Anyone have a simple getting started using NIP-07 for window.nostr JavaScript web apps?

Literally a basic, check extension exists and is working, log pubkey to console (after consent), then sign a generic hello Nostr kind 1 event and return the signed json object / string.

The websocket stuff isn’t needed. Just the async JavaScript plugin interactions.

Your latest metadata event no longer has the deleted=true. It’s pushed to 75+ relays now.

Unsure exactly which Damus checks out of the box, but if still issues I’d remove the app entirely and try again. https://api.nostrgraph.net/beta/events/7507456bd9aa32e28ccef7d84917c499b05e159628e3c65583a78778299903df.json?pretty=true

I have 1,600 events back from 2022-12-25 21:25:00.000 +0100 which seems to be your first profile update. Can dump an export if you like.

*bug

I just broadcast this contact list (from 2023-03-02 19:49:48.000 +0100 following 290) to around 30 relays - most had it - which is the newest one I found.

Incidentally I found a long running but that had been on the loose - I was dropping events based on payload size - a size I set months ago before Nostr exploded and I’ve been dropping large events like you latest contact list.

You just need a new metadata event without deleted=true and broadcast that, and you're good. That only impacts Damus (as far as I know), and other clients just ignore it.. so your account looks fine.. but likely just Damus login is broken.

https://api.nostrgraph.net/beta/events/713ebaf0c3ee457108dd23e0a16d5453d40beb486f1036943296ea1682bc79b1.json?pretty=true

Here is likely the newest contact list you had. From march 2. I don’t have a copy on my servers.

If you just delete that metadata deleted=true (and publish that event widely) and use the same relays as before, then I think you’re good.

https://www.nostr.guru/e/713ebaf0c3ee457108dd23e0a16d5453d40beb486f1036943296ea1682bc79b1

Yes. The photo above is sorted defending by created at. There are 9 after the initial deletion.

People who follow you haven’t been dropped. It’s just relay selection causing data gaps.

Your latest contact list, that I have, was from 21/12/2022. It was only posted to Damus’ relay - so it likely never replicated well (unless you have a newer one since)

Here is that latest contact list I have for you. Again, I can’t see any deletion events (kind 5).. so it just wasn’t broadcast well.

https://api.nostrgraph.net/beta/events/03f32f7a904c673257afd3453e8ff5acfdbf22c00de92110af4b8fde61861631.json?pretty=true

You both are the largest aggregation relays by volume atm. After fixing spam, you’re next headache is all the bloat events from spammers not of kind 1/42 😄

Nostr is the modern day Wild West atm.

Awesome. I noticed way more spam coming in.

And here is the event before the first deletion event, from 2023-02-14 01:38:56.000 +0100.

https://api.nostrgraph.net/beta/events/aafeba7ec0c94562533661af130cf92c212bd2b145b37f39fbb5f933e7ad113f.json?pretty=true

Oh. Here is what you’re after.

Here is your deletion event and all relays that have seen it (that I know about).

And your second issue is your latest metadata events still have a JSON deleted=true in them. I’d bet that needs to be removed.

https://api.nostrgraph.net/beta/events/7c379ff370607982be284d9d263e0de734392ff2d5ede6d133521a727f7496a6.json?pretty=true

Just swap out the null event id in the url. Shows seen by relays.

https://api.nostrgraph.net/beta/events/fafe2af8f2df810fb36b58cf0620b88488750d8b40bfd45a67ffff0e0f37a498.json?pretty=true

I made this concept blog. Single static page, all cdn JS imports. I’d change the blog page identifier to include a target relay/s (with pubkey + 30k kind). It loads from profile/meta for sidebar content. Supports markdown.

Easy enough to add an editor using window.nostr. You’d still want to trust the static file host perhaps.

Is this the direction you’re thinking?

https://cdn.nostrgraph.net/public/blog.html#d5d58a7473a2a719a0fb41aa7fe68b7e396ef0a881d8fb4eaafc6e0d96e743a9:11001

This impacts client apps validating in real time as well. All previous ZAPs wouldn’t pass validation and only newer ones would validate successfully.

Unless I’m confusing it, one major downside of the current ZAPs NIP is that if I change my LNURL (lud06), it prevents unprocessed (or future processed) ZAPs from being validated.

How? Well, part of validating a ZAP is decoding your LNURL and doing a get request to look for the nostrPubkey JSON property. The sender of the 9735 event MUST match.

If you change your LNURL (like change wallets), that pubkey now won’t match.

I’m using Redis as an event steam, which has workers that persist to DB. Helps manage spikes and even out DB writes.

It’s working well.

Maybe some confusion. Most of those top accounts are spam and automatically generated.

I included them as it’s important to compare spam usage vs normal humans - so we can better detect and address it.

I’m certainly not claiming above maybe 30k events so far is human. I’m only at 2k. But certainly 10-25k events is achievable - when including stale kind 0/3.

Quick and dirty Rust Nostr ZAP Validation

https://gist.github.com/blakejakopovic/76714633f844e1a8b343805997fea5d7

Database storage size (event json string only) for top 32 pubkeys.

Still some spam accounts to filter/delete. Includes history meta and contact list event data (which adds maybe 40%).

Interestingly 1GB relay storage would appear to get you pretty far. Obviously actual storage with indexes, metadata and such is at least 2x.

I have 1,379 identities in my 2nd degree Nostr network (followers of my followers, uniqued from 5k), from following 129.

Too tired calculate, but I imagine someone’s general reach may be the followers of their followers, as client apps start to use 2nd degree filtering.

Perhaps not censorship.. but feels like it kills the ‘town square’ ideal for Nostr. Needs more thought…

Maybe views solve this.. and a more personal feed ok, as long as we don’t forget the world outside our 2nd degree bubble.