Have you seen this post?
Please don't give TNBTC to people who are just hoovering it up to paid by defi scams based on it.
--snip--
Your calculation on the cost of mining Testnet
0.001 BTC/tBTC shows the system isn't working anymore. I don't get it: why would anyone spend that much money on mining tBTC anyway, increasing the hash rate to rediculous heights?
I have abuot half a tBTC (from various faucets), and I've used it to play around with LN channels on
Eclair Mobile Testnet. I haven't used it in a while but still keep it anyway.
I can honestly say that, no, i didn't see gmaxwell's post (until you quoted it). What he says makes a lot of sense... On the other hand, I did get a couple of requests with a reasonable explanation as to why the person requesting a lot of tBTC would need them...
For example, if you're coding a complete exchange or a mixer: i know you can also do your tests with altcoins, or with very small values, but I do understand why you'd want to start with 10 tBTC when you want to simulate a couple hundred mixes and analyse taint afterwards.
This is why my loan thread requires a decent explanation: if somebody wants to borrow tBTC but his/her reason is a bit flaky, i"ll probably just deny the application... This isn't 100% foolproof, but it's better than nothing i guess.
As for the second issue discussed in your post: I couldn't agree more... Testnet is broken (IMHO). Testnet should be about allowing developers to test their application(s) and to allow endusers to experminent with new wallets or technology (like the LN). In my honest opinion, everybody with a half-decent CPU should be able to mine at least 12,5 tBTC within minutes by following a walktrough that's simple enough for a normal, tech-savvy person. If this can't be done, the testnet is missing it's purpose (IMHO).
I would be in favour not only to reset the testnet, but also change it's parameters (for example, remove the difficulty retarget from the testnet completely... Set the difficulty very low and keep it there). Also, I'd be in favour of setting a fixed coinbase reward without deflation: Always 12,5 tBTC/block. This would probably de-incentivise people pushing the hashrate so incredibly high on the testnet: somebody doing this would be stuck with billions of tBTC, i'm pretty sure those DEFI scams would not be able to cope with this. The biggest downside would be the fact that the testnet size could grow quite fast if somebody did push a latest gen ASIC on the testnet, but this could be solved by resetting the testnet more often. I know there are other downsides to, like the fact that if these changes would be implemented, the testnet would no longer behave the same as the main net... But let's be honest, as it currently stands, the testnet is becoming unusable anyways...
Last point: yeah, i also think that it's ok if people hold on to some tBTC for testing... I would never pass out my last tsat either