GNUnet

We spoke at length with Christian Grothoff of `GNUnet`__. He has moved himself and the project from TU Munich to `Inria`__ in France. He has a large number of `open positions`__. This is a great opportunity to get paid to work on open source anonymity tools, we encourage everybody to contact him about it.

The prospect of an invigorated GNUnet with a large amount of new funding is quite interesting. We discussed more ways to work together. In early 2014, we worked hard to understand the GnuNet DNS replacement, but we were unable to figure out a good fit for it in I2P. One of his new ideas is a distributed, anonymous statistics gathering subsystem, for detecting problems or attacks on the network. We'd definitely be interested in that.

We also discussed the `Special-Use Domain Names of Peer-to-Peer Systems draft`__. A new, greatly simplified version 3 was posted in December. The prospects for approval remain unclear. The best way to monitor or participate in the discussion is via the `IETF DNSOP WG mailing list`__. We will attempt to do so on our side, and also give Hellekin a new point-of-contact for this topic.

We apologized to Christian for not being organized enough to have a talk at his `We Fix The Net assembly`__. One of our biggest failures as a project is our seeming inability to submit talks at conferences. We'll have to do better in the new year.

Debian

Iain Learmonth, ein Debian Teilhaber, hat vorbeigeschaut. Er möchte I2P, zusammen mit anderen Anonymiserungstools, in sein neues Debian "Superpaket" packen und fände es großartig das noch 2015 zu bewerkstelligen. Er versicherte der Ablauf sei jetzt ganz leicht, folgt einfach der anleitung. Wir meinten das sei lustig weil wir in dem Ablauf schon über 7 Jahre feststeckten.

He said, well, try the new process, it works great, should be no problem at all if your package is in good shape. The people in Debian that run this process are eager volunteers who want nothing more than to get more packages in. We said our package is indeed in fantastic shape, and we would try out the new process as soon as possible. If all this is true, we will be in the next Debian release in late 2015. This would be very very cool.

Tails

Wir hatten eine nette Diskussion mit BitingBird von Tails. Sie sind sehr glücklich über unser schnelles Reagieren auf die `Angreifbarkeits Offenlegung´ letzten Sommer, welches in unseren 0.9.14 release gemündet ist. Unsere Angreifbarkeit war ursprünglich Tales zugesprochen worden die sich dadurch und ebenso durch die fehlende vertrauliche Behandlung des Themas von Seiten Exodus Intel `stark angegriffen´ fühlten. Wir dankten ihnen dass sie, auch für uns, in den Ring gestiegen sind und eine passende Antwort gegeben haben.

BitingBird also handles support, and she tells us the number one issue is how long I2P takes to start up and be useful for browsing I2P sites. Her standard answer is "wait ten more minutes" and that seems to be effective. I2P is particularly slow to startup on Tails since it does not persist peer data by default. It would be nice to change that, but there's also things we can do on the I2P side to make things start faster. Expect some improvement in our 0.9.18 release.

Onioncat

Ein langjähriger Freund von I2P, Bernhard Fischer von `Onioncat´, hat vorbeigeschaut. Die zukünftigen Veränderungen bei Tor Hidden Services bedeuten dass ihre Schlüssel nicht länger in einen Block einer IPv6 Adresse passen, er arbeite aber an einer Lösung. Wir erinnerten ihn dass dies schon immer der Fall für I2P war (mit "GarliCat"), das ist keine neues Problem. Er zeigte uns `eine Präsentation´ seines Plans. Er beinhaltet das Speichern eines extra Eintrags im hidden service Verzeichnis (equivalent zu einem I2P network database leaseset). Es war nicht ganz klar wie das funktionieren würde, oder ob wir es als Missbrauch des netDb betrachten. Wir werden das mt ihm weiterverfolgen sobald es darüber Neuigkeiten von seiner Seite gibt.

Neue Benutzer

Neue Benutzer
---------

Wir haben Stunden über Stunden damit verbracht den Leuten, die an unserem Tisch Halt gemacht haben, I2P zu erklären. Manche hatten von I2P schon gehört, andere nicht; Jeder hat aber von Tor gehört und hatte zumindest eine wage Idee was hidden services sind. Wie üblich, den Leuten I2P näherzubringen ist schwierig. Am Ende des Kongresses kamen wir zur Überzeugung dass das Problem in Unterschieden in der Terminologie liegt. Vor 12 Jahren, als I2P und Tor beide an den Start gingen, führte jeder von uns bestimmte Termini, für die jeweiligen Bereiche unserer Systeme, ein. Heute ist die Tor- Terminologie, wie etwa "hidden services", verstandenes und anerkanntes Gemeingut. Die I2P Terminologie, wie zum Beispiel "eepsite" ist es nicht. Wir sind übereingekommen dahingehend unsere Unterlagen, router console und weiteres auf Möglichkeiten zur Simplifizierung hin zu überprüfen und allgemeingültige Begriffe zu verwenden.

I2P project topics

• Spending money: We discussed several ways to effectively use our resources in 2015, including more hardware for testing and development. Also, we plan to increase reimbursement levels for conference attendees.
• Toronto meetup: CCC is such a productive time for us, and it seems that a second meetup in the year would be quite helpful. We have proposed it for August 2015 in Toronto, Canada, in conjunction with `Toronto Crypto`__. It would include developer meetings together with presentations and tutorials, all open to the public. We are attempting to gauge interest and research possible venues. If you are considering attending, please let us know by `tweeting @i2p`__ or posting `on the dev forum thread`__.

• We discussed Meeh's workload and the state of the various services he is running. We made some plans to reduce his load and have some other people help out.
• We reviewed our critieria for placing links to `i2pd`__ on our download page. We agreed that the only remaining item is to have a nice page on the `Privacy Solutions web site`__ or elsewhere with binary packages for Windows, Linux, and Mac, and source packages. It's not clear who is responsible for building the packages and where the "official" version is. Once there's an established process for building and signing packages and an official place to put them, we're ready to link to it. If it is not feasible to host it on the Privacy Solutions website, we will discuss alternatives with orignal, including possible migration to our download servers.

• Lots of people coming by the table asked if we had a non-Java version. It was great to finally answer "yes" and we're eager to get the word out and get more users, testers, and developers on it.
• `Vuze`__ continues to make good progress on their I2P integration. We look forward to working with them in the new year on a managed rollout to more users.

• We discussed the state of Meeh's and Sindu's reseed servers. They made several improvements while at the congress and are investigating migration to `Matt Drollette's Go implementation`__. The security and reliability of our reseed servers is vital to new users and network operation. `User 'backup'`__ is doing a great job monitoring and managing the pool of reseed servers.

• We agreed to purchase a second root server for development, testing, and services. Echelon will be adminstering it. Contact him if you would like a VM.
• We reiterated that we have funds available to purchase test hardware, especially for Windows and Mac. Talk to echelon for details.
• We met with Welterde about the state of his services including his `open tracker`__. These services are not being adequately maintained and will soon become inaccessible due to crypto changes if they are not upgraded. He committed to upgrading them soon.

• We met lots of people interested in our `Android app`__. We passed several ideas and bug reports back to str4d. We plan to make a big push to give the app some development love early in the year.

• Regrettably, we didn't get to see too many talks at the Congress, as we were so busy meeting with people. We plan to catch up and `watch them online`__. As usual, Tor's "State of the Onion" talk was excellent, and Jacob's talk was great. We hear that the cryptography talks were good as well.