Open leo816 opened 3 years ago
- API > What help do you need to reduce the time of this task?
@ripcurlx has been using the api for personal stuff and that helps with testing. @sqrrm has done all of the api reviews. It will help if he continues. They both have their own chores, so I'm not hoping they can take on much more than in the past. Other devs said I was making steady progress before api work was put on hold. I think I can return to that speed again pretty soon.
A fully functional api containing most/all features of UI is a long term task, and involves some refactoring (indirectly related to api, and not visible to end-users) to avoid duplicating code. No ETA on delivering the "finished" api. And adding some trading bots to the release is something I want to do, but we have to be cautious about it. Up till now, the consensus was to let end users build and run their own bots at their own risk. It sounded kind of straight-forward and easy in the dev call, but much more discussion and prototyping is required.
In the short term, I'm making sure the api will support creating swift payment accounts soon after @jmacxx's PR is merged. I have it working in a local copy of his branch, and have asked him to review the swift account json-form (api analog to the ui's swift acct form).
I intend to start work on XMR <-> BTC trading w/ api tomorrow or Monday (06-Sept).
Hi @leo816
I would like to add 6-10 payment methods over the next couple of months.
I need help of a dev that is available to do the coding work required for adding them.
@jmacxx has done three payment methods, I am unsure as to how many more @jmacxx can take on? @BtcContributor has done one but is no longer able to take on any work
It would be great if one or two more devs were available.
1. Devin's Proposal
hybrid app development approach, a single codebase that could support both, Swift app for IOS and java for android. Possible developer candidate reached out.
concerns from Christoph: too many dependencies, is it easy to do an audit? potential security issue?
2. API
@ghubstan : Update: about 3/4 done with work to support XMR/BTC trading pair in API.
3.How can we attract more developers? bisq is not running like a business so it's hard to hire people as if it were a business.
Stan: Maybe we could narrow down the search for web UI devs. I can build a webapp over the core-api, and server Json. What could convince me to work on Bisq (beyond interesting content of the work) is confidence in the core devs' integrity (not scammers), and that I will be able to pay the bills. Not necessarily a guaranteed paycheck, but a high level of confindence I will earn enough ducats to live on if my work is being accepted.
Also a consistent road-map... not too many detours on new projects that get suddenly dropped. I know it can't be helped sometimes. But detours should be taken thoughtfully.
@Pazza has some great ideas, No one is assigned yet, feel free to put ideas out there. --> Where can I read up on the ideas? Would love to know more ...
I appologize for posting this here, I can not find another way to reach out to @pazza83. For several months now, it appears as though you are getting tagged as @pazza which is notifying me of everything issue/comment/PR. Just an FYI to your users (as well as yourself who may not be receiving notifications).
I appologize for posting this here, I can not find another way to reach out to @pazza83. For several months now, it appears as though you are getting tagged as @Pazza which is notifying me of everything issue/comment/PR. Just an FYI to your users (as well as yourself who may not be receiving notifications).
Thanks for reaching out much appreciated, sorry that you are being notified.
Hey everyone please can you tag me in to any discussion here as @pazza83
My user name on Keybase is pazza but on here it is @pazza83
@Pazza has some great ideas, No one is assigned yet, feel free to put ideas out there. --> Where can I read up on the ideas? Would love to know more ...
Hi @flybylow many thanks I will summarize the UX posts I have proposed yet to be implemented shortly and post here.
1. Bisq on mobile Devin discussing the debate about what language to use for his proposal (Kotlin) the difficulty finding the right devs for this. m52go would like to see a firm timeline for the Mobile project to have interop with the Bisq network natively (e.g. the client/server access to offer book that chimp mentioned).
2. Release 1.7.4 Emzy just pushed a pull request for the tor v3 tor v3 works. Have used it for a week. tomorrow Chirstoph will likely merge. Friday likely prerelease.
3. API and Daemon update @ghubstan API (gRPC Daemon+CLI) update:
I posted a warning in #general & #api about an API bug that could have shown a changing "buyer cost" value in Fiat/BTC trades (market-price-margin offers) in the CLI console's 'gettrade --trade-id' output. So far, no API user has claimed they paid too much, but it's early.
I posted a fix with an explanation of the problem in
https://github.com/bisq-network/bisq/pull/5704
API (gRPC Daemon+CLI) 'editoffer' support has been merged into the master branch; it might be included in the next Bisq release.
Consensus among core devs about what I should work on next would determine my short term priority. Otherwise, I work on a new 'gettrades' API method.
4. Attracting new developers (who may want to contribute on their free time but don't know about bisq).
5. Review of priority list
Hello everyone, this is Andres, author of JReleaser. In your own words, what's required to consider JReleaser as ready? Please feel free to post a discussion topic and/or issues at https://github.com/jreleaser/jreleaser
1. X @cbeams Q&A
2. API update @ghubstan API update:
I am working on a new API CLI getrades --category=<open|closed|failed> method. Submission of a PR for that will be preceded by some preliminary refactoring work in a couple of desktop module trade views and models (another PR). I think it will be ready ~6-SEP, +/- a couple of days.
After that (bugs or other priorities permitting), I was planning to build a Bisq webapp POC that depends on Core API (no gRPC), and will serve json to thin web/mobile clients. The POC would respond to browser HTTP REST requests by serving them raw json. I will not be building a UI.
API (gRPC Daemon+CLI) 'editoffer' support has been merged into the master branch; it might be included in the next Bisq release.
Consensus among core devs about what I should work on next would determine my short term priority. Otherwise, I work on a new 'gettrades' API method.
3. Review of priority list
Priority list
1. BSQ<->BTC swap
2. API
I am working on a new API CLI getrades --category=<open|closed|failed> method. Submission of a PR for that will be preceded by some preliminary refactoring work in a couple of desktop module trade views and models (another PR). I think it will be ready ~6-SEP, +/- a couple of days.
After that (bugs or other priorities permitting), I was planning to build a Bisq webapp POC that depends on Core API (no gRPC), and will serve json to thin web/mobile clients. The POC would respond to browser HTTP REST requests by serving them raw json. I will not be building a UI.
3. Messaging Problem
4. Mobile app and UX
6. XMR
7. Payment methods
8. Adding Tether
- Topics for next call:
1. BSQ<->BTC swap Almost done, next release will include it. @ripcurlx : making the whole process easier (specially for non users)
2. Payment Methods @jmacxx added 14 new payment methods.
3. API update @ghubstan API update:
4. XMR/BTC swap Currently reviewing the process to implement it.
5. Review of priority list
Bisq mobile: new potential contributor to be put into contact with Devin.
BSQ<->BTC atomic swap PRs and refactorings done, being reviewed by @stejbac, @ghubstan, @ripcurlx et al.
DAO performance improvement / refactor PRs ready for review & testing. Action all devs.
v1.8.0 - We should allocate more time for testing this cycle due to the major changes. Several more days would be good, so that full regression testing can be performed. Action @ripcurlx @BtcContributor
@ripcurlx now has immunity to all viruses (:crossed_fingers: )
API: @ghubstan to continue.
Running Bisq on Umbrel - @ghubstan to consider investigating.
Discussion of https://github.com/bisq-network/proposals/issues/344 Action item for all: give feedback. See if a small scale project can be defined, with suitable budget. Project better focus than 'role'. @pazza83 - One possible project idea would be to solicit input from XMR users.
Payment methods - not mentioned in the meeting, but there are a few more left to add. Action @jmacxx
@jimmyneutront expressed the desire to hear more about X Protocol progress.
1. BSQ<->BTC swap PRs and refactorings done, being reviewed by @stejbac, @ghubstan, @ripcurlx et al.
2. Payment Methods @jmacxx : working on the last 4 payment methods for @pazza; started on Grin re-listing tasks; and the usual bug fixing / support.
3. API update
@ghubstan API update:
re-submitted some API related work in the last few days:
(1) some minor adjustments in the API to the new bsq-swap support,
(2) improvements to the API CLI's output formatting code, to make it easier to use as new API methods are added.
(3) BSQ swap testing, looking at umbrel
4. XMR/BTC swap @ghubstan XMR/BTC trading pair support in in the API -- adjusted for the new bsq-swap feature.
5. Research & Connectivity Chimp - waiting for PRs to be merged; did some research into Misq & Kotlin / async. Looking into tor connectivity issues.
6. Review of priority list
1. Provide ASIC resistant PoW scheme for BSQ swaps Already merged by Steven. Now he will focus on bug fixing https://github.com/bisq-network/bisq/pull/5858
2. sqrrm started reviewing the API PRs
3. Messaging problem Is the core to a large percentage of all failed trades and yet no one is looking into it besides James
4. Review of priority list
1. Misq brainstorm and updates Update of current plans and review of in progress work from Chimp. It would also be great to have someone developing the mobile app simultaneously. No updates from Devin yet.
2. Recommendation for power users for big bisq users, using a local bitcoin node will really help and avoid different bugs. creating a lot of offers loads too much your client. 20/30 offerrs at once is the limit as to what could be generating problems.
3. We need a full time UI developer working for the next 4-5 months It would also be great to have someone develop
4. Growth BSQ/BTC Swap Have been adopted nicely but no marketing and virtually no one knows about it. Team lead for growth is key.
5. Review of priority list
1. Bisq2.0/Misq brainstorm and updates Update of current plans and review of progress work from Chimp. Connecting from i2p y Tor simultaneously. Liquid for multisig and swaps will work. social aspect - moderators will have to burn btc and will be compensated.
2. We need a full time UI developer working for the next 4-5 months It would also be great to have someone develop the mobile app.
3. Growth/marketing BSQ/BTC Swap Have been adopted nicely but no marketing and virtually no one knows about it. Team lead for growth is key and people engaging more on platforms like reddit is key.
4. Liquid implementation Alva already implemented it this week.
5. Updates from last call "API examples documentation. to show that the API is fully functional, have 3 examples that companies can embrace." first examples from ripcurl: https://github.com/ripcurlx/bisq-api-examples @ghubstan : I'm still working on api reference doc generations, then commenting api protobuf files (api reference doc content), then snippets for api calls, then the "3 examples" (complete bot trading scripts)
6. Running the numbers Pazza's monthly P/L review: https://github.com/bisq-network/roles/issues/111#issuecomment-1029551837
6. Review of priority list
1. Comments on Tasks/priorities ripcurlx priorities: review/merge PRs, esp. API to "market" it with users; then UI + contact with UI devs (no leads for now); API is good enough for trading bot, not for mediation/arbitration; might be issue where API daemon needs restarting with heavy usage.
2. Next release ripcurlx: release 1.8.3 close as it just needs some work on the DAO state fix/restart situation.
3. Growth/marketing BSQ/BTC Swap Have been adopted nicely but no marketing and virtually no one knows about it. Team lead for growth is key and people engaging more on platforms like reddit is key.
4. Concerns i. jmacxx is concerned that the messaging problem will be inherited by bisq2. reports issue with messaging over network failing in ~1% of cases, maybe Tor problem, no clues as of now how to solve; alternative trading implementation, robosats over LN. ii. pazza: citing m52go, with bisq2 it's important to keep contributors' privacy, but maybe also keep private trades information (and profit/loss info as well?) (Create a guideline on how to "keep" a private profile as contributor.) w000000t: agreeing to write a wiki article about privacy for contributors
5. Review of priority list
1. Comments on Tasks/priorities Bisq2.0 update with chimp. will take about 2 months to have a first prototype ready. alcom, alva will be joining as devs. the chat based area should be very feasible to adapt to mobile. promoting bisq is crucial. chimp: API should still be n1 on the priority list.
security model in terms of how the offers are secured, reputation concept based to bring everything 0 and 1. this number will be calculated directly in the background, when you create an identity it has a timestamps and that is a small part of reputation.
2. Growth/marketing We have been in contact with many people to try and help bisq grow. We are still looking for someone to speak in the name of bisq.
3. Review of priority list
1. Comments on Tasks/priorities
Tor issue seems partly solved but future problems will likely arise as there is no way to solve it from our side. Useful to implement at least an additional tor monitor to have multiple views of the network status, but long term solution is having multiple supported networks(We have to make sure we always have the latest release)
Emzy: Price nodes Issue, some exchanges had some weird prices. not sure how to manage that. If its off more than 2% then we should do something to prevent that. We had an experience where the monero price was diverging but onnly by a small percentage, but brazillian dollar was way off. Manfred suggests that we can implement some code to detect outliers.
tor issues and seednodes not working and then working again by themselves; implement seednodes switching when one stops responding
Haveno stopping UI dev
2. Bisq 2.0 Chimp showing developments on bisq2. and giving a tutorial on how the new version will work. Full flow from login/user creation up to trading and inviting mediator, inline knowledge base; possibility of electrum wallet integration Bisq 2.0 looks amazing, looking forward to the release.
3. Growth/marketing We have been in contact with many people to try and help bisq grow.
4. Review of priority list
Dev call 2021-09-03
Topics
We all agree on Misq/Bisq's 2.0 long-term vision. for that to happen, we need the right developers with enough experience on this area. Bisq contributors should do all they can to attract the right talent.
With this in mind, these are Bisq's top priorities in the short-term:
1. Tor. v3:
2. API
3. Messaging Problem
4. Mobile app
5. UX
6. XMR
7. Payment methods
8. Adding Tether
- Topics for next call:
Click to see the example
what it would look like: At 00:00, node A creates a SELL swap offer and publishes their public key, and then disconnects from the internet. At 01:00, node B connects to the internet, scans the blockchain for new swap offers, and discovers node A's swap offer. Node B takes node A's swap offer, publishes their (node B's) public key, and then posts a message in the Matrix room containing payment method info, encrypted with node A's public key and signed with node B's private key. Then node B disconnects from the internet. At 02:00, node A connects to the internet, realizes their offer has been taken by node B, and then gets all unread Matrix messages, searching for the payment method info message from node B, decrypting it with their own public key and verifying the signature with node B's public key. Upon success, node A posts a message in the Matrix room containing their payment method info, encrypted with node B's public key and signed with node A's private key. Then node A disconnects from the internet. At 03:00, node B connects to the internet and gets all unread Matrix messages, searching for the payment method info message from node A, decrypting it with their own public key and verifying the signature with node A's public key. Upon success, the operator of node B can send fiat payment according to the payment method details specified by node A, and then can notify that payment is sent. Then node B disconnects from the internet. At 04:00, node A connects to the internet, discovers node B has indicated that payment has been sent, and the operator of node A verifies that fiat payment has been received. Then, node A confirms that payment has been received, allowing each swapper to close their half of the swap and receive their stablecoin whenever they wish. In this way, a swap can be completed from start to finish without the swapping parties ever being online at the same time, and without Bisq/Commuto devs having to design/test/maintain complicated P2P infrastructure.