home
***
CD-ROM
|
disk
|
FTP
|
other
***
search
/
ftp.umcs.maine.edu
/
2015-02-07.ftp.umcs.maine.edu.tar
/
ftp.umcs.maine.edu
/
pub
/
WISR
/
wisr6
/
proceedings
/
ascii
/
conn.ascii
< prev
next >
Wrap
Text File
|
1993-10-19
|
8KB
|
190 lines
Impediments to the Software Reuse Industries
Richard L. Conn
TheMITRE Corporation
145 Wyckoff Road, Suite 201
Eatontown, NJ 07724
Telephone: (908)389-6762
FAX: (908)542-3679
Email: rconn@mitre.org
Abstract
The software industry and the reuse industry are facing far too many impediments to become
viable without better government support. This support must materialize in several areas:
eliminating or better documenting and controlling software patents, improving the national
information highway, establishing new laws to wiselyprotect the free exchange of information
between our universities, and reducing barriers to trade, particularly export controls. With
these impediments, far too much energy in the software industry as awhole is spent in meeting
the letter of the law and going to court. This energy could be b etter sp ent in producing new
products and advancing the technology.
Keywords: Software patents, Software reuse component industry, Export controls, National
Research and Education Network, Ada Software Repository, Public Ada Library, information
highway, freedom of exchange, reuse economics
Workshop Goals: Learning; networking; start promoting a correction to the existing legisla-
tion pertaining to the software industry in the United States in order to remove impediments
to the operation of the industry
Working Groups: Reuse management, organization and economics; Education; Reuse hand-
bo ok
Conn- 1
1 Background
I have managed the Ada Software Rep ository (ASR) on the Internet since November, 1984. The
ASR started as a voluntary activity, was funded by the DoD for six years, and reverted to a
voluntary activity. I currently manage the Public Ada Library (PAL) at WashingtonUniversity in
Saint Louis, which I started in June, 1993. The PAL iscurrently a voluntary activity, but DoD
funding is anticipated. I have performed research in the broad field of software reuse, including both
the technical, political, and DoD issues, for the MITRE Corporation. I have authored a tutorial
on software reuse as a result of this research and am in the process of completely documenting
this research in a detailed paper. I have developed reusable components for both commercial and
military applications in Ada, C, and assembly language for over 10 years. I have developed reusable
components in C++ for over 3 years. I have authored a book on the ASR and the Defense Data
Network and several papers on the ASR, reuse, object-oriented design, and software engineering.
I have taught courses in software engineering, object-oriented programming, and object-oriented
design as an Adjunct Professor at the University of Cincinnati (Ohio) and Monmouth College
(New Jersey) for over 5 years. I participated in the review of the do cument Ada Quality and Style:
Guidelines for Professional Programmers by the Software Productivity Consortium.
2 Position
The laws pertaining to software patentsmust be reworked extensively so the patents are clearly
understood. The issuance of software patents, particularly for algorithms and "look and feel" con-
cepts that can be readily rediscovered/invented, must be stopped or at least more wisely controlled.
The ability to issue a patent on a well-known algorithm that has been known to the public for some
time must be denied. The patent approval process, wherein it can take up to five years to issue
a patent in an industry that is evolving radicallyover the span of a single year, must be replaced
with something more timely and effective. To spend years working on a program only to find that
it violates some patent that was in the process while you were working on the program and could
not possibly have known about it is ridiculous. It may b e best to abolish the idea of patenting
software, using the copyright as the only means of rights protection. Copyrights on "look and feel"
should not be allowed; copyrights onlyon actual software source code should be allowed.
The national information highway is currently based on the Internet, and efforts are underway
to improve a portion of the Internet by forming the NREN (National Research and Education
Network). Technical barriers to making this happen are already overcome to a large extent. Political
barriers are far more difficult to overcome,particularly with special interest groups lobbying in their
own interests instead of in the national interest. Consequently, an industry-based reuse council and
an education program for Congress must be set up to look out for the national interests and to
keep Congress informed of the issues.
A similar university-based council and education program for Congress must be set up to look out
for the national interests in terms of the free exchange of information between our universities.
Too little attention is paid to the economics of software reuse, and a "standard" economic model
for software reuse should be developed. The cost of reusing software must be less than the cost of
developing new software, but much of the reuse efforts today do not consider the cost factor.
Finally, export controls placed on the software industry have been ineffective and served mainly
to hurt the software industry as opposed to affordthe protection they were originally intended to
Conn- 2
provide. These export controls must be removed.
3 Comparison
Richard Stallman [1, 2] addressed several of the copyright and software patent issues. Itend to agree
with his perceptions, except that they are somewhat extreme and the actual solution is probably
something somewhat more moderate.
The Central Archive for Reusable Defense Software [cards93a] addressed many of the essential issues
for implementing reuse on the information highway. The CARDS execution is lacking, however,
and many of the concepts promoted by the Public Ada Library [3 ] should be used to augment the
concepts of CARDS.
The Reuse Economics Model [4] of the SPC makes a good attempt at defining an economic model
for software reuse, but it is not complete. The economic model presented by the author [5] is
also incomplete. Issues such as these models should be brought into the spotlight of the reuse
community and a viable "standard" model should be drafted.
References
[1] R. Stallman,"Keynote Speaker Presentation: Against User Interface Copyright,"in Proceedings
of Tri-Ada 92, pp. 9-13, Tri-Ada 92 Conference, November 1992.
[2] R. Stallman, "Against Software Patents," in Proceedings of Tri-Ada 92, pp. 14-27, Tri-Ada 92
Conference, November 1992.
[3] R. Conn,"Public Ada Library Frequently Asked Questions," Tech. Rep. July 1993, Public Ada
Library, July 1993.
[4] B. Jenkins, "Reuse Economics Spreadsheet Model User Manual Version 01.00.00," Tech. Rep.
SPC-92068-C, Software Productivity Consortium, May 1992.
[5] R. Conn,"Ada Software Repository Experiences and Lessons Learned," in Minutes: Ada Board
Meeting, December 4 and 7, 1992 (A. Board, ed.), p. Attachment 4, Ada Information Clearing-
house, 1992.
4 Biography
Richard L. Conn is a Member of the Technical Staff for the MITRE Corporation, Eatontown,
New Jersey, a Research Professor for Monmouth College, West Long Branch, New Jersey, and
a member of the Federal Advisory Board for Ada, Undersecretary of Defense for Research and
Engineering. He is performing research on software reuse and is developing Ada and C source code
analysis tools. He teaches courses in Object-Oriented Design, Software Tool Construction, and
Software Engineering for Monmouth College. Richard currently manages the Public Ada Library
at Washington University in Saint Louis. He received a Master of Science in Computer Science
from the University of Illinois in 1978.
Conn- 3