Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

I am endlessly fascinated with content tagging systems

twitter.com

Post reply on HN
Print view View original

141–150 of 279 posts

Re: I am endlessly fascinated with content tagging systems

#141

Post by vonseel »

lofatdairy wrote: post #91
jrochkind1 wrote: post #14

> I can't find anything on how to design and implement anymore more than the barebones basics of a system. All of this stuff (horse/horses etc) is extensively discussed, maybe look under "taxonomy" or "ontology". Now, whether you want to use any of those solutions or not or find the discussion useful or not... if you aren't finding anything about it at all, you aren't looking in the right places. (I learned about it…

To be fair to OP, the biggest hurdle in learning anything is knowing what questions to ask. When you don't have ontology as part of your vocabulary it's hard to find literature regarding, say, "comparison of ontologies for user-generated text content". I suppose this flows back into library science, which is all about systematizing where to look for answers to questions, but I'm always astonished to find that there's…

I think OP is referring to finding software-engineering related design discussions surrounding tagging systems, but yes, I’m sure there is a great depth of ontology material and librarian knowledge that could add to software system designs.

Re: I am endlessly fascinated with content tagging systems

#142

Post by at_a_remove »

PuppyTailWags wrote: post #132
at_a_remove wrote: post #117

Earlier quoted context omitted.

"no regenerating limbs those arms are toast sorry QA despises them" just isn't useful if I want to locate a particular text, other than "I'm liable to get a Tumblr-stink off of this crap." And your defense of this is really ... internal , as in, this all looks like a lot of in-jokes to an outsider who is new to AO3, or even new to a particular fandom. If someone doesn't know the slang, the in-joke reference, it's sti…

> "no regenerating limbs those arms are toast sorry QA despises them" just isn't useful if I want to locate a particular text, other than "I'm liable to get a Tumblr-stink off of this crap." Yeah, but you're not looking for that tag, and that tag wouldn't affect your search in any way. That's the thing. You're approaching tags like they can only only ever be used one way, and yes they can be that, and also other thin…

You have it backward: I found the fic through other means entirely and eventually dropped it. When I encountered it again on AO3 (it was a cross-post), I said "Oh, look at those horrible tags." It was notable in the fact that I said "I need to keep this one handy the next time I end up having yet another conversation with someone about how much tagging sucks on AO3." Because this isn't the first time someone has brought it up to me.

They just crap up the results if I am searching for "regeneration" or "limbs." If something is used more than one way, yes, it does affect my personal use because it means "more stuff I have to filter through." When you search, what you do not want is extraneous results. That's the whole point of searching! And I guess my library experience is showing, but AO3 just reeks of amateur hour shenanigans. I predict that at some point there will be a movement to clean up that kind of junk.

Re: I am endlessly fascinated with content tagging systems

#143

Post by AtlasBarfed »

Eh, the diamond problem and transitive issues don't exist because what is being reduced to is simply a set and membership. if expansions / aliases / synonyms / multi-membership produce overlaps, who cares, it's a set of hashs. The overwrites only represent wasted computation.

Really this is a simpler version of multiple inheritance. You don't have the issue of conflicting method signatures and implementations, only names.

The only danger is names meaning different things. You need your tags to be relatively unique to the meaning.

Re: I am endlessly fascinated with content tagging systems

#145

Post by all2 »

ok_dad wrote: post #136
travisjungroth wrote: post #111

Earlier quoted context omitted.

The issue is that system has nodes and edges, but no concept of distinct graphs. That leaves you trying to fit all notable human knowledge onto a single graph, which is non-optimal. Whether it’s also a DAG, tree, or something else doesn’t even matter. Ontologies are like languages. There is no correct one. What matters is how good a fit it is for the problem at hand and that you’re all using the same one! If half the…

Isn't this literally just saying we need another layer of categorization on top of the categorization layer?

Perhaps "adjacent to" rather than "on top of"? I've started looking at this kind of problem in terms of DB queries or set relations. Even "organization" can be a set relation if there are the right bits of metadata in place.

Re: I am endlessly fascinated with content tagging systems

#146

Post by travisjungroth »

ok_dad wrote: post #136
travisjungroth wrote: post #111

Earlier quoted context omitted.

The issue is that system has nodes and edges, but no concept of distinct graphs. That leaves you trying to fit all notable human knowledge onto a single graph, which is non-optimal. Whether it’s also a DAG, tree, or something else doesn’t even matter. Ontologies are like languages. There is no correct one. What matters is how good a fit it is for the problem at hand and that you’re all using the same one! If half the…

Isn't this literally just saying we need another layer of categorization on top of the categorization layer?

It’s saying you need support for multiple types of categories. You could use the same system to organize itself. No need for a meta layer.

Re: I am endlessly fascinated with content tagging systems

#147

Post by RussianCow »

dahdum wrote: post #71

I adore tagging systems and have worked on them in several different applications and implementations, but there are always pitfalls and trade offs, and it’s possible to bury yourself Nowadays I nearly always store the assigned tags as an integer array column in Postgres, then use the intarray extension to handle the arbitrary boolean expression searches like “((1|2)&(3)&(！5))”. I still have a tags table that stores…

The tradeoff here is that you lose the foreign key constraint, correct? So if you delete a tag, there is no way for the database to automatically remove all references to it. Or is there some way to do this now?

Re: I am endlessly fascinated with content tagging systems

#148

Post by laszlokorte »

xg15 wrote: post #82

I worked with the Wikipedia category system a few years ago, and you could see the problems with hierarchical tagging systems right in action back then. (Though it may have gotten better in the meantime) The system appeared simple: There were just two relations, "Article A is a member of category B" and "Category X is a subcategory of category Y". However, in practice, the community was using this system to represent…

Clearly the solution to all of this would be the category of all those categories that do not contain themself.

Re: I am endlessly fascinated with content tagging systems

#149

Post by aaroninsf »

xg15 wrote: post #82

I worked with the Wikipedia category system a few years ago, and you could see the problems with hierarchical tagging systems right in action back then. (Though it may have gotten better in the meantime) The system appeared simple: There were just two relations, "Article A is a member of category B" and "Category X is a subcategory of category Y". However, in practice, the community was using this system to represent…

An exactly analogous problem exists in the Collections hierarchy at the Internet Archive, of uploaded/digitized material (not the Wayback Machine web captures).

A single graph is applied locally with very different semantics; and absent a distinct tagging systems, collection membership is sometime used to mark material for treatment in some way.

Re: I am endlessly fascinated with content tagging systems

#150

Post by pphysch »

My current solution to this problem is just putting a JSONB column in relevant tables. GIN indexes do the heavy lifting as needed.

This lets us implement arbitrary, queryable ontologies on top of the data without requiring further database instrumentation (aside from creating an index now and then).

Post reply on HN