I think the theory is correct - in fact I think I mentioned something similar in another thread - but in practise it’s not that practical. On top of this, the thought of thousands of files littered around my machine would trigger my OCD something terrible.
If everything could be searched without structure there’d be no need for Lightroom, Photos, Music etc, but the fact they exists demonstrates it’s not there yet.
Having said that, most people do close to this with the web. With the exception of the dozen or so sites I visit on a daily basis, everything else is found via Google. I can’t remember the last time I actually bookmarked something or even visited my bookmarks page - it seems just as fast to search for things I need.
I use tons of bookmarks. In Safari entering the first few letters is usually enough to get the bookmark to show up. Hit return. Done. I’m not sure there’s anything faster. Before iPhone and Safari bookmark syncing, I had a script that exported my Safari bookmarks to an html file that I dumped on my personal web server just so I could access my usual sites from afar.
I cringe when I see people go to Google to enter target when all they could be doing is enter target.com.
Or use a decent browser like Safari (with Include Safari Suggestions enabled) and then typing tar in the address bar is already enough. The less use of Google, the better IMHO. DDG all the way here. There’s always !g after all.
I find tags very useful when working on projects to note the state of certain files. For instance, any PDF where I’ve highlighted certain sections during my research will get tagged with annotated. Or if I’m processing files for another reason I’ll use tags to mark what stage they are in.
The other way I like to use tags is when I want to mark files for future action. If I’m travelling, I might use a ‘to print’ tag which is applied to files in all sorts of locations in the filesystem. I’ll create a saved search that has the tag as its only condition. That way I have a ‘folder’ on my desktop containing all the files, and it’s easy to deal with them at the appropriate time. Here’s an example from the last time I did a clean install of MacOS:
I’ve also used tags+saved search to create a folder of disparate items inside a project folder, as @eyeless does when keeping track of photos without moving them from their location.
I don’t agree with this theory either. I think it works on a small scale. I mentioned “data soup” above, which users from the swinging '90s will recall is how Apple Newton stored information. It worked great in that relatively small pot, but I do wonder how sufficient the search capabilities would need to be to retrieve disparate chunks of data on an immense scale like we have today.
I hadn’t thought of Google as an analogue to Apple Photos, Music, iMovie, Final Cut Pro, etc. But with all three of those, as @Shamino alluded to above, they are really just catalogues of files, with auxiliary pointers to fragments of those files. Google is the most transparent example, because it points directly to URLs, which are either static files residing on a server or potential files that are generated on demand by a CMS.
The suites on my local Mac are also fancy catalogues, each of which operates on a subset of discrete files and generates a visual representation of what’s there. (Even FCPro is at heart a catalogue of files, and uses edit decision lists to generate a composite representation of those files—which is really all a video is.)
Whether Google on the Web or Apple Music, you are always looking at a discrete file, which you can prove by surfacing the original. A data collection à la Raskin couldn’t be pulled out that way in an understandable form…you’d need a search query to assemble what you’re seeking.
Oh! Those five colors. I have been using them from the start. (I’ll see if I can change my vote. But read on.)
Let me be clear: They are simply colors on line items (sadly, merely colored dots when “Boy-do-we-look-slick” design). I was blissfully unaware of other uses. Nor, having read a very small number of the preceding posts, would I invite misery, by trying other uses.
Four schemes:
Regular user account:
o Pathfinder: Red Important
o Thunderbird: Red Important, Orange ToDo, Green Money
Confidential user account
o Pathfinder: Red Important, Blue (if not in the separately encrypted volume) Confidential Work
o Thunderbird: Red Important, Orange ToDo, Blue Agency matter, Purple Strange
(I ignore the predefined labels. And I much prefer the neat use of the number keys in Thunderbird; hat the jumping around when I accidentally type when (rarely!) in Finder (I forgive Pathfinder for being compatible with Apple, one of its many virtues; and praise them for using the full line color, instead of the cutsey-tiny dots.))
I agree in general (twice, once for using Google and once for not adding .com to the search term), but I have a specific counter-example (which could be avoided by using “tons of bookmarks.”)
I have read (but have never seen firsthand) that some sites take a URL that is one character different (or a pair of characters swapped) from some URL where people might enter sensitive information. For a situation where I might enter sensitive information and I do not have the site bookmarked, I might type site into a search engine and then click on site.com to avoid the possibility of typing stie.com in error and then divulging information inappropriately.
My approach to filing took a turn when the why-organise-use-tags era arrived.
I did look again at my file naming strategy which is consistent searchable and informative (to me) and used steadily since the late Nineties and realised that I didn’t need the endless and at times less than stellar cascade of folders which projects ended up being.
So I use Hazel to automatically add all text files in one folder, spreadsheets in another, movies in another etc. I don’t tag but the filename contains the data I need for searching.
When it comes time to archive I add them all to a DevonTHINK database per project and tag within that for classification of various types of projects.
Your example of websites run on CMS is actually exactly that – every URL is essentially a search query which the CMS then uses to assemble the page by pulling pieces from different databases and files on disk. So there is no discrete file behind the page. This page is a good example of this. There is no file containing all our musings, instead a collection of databases and image files.
I use them routinely in my work, identifying my role in jobs, so I don’t accidentally take on someone else’s task.
The change to tiny coloured dots was annoying until I found ColoFolXS (https://trollin.loos.li), which is invoked with a right-click and applies your choice of colour (user-definable) to the folder icon as well, making the tag stand out better.
I am used to tags from using them for many years in my wiki. I like to use them in MacOS too, but I find it uncomfortable cause of the lack of tools and the smooth integration in the OS. But I still use it with hazel, obsidian and with the extension MailTags. And with these tools I use them extensively. I would like to see a software or better support for the daily use with files. Tags are great to organise my stuff.
I use them in a very limited fashion. For example I keep file notes on my desktop indicating important appointments coming up and use a colored tag to indicate types of appointments (being old most are medical lol). This gives me a quick view since they are listed by date. Other than that I sometimes use them with multiple tags attached to a given file for quick finding folders I regularly use.
I too miss the ability to have the whole file or folder name highlighted by the color.
Interesting discussion - given me some ideas. Thus the need for TidBits discussions!
I use them at tax time to tag each slip or document when I’ve entered it into the tax software. And monthly I use them to tag bank statements as reconciled.
I’ve used tags regularly for years. But perhaps my uses are just specific to me. I only use column view and tags give me an easy visual indication about a file. For example, I scan a lot of documents. Some of those documents have to get uploaded to a financial site, but I don’t want or need to separate them out. Once uploaded, I tag the file as uploaded to [site name]. Then I don’t have to do other kinds of organization with the files. It makes it really easy.
I like to use tagging with the photos I take. I organize my photos by month. Tagging lets me easily know favorites, etc in column view. Also, GraphicConverter shows finder tags so that really helps me.
A sort of silly use of tags is marking youtube videos I’ve downloaded (shout out to Softorino’s downloader) as watched or partially watched. I can’t always remember what I’ve watched otherwise
Interesting. No, it’s not working for me, but I use Maestral to sync, not the actual Dropbox client, so maybe that’s a limitation of Maestral.
(I switched to Maestral a while back as the Dropbox client had gotten so bloated and buggy it just wasn’t worth using any more. I’d actually considered canceling Dropbox until I discovered Maestral. Maestral isn’t perfect and has a few limitations, but it’s way better than the Dropbox client.)
I use tags routinely, but only for identifying the source of the application or something similar. I divide the sources into Apple, App-store, Microsoft, 3rd-party written, shareware, shareware-not-yet paid-for, freeware, unknown, etc. Unfortunately, Apple won’t allow their applications to be tagged. Why?
