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

Yet another hot take on “folders versus tags”

eleanorkonik.com

Post reply on HN
Print view View original

151–160 of 181 posts

Re: Yet another hot take on “folders versus tags”

#151

Post by marginalia_nu »

foxfluff wrote: post #147
marginalia_nu wrote: post #137

Earlier quoted context omitted.

It is a bit like the original Myst, which was made of a bunch of "places" that you could walk between along pre-defined paths. But that's also how we conceptualize space. One place is my bedroom, it connects to a space that's a corridor, which connects to a few spaces like the kitchen, the bathroom, the living room. In one of these spaces is a bookshelf, which contains a shelf for sci-fi, a few shelves worth of compu…

Yeah these hierarchies are something that really slow me down. Was it arch/aarch64/dts? Arrgh no, arch/aarch64/boot/dts? No, actually arch/arm64/boot/dts.. Similarly, I couldn't decide on a hierarchy for my apartment. What's above living room? If root is your entire house, does that mean you can skip into any room without going through another? I can't go to the bedroom at the other end of my apartment without going…

You do get into delineation problems when you investigate the conceptual notion of space, as well as overlapping concepts, but I don't really see this as a problem. The world as it is, and the world as we conceptualize it is not the same thing. Our conceptual view of the world does not play by the same rules as the things in themselves.

Anyway, given a google maps view of the entire world, completely unstructured, Where's Waldo? You get exactly the same problem you had with arch/aarch64/dts, except now you don't even have a hierarchy to go on. He could be in Tokyo among 13.5 million people, he could be in Siberia, he could be on a boat on the ocean. Heck, he could be in a mine underground. He could be on the International Space Station.

Re: Yet another hot take on “folders versus tags”

#152

Post by daenz »

hocker_news wrote: post #140
daenz wrote: post #100

Earlier quoted context omitted.

I'm really trying not to overly promote Supertag[0] in this thread (disclaimer: I am the author) , but it does exactly what you describe. 0. https://amoffat.github.io/supertag/

This looks great! Am I understanding correctly that all tagged files need to have a unique filename to be addressable?

No, but that was a challenge to overcome. See #2 here[0]

0. https://amoffat.github.io/supertag/faq.html#why-are-my-files...

Re: Yet another hot take on “folders versus tags”

#153

Post by Buttons840 »

bityard wrote: post #67

I keep my personal notes in an app that I wrote which is backed by SQLite. It resembles a wiki. First, I tried tags. This seemed like a good idea and was a lot of fun at first, but eventually I got _really_ tired of having to curate tags for all of my notes and boy there were a lot of them! It's not something you can just do once either, because every time you update a note you have to remember to change the list of…

That's a really valuable insight and well said, thanks for sharing.

I wonder if tags could work like spam detection. For each tag mark a few things as "tag" and "not tag" and let an AI figure out everything else.

If there's a tag you never look at, and so it's poorly trained, who cares? You never look at it. Commonly used tags will be better trained, because you will have marked more things as "yes tag" or "not tag".

Re: Yet another hot take on “folders versus tags”

#155

Post by alpaca128 »

fouc wrote: post #150
alpaca128 wrote: post #124

Earlier quoted context omitted.

Also this limits the number of tags that can be assigned to a file, with the limit varying between operating systems. But if you don't use filenames you have to store this tagging information externally, which has to stay in sync with file locations and ideally not be reliant on one specific program. There's folders + symlinks but I imagine that might be complicated to manage as well.

I think it's possible to avoid using symlinks and just use plain folders to represent tags only. By using a special naming strategy on the folders themselves, it would be possible to have a set of scripts for "tagging" files and operating on them. Tagging would be equivalent to just moving the files into the relevant nested tag directories. No duplicate files or symlinks necessary.. Can make scripts like tagcd and ta…

I don't think that would scale. The nested directory structure could reach a depth equivalent to the number of tags to support all combinations. Meanwhile with symlinks you only need one folder per tag and it can be comfortably used with file browsers.

Re: Yet another hot take on “folders versus tags”

#156

Post by smcnally »

mason55 wrote: post #143
bityard wrote: post #67

I keep my personal notes in an app that I wrote which is backed by SQLite. It resembles a wiki. First, I tried tags. This seemed like a good idea and was a lot of fun at first, but eventually I got _really_ tired of having to curate tags for all of my notes and boy there were a lot of them! It's not something you can just do once either, because every time you update a note you have to remember to change the list of…

This is how I tell people to use Confluence. There’s no hope of ever having the general set of all articles organized in a way that makes sense so just try to make the article searchable. If it’s about an error, make sure the error message is in there. If it’s about configuration of a feature make sure the feature name and “configuration” is in the title. Also make sure to write a paragraph describing what the articl…

Making docs discoverable is good, and this works well for retrieving mostly static documentation and tickets. It gets complicated when when resulting items are used for troubleshooting or new documentation and tickets.

When results, e.g., include multiple tickets and documents referencing the same error or configuration, which do you use for troubleshooting? Which results are the authoritative / canonical / source of truth for documenting the latest configuration directives?

Versioning used well for the input helps. Organization and ontology is additional context for using search results well.

Re: Yet another hot take on “folders versus tags”

#157

Post by yellowapple »

> Take my word for it — it’s much easier to tell a computer “do this to every file in this folder” than to tell it “first, find everything with this tag, then put those into a string, then do a thing to every file in this string…”

The solution here would be a shell (or command) that's able to "do this to every file with this tag", right?

In the GUI world, Haiku's Tracker is an example of how such queries might look (https://www.haiku-os.org/docs/userguide/en/queries.html); a command-line tool (perhaps with nicer syntax if possible) could readily do the same (and then perhaps do something with the results). Haiku's advantage here is that it uses attributes instead of tags, so it's a bit of a richer experience; also, most Haiku/BeOS software is already aware of attributes, so that also makes it easier to rely on those attributes being actually used.

Re: Yet another hot take on “folders versus tags”

#158

Post by yellowapple »

Quekid5 wrote: post #114
em-bee wrote: post #110

Earlier quoted context omitted.

for me, folders are just a limited form of tagging. since i switched from mutt with its folders to sup which uses tags for email, i don't want to look back. first of all, my extensive list of mutt folders was trivially translated to equivalent tags, so i could continue as i was used to. but then, tagging allowed me to create additional tags as i needed them. i didn't tag everything, and sup also indexes all mails and…

> for me, folders are just a limited form of tagging. I understand that this is a very subjective thing... but folders offer an exclusion which isn't immediately available via tagging. Tagging is implicitly an OR operation. Now, your search engine might offer ways to add "not X", but it's a tradeoff. It you have highly-compartmentalized bits of info you end up with the problem of "how much do have to explicitly exclu…

> Tagging is implicitly an OR operation.

Doesn't have to be. I can't think of many tag-based search systems that default to OR; usually the default is AND, and that's pretty sensible.

There are plenty of text-based search systems that default to OR, though (see also: pretty much every mainstream search engine), and it makes searching a fucking nightmare, so I definitely get the aversion.

Re: Yet another hot take on “folders versus tags”

#159

Post by daenz »

alpaca128 wrote: post #117
daenz wrote: post #100

Earlier quoted context omitted.

I'm really trying not to overly promote Supertag[0] in this thread (disclaimer: I am the author) , but it does exactly what you describe. 0. https://amoffat.github.io/supertag/

Does this also allow looking for e.g. files that are either in tag A or B or both? Intuitively I'd guess this is a limitation when you use a path to filter files, though I might be wrong.

Yeah the path parts are AND, with no ability to do OR, but you can do NOT. If you wanted to do your use-case, you'd need to look in 3 separate places... /A, /B, and /A/B (or /B/A)

Re: Yet another hot take on “folders versus tags”

#160

Post by bccdee »

sseagull wrote: post #113
endisneigh wrote: post #75

What can you do with folders that you cannot with tags?

It’s not about ability, but mindset. Folders are tree structures where each file lives in only one place (in practice, hard links are uncommon, and symlinks are more common but obvious). That abstraction is useful when navigating, looking for related files, etc. Tags are more general. Yes, they could implement the features of folders, but the connotation of tags is that they are flat and unstructured.

Sometimes programs implement hierarchical tags, and that appeals to me the most. That way, you can navigate hierarchies of tags the same way you would navigate hierarchies of folders, but there's more than one hierarchy that can be navigated. And because you don't need to worry about encoding every piece of information into a single path, the hierarchies can be shallower and more convenient.
Post reply on HN