Page 1 of 3

#namecoin-meeting Recap/Notes

Posted: Sun Feb 08, 2015 4:28 am
by biolizard89
Here are the meeting notes from today, for those who missing it (or who just have a poor memory and want to refer to something). Someone (not necessarily me) will post them in this thread after every meeting. Cheers!

Code: Select all

#namecoin-dev meeting notes, 2015 02 07



Namecoin Core

(This was Daniel's first #namecoin-dev meeting.)
Namecoin Core is mostly working.
Daniel is rebasing regularly on the latest Bitcoin Core code.
DNS seed seems to be working fine for mainnet.
    Need more DNS seeds and hardcoded nodes
        fluffypony volunteers to host a DNS seed
Daniel suggests pushing to a branch of namecoin/namecoin; after that, merge Bitcoin Core commits as instead of rebase.
    Jeremy concurs.
Daniel suggests geting Windows and OS X binaries released for testing.
    Jeremy concurs.
    We need to test the Gitian build scripts.
        Travis CI builds properly, so Gitian is likely to work as well.
    phelix reports that EasyWinBuilder doesn't work, but probably due to upstream Bitcoin compatibility problems.
    fluffypony volunteers to test OS X
Rebranding the Qt GUI needs to be done.
    Even without name management, it is useful, e.g. Coin Control.
    Daniel doesn't consider porting name management GUI to Namecoin Core a high priority for him.
        Daniel suggests alternative GUI's such as Armory
            Jeremy is playing with a name JSON generator for Armory; blocked by NMControl development
We need to create new alert keys
Various softforks (and one hardfork) planned (happened in Bitcoin after Namecoin forked)
Jeremy and Daniel got the Travis CI tests running for Namecoin Core
    everything works, except Daniel had to disable the BitcoinJ comparison code.
        Jeremy suggests readding it using HashEngineering's NamecoinJ later (but low priority)
            Daniel concurs

Armory for Namecoin

Joseph has been trying to get Armory to build in Gitian; determinism hasn't been achieved yet.
Joseph thinks the issue relates to timestamps.
    Jeremy and fluffypony suggest faketime
Joseph is planning to work on Armory compatibility with Namecoin Core.
    Anticipated to be mostly minor things.
Joseph is having trouble with the tests
    Jeremy places $25 bounty on it to encourage a 3rd party to assist


Daniel will start the registration process when GSoC opens on Monday.
A few issues remain on the ideas page.
Daniel is available as a mentor for anything (though not first choice for all)
Daniel wants to keep the encryption project for GSoC
    Both NMControl-level encryption and namecoind-level hashing are okay, although a student could choose to only do 1.
    Students should be aware that namecoind-level things would probably be implemented on a testnet, and enabled on mainnet after GSoC concludes
Jeremy is concerned about the example spec for file signing (scalability concerns)
    Daniel is okay with the project consisting of a wrapper UI around GPG and Namecoin (just using an id/ GPG key) -- not too small for GSoC
    Daniel also suggests integrating id/ into Enigmail
    Jeremy suggests sidestepping controversy by not giving any specific example spec on the GSoC page and letting students propose something
        Daniel concurs.
We need to fill in expected outcomes and requirements for couple items.


Jeremy brings up that Python 3.4 or 2.7.9 appears necessary to get top-notch TLS security (e.g. TLS 1.2).
    However, the cipher suites can be secure with older versions
    Joseph points out that 3.4 is in Debian Jessie, but 2.7.9 won't be, so upgrading to Python3 would be helpful here
        Jeremy concurs, but mentions that we need to wait for mitmproxy to make the transition
Jeremy would like to rebase on requests library
    Problem is that we would need to wait for next release of urllib3 (which will add SHA256 fingerprint support)
        Joseph says this probably won't be in Debian Jessie
            Jeremy suggests trying requests, and falling back to Daniel's code if it fails (should just be a simple try clause)
                Daniel concurs


Jeremy is trying to build libcoin on Yosemite; having some issues with cmake not finding openssl
    Cassini will try to help debug tomorrow
Cassini asks if we should release 0.3.80 for OS X or wait for Namecoin Core
    Jeremy says probably release, if it's not much effort

Single-Click-Install .Bitcoin

Cassini asks what status is
Jeremy wants to make a Namecoin Bundle Installer that installs Namecoin-Qt, NMControl, and other stuff like ANTPY/DyName/NameID
    Daniel concurs
Jeremy needs to talk to Phelix about Innosetup vs NSIS... Bitcoin is using NSIS, while Phelix is using Innosetup

Web sites

Cassini notes that is down
    Jeremy's VPS is having issues, will hopefully be back soon (might be after Zach's grant is finished)

Light Clients

Joseph is looking at porting Multibit
    Question on whether to use Multibit Classic or Multibit HD
    Looks like they're using a fork of BitcoinJ that isn't upstreamed yet, which could complicate things
        Jeremy would prefer not to have any more forks than we already have if we can avoid it
            Joseph will ask the Multibit team about when the fork might be upstreamed

fluffypony asks which RR types can be stored in a Namecoin record
    Hugo answers


fluffypony asks what to do about devices like non-jailbroken iOS devices who want to use .bit
    Jeremy suggests something similar to how Tor handles it
    Jeremy notices that there are licensing issues with LGPL and iOS, meaning it's unclear if NMControl can be on iOS
    Jeremy also notices that EFF's analysis of the iOS developer agreement is *very* negative
        We couldn't release on Cydia if we use Apple's SDK
        We would also be under an NDA regarding Apple's SDK
        Jeremy considers these dealbreakers
            Hugo concurs
namecoind errors

fsmv suggests improving errors in namecoind
    e.g. searching for a name that isn't registered gives a cryptic error, not "name not yet registered"
    Jeremy concurs, suggests filing a GitHub issue for it

Who Will Work On What:

fluffypony:     DNS seed
fluffypony:     test OS X Namecoin Core Gitian builds
Joseph:         try faketime on Armory Gitian
Joseph:         Armory / Namecoin Core compatibility
Daniel:         GSoC registration
Jeremy:         finish cleanup of GSoC ideas page
Jeremy:         improve security of Daniel's NMControl REST pull requests
Cassini:        try to debug libcoin OS X build
Cassini:        try to release Namecoin 0.3.80 on OS X
Jeremy:         talk to Phelix about Innosetup vs NSIS
Jeremy:         fix VPS
Joseph:         ask Multibit team about BitcoinJ fork and porting to Namecoin
fsmv:           file GitHub issue regarding namecoind error messages

Re: #namecoin-dev Meeting Recap/Notes

Posted: Wed Feb 11, 2015 11:58 am
by phelix
Thanks for the notes, very helpful.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Sun Feb 22, 2015 10:08 am
by biolizard89

Code: Select all

#namecoin-dev meeting notes, 2015 02 21



Namecoin Core

Phelix has forked Daniel's Namecoin Core repo to the namecoin user on GitHub.  Development will proceed there.
Daniel says we could use more DNS seeds (we only have 1 for mainnet, and 0 for testnet).
	Also Daniel says we could use some hardcoded .onion links.
	fluffypony got a DNS seeder working on IPv4; currently working on IPv6
Jeremy and Joseph tested the Linux Gitian descriptor; got same hashes, and Jeremy successfully ran the resulting x64 binary.
	The Gitian descriptor needed minor changes to avoid colliding with Bitcoin's output filenames; Jeremy will send a PR shortly.
	Still need to try Windows and OS X
Daniel has extended the getblock output to to represent auxpow objects in JSON.
Luke-Jr informs us that TLS support for Bitcoin Core's RPC port will probably be removed.
	Luke-Jr says this is because it's not useful over trusted networks, and Bitcoin Core's RPC implementation isn't safe or recommended over untrusted networks
Hugo proposes a name_sync RPC command for allowing external databases to sync with Namecoin Core's name database.
	Daniel suggests that this would potentially be useful for Bitcoin as well, maybe merge upstream.
	Daniel mentions that Bitcoin Core supports a notify command feature for new blocks
		Jeremy suggests that NMControl should add support for this, so that the name cache refresh rate is more intelligent than every 30 minutes.
		Jeremy suggests a namenotify command
	Hugo says that a transactional method is more robust than the current Bitcoin Core notify commands.
	Hugo will write a proposal for Bitcoin Core.
indolering asks how many miners are on Namecoin Core at the moment.
	Daniel writes a script and finds out: In the last 1000 blocks, we had 536 with version 1 and 464 with version 2.
brand0 asks about the repo structure (right now Namecoin Core is in namecoin/namecore and NamecoinQ is in namecoin/namecoin).
	Daniel suggests renaming the repos, so that Namecoin Core is namecoin/namecoin and NamecoinQ is namecoin/namecoin-legacy.
		Jeremy and brand0 concur.


Joseph and Jeremy got Armory building reproducibly on Linux
	Dependencies aren't built, they're just pulled using apt-get, so if the Ubuntu repos update, the hashes might change.
		fluffpony suggests installing from a signed .deb, which will be more deterministic than using apt-get
		Jeremy mentions that Debian is aiming for reproducibility, so if Gitian supported Debian guest, that would be helpful long term.
		Joseph asks if it's particularly difficult to do Debian guests in Gitian.
		Jeremy isn't sure, but recalls that Mike Perry mentioned it in his 31C3 talk and it's apparently difficult.
		Joseph will try to make Debian guests work.
			Later Joseph informs us that vmbuilder only supports Ubuntu, so some alternative like grml-debootstrap would have to be used instead.
	Joseph says Gitian for Windows or OS X will be much trickier.
Armory 0.93 was released roughly during this meeting.
	Supports Bitcoin Core 0.10 with headers-first, so Namecoin Core's blockchain should be easily supported as well.
Joseph noticed that there's a branch involving Py3/Qt5
	Joseph asked Doug from ATI about it on the Armory forum; Doug says it's just an experiment for now.
	Jeremy says this has relevance to Gitian builds if we're going to be building dependencies.


Daniel has registered us for GSoC, hopefully we'll be accepted as a mentor organization.
	We'll hear back on March 2.
Ideas page is basically finished; we're waiting on Ryan and Phelix to confirm which projects they can mentor.  (Ryan confirms he can mentor a project.)

NMControl and REST

Jeremy will stop testing against the REST TLS feature that Luke-Jr says will be removed soon.
Jeremy will submit a pull request for REST support soon.


Jeremy fixed the libcoin build issue on Yosemite, turns out that cmake is broken on OpenSSL 1.0.2; downgrading OpenSSL fixes it.
	Cassini suggests posting an issue on libcoin's GitHub, since this isn't documented anywhere by libcoin.
Jeremy asks what the status is on getting an 0.3.80 OS X release out there.
	Cassini can build the release; we need Phelix or Ryan to upload it.

Windows installers

Jeremy says that phelix appears to be fine with using NSIS for new Windows installers (same as Bitcoin).
	Hugo thinks the NSIS compiler works better on Linux than Innosetup does.

Light clients

Joseph will try to do a Multibit port as a GSoC project.
Jeremy is working on a blogpost about the FN36-UNO-S libcoin client; draft should be ready for review soon.
kefkius is working on an Electrum port.
	PoW isn't properly handled yet, so not secure enough for standard use yet.
Hugo has a proposal for updating the wire protocol with a getname command.
	Hugo asks if the txid is stored in the name index; Daniel says it's present.
	Hugo asks if it's safe to assume that txid's are unique; Daniel says it should be fine.
	Hugo notices that the getname command response needs to contain the txid as of 12 blocks ago -- but those are already pruned by the full node by default.
		Hugo will ask on the forum about this.


Jeremy asks what people impressions of Lighthouse are.
Cassini asks if any non-IT people know Lighthouse.
Jeremy says that right now, probably very few -- but most Namecoin donors are probably technically savvy and familiar with Bitcoin.
Jeremy lists 3 advantages that Lighthouse has over BountySource.
	BountySource takes a 10% cut of awarded bounties.
	BonutySource is subject to AML rules because they hold funds.
		"As part of the cash out process we require a full name, address, and email address. We may also require that you fill out a W-8/W-9 form for tax obligations (see "Do I have to pay taxes on bounties I collect?") below."
	BountySource could, hypothetically, censor us or seize funds.
Jeremy suggests a small-scale trial of Lighthouse to see if it meets our needs.
Unanimous consent to continue the Lighthouse discussion in the forum.

Who will work on what:

Jeremy:		Send PR with fixed Namecoin Core Linux Gitian descriptor.
Jeremy: 	Try building Namecoin Core in Gitian for Windows and OS X.
Hugo:		Write a proposal for a transactional sync mechanism for Bitcoin Core.
fluffypony:	Get IPv6 DNS seeder working.
Joseph:		Try to get Debian guests working in Gitian.
Jeremy:		Submit a pull request for NMControl REST.
Jeremy:		Post a GitHub issue for libcoin's build issue.
Hugo:		Ask on the forum about the 12-block depth issue for getname.
Jeremy:		Start a forum thread about Lighthouse.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Sun Feb 22, 2015 1:37 pm
by phelix
Thanks for the notes.
Daniel suggests renaming the repos, so that Namecoin Core is namecoin/namecoin and NamecoinQ is namecoin/namecoin-legacy.
Hmm this is difficult imho as normal users probably want the GUI. Maybe only a readme should live at /namecoin explaining the versions.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Mon Mar 30, 2015 10:51 pm
by biolizard89
Apologies for the delay getting these up.

Code: Select all

#namecoin-dev meeting notes, 2015 03 07



1) ... bug=779686
Friend or Foe for Namecoin?
Pro: Publicity for free
Neutral: Packaged software soon out of date. However, if users encounter problems they'll most likely discover and then download the latest version.
Con: Maintainer can add iffy code, as happened to Bitcoin, see

2) Possible memory leak in Namecoin-Qt. Observed by breathingdog in OSX. Same for MS-Windows Namecoin-Qt? How to detect memory leaks in Windows 8?

3) We should motivate Daniel to go ahead with the PoW-secured API query approach, see

4) st12out_of_range pop-up window. An incompatibility phenomenon to keep an eye on.

5) User samurai777 reported difficulties with merge-mine NMC setups. While he may have mixed up a few technical aspects, we should discuss the main question that evolved: which merged-mining pools are recommended for new miners, i.e. easy to setup, reliable long-term operation and acceptable variance?
Eligius: a bit of a challenge for newbies
BTCGuild: easy setup, around since 2011, but mine empty NMC blocks and not prepared to update namecoind
mmpool: easy setup, around since 2011, but huge variance
Discus Fish?

6) Watch jbisch's Armory Gitian port. Should be ready around March 15th.

Code: Select all

#namecoin-dev meeting notes, 2015 03 21


Phelix (participated via email, due to IRC technical issues)


Jeremy's REST backend code is still undergoing review by Phelix.
    Phelix is getting a weird PyOpenSSL error that Jeremy can't reproduce, and which doesn't make any sense given the documentation.
    Jeremy suspects it may be specific to Windows (which Phelix uses), or the specific version of PyOpenSSL that Phelix is using.
    Cassini suggests that maybe it's an OpenSSL 1.0.2 issue.
    Jeremy will follow up with Phelix.
Jeremy has been working on a Bottle HTTP server plugin for NMControl.
    Basically works in terms of functionality.
    However, security issue: a malicious website could issue an HTTP request on localhost and call an NMControl method.
        Same-origin policy will prevent the website from getting the response, but we don't want NMControl doing anything stupid as a result.
        "Anything stupid" would include starting/stopping plugins.
        Might be additional security-sensitive methods added in the future.
        Jeremy suggests making each plugin provide a whitelist of methods that can be called by HTTP.
        This would need to be updated whenever we add more public methods to a plugin later.
        Joseph notes that with a blacklist, a plugin developer might forget to blacklist dangerous methods.
        Jeremy agrees, says that it's better for a method to not work, than for a dangerous method to work when it shouldn't.
        Cassini and Joseph concur.
    Jeremy will add the whitelist and submit a PR.
Phelix has a staging branch for some of his proposed changes to NMControl, called Hyperion.
    Phelix requests testing/feedback, as he hasn't gotten much feedback on those changes yet.


Doug from ATI mentioned the Py3/Qt5 port again on March 12.
Joseph now has an internship with ATI, working on Gitian support -- congrats Joseph!


Jeremy asked Daniel yesterday if he was interested in porting Namecoin-Qt's GUI to Namecoin Core.
    Daniel said that he didn't think the task was a good fit for him, and he preferred that effort be spent on other GUI's.
    Daniel is fine with answering questions for anyone who wants to do the port.
Jeremy notes that Bitcoin Core and Armory appear to be the best Bitcoin GUI's at the moment.
    Joseph agrees.
    Jeremy strongly opposes coding a custom Namecoin GUI from scratch, as that should be done by Bitcoin developers.
    Cassini believes Armory is the most promising option for a Namecoin GUI.
    Jeremy has $400 + 0.42 BTC available for bounties / contract work, asks how much would be reasonable for Namecoin Armory (display names, and issue name transactions).
Jeremy suggests that a future GUI (e.g. Armory) should, by default, purchase a d/ name and a dd/ name at the same time, 
automatically make d/ import from dd/, and move the d/ name to a different wallet after registration.
    This way users are encouraged not to do stupid things with keys that can steal a name.
    Cassini asks if moving between wallets could cause additional anonymity issues.
    Jeremy agrees that this has to be done carefully, but users could create a new wallet for those names via Armory GUI.
    Also Armory has Coin Control, so idenitity linking is less of an issue.
    Cassini and Joseph think this sounds okay.
Who will work on Armory?
    Joseph can work on it some, but is relatively busy with school and ATI internship.
    Jeremy doesn't have any experience with the Armory codebase, so this isn't a great fit for him.


Cassini asks if there is any documentation yet on anonymity-breaking mistakes with Namecoin.
Jeremy believes there is not yet.
Jeremy suggests contributing Namecoin-specific code to libitpriv by Open Bitcoin Privacy Project once their code is released.


Cassini asks if we should encourage a pay-to-name feature in Monero.
Jeremy points out that Daniel did contract work for Isracoin, and no one gave us grief about it.
Cassini says that transparency also helps.
Jeremy ACKs talking to the Monero devs.


Cassini will try to contact owner of
Cassini asks if there are other options for a stable block explorer.
Jeremy says that if open-sourced their code, it would be a good contender.
Cassini says that is FOSS -- only thing is that it's in Ruby, and most Namecoin devs don't code in Ruby.
Cassini will talk to phelix about a possible bounty -- Jeremy is willing to fund part of it if phelix is okay with it.

Meeting Logistics

Jeremy notes that 5PM UTC is now an hour later due to daylight savings time.
Jeremy asks if we should move to 4PM, or stay where we are.
Cassini says Phelix might prefer 5PM.
Jeremy will check with Phelix on this.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Sun Jun 07, 2015 7:21 am
by biolizard89
Here are the notes from June 6's meeting:
#namecoin-dev meeting notes, 2015 06 06

# Present:


# Funding

Bounties funded by NMDF and FSM are public, thanks to Phelix
We already awarded one bounty.

# NMControl REST Backend

REST Backend has another reported HTTPS problem from Phelix
Jeremy hasn't yet confirmed the bug.
HTTP works fine
Jeremy suggests merging as-is and debugging later, since almost no one uses HTTPS for the namecoind RPC port
Cassini and fluffypony concur
Jeremy will give 48 hours for a NACK on GitHub, and merge otherwise

# NMControl DNS Server

Hugo and Jeremy have decided that it's a better idea to use PowerDNS's Remote backend to talk to NMControl's REST frontend, rather than using Unbound to talk directly to NMControl via the Python module.
This would reduce implementation complexity, eliminate the need to build Unbound's Python module on Windows, and allow us to use non-Python languages in the future if we want.

# Jekyll port of

Jeremy has been handling this.
Current status at
Home/exchange/team pages are working.
Home page needs some review since it's slightly changed from the public site.
Jonas expresses endorsement of the Jekyll port, due to lower threshold to make changes

# Video tab on

The video on is not made by us, and advertises a third party website.
That third party website has misleading information, and links to a scam application
Jeremy suggests removing the video until those issues are dealt with.
fluffypony and Joseph concur.
Jeremy will remove the Video tab from the Jekyll rewrite, so when it goes live, the video will disappear.
Jeremy doesn't think it's worth editing the PHP-based repo.

# Anonymity documentation

Jeremy says that while it's well-known that Namecoin isn't anonymous, some people seem to be misconstruing that to mean that Namecoin isn't useful for Tor-related things.
Use cases include location anonymity of name owner (which we have no problem with if Whonix is used) and lookup anonymity (which we'll be fine on once we have proper TorBrowser support).
Jeremy proposes documentation explaining these use cases, how they differ from overall anonymity use case (which we don't support), and why Namecoin is useful for Tor despite not being anonymous.
Jeremy also proposes an experiment for fully anonymous name purchases: mine a CPU-minable coin for some time, trade for NMC using ShapeShift, and buy a name, all from within Whonix.

# Linux packages

Jeremy is concerned that the pubkeys for Linux OBS packages are distributed over HTTP.
Jeremy asks if Peter could post the keys on so we can provide them.
Joseph points out that OBS might hold that key.
Joseph mentions that he's beeing working on packaging Namecoin Core for Debian without OBS.
We'll ask Peter about it.

# TLS Firefox extension

Jeremy has tentative approval to use the extension as his midterm project in his Network Security class.

# Tor/I2P proxy

We had been planning on using mitmproxy as the proxy between the browser and Tor/I2P.
(Originally because it did TLS interception, which we've scrapped.)
Yawning from Tor has just released a SOCKS proxy specifically designed to sit between TorBrowser and Tor/I2P.
Built in support for stream isolation, basically everything we need for Tor/I2P support.
Written in Go; Jeremy doesn't really use Go.
Hugo does use Go.
Hugo will look at the proxy; Jeremy will provide details.

# d/ spec

Hugo and indolering and Jeremy have been working through the review process.
3 things remaining:
(1) allowing import to limit types that are imported
(2) getting Ryan's feedback on TLS
(3) getting Daniel's feedback on import
Also figure out what we're doing about .onion signing .bit.
(Jeremy has a message pending with the GlobaLeaks guys asking exactly what they need here.)
Everyone who hasn't read the spec yet -- please review it ASAP.

# Freenode and Tor

Freenode isn't very Tor-friendly, Jeremy asks if there's any interest in switching servers.
OFTC is more Tor-friendly, but their TLS cert isn't trusted.
Hugo suggests staying at Freenode for now.
Jeremy concurs.

# Windows installer

Hugo has a Windows installer that installs Unbound/ncdns/namecoind.
Currently forwards normal DNS to Google public DNS.
Requires manually setting DNS settings.
Using WiX now.
Jeremy suggests that NSIS might be better suited for reproducibility.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Sat Jul 18, 2015 7:38 pm
by biolizard89
#namecoin-dev meeting notes, 2015 07 18

# Present:


# NMControl REST Backend

Phelix is concerned that using urllib3 (which is used by requests) might make Tor support more difficult.
However, Phelix's suggestion of urllib2 requires monkeypatching, which Jeremy describes as hacky and potentially unsafe.
urllib3 can reportedly be monkeypatched too. ... t-11360808
urllib3 is willing to accept pull requests for native SOCKS support (and has been working on it).
urllib3 is more actively developed (26 commits between July 1 and July 18 2015) compared to urllib2 (20 commits in past 2 years).
Consensus: stay with requests/urllib3 unless new issues develop.

# SPV Client

A Reddit user is interested in working on an SPV client. ... s_resolve/
We've reached out to him/her, and are awaiting a reply.

# Namecoin Core

Cassini is working on improved instructions for migrating NamecoinQ to Namecoin Core (particularly regarding blockchain format).

# Mining

Cassini is working on documentation of setting up merged mining for pools, to try to get more miners doing Namecoin.
Cassini had a few questions on setting up a mining system.
Jeremy suggested asking Luke-Jr or the F2Pool operators if questions arise.
Joseph suggests porting Testnet-In-A-Box to Namecoin.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Mon Jul 20, 2015 7:48 am
by phelix
Thanks for the notes. [discussion about method of communication moved to new thread]

Re: #namecoin-dev Meeting Recap/Notes

Posted: Sat Aug 15, 2015 8:35 pm
by biolizard89
#namecoin-dev meeting notes, 2015 08 15

# Present:


# OpenSSL Consensus Bug

100% version 3 in past 1000 blocks as of 2 days ago.
Jeremy plans to write a blogpost summarizing the bug and our response to it.
0.3.80 clients are on semi-SPV security and should wait for extra confirmations.
Namecoin Core clients are on full security.
Both 32-bit and 64-bit Namecoin Core are now safe for mining.

# Mining

Cassini noticed that Antpool merge-mined around February to April and then stopped.
Cassini also noticed a TangPool in May, which didn't min a lot of blocks.
Cassini wonders if we have anyone in touch with AntPool.
Cassini and Jeremy think snailbrain might.
Squidicuz is interested in making merged mining sharechains work in P2Pool.
Jeremy suggests asking forrestv what changes are needed to support this.
drazisil is also interested in this.
Squidicuz and drazisil will work on this and let everyone know how it goes.

# Website

Jeremy is going to try to finish up the Jekyll port of
Jeremy is hoping more people can ACK the reworded main page (Phelix has ACKed it).
Cassini will take a look at it.

Re: #namecoin-dev Meeting Recap/Notes

Posted: Mon Sep 07, 2015 5:25 pm
by biolizard89

Code: Select all

#namecoin-dev meeting notes, 2015 09 05

# Present:


# Namecoin-Qt

Jeremy asks whether anyone is interested in moa's bounty for porting the Manage Names tab to Namecoin Core.
    No takers.
Drazisil mentions outdated Namecoin Core readme.
    There's a pull request pending from Ryan Singer which should improve this.
    Jeremy will try to get that pull request merged.

# Website Migration

Jeremy has redirected to's FreeSpeechMe page.
    Jeremy will clean up the content from the old site later.
Jeremy suggesrs explicitly marking the FreeSpeechMe GitHub repo as deprecated.
    New Firefox extension is coming soon.
    Convergence codebase doesn't work on current Firefox releases.
    Jeremy doesn't want random people finding the repo and thinking it's usable code at this point.
    Drazisil and Cassini concur.
Jeremy has dumped into Jekyll format.
    Still need to merge it into the Jekyll site and verify that everything came through okay.
Cassini will file 1 or 2 wording issues on GitHub Sunday or Monday.
Drazisil suggests taking cues from

# Mozilla

Mozilla is deprecating XPCOM.
    This affects the Firefox extension.
    It's not clear whether Mozilla will support our use case with a new API.
    Mozilla gave Jeremy the run-around when he asked.
    Shouldn't affect Firefox end users for about 11-17 months; gives us time to come up with a new strategy.
Hugo and Jeremy talked to the Pale Moon developers.
    Pale Moon does not plan to deprecate XPCOM.
    Pale Moon devs are pretty sure they don't support the API that we're using, due to it being introduced to Firefox after Pale Moon forked.
    Jeremy is not confident in Pale Moon's security budget, given that their downloads are over HTTP.
    Jeremy isn't sure what Pale Moon's market share is, but guesses it's much lower than Firefox.

# BountySource

BountySource emailed Jeremy this week; Salt is no longer invite-only.
We can set up a fundraiser at our convenience.
Salt supports recurring donations in USD.
    Jeremy will ask BountySource if recurring donations are possible in BTC, and if so, how that's handled.

# OBS Linux Packaging

Peter posted a signed repo key on GitHub.
We should add this to our downloads page.
HTTPS would still be better.
And of course we should start using Gitian.

# Gitter

Freenode doesn't allow Tor; this sucks.
Phelix reports that Gitter works fine over Tor.
Jeremy suggests running a bridge bot between Freenode and Gitter.
    That way people can choose what they want to use.
    Unregistered users can use Freenode.
    Tor users can use Gitter.
    Registered clearnet people can use either.
    Joseph and Drazisil concur.

# Armory

Joseph merged upstream changes from past half year into Namecoin branch.
Joseph asks if he should push to the Namecoin fork.
    Jeremy says yes.