tagging as a faceted classification system that allows for unique and poor terminilogy, but as long as you are familiar with your culture's colloquilisms, you can use.
I love more than anything else, that possibility of tagging to establish links among people and unexpected links among objects. Professional systems don't allow for as much chance (serendipity) because they are so accurate.
posted by kingfisher, his musclebound cat at 1:30 PM on November 16, 2005
In general, I am in favor of a microformat for tagging that can be passed among sites or aggregated as needed. I would also imagine that tagging systems need to allow for the formation of groups where greater specialization might lead to more accurate tags, or at least for groups where I might be more likely to find interesting or relevant results.
Now you answered your own question as to why tagging is . . . different from taxonomy (note I removed "fundamentally" because I am not sure what you mean by that). Taxonomy tells you what people "ought" to think about something; whereas tagging gives you an idea of what they do think about something. It's also very interesting for tracking interest over time, or it has value for casual terminology (e.g., "cool") that do not appear in professional taxonomies.
Once again, I don't see the two as opposing systems. I don't know why they must be. Each has advantages -- if I want to find something that a lot of people similar to me (based on tags and overlap) have found interesting, a tag system will do that. It helps in situations where there is so much information (e.g., the freaking internet) that professional classification is impossible. Think of tagging as a faceted classification system that allows for unique and poor terminilogy, but as long as you are familiar with your culture's colloquilisms, you can use.
I love more than anything else, that possibility of tagging to establish links among people and unexpected links among objects. Professional systems don't allow for as much chance (serendipity) because they are so accurate.
posted by kingfisher, his musclebound cat at 1:30 PM on November 16, 2005
kingfisher: Taxonomy tells you what people "ought" to think about something; whereas tagging gives you an idea of what they do think about something.
Well, I'll just repeat the same criticism as before. This just seems like attaching a bunch of unnecessary baggage to taxonomies for the purpose of inventing a contrast that probably isn't really there.
Taxonomies start from communities of people putting objects into categories, and only solidify through the process of consensus.
It's also very interesting for tracking interest over time, or it has value for casual terminology (e.g., "cool") that do not appear in professional taxonomies.
More baggage. Let's drop it. When you say, "X is an example of Y" you create a taxonomy.
posted by KirkJobSluder at 1:41 PM on November 16, 2005
Well, I'll just repeat the same criticism as before. This just seems like attaching a bunch of unnecessary baggage to taxonomies for the purpose of inventing a contrast that probably isn't really there.
Taxonomies start from communities of people putting objects into categories, and only solidify through the process of consensus.
It's also very interesting for tracking interest over time, or it has value for casual terminology (e.g., "cool") that do not appear in professional taxonomies.
More baggage. Let's drop it. When you say, "X is an example of Y" you create a taxonomy.
posted by KirkJobSluder at 1:41 PM on November 16, 2005
lodurr writes "How is tagging fundamentally different from taxonomy? "
Heck if I know, and I don't know what Drupal is. I know this though, stuff like the original Yahoo or Dmoz doesn't work for me nearly as well as del.icio.us even after the hassle of setting up a directory for my own personal use. The structure means I have to make hundreds of decisions about where every page should go, create cross links so I can find stuff, and still constantly not be able to find a page I read last year because I'm in a different frame of mind this week.
I've had to break my self of the habit of strictly classifying stuff and instead just letting key words flow. And amazingly it works, my del.icio.us page isn't yet as large as my bookmark file but still I never can not find something I know I put in del.icio.us. The same can not be said about my bookmark file.
posted by Mitheral at 1:44 PM on November 16, 2005
Heck if I know, and I don't know what Drupal is. I know this though, stuff like the original Yahoo or Dmoz doesn't work for me nearly as well as del.icio.us even after the hassle of setting up a directory for my own personal use. The structure means I have to make hundreds of decisions about where every page should go, create cross links so I can find stuff, and still constantly not be able to find a page I read last year because I'm in a different frame of mind this week.
I've had to break my self of the habit of strictly classifying stuff and instead just letting key words flow. And amazingly it works, my del.icio.us page isn't yet as large as my bookmark file but still I never can not find something I know I put in del.icio.us. The same can not be said about my bookmark file.
posted by Mitheral at 1:44 PM on November 16, 2005
KirkJobSluder -
Taxonomies start from communities of people putting objects into categories, and only solidify through the process of consensus.
and
Taxonomies start from communities of people putting objects into categories, and only solidify through the process of consensus.
then
More baggage. Let's drop it. When you say, "X is an example of Y" you create a taxonomy.
and you're not grasping the irony of this are you?
but then you didn't really grasp the nature of distinctions, which is key to categorization, right? Now I never said tagging was not a taxonomy, I was playing off Lodur's criticism (implicit that tagging allowed for "incorrect" though) and was quoting him.
But to follow up, why is the fact that tagging represents group activity, interest, language use over time "bagging" and not fundamentally interesting information that is not created as rapidly by professional taxonomies. You are right, professional taxonomies are also flexible and evolve over time, but they opt for standardization. The madness and speed of a folksonomy is not baggage, but a recognizable quality of the system, and I think it is useful and fascinating data, not "baggage". Heck if you cast aside qualities that do not fit your presupposed classification scheme, you are doing, I think, a woeful and biased job of classification.
posted by kingfisher, his musclebound cat at 2:41 PM on November 16, 2005
Taxonomies start from communities of people putting objects into categories, and only solidify through the process of consensus.
and
Taxonomies start from communities of people putting objects into categories, and only solidify through the process of consensus.
then
More baggage. Let's drop it. When you say, "X is an example of Y" you create a taxonomy.
and you're not grasping the irony of this are you?
but then you didn't really grasp the nature of distinctions, which is key to categorization, right? Now I never said tagging was not a taxonomy, I was playing off Lodur's criticism (implicit that tagging allowed for "incorrect" though) and was quoting him.
But to follow up, why is the fact that tagging represents group activity, interest, language use over time "bagging" and not fundamentally interesting information that is not created as rapidly by professional taxonomies. You are right, professional taxonomies are also flexible and evolve over time, but they opt for standardization. The madness and speed of a folksonomy is not baggage, but a recognizable quality of the system, and I think it is useful and fascinating data, not "baggage". Heck if you cast aside qualities that do not fit your presupposed classification scheme, you are doing, I think, a woeful and biased job of classification.
posted by kingfisher, his musclebound cat at 2:41 PM on November 16, 2005
I haven't signed up, but RawSugar's links for "iPod" looks like this: http://www.rawsugar.com/collections/dotcoma?__ss=T&__cg=748&qrtp=54427
whereas del.icio.us links look like this:
http://del.icio.us/tag/ipod/ or http://del.icio.us/popular/ipod/.
Game over.
posted by spock at 3:04 PM on November 16, 2005
whereas del.icio.us links look like this:
http://del.icio.us/tag/ipod/ or http://del.icio.us/popular/ipod/.
Game over.
posted by spock at 3:04 PM on November 16, 2005
kingfisher: but then you didn't really grasp the nature of distinctions, which is key to categorization, right? Now I never said tagging was not a taxonomy, I was playing off Lodur's criticism (implicit that tagging allowed for "incorrect" though) and was quoting him.
Not in so many words but you did say, "tagging is . . . different from taxonomy (note I removed "fundamentally" because I am not sure what you mean by that). Taxonomy tells you what people "ought" to think about something; whereas tagging gives you an idea of what they do think about something."
So at this point, I'm a bit baffled. Are you saying that tagging is different from taxonomy? If so, what makes it diffferent from taxonomy? Or are you saying that tagging is a taxonomy, in which case I'm not certain as to why you have a conflict with what I'm saying?
And I think you misunderstood what I mean by "baggage." You appear to be making the claim:
Taxonomies are different from tags in that they are:
a) professional
b) "presupposed"
c) "opt for standardization"
d) what people "ought" to think vs. "do" do think
My point is that none of these things are necessary, required, or even associated with taxonomies. They are added "baggage" that you've chosen to add to the definition of a taxonomy for the sake of argument. They ignore the rather messy process of how taxonomies come to be adopted within communities. Further, all of the above can also apply to some types of tagging in practice. As an example, in the frontmatter of most books, you will find a list of Library of Congress subject tags that meet all of the above criteria.
Personally, I think you shoot yourself by creating such a crude straw man of taxonomies to disassociate "tagging" and "folksonomies" from. What is described here, looks, smells, and talks like a taxonomy.
posted by KirkJobSluder at 3:30 PM on November 16, 2005
Not in so many words but you did say, "tagging is . . . different from taxonomy (note I removed "fundamentally" because I am not sure what you mean by that). Taxonomy tells you what people "ought" to think about something; whereas tagging gives you an idea of what they do think about something."
So at this point, I'm a bit baffled. Are you saying that tagging is different from taxonomy? If so, what makes it diffferent from taxonomy? Or are you saying that tagging is a taxonomy, in which case I'm not certain as to why you have a conflict with what I'm saying?
And I think you misunderstood what I mean by "baggage." You appear to be making the claim:
Taxonomies are different from tags in that they are:
a) professional
b) "presupposed"
c) "opt for standardization"
d) what people "ought" to think vs. "do" do think
My point is that none of these things are necessary, required, or even associated with taxonomies. They are added "baggage" that you've chosen to add to the definition of a taxonomy for the sake of argument. They ignore the rather messy process of how taxonomies come to be adopted within communities. Further, all of the above can also apply to some types of tagging in practice. As an example, in the frontmatter of most books, you will find a list of Library of Congress subject tags that meet all of the above criteria.
Personally, I think you shoot yourself by creating such a crude straw man of taxonomies to disassociate "tagging" and "folksonomies" from. What is described here, looks, smells, and talks like a taxonomy.
posted by KirkJobSluder at 3:30 PM on November 16, 2005
KJS, in the first cited passage, I was referring back to Lodur and using "taxonomy" as it had been meant above, as shorthand for "professional taxonomy" -- I agree that tagging is a form of taxonomy. I do not agree that a distinction cannot and should not be made, which I took your position to be.
I do not think point 1-4 are neccessary for taxonomies, but that 2,3, and 4 describe a professional taxonomy. Now I am happy to play around with those terms and agree that they are problematic. For example, you could argue that a lot of our classification systems for the natural world were started by gentlemen scientists who are not professionals. Points 3 and 4 also would tremble a bit under careful analysis.
I do think that they are good starting points for hammering out a distinction between tagging -- an ad hoc and cummulative taxonomy created by amateurs who often are not concerned with the system as a whole, but just immediate and personal descriptions -- and professional taxonomies.
Sorry if my use of "taxonomy" was misleading. It certainly would be without the following context . . .
If you design a taxonomy, you have the opportunity to describe how people ought to understand something. Sometimes you want to do that. For example, a user-specified tagging system could easily end up classifying 802.11g under "health risks".
(above)
posted by kingfisher, his musclebound cat at 3:51 PM on November 16, 2005
I do not think point 1-4 are neccessary for taxonomies, but that 2,3, and 4 describe a professional taxonomy. Now I am happy to play around with those terms and agree that they are problematic. For example, you could argue that a lot of our classification systems for the natural world were started by gentlemen scientists who are not professionals. Points 3 and 4 also would tremble a bit under careful analysis.
I do think that they are good starting points for hammering out a distinction between tagging -- an ad hoc and cummulative taxonomy created by amateurs who often are not concerned with the system as a whole, but just immediate and personal descriptions -- and professional taxonomies.
Sorry if my use of "taxonomy" was misleading. It certainly would be without the following context . . .
If you design a taxonomy, you have the opportunity to describe how people ought to understand something. Sometimes you want to do that. For example, a user-specified tagging system could easily end up classifying 802.11g under "health risks".
(above)
posted by kingfisher, his musclebound cat at 3:51 PM on November 16, 2005
Tagging's been with us since the beginning, but the only early player to take advantage was Amazon.
Oh, right, that's why they're just implementing it now. Uh?
Tags -- keywords -- existed in e-mail and newsgroup headers long before the web, and the first large site to use them that I know of was the IMDb. One of the first attempts to use keywords semantically was the doomed proposal for Usenet 2.0, which would have abolished newsgroups and formed "conversations" based solely on content.
Still, tags are just one means of handling data, and merely using it to create a new view isn't all that revolutionary. I like "tag radio" at Last.FM, though.
If social bookmarking is the shiznit, though, I wonder why nobody's snatching up this del.icio.us clone.
posted by dhartung at 4:25 PM on November 16, 2005
Oh, right, that's why they're just implementing it now. Uh?
Tags -- keywords -- existed in e-mail and newsgroup headers long before the web, and the first large site to use them that I know of was the IMDb. One of the first attempts to use keywords semantically was the doomed proposal for Usenet 2.0, which would have abolished newsgroups and formed "conversations" based solely on content.
Still, tags are just one means of handling data, and merely using it to create a new view isn't all that revolutionary. I like "tag radio" at Last.FM, though.
If social bookmarking is the shiznit, though, I wonder why nobody's snatching up this del.icio.us clone.
posted by dhartung at 4:25 PM on November 16, 2005
del.icio.us is an evolution of memepool -- in fact, they were created by the same guy.
posted by seanmpuckett at 4:33 PM on November 16, 2005
posted by seanmpuckett at 4:33 PM on November 16, 2005
When will the tagging craze die down?
Everything old is new again. Anyone heard of keywords?
RawSugar is an ugly site too. In fact all of them blow goats except for del.icio.us.
posted by sjvilla79 at 4:48 PM on November 16, 2005
Everything old is new again. Anyone heard of keywords?
RawSugar is an ugly site too. In fact all of them blow goats except for del.icio.us.
posted by sjvilla79 at 4:48 PM on November 16, 2005
I think tagging is a very powerful concept and we have just seen the very beginning of it. It will not die down, I am very confident of that.
It just makes it far faster, simpler, and easier to quickly access your sites. The whole social bookmarking site is another very nice bonus and as we have more user generated content, tagging will continue to rapidly expand.
However, why do you say all of the social bookmarking sites blow goats except for del.icio.us? Have you checked out BlinkList?
posted by mreining at 5:53 PM on November 16, 2005
It just makes it far faster, simpler, and easier to quickly access your sites. The whole social bookmarking site is another very nice bonus and as we have more user generated content, tagging will continue to rapidly expand.
However, why do you say all of the social bookmarking sites blow goats except for del.icio.us? Have you checked out BlinkList?
posted by mreining at 5:53 PM on November 16, 2005
However, why do you say all of the social bookmarking sites blow goats except for del.icio.us? Have you checked out BlinkList?
The simplicity and originality of del.icio.us is wonderful. Thus, clone sites just humour me.
posted by sjvilla79 at 6:04 PM on November 16, 2005
The simplicity and originality of del.icio.us is wonderful. Thus, clone sites just humour me.
posted by sjvilla79 at 6:04 PM on November 16, 2005
Clay Shirky wrote a pretty good piece on how tags differ from taxonomies, and why tags are often better.
posted by scottreynen at 6:06 PM on November 16, 2005
posted by scottreynen at 6:06 PM on November 16, 2005
Clay Shirky wrote a pretty good piece on how tags differ from taxonomies, and why tags are often better.
That article by Clay Shirky is very good.
I don't know, I got caught by this mistake:
All the way over in the right-hand column, the pink column, are noble gases. Now noble gas is an odd category, because helium is no more a gas than mercury is a liquid. Helium is not fundamentally a gas, it's just a gas at most temperatures, but the people studying it at the time didn't know that, because they weren't able to make it cold enough to see that helium, like everything else, has different states of matter. Lacking the right measurements, they assumed that gaseousness was an essential aspect -- literally, part of the essence -- of those elements.
ven in a nearly perfect categorization scheme, there are these kinds of context errors, where people are placing something that is merely true at room temperature, and is absolutely unrelated to essence, right in the center of the categorization.
Which seems odd to me because I don't recall anybody making the claim that any of the noble gases are "fundamentally" a gas, in the same way that calcium and lithium tend to violate many of our common beliefs about metals. The essence of the "noble gas" classification has not a damn thing to do with gassness. (And in fact, Helium was first discovered in plasma form, not gas form.) Shirky seems to be making an idiosyncratic demands of an ontology: that a descriptor should be more than a generalization, but be true for every possible throughout the entire range of conditions in the universe.
Let's say I need every Web page with the word "obstreperous" and "Minnesota" in it. You can't ask a cataloguer in advance to say "Well, that's going to be a useful category, we should encode that in advance." Instead, what the cataloguer is going to say is, "Obstreperous plus Minnesota! Forget it, we're not going to optimize for one-offs like that." Google, on the other hand, says, "Who cares? We're not going to tell the user what to do, because the link structure is more complex than we can read, except in response to a user query."
At least to me, this is just the creation of an ad hoc ontology. Personally, I'd want to go one step further and ask "what are the cognitive concepts behind the keyword search?" or "why is a person thinking of seaching on the word 'obstreperous.'"
Tags are just another way of creating ontologies. But, of course, the best way to make a name for yourself is to loudly and boldly claim that everything that was done before is wrong, and what is happening now is different and new.
posted by KirkJobSluder at 4:03 AM on November 17, 2005
That article by Clay Shirky is very good.
I don't know, I got caught by this mistake:
All the way over in the right-hand column, the pink column, are noble gases. Now noble gas is an odd category, because helium is no more a gas than mercury is a liquid. Helium is not fundamentally a gas, it's just a gas at most temperatures, but the people studying it at the time didn't know that, because they weren't able to make it cold enough to see that helium, like everything else, has different states of matter. Lacking the right measurements, they assumed that gaseousness was an essential aspect -- literally, part of the essence -- of those elements.
ven in a nearly perfect categorization scheme, there are these kinds of context errors, where people are placing something that is merely true at room temperature, and is absolutely unrelated to essence, right in the center of the categorization.
Which seems odd to me because I don't recall anybody making the claim that any of the noble gases are "fundamentally" a gas, in the same way that calcium and lithium tend to violate many of our common beliefs about metals. The essence of the "noble gas" classification has not a damn thing to do with gassness. (And in fact, Helium was first discovered in plasma form, not gas form.) Shirky seems to be making an idiosyncratic demands of an ontology: that a descriptor should be more than a generalization, but be true for every possible throughout the entire range of conditions in the universe.
Let's say I need every Web page with the word "obstreperous" and "Minnesota" in it. You can't ask a cataloguer in advance to say "Well, that's going to be a useful category, we should encode that in advance." Instead, what the cataloguer is going to say is, "Obstreperous plus Minnesota! Forget it, we're not going to optimize for one-offs like that." Google, on the other hand, says, "Who cares? We're not going to tell the user what to do, because the link structure is more complex than we can read, except in response to a user query."
At least to me, this is just the creation of an ad hoc ontology. Personally, I'd want to go one step further and ask "what are the cognitive concepts behind the keyword search?" or "why is a person thinking of seaching on the word 'obstreperous.'"
Tags are just another way of creating ontologies. But, of course, the best way to make a name for yourself is to loudly and boldly claim that everything that was done before is wrong, and what is happening now is different and new.
posted by KirkJobSluder at 4:03 AM on November 17, 2005
Disclaimer: I work at RawSugar but was a MetaFilter member for several years first; in fact the reason for my absence the last half year is being busy at work and no longer have the hour or two a day I used to spend here.
I don't think it's my place to defend our service or criticize Joshua or his efforts (or Blinklist, etc...) ;) In fact I was fairly happy using Delicious before joining RawSugar and only wish that crew success.
I would, though, like to use my membership to clarify a couple of points here. First, sjvilla79, LOL thanks for the laugh. Second, Spock, we recognize our URLs are too complex and a new scheme without all the parameters is currently in the QA phase of our next release which, should we meet schedule, will be public in about 10 days. Third, we agree that our current page design can be improved, which is why we posted to Craig's List for a GUI designer recently.
With that said, I believe tagging adds value in many instances but is still very early days and the question of how and where it best fits is still open. One point rarely mentioned in these discussions, but significant to me, is that (assuming proper safeguards are in place) some person has determined a particularly page (or photo, etc.) is useful/interesting enough to be worth taking the time to tag. IOW, tagging is much less susceptible to gaming than machine algorithms. This is independent of the tags chosen for the tagged object. Again, systems must be on guard to protect themselves but since the account (person) doing any particular tagging is known if that account is judged to be a 'bad actor' then any or all of his 'contribution' can be rolled back.
Some comments have questioned the value of uncontrolled vocabularies and access. This is actually a strength of tagging because it surfaces the different ways people think of the same thing. A simple example Clay Shirky gave is the way different people (or the same people at different instances) will use the words movie, film and cinema. Physically all three refer to the same thing but (I hope) you'll all agree they have subtly different meanings. By allowing anyone to tag, say, a movie's official website as they wish the number of distinct taggings will wash away the likelihood that the first few people might use only the one they prefer (film) and so searchers (using movie) won't find it.
Anyway, that's long enough.
posted by billsaysthis at 10:50 AM on November 17, 2005
I don't think it's my place to defend our service or criticize Joshua or his efforts (or Blinklist, etc...) ;) In fact I was fairly happy using Delicious before joining RawSugar and only wish that crew success.
I would, though, like to use my membership to clarify a couple of points here. First, sjvilla79, LOL thanks for the laugh. Second, Spock, we recognize our URLs are too complex and a new scheme without all the parameters is currently in the QA phase of our next release which, should we meet schedule, will be public in about 10 days. Third, we agree that our current page design can be improved, which is why we posted to Craig's List for a GUI designer recently.
With that said, I believe tagging adds value in many instances but is still very early days and the question of how and where it best fits is still open. One point rarely mentioned in these discussions, but significant to me, is that (assuming proper safeguards are in place) some person has determined a particularly page (or photo, etc.) is useful/interesting enough to be worth taking the time to tag. IOW, tagging is much less susceptible to gaming than machine algorithms. This is independent of the tags chosen for the tagged object. Again, systems must be on guard to protect themselves but since the account (person) doing any particular tagging is known if that account is judged to be a 'bad actor' then any or all of his 'contribution' can be rolled back.
Some comments have questioned the value of uncontrolled vocabularies and access. This is actually a strength of tagging because it surfaces the different ways people think of the same thing. A simple example Clay Shirky gave is the way different people (or the same people at different instances) will use the words movie, film and cinema. Physically all three refer to the same thing but (I hope) you'll all agree they have subtly different meanings. By allowing anyone to tag, say, a movie's official website as they wish the number of distinct taggings will wash away the likelihood that the first few people might use only the one they prefer (film) and so searchers (using movie) won't find it.
Anyway, that's long enough.
posted by billsaysthis at 10:50 AM on November 17, 2005
First, sjvilla79, LOL thanks for the laugh.
It wasn't a joke. I guess ignorance is bliss though.
posted by sjvilla79 at 10:34 PM on November 17, 2005
It wasn't a joke. I guess ignorance is bliss though.
posted by sjvilla79 at 10:34 PM on November 17, 2005
« Older The Heart of Silicon Valley beats for WiFi | The history of puppetry. Newer »
This thread has been archived and is closed to new comments
What? That's a dumb question. When will did the Industrial Revolution craze die down? Uh, it didn't, it just evolved into the Information Age.
Yeah, that's right, I compared tagging to the Industrial Revolution.
posted by Plutor at 11:30 AM on November 16, 2005