Your second brain

Hey all, recently I was having a discussion with some colleagues about the best way to store/retain/archive useful information and as you might imagine, there seem to be a innumerable amount of ways one might go about doing it. From apps like Roam and Obsidian, to using emacs org-mode¹, to simply storing everything into a single large document, leveraging some Johnny Decimal system to easily traverse the soon-to-be gargantuan document. I was even made aware of courses which one might take to be better able to retain and leverage such systems, one² such that I was made aware of. Despite all these suggestions, I still think my tool of choice needs mention and it's what I like to call 'flatpack HTML'.

To preface this, I don't beleive this to be a novel approach to archiving/ retaining knowledge. I do vaguely remember reading something about using Print to PDF many years ago, and whilst Print to PDF did offer some fantastic attributes: indexable, searchable and available offline, it also does fall short in one major aspect: many websites have an incredibly broken print layout - which should be of no surprise to anyone. Despite working in this industry for the better part of a decade, I can recall only a handful of times where in which the print.css was made mention in the design brief. Often what you find in the wild are print layouts which are barren of the very text they are supposed to display. Also in the context of a easily accessible data source, PDF's are notoriously difficult to work with and don't work very well with the standard set of unix tools.

Saving the relevant HTML of a page would do the job, albeit it would also produce many artifacts: including a folder which houses all the relevant images, CSS, JS etc. This is where this fantastic plugin comes into play: SingleFile³. In effect, all it does is convert images into base64 and inlines CSS/JS, however what this does is it creates a semantically correct, beautiful layout in one neat document and because the text is also available in a HTML document, it becomes trivial to find relevant articles when the need arises. I often leverage the standard set of unix tools to easily find what I'm looking for:

~/Documents/Archives » grep -rnw . -e "mongo"
./Why You Should Never Use MongoDB « Sarah Mei (2020-08-02 15_23_16).html:492:<p>Other than that, there _are_ relational stores provided as a service in the cloud. You don’t have to roll your own. (Then again, you don’t have to roll your own mongo either.) </p>

To date I have around 1044 documents, some of which have changed my outlook as a Software Engineer and are no longer available on the WWW. As such, I'm a big proponent of saving these out when they enlighten/inspire you because they might not be available tomorrow.