Data is key.
We discuss the idea of network effects are great deal. The reason is, when building a digital platform, network science is at the core. We are dealing with a system that following some laws that appear to be natural. Researchers have seen similarities to other networks including biological, digital, and economic.
When looking at LEO, it becomes clear to the focus should be.
Of course, the easy answer is more users. This is usually the extend that most go to.
in this article we will explore how this all works and offer some ideas of what we can do to kickstart things.
Why so much attention on data?
Basically, it is the key to the last 30 years. Everything we do online is tied to data.
That said, it is something that few of us ever really considered. Sure, we all are aware of the phrase that "data is the new oil". It is evident that data has value based upon the revenues from companies like Google and Facebook.
There is a method to what takes place. In fact, it is rather simple, at least conceptually. It is from here that everyone involved in Leo can make an impact.
Here is a diagram of how the feedback loop for digital platforms works:
This should not be a major surprise to most reading these words. The problem is everyone starts at the top with the more users. What gets overlooked is the fact this is something that can be entered from any point.
What this means is that we can move a bit further down the scale and concentrate on the data. We get the same results simply be creating more data.
Therefore, instead of trying to bring people to an inferior product, we can build more data to feed the algorithms, leading to a better product. When users are presented with this, they will be more apt to join.
It is also something that is not dependent upon external factors. It simply can be handled by those involved already.
Over the last few months the conversation was around "stickiness". It is essential to fill a database with data that can be reused and doesn't simply appeal "in the moment". This is the reason why so many initiatives were undertaken. Music on Leo is one such example.
Here is an example using Threads for songs.
As we can see, there is the embedded song along with a poll. Now this is something interactive.
All these songs are captured on the singles page for the artist, in this case Foreigner.
From here, the songs were placed into a playlist, with the threads (songs) embedded in there.
Playlists are placed on a page of with the rest created that month.
Of course, we can always simply rethread the song creating another version of it.
The point is we are laying the data. What starts small grows as we place more on top. The monthly playlist pages, as an example, will end up feeding into a page listing all playlists for 2024.
Going in the other direction, the individuals songs are being used by different people in the Party Room that is set up each day by 
We were told about LeoAI. For NOW, the specifics are a bit scarce so we do not know how that will work exactly.
One guess is that we are dealing with some LLM, most likely open source. From there, it is forked, providing the basics of what is contained in the model. The data generated on Leo is being fed in, to contour it to Leo.
Once again, we see data. Look at the image at the start of the article. If we start with data as the entry point, we can see how it feeds to algos (in the LLM), resulting in a better product. Expanding this enough could move us to the point where Leo has a product to market.
Hence, whatever is brought in to the database is utilized. In addition to creating data, we can simply pull what is out there.
One example is what 
Here is the summary for an episode of the Lions Den.
This is about a 500 word summary of the show. It covers some of the highlights, placing them in Leo's database. Even though this took place on Spaces, it provides value to Leo.
In scrolling down, what do I see at the bottom?
The recommend algo is offering up another show that was summarized in the same manner. Now the site is starting to deliver more data through the feeding of the algorithms.
Again, look at the first image. We can see how this design is being played out on Leo. This is no surprise since this is what digital platforms do.
It is also why it is imperative that everyone involved feed the database as much as possible. We have all heard the talk of getting more users. How has that worked? For 6 years, on Hive, we hear the same thing. Yet, all the while, how many offered up the suggestion of more data?
For most applications, it means nothing since the development isn't in that direction. Leo is different since there is a team that is focusing upon these processes. We have a recommend algorithm, even if archaic. We know there is some type of AI being developed. The UI is being worked on in an effort to remove bugs while also adding more features.
Then we have a number of monetization paths that are feeding the tokenomics.
If data is the new oil, how come we aren't focusing upon it?
It is time we changed that.
This starts with creating data and then repurposing it.
