|
DataMuseum.dkPresents historical artifacts from the history of: DKUUG/EUUG Conference tapes |
This is an automatic "excavation" of a thematic subset of
See our Wiki for more about DKUUG/EUUG Conference tapes Excavated with: AutoArchaeologist - Free & Open Source Software. |
top - metrics - downloadIndex: T r
Length: 9737 (0x2609) Types: TextFile Names: »real.tex«
└─⟦2d1937cfd⟧ Bits:30007241 EUUGD22: P.P 5.0 └─⟦dc59850a2⟧ »EurOpenD22/pp5.0/pp-5.tar.Z« └─⟦e5a54fb17⟧ └─⟦this⟧ »pp-5.0/doc/manual/volume0/real.tex«
\input{dist-params} \chapter {Overview} \section {What is PP} PP is a Message Transfer Agent (MTA), which supports a number of Message Transfer Protocols, including: \begin {itemize} \item X.400 (1984) P1 \cite{MHS}. \item X.400 (1988) / ISO 10021 P1 \cite{CCITT.MHS} \item A number of RFC 822 based protocols \cite{RFC822}. \end {itemize} It is based on experience with previous MTAs, and has the following aims: \begin {itemize} \item Use for switching large volumes of messages in a service environment \item Management features, suitable for complex sites. \item Protocol conversion facilities, particularly for mapping between RFC 822 and X.400 according to RFC 987 \cite{ARPA.MHS} and RFC 1148\cite{RFC1148}. \item Body part format conversion. \item Support for development of advanced User Agents, particularly those wishing to use X.400 and/or multimedia capability. \end {itemize} For a technical overview of PP, the paper by Kille is recommended \cite{IFIP.PP}. PP is implemented in `C', and runs on a range of UNIX and UNIX-like operating systems. The current release includes the following modules: \begin{itemize} \item X.400 (1984) P1 protocol \item JNT Mail (Grey Book) Protocol \item Simple Mail Transfer Protocol (SMTP / RFC~821) \item UUCP Mail transfer \item Distribution List expansion and maintenance. \item RFC 822 based local delivery \item Delivery time processing of messages \item Conversion between X.400 and RFC~822 according to RFC~1148 and RFC~987 \item Conversion support for reformatting body parts and headers. \item X-Windows based management console. \item Message Authorisation checking \item X.400 (1988) P1 protocol (beta test) \item X.500 based distribution list facility using the QUIPU directory (beta test) \end{itemize} \section {The PP Documentation} This document refers to PP version \ppversion/ of \ppdate/. The PP Manual is divided into 3 Volumes and this covering document, which introduces PP, and contains general information relevant to all of the other volumes: \begin {description} \item[Volume 1] Installation and Operations. This describes all that is needed to install PP and how to administer PP as part of an operational service. \item[Volume 2] Programmer's Guide. This describes programming interfaces for submission of messages into PP, and for channels which deliver messages locally or remotely. \item[Volume 3] User's Guide. In general, PP should not be visible to the end user. There are some associated tools, which are needed, and are described in this volume. \end {description} \section {A brief history of PP} PP began in 1985, when several of us wished to produce a system to support X.400. It was clear that MMDF (our favourite MTA at that time) was not going to be extensible to do all that we wanted. Steve Kille had looked at the EAN MTA at the end of 1984. Whilst this was an impressive development at the time, it did not really meet our needs --- although our initial plan was to use the OSI layers from EAN. Two meetings were held in September 1985 to consider the possibilities of future MTA development: \begin {itemize} \item At CWI, discussion took place between: Doug Kingston (CWI/BRL), Piet Beertma (CWI), Miriam Amos (DEC/UCB), Kevin Dunlap (DEC/UCB), Daniel Karrenberg (Uni Dortmund) and Steve Kille. \item A UCL, discussion took place between: Phil Cockcroft (UCL), Doug Kingston, Steve Kille and Julian Onions. \end {itemize} These meetings, and following messages, resulted in some of the initial PP design. As a result of the second meeting, Phil Cockcroft and Julian Onions started to write large chunks of code, and Steve Kille wrote some small chunks of code. By April 1986, a system was starting to come together, which was being called PP. Marshall Rose was starting work on his ASN.1 tools, which were clearly going to be a part of PP! In July, the mechanisms for representing multimedia messages in the queue were designed and evolved. By October 1986, it was time to fully design the QMGR. Steve Kille had been at the infamous CCITT Directory meeting in Munich in February, and so specified a QMGR design in the new ASDC (Abstract Service Definition Conventions). Phil Cockcroft left UCL at the end of 1986, which was a big loss to the effort. In February 1987, there was a basic system working, which talked SMTP and X.400. Filling in the holes took much longer than we expected! The first priority was to replace the shell script QMGR. Julian Onions was then working on the ISODE ROS tools (ROSY and POSY), and so it was decided to use ROS to implement the QMGR. The QMGR and the tools proceeded in parallel. In 1988 Alina daCruz (UCL) rewrote the access libraries, Pete Cowen (Nottingham) worked on a range of formatting tools, and Bruce Wilford (UCL) redid all of the channel binding and domain handling. Mike Roe (UCL) started work on LISP code to produce a first cut an the 1988 X.400 protocol. Discussions with Piete Brookes (Cambridge) led to the initial design of the JNT Mail channel. In October 1988, Steve Kille presented a paper on PP at the IFIP WG 6.5 conference, but there was still no PP ready for release. By the end of 1988, there was a PP which had been radically changed and looked much more like the final system. PP received its first public airing at CEBIT (Hannover Fair) in March 1989, where it carried ODIF over X.400, and converted into Diamond format by use of private conversion tools. It interworked with a number of X.400(84) systems at this point. This was as a part of the Esprit PODA project. In 1989, the X MTA Console (Pete Cowen) and authorisation (John Taylor) were added in, and the internal structures fully upgraded to X.400(88). PP 3.0 was cut in July 1989, and was used as the basis for UK to full X.400 relaying. This system was released outside UCL and Nottingham on a very selective basis. PP 4.0 was released at the very end of 1989 to 15 beta sites. Three more beta releases were made during 1990 to an increasing number of beta sites. During this time, the upgrade to RFC 1148 was made, and initial implementations of P1(1988) and directory based distribution lists (Colin Robbins) added. PP 5.0, The first openly available release, was cut in September 1990. \section {Etymology} PP is not an acronym. There is no truth in the rumour that PP stands for ``Postman Pat'' --- a famous British postman. \section {Acknowledgements} PP funding has come from a number of sources. In particular the UK Joint Network Team (JNT) has funded a significant part of the PP development. Jim Craigie of the JNT has been helpful and encouraging in this. The Alvey sponsored COSMOS project funded a major part of the initial development of PP. The Alvey sponsored Locator project funded work on PP to provide hooks for security features. Use of PP as infrastructure in the Esprit PODA project has led to many improvements in the current release. Prof. Peter Kirstein of University College London (UCL) and Dr. Hugh Smith of Nottingham University have been particularly tolerant of the excessive resources consumed by this development. The major work of coding PP (apart from the authors) has been done by Phil Cockcroft, Alina da Cruz, and John Taylor of UCL and Pete Cowen of Nottingham University. Code has also been written by Adrian Joseph, Mike Roe, Colin Robins and Bruce Wilford of UCL. Irene Hassel of UCL helped to write documentation. John Andrews of UCL has helped to bring PP into service at UCL. Piete Brooks of Cambridge Computer lab has been a faithful alpha test site, and has contributed code to integrate UNIX-NIFTP and C-NRS. He has made many useful comments and contributed a number of bugfixes. Andrew Macpherson of STL has contributed a number of bugfixes and tools to help Sendmail sites convert to PP. The ISODE from Marshall T. Rose of NYSERNet is an essential component of PP. Useful comment and fixes have been provided by our beta test sites. Particularly useful help has come from: Andrew Findlay of Brunel; Juha Heinanen of FUNET; Simon Poole of SWITCH; Ole Bj{\o}rn Hessen of University of Oslo; George Michaelson of University of Queensland; and Peter Yee of NASA. \chapter {Using PP} \section {Conditions of Use} PP is an openly available suite of software. There are no restrictions on its usage. The authors accept no liability resulting from use of PP. \section {Comments and Discussion} Comments and problems on PP should be sent electronically to the PP support mailbox. The RFC 822 form of this is: \begin {quote} \verb|<pp-support@cs.ucl.ac.uk>| \end {quote} Bug reports (and fixes) to this address are welcome. These reports will be dealt with on a best effort basis. The X.400 form is: \begin {center} \begin {tabbing} Organisation \= UCL \kill Surname \> PP-Support \\ Org Unit \> CS \\ Organisation \> UCL \\ PRMD \> UK.AC \\ ADMD \> Gold\verb*| |400 \\ Country \> GB \\ \end {tabbing} \end {center} \pagebreak[4] If this is not possible, comments may be sent to: \begin {tabbing} Postal Address: \= Steve Kille \\ \> Department of Computer Science \\ \> University College London \\ \> Gower Street \\ \> WC1E 6BT \\ \> UK \\ \\ Telephone: \> +44-1-380-7294 \\ \end {tabbing} There is an electronic discussion list: \begin {quote} \verb|<pp-people@cs.ucl.ac.uk>| \end {quote} All interested parties are encouraged to join this list by sending to: \begin {quote} \verb|<pp-people-request@cs.ucl.ac.uk>| \end {quote} \chapter {Obtaining PP} The PP version \ppversion/ software may be obtained in the following ways from these places: \input{distrib} \bibliography{bcustom,pp,networking} \bibliographystyle{alpha}