PaperCut

PaperCut is an NNTP server written in Python.

http://pessoal.org/papercut/index.php

Its main objective is to integrate existing web based message board software (Phorum on this case) with an Usenet front-end. However, its extensibility enables developers to write their own container for the storage of the Usenet articles (messages). That means that the code is extensible enough that you could write new containers to integrate the news server with other web message board projects or even other ways to store the messages. Jon Udell and Michael Sippey have sometimes wondered whether the replication mechanism of NNTP should be applied to the BlogWeb. Would widespread availability of a lite NNTP server make this model more viable?

Hi, I'm the author of the Papercut project. Thanks for the pointer by the way :)Care to give more details about how an NNTP server could be applied to weblogs ? I would love to hear more.

reply (Bill Seitz, Mar13'02):

• Michael Sippey's thoughts are here .

• Jon Udell has used newsgroups as an Intranet repository much like a weblog (or, as John Robb would call it, a k-log).

• If all writers and readers of weblogs each had their own NNTP server, then

  • each weblog would be its own newsgroup. A writer would have each weblog entry be a UseNet posting.

  • readers would have their NNTP servers subscribe to the newsgroups/weblogs they wanted to track. Efficient replication comes for free.

  • Weblogs could then be read offline.

  • Each blog entry would be the basis for a UseNet Group Discussion, which would replicate across servers.

(Joao Prado Maia)

Hmm, interesting indeed. It sounds very doable, but the problem now is related to how the Usenet articles would be stored on the user machines. As you know, right now Papercut is only bridging messages posted on Phorum which are stored on a simple MySQL database. However, to implement something like a distributed network of Papercut servers each running on the user's machine, the requirement for a MySQL database would be a little too much.

I wonder if storing the messages in XML format (or even plain unix mailbox format) and having a simple dbm database to index the files would be good enough. All of that without mentioning the requirement of the users to install Python, but I guess Gordon Mc Millan's Installer stuff could be used to create an .EXE files for Windows users.

Very interesting stuff, I will keep brainstorming about this.