* Bookmark
  http://gnobog.sourceforge.net

* General Info
  - http://www.emailman.com/conversion

* Outlook Express -> mbox
  - http://www.gpl.no/liboe
    pros: 4.x 5.0
    liboe is a C library that reads Outlook Express 4 and 5.x mailboxes and
	returns a standard Unix mailbox to a function specified as a parameter
	to the converter. oe2mbx is a C program using liboe to convert Outlook
	Express messages to the Unix mailbox format

  - http://www.arrakis.es/~fperezma/OtlkToNs/OtlkToNs.htm
    pros: 4.x
	cons: no 5.0 support
    This software is suited for anyone who wants to move his e-mail files
	from OutLook Express (mbx format) to Communicator v4.61
    It can't process files from OutLook Express 5 because this release stores
	e-mail data in files with a format (dbx) that OtlkToNs can't read.
    It is still in Alpha stage and it may not work well with some *.mbx files.
	Also, the format specifications for the *.mbx files used in this program have
	been recopiled from the Internet and may not be complete.
    OutLook Express uses some index files (idx) in order to speed up access to
	e-mail data. These are not used by this program because e-mail bodies keep
	stored into *.mbx files.

  - kmailcvt
    pro: 4.x 5.0 really well done and programming
    cons: C++, depends on qt and kdenetwork

  - MS Outlook to Unix Mailbox Conversion mini HOWTO
    http://www.ibiblio.org/pub/Linux/docs/HOWTO/mini/other-formats/html_single/Outlook-to-Unix-Mailbox.html
    use windows and an IMAP server

  - http://www.wirejunkie.com/freestuff/devout/
    pros: 2000 !!!!!!!!!
    it's a VBA utility that will export Outlook 2000 mail to Evolution mbox format.  It now does Base64 and
    Quoted Printable encoding (courtesy of a DLL), and thus handles attachments
    and HTML messages.

  - http://mbx2mbox.sourceforge.net/
    cons: no update since 1997

-------------------------------------------------------------------
07/11/01 - Christian Belisle

Goal of the project: Migrate mail, data and useful informations from
Windows to Mandrake Linux

Features:
- Mail & mail account migration from Netscape, Outlook Express, Outlook
and Eudora.
- Data migration from folders specified by the user.
- Migration will be in a format that all the programs can read. Need to
be determined.
- Data from MS SQL Server migration

How it will work:
- The first step will be in the Mandrake First Time Wizard. The program
will detect Netscape, OE and Eudora accounts and will ask to the user if
he want to transfer his account(s) in Linux. And if so, in which
program. After, the user will have to choose if he wants to transfer his
emails or not. And finally, he will have to tell if he wants to transfer
his data, and from where.
- After, an application will be available on the desktop to allow the
user to synchronise with his existing Windows partition.
- The two parts will be depending on a library with all the necessary
functions to allow migration. Only the ergonomics features will be
developped in the applications.
- If the user choose so, a boot-only program can be installed to
synchronize Windows and Linux each time he boots (will mainly be
symbolic links to the files in the windows partition).

Priorities:
- Research for information about Netscape, OE and Eudora formats
(mailbox and accounts)
- Build the mail part of the library
        - OE migration (high priority)
	        - Netscape migration (middle priority)
		        - Eudora migration (low priority)
			- Build the data part of the library
			- Build the standalone interface first to crashtest the library
			- Build an embedded interface to put it in Mandrake First Time
			- Build the synchronizer

			Timeframe:
			- Research: 1 week
			- Mail Part of the Library: 5 weeks
		        - OE migration: 2 weeks
		        - Netscape migration: 2 weeks
		        - Eudora migration: 1 week
			- Data Part of the Library: 2 weeks
			- Standalone interface: 3 weeks
			- Embedded interface: 2 weeks
			- Synchronizer: 2 weeks

Risk Evaluation:
- The first risk will be found in the eudora part of the library. That's
why I put only 1 week. If it's not working, it's not enough popular to
give more time.
- For OE and Netscape, there should be any stuff on the internet to
build something, so no risk here.
- For the data migration part, there's no risk either... we just have to
think about a good way to manage the files (copying or linking in the
windows partition) and once done, the development will be easy.
- In the standalone interface, there will be risks that the library is
not working correctly. That's why I put 3 weeks for the standalone and 2
weeks for the embedded.
- The synchronizer will be like the embedded part... meaning no major
risks.

Technologies used:
- Library: C or Perl (Perl is faster to develop. But if I loose time
coding the existing C source in Perl, I will finish the library in C.)
- Standalone interface: gtk (perl or C... will be determined soon)
- Mandrake First Time: using the same language than the program itself.
- Synchronizer: perl (to ease the development, we don't need big speed
here)

Many things still need to be determined during the research timeframe.
So a second version of this file will be out after the research week.