Ok, i went back and reviewed the previous post and I think I understand it now.
-
At the time, polkassembly had completed the work and based on hours spent, gave a cost of $20K USD for an OpenGov implementation for Moonriver and Moonbase. This was not priced out by network but rather based on number of hours across different aspects and resources according to this chart
-
An additional cost for Moonbeam was not quoted at this time.
-
There was a request to take into account the 80/20 rule for the cost of infrastructure that is for the benefit of all 3 networks. Polkassembly computed this by splitting the cost into 2 x 10K USD amounts, one for each network (moonriver and moonbase) and then including 100% of the cost for Moonriver and 20% of the cost for Moonbase. The other 80% for Moonbase to be requested in a later GLMR proposal along with the implementation cost for Moonbeam.
In reviewing these facts, I guess I understand how we arrived here but I must say that the MR4 request did NOT leave me with the impression that there would be an additional 10K USD ask for Moonbeam - that was never stated and if that was indeed known at the time, it would have been much better to present it at that time than 6 months later.
Moreover, although the chart has a note that “costs for each platform can be considered as 50% of the total”, presumably this is because a lot of the work is equally applicable to any Moonbeam environment. For example - 65 hours were spent on “product features”. As Polkassembly was writing the code for the features at the time and this code was to be deployed for both Moonbase Alpha and Moonriver, I guess it seems reasonable that 50% of the cost be applied to both networks. However, it also would seem reasonable that once this is done, there would be $0 cost for Moonbeam since the features have now been implemented for the general moonbeam stack.
The other activities include “OpenGov Deployment” and “technical migrations”. I find it very difficult to accept that after having completed the work for these two initial networks, it would take an equal (in proportion) amount of time to complete these activities for Moonbeam. So basically, the claim is that there were no lessons learned, tasks that were automated, etc that would make the implementation for Moonbeam go more quickly.
Overall, I would have expected a discount to apply similar to the one you mentioned above related to some of the other features:
“Instead of 3x the cost it is roughly around 1.75x the cost for one individual chain in this case.”
So my back of the napkin logic would look something like this. Take the original chart and:
- Remove product feature development
- Divide all number of hours by 2 since it’s one network and not 2
- Give a 50% discount to the remaining activities since it would be much easier to implement for the 3rd network
To arrive at $3,763 for the Moonbeam implementation (instead of $10K USD).