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/
Simple general rule of digital marketing: No one reads website copy. They look at pictures and scan headings.
Another simple rule: I inherently don’t care about your product if I’m reading your site for the first time, so don’t spend time describing your capabilities. Simply describe how my life will be better if I use your product, and show me the product doing those things.
Less words more pictures. Less words more value.
(Sorry for being blunt, just trying to help with your conversions)
Funnily enough I apply some of these things in a different context to READMEs for my popular GitHub projects[1], but whenever I see them applied on a product page I often click away very quickly because I associate with snake oil (it wouldn't surprise me that I'm in the minority here).
Also, the line length (in characters) is crazy long. You ought to constrain the width of your text by putting it in columns or blocks, because you're trying to sell me something and I'm too lazy to resize my browser window just to get a not-unpleasant reading experience. I'd do this for https://danluu.com, but not yours yet.
Online recipe sites tend to either be horrible ad-clogged messes, have unreliable URLs (or hosts), or both. Notebook let me select the text of a recipe (ingredients and instructions) and save that along with the URL, something I made frequent use of during its life.
Since Notebook died, I... Print recipes. On paper. Which also hosts the content, annotates with the URL, and allows me to easily take notes. And also doesn't cost me much if I spill things on it. But there are certain downsides as well.
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.
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].
To be completely honest, even if I didn't have a single paying subscriber I would still happily pay to host it out of pocket (and for a long time, I did) because it is the perfect tool for my own needs and it is so deeply integrated into all written knowledge consumption in my life[1] - I will use it until the day I die (and then my wife will open source it).
[1]: I use it to save comments from HN, Reddit, Twitter etc., I use it to save highlights from web articles, I use it to save/import my Kindle highlights, I use it to highlight parts of newsletters in my email inbox - the list is endless
Text is one kind of content. There are many more.
Also known as my 500-ish tabs in a single window.
Yes, there are multiple windows.
I've built a few browser extensions, but not found a super simple way to monetize them. Which technical solution do you plan to use for a paid version of your extension? Authentication? Other?
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.
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.
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.
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" :)
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.
With my app, BrainTool ( https://braintool.org ), I emphasize a visual hierarchy, but also allow notes and full text incremental searching across all saved content. Along with comprehensive keyboard commands, this enables a workflow where you can start typing what you are looking for, iterate through matches and then hit enter when you find it to open in a new tab, tabgroup or window.
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.
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.
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.
[InternetShortcut]
URL=https://www.afewthingz.com/browserbookmark
<?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>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.
(I've not used it since then.)
I also use the Firefox css to hide the top sidebar, so I get maximum screen usage.
Their bookmark feature is pretty awesome too.
My point is: the file explorer seems to have (at least for me) a far better UX than the browser's bookmark bar.
Example: you accidentally renamed a bookmark in the bookmark bar. Can you do CTRL-Z? No! With files in file explorer, you could.
Also im a avid user of keyword bookmarks in Firefox, so I need to store those in Firefox anyway for them to work
I fail to see how bookmarks have vendor lock in. Every browser I've used has bookmark import/export to a format understandable by other browsers, like HTML.
Somewhere along the way it just feels like a backup makes more sense.
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 :)
Not saying there's a right or wrong. Just down to how people treat bookmarks.
In your example, how do you add/change tags for the same url if you have multiple files (and you don't even know how many)?
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 wouldn't be surprised if 95% of people who get a new phone, for instance, never create a bookmark on its web browser.
Possibly the % is higher on desktop, but then I would guess the number of bookmarks is still probably in the magnitude of less than 5, and they could be considered more like quick launch shortcuts than a true hierarchal bookmark organisation system.
I don't even know how to create bookmarks on my phone.
On a bunch of computers, though, I have thousands of them, going back years... if not decades.
(So maybe I'm not all that normal.)
Mostly, I just leave the tab open. I have... Many tabs.
- 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:
- Comments? Put them in the filename
- Tags? Put them in the filename
- Sync? Many of us already sync our devices in some way (dropbox/gdrive/syncthing/...). I see it as a plus - it puts me in control, not "the cloud"
- Import / Export? `mv` & `cp`. You can take your export on a USB stick, send it over email, you name it.
- Users? /home/bob/bookmarks, /home/alice/bookmarks
- Can be accessed from any local browser.[1]
- Can be edited with any local text editor.
- Can be liberally annotated.
- Can be readily searched (Ctrl-F, grep, etc.).
- Can be version controlled.
- Can be rsynced to other systems, or served over a local LAN, or privately-managed VPN, should that be necessary.
Within the homepage I can set up various categories, projects, date-oriented classifications (which can be annotations themselves), and of course a healthy and growing "misc" category.
________________________________
Notes:
1. This is occasionally not the case, as file:/// URIs are deprecated. In which case one can serve the file locally e.g., with Python (python3 -m http.server), netcat, etc.
https://github.com/shepherdjerred/sjer.red/blob/main/src/lin...
That said, you'll still find old-school public homepages with lists of links on them, and they really are goldmines.
There used to be an excellent service that allowed you to save downloaded versions of entire pages to your account, it was called furl.net IIRC. The service was well ahead of its time as it included search capability within the content of the saved documents. It was extremely handy for building supporting documentation for all kinds of research. From time to time I entertain the idea of recreating furl and testing if it would catch on this time around.
I've noticed this. The worst part is if you are looking for some specific piece of information similar to other links that are still valid it's hard to tell if you have the correct information at hand or not.
Chrome can take a full page snapshot of a webpage but the image is not high res.
Putting tags in the file name with a hash mark feels “ick” and like the Wrong Way to solve this problem. Using folders and symlinks goes with the “grain” of a file system based solution.
I respectfully disagree. If we were speaking about millions or billions of data points, yes, performance would be important, and we would look for the "Right Way" to do it, either with a DB or with files+symlinks, as you describe.
But here simplicity and portability is key. "Title of the page #tag1 #tag2.url" does the job: easily searchable with the OS search. Why complicate this with symlinks and folders for tags for just a few 10k bookmarks?
I’d prefer the folder approach. Instead of running a search you’d just open the folders.
Not hard either, on macOS you can option drag to make an alias.
The several sites I use daily are in mah browsers' cache, so I just type a few letters in the address bar. (Yeah, not "mah browser's"; all mah browsers.)
No but you need an additional app to search/manage them (the file browser)?
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.
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...
IMHO filesystem efficiency questions never arise for bookmarks of a user of a computer. If one day you want to do some data mining on your 10k bookmarks, it will probably take < 1 second, even if done with Python.
Do you see a real-life situation for which reading a .url in 1 µs instead of 100 µs would make any difference?
(If you're speaking about search/querying, then the OS search feature does it for us)
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.
See also: <https://wiki.mozilla.org/Permafrost>
Ok, how do I see the full list of tags and be able to rename a single tag and let it propagate to all bookmarks? Not trivial
> See the video, the drag-and-drop creates a .url shortcut file:
Dragging is worse UI vs a shortcut.
Also, how do you sync with a smartphone?
And what about drag/open on Mac vs Windows where url file formats differ?
SQL-based system, many-to-one links, redefine the tag name in the Tags table without touching the numerical primary key, the bookmarks keep their numerical reference (via a many-to-many BookmarkTags table) to it. That's how you do it.
Stuff like that is probably why browsers use SQL- in stead of file-based bookmarking systems nowadays.
Also, I personally miss good old [del.icio.us][2]. It was way ahead of time.
[1]: https://divinedragon.github.io/saves/
[2]: https://en.m.wikipedia.org/wiki/Delicious_(website)https://chromewebstore.google.com/detail/one-page-favorites/...
https://microsoftedge.microsoft.com/addons/detail/one-page-f...
https://addons.mozilla.org/en-GB/firefox/addon/one-page-favo...
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.
With multiple synced devices, I should be able to see all synced tabs, and all bookmarks, and manage and search them, all from one unified interface. The Firefox local cache makes this possible.
[1] https://www.reddit.com/r/firefox/comments/1eqjl70/major_issu...
I come from the world of Delicious and Pinboard (lifetime license). I have also tried many other services, such as Instapaper, Pocket, Raindrop, and other self-hosted ones. I currently run Readeck[1] for less than $2 a month on Pikapods[2]. I like it so far; the readability is superb. Now, I need to figure out if I can make some of the bookmarks perpetually public (currently limited to 24 hours).
I’m also not worried if everything gets lost. I might end up with one of the services, so I won't have to worry about it at all.
When organizing shortcuts on my desktop into folders it was sometimes appropriate to reduce a topic to a single html document. For example if the folder has only 3 links in it and is unlikely to grow or if the topic is not really as interesting as the number of links gathered (like a level in a game you've researched years ago) Sometimes I would drop the link lists in the ftp client and send someone a pile of links.
I just noticed that one cant select multiple links > open on the windos desktop nor drop multiple on a browser. It was long ago but I think that worked once upon a time(?)
Will need browsers to support this but doesn't sound too difficult.
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.
Now searching for something in the address bar is much quicker because it will be populated only by sites important enough to warrant a bookmark.
I have tons of keywords in muscle memory now to trigger queries on many sites.
My bookmarks are also curated very well because I actually need them to be.
Working for me with Mint (21.3) + Firefox (130.0.1). However, Nemo seems to treat the resulting .desktop files specially (reporting them as 0 bytes in size and a text/html MIME type), and trying to open them with a text editor doesn't work from Nemo (but does from a terminal).
So I run a standalone bookmark server instead.
Instead add hashtags to the end of the URL and bookmark them like normal. This way you can search them based on context without having to faff about with files and folders.
Of just email the links to an email address and add the hashtags in the body of the message.
https://news.ycombinator.com/item?id=41695840
(on HN's frontpage today)
Not the recent .com HD remaster.