Welcome to This Month in Graph Indexing where we look back at the last month, highlighting the most interesting happenings we have found in the Graph Indexer community.
Network Launch 
The Graph mainnet launched to great fanfare in late December thanks to a mammoth push from the Edge & Node team. Now at the end of January, the network has 1.9B GRT staked and 145 active Indexers that are helping to bootstrap the network in preparation for bringing more subgraphs onto mainnet.
The Graph Network dApp is Live 
The Graph Network dApp, created by Edge & Node, is live here and available for checking the state of the network and to enable Delegators to manage their delegations. The dApp recently received an upgrade to support vesting contracts in the last week of January. Curator functionality coming soon. If you have any feedback on the Graph Network dApp you are welcome to make a post on the feedback section of the forum.
Community-made Indexer/Delegator dashboards 
An impressive number of web dashboards and tools have been created in the short time since the network launched. These tools can help both Indexers and Delegators make decisions about their role in the network and have been created exclusively by community members. Links to these tools can be found on Discord in this pinned message.
Indexing Documentation is live courtesy of @Ford 
Interested in running an Indexer? Check out the new Indexer docs here and if you have any questions, jump in the #indexer-software channel on Discord.
The new Graph Testnet is live 
The permanent Graph testnet is ready to accept new Indexers - this is an excellent way to get familiar with Indexing and make contributions that will be noticed by the community. Check out the announcement here
Delegators are watching Indexer Behavior Closely! 
Now that Delegators have some excellent tools available for monitoring their delegations and the network as a whole, they have picked up on potentially malicious Indexer behavior and made it known to the rest of the community.
By monitoring changes in Indexer cuts using @stakemachine 's dashboard a delegator, with the assistance of @indexer_payne, picked up on a potential Sandwich Attack 
, first described and named by Indexer @ellipfra. A genuine Sandwich Attack is a malicious attempt by an Indexer to pretend they will share the majority of rewards with their Delegators, but then take all the rewards for themselves and hide the fact that they did so. So what does this look like in the real world?
- Indexer has a low reward cut e.g. 20% (Indexer takes ~20% of rewards, Delegators get ~80% of rewards) which attracts new delegations
- At some point the Indexer changes the rewards cut to 100%
- Minutes later the Indexer settles on-chain and takes all pending rewards for themselves, Delegators get nothing
- Indexer quickly sets reward cut back to 20% to look appealing to new Delegators
- Delegators are either totally unaware they are being deceived, or are left wondering where their projected yield has gone
In a truly malicious Sandwich Attack, the above actions are performed in a short period of time in an attempt to hide the malicious activity from Delegators, and 100% of the rewards are taken. In the case that was seen on mainnet, the Indexer moved their cut from 20% to 80%, settled, then set their cut back to 20%. In explaining why, when confronted by their Delegators, the Indexer explained that they use this strategy in order to provide a fixed return to their Delegators. Does this seem like honest Indexing? The answer is up to you.
Highlights from The Graph Forum 
Indexer rewards should be withdrawable to a separate address - link
- Indexers are discussing an important issue on Indexer revenue management. Currently, Indexers cannot extract any revenue from their operation unless they down tools and take their self stake offline for 28 days. This problem is due to be fixed with a smart contract upgrade and the last update on progress was provided here .
Only allow Indexers to become over-delegated by up to X% - link
- Currently it is possible for an Indexer to be delegated beyond their max capacity. When this happens, the rewards that a Delegator will see are diluted by the over-delegation and the over-delegation is not usable in the network (so provides no additional economic security). This is an important mechanism within the protocol, however some argue that there should be a percentage cap on over-delegation in order to minimise the reward dilution and to try and promote delegation to under delegated Indexers. Do you have an opinion on this topic? Make your voice heard in the discussion.
Indexer Performance Workshop - link
- @martintel has proposed an Indexer Performance Workshop as part of workshop planning for the next few months - do you have some ideas of what you would like to see in an Indexer workshop? Get involved in the discussion.
That’s the highlights for the last month. If you have any feedback for the “This Month in…” updates feel free to post below or get in contact with me directly. See you in the next one.
