Thoughts on HPI and future of cl-ownpress

Reflecting on similarities and differences b/w HPI and what I want cl-ownpress to build into. Thinking out loud.

Recently I came across HPI again. HPI has a lot of similarities to what I want cl-ownpress to be. Its goal seem to be enabling a user to take ownership of their own data that's locked in online platforms. This is exactly what I want cl-ownpress to do for me. This realization gave me pause, and here I am rationalizing continued existence of cl-ownpress.

cl-ownpress is supposed to be a part of a bigger ecosystem for personal computing. By personal computing I mean hooking into every aspect of my personal life. My thoughts (daily journal, captured ideas), finances, things I own, people I interact with, work I do and everything else. In my ideal personal computing ecosystem, different independent pieces interact with and build upon each other.

For example, I want to have a CRM to manage people and my relationships with them. I want the ability to mention people from my CRM in my daily journal, and then be able to make queries like "all interactions with person <X>". Similarly, I want to utilize my CRM to also mention people in my blog-posts. These mentions however should get intelligently published. e.g on my personal website, mention of a person should become a link to their own personal website. When publishing to mastodon, it should become a mastodon mention.

Another example, I'd like to maintain an accounting ledger of all the money I spend. I want to utilize this to also maintain a list of things I own. I'd like to utilize this in turn to review things I use, and publish my reviews to my personal website and/or where I purchased it from.

Publishing a subset of all the data in my personal computing ecosystem is where cl-ownpress comes in.

At present, cl-ownpress happens to be the most complete component of my ecosystem. I have a few other pieces in pretty-much abandoned state while I have replaced them with some untracked lisp code (mostly in my Emacs setup).

HPI seems to provide same functionality as cl-ownpress's providers. So why not replace provider-s with HPI and build on already done work? Well, things I wanted so far (converting my denote notes to blog posts while using my Emacs configuration to do the conversion from org to html) don't have a replacement in HPI. In future, when I want to deal with an external data source and there is an HPI module available for it, perhaps I'll create an hpi-provider.

That settles one aspect of relationship of cl-ownpress and HPI. They're complimentary instead of contradictory. HPI is essentially a set of cl-ownpress provider-s. But I believe there is a philosophical difference in how both these projects approach gaining ownership over user's data.

HPI accepts that data is going to be produced on walled platforms, while cl-ownpress establishes user's own platform where data is originally published. Cl-ownpress then uses the walled platforms for interacting with people these platforms have managed to hold hostage.

Admittedly HPI has a broader scope, it deals with more than just getting data back from walled platforms, but in the aspect of 'publishing', I believe this holds true. With cl-ownpress, I want to invert the relationship. Instead of "the platform <X> uses me and locks in my content", I want it to be "I own my content, and use the walled garden to interact with its hostages".

The subtle difference is important to me. This is an attitude I'd like to have towards life, and what is my personal computing ecosystem if not its reflection.