Replying to Avatar vinney...axkl

I've considered a threshold like that but it seems a bit overly complicated for now. I'm not sure "where" that multisig would go.

Plus: I think there is a lot of value in having Arbiters as sinks of reputation and expertise. For example: say as a Patron I want to have Y software built, but I know little about the domain. I can find an Arbiter who is a domain expert, and trust that they will either disperse funds or refund me depending on the satisfaction of the task. They will evaluate the work and determine if it truly meets the spec of whatever Y Software was intended to be.

That is good for all parties involved in the transaction **and** the market, as end-users of the resulting product are more likely to have a good experience if a knowledgeable Arbiter helped guide it. (And the Arbiter gets a cut, so being an expert/curator like this is profitable)
Avatar
DanConwayDev 11mo ago

It some circumstances, the arbiter may need an incentive to participate due to the work involved in review, so they could be assigned a split. The ecash could be locked to the threshold in the same way that cashu ecash can be locked to a nostr pubkey.

I agree with not introducing complexity until there is a solution that validates product market fit. nostr:nprofile1qqsdqnk0xw3s8fvc2t7mdq0d3dqjyqd6shvdyxv6a3eukcngr4324yqpzamhxue69uhhyetvv9ujuurjd9kkzmpwdejhgtcppemhxue69uhkummn9ekx7mp0qythwumn8ghj7cnfw33k76twv4ezuum0vd5kzmp0c6dswr with satshoot is taking that approach by just focusing on coordination and not payment yet.

Reply to this note

Please Login to reply.

Discussion

Avatar
vinney...axkl 11mo ago

Current design just has Arbiter getting a few on the total amount paid, but I agree that is likely too naive for the use-case we're describing here.

...then again, if the Arbiter's split is too attractive, he could greenlight garbage just to get paid. Though I would expect public eyeballs to ding his reputation in that case.

Might be the case that a market like this can't really work for high value until Nostr has proper reputation primitives beyond manual pubic comment.

👀 nostr:nprofile1qqsw2feday2t6vqh2hzrnwywd9v6g0yayejgx8cf83g7n3ue594pqtcpz3mhxue69uhhyetvv9ujuerpd46hxtnfduq3kamnwvaz7tmjv4kxz7fwvf5hgcm0d9h8qctjdvhxxmmdqyt8wumn8ghj7erpwe5kgtnwdaehgu339e3k7mgxeu5tj nostr:nprofile1qqs0dqlgwq6l0t20gnstnr8mm9fhu9j9t2fv6wxwl3xtx8dh24l4auspz4mhxue69uhkzat5dqhxummnw3erztnrdaksz9rhwden5te0wfjkccte9ejxzmt4wvhxjmcpzemhxue69uhhyetvv9ujumn0wd68ytnzv9hxgxmpqkh

Avatar
DanConwayDev 11mo ago

The threat of reputational damage maybe enough. Also the arbiter might have other incentives. For example, the patron might want a feature in a library and the arbiter might be the library maintainer.

Avatar
vinney...axkl 11mo ago

All good points, yes.

In general I want the protocol to have as few opinions as possible and just provide the tools that the market can use to find the best solutions to these challenges.

No way I'm going to have all the answers up front!

Avatar
DanConwayDev 11mo ago

Yes. Simplicity, extensibility, flexibility.

Avatar
david 11mo ago

Is there any feedback or “rating” mechanism for arbiters? If so, GrapeRank will be able to it to good use and make sure that spammy feedback gets ignored. Format of the feedback can be anything: reactions, numeric rating, context or no context, etc, which means you can optimize for UX. Even just the fact that an arbiter was hired for a job and got paid can be used as input, if that data is available.

Avatar
vinney...axkl 11mo ago

The Arbiter and Patron proposal and acceptance events, and all the payments are all tracked via Nostr events so yes definitely available data.

And a project being successfully completed or rejected it also tracked via events - which would be good for reputation of the Free Agents who perform work.

Could definitely look into adding more feedback and rating mechanisms, I'd love for this to be synergistic with GrapeRank

Avatar
david 11mo ago

Do you have in mind a mechanism to flag dishonest / scammers / bad actors? NIP-56 reports or NIP-32 labels would be a possible place to start.

I suppose it would be simple enough to calculate the number of successfully completed projects score, something similar to what eBay does, but do a GrapeRank-weighted sum rather than a simple sum to screen out fake projects.

This reminds me that I’ve been wanting to export WoT scores using nostr:npub1gcxzte5zlkncx26j68ez60fzkvtkm9e0vrwdcvsjakxf9mu9qewqlfnj5z ‘s NIP which you could use to import GrapeRank scores for the above calculation.

Avatar
vinney...axkl 11mo ago

You could definitely query for all the rejected projects where the Arbiter returned the funds to the Patron because the work was crap or fake. But that tells you the bad Free Agents.

There sort of wouldn't be "fake projects". A Patron has to pay the project amount in escrow to the Arbiter before a Free Agent worker can be assigned , so once that happens there is skin in the game.

I suppose you could check for Patrons who initiate tons of spam projects and never finalize the escrow... But I don't see that happening a ton.

The real thing you'd want to be vigilant for is Arbiters who abscond with the escrow! You'd query that by looking for work submissions that don't ever have a correspondening payment NOR refund.

TL;DR: I'd like if the whole reputation picture can just be gleaned from the event trail without requiring reports or anything like that. Maybe an Arbiter somewhere sets up a DVM that does these queries routinely in order to prove to the market that they are among the most trustworthy. He could charge for that DVM - almost like a licensing board, but nobody has to pay attention to it if they don't care.

Avatar
Five 11mo ago

SatShoot concentrates on finding the best client or freelancer with minimized reliance on third parties:

- reviews

- public zaps as payments

- outbox model

- client-side wot

- app-scoped follows

It is a simple scenario whereby freelancers bid on problems, not a bounty type of flow. Specs available [here](https://github.com/Pleb5/satshoot/blob/main/apps/satshoot/EVENT_STRUCTURE.md) .

Payment flow in SatShoot is very simple, you always pay as much as you want but cannot reverse it(LN, nip60 Cashu ecash). There's option to pledge a split from freelancer earnings which is tied to reputation.

Escrow could play a part in it, it's just not a focus yet. Talked about this with nostr:nprofile1qqs8rheprycaymhyzysa99dag09u0cuz2p0rxw6uz02qzm8dj4pdn4cpzemhxue69uhhyetvv9ujuurjd9kkzmpwdejhgqg5waehxw309aex2mrp0yhxgctdw4eju6t0qyf8wumn8ghj7ur4wfcxcetsv9njuetnrptj7x before, he has a Cashu escrow service project going.

More on SatShoot design and plans:

It is hard to do nostr right and get a decent UX at the same time. It's doable but takes time and effort.

The core experience needs to be nailed first and foremost which is going to be there with the next version soon enough. Many details of the deal flow is pushed to the margins where parties can flexibly agree on exact terms.

The reason I did not go the bounties direction is that most of the time teams need to invest heavily without knowing that the work will be compensated. There are exceptions (eg bug bounties) but I wanted to solve something that seemed as a more pressing problem to me:

An unstoppable freelance market for earning sats. The strongest use-case is software development. So there's a natural incentive to integrate SatShoot with nostr-git and nostr-based project management like kanbanstr. Pledge splits can be extended to multiple parties that integrate SatShoot-like job posting.

That's my plan in a nutshell.