Slashdot Mirror
Folksonomies In Del.icio.us and Flickr
Ian@falsepositives.com writes "Lots of discussion going on about 'folksonomies' -- bottom-up taxonomies that people create on their own -- as used in Del.icio.us and Flickr: Adam Mathes has a thesis on Folksonomies; IFTF's Future Now makes a point about problems with folksonomies: no synonym control ( "mac" and "macintosh" on Del.icio.us); no hierarchy and content types; and only simple one-word tags. Joho the Blog notices a discussion about what to call it in Mob indexing? Folk categorization? Social tagging?, and John Battelle links into Taggle and "federated tagging". I wonder if a Google Suggest like system might reduce 'lazy tagging' ,and maybe synonym control when the federation appears. Tag, you're it!"
I hate this term, there is no lazy tagging, only different tagging. Tagging using too precise a description, thus too many words is as useless as tagging with too few.
That was the single most incoherent paragraph I have read in awhile. I'm afraid to RTFA because it'll probably result in me contracting brain cancer somehow.
Wha.thef.uk
I must need more sleep.. that looked like complete gibberish to me.
"Consider how lucky you are that life has been good to you so far. Alternatively, if life hasn't been good to you so far
Meanwhile, this is pretty much what happens in any ad-hoc metadata system, and not all of us have the luxury of paying someone to manage our indexes. The place I used to work is just the same. At least it's better than nothing.
A study of tagging on del.icio.us .. "A mini-ethnography of social practices in a distributed classification community"
this list included much of what one might expect as common subjects of photos: cat, friends, dog sky, sea, park, kids, garden, baby, building, flower, flowers signs, sculpture, city, vacation.
From the folksonomy thing. What's a "dog sky"?
I don't need a signature.
It seems that in a system like del.icio.us which is very open social practice will reinforce some standard way of dealing with things like synonyms. maybe it can be expected that people will eventually use the most common words in the system for tagging if they want to make full use of the social benefits that the system offers.
if on the other hand people use del.icio.us and the like only for their personal benefit or for a small group then there is nothing wrong with using different words than some other subgroup.
I read the article, I've read the paragraph.
Maybe I'm just having a bad hair day, but it *sounds* so stupid that no one can even explain it coherently.
You obviously have a firm grasp of deciphering Slashdot articles.
I didn't know either so I looked it up
...more info at http://www.adammathes.com/academic/computer-mediat ed-communication/folksonomies.html
Del.icio.us http://de.licio.us/ henceforth referred to as "Delicious") is a tool to organize web pages. A description online states it is: "a social bookmarks manager. It allows you to easily add sites you like to your personal collection of links, to categorize those sites with keywords, and to share your collection not only between your own browsers and machines, but also with others" (Schachter, 2004)
Flickr http://www.flickr.com/, a photo management and sharing web application, has a similar system of free-form tagging for photos that was adopted and modeled after Delicious. It too requires users to create a user account, and is free to join.
Bringing your mosaic ideas to life. Mosaiclegs
Instead, these systems works because there are so many participants, it doesn't matter if you miss 50%, 80 or 90% of them because of differing tag names.
This is the first slashdot blurb I've ever read that left me feeling like I had no f'n clue what they're talking about. It was like reading the mental vomit of an ADD kid.
Blessed be he who reads this post, Cursed be he who tells my boss.
no synonym control ( "mac" and "macintosh" on Del.icio.us)
Aren't words what people make them to be? I mean, if many people, from the bottom up, decide that "Mac" should be primarily a synonymous of "Macintosh" (which it is, de facto), then secondarily an acronym for an ethernet card address, then for TV addicts a short for Duncan McLeod, so what? Who's to define what words mean if it's not the people who use them?
I mean look at the French: they have something called the "French academy" that makes up a bunch of words willy-nilly every year, after much discussion, to be added to the "official" french language, but without consulting the potential users (the French). Well guess what: most of these words aren't known, let alone used, with precious few exceptions.
So I say great: if grassroot efforts end up redefining the language, and help consolidate new words into the core language, and help create new words and expressions, I say fine. That's what defines a living language that people like and use.
Please try to keep up.
This all looks like nothing more than a filing system for the anally retentive.
I'm sorry, but this is not a very good idea. After it has got popular enough to attract attention, it will be ripe for abuse.
You can just imagine what the bots will be tagging the viagra ads and nudie pics with...
Sure, we can start with bayesian filtering and manual deletion all over again, just like with wikis and blogs. But isn't it time that we start caring about these issues before we jump on every new product?
I didn't get the article but I think that's not it.
Add an entry "mac" and entry "macintosh" and point both to Apple and you have the synonyms problem solved. Many words to describe the same thing, multiple entries describing the same page.
Worse about homonyms, where one word has several meanings. Wikipedia solves that by "disambiguation" pages.
Anagram("United States of America") == "Dine out, taste a Mac, fries"
..H. Christ on a pogo stick. My brain stem just fell off.
I'm sorry, I can't hear you over the sound of how awesome I am.
It's not that complicated a concept - systems have arised which allow you to categorise your own information (bookmarks and photos in the two examples given). Because everyone can use whatever categories they find useful for themselves this means that I can tag my Mac stuff "mac", you can use "Macintosh" and someone else can use "Apple", leading to miscommunication.
My Journal
1) Most incomprehensible /. blurb EVAR
2) Adam Mathes is one of those guys I always though really understood the internet as a distributed ad hoc metadata generation system. He's also pretty funny. He was one of the cofounders of the snarky webzine Uber.nu (which I used to write for). He combined the two and invented googlebombing, which earned him a certain degree of noteriety.
3) I think there is nothing new in these criticisms of distributed ad hoc systems. It's the same with google, and wikipedia. You sacrifice some depth and accuracy for breadth, And as your algorithms become more sophisticated they can organically uncover the inconsistencies and work around them.
Okay smartass moderator. If you're so clever, explain this. Slapping -1 overrated and acting that for you it's all perfectly clear and understandable is easy.
The ability to also merge tags in a search is particularly useful, such as in the case of a search for Python packages (http://del.icio.us/tag/python+packages) as opposed to Python movies (http://del.icio.us/tag/python+movies).
As the site gains more and more users, I have to wonder about the S/N ratio, although merging keywords in searches will help.
Hopefully some scumbag won't figure out a way to make his H3RB4L V14GR4 page come up no matter what keywords you enter.
Please, someone explain how a "folksonomy" is different from a "vocabulary"! The article makes this big deal about how a folksonomy is a weird kind of taxonomy without explicit relations between words, where you can have synonyms, where people disagree about exact meanings.
Isn't this just a "vocabulary" like we use every day when we talk??? Why is this a strange new thing???
A few months back I wrote avar.icio.us, which does autocompletion and dynamic suggestion of tags based purely on your own tag coincidence statistics. You can see it in action on Bowen Dwelle's site: try typing "ja" in the tags field - it should autocomplete "javascript" in the textbox and then suggest a couple more tags beneath. (This is all based on Bowen's own tags) Note that autocompletion only works in Mozilloids and IE/Win at the moment.
A popular add-on that makes suggestions from other user's tags is Greg Sadetsky's nutr.icio.us (unfortunately unavailable at the moment). Joshua is building some of those features into del.icio.us core.
... and is modded up using the collaborative karma system.
Anyone else giggling about this?
Or are you all waiting for a post that everyone sane can understand, like how to modify your Gentoo PPC install to use both OSS and ALSA without frying your SBLive?
*sighs wearily*
I always thought that OPML and Google-like search powers was the beast for this job. Is it being used? It would certainly gather together the disparate threads in a self boot-strapping manner.
Dave Winer (of Scipting News fame) always had a bee in his bonnet about this subject and on this he makes sense.
Patriotism is a virtue of the vicious
All that's needed is a decent design to be built into one piece blogging software. A few people are working on it for Drupal. Once one popular blogging tool has a simple and elegant solution others will adopt it, just like trackbacks.
Personally I think the central server(s) should use something like WordNet to determine common synonyms based on context and build from there. I think the fact that the keywords come from so many people is a good thing. Instead of a few people thinking hard about how to organize, general concensus will help it work itself out, especially as it become much more popular.
Developers: We can use your help.
/* Note: this is going to be off topic, so I don't mind if it gets modded that way */ I read the damn thing at least 3 times... not that I didn't understand for first (I know about it all over, and the linked stuff) but for the plain reason that I just couldn't believe my eyes someone could put together a paragraph which sounds so totally out of language non-human gibberish all over. My head just hurts. Indeed.
I am putting myself to the fullest possible use, which is all I can think that any conscious entity can ever hope to do.
Enable nonstochastic communication modalities by utilizing post-Bayesian non-deterministic linguistic differentiation mark-up in a non-denomenational plurisitic non-denominational rainbow-like blur of buzzword hyperbole.
The problem is that many modern categorisation systems assume that people know how they want to categorise their own data. They therefore aalow individuals to use whatever word/phrase they want to tag their data with. These tags can conflict both with other users (for instance one user could use "Mac" to refer to items related to Apple Macintoshes while another uses "Macintosh" and a third uses "Apple") and with themselves (when a user's nomenclature changes or they mistype).
There are a few obvious solutions to this:
1) Reuse: Help the user to re-use their old tags by offering them a list of previously used tags - this will prevent typos and unintentional changes.
2) Synonyms: Help users to lump tags together by stating that "Mac" and "Macintosh" mean the same, as far as they are concerned. When they look for tags in the same category as "Mac" the search will automatically be broadened to include similar ones.
3) Build categories from the most commonly used tags. This returns to the top-down imposition of structure, but builds it from the tags that people actually use. If a tag is used by more than x% of the population then categorise it and assign it a detailed description. For instance, if more than 1% of people are using "Mac" as a tag, then "Apple Macintosh Computer" could be assigned as a detailed description. Users could then choose to use the 'official' tag. Synonyms would also exist, so that "Macintosh" and "Apple" would both link to this single 'anchor'.
The use of more-defined descriptions would allow multiple meanings for the same tag to exist, so that someone using "Apple" as a tag could be offered the choice of attaching that tag to the definition "Apple Macintosh Computer", "Apple Fruit" or "Apple Music Corporation". The user could obviously also attach it to any other definition or leave it definitionless.
I am, of course, assuming that most people would find utility in using common definitions, as it would allow them to find things that used the same tags, whilst leaving them the freedom to use any tag they like for their own use.
My Journal
MusicBrainz/Audioscrobbler have a system that lets users vote on synonyms for metadata. For instance, "The Beatles" and "Beatles, The" both point to the same group tag.
Del.icio.us is a bookmarking system. Joshua Schachter programmed it to have a bookmarking system and as far as I know, he did it for himself, not for the public at first.
So, _you_ add a bookmark, _you_ tag it, so _you_ can organize your links in the way you like it. There are many ways to categorize bookmarks and the tagging system allows you to use multiple ways in one.
I recreated del.icio.us for porn (porn-a-licious.com) and something interesting happend: In the beginning, people tended to tag their posts in the usual way (hardcore, softcore, etc.). Then came people tagging their bookmarks using their favorite porn star names as tags (luba, marketa, etc.). And than came a guy starting to tag them using tags like f, ff, fm, ffm, etc. And now, most people tend to combine all or some of these types of tags.
there is no horizontal, vertical or other buzzword-way to tag. You just start to organize your bookmarks in the way you like it. And most people may adopt the most useful tag-styles creating a huge, very well organized link list.
You don't need a synonym control if you have enough users because if the link is important there will be someone who will post that link with tags assigned to them you would use, too. Porn.a.licious is bookmarked often on del.icio.us, and since some users still try to hide their porn-bookmarks, not all tags used were really useful (sometimes, porn.a.licious was tagged with 'cars' or something like that).
Lots if discussion going on about fragglemat. Toxic taxidermists tipptoe on people creating their own. As seen in Flippsonomatic De.li.ri.um.
Flicker, flicker, *wink* *wink*. ITVTVTT-TV WTF?
Future Now makes a point in being later than yesterday. No synonyms controll mac for macintoshes. Herarchy one-word-tagged content-types.
Jojo-Joohohoho - The Blog! Notesdiscussion What-about-what?
Mobsinjection? Folksoflippsonomy-Calegari?
Taggletaggle (the federated social one)?
Wonder, wonder, google, google.
Makes me lazy, makes me hazy.
Tag! You are it.
--
I allways had the impression that slashdoters and the slashdot editors were stoned beatiks, but this guy obviously double dosed his morning share today.
We suffer more in our imagination than in reality. - Seneca
eschew oxymoronic obfuscation
*g*
______________________________________________
sigamajig...
What's a taxonomy without hierarchy? This is just simple classification for indexing and it's a shame to misuse the terminology and make all the ignorant responses above right about the whole article being a blur of buzzwords (... and, God damn it, not programming jargon!)
'nuff said.
God I hate the quantization of internet culture. There's nothing more lame than some academic know it all spewing junk about how web facilities are creating fun new words.
Unfortunately, Amazon's shopping cart is painful to browse when it reaches that size. Also, Amazon distinguishes between the current cart and items "saved for later", and moving between them is also awkward. There's also no way to move an item from my UK shopping cart to the Amazon.com shopping cart, for example.
Recently, I started tagging items with del.icio.us instead of adding them to the shopping cart. Voilà -- a portable shopping cart.
Since I can get the RSS for a collection of tags, I can now easily have a bookmarklet or script that, given a tag, adds each item (based on its ISBN) to a shopping cart -- any shopping cart, not necessarily Amazon's.
Take it one step further: Have a script that looks for the cheapest stores based on the items in the RSS. This assumes the URL contains the ISBN and the script knows how to find it, but that's easy.
Since my portable shopping cart is available for all to see, it doubles as a gift wishlist. I've started a read list, too.
Why don't we just call it what it is, meta data?
How hard was that !
Arguing about the name of the thing, 'Tags', 'Folksonomies', etc. is all a load of BS as far as I am concerned. The real issue is that we have a means of attaching meta data to other datum in a way that is easy to use and easy for computer systems to digest and parse.
There is already a standard that allows this - and even allows you to extend it as needed: XML. What is lagging behind are the tools to make that an easy process for the end user, as well as a flexible data storage medium that does not have the current limitations RDBMSs [relational database management systems] incur - (at this juncture it is probably useful to bring up that there are object oriented database systems that can be useful in this vein - Zope is one example, and there are others - far more flexible and extensible than traditional RDBs which require the attentions of a database administrator for even minor changes).
Something else to consider: with the exception of a small subset of the total information store at any given moment, it is not possible to effectively systematize the classification of all information in a way that is useful to everyone. What needs to take place is for the end users of the information to also overlay their own classification system upon disparate datasets - tieing them together in a meaningful way for them. Just as every individual thinks differently and approaches various activities in different ways, so too should their information systems.
Lodragan Draoidh
The more you explain it, the more I don't understand it. - Mark Twain
Or are you all waiting for a post that everyone sane can understand, like how to modify your Gentoo PPC install to use both OSS and ALSA without frying your SBLive?
*sighs wearily*
I would take practical advice any day over these meta-abstract pseudo-intellectual discussions that self appointed experts like to get into. It seems every week, there is some new Paradigm That Will Change The Way We Process Information. This one looks just as stupid as all the rest.
It's in danger of becoming a bit like the Emperor's new clothers. Tagging has been around for a long time, it's just that we all got bored of doing it - meta tags that is.
For evey page on your website you'd create a bunch of meta tags and then cross yourself three times in the hope that a decent search engine would make sense of it all.
Of course, then Google came along and made the content of the page much more important than the author's chosen keywords, which is right, in my opinion.
No, I understand the difference between an author's chosen tags and these folksonomies and I'm for the folksonomy but it's hard work - I don't see a future in user-defined tags. I do, however, see a future in scripting all this - be it by a personal proxy server or a branch of Google.
A personal proxy server written in Python
The problem with tagging
Playaholics : Great online games
Suttree, a weblog about casual games development
these meta-abstract pseudo-intellectual discussions
... and I use this kind of tagging every day. People are building it into new applications as we speak. It's not abstract, it's working and useful right now.
"big words I don't understand and can't be bothered to click on"
self appointed experts
"people actually learning about things and explaining them"
It seems every week, there is some new Paradigm That Will Change The Way We Process Information.
This one's been around for months. Tens of thousands of people are using it already. That's worth commenting on, isn't it?
"I would take practical advice any day"
"big words I don't understand and can't be bothered to click on"
No, "made up words". Or does your Websters or OED contain the word "Folksonomies"? I can make up important sounding words too; that doesn't make me an "expert" nor does it increase my IQ.
This one's been around for months. Tens of thousands of people are using it already. That's worth commenting on, isn't it?
No. Just as AYB, Dancing Hamsters and Napster were passing fads, so is this. It will quickly fad, shortly followed by "everyones" favourite "the blogoshpere". It will slip from grace to be replaced by next weeks New! Shiny! Paradigm! of crap, and we'll get a whole new bunch of self-agrandising "experts" who'll invent a whole new bunch of psuedo-intellectual bullshit and more made up words under the guess of "jargon".
I find that it's best not to use abbreviations when labelling things for later search retrieval, because abbreviations can have multiple meanings and words can have multiple abbeviations: Does "script" mean "shell script" or "prescription?" If there is a commonly abbreviated phrase, I will include the initialism as well as the spelled out phrase: "Local Area Network (LAN)" and "Random Access Memory (RAM)" I also include common synonyms when labelling things like files or e-mail subjects even if it seems redundant: "Measurement Conversion Factors Equivalents List"
This is an interesting problem. FotoFlix uses a different model than "keywords" ... they use labels. So "Mac" and "Macintosh" are completely different. They also use icons associated with each label. So with that system you can aggregate the properties placed on a given photo based off the name or the icon.
They're working on a way to synchronize labels between users in groups as well. That way you share not only photos but organization as well. Definitely check it out...a very cool approach.
FotoFlix
Better than Flickr - Manage, Share, Archive
Its great that people are at last forcing/levering/coaxing the issue of the organization of information from the hands of geeks who pride them selves on never forgetting anything (but couldn't tell you what any of it means if you put a gun to their temple.)
Remember these are the same folks who insisted on 8.3 file names and who only recently discovered desktop search.
These are the same people who insist on interning references instead of describing relationships between data objects as exactly that,relationships between data objects. (Too complicated. It'll lead to slow retrieval. [It doesn't.] That will never work! [It does.] It'll break referential integrity! [Like you turn it ON for your database!] Retrieval will get too complicated. [Not if you use your schema properly and create wiews.]
Once somebody collects there and starts a meta meta project with the data, I think we'll actually get somewhere.
MSBPodcast.com The opinions expressed here are my own. If you don't like 'em... Think up your own stuff.
... systems have arisen which allow you to categorize your own information ...
I'm not usually a grammar nazi, but really... how can we trust the categorization skills of the public, when the vast majority of the users don't even bother to check if their statements make sense! Are you going to be "categorising your gamez"? What, if any, value is there to your opinions on a subject (especially keywords for an object), when I have to fight to understand you because of your poor grammar and spelling skills?
I don't get it... You add bookmarks, categorize them with your own categories and then optionally share them with others, doesn't that just make this pretty much the same as Yahoo's system but just implemented differently?
How many people still use the Yahoo method to find stuff anyway? I find it much easier to just do a search usually Google for stuff I want and only bookmark specific things that I will need to refer to later.
I think their solution will be to end up using defined categories and hence become even more Yahoo like.
WTF if a Folksonomy anyway? It wasn't even listed in dictionary.com, oh a Google reveals it is just a made up word to collabratively categorize stuff, sounds like one of those big words people use to show they are smarter than you...
In the drops - An Aussie's musings on all things cycling
Me: these meta-abstract pseudo-intellectual discussions
... and I use this kind of tagging every day. People are building it into new applications as we speak. It's not abstract, it's working and useful right now.
You:"big words I don't understand and can't be bothered to click on"
Response: I didn't say that I couldn't understand them (although that is also an issue), I said that there were:
(1) "meta-abstract", meaning that they are discussions about discussions and seperated from actual implementation, and
(2) "pseudo-intellectual", meaning that are carried out with a intellectual attitude (big words, big principles) but they are lacking the actual academic rigour that would make them truly intellectual.
In response, you basically called me stupid and lazy.
Me: self appointed experts
You:"people actually learning about things and explaining them"
Response: I have no problem with people learning and explaining things. Still, I am correct to refer to those who are inventing this folksonomie thing as "self appointed experts".
Me: It seems every week, there is some new Paradigm That Will Change The Way We Process Information.
You: This one's been around for months. Tens of thousands of people are using it already. That's worth commenting on, isn't it?
Months? Wow. That's almost as long as the dot-com boom lasted. Sarcasm aside, I agree that it is worth commenting on. My objection is to the scientific veneer placed over a fairly interesting, but simply problem: how to make classification of information simple and straightforward.
"I would take practical advice any day"
The implementation is not abstract - I use audioscrabbler, so I know it is useful. Again, I am objecting to the meta-levels that bloggers apply to a simple discussion. Instead of all of this bullshit about taxonomies and synonym control, how about we see some code, or even pseudo-code?
I really don't.
"It's not that complicated a concept - systems have arised which allow you to categorise your own information "
Oh lovely. We're going to RDF the web.
The other key part of social categorization is that there is a *feedback loop* based on tag popularity that reinforces common tags - the more people who use a tag, the more prominence it gets in the system, encouraging people to use the common term.
Flickr and 43things use bigger type to show tag popularity.
Social categorization is useful because it is fuelled by self-interest - people tag info in these systems to find it later themselves - but it has a public benefit in finding related information.
It's also no silver bullet - but it's useful as part of a bigger information architecture effort (my business partner started the whole discussion about social categorization, and we're starting to use it on some projects)
The internet is the global network, most-but-not-all of which uses TCP/IP, and the services that run on top of it. It's a common noun in common usage, and never you mind what your pet style guide says.
You can perfectly well submit stories about topics that are technical (hey, this *is* news for *nerds*) but it's all in the language.
I submitted that Half-Life-2-on-Linux-with-Transgaming story a while back, and while I figured everyone probably knew about all those pieces, I still backed all the way up and explained that Cedega was a game-enhanced version of Wine, a windows porting layer, and that Transgaming was it's creator. They had just released a new version that supported a *major* game release in the first person shooter market called Half Life 2. I worked pretty hard to make sure it was a clear as possible, even if you weren't familiar with the subject matter.
The guy who submitted this piece of trash probably would've written something like:
"Codeweaver's Wine extention Cedega 4.1.1 by Transgaming is out with support for Valve's Half Life 2. Go get it!"
Even for people that know what's going on, that can be an odd (if not confusing) statement.
The moral? Feel free to post techno-topics, but work to make sure that even a uninitiated nerd/geek would be able to figure out what the heck you're talking about.
I hadn't heard about this before, so it's kind of interesting, but only after I read a few comments figuring out what it was about.
I jumped on the del.icio.us bandwagon & my biggest pet peave is that it doesn't search for other capitalization. I run a page of LaTeX tips. I must monitor 'latex,' 'LATEX,' 'Latex,' and 'LaTeX' in order to cover most of the relevant topics. Multiply this by the number of topics you're interested in & it can grow quite annoying quite fast.
It also doesn't differentiate between LaTeX the typesetting language & latex the emulsion of rubber or plastic globules in water. There is a high geek population on del.icio.us, so I manage to get very few 'hot liquid latex sexxx' sites. Such inability to disambiguate a word with homonyms is also an annoying feature. In this case, it would normally work well for me, but would work quite poorly for any latex fetishists out there.
That description made absolutely no sense whatsoever. What the fuck is delicious and that other thing, and what the fuck is this folk thing, and why should anyone care?
Do editors even read this shit anymore? I should start submitting hoaxes with random abbreviations to see if it gets posted.
It only seems to hold as long as the controller/owner of the system succeeds in keeping porn or other aggressively commercial media out of its systems.
When that happens, popular keywords will soon start referring to porn and such media and the designers will need to think of other ways to determine relevancy of terms/keywords/tags to an object.
The article is interesting and relevant to any "unspoiled" community tag-database. But imo, it has little value when talking about systems that have been open for some time to the commercial scum, that seems to succeed in filling every nook and cranny of the internet.
The premise of realism suggests that the significance of the observer is deconstruction. Thus, the characteristic theme of social tagging and mob indexing is the role of the participant as artist.
Anyone understand that? No? Neither did I. Didn't understand the OP either. Is /. receiving crossposts from alt.postmodern?
WHAT?
You didn't explain it. You gave me an example.
It looks like the kind of thing that D&D players do because they're nerds. There's absolutely no use for this.
Its basically a private namespace to describe stuff because (a) people think its clever (b) they're too lazy to learn the actual names of things. See? I came up with the only reasonable definition of this crap.
It like people who call tools or gadgets they don't understand a "Johnson". Oooh. clever.
(1) "meta-abstract", meaning that they are discussions about discussions and seperated from actual implementation, and
(2) "pseudo-intellectual", meaning that are carried out with a intellectual attitude (big words, big principles) but they are lacking the actual academic rigour that would make them truly intellectual.
In response, you basically called me stupid and lazy.
Okay, I apologise. I didn't fully appreciate what you were saying.
I feel picky about your point 2, though: just because a thorough empirical study hasn't been applied, it doesn't mean that the comments aren't intellectual (and more importantly, intelligent).
Response: I have no problem with people learning and explaining things. Still, I am correct to refer to those who are inventing this folksonomie thing as "self appointed experts".
I think you're being unfair by implying a hubris that I just don't see here. The commenters are expressing opinions, but not claiming to be experts. And it is perfectly possible, given the youth of the field, that people like Merholz have read most of the major essays on the topic.
My objection is to the scientific veneer placed over a fairly interesting, but simply problem: how to make classification of information simple and straightforward.
Ah, except you're missing a huge chunk of what this is about: Not just how we can classify things, but what happens if you give such an open, simple and flexible classification system to a diverse, separated bunch of users and then aggregate the results.
Instead of all of this bullshit about taxonomies and synonym control, how about we see some code, or even pseudo-code?
But there's code already. del.icio.us and Flickr have been around for a while. If you look at the Taggle suggestion, there are already several implementations in the comments. This is about post-implementation discussion and analysis.
The distinction between formal and informal taxonomies are about value-add, inclusion, utility, and control.
It isn't about whether words are "redefined". It is about the fact that, without standards for synonyms, taxonomies lose value, because they end up with "semantic forks", if you will, with redundant data in some places, missing data others, reduced search value, and general lack of confidence overall.
Now, whether the inclusiveness and volume of assistance outweighs the cost of these weaknesses, I think depends on the particular taxonomy and use.
I think it kicks ass in Delicious, but I wouldn't want volunteer contributors in the Dewey Decimal System or a legal knowledge management system.
I forget what 8 was for.
In the absence of strong identity verification for "voting" systems that presume to measure "popularity", all metadata matures to spam.
Now there's a useful term to come out of this discussion. Thanks!
talk of tagging...did the author use the word flickr at all besides in the title ?
Somewhat legitimate gripe. One thing I would like to see in del.icio.us is the ability to rename a tag. As it is, you have to create a new tag and then retag all your items with it.
no hierarchy and content types;
No hierarchy is exactly what I like about del.icio.us and GMail and similar systems. Hierarchies take work to maintain and your stuff never fits neatly into them. As for content types, make that part of your tagging system if you care. For my GMail account, a simple tag "Content" is enough for me to use on any emails containing "nontrivial" attachments.
and only simple one-word tags.
Again, exactly what I like. The emphasis is speed and ease of access. If you have to remember a long and complex tag, it's going to take too long to add a new entry.
Things are more like they are now than they ever were before.
The submitter assumed you had heard of Del.icio.us and Flickr
Based on most of the responses to the article, this was a very bad assumption.
and whated to learn more about "folksonomies" defined in the article as : bottom-up taxonomies that people create on their own.
This definition just shifts the confusion to somewhere else. How about defining "folksonomies" as "using the meta-data that users apply to their own data to organize a general system of classification"? At least that gives a hint of what is actually being done. Telling me that it is a bottom up taxonomy is next to useless.
It was also assuned you had a better than grade 10 level of reading and/or education, and could figure things out.
Once again, this is a pretty bad assumption, considering that it was submitted to slashdot, where the average intelligence is about 3rd grade. Ignoring that bad assumption, the point of an article submission should be to explain in laymans terms what the article is talking about, so that readers can decide whether or not they want to read the article.
Physicists don't submit slashdot articles that read "Contrary to GJNM-3 predictions, the free space corrector to the assumed universal fluence theory, which predicts the effect of non-dispersive media on the transport of energy, has been measured by almat 5 to be over 1.2 times greater than analytic theory." Why do made-up fields, such as folksonomie get to do it?
Things are more like they are now than they ever were before.
...suffer similar problems. It seems they are only suited for the ephemeral and personal; for any serious topic it's just a kiddie version of Usenet - although many people optimistically attempt to create "serious" categories for discussion.
you had me at #!
Over the last few weeks I've implemented tagging for Trac (the wiki side of things for now). Here's a relevant blog entry: http://muness.blogspot.com/2005/01/tags-for-trac.h tml .
In wikis, tags can be "linked to the wiki entry of the same name, allowing you to describe them explicitly under the wiki entry of the same name. This allows for a flexible means for establishing the context of wiki entries." This makes them a lot more expressive than in del.icio.us or flickr.
Also, querying tags is weak right now. Full set operations should be supported. Also, most tag systems seem to ignore hierarchies. There's no reason for that.
With all the discussion about the fallibility of user tagging, with regards to spam/porn - I can't help but suggest that some of you take a look at http://www.stumbleupon.com/ It gets comparisons to del.ici.ous, and has similarities, but in SU the general tendency seems to be based on auto-classification, with some human oversight where users can update blatantly miscategorised sites. The only problem I've seen so far is a lack of breadth of categories.
I think a lot of people are missing the real reason why this form of metadata, namely the non-hierarchical, user-created tag system that del.icio.us and flikr use, is interesting. Namely, it's that people are actually using it on a daily-basis and, what's more, making it do useful things.
I think that this isn't so much a proof that it's a great system, but rather an indictment of existing metadata systems. It demonstrates that the key to useful metadata is to make its use easy and natural. Thus, creating a successful metadata system is as much a question of HCI as it is a question of abstruse ontologies.
You've obviously never used del.icio.us. Not to mention that *every* word is a made up word.
Theres pluses and minuses with each approach. Have a read of the folksonomies article which once you get to the core was not to dificult to understand.
There are four sorts of people in the world: fools, lunatics, idiots and morons. - Umberto Eco, Foucaut's pendulum.
When you create an index or taxonomy after the content has been produced, that's called Post-coordinate indexing (as opposed to pre-coordinated indexing, eg faceted taxonomies).
No new term needed. Thanks again to library science.
I posted on this very same matter back in November, though with an eye more towards sorting out such taxonomical discrepancies with code. I guess I should have hacked something up then. Still, nice to see folks thinking about the same issues.
Do I have to click all of the links just to figure out just what the heck this article was about and whether or not I _wanted_ to click the links?! That post made absolutely _no_ sense whatsoever...
*sheesh*
+++++++
"Look, dear, it's a crazy hairy scary man!"
They call me Mr. METADATA!
At last - one person who groks that this isn't a new problem (forget history...repeat it) This is old, old news to librarians, archivists, taxonomists and the like. The bloggers and taggers will probably rile at the idea of authority control (because of the name, most /.ers probably won't try to grasp the concept), but it's an inevitable problem that must be tackled.
Folksonomies (the first meme of 2005?) is attributed by Wikipedia to Thomas Vander Wa.
Adam Mathes has a thesis on Folksonomies which examines user-generated metadata as implemented and applied in two web services - Del.icio.us and Flickr - designed to share and organize digital media to better understand grassroots classification.
IFTF's Future Now makes a point about problems with folksonomies: no synonym control ( "mac" and "macintosh" on Del.icio.us); no hierarchy and content types; and only simple one-word tags. Are these features or bugs? Consensuss says 'feature'. Andrew Ducker has a suggestion for synonyms and a modest proposal
Joho the Blog notices a discussion about what to call it in Mob indexing? Folk categorization? Social tagging?,
John Battelle links into Taggle and "federated tagging".
I wonder if a Google Suggest like system might reduce 'lazy tagging' ,and maybe synonym control when
the federation appears.
New: In Beyond Laser Tag and Telephone Tag, JC Francois wonders if "2005 will be the year of tagging".
Will Folksonomies lead to the nirvana of the Semantic Web, or at least Semantic web light? (see : ftrain.com August 2009: How Google beat Amazon and Ebay to the Semantic Web)
Tag, you're still it!"
That's the problem with flickr, if you have a photo of a stop sign, which tag do you assign to it?
...and more.
sign
signs
stopsign
stopsigns
(tags like "stop sign" are converted to one word)
what about my camera phone images?
there are tags of:
nokia
nokia3650
moblog
cameraphone
I don't have all day to assign multiple tags and to research what tags others are using.
I think "lazy tagging" is stuff like mispelling a tag or starting a new tag that no one used before whena better one is available, like starting stopsigns when stopsign alreay exists.