Steve Snyder transcribed 0.6K bytes:
Looking at the Tor Metrics page, I can see the number of bridges and the number of users connecting via bridges, but that's not enough information to determine satisfaction of demand.
Are there now enough bridges to comfortably satisfy demand? Enough bridges with a particular PT type? If not, what kind of resources are lacking?
How does one determine where the need for more bridges, or PT type, is greatest? (Assuming that there is any unmet need at all.)
Hey Steve,
We're currently lacking several important bridge metrics, such as:
* bridge bandwidth measurements and usage, * number of bridges per PT type, * number of clients per country per PT that a bridge sees, * number of requests to BridgeDB for a specific PT per time period, * number and type of bridges handed out by BridgeDB, etc.
As such, it's currently a bit difficult to tell when a very particular need is not being met. :(
However, in general, we have a lot of bridges which have no PTs at all. In fact, currently only about 25% of the bridges in BridgeDB offer PTs. So, one easy(ish) answer to your question is that we always need more stable bridges running the latest-and-greatest PTs (currently: obfs4, obfs3).
We also often need bridges with PTs, extremely high bandwidth, and large allowed network transfer to act as the default bridges in Tor Browser.