Hacker News Reader

Comment by codemac | Hacker News Reader

This is why you should bookmark by copying important information into your own knowledge database.

Would be great if we could download a webpage and guarantee it will always work. These days, however, I imagine lots of pages require runtime fetching of data via javascript. The alternative is whole-page screenshots but you lose the accessibility of the text within. I kind of miss the days when the web was just HTML.

Right click > Save as... in chrome is pretty good.

There's also the issue of plugins compatibility.

Back in the days when the web was just Flash, it wasn't as easy as Ctrl+S to save content you wanted to keep. In a few years, you may need to spin a VM with an older OS to launch a flash-able browser. (https://en.wikipedia.org/wiki/Adobe_Flash#End_of_life)

I wonder if it's easier with java applets.

It's not much easier with Java applets since Oracle tightened applet security (requiring everything to be signed, which the old applets aren't), and deprecating the plugin altogether.

I had to jump through hoops to get an applet I wrote only 10 years ago running.

You can save the current dom you're viewing no matter how interactively it got there, to something like maff / mhtml, or print to pdf and you'll have access to the text & images.

I personally prefer mhtml, though with the recent breakage with firefox, I've been doing PDF.

Archiving something interactive.. that's a different story.

For some URLs given one of their old Wayback Machine entries it would be nice to be able to render those in their contemporaneous browser environments (e.g., going way back maybe Mosaic 3.0 or IE 10) to maybe then have a chance of working with some close facsimile of the original interface and functionality.

Do you have a recommendation for "my own knowledge database", and also, do you have a recommendation for one that benefits everyone, not just me? When appropriate, of course.

Personally, I use zotero to archive, tag, index, and comment on page snapshots

- For you: Depends on what your software preferences are. Evernote is the obvious choice for most consumers, but I personally use org-mode in emacs + recoll for indexing.

- "one that benefits everyone": probably wikipedia+scihub, but I think you would have already guessed that. A knowledge database that benefits everyone looks a lot like the internet, minus the serving issues. So probably something closer to ipfs.

Reply on news.ycombinator.com