GITBOOK-132: No subject

Author: 0xngmi

analysts/data-definitions.md

@@ -89,3 +89,7 @@ Number of unique github accounts that made at least one commit to a repository i
 ### Developers
 
 Number of unique github accounts that made at least one commit on three distinct months to a repository in the github organization of a given project.
+
+### Assets (for CEXs)
+
+This includes all the assets that the CEX has under their custody, excluding any assets that are under a different custodian but are credited inside the CEX for use as collateral.

faqs/frequently-asked-questions.md

@@ -37,3 +37,7 @@ description: Answers to the most Frequently Asked Questions
     * Yes. Our API is an open API and it free to use. Citing DefiLlama as the source is much appreciated. Find the API docs [here](https://defillama.com/docs/api).
 13. **How can I download DefiLlama data?**
     * DefiLlama data is available to download in csv format. You can find a “download” .csv button on various areas of the website: protocol page, chains page, overview and the bottom of the side menu bar. The amount of data retrieved depends on where the button is used, using it on the protocol page retrieves data for the specific protocol, using it on the chains page retrieves info by chain, the others retrieve all TVL data on the website.
+14. **Why don't we include assets in other custodian under a CEX's assets?**
+    * Our users use our CEX asset metrics in order to see how much the market trusts a CEX with their capital, which is used as a proxy for how safe a CEX is to leave their money in. Deposits in a different custodian are completely independent from this and don't provide that safety market signal.
+    * If we included custodians it'd open the door for our metric to be abused, for example by hypothetical CEX with 100m and 2bn in custodian vaults, displaying their assets as 2.1bn would give the signal that 2.1bn trust them with their money, when actually the opposite is true and 95% of their users trust them so little to keep the custody safe that theyre paying extra to have the custody segregated. If that CEX got hacked or had some issue and 95m was withdrawn, our metrics would show a 4.5% outflow, giving the signal that outflow is minimal and there's no danger of a run on the bank, when 95% have been withdrawn and actually the CEX is on the late stages of a bank run.
+    * TLDR: by including custodian assets we'd end up giving the opposite signals for our users, and our goal is to provide data that helps in their decisions, not the opposite.