|
|
Computer Programming
In the seventies, we asked: "Can we have a computer on every desk?" We now know this is possible, but those computers haven't necessarily empowered their users. Today's computers are often inflexible: the average computer user can typically only change a limited set of options configurable via a "wizard" (a lofty word for a canned dialog), and is dependent on expert programmers for everything else.
"What will happen if users can program their own computer?" We're looking forward to a future where every computer user will be able to "open the hood" of their computer and make improvements to the applications inside. We believe that this will eventually change the nature of software and software development tools fundamentally.
We compare mass ability to read and write software with mass literacy, and predict equally pervasive changes to society. Hardware is now sufficiently fast and cheap to make mass computer education possible: the next big change will happen when most computer users have the knowledge and power to create and modify software.
The open source movement claims that peer review of software by thousands can greatly improve the quality of software. The success of Linux shows the value of this claim. We believe that the next step, having millions (or billions) of programmers, will cause a change of a different quality--the abundant availability of personalized software.
The tools needed for this new way to look at programming will be different from the tools currently available to professional programmers. We intend to greatly improve both the training material and the development tools available. For example, non-professional programmers should not have to fear that a small mistake might destroy their work or render their computer unusable. They also need better tools to help them understand the structure of a program, whether explicit or implied in the source code.
Our plan has three components:
- Develop a new computing curriculum suitable for high school and college students.
- Create better, easier to use tools for program development and analysis.
- Build a user community around all of the above, encouraging feedback and self-help.
These components come together in the scientific exploration of the role of programming in next generation computing environments.
To undertake a research effort called Computer Programming for Everybody . This effort intends to improve the state of the art of computer use, not by introducing new hardware, nor even (primarily) through new software, but simply by empowering all users to be computer programmers.
Recent developments in computer and communication hardware have given many people access to powerful computers, in the form of desktops, laptops, and embedded systems. It is time to give these users more control over their computers through education and supporting software. If users have a general understanding of computers at the level of software design and implementation, this will cause a massive surge in productivity and creativity, with a far-ranging impact that can barely be anticipated or imagined.
On a shorter term, the quantity and quality of available computer software will improve drastically, as the imagination and labor of millions is applied to the problem. Inventive users will be able to improve the software that supports them in their tasks, and share their improvements with their colleagues or--via the Internet--with others far away who are faced with the same tasks and problems. The ability to modify or customize software is important in crisis situations, when experts cannot be appealed to for help. It is also important for day-to-day activities: some at 200,000 to 400,000 currently estimates
The two major research goals are the development of a prototype of a new programming curriculum and matching prototype software comprising a highly user-friendly programming environment. We envision that the typical target audience will consist of high school and (non-CS major) undergraduate college students, although younger students and adults will also be considered. Course and software will normally be used together, so they should be tightly tuned to each other; each will also be usable on its own.
We must also explore the role of programming in the future. We are rapidly entering an age where information appliances, wearable computers, and deeply networked, embedded CPUs in everyday objects offer users control over their physical and information environments. End-user programmability will be the key to unlocking the potential of these technologies.
In the dark ages, only those with power or great wealth (and selected experts) possessed reading and writing skills or the ability to acquire them. It can be argued that literacy of the general population (while still not 100%), together with the invention of printing technology, has been one of the most emancipator forces of modern history.
We have only recently entered the information age, and it is expected that computer and communication technology will soon replace printing as the dominant form of information distribution technology. About half of all US households already own at least one personal computer, and this number is still growing.
However, while many people nowadays use a computer, few of them are computer programmers. Non-programmers aren't really "empowered" in how they can use their computer: they are confined to using applications in ways that "programmers" have determined for them. One doesn't need to be a visionary to see the limitations here.
An even more radical change is the introduction of computing and communications embedded in home and office systems. The number of devices that will contain programmable elements is expected to grow dramatically in the coming years. We must learn how to expose this programmability to users in a meaningful way and to make it easy for non-programmers to control and program these devices.
In this "expedition into the future," we want to explore the notion that virtually everybody can obtain some level of computer programming skills in school, just as they can learn how to read and write.
There are many challenges for programming languages and environments to be used by a mass audience. If everybody is a programmer, poor programmers will surely abound. Coping with this situation adequately requires a rethinking of the fundamental properties of programming languages and development tools. Yet, we believe that there should be no clear-cut distinction between tools used by professionals and tools used for education--just as professional writers use the same language and alphabet as their readers!
Given the ever more pervasive use of computers and software in every aspect of society, we expect the need for programming skills will only increase. While professionals will produce most quality software, there will be a need for more programming and customizability by end users.
Examples of this drive for flexibility can be seen in both present-day computing and its likely future:
- Increasingly powerful applications for desktop and laptop computers use scripting and macro facilities.
- Growth of the Internet has led directly to greater need for programmability to create active and interactive Web content.
- End-user information appliances and networks of CPUs embedded in everyday objects--both will demand user control and personalization.
- Mobile and intelligent software agents will be commonplace and require customization by users.
In the future, we envision that computer programming will be taught in elementary school, just like reading, writing and arithmetic. We really mean computer programming--not just computer use (which is already being taught), for example, has shown that young children can benefit from a computing education. Of course, most children won't grow up to be skilled application developers, just as most people don't become professional authors--but reading and writing skills are useful for everyone, and so (in our vision) will be general programming skills. For the time being, we set our goals a bit less ambitious. We focus on teaching programming to high school and (non-CS major) college undergraduates. If we are successful here, we expect that the lower grades will soon follow, within their limitations.
In addition to the goal of teaching how computers work, a course in computer programming will return to the curriculum an emphasis on logical thought which was once the main benefit of teaching geometry.
Two general computing trends of particular interest are the move towards information appliances and the growth of embedded CPUs in everyday machines and appliances--whether in the military or the civilian sector. The decreasing size of computing and the increasing reach of networking, particularly wireless networking, make it possible to share information between devices and to interact with them. The ability to program will greatly improve users' ability to control these devices. Imagine that users could make their own changes to the software embedded in, say, their GPS receiver or handheld organizer, rather than (or in addition to) downloading upgrades from a vendor or buying "canned" add-on applications from third parties. This would greatly empower people to improve their life by programming their personal tools to do exactly what they need them to do.
If we are successful, non-experts will be able use their computers and other intelligent devices much more effectively, reducing their level of frustration and increasing their productivity and work satisfaction. (New leisure possibilities will undoubtedly ensue as well!) Computer users will be able to solve their own computer problems more often, reducing the need for technical support.
Even if most users do not program regularly, a familiarity with programming and the structure of software will make them more effective users of computers. For example, when something goes wrong, they will be able to make a better mental model of the likely failure, which will allow them to fix or work around the problem. They will also be able to assess better when they can make the changes themselves and when they will need the services of an expert. They will be more able to converse with experts, since they will now share more of a common language. An analogy is obtaining basic literacy in automotive maintenance: you know enough to check your oil and add a few quarts if necessary, but you also know that you shouldn't try to change your own brakes. When the mechanic says "your rotors are warped and you need new pads," you understand what he is talking about.
If this effort is successful, there could be many millions, eventually billions of computer programmers, at various levels of proficiency. The effects this will have on the state of the art of software development is hard to imagine. The nature of software will change to accommodate the needs of these programmers, allowing customization through source code modifications--and personalization's will be plentiful.
The effort could also have a major impact on getting women and minorities into computer programming--currently, these groups are vastly underrepresented.
The recently popular open source movement [OpenSource] is promising to improve the quality of key software packages through the peer review of thousands, as well as the ability for programmers to "scratch their own itch" (i.e., tweak the software in a minor way that only one individual cares about). We expect that moving from thousands to millions or billions of programmers will further change the nature of the software development process. Personal programming will become more important (and feasible) at this scale, while mass peer review will become relatively less important, due to diminished returns (the logistics of integrating bug fixes from thousands of sources is already a formidable task).
But most current software, open source or otherwise, is too complex to allow anyone to do personal customization without first investing a serious amount of effort and time into understanding the software they're using. We are interested in changes to the whole software development process that will fix this as well--in particular, development tools.
In addition, by enabling the programmability of applications by anybody, we will leverage economies of scale without sacrificing the desire of users for highly personalized software. Applications can be mass-produced, without forcing everyone to fit the same mold in their use of the software (or into just those eddies of customizability planned by the developers). Users will want to personalize their systems for a number of reasons; these include becoming more productive, solving a problem peculiar to their needs, or just expressing their creativity and setting themselves apart from their peers. They will be able to achieve this if they have the basic programming literacy we envision.
Some broad questions help frame our specific research goals, such as: Will the programming language taught in schools resemble the programming languages we know today? Will it even be called a programming language? How will we teach it? Will there be only one language? What other tools are essential to the teaching and use of this language? Is it even possible to have a language and tools that are both good for teaching and useful for experts?
Just as interesting are questions like these: How and for what purposes will people use their programming skills? How will a near-universal ability to read and write computer programs change the structure and utility of computer software? (This is especially interesting in combination with future versions of the Internet, which promise high-speed ubiquitous access to computing and storage elements.) Will people be motivated to actually program their systems once they have the confidence that they can? Will they even be interested in the first place?
A clear concern is the expectation that, if most people are programmers, many of them will most likely be poor programmers. People who can't write understandable sentences in their native tongue or balance their checkbook are unlikely to write well-structured computer programs! Our intent, however, is to make programming accessible, if not easy, for everyone. Some users will employ or contract a third party programming and customization service. This is much like a homeowner contracting out for a remodeling job.
We therefore need to investigate ways to improve the quality of the interaction between the programmer and the system, to help even poor programmers get the most out of their computers. For example, you might want to write a program to customize your PDA or toaster, but you might be discouraged if a small mistake could wipe out your address book or set your house on fire. Safeguards against disasters are needed, as well as ways of backing out of unwanted changes to a system as a whole. ("Undo", while very powerful, usually only applies to one file at a time. Backing out of unwanted global system changes typically requires a reboot or even painful data restoration from back-up media.)
Another concern regards configuration management. Without superior configuration management, businesses are going to find themselves either unable to correct problems, or held hostage by programmers who have modified the operating system or applications in a manner that precludes either upgrading or making other changes. In general, all locally made changes to large software systems are currently in danger of being incompatible with future upgrades of the primary product. Even locally produced software may be rendered unusable when the primary developer leaves, due to a number of reasons including lack of testing or documentation.
Apart from the fear that something might go wrong, another concern for beginning programmers who are interested in customizing their computer is the daunting task of trying to understand a large piece of existing software. We need to look into user-friendly tools for program analysis; more about this later. Another intellectual challenge is visualization of (application-generated) data in ways that help novices. Spreadsheets are of great value here, but not all data fits the matrix form.
Scripting languages are growing in popularity among professional programmers but questions remain about performance, software reuse, and integration with components written in other languages. We can address these challenges by enhancing the facilities of Python dialect seamlessly integrated with Java, and SWIG, an interface generator that creates interfaces between scripting languages and systems languages like C or C++.
It is well understood that there is something of a dichotomy between "general" programming languages on the one hand and "domain-specific" languages on the other. For this discussion, we use the term "general" in a broad and loose sense, to include functional programming languages and possibly even logic programming languages, to the extent to which they are usable as a general programming tool. Turing-completeness is the key concept here.
The domain-specific category then contains everything else, from command line argument syntax to email headers and HTML. The distinguishing factor here is the presence of a relatively narrow application domain. In this category we also place things like Microsoft's "wizards" (really just sequences of predefined dialogs connected by simple flow charts) and the controls and dials on microwave ovens or nuclear reactors.
A typical property of domain-specific languages is that they provide excellent control in the application domain for which they were intended, and (almost) no freedom in unanticipated areas. For example, HTML has no inherent ability for conditional inclusion of text, or for variable expansion. (The fact that such features have been added many times as incompatible extensions merely proves this point.) General languages, on the other hand, usually aren't as good in any particular domain. For example, it is much harder to write a program in a general language to format a paragraph of text than it is in HTML.
However, general languages make up for this through their Turing-completeness, which makes it possible to solve any problem that might come up (assuming availability of sufficient resources). General languages are therefore ideal when used in combination with domain-specific languages. For example, if cell phones were programmable, one would still use the regular domain-specific interface (the keypad) to dial a specific number, since that's the most convenient way to access that specific functionality. However, without programmability, there is no way to make it try several different numbers for a particular friend until one is answered, unless the cell phone vendor anticipated this particular feature.
|
|