I finally got reactions and reposts (NIPs 25 and 18) implemented in the Pretty Good Apps desktop client. Next up: zaps. Probably create another release at that point. Then back to work on a basic lists (NIP-51) implementation, which is partially working. (Took a break from that this weekend to work on reactions and reposts). Roadmap after that, in broad brushstrokes: try to expand the use-cases for NIP-51 lists, probably ones that involve "composite" lists via importation of one list into another. (Composite mute lists? composite people lists, for use in nostr feeds? composite bookmarks lists?) Then add more list management features (beyond NIP-51; maybe mix in some NIP-32, maybe add some nonstandard protocols if I have to) until we arrive at DCoSL: decentralized curation of simple lists. Then build on list curation to achieve DCoG: decentralized curation of graphs, which we can do bc a graph is made of two lists: one list for nodes and one list for edges. Then build on that to achieve DCoS: decentralized curation of standards, which we'll be able to do because things like schemas, verified credentials, etc can be represented as graphs. Then pick one or a handful of w3c standards and see if it is possible manage them using these tools rather than the committees that are today's standard. That still won't get us all the way to the self-sustaining, atomic chain reaction of the decentralized web that I mentioned in a note yesterday. But it will get us a good bit of the way there. DCoP and/or DCoR: decentralized curation of protocols and/or of repositories will have to be conceptualized and achieved at the same time or maybe after DCoS. Then DCoS, DCoP, and/or DCoR can be used to manage the NIPs and other protocols that got us to this point in the first place. At that point, we can start to think about feedback loops: self-sustaining chain reactions of decentralized curation of data in all its forms.

NIP-51 allows users to create a list which effectively imports items from one or more other lists using their naddr identifiers as pointers. Cool feature, I think! I wonder: Are any clients or apps actually utilizing this feature yet? I know listr.lol allows me to add pointers to a list, but I can’t get it to do the actual import, i.e. import the actual items and show me the final composite list. Unless I’m just missing it. If not, anyone planning or thinking about utilizing / building on this feature?

The decentralized web is (will be) like an atomic reaction. And the nostr note is the neutron. We have not yet reached critical mass. But we will.

You know those pics of rubbery-looking blood clots that pathologists have been supposedly pulling out of blood vessels ever since the vaccine? One day my decentralized web of trust will help me zero in on multiple trustworthy sources — pathologists from a variety of settings, ideally — to share their assessment of those photos. How many are seeing the same thing? Have they ever seen something like it before? I’d be willing to tip a few sats for their input. Wouldn’t it be nice to have sourced info on any and every random topic at our fingertips?

I’m happy with my roadmap. Except that it needs to be finished yesterday.

Legacy social media uses advertising to finance the design, implementation and maintenance costs of: 1. platform UX 2. content storage / archive 3. content delivery to end user 4. content curation algos With nostr: FOSS is handling (1). Nostr uses relays for (2) and (3), although unclear how relay monetization will play out. Also unclear how (4) will work with nostr. The ability to make lists is a start. But long run: - Who designs the algos? Does someone need to be paid for this or can they be implemented at no cost via a combo of FOSS + user input that is provided for free (e.g. using lists)? - Who implements / executes the curation algos, the computations & data processing? Will these costs be trivial compared to (2) and (3) or will these costs be significant? My thoughts: once we get into web of trust, which will probably require high quality data + computationally expensive graph analysis, the costs associated with high-quality content curation will be nontrivial. In the long run users should have a mechanism to be compensated for their contributions to content curation. Legacy media expects users to provide things like ratings for free, but the legacy monetization model needs to die. Have mechanisms to pay your trusted users a few sats for their ratings, not just for their “content”. Let your WoT help you decide which users to pay for ratings in any given context. This is how we replace the legacy advertising-based monetization model.

Legacy social media typically don’t tend tend to give users very good access to social graph data. Example: I don’t know of any widely-used tools that would allow me to visualize the Facebook social graph and do things like discover interesting connections between different friend groups that I didn’t know existed. Is it because most users simply aren’t asking for sophisticated data analysis tools like this? Or bc Big Tech doesn’t want us having too much access to their proprietary data?

Anyone read the Three Body Problem by Cixin Liu? The book opens in 1967 in the midst of the cultural revolution in China. If anyone needs a reminder of why bitcoin and why nostr, read the first two pages. Absolutely devastating.