Year 2011

We implemented and released libpsyc: The lightning fast PSYC library in C (see the benchmarks how PSYC compares to XML- and JSON-based protocols).

We finalized the PSYC 1.0 new syntax.

We are experimenting with an F2F (enhanced peer-to-peer) version of PSYC to maximize user privacy (see also Saikound).

Year 2010

We combined PSYC with RTMP for high scalability web-based video chats.

Year 2007: ...

See News.

Year 2006: the web via PSYC

Wizards of OS

See WOS.

Decentralized Social Network & Web of Trust

At a certain point el and lynX sit in the park and decide that it is time to get certain things working, that we had been thinking about forever. Within a few months el gets the web of trust working while I hack up the HTML profile features that make psyced a truly decentralized Social Network. Later Monkey and lynX even fix up this HTML to be delivered to PsycZilla via PSYC protocol, not HTTP. We get to experience watching a web page with a psyc: URL in the location bar. And it is real. Breathtaking.

PsycZilla: Web 2.0 here we come

Monkey joins our team and revolutionizes the way we have been thinking about Clients. His PsycZilla firefox extension and XUL application looks fabulous, has spectacular potential by its mozilla integration and grows better each week, as it is a rapid development dream come true. We still have a huge list of things to do to make it a real PSYC clients with stream multicasting and P2P abilities, but it's great we are getting even close.


For all the missing years, see also the history page on which is obviously tightly coupled with PSYC's history.

Year 2000: symlynX

See symlynX.

Year 1999: politik-digital

See symlynX.

Year 1998:

See wahlkampf98.

Year 1997: PSYC in action

The Stern Webchat instantly makes it to the 5th place in the popularity of webchats in Germany. Unfortunately the SGI it was running on started to show problems with the load. I got nervous if there was something wrong with my work, but it turned out there was a bug in IRIX networking which only got fixed the following year!

Year 1996: We need a webchat

The CMS for that I had hacked up in a very stressful month was stable and Gruner+Jahr was producing pretty much all its web sites with it. Since the websites were running fine I went up to my boss and said we need a webchat. And since I was already known for being a chat person I was granted to do the webchat as a research work, leaving all the rights to me and the agreed intention to release the PSYC parts of it as open source. Gruner+Jahr would merely enjoy a royalty-free license. Thus I started developing a PSYC based webchat.

Year 1995: Specification

lynX publishes Internet Drafts for the PSYC protocol on

Mark Pesce, the man who made VRML a reality, was inspired by PSYC, but PSYC wasn't ready yet.

Year 1994: binary PSYC

I implemented a PSYC prototype in C as a Studienarbeit, using a binary version of the protocol. The implementation had the funny detail of being downward compatible with an ancient messaging protocol called MSP. I spent too much time getting the data structures straight (hash tables blah blah), that was annoying. And the binary encoding was efficient but not flexible. Out of that experience arose the principle to allow inheritance as much as possible.

Also it struck me that Tim's idea of uniform resource locators was the right thing to do for human identificators, too. Consider that at the time it was normal to address users at an IP address, so coming up with the idea of having a server that will act as a relay and answering machine was an important step to take to move forward in the design of the protocol.


Year 1991: Nemesis

I wrote an IRC#IRC server variation that didn't make it. But the dreadful experience of the IRC War made me mentally move on and seek a technological solution which would not require oligarchic political structures as IRC does, and address all the well known issues.

At the same time the Nemesis MUD was going strong. I rewrote half of the backend library at the time, which made me very fluent in LPC, event-oriented systems and object orientation.

Year 1990: /me

I published an improved IRC-like protocol called Cafe. At the same time I added a lot of things to ircII, one of the least important ones was a command I chose to be /me.

I also briefly ran a MUD called ChatLand, but later merged it into Nemesis.

Year 1989: Finally on Unix

Actually we started learning unix the year before, but on machines which weren't connected to the university network (can you imagine!?). In this year we got our logins on graphical unix workstations. At the same time the Internet got through to us, and suddenly the policy of network usage being a crime and strictly for research was dropped. I looked at the IRC protocol and went "hell, this can be done in a better way." Still, IRC was our greatest playground and I spent time working both on ircd and ircII source code.

Year 1988: Servers are forbidden

With the VM Shell I implemented a P2P chat system. It just turned out that way because our university was keeping us from running server applications.

Year 1987: Chatting is for hackers

It was Maex who first showed me the Bitnet Relay. I was instantly amazed. Too bad that using the network for casual conversation was considered a crime at the time, at least at my university. I spent some time on a BBS chat system later that year.