The problem is that major browsers don't offer a way to offer a view into this data model. The integration is important for an efficient workflow.

Firefox and Chrome both will display a local directory. In fact I exploit that for my no-software written personal wiki, github.com/a3n/miki. I bookmark the directory of the top of my wiki, and drill down into directories to get at my pages, except the pages I've bookmarked.

I see no reason why a browser couldn't do that with URLs in local directories.

EDIT: Unfortunately I don't think the first part, drag/drop from browser tab to file manager folder, works in Linux. Tried it with nemo, dolphin and nautilus. But viewing and clicking on directories and contents from the browser works fine.

EDIT EDIT: Dragging a tab didn't work for me. But dragging the icon from the address bar worked fine. Except in Dolphin, which gave me something weird. nemo and nautilus were good. And the "files" clicked in the file manager opened in the browser. And of course, opening them from the browser directly also works.

That's a pretty neat use-case, thanks for sharing. I could see it being useful for my own personal knowledge database as well. Unfortunately it doesn't help with my regular browsing workflow, which revolves around a bookmark bar filled with textless favicons arranged by color.

Hmm, I'm not sure what the anti-pattern you see is.

I just go to the folder that represents the topic (typically /Vendors/Mgfr/Product/) and there along with spec sheets, user guides, etc., are the URLs I've saved.

Then I {wait for it} open a blank tab (ctrl+t) and drag the URL 'file' back and drop it on the blank tab -- Presto, there's the page.

Considering I probably find four or five worthwhile web pages for every hardware purchase, every semi-tricky JIRA problem, every new tech to master ... that'd just be unwieldy in a bookmarks app. And this way it's detached from any given browser.

It's not super portable but I suppose I could start saving stuff in OneDrive (forBiz) or the like and they'd be more accessible from various networks (home/work/travel/what-have-you), but so far it's done the job for me.

That is quite a few more steps than just clicking a drop down folder on my bookmarks bar or using the omnibar/awesome bar to quickly access previously visited pages. Prohibitively more steps for casual internet users and needlessly more steps for a power user like me who can just further optimize the steps I find myself repeating often.

It's not that maintaining a database of useful links in a filesystem structure is an antipattern, I just find it too limited and hierarchical for most purposes. Knowledge databases like Zotero[0], or mindmapping software like Freemind[1] are good for when I need to build link databases for specific topics and projects, and a combination of web scraping and offline data consolidation and organization provided by such software has so far been the best approach for me.

If I find myself visiting a static page more than 2 or 3 times, it really makes no sense for me not to just archive the page on my hard drive in case it later becomes inaccessible and then tag it for easy searching.

[0] https://www.zotero.org/

[1] http://freemind.sourceforge.net/wiki/index.php/Main_Page

Makes alot of sense