The Current Vision for Re-Public

Ideas
#1

I missed last week’s town hall, but if anything that seemed to raise the level of discussion. There was a lot of guessing about what Keith’s Vision is so I’m trying to answer those questions here.

Also, I’d like to encourage people to say “Re-Public should do this …” and not be worried if that aligns with the vision in my head. If we only do the things I think we should, I don’t think we’ll get to where we need to be.

The end goal is to seamlessly enact a person’s will in the dataverse in an automated way. People set and forget preferences and then Re-Public negotiates deals on their behalf and helps them manage their digital assets. A non-obvious component to that is just showing people what they have and help them make sense of it, so included in that vision is a lot of data visualization and extracting of insights.

Like Sumit said in the town hall, people basically now have the need for what enterprises needed 10 years ago with regards to their data. We can imagine what CEOs were asking for, are asking for, in order to get visibility into their business and operations, and that’s what individuals need now for their personal data.

The road to get there is filled with complex and hairy problems. The vision can seem vague of disjointed depending on which area we focus on: Licensing, UX, Data Visualization, Storage, Security, Privacy, Distribution, Compensation, etc.

As a solo-developer I’m always adjusting to working on the thing that I think will bring the highest reward for the minimal amount of effort. I don’t always get it right … maybe I never do, there’s no way to know.

If you ask me today what I think we should be working on from a product perspective, I would say some sort of compelling data visualization that brings in two silo’d datasets and puts them together. Like folks were talking about on the town hall, Re-Public is in a unique position to take something like Strava data and Runkeeper data and show some new insights.

What do you all think? The more discussion we have, the further we get, faster.

Some focusing statements for Re-Public:

A Dashboard for Your Dataverse

A Global Data Utility

1 Like
#2

I fully agree with this vision.

2 Likes
#3

Agree with this as well - but this means writing connectors/scrapers to pull down a user’s data from various services. That’s very different from what Re-public is currently doing. However, I wonder if building these connectors can be a community activity. When a developers templates are used to pull down data, he/she earns from $Data or other incentive.

2 Likes
#4

Yeah, I like the incentive/bounty approach. Hadn’t thought of it until you suggested. Although I’d like to do a few myself to see if there’s anything that can be standardized and offered to devs to make it easier. At least docs on what Re-Public needs as far as formats and auth flows, etc.

We had actually identified Strava as a first target until you pointed out what a pain it was :sweat_smile:

1 Like
#5

Long time lurker, first time commenter… and former Zapier employee / API guy.

There are several public attempts at capturing API use cases, which might help you with standardizing an approach for data gathering. Most recently Pipedream created repos for its connectors. There are others, but it’s been awhile since I’ve researched the topic.

3 Likes
#6

Keith, picking which dataset to download is not easy. I am sure there are lots of ways to think about how to select the right dataset. If I had to pick three criteria for where to focus precious resources in terms of downloading a dataset, I’d offer up:

  1. greatest number of users have the target datasource
  2. dataset is available in a ready-to-consume format
  3. path to creating value for users is relatively straightforward (either functional or financial value)
2 Likes
#7

If I run that set of three criteria against a bunch of target datasets, the site I come up with is Amazon.com.

  1. % overlap between [people who might download re-public] and [people who use amazon] is likely very high
  2. amazon dataset is readily available here: Amazon.com (very easy to instrument this assuming we have a browser control or some other mechanism to access the site)
  3. creating monetary value from these datasets should be pretty easy (amazon purchase history is a gold mine)
  4. Functional value is an open question. Would users care about their visualizing their Amazon purchase history? They could see their most expensive purchase, oldest purchase, most frequent purchase, etc. This might be more of a toy than something that creates enduring value (per what Emre is saying)
2 Likes
#8

I love this! Sounds like a future in which I would like to live.

1 Like
#9

Personally, I would love visualization and insights about my productivity. Especially now that I’m working from home. How does my sleep, exercise, time on social media, brakes, etc impact my productivity? What do I do on the days I’m the most productive?

2 Likes
#10

Sounds very helpful! I can imagine how to measure the “inputs” like time spent reading hacker news, time spent walking or moving, etc etc. How would you envision measuring the “output” which is productivity. Would that be a user-submitted rating at the end of each day? Or inferred from things like usage of certain apps like email or docs vs twitter and facebook?

Very cool idea!

1 Like
#11

Very helpful, thank you! Wonderful surprise to see you chime in here :slight_smile:

#12

I’ve been thinking about how to make Amazon data interesting, especially with a map. Kind of a tangent … I wonder if Fedex, UPS, USPS delivery routes are publicly available anywhere. Might be interesting to see routes around your house … somehow cross reference that with your delivery times … not sure what we’d learn from that.

#13

I love this use case but I might suggest putting it off until we get our Desktop app out, since that will make it easier to give you a sense of your day with both your phone and your laptop/desktop.

#14

I think we should brainstorm some use cases which require just 1 source of data, or at the very most, two. That’s probably complex enough for the first version.

1 Like
#15

Thanks for sharing

awesome :slight_smile: also really enjoyed reading the replies and brainstorming thats going on :smiley:

1 Like
#16

same here :slight_smile: maybe this is something we could be building with Swash data?

1 Like
#17

Actually, what I’m envisioning is very similar to Zapier, but instead of APIs, its a user’s types of data. Like their specific Twitter activity or Spotify activity. And instead of being made for developers, it’s made for normal people who just want to mix and match data to see what they can learn about themselves and their lives.

1 Like
#18

Love the zapier idea.

1 Like
#19

What data does Swash have? I haven’t had a chance to try Swash yet. Would be curious to learn more.

1 Like
#20

they collect browsing history i believe every 2 mins a snapshot is taken and streamed to their DU, im not sure if we could access that data for free though or if there would be an integration for their members to enable it without having to pay for their data

1 Like