Experiment MILESTONE 3

Continuing the discussion from A new experiment: Breaking The Blockchain Trilemma in preparation for the GCP:

THIRD MILESTONE REACHED!

TLDR - the blockchain trilemma has not been broken, BUT we’ve identified a path to do it, and we’ve brought a new star into the Arbitrum gaming constellation.

OK So - work is wrapping up on an intense, informative and rewarding experiment with the Thrive team.

For the sake of full transparency - we ran into some issues. Certain assumptions I made about core components were wrong, certain parameters were not reached.

But to quote Rocky - “It ain’t about how hard you hit. It’s about how hard you can get hit and keep moving forward. How much you can take and keep moving forward.

Rocky-001

So with that said, let me run you through what happened in this experiment.
I’ll start with what went wrong, how we pivoted, and finally, what went right.

What went wrong

To recap, our experiment involved using our software, HoloNET, to port a game project through to Holochain, instead of AWS.

After integrating Stellar Gate with HoloNET, we hit a block:

AWS has a feature called SignalR, a library for ASP.NET that enables real-time web functionality in applications, allowing server-side code to push content to connected clients instantly. In the context of AWS SignalR facilitates real-time communication in web applications hosted on AWS infrastructure.

Holochain also utilises Signals - but, due to the difference in Holochain’s distributed architecture we found it is not yet sufficiently developed for real time web functionality. Consequently our ‘experiment’ phase of the grant produced a negative result.

How we pivoted

At this point, we asked ourselves three questions:

1 - What can we do NOW, given the existing restrictions?
2- What can we do NEXT, to remove the restrictions?
3- How can we ensure that no matter what, this experiment is a win for Arbitrum as a whole?

1 - What can we do NOW

We built a HoloNET API to load + save JSON data to Holochain, easily

Why:

Just as photographs came before film - while we can’t currently provide ‘real time functionality’, we can take a snapshot, and build from there. This API can be called at the end of the game, to ‘save’ game state data from AWS, and load it at the beginning of the next in a cheap, secure way. Additionally, this API abstracts away the complexities of Holochain, creating a highly accessible API that any developer can call with just a single line of code. We are aiming to have this demonstrated by Battle 4, which is on October 4th.

2- What can we do NEXT

We mapped out how to develop a working Signals instance for Holochain. This will be called HyperNET, an upgraded version of HoloNET.

Why:

This feature will achieve the positive result we were looking for in the experiment, as well as potential benefits such as infinite players.

Creating a docker to port OASIS / HoloNET through to Threefold

Why:

Holochain is just an architecture, it is not a storage host in its own right. Threefold offers a decentralised, distributed node network which we can host our software on - which, when functional, will offer an incredibly powerful alternative to AWS. Because OASIS is an abstraction layer across the entire open web, attaching it to a node network will give developers an easy way to develop and host powerful, scalable cross-chain dApps.

3- How can we ensure that no matter what, this experiment is a win for Arbitrum as a whole?

While we are very grateful for Arbitrum and Thrive letting us conduct this experiment, we were mindful that there are certain on-chain metrics which we also wanted to ensure we delivered on, to help grow the ecosystem as a whole. So while David focused on the pivots, I shifted my focus to working with Jorge from Stellar Gate, to build out the game world and make the battle series come off as a success. This part, we feel, certainly “went right”.

What went right:

Without bigging ourselves up too much here, a huge amount of work went into this section, and we are proud of what we achieved in a short window of time with limited resources.

For a detailed look at the work that we did, please read this document:

But to give a snapshot of what we did:

  • Stellar Gate brought across from Phantasma to Arbitrum
  • 25,000+ impressions on X
  • 58 Unique wallet registrations
  • 7 Smart Contracts written
  • TAZZ token launched on Arbitrum
  • 3 multiplayer battles hosted, with a final this coming Friday (October 4th)

Assets

  • Trailer film created using runway, 50+ new visual assets created
  • Military promotions system automated

Closing remarks.

It’s been said many times before, but building in web3 is like the Wild West.
In amid the gunslingers and card sharks, there are also those of us engaged in the serious task of building a new world. A lot of this territory is uncharted, there are dead ends and blocks aplenty. While we didn’t reach the goal we were looking for in our experiment this time round, we did get a lot of learnings which will make us better equipped next time around. Crucially for myself, the idea that this is a shared journey, with a shared destination in mind, is one which matters - while alone we will quickly get lost, together we will make it. To that end, many thanks again to the Thrive team who gave us the chance to explore.

ONWARDS!

Experiment MILESTONE 4 -

The final milestone in this series is a compiled report containing findings from this experiment.

It can be found here

As many of the technical findings were already explained in forum post 3, this post will focus on our experience of launching on Arbitrum.

In some ways this process has felt like the first day in a new school - exciting and a little scary at times. While we did not do see numbers this time around, the groundwork has been done, and we have a plan for rapidly ramping up metrics moving forwards.

Charts

Clear differences in performance levels were recorded, showing that skill matters in the game, as well as tactics (deploying artillery, emeralds, scans)

Actions per player increased as the series progressed - showing that the players got better at the game with time.

Player numbers ranged from 5-10. This is a low number but those players are very engaged, making adjustments to their schedules to show up and play.

Findings

These are our own personal take-outs from conducting this experiment. Some of them may be specific to us, but if others reading this post find them helpful then that’s a plus.

1. Rolling events

We opted to run 4 battles, thinking that having just 4 would increase focus. However, it would be better to run ‘hourly’ battles over an extended period. This will increase the server costs to give a better dataset, as well as bring more opportunities for network effects and ecosystem growth.

2. We struggled to attract an Arbitrum community

No fast way to plug into the ARB community was identified. As newcomers to the Arbitrum ecosystem without a budget for KOLs or gaming guilds, we were a little isolated and feel that either A) we need to do more research to find the resources out there or B) there could be more of a dedicated ARB gaming community builder, especially in the runup to GCP.

3. More time

More time would have been fantastic on this assignment, especially during an experiment.
We worked hard to end the 4 battles by October 4th. But more time would have allowed us to increase network effects, bringing in more players, and stand a better chance of fixing the obstacles posed by working with Holochain.

4. The biggest pain point was wallet registrations, especially for non web3 natives.

The gameplay experience was straightforward, but the steps of creating a wallet and crediting it with ARB ETH was a major drop-off point for newcomers. A lot of time was spent sending small amounts of ARB ETH to newcomers - begging the question, would a pre-loaded gaming wallet be a good idea in the runup to the GCP, to onboard web2?

5. A ‘swarm’ type experience with other game projects would be beneficial.

Assuming that other projects experience the same problem, it would be helpful to team up with other game projects for interoperable, cross-marketing or cross-incentive opportunities, while remaining loyal to Arbitrum. To that end, the creation of a ‘Game Season’ type of event would be excellent.

6. More can be done with wallet addresses

Post-match game data is able to identify the best players. In a ‘Game Season’ type of event, you could create a ‘single-player’ mode where the best performing wallet receives a ‘winner takes all’ prize pot. Marketing opportunities can be found in spotlighting player talent - for instance, using the ‘Admiral Broadcast’ format developed during our experiment.

Conclusion

We’d like to thank Arbitrum for the chance to conduct this experiment. While it’s early days our ambition is in the stars with this project, and we are looking forward to more wins on Arbitrum.

Thank you!

Max Gershfield