Let me just give a quick heads up on the Travis issue(s): I'm not sure what exactly Travis runs, but from my own experience (only on GNU/Linux, of course) the regression tests should "mostly" work now after fixing the BIP9-related test. The only thing that I'm aware of is that "p2p-fullblocktest.py" works sometimes and sometimes not. This seems to be related to upstream Bitcoin rather than Namecoin, see https://github.com/bitcoin/bitcoin/issues/7978. Roughly speaking, I added more mempool consistency checks in Namecoin to make sure everything is working fine, and one of them may fail for this test randomly, even if they are enabled on the pure Bitcoin code alone. I'm not sure why exactly that happens, and if it is really supposed to be like that as the one guy in the bug report stated.
BTC: 1domobKsPZ5cWk2kXssD8p8ES1qffGUCm | NMC: NCdomobcmcmVdxC5yxMitojQ4tvAtv99pY
Use your Namecoin identity as OpenID: https://nameid.org/
Use your Namecoin identity as OpenID: https://nameid.org/
Thanks for the info Daniel. There is at least one other issue related to Travis, which I'm currently investigating, and will post more info on when I know more. (Maybe move discussion of this issue to a different thread?)domob wrote:Let me just give a quick heads up on the Travis issue(s): I'm not sure what exactly Travis runs, but from my own experience (only on GNU/Linux, of course) the regression tests should "mostly" work now after fixing the BIP9-related test. The only thing that I'm aware of is that "p2p-fullblocktest.py" works sometimes and sometimes not. This seems to be related to upstream Bitcoin rather than Namecoin, see https://github.com/bitcoin/bitcoin/issues/7978. Roughly speaking, I added more mempool consistency checks in Namecoin to make sure everything is working fine, and one of them may fail for this test randomly, even if they are enabled on the pure Bitcoin code alone. I'm not sure why exactly that happens, and if it is really supposed to be like that as the one guy in the bug report stated.
Code: Select all
#namecoin-meeting notes, 2016 07 03 # Present: Jeremy Jonas Samurai pigeons midnightmagic Decentralized Web Summit Jeremy attended DWS. Blog post summary is in namecoin.org PR. https://github.com/namecoin/namecoin.org/pull/64 Note the new News layout / links, the Earlier News link, and the new post from June 20. Jonas ACKs. Namecoin Core Merged Mining Jorge Timon suggested replacing the MM chain ID with the BIP 122 chain ID. https://github.com/namecoin/namecoin-core/pull/75#issuecomment-228597157 OS X Gitian Builds Jonas posted an OS X Gitian binary of Brandon's name tab PR Jonas will look into submitting a PR. Look into deterministically generating SDK input. PGP keys Jonas reminds everyone that we need to get our PGP keys into GitHub for Gitian. AAA hardfork Samurai asks if any problems expected short-term. Jeremy is pretty sure we're fine until Bitcoin schedules bit 0x100 for VersionBits. SegWit is bit 0x2. SPV Jeremy sent PR to libdohj. Currently undergoing review. https://github.com/dogecoin/libdohj/pull/18 Specs Jeremy posted draft Dehydrated Cert spec. https://github.com/ifa-wg/proposals/pull/24 Review would be appreciated. ncdns Jeremy noticed a bug in ncdns's JSON-RPC library that breaks connecting to the SPV client. Jeremy fixed it, but wonders why we're even using that library. It appears to be a very old library from Conformal that was abandoned over a year ago. Bring this up with Hugo (he didn't attend this meeting). Public Relations A bunch of website PR's have been ACKed by Jonas and merged by Jeremy. https://github.com/namecoin/namecoin.org/pull/59 https://github.com/namecoin/namecoin.org/pull/60 https://github.com/namecoin/namecoin.org/pull/62 https://github.com/namecoin/namecoin.org/pull/65 https://github.com/namecoin/namecoin.org/pull/66 https://github.com/namecoin/namecoin.org/pull/67 https://github.com/namecoin/namecoin.info-dotbit/pull/10 Jeremy added new "Donations" page to website. https://namecoin.org/donate/ Jeremy added new FAQ to website. https://namecoin.org/docs/faq/ Funding libcoin-related bounties have been refunded by BountySource. BountySource balance: $5/month from 1 recurring supporter Total funds available: $299 USD NMDF incoming funds: 98.4 mBTC received since 2016 05 15 0 NMC received since 2016 05 15
Code: Select all
#namecoin-meeting notes, 2016 08 21 # Present: Jeremy Jonas Brandon Midnightmagic Pigeons Joseph Namecoin Core The wallet name operation bug in Brandon's name tab PR has been isolated by Jeremy. Appears to be a bug in upstream Bitcoin Core (or perhaps Namecoin Core master branch) that has been fixed. https://github.com/namecoin/namecoin-core/pull/67#issuecomment-231316792 Whit Jackson submitted a documentation PR for building on OS X. https://github.com/namecoin/namecoin-core/pull/97 If anyone can test, please do so! Jonas will test this. Jeremy noticed that Travis CI accidentally flagged our repo as "potential abuse detected". Jeremy contacted Travis CI support and got our account whitelisted. Jonas notices that the Travis CI builds for our 0.13 branch are failing at the moment. Jeremy suggests filing a GitHub issue. Jeremy asked Jonas to submit a PR for Gitian builds for OS X. https://github.com/namecoin/namecoin-core/issues/38#issuecomment-236101580 DrHaribo of BitMinter requested that getblocktemplate be re-added to Namecoin Core. https://github.com/namecoin/namecoin-core/issues/98 Jeremy points out that since nVersion=4 blocks are nearing the lock-in point, we should deal with this sooner rather than later so that BitMinter doesn't get kicked off the network. Brandon asked on GitHub about how wallet unlocking should be done with the name tab GUI. https://github.com/namecoin/namecoin-core/pull/67#issuecomment-240852324 Upon discussion, Brandon currently plans to try using the raw transaction API for this. Daniel submitted a PR to temporarily disable the low-S standardness check. This should improve confirmation times for the old 0.3.80 clients. The check will be restored after AAA activates. https://github.com/namecoin/namecoin-core/pull/100 Test reports would be great. Jeremy points out that this needs to be done before nVersion=4 blocks are locked in. Jeremy wonders whether we should cancel plans to release 0.12 as stable, and focus on 0.13 branch. Tentative plan: keep maintaining 0.12 until Bitcoin Core releases v0.13.0; backport the name tab to whatever stable branch exists at the time that it's merged to master. SPV Jeremy's initial PR for Namecoin support has been merged to libdohj. https://github.com/dogecoin/libdohj/pull/18 Jeremy still needs to rebase the bitcoinj-addons code based on the libdohj changes made during review, and then submit a pull request. In the meantime, Jeremy's latest bitcoinj-addons and libdohj code is posted. https://forum.namecoin.org/viewtopic.php?p=16788#p16788 Someone please test it? Jeremy asks if he should post a link on Reddit. Jonas says yes. Jeremy will do so. ncdns Jeremy asked about the old fork of a Conformal library (see previous meeting); Hugo doesn't recall details but thinks it had to do with pre-Namecoin-Core not following spec properly. Hugo would be happy to accept a PR to make it use the current Conformal libs. https://github.com/hlandau/ncdns/issues/12#issue-163964219 Jeremy pointed out that sometime in the future we should fix the Extended Key Usage Critical flag on the dehydrated cert template. Go's standard library doesn't have a built-in way of setting that flag, which is why it's not in the first draft. Jeremy and Ryan can't think of any plausible attacks where it matters. Jeremy suggests a 1.0 BTC bounty for an ncdns NSIS installer. Possibly funded 50/50 by NMDF and a fundraiser. https://forum.namecoin.org/viewtopic.php?f=18&t=2566 Mining Cassini noticed that BTCC and ViaBTC have started mining Namecoin. https://forum.namecoin.org/viewtopic.php?p=16703#p16703 This is good news for mining diversity. F2Pool's share of Namecoin blocks is down to circa 43% as of July 26, 2016. Public Relations Cassini represented us at GETD#4 in Berlin July 22-23. Jeremy mentions that we should think about translation workflow. https://github.com/namecoin/meta/issues/35 Jeremy notes that Bitcoin Core is using Transifex. https://github.com/namecoin/namecoin-core/blob/master/doc/translation_process.md Jeremy says that whatever works for Bitcoin Core, should probably work for us. Midnightmagic concurs. Joseph believes that Armory (post-ATI) is trying Transifex. Brandon asks what the pricing looks like for Transifex. Joseph believes it's gratis for open-source projects. Jeremy notes that Wikipedia confirms this. Jeremy wonders what safeguards are in place on Transifex to reduce risk of malicious translations. Joseph isn't sure. Jeremy points out that the way we organize the list of exchanges doesn't make sense for decentralized exchanges like Bitsquare. https://github.com/namecoin/namecoin.org/issues/70 Jeremy suggests listing decentralized exchanges at the top of the list. Jeremy notes that this might annoy the centralized exchanges who pay us for placement, but that this doesn't bother him at all. Midnightmagic concurs. Pigeons mentions that he's seen an increase in Namecoin offers on Bitsquare recently. Jeremy will have a table at the OU CS welcome party on Sept. 9. Let's try to recruit some developers! Brandon asks what the status is of the students from last year. Jeremy is sad to report that the SPV student and the UX student from last year are not participating this semester. Pigeons says a "Namecoin vs Blockstack" FAQ entry would be helpful. Jeremy will work on it. Midnightmagic asks if a meetingbot would be welcome. Jeremy says yes. Midnightmagic will look at setting one up. Funding Jeremy has filed a complaint with Tip4Commit about their usage of CloudFlare CAPTCHAs. https://github.com/tip4commit/tip4commit/issues/300 Tip4Commit has not responded after 25 days. BountySource balance: $5/month from 1 recurring supporter Total funds available: $304 USD Pigeons suggests looking into Patreon. Jeremy notes that they have CloudFlare CAPTCHAs. Jeremy notes that Wikipedia says they take a 5% commision, which is half of BountySource's 10%. Jeremy notes that some free software projects use them currently. https://www.reddit.com/r/linux/comments/33dp6n/should_open_source_projects_use_patreonkickstarter/cqkrlkf Jeremy will look into this in more detail later. NMDF incoming funds: 0 mBTC received since 2016 07 03 0 NMC received since 2016 07 03 Pigeons suggests looking into Flattr. Jeremy notes that they have CloudFlare CAPTCHAs. Jeremy notes that Wikipedia says they take a 10% fee (same as BountySource). Jeremy will look into this in more detail later.
Sorry for the delay posting this one.
Code: Select all
#namecoin-meeting notes, 2016 09 04 # Present: Jeremy Joseph Hugo Jonas Brandon Consensus Jeremy wrote up a draft proposal for a future consensus change: the "Soft Hard Name Fork". https://github.com/namecoin/meta/issues/38 Inspired by Luke-Jr's SegWit and AuxPoW forks. Would be great to have some people poke holes in it. Jonas suggests that testnet2 could be included when we fork away from 0.3.80. Jeremy probably wouldn't consider it a blocker, but probably wouldn't have any objection to including it. Namecoin Core Cassini has tested and ACKed Daniel's PR for disabling the low-S standardness check. Daniel has merged it. Daniel would like to release 0.13.0 based on Bitcoin Core 0.13.0, with the low-S standardness check disabled, and wait until 0.13.1 for the name tab GUI. https://forum.namecoin.org/viewtopic.php?f=8&t=2603 Jonas volunteers to do Gitian builds. Jonas ACKs releasing 0.13.0 without the name GUI. Brandon has added wallet unlocking to the name tab Qt PR. The raw transaction API turned out to be too complex for our timeframe. We'll look into adding it later. Also fixed all Travis CI failures in the PR. Also fixed the iCCP errors. Jeremy noticed a blocker bug that occurs when 2 people try to register a name at the same time. Brandon has pushed a fix, needs testing. Jeremy will try to test the fix. Jeremy asks if Jonas has had any luck submitting a PR for Gitian builds for OS X. https://github.com/namecoin/namecoin-core/issues/38#issuecomment-236101580 Jonas will try to submit a PR. TLS Jeremy notes that we need feedback on the dehydrated cert spec. https://github.com/ifa-wg/proposals/pull/24#issuecomment-232153966 Other clients Joseph mentions that there's an Electrum fork for Namecoin that seems dead. Jeremy notices there's more than one of them. After some confusion, Jeremy realizes that 2 of the Electrum forks are forked from each other. https://github.com/Kefkius/electrum-nmc-server https://github.com/Kefkius/electrum-nmc https://github.com/mazaclub/electrum-nmc-server https://github.com/mazaclub/electrum-nmc Joseph mentions https://jonasnick.github.io/blog/2015/05/09/fuzzing-bitcoin-consensus/ Joseph thinks it might be interesting to try with Namecoin clients. Public Relations The Google search result for namecoin.org is now fixed. Reminder that Jeremy will have a table at the OU CS welcome party on Sept. 9. Funding BountySource balance: $10/month from 1 recurring supporter Total funds available: $314 USD NMDF incoming funds: 0 mBTC received since 2016 08 21 0 NMC received since 2016 08 21
As of the meeting 2 days ago, midnightmagic is providing a MeetBot that automatically generates minutes and logs for us. The logs, starting with the meeting from 2016 09 18, are currently living at https://github.com/midnightmagic/namecoin-meeting/ . Huge thanks to midnightmagic for setting this up.