GM everyone.
Kusama is supposed to be a place for experimentation, where we should try new things and build crazy stuff. Some of this crazy stuff might end up being nonsense, it’s not going to work. Some of it, though, is going to be amazing. Experimentation in and of itself is valuable - Polkadot cannot be that experiment, it must be a reliable protocol for it to have its credibility as a leading platform, but Kusama doesn’t need to hold itself to that conservative ideal. - Gavin Wood, Polkadot Decoded 2022

Kusama is supposed to be a protocol with a low barrier to entry for parachain deployment so that early-stage startups can experiment with new ideas.

The GM Parachain team has noticed some inconsistencies with how certain deposit fees on Kusama are handled, and we believe that it is in the best interest of our ecosystem as a whole for some of these fees to be reviewed.

At a high level, we propose to lower the deposit fees associated with the following:

1. registrar(reserve) - Reserve a Para Id on the relay chain.
2. registrar(register) - Register head data and validation code for a reserved Para Id.
3. crowdloan(create) - Create a new crowdloan campaign for a parachain slot with the given lease period range.
4. hrmp(hrmpInitOpenChannel) - Initiate opening a channel from a parachain to [the] given recipient channel.
5. hrmp(hrmpInitAcceptOpenChannel - Accept a pending open channel request from the given sender.

TLDR Version

Kusama is the experimental network of the Polkadot ecosystem. It should be used for innovation and experimentation, and we should not be afraid to push it to its limits nor stifle this innovation through high barriers to entry (i.e. big deposits). This is why we are proposing the Kusama Innovation Cost Reduction Act 2022 (KICRA 2022) - we believe the reduction of the extrinsic deposits outlined below should be an order of magnitude lower than the equivalent deposit for Polkadot.

registrar(reserve)

registrar(reserve) reserves a parachain ID on the relay chain. It is a mandatory extrinsic that must be executed in order to connect properly to the Kusama relay chain.
The current deposit fees associated with issuing this extrinsic are too high for a canary network, and can be prohibitive for new teams who may not be well funded yet want to experiment with a niche idea on Kusama.

We can see that typically the deposit fees associated with this extrinsic on Kusama are ~40 KSM.
For Polkadot, the deposit fees associated with the same registrar(reserve) extrinsic are 100 DOT.

On Kusama, this works out to be:

• ~40 KSM
• $2,344 (30D MA)
• $8,132 (1Y MA)

On Polkadot, this works out to be:

• 100 DOT
• $832 (30D MA)
• $2,184 (1Y MA)

Note: Regarding crypto to fiat conversion, the following figures were used throughout this document - as of 25-Aug-22:

• KSM 30D MA = $58.6
• DOT 30D MA = $8.32
• KSM 1Y MA = $203.3
• DOT 1Y MA = $21.84

Workings:
https://docs.google.com/spreadsheets/d/1uPSd0MWvPSiJTsqMwAwkXlN1daKSoQzYJsTRuzQvxxU/edit?usp=sharing

We at GM Parachain are of the opinion that the deposit fees associated with executing this extrinsic are too high, and should be considerably lower on Kusama than on Polkadot (by an order of magnitude at least), especially due to its sole purpose of being the canary network for Polkadot.

registrar(register)

registrar(register), at a basic level, allows the parachain to upload its runtime and some other information to the relay chain once it has a reserved Para Id from the registrar(reserve) extrinsic (it’s a bit more nuanced than that of course). This is a mandatory extrinsic that must be executed in order to connect properly to the Kusama relay chain.
Like the registrar(reserve) fees above, these are also too high and should be reduced to facilitate greater experimentation for teams with unique use-cases but lack capital.

We can see that typically the deposit fees associated with this extrinsic on Kusama are ~23.81 KSM.
For Polkadot, the deposit fees associated with the same registrar(register) extrinsic are ~537.2 DOT.

On Kusama, this works out to be:

• ~24 KSM
• $1,395 (30D MA)
• $4,841 (1Y MA)

On Polkadot, this works out to be:

• ~537 DOT
• $4,469 (30D MA)
• $11,732 (1Y MA)

Workings:
https://docs.google.com/spreadsheets/d/16AZwJLf7qBoezzAztai787Q3b1akH5HWiJivAs6jX9k/edit?usp=sharing

We at GM Parachain are of the opinion that although the fees are less than the DOT counterpart in this instance, the deposit fees required to enact this extrinsic on Kusama are still too high. However, there should be a lot of care taken when considering how to lower the deposit fees for this particular action (changing constant, instead of changing deposit per byte, for example).

crowdloan(create)

crowdloan(create) effectively opens your crowdloan based on the lease period range that your team selects. It is a mandatory extrinsic which must be executed in order to open a crowdloan for a parachain slot auction.
Much like the previous extrinsics, the high deposit fees associated with executing crowdloan(create) can be prohibitive for a team that does not have significant financial backing.

We can see that typically the deposit fees associated with this extrinsic on Kusama are 10 KSM.
For Polkadot, the deposit fees associated for the same registrar(register) extrinsic are 500 DOT.

On Kusama, this works out to be:

• 10 KSM
• $586 (30D MA)
• $2,033 (1Y MA)

On Polkadot, this works out to be:

• 500 DOT
• $4,160 (30D MA)
• $10,920 (1Y MA)

Workings:https://docs.google.com/spreadsheets/d/17o9uwn4WnqKV1TPUdoWh-cdIU6BTN59eWSjELIV_P78/edit?usp=sharing

We at GM Parachain are of the opinion that, while we note these deposit fees are already lower than the Polkadot versions, there are already a lot of fees associated with setting up a parachain candidate project prior to even opening the crowdloan itself.

hrmp(hrmpInitOpenChannel) / hrmp(hrmpInitAcceptOpenChannel

hrmp(hrmpInitOpenChannel) & hrmp(hrmpInitAcceptOpenChannel) must be initiated from both teams involved in opening these communication channels. These communication channels are one of the current key features of our ecosystem, and demonstrate the interoperability of parachains.

On Kusama, the deposit for each extrinsic is 5 KSM. That means that Project A must deposit a total of 10 KSM, and Project B must also deposit a total of 10 KSM.

On Polkadot, the deposit for each extrinsic is 10 DOT. That means that Project A must deposit a total of 20 DOT and Project B must deposit a total of 20 DOT.

There are a limited number of these channels that can be open, and this number is controlled (and can be changed) by the relay chain. We are of the opinion that on Kusama, we should be actively testing the limits of the network, and be pushing towards opening the maximum number of channels available to test the robustness of the network.

Kusama is the canary in the coal mine; it should be at the vanguard, leading the way and actively testing the limits of our ecosystem. This is how we can ensure the stability of opening additional channels on Polkadot.

On Kusama, this works out to be:

• 20 KSM (both teams combined)
• $1,172 (30D MA)
• $4,066 (1Y MA)

On Polkadot, this works out to be:

• 40 DOT (both teams combined)
• $332.80 (30D MA)
• $873.60 (1Y MA)

We at GM Parachain are of the opinion that, similar to the first instance, we are now in a position in which teams will likely find these deposit fees a financial burden that will only create friction in developing novel ideas. As noted in all four instances, the deposit fees associated with executing this extrinsic on Kusama are too high. These extrinsics should be considerably lower than their Polkadot counterparts (by an order of magnitude at least).
In contrast to the previous three extrinsics however, this extrinsic is required for each additional channel that the parachain opens - and therefore the associated deposits will continue to accumulate as projects become more innovative and outward looking.

Typical Summarization of Deposit Fees

Below shows the typical deposit fees for a parachain team launching on Kusama / Polkadot and also opening 5 sets of HRMP channels (note: the 5 sets of HRMP channels is a purely arbitrary number, it is pretty realistic for the number of channels that a parachain may want to open fairly quickly after deployment onto Kusama / Polkadot):

On Kusama, this works out to be:

• 174 KSM
• $10,196 (30D MA)
• $35,374 (1Y MA)

On Polkadot, this works out to be:

• 1337 DOT
• $11,124 (30D MA)
• $29,200 (1Y MA)

We can see that based on the 30D MA -** it is only ~9.1% more expensive to launch on Polkadot versus on Kusama.**
We can see that based on the 1Y MA - it is actually ~21.1% more expensive to launch on Kusama versus on Polkadot.

We would like to encourage all of the Polkadot / Kusama ecosystem participants to discuss below what you believe the appropriate deposit fees should be for each extrinsic type, and ultimately the total deposit fees associated with setting up a Kusama parachain and opening 5 sets of HRMP channels (as an example).

Summary

Taking into consideration the arguments presented above, along with the fact that Kusama was envisioned to be the canary in the coal mine - the incentivized testnet - we recommend making these changes to unleash the untapped innovative power of Kusama. We need to push Kusama to its limits. We should not be concerned about “reaching the cap” of open HRMP communication channels; instead, we should be actively trying to reach that limit, so that the community can monitor the stability of the network. If it is stable, we could potentially push it even further or see the theoretical maximum number that can be deployed safely on Polkadot, which will complete the interoperability vision. We can facilitate this by lowering the barriers to entry - such as exorbitant deposit fees.

Ultimately, the GM Parachain Team believes Kusama should not be gatekeeping teams from being able to participate in the Kusama ecosystem with over-priced deposits, especially when compared to the more conservative Polkadot that has, at times, significantly lower deposit fees - despite being the enterprise solution for parachain teams.

We would like to make reference to a previous referendum that was passed on Kusama ~10 months ago by the HydraDX / Basilisk team - Snek slithered, so that GM could walk, and we want Kusama to fly. (Link to that discussion: https://kusama.polkassembly.io/post/1056 and the subsequent referendum: https://kusama.polkassembly.io/motion/391)

As a closing thought, here’s a clip of Gavin Wood talking about his vision for Kusama in an interview with our friends at Wagmedia: https://drive.google.com/file/d/1W9_2ZCUTGtI4WOQKGXg_Vblo54OxcdGU/view?usp=sharing

Special thanks to @lolmcshizz, @ThePhunky1, @pixeltrip, @TinkerGabe and @xcBirdo for their input into the KICRA campaign, and also to the rest of the GMorDIE DAO team members.