I had some requests---Entrez/VMS summary (very long)
Rick Westerman
westerm at aclcb.purdue.edu
Wed May 19 17:03:49 EST 1993
Thanks Steve for the 1500+ line posting of all the messages you have
received about making Entrez available to users. To summarize:
On VMS
1) You must use DECwindows/Motif.
2) If you wish to use the CD-roms, you must have a special CD driver to
read this disks; this driver is easy to obtain.
3) If you wish to use the NCBI-network version, then you must purchase
the Multinet TCP/IP software. There are no public domain solutions.
4) If you wish to use the line-oriented Entrez (e.g., "ANGIS") instead of
DECwindows/Motif, then it is not available (see note 1).
On Unix
1) You must use Xwindows/Motif.
2) If you wish to use the CD-roms, then there are no problems.
3) If you wish to use the NCBI-network version, then there are no problems.
4) If you wish to use the line-oriented Entrez (e.g., "ANGIS") instead of
Xwindows/Motif, then it may be available (see note 1).
On Dos
1) You must use Windows.
2) If you have your own CD drive, there are no problems.
3) If you wish to use the NCBI-network version, then you must purchase
the PC-NFS TCP/IP program or a WinSock-compatible TCP/IP program.
There are no public domain solutions.
4) If you wish to use a network mounted CD-rom, then you must have a
network in place (Novell, Pathworks, NFS-mounting); all of this
requires purchase of software for the PC. There are no public domain
solutions.
On Macs
1) You already have a windowing system, no problem.
2) If you wish to use the CD-roms, then there are no problems.
3) If you wish to use the NCBI-network version, then there are no problems.
4) If you wish to use a network mounted CD-rom, then you must have a
network in place (Appletalk, Pathworks, NFS-mounting); the Appletalk
network is built in, Pathworks/NFS will need to be purchased. If the
CD-rom is on UNIX or VMS, then software will need to be purchased.
Note 1:
After struggling with porting the line-oriented version of Entrez (aka,
ANGIS or entrez-2d) for the last 4 days, my conclusions on the package
are:
1) It is nice.
2) It doesn't seem to have too many bugs (I only found one so far). It
does have several bad coding practices that cause the various
compilers to complain. A bug report is being sent to the developers.
3) It is extremely UNIX SysV oriented.
4) It probably will *not* compile cleanly on anything but Unix SysV.
5) You can get it to compile on Ultrix (version 4.2A, DECstation 3100)
with some tweaking.
6) You can not get it to compile on older Unix SysV systems (release
3.1.1 AT&T 3b2) nor on other older Unix systems (SunOS 4.0).
7) The code has so many UNIXisms and VMS's "curses" package is so old
and non-SysV compatible, that porting the program to VMS will be a major
task. Expect to spend two people-weeks doing the port. It may be
possible to port the program using VMS's POSIX implementation, however
you would have to put up with POSIX's slowness.
-- Rick
Rick Westerman System Manager of the AIDS Center Laboratory
westerm at aclcb.purdue.edu for Computational Biochemistry (ACLCB), BCHM
Phone: (317) 494-0505 bldg., Purdue University, W. Lafayette, IN 47907
FAX: (317) 494-7897
More information about the Bio-soft
mailing list