6 After a few false starts, I (Tony Rossini) finally know how I want to
7 proceed. First, we update CLS. Then, continue with TSL. Separation
8 is needed, so that we can make everything separate.
10 We are working with SBCL and CLISP implementations to keep a wide
11 range of target platforms.
17 File Dependencies from 9 Feb 2007
19 Package lisp-stat-object-system
22 Package lisp-stat-fastmap
25 Package lisp-stat-basics:
42 Original, dating to Oct 30, 1991.
44 This is Lisp-Stat 1.0 Alpha 1, a first attempt at producing a Common
45 Lisp version of Lisp-Stat. This version contains NO graphics, but
46 should implement all the non-graphical facilities of Lisp-Stat.
48 The implementation uses C code from XLISP-STAT for linear algebra and
49 probability distributions, so this code is dependent an a CL's foreign
50 function interface. At this time, three CL's are supported: AKCL (at
51 least verision 1-600) for UNIX systems, Franz' Allegro CL for UNIX
52 systems, and Macintosh CL (version 2.0b1). Separate README files
55 On A DEC 3100 or 5000 the AKCL version of Lisp-Stat runs about half as
56 fast as xlispstat on a standard battery of tests if xlispstat is given
57 an (expand 30) command. I seem to recall that on a sun3 this the AKCL
58 code and xlispstat code run at about the same speed. I have not yet
59 confirmed this. If so, then the relative performance of the AKCL
60 version to xlispstat may be quite hardware-dependent. The Allegro
61 version on a DEC 5000 is considerably slower and larger than the AKCL
62 version, but I have not yet figured out how to tune Allegro's memory
63 management. The Macintosh CL version seems to run at least as fast as
64 xlispstat on the Macintosh.
66 To port this code to another CL, you need to
68 Edit defsys.lsp to add any necessary definitions
70 Add a top level to lstoplevel.lsp (this is only needed if you
71 want to recover the history mechanism, which is broken by
74 Write versions of the lisp and/or C glue files to interface to
77 Experiment with tuning the memory management to run reasonably
78 in statistical applications.
80 If you do port this code to another CL, please let me know so I can
81 add it to this distribution.
85 University of Minnesota
87 luke@umnstat.stat.umn.edu