Data processing: software development – installation – and managem – Software program development tool
Reexamination Certificate
1999-04-09
2002-07-23
Chaki, Kakali (Department: 2122)
Data processing: software development, installation, and managem
Software program development tool
C717S114000
Reexamination Certificate
active
06425119
ABSTRACT:
FIELD OF THE INVENTION
The invention is directed to the field of computer methods to produce application oriented languages.
BACKGROUND OF THE INVENTION
An application-oriented language (AOL)—what Bentley (1089) calls a “little language”—is a good way to solve many a problem. An AOL “knows” the nitty gritty and steps needed to solve problems in an application domain, and applies its knowledge automatically as a document written in the language is processed. An AOL frees us to deal with a problem at a higher level of abstraction, and relieves us to the mindless drudgery that is often the cause of errors.
Unfortunately, it takes a lot of expertise and effort to make an AOL. Although tools such as lex (Lesk & Schmidt, 1990), yacc (Johnson & Sethi, 1990), and
MetaTool
(Cleveland & Kintala, 1988) help, they are no panacea and no substitute for expertise in the art and science of designing and implementiing programming languages. This puts AOLs out of the reach of most domain experts who have the need for them but lack the skills to make their own. If domain experts could make little languages easily, more of their potential to enhance productivity would be realized.
To save on the effort of making an AOL, certain features are often left out; the result is a compromised and less effective AOL. For example, a typical AOL doesn't have a debugger, can't be easily extended or customized by its users, lacks scoped variables and data structures, and lacks parameterized functions. The cost of putting these features into an AOL is usually prohibitive given the limited use of each individual language. The hidden cost is that, as users, we must suffer an inadequate AOL over the long haul.
Like many good things in life, too many AOLs may be bad. As they proliferate and our use is spread thinner and thinner over more and more AOLs, we find it harder and harder to achieve mastery of any one, and thend to forget those we use rarely. Their proliferation also creates appliction “islands” that can't talk to each other. The AOLs that fafcilitate problem solving in their respective domains become barriers for solving large problem spanning multiple domains and requiring communication and coordination among partial solutions written iin different languages. Although people can be multilingual, AOL compilers and interpreters are decidedly monolingual. So we end up with many AOLs that can't work together.
We propose here a way to reap the benefits of AOLs without incurring many of their costs. The proposal is that AOLs be realized as jargons. Jargons are a family of AOLs that share many features in common, and are distinguished only by the expressions they contain, just like jargons of a natural language like English. The syntax, execution semantics, and other features that jargons share make them easy to learn and remember, and also make it possible to combine different jargons into a new, hybrid jargon capable of solving bigger problems.
We also present the infocentric paradigm and its realization by the Info Wiz system as an easy way to make a jargon. The infocentric paradigm enables someone with no expertise in programming language design and implementation to prototype a jargon in a day and finish it in a week. The infocentric paradigm represents a sea change in how we go about representing and processing information. The change comes about because InfoWiz makes it so easy to solve a problem indirectly by first making a jaragon, and then using the jargon to represent the solution. This approach is called the infocentric paradigm, because it is centered on information modeling as the key to problem solving, in contrast to the conventional algocentric paradigm that is centered on modeling algorithms or procedures.
The infocentric paradigm makes information reuse practical. Iniformation reuse is realized when an AOL document (i.e., “program”) is kept fixed, but the semantics of the expressions of the jargon are changed on each processing of the document in order to generate different products. This aspect of the infocentric paradigm shows that an Info Wiz document is really a program, but not of the familiar kind. Unlike conventional programs whose free parameters are vairables that take on different data values, the free parameters of an Info Wiz document are expressions that take on different semantics with each distinct reuse.
The original inspiration for Info Wiz was Sharon Murrel's monk text formatting system (Murrel & Kowalski, 1984). Monk introduced the important idea of using a high-level programming language for writing actions that was different from the base language.
following is a bibliography of prior aork in the field of the invenion:
Anonymous, MetaTool Specification-Driven-Tool Builder. North Andover, Mass.: AT&T Bell Laboratories, 1990.
Bentley, J. L. Little Languages for Pictures in Awk. AT&T Technical journal, July-August 1989,,p..
Cleveland, J. C. & Kintala, C. Tools for Building Application Generators. AT&T Technical Journal, July-August 1988,,p..
Devanbu, P. GENOA—
A Customizable, Language
-
and Front
-
End Independent Code Analyzer.
AT&TBell Laboratories Memorandum 11262-910816-22TM, Aug. 16, 1991.
Emerson, S. L. & Paulsell, K. Troff Typsetting for Unix System. Englewood Cliffs: Prentice Hall, 1987.
Goldfarb, C. F.
The SGML Handbook.
Oxford, England: Clarendon Press, 1990.
Greer, R. & Belanger, D. G.
backtalk: A Text Generator, Tree Manipulator and Macro Processor.
AT&T Bell Laboratories Memorandum 112687-931115-18-TMS, Nov. 15, 1993.
Johnson, S. C. & Sethi, R. yacc:
A Parser Generator.
Anonymous (Ed.)
Unix Research System Papers. Tenth Edition.
Murray Hill, N.J.: AT&T Bell Laboratories, 1990.
Knuth, D. E.
The TEXbook.
Reading, Mass.: Addison-Wesley, 1984.
Ladd, D. A. & Ramming, J. C. A*:
a Language for Implementing Language Processors.
AT&T Bell Laboratories Memorandum BL0112650-930924-17-TM, Sep. 24, 1993.
Lesk, M. E. & Schmidt, E. Lex—
A Lexical Analyzer Generator.
Anonymous (Ed.)
Unix Research System Papers, Tenth Edition,
Murray Hill, N.J.: AT&T Bell Laboratories 1990.
Murrel, S. L. & Kowalski, T. J.
Overview of Monk
0.2:
Typographical Database.
AT&T Bell Laboratories Memorandum 11229-841210-12TMS, Dec. 10, 1984.
Nakatani, L. H. & Ruedisueli, L. W.
FIT Programming Language.
Murray Hill, N.J.: AT&T Bell Laboratories, 1991.
Nakatan.. L. H. & Ruedisueli, L. W.
FIT Programming Language Primer.
AT&T Bell Laboratories Memorandum 11264-920301-03TMS, Mar. 1, 1992.
Ousterhout, J. K.
Tel and the Tk Toolkit.
Reading, Mass.: Addison-Wesley, 1994.
Reid, B. K. A High-Level Approach to Computer Document Formatting. In
Proceedings of Seventh Annual ACM Conference on Principles of Programming Languages,
New York: ACM, 1980.
SUMMARY OF THE INVENTION
Jargons are a family of application-oriented languages well-suited for representing a processing complex, hierarchically structures information. A system is presented that automates most of the work of making a jargon, so practically any programmer can make a simple one in a few days. Every jargon has the same syntax, is processed with same ready-made base interpreter, and comes complete with a suite of “deluxe” features: debugger, error handler, function definition, associative arrays, varibles, incremental loader, among others. The application-oriented and declarative nature of jargons make them usable by domain experts who are no programmers. The commonalities among all jargons, especially their syntax, make them easy to learn and remember, and make it possible to combine different jargons to solve bigger problems. Jargons facilitate information reuse, whereby the same information document is repocessed to generate multiplocity of products.
One aspect of the invention is a method for automatically producing an application-oriented language for processing hierarchically structured information. The method comprises the steps of:
providing a general-purpose information language for writing expressions associated with a domain of application;
providing a general-purpose programming language f
Jones Mark A.
Nakatani Lloyd H.
AT&T Corp
Chaki Kakali
LandOfFree
Method to produce application oriented languages does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method to produce application oriented languages, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method to produce application oriented languages will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2825080