I want to shed some light on a topic that seems to me a bit overlooked at the moment. We are now live with the new Giveth DApp for quite some time. During all this time we also kept Giveth TRACE around and it still works surprisingly well, considering that it is still using the Rinkeby network that technically belongs to Ethereums past.
Experience shows us that a system like this will not run indefinitely without maintenance and of course during its runtime it will accrue costs for infrastructure - billing is a bit hard to isolate on Azure where most of the infrastructure is hosted, but the estimate is that running TRACE costs the DAO ~300 USD on Azure for production and ~45 USD on Digitalocean for dev environments per month.
In the past year we had at least one incident where TRACE was negatively impacting the main DApp due to shared resources (which was solved consequently).
If something breaks - depending on severity - we might not be able to fix it in time (use dev resources for other more important things) or not at all (afaik Amin and to a good deal Mohammad are the only ones proficient enough to fix i.e. a problem with feathers).
My hunch is that TRACE makes no problems, because it has no users. The main purpose now (as always has been) seems to be an onchain payment processor (which is a different product than a donation platform).
Because we have to justify using resources we started to debate this within devops and with some developers, but I would like to hear from the larger community what others are thinking about this.
As a starter I see following options are on the table, but of course there are many ways we can go about this:
we change nothing, leave everything on and see how long it keeps running with normal maintenance
we reduce it to the bare minimum and delete some dev environments
we develop it further, make sure it does not break with chain upgrades
we develop it into something different, like that payment processor and try to make a service out of it
I think it might still be being used for commons stack payments or something similar. We should definitely have a look at who is still using Giveth TRACE & involve them in the discussion of it’s retirement.
I think moving towards retirement is a good idea though. When Rinkeby is deprecated (which I think will happen in less than a year), it will not be possible to use anyway.
I’d like to explore this option further. Trying to understand what makes TRACE better than any competitors out there, if such exists, and what we could potentially build on top of it, rebrand it, package it and offer to the market.
While it could be developed by some creative miracle into a payment handler, it is draining resources for basically no benefit to Giveth nor does it bring any revenue. It would take a lot of bandwidth and debating to take a product that was already on it’s last legs in 2021 but revived with no real purpose or goal to then revamp it again, still with no clear roadmap or promotion.
If it’s only benefit is to provide a payment system for the Commons Stack or other Orgs then maybe it’s time to pass the bill to those organizations rather than drain our own resources for no real benefit.
It’s a historical anecdote and was novel for it’s time but with gnosis chain and cheap gas, the rinkeby bridge is not really a necessary feature anymore.
Ironically, it was mainnet gas fees that led to TRACE being built on Rinkeby, and it was mainnet gas fees rising again that led to users abandoning - we imposed a minimum amount on withdrawals that made small donation collection impossible.
I’m a staunch believer in TRACE because of the robust feature set that Giveth was once upon a time built upon. Since launching Giveth.io with the greatly reduced functionality, the direction has gone towards building out the economy instead of advancing functionality and that’s great too - we’ve differentiated ourselves through capitalizing on current trends and decentralized finance development.
That said, maintaining two different products does not make sense for Giveth, now or even in the future really, even when the economy turns around and we could fund further development. There are so many collaborators that have complementary products to Giveth such as DoinGud… and soon Gaia.Gives will finally be launching this year. They have a new CTO and backers making it happen and it’s built on the original designs of TRACE with all the functionality it had plus a lot of what Giveth has advanced.
I see collaborators not competitors, and Giveth will be well served to gracefully twilight TRACE and focus on our strengths. It will need to be a coordinated effort that allows past users to collect remaining funds - there is a comms opportunity in here to put Giveth in a spotlight while we run a consolidation and closedown. Users and interested parties of old (that may still be holding GIV or never collected their airdrop) can be advised (and incentivized) to move their project and funding to a new project on Giveth or consider other partners in the space.
Vote is looking quite in favour of sunsetting Giveth TRACE & I’m on board.
Next steps I think we should take (after talking to @griff & @amin):
Stop the ability to accept more donations by removing the donating button from the UI
@amin & @freshelle to look at which projects have money & where it is allocated to them - it’s possible it is mainly the commons stack
pending info from 2, if there are other projects with unclaimed funds, we should start a comms campaign explaining that they should claim their funds by x date (TBD). Any unclaimed funds after that date will be withdrawn from the bridge and sent to the Giveth Matching Pool.
Fortunately, GivethBridge (deployed on Mainnet) and ForeignGivethBridge (deployed on Rinkeby) are both pausable contracts. When we put each one on pause, on
GivethBridge: Users cannot donate anymore on the Mainnet network, and also we cannot transfer money withdrawn on Rinkeby too. It means no money can go in or out of the vault, except through an escape hatch process that can be run by administrators.
ForeignGivethBridge: Users cannot withdraw money.
Note: We can pause and unpause a pausable contract
I believe we must pause contracts alongside removing donate button from the UI
It would be nice if users will still able to withdraw money, i.e. claim funds donated to them. It sounds like if we pause GivethBridge, the users won’t be able to do that… am I right? @amin
I think it would be nice if we start a comms campaign to tell people “withdraw your funds now or else it will go to the giveth matching pool”
Yeah, they cannot claim their fund. As the withdrawal process consists of two transactions, on one Foreign Bridge (Rinkeby) and one on the Home Bridge (Mainnet), we must pause both bridges almost simultaneously. Otherwise, if we pause only one, withdraw or donate actions makes imbalance between two sides of the bridge.
@amin shared this report showing the balances as of October 24, 2022. It’s categorized according to type (trace, campaign, community). MAJOR PRAISE TO AMIN FOR THIS! I’m just sharing what he sent me.
We’re in the process of shutting down the resources that keep trace alive. The reason we are doing it is to save a lot of infrastructure costs on our obsolete applications.
But before we do that I want to share with you the plan we are following in this process: