Freelance work I've been chipping away at this year has revealed this:
I regret we often have to tether ourselves to things that cause burnout, to get to the place where we can build truly useful things. Can't change the game yet, so just have to win it :)
1
144
curion retweeted
Wasted a lot of time researching wallets, they're all bad.
Metamask - bugs, unreliable HW support, bad UX.
Rainbow - no custom RPC support, hides metadata too much.
Rivet - no ledger support, dapp-dev focused, stagnant 2+ mo.
XDefi - Swaps I don't need, more defi than tech.
Keplr - good multi-chain, bad multi-account, why?!
GridPlus - expensive hardware, doesn't fix browser UX.
Status - social stuff I don't need in a place that shouldn't have it.
MyCrypto - joined metamask, RIP.
Ambire - not an extension, walletconnect, no custom networks.
MEW Enkrypt - better metamask, still too non-technical.
Frame - github stagnant 2+ mo. Window management issues.
Rabby Desktop - seems ok, very recent, trying it out.
Can I have a wallet that doesn't hide what it is doing?
How am I supposed to trust if I can't see?
Most users might not be interested in what happens under the hood, but I am.
And where are the real technical features?
Wishlist:
- Log of RPC calls, tx hashes, etc?
- Account management that isolates each account?
- Introspection of signing requests?
- Network customization that actually works?
- Different local network configs with the same port?
- Different RPCs for the same chain.
- Network-selection not stuck when current RPC is offline.
- User-choice for insertion as window.ethereum or EIP6963.
- Wallet-connect support as a last-resort, not as a default.
- Choice of different RPC endpoints to serve different RPC methods?
- Blocking RPC requests until manual ACK.
- Allows more than 1 chain to be active, across different windows?
- Dumping data to JSON / CSV as first-class feature.
- Makes transactions readable by decoding ABI calls.
- Allows for simulation of the transaction, with clear state-changes, before signing.
- Doesn't try to override the fee settings of a tx after you modify them.
- Provide cast commands to inspect/debug data?
- Logs state-changes along with transaction data, before it becomes archive-only data.
- Allows me to copy transaction/receipt contents without going to etherscan.
- Provides easy etherscan/blockscout/etc. explorer links.
- Provides me with a clear idea of the canonical chain.
- Supports light-clients or cross-RPC checks to not get trapped by lying RPC provider.
- Provides me with clear confirmation-distance / finality info.
- Allows you to label addresses that are not yours. And export/import label books.
- Allows you to pull address labels from etherscan/nansen/other.
- Can show you history as a graph of transactions.
- Makes transaction-nonces very clear, and allows you to override them.
- Understands queues of transactions.
- Allows you to re-submit a tx, even to a different RPC, without signing it again.
- Shows you clear historical basefee data.
- Shows you clear historical priority-fee data.
- Shows you clear historical block gas-usage / capacity data.
- Allows you to turn a dapp tx request into a multi-call, to append side-effects.
- Has "dotfiles" you can export/import for common chain interactions, like WETH wrap/unwrap.
- Allows you to *record* dapp interactions and turn them into a forge script for repeat/customized execution.
- Hides web3 accounts from websites you visit
- Provides read-only mode, does not require online keys.
- Doesn't rely on centralized data indexers.
- Runs outside browser as much as possible.
- Tells me if a signing/RPC request was closed by the dapp and got stale.
- Shows a clear overview of in-flight transactions.
- Supports monitoring of tx-hashes and nonce-increments across different RPCs, to get extra quick confirmation.
- Doesn't hide blockhashes / account addresses / technical metadata.
- Doesn't have a big supply-chain attack-surface.
- Doesn't auto-update with some app-store.
- Minimal javascript reliance. Not an electron app.
- Is very clear about local storage usage.
- Does not have a token.
- Doesn't have telemetry.
- Does not sell my orderflow.
- Does not sell my data.
- Does not force in their bridging.
- Does not force in their swaps.
- Is not owned by mega corp.
- Is open-source.
I don't care about an NFT gallery in my wallet.
I don't care about auto-updating ERC20s in my wallet.
I care about transparency and dev-level ownership over my own wallet.
Is there some wallet out there I don't know about?
How big was the opportunity cost of focusing on the protocol, and not fixing the above?
How much defi did I miss? How many airdrops? Degen'ing feels a lot a safer with a good wallet.
Should I just create my own web3 wallet? I know I can build all of the above, given enough weekends.
And no, I'm not contributing to your wallet if it's JS/TS/mobile.
Is there really nothing like this?
curion retweeted
This is unacceptable, should not be possible
Made possible by @_nonfigurativ_
this was fun. flashed an old PC with Colony OS from @NativePlanetIO and am now self-hosting my urbit planet (~risdev-dandel will be my initial planet if you want to connect)
it just works. ezpz. now to learn app development.
Total urbit ships booted is 11k, not even approaching planet capacity of a single star, I think it'd be a long time before that is part of the value proposition to buy a star. As far as NFTs go, it's a use case, but I don't see why someone would get one for that purpose.
2
181
Embrace Absurdity (Optimistic Nihilism)
This can be extended to just include an optimistic sense that we don't understand life's purpose and meaning but we can still do some cool things and reach our potential
I would find it more limiting if we did have a single purpose tbh
97
Agreed - very built out, maybe more focused on the social part and less on the part where you speculate on the future value of other keys
Status (discord-like communities) + Friendtech seamless key-o-nomics =🤯
been researching alternatives to FT
status wallet lets you use it in almost the same way as friend tech but arguably better
mint a token thats required to access the group chat (no bonding curve tho)
1
2
165
To leave a token-gated community in Status, I need to navigate to the wallet, swap from A to B, then join Community B's token gated chat. Subtle UX changes but I think enough friction to prevent a friendtech type experience. They could pivot though and bring a lot of expertise.
1
1
79