karlicoss/HPI

Social Media - Aggregate Updates

7flash opened this issue · 3 comments

7flash commented

I want to have daily updates aggregated from all creators I am following in different social networks, including Youtube and Patreon, - how can HPI help me to achieve this?

seanbreckenridge commented

Currently the way this gets youtube data is through google takeout, (can see the parser for it here), which is typically done on a monthly or so basis. That does have a list of your subscriptions, but it doesnt have creator data, its just what you've already watched.

Theres no patreon module here, thats also sort of creator focused.

Both of these could be achieved through new modules, it would likely be some script that runs in the background like one of these exporters:

But those as well, more focus on your own data -- videos you've watched, posts you've made, rather than what all the people youre following are doing

So, to do what you want, would have to create two separate sort of projects, one for youtube, one of patreon, which downloads all of the data from people you're following. All possible, there just hasn't been a desire to do any of it yet, since it feels a bit out of scope of HPI -- HPI helps you organize your own data, not data from people youre following (though sometimes, it does have that data). I may be wrong in my understanding here though, feel free to correct me @karlicoss

Sure it would be added if available, just hasnt been done yet

To do the daily updates it would probably be some sort of python script that batches the updates/you run periodically/put in cron to send an email or something like that

  • 👍1
  • ❤️1
seanbreckenridge commented

Personally I would probably solve this with RSS, but may not be the kind of experience youre looking for

  • 👍1
karlicoss commented

Yup, in addition daily updates to Youtube would be particularly tricky with HPI since Google doesn't really expose nice APIs, and it's not feasible to request a takeout every day. Although once day I certainly hope to scrape it or extract from the browser cache somehow...

Another alternative to RSS could be Fraidycat, I believe it's specifically focusing content aggregation

  • 👀1