ASIC are good as long as you don't need hefty amount of RAM, as it is for Argon2. See this paper: https://www.cryptolux.org/images/0/0d/Argon2.pdf
Basically, the point of memory-hard hash function is to make parallelization done by ASIC expensive due to memory usage.
10minutemail/tempmail is what I consider "one-time use". Yeah, one literally uses it for short period of time to verify an email address for example and to not use it in future. However, many such services offer you an ability to open that mailbox later, but with a constraint of deleting letters one hour after receiving them.
Proof of Work can be done in a way that is extremely expensive to scale - see Argon2 or RandomX algorithms. Basically, the idea is to resist ability to offload computation to GPU/FPGA/ASIC. Without them and with a memory-intensive algorithm, cost of an attack grows fast.
I think it is worth to add that no solution will eliminate 100% of spam. Well, you can filter 100% spam only if you are not accepting any event at all, which is not practical approach.
The key is to filter most of the spam and leave only small amount of false-negatives (spam that made it to your feed) in favor of not having false-positives (legitimate event/post being blocked/filtered) at all.
Isn't LN URL deprecated in favor of LN Address?
I am big fan of filter.nostr.wine
It is paid relay, but at least you can see for what you are paying
You can just use algorithm that is not vulnerable to GPU/FPGA/ASIC offload. See Argon2 or RandomX.
10minutemail/tempmail? email alias/simplelogin/anonaddy? There are bunch of services available to accomplish this and similar approaches
Unsure about **best**, but I use Thunderbird on desktop
I like this idea. Huge PoW to register on a relay might make using disposable npubs useless (too expensive) for an adversary (spammer)
While you are completely right, there are algorithms that are harder (more expensive) to offload to ASIC/FPGA - for example, Argon2 or what Monero uses - both are pretty resistant against such offload.
But while normal users post a few posts per day, the spammer does multiple per second? Isn't it different magnitude of required computational power, making spam way more expensive?
Nowadays email is mostly used for registration at websites, a bit less often to receive notifications for them, and (I believe) most often for B2B communications
For throughput more sticks is better, but not any CPU (their memory controller) is able to utilize it. Server CPUs are ones better optimized for parallel memory access
What shenanigans are yoy talking about? If you are referring to their spam filtering, they are pretty responsive at their abuse_rbl email.
I self-host email server to be able to manage it actually, I can't trust a 3rd party with it
I highly recommend you to have both incoming and outgoing liquidity - this means being able to send and receive sats
lightningnetwork.plus or amboss.space
Personally I recommend you to look at https://lightningnetwork.plus/swaps - this will allow you to get in+out liquidity. Triangular swap is good to start with. Also, it is fine to start as low as 100k sats - if you will need to increase it you can open new channels to different nodes you, preferably ones that are not reachable in 1-2 hops from you to enhance your hopness metric.
For my business I am operating this node: https://lightningnetwork.plus/nodes/02f16437c2b9bf9f38bc4a57d7f32d7c36633725392cebf1faadea11b4ea855a7b
I think ratelimit would not help for disposable npubs - I mean ones that are generated for single event. If network is flooded with unique npubs ratelimiting is not really useful