I'm currently using @Minibits mint through @Keychat , and I'm reaching out with a question. I'm personally developing a service that requires payment via hold invoice, and during development, I was testing the expiration/cancellation behavior by paying a hold invoice using the Keychat Cashu wallet. The payment attempt initially seemed to fail—although the payment did reach the recipient side, it remained in a "pending" state since it wasn't settled, which is expected. I didn't mind the unclear UI indication, since most wallets don't clearly display pending states anyway. However, the next day, after the invoice expired as intended and the payment properly failed, I noticed that the deducted balance in my Keychat wallet wasn't restored. Normally, when a hold invoice is canceled, the mint's Lightning node should recover the funds and reflect that credit back to the user's balance. If this were my own Lightning node, the balance recovery would have happened automatically. But with the Cashu wallet, since I had already paid ecash to the mint to initiate the payment, my balance was deducted up front. From the mint’s perspective, once the ecash was spent, the balance was marked as spent—even though settlement never occurred. And since the control had already been effectively transferred, the mint likely considered the payment as completed on its end. It doesn’t appear to track the subsequent cancellation status. Moreover, due to the anonymous nature of ecash, the mint likely can't identify who made the payment—even if they wanted to issue a refund, it might not be possible. I can provide proof of the hold invoice and confirmation that it was ultimately canceled (via DM). Surely a refund shouldn’t be impossible? I understand it might not be easy to verify, and while I understand if no refund can be issued (it's only 5,028 sats, not a large amount), this highlights a bigger issue: going forward, such payments need to be properly handled. Otherwise, it effectively means users permanently lose funds to the mint even when payments fail. This isn't just an issue with Minibits, but likely all #cashu mints. A proper mechanism needs to be developed to handle these cases.

There's a feature in Keychat wallet where sending "pay ecash" returns an ecash token, but sometimes it comes with a QR code and sometimes it doesn't. What's the difference? I'd like to provide a smooth experience where users can simply pass along a QR code, but when it only returns a token, I have to send it via DM, which is a bit inconvenient. #asknostr @Keychat

Sometimes, while using a VPN, Nostr-related apps occasionally fail to work properly. Since they always start working again as soon as I turn the VPN off, it's nearly certain that the VPN is the apparent cause. I initially thought the VPN might be blocking WebSocket connections to relays, but since the apps often work perfectly fine even when the VPN is on, that doesn't seem to fully explain it. What's more puzzling is that @Amethyst always works without any issues, regardless of whether the VPN is active or not. I can't figure out the reason for this inconsistency.

I believe providing economic incentives to relay operators is essential for the nostr ecosystem, and among the possible approaches, attaching e-cash when publishing events seems optimal to me. I’ve been skeptical of subscription-based payment models, and more importantly, with standards like NIP-17, users send their DMs not from their main pubkey but a temporary one—making subscription models fundamentally incompatible. Recently, I noticed increasing spam accumulating on my personal relay, which became quite unpleasant. I thought there must already be a NIP addressing this, so I went looking—but found nothing. Hmm? I decided to try implementing a solution myself, even if just according to my own idea. But immediately after starting, I understood why such a standard hasn’t emerged yet. It turned out to be too complicated. When publishing an event, devs familiar with the nostr ecosystem typically use an outbox model, publishing across different sets of relays dynamically. However, most clients don’t publish to individual relays one by one, but broadcast to all target relays via nostr-tools(or something like that). This means if we want to attach e-cash to a *specific relay only*, implementation complexity spikes dramatically. Moreover, what happens if the relay’s price changes? How should the relay behave if it receives the same event again (e.g., due to a re-synchronization or backfill)? Should already-sent e-cash be refunded, and if so, how? And if refunded, must that e-cash be redeemable again? When you dig into all these edge cases one by one, the feasibility of such a system becomes questionable. The most practical implementation so far might be keychat approach, but even that was designed for a messaging app rather than a social network—and still doesn’t integrate well with the idea of an inbox-relay model for DMs.

Being at a resort with a poor internet connection has made the performance differences between Nostr clients strikingly obvious. nosotros seems to be the best. Of course, this might require a deduction based on how well it adheres to the Outbox model, so a simple comparison based on initial loading speed alone might not be a fair assessment. ​ Originally, the top performer in this area was @Amethyst but it slowed down a bit after the Outbox model update a few months ago. ​In reality, we can't really know how perfectly nostr clients construct the feed using the Outbox model. Since most people designate at least one large, popular relay as their outbox relay, it would be difficult to notice if a client doesn't actually support the Outbox model. ​As a side note, @Damus android was the only one that worked even when the internet connection was completely severed. This must be due to it fetching the feed from a local database rather than a relay, building the feed, and continuing to accumulate data in the local DB in the background. It's fast, too.