| Hey! Following the advice dang left in the comments, here’s the backstory for this workflow: I've been taking plain text notes on my computer for almost 20 years. I’ve been keeping bookmarks for longer than that. A few months ago, I subscribed to Pinboard and imported all these 1000+ bookmarks, only to find that hundreds of them were dead links. I could remember many of those websites and pages, I could recall how and why they were important to me; I just couldn’t see them. And never will again. It was depressing and alarming. So, weeks later, when I stumbled upon Jeff Huang’s manifesto “Designed to Last”, it deeply resonated with me. I was setting up a blog, then, and even considering a new note-taking setup, and future-proofing became an even greater concern for me. Like I said, I keep plain text notes. I use markdown a lot, and feel comfortable migrating my notes between different apps and platforms. For example, I’m currently moving from Bear to Roam. The text itself is never a problem — and probably will never be — but linked content is a whole different beast. Each time I switch apps, I either lose inline images or have to relink them, sometimes manually. And URLs often go bad, so the note is suddenly worth a lot less, if anything at all. Being an enthusiastic Alfred user (it's the single most important app on my macs), I started to play around with the idea of copying every image to a single "resources" folder in my Dropbox, before using it inside my notes. With a single keyword or keyboard shortcut, Alfred would take care of uploading an image via DB's API, get the share link for it and then put it in the clipboard, properly formatted with markdown. The same principle applied to any file attachment. Half of the problem was solved. But what about web pages? I considered using wget to scrape the page at any given URL, store it on Dropbox and do the same routine. I still entertain the idea. But then there's the Internet Archive's Wayback Machine. And I love it so so much, that I wanted to contribute. So I went with it instead: Alfred takes the URL, feeds it to WBM through curl, and then returns a link to its time stamped snapshot. Actually, it returns both the live and the snapshot links, the latter with a cool hourglass emoji as anchor. So, now I don't fear losing track of linked files or worry too much about websites going down. Once I had these two bases covered, I thought to myself "this is pretty powerful. I should share it." And so I did. I hope more people find it useful, and would appreciate any comments, ideas or suggestions. Cheers! |