http://home.cc.umanitoba.ca/~psgendb/birchadmin/addprog/localdocs.html
Below is a more general description of the most recent
release of BIRCH
---------------------------------------------------------------
The word 'system' implies many parts working together in a coordinated fashion.
While installing programs and databases can be a major effort, transforming them
from a mere collection into an integrated system is far more difficult.
The BIRCH system (http://home.cc.umanitoba.ca/~psgendb) makes it easy to find and use
almost any program, data or document. The core distribution includes software for DNA,
protein, and molecular marker data.
For end users:
* The GDE graphic interface [Smith et al., 1994] makes it easy to create ad
hoc data pipelines using output from one program as input for another.
* Locally-installed programs and databases can be added transparently,
without re-compiling GDE.
* Web documentation from the BIRCH core and local documents are unified in a
single database.
* Tutorials are task-oriented, rather than program-oriented.
For the sysadmin:
* Install wizard automates installation, configuration and updates.
* Local customizations automatically re-integrated into BIRCH during updates.
* Central configuration: everything works the same for all users.
* OS-independent datafiles, documentation and scripts are installed separately from
OS-dependent binaries. BIRCH automatically finds the binaries for each host (eg.
Solaris, Linux)
* All components are freely-available software.
* Scaleable from a single PC to a campus-wide system
Web applications are awkward to use and are intractable to automation for large projects.
When BIRCH is installed along with desktop and office applications, the result is a
seamless system on a single desktop. Because BIRCH is network-centric, any user can do
any task from anywhere.