B.8.1 Company Description

This project was conducted at Xerox's world-renowned Palo Alto Research Center (PARC).  PARC was founded in 1970 with a mission to create the "architecture of information," and over the past 30 years has invented graphical user interfaces, Ethernet networking, the computer mouse, object-oriented programming, and many other path-breaking technologies.  Xerox is one of the 100 largest corporations in the United States, with 1999 sales of $19.2 billion.  The Xerox corporate headquarters is in Stamford, CT,

B.8.2 Technology Description

Goals of the ATP Project

When Xerox first started its ATP-funded project they realized that the state of technology of component-based software was already advanced.  Numerous companies had already produced components that customers could purchase, install, and use within their production process.  However, an infrastructure was needed that did a better job of supporting plug-and-play compatibility.  It was difficult to determine which components fit together, how they interacted, or whether they would even work together.  Components that are unable to fit together or work together have minimal, if any, value; their value occurs when they are tied together within one program and used as a group to achieve a greater purpose.  Xerox's goal was to change the nature of component-based software development by incorporating a new model of software development, aspect-oriented programming, into software tools.  These tools would separate the semantic details (functionalities) of a component from the implementation details to increase the plug-and-play compatibility of components. 

Technical Accomplishments

Xerox developed technology that fully accomplished the goals of the ATP-funded project.  Their tool AspectJ embodies this technology.  This tool operates within object-oriented programming and Java programming environments.  AspectJ allows more components to be plugable into a software program, which greatly reduces the cost and time of installing components and increases the selection of available components that a software developer can use within its production process.

Customer Use of the Product

The use of AspectJ is simple and straightforward.  After installation, the product blends with the other Java tools, as well as the compiler and other development tools that software developers use.  Xerox designed the product to be as simple out of the box as possible. AspectJ can be put to use as soon as it is installed, and software developers who use the product can put together componentized code much more easily and efficiently.

Future Products

AspectJ is currently in beta testing.  It will continue to evolve and improve through time.  No other products are available based on the same technology, although other researchers at the PARC lab most likely benefited from the ATP award, according to Gregor Kiczales, who manages the project.  Furthermore, AspectJ researchers learned about new technological innovations from discussing the ATP-funded project with other researchers at PARC, and may have applied some of those lessons to other products not benefiting from ATP-funded research.  However, there is no direct link between the ATP funding and the impact on the non-ATP products.  A benefit most likely existed, but it is excluded from this analysis.

B.8.3 Project Performance

This section estimates the economic performance measures for the ATP-funded Xerox PARC project.  From structured interview information on product life spans, actual and projected sales data, and total project costs, we derived demand curves for AspectJ during its expected life and estimated consumer and producer surplus and R&D expenditures.  Based on these estimates, we estimated the benefit-cost ratio, net present value, and internal rate of return for the ATP-funded project.

Product Life Spans

Xerox first released AspectJ in 1998.  Since that time, it has been available for downloading from the Internet and used by thousands of software developers.  Xerox will continue to release the product and build on it well into the future; however, to maintain our conservative bias we only estimated benefits through 2002.  This is in fact a very conservative approach.  Xerox is still beta-testing the product.  Once this process is complete, they will release a new product that is expected to have a life span that lasts several years past 2002.  These future benefits are excluded from this analysis.

Xerox chose to release the product free over the Internet as part of their pricing strategy.  They believe that to turn aspect-oriented programming into a new standard programming methodology, AspectJ must have broad acceptance.  Once the methodology has been widely accepted, Xerox could then make a profit from the sale of complementary products, consulting services, and training.  This strategy has been pursued by other information technology companies such as Adobe and Netscape.

Competing Technologies

The only competing technology that exists is manually installing components in a program.  The costs of doing this are so prohibitively expensive that it is not often done.

R&D Expenditures for the Technology

Since the start of the ATP project, Xerox has spent roughly $500,000 dollars per year on R&D related to the AspectJ technology.  Since the end of the ATP project in 1999, Xerox has received additional funding from the Department of Defense that has equaled roughly $500,000 per year in 2000 and 2001.  We estimate the total private expenditure to be $5,000,000 over the entire project horizon.  ATP contributed $1,670,000 to the project, from 1995 to early 1999.

Estimation of Performance Measures

Because Xerox is planning to offer AspectJ at no cost during the entire span of our analysis, the marginal production costs are very small, and Xerox will earn zero revenues and therefore no producer surplus.  However, those firms and individuals who use the product will be earning a consumer surplus, which we estimated from data shared with us by Gregor Kiczales.  We estimated the benefit-cost ratio, net present value, and internal rate of return and present them in the table below.  The net benefits and NPV from the project are very small as would be expected.  

B.8.4 Qualitative Benefits Attributable to ATP

General Impacts of ATP Funding on the Company

Without the ATP funding, Xerox would never have been able to develop AspectJ.  The idea existed at Xerox and many people agreed it could be a successful product.  However, Xerox was unable to muster the resources internally until the ATP funding was put in place.  Xerox scaled up their entire production process to match the flow of ATP funds.

There was no external halo effect from the ATP funding.  No mention of the funding is included with any of the information that Xerox presents about the product.  However, there was a significant internal halo effect within PARC about the project.  The ATP funding allowed resources within PARC to be readjusted to allow the development of AspectJ.

General Impacts of ATP Funding on the Market

The Xerox company representative noted that ATP funding of the component-based market has allowed companies to focus on some of the longer-term technologies that are needed to fully develop this market.  Companies can be profitable in the short run based on sales of specific components, but the tools and infrastructure that are needed to develop a full-scale market take longer to reach profitability.  Because of this, the private sector generally lags in the development of these infrastructure and tools.  In the view of Dr. Kiczales, ATP encourages companies to focus on these issues, which has increased the potential for component-based software.

Return to Table of Contents

Date created: December 3, 2002
Last updated: August 2, 2005

ATP website comments: webmaster-atp@nist.gov  / Technical ATP inquiries: InfoCoord.ATP@nist.gov.

NIST is an agency of the U.S. Commerce Department
Privacy policy / Security Notice / Accessibility Statement / Disclaimer / Freedom of Information Act (FOIA) /
No Fear Act Policy / NIST Information Quallity Standards / ExpectMore.gov (performance of federal programs)