OP describes drag-and-drop creation of *.url files in Windows:
[InternetShortcut]
URL=https://www.afewthingz.com/browserbookmark
In macOS, selecting URLs and dragging to Finder creates
*.webloc files:
<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd">
<plist version="1.0">
<dict>
<key>URL</key>
<string>https://afewthingz.com/browserbookmark</string>
</dict>
</plist>
macOS (Sequoia 15.0) also handles *.url files appropriately: file type is identified as Web site location and opens with the default browser.
My monthly opportunity to put out the idea that bookmarks should be centered around content and not metadata (links).
I've written a lot about this, and I got so annoyed with bookmarking and highlighting services getting it so frustratingly wrong[1] that I wrote my own solution from the ground up in 2020[2], and I have never looked back to browser bookmarks or services like Pinboard, Instapaper, Readwise etc. which are built around bookmarking metadata instead of content.
It's amazing once you get the mental model, and if you aren't interested in using a service you can easily build something that suits your own needs over a few weekends.
My favourite part of this mindset switch is that it makes bookmarking user generated content[3] both sane and easy, and automatically enriching those bookmarks with additional metadata a breeze.
[1]: https://lgug2z.com/articles/the-bookmarking-data-model-is-wr...
[2]: https://notado.app
[3]: https://lgug2z.com/articles/best-of-hacker-news-comments/
I don't know... when I bookmark a page, it's because I want to get back to that exact, specific page in the future.
Do you mean you want to go to back to the exact specific sentence(s) on the page you remember reading?
... which is what happens when you click either the link to the source which is stored as metadata or the link to the automatically archived website copy.
Well, of course.
I was responding to this:
> bookmarks should be centered around content and not metadata (links)
Perhaps I'm not understanding exactly what "centered around" means, but to my ears, that statement sounds like it would not be an improvement for my particular use case.
Not saying the idea is a bad one at all -- just saying it doesn't sound appealing to me personally. But I also suspect I may not really be understanding it.
There is also a self hosted solution called Wallabag https://wallabag.org/
Same concept its about archiving rather than just the link, given how quickly links often die its often what you want depending on why you bookmarked it.
> Same concept
Unfortunately it looks like Wallabag has the same fundamental issue of treating links as primary entities and scraped content as additional metadata that I described in the first article linked in the parent comment.
Especially when it comes to long form articles which cover multiple topics or are by their nature inter-disciplinary, it is essential for highlights or slices of content to exist independently of their source, while retaining their source as metadata, and allowing them to be linked independently (via tags, collections, feeds, titles etc.) to other slices of content (ie. commentary on the same article).
Archiving is an important step forward though, especially for a self-hosted solution, and especially after so many people have been burned by Pinboard's failure to deliver on its archiving promises for a paid product. I ultimately took a different approach to this and instead of maintaining my own scraping/archiving product, built an integration with the Wayback Machine[1].
[1]: https://lgug2z.com/articles/notado-07-2023-update/
The internet changes too much, that bookmarking for content is sometimes a futile effort. And even then, a browser like Firefox lets you tag liberally, if that's your thing.
Just a shout out to https://github.com/mbnuqw/sidebery. My favorite productivity extension. I'm a tab hoarder, this makes my life manageable and gives my Firefox all the screen real estate by using keyboard shortcuts to open/close the tabs easily.
I also use the Firefox css to hide the top sidebar, so I get maximum screen usage.
Their bookmark feature is pretty awesome too.
Thank you! I just learned I don‘t need Arc to achieve this
I appreciate how universal and decoupled this approach is, but it doesn’t fix my main problem with browser bookmarks which is that management overhead gets to be problematic and makes me want to not bookmark things unless there’s adequate “justification” for doing so.
This is what fuels a lot of my tab hoarding. Tabs are quicker/easier to clean up. This has led some browsers (like Arc) to blend tabs and bookmarks into the same thing, but I’m not sure how that this is the right approach either.
I’d like to explore bookmark manager design/UX in a project of my own at some point. It’s not something that’s gotten much attention in browsers in something like a couple of decades, and while plenty of external managers are out there none I’ve seen really nail it IMO.
For some time now I had a similar tab hoarding problem. My stop-gap solution is crammming a locally hosted markdown text editor in a new tab page. This way my bookmarks live as a Markdown file on my computer and I can easily add or remove links with as much additional comments as I like.
What I would like to add to it besides tons of polish is for it to be an extension that would also expose those bookmarks back to browser in form of bookmark folder syncing with the underlying markdown.
For me, it totally fixed the problem you mentioned: each time I find something really interesting, I drag and drop the bookmark to either a folder "MISC" (unsorted) or to a dedicated folder if it's specific to a project I'm working on.
Since the shortcut's file name contains the Page title, I can later search with my OS's search tool "curve fitting .url" => it finds the right bookmark.
If I use it in a particular project, I can copy/paste this .url file into the project folder, etc.
Having thousands of bookmarks creates no real problem: you end up with thousands of 1KB file in various folders, there is no mental burden in that: it doesn't add "weight" to the UX of a particular browser extension, since they are only files.
Drag-drop only takes 1 sec, there is no friction, no prompt.
Management overhead must be self imposed? I have tags on some of my bookmarks and put them in folders sometimes but that's it.
Perhaps. In my case bookmarks tend to be append-only because going through and reviewing them for relevance, link rot, etc is tedious, which then makes finding bookmarks later more difficult, particularly when it’s been long enough to not remember the title/address of the bookmark. Tags can help but like folders lose effectiveness with number of bookmarks.
and if you really aren't sure you can put it in "unsorted" which is no worse than a forever tab
Filesystems often aren't very efficient at lots of small files.
If they could handle compressed archives transparently then an array of files, maybe extended from the old windows URL= style files, might work.
An SQLite file also sounds like a great way of handling URLs, which Firefox does:
https://stackoverflow.com/questions/464516/firefox-bookmarks...
It's an interesting idea but missing vital features for me. For example, the star in Chrome tells me that I have bookmarked this page in the past so I avoid having duplicate bookmarks even after editing the name. The standard synchronization also makes it easy to bookmark a link on my phone and then deal with it once I'm back at my computer. Now I would have to figure out a way to somehow download the URL as a file on my phone so it syncs to my computer. The favicon is another neat thing to have on bookmarks.
Somewhere along the way it just feels like a backup makes more sense.
What is the real problem on having duplicates? They are only 1 KB files.
Having duplicates with different names is even better, and helps to find it more easily in the future: let's say I have bookmarked 2 times this question https://stackoverflow.com/questions/19165259/python-numpy-sc...:
1. "python numpy/scipy curve fitting"
2. "scipy.optimize.curve_fit question"
Later I can find it with query="curve fitting", or I can also find it with query="optimize". So it increases the chance of me finding it again :)
It's probably tied to how I use bookmarks but I strive for quality over quantity. Some are just temporary until I get the chance to write down the important information or watch the video - others are more constant, but when I go through and purge bookmarks I want them gone. They will just clutter my bookmarks, waste time and make it difficult to find things. In your example about bookmarking the question I would instead transcribe the knowledge to Obsidian and link to the source.
Not saying there's a right or wrong. Just down to how people treat bookmarks.
I think it's mostly a very good idea, but much less accessible compared to the omnipresent bookmarks bar, so I will keep using the bookmarks bar (whose primary downside is the vendor lock in, imo).
My own solution is along these lines. I have a static html page on my personal server; that's the home for all my browsers. (It's under git, of course.) Just flip to my ongoing mosh session to my server, and a trip into vim can add/move/delete anything desired. It's currently an HTML table, which tells you how long this technique has been serving me well.
Interesting concept, trick, but no.
- Can I write comments about some bookmarks?
- Can I tag bookmarks?
- I cannot self-host it, hence you have to sync things between devices, which is stupid
- Can it automatically do import / export?
- Can it support multiple users?
I am using my own bookmarking system, which solves these issues for me, but again, it is not a jack of all trades. I do not see your aunt running it in portainer. I am still developing it, so it is not super stable. Even with these shortcomings this is how I consume internet now.
It is "bookmarking system" x "rss reader" x "simple search engine"
Link:
https://github.com/rumca-js/Django-link-archive
Bah... just give us the good old del.icio.us.
Not the recent .com HD remaster.
I vaguely remember an older browser just creating files in a folder for its bookmarks.
I wish I could find this folder on my work computer: I only have one work computer, so I don't sync work bookmarks with other devices.
Didn't Internet Explorer do that? It was just the "Favorites" folder c:\users\username.
In Firefox, it's an SQLite database in the profile folder, readily accessible by normal SQLite tools. The profile folder is accessible through the Help menu, if you don't like to dig for it in a file manager.
You can also export them to json from the bookmark manager, I do that semi-frequently so I can "spring clean" my bookmarks.
My old ones are still backed up and can easily be grepped with no external tools if I need them.
Firefox/Netscape used to do that.
This acrticle completely ignored mobile browsers. You can not drag and drop a url file here.
This is a pretty good idea but I feel like it exposes some of the shortcomings in our modern UI stack and file browsers. Users using the tools the OS provides to solve problems like this should be encouraged, however the separation between the file browser UI and the web browser UI feels like it creates a certain amount of inertia to using such a solution. If my UI had enough customizability that I could easily do something like attach a slide out drawer of a file browser view to my web browser windows, I feel like I'd be much more free to experiment with mixing and matching the various tools at my disposal and using my own solutions to problems like this.
I completely disagree.
If the built-in bookmark systems in browsers could support tags, then I would say yes. However, it currently only supports a basic tree concept, with "folders" for links.
This is very one dimensional. I read loads of articles that talks about multiple topics. Especially Hacker News type articles :). An article can talk about, say geo-politics. As an example, perhaps an article on the recent pagers that exploded in Lebenon. This article may also be discussing some cybersecurity topics too. In this case I may want to tag it with 1->n tags.
I currently use Raindrop.io. It kinda works, but it doesn't really have what I have in mind. It also has more features than I think I need from a bookmarking app.
I kinda feel that Digg (wayback, it was one of the first 'Web 2.0' sites had a model that could work.
If I had enough motivation, I think I could probably produce a simple app that does tagging, and only tagging, with bookmarks.
Not sure about other browsers, but Firefox's built-in bookmarks support tags - no need for external apps.
I would like to give Firefox a try. I currently use only Chrome/Edge/Safari.
Let me check if it works in IOS.
Firefox can store bookmark tags, but they don't save with the bookmark export without reading the SQLite database with a different tool: "Allow reading and writing bookmark tags" (9 years ago) https://bugzilla.mozilla.org/show_bug.cgi?id=1225916
With bookmarks as JSONLD Linked Data, it's simple to JOIN with additional data about a given URI.
The WebExtensions Bookmark API does not yet support tags.
Firefox's bookmarks manager exports tags just fine (whether to JSON or the bookmarks.html format). WebExtension APIs are a completely separate issue.
Not only that...but Firefox Mobile doesn't support tags either!
This looks very interesting. Thank you for the link.
It doesn't support Safari as far as I can see. An extension for Safari (especially on IOS), is quite important. This is perhaps only for me, because my general workflow tends to be quickly scanning a couple of articles that I would want to read later, and I would like to easily bookmark them from Safari.
Secondly, its self-hosted only. This is perhaps not so bad - it just means I have to put some thought into where I would host it.
But again, thank you so much for linking linkding :). I am definately quite interested in trying it out.
But this file-based bookmarking system totally support tags :)
Example:
- Let's say you bookmark my article https://afewthingz.com/browserbookmark
- drag and drop => it creates a file "The best browser bookmarking system is already built-in.url"
- you rename this file into "The best browser bookmarking system is already built-in #tag1 #tag2 #productivity.url" in 2 seconds
- later when searching, you search with query="bookmarking #productivity", bam, you find it with tag "productivity" :)
You can put all .url files in a single folder with "#tags" in the filename. It works exactly like a tagging system, no more, no less.
On paper, tagging is objectively better for the reasons you describe. But in my experience, the human brain has an intuition for location and object-permanence which is confused by having the same thing in multiple places.
Every once in a while I go down this bookmark rabbit hole. Tags is the correct solution (for all the reasons your mentioned). I hate the standard folder / tree based bookmark system that browsers and most 3rd party bookmark managers use. Firefox supports tags, but Firefox Mobile doesn't. Raindrop is clunky as hell. And...for along time, that was it.
Luckily, a few years ago I discovered xBrowserSync, which turned out to be exactly what I'd been looking for. It's a stupidly simple tag only based system that syncs across devices. The browser extension makes bookmarking easy. Your data is locally encrypted then synced. It has a phone app. It's open source. And I can self host a server if I want to. There is no "organizing" or sorting of anything. Bookmarks live almost ethereally in the plugin (tho they actually live in your browser's built-in bookmark manager too...but we never need to visit that place).
My only concern is that it hasn't been updated in forever (not that it's ever been broken for me). But I fear the day it does break and wonder if anyone will be around to fix it.
Someone in the comments below mentioned Linkding, which looks like it could work (if the browser extension or bookmarklet turn out to be mobile friendly). I'm definitely going to give that a run and see how it fits. Anyways, enough shilling for xBS (I swear I'm not affiliated with them). Good luck in your search.
I've been using Trello for a while to organise bookmarks and other snippets, but with the recent force-in to rich text instead of markdown and links displaying as "preview" by default the UX has got a lot worse. Yes, there are extensions that make it almost as good as before, and I'm using one, but still.
From the article, I gather that it turns out that filesystems are a good way to organise vaguely hierarchical information. SQLite isn't terrible though either, people should be able to write third-party tools to help manage that.
Out of curiosity, do some of you also use this bookmarking technique?
I use raindrop.io, which has the advantage that it is easy to share bookmark lists with others, which I use fairly frequently.
Seconding raindrop.io. Went from being a tab hoarders with Tab Outliner, but the extension finally broke and isn't supported. Thankfully I managed to import my huge list by munging the JSON file into a CSV. Hate that it's stored on the cloud, but I just export it out into CSV so if I have to move again I can. That said the autotagging and recommendations are great. I have a ton of tabs I didn't organize and it automatically suggests folders to move them to that are correct 99% of the time.
I would mention how many tabs I migrated to highlight how good the performance is, but I'm embarrassed to admit how many I saved...
I don't but I also avoid data silos whenever I can, so I paste links and other things in a text file.
For me, it works well for temporary bookmarks. For ones I want to keep long-term, though, the bookmarking systems provided in the browsers is not adequate. It's too difficult to use bookmarks from other places and browsers, and I find the support for organization to be lacking.
So I run a standalone bookmark server instead.
I have bookmarked and highlighted nearly everything I've read, by topic for the past 10 years.
I agree bookmarking could be files, but the reason for keeping the bookmarks is important to consider and important not to lose.
The piece that makes bookmarks hyper valuable, is remembering why or what was important about them. Annotation-centric bookmarking for me is really valuable. That usually means highlighting.
There's some nice options listed in the comments, I use diigo.com for a while as a paying customer and it's quite capable. Every so often I want to see what's out there, appreciate the links
In my mind I don't bookmark a page, as much as a sentence on it.
First step is am I just keeping it, or reading it. If I read it, I don't want to lose that time to have to spend it again in the future. If I read, I always highlight as I go anything. It kind of makes a journal, and also helps you reinforce if what you're reading is applicable to something you're currently needing to do.
The unfair advantage? When I come back to look for a link, I'm often actually looking for a sentence, phrase, or something I highlighted. I might occasionally put notes on the highlights. You can end up with dozens or hundreds of snippets explaining in and around a concept.
Annotating web pages, creates a feed of those by tag, which can then be fed to other things like sharing topics with people easily. There are other tools too like Readwise that help a lot to extract the insights.
“Browser built-in bookmarking system is good enough”, proceeds to not mention it again and talk about the filesystem instead.
I used to add my links to a github page :)
Now I add all my frequently used links to easyy.click
> No browser extension needed.
No but you need an additional app to search/manage them (the file browser)?
You can use your OS file search tool (even the built-in File Explorer search), that you use for your everyday file searches.
I have to be honest, I search for files about once a year. Tops.
Now then in the browser, I start typing a URL and it's auto-completed from my bookmarks (and/or history). Even the most casual users do the same, just using search results instead of bookmarks.
The idea is fairly sound, but it relies on a bookmark usage pattern which I think is becoming more uncommon.
With this scheme, you can't automatically sync bookmarks across multiple machines though.
Put the folder in OneDrive, iCloud, DropBox, Google Drive, Syncplicity...
Syncplicity is a name I haven't heard in a while. It was the best (free for home) from all those names in the list 10 years ago. It seems they have abandoned the home user space nowadays.
That’s going to mess up when I use both computers simultaneously.
Why? Unless you’re adding and deleting bookmarks every few seconds, I don’t see why it wouldn’t keep up.
The network isn't that reliable. It could easily take a couple minutes for sync to succeed, which is far too much chance of a conflict.
What I want from bookmarks it's not manage them as files, since those files are just links, I'd like to have eventually collected snapshots (like Zotero does), eventually DIFFING through them (because often articles get modified, without changing title/URL etc), instead of a full snapshot maybe just the "Firefox Reader" version saved so I can avoid wasting space in useless bits, check their on-line status slowly and regularly so when a bookmark is broken I got a small alert and I see it "greyed out" and appear in a dedicated "broken bookmarks" page I can try to update (often the same bookmarked page exist but under a different URL and thanks to the cached copy I can look for the new version or a mirror with a search engine).
Files for UIs was an ancient concept trying mimicking paper files, it's about time to use textual pages and search&narrow UIs more than files for many, many things.
That would compete with Pocket (Mozilla's proprietary, commercial bookmarking service). So there's little chance of that happening.
See also: <https://wiki.mozilla.org/Permafrost>