TL;DR
This 1990 paper envisions a "Personal Information Portal" (PIP) – a portable, wireless device that acts as a gateway to a distributed information infrastructure, aiming to replace a multitude of current devices. It explores the hardware, software, and networking requirements for such a system, anticipating a future where users access information and computation services remotely rather than relying solely on local storage and processing.
What Problem Does This Paper Address?
The authors identify a growing need for portable computing devices that offer the convenience of personal machines without sacrificing the power and accessibility of networked resources. At the time, portable computers were either underpowered or lacked robust connectivity, while workstations offered power but lacked portability. The paper addresses the challenge of creating a device that bridges this gap, envisioning a "personal dynamic medium" that integrates with a larger information infrastructure, moving beyond the limitations of standalone PCs and terminals. They draw inspiration from Vannevar Bush's "memex" and Alan Kay's "Dynabook" concepts, aiming to realize a more practical and integrated version for the professional and student communities.
Methods
The paper outlines a conceptual design for the PIP, detailing its envisioned hardware and software architecture. Key methodological aspects include:
- Hardware Design: Proposing a device roughly the size of an A4 sheet of paper, weighing under two kilograms, with a high-resolution display (megapixel range), audio capabilities, and various input methods (stylus, touch, keyboard). A crucial aspect is the emphasis on being diskless, relying on a wireless network for storage and computation.
- Software Architecture: Suggesting a true multitasking operating system with segmented virtual memory and network file system capabilities, drawing parallels to systems like CMU's Mach or OSF kernel, and AT&T's Plan-9.
- Networking: Advocating for a high-speed wireless network interface capable of speeds approaching current Ethernet, enabling seamless access to the distributed infrastructure. They discuss physical and data-link layer considerations for robust wireless communication.
- Infrastructure Integration: The core idea is that the PIP is not a standalone device but a portal. The infrastructure would provide storage, computation, and communication services, managed by servers.
Datasets
Not discussed in the paper.
Key Insights and Ideas
- The "Inforum" Concept: The PIP is envisioned as a tool for a new class of users to participate in an "information forum" – a computing milieu built on electronic information exchange.
- Shift from Local to Distributed: A fundamental shift from storing all data locally to leveraging a wireless connection to an infrastructure of servers for storage and computation.
- Beyond PC/Workstation/Terminal: The PIP is positioned as a distinct category of device, not simply a smaller PC or a dumb terminal, but a personal gateway to a larger information ecosystem.
- Importance of User Interface: Significant attention is paid to display resolution, input methods (stylus, touch, voice), and the potential for immersive interfaces like private-eye displays.
- Wireless as a Primary Enabler: The success of the PIP is heavily dependent on the availability of high-bandwidth, reliable wireless networking.
- Research Opportunities: The paper highlights numerous areas for future research, including low-power hardware, advanced display technologies, efficient networking protocols, and the division of functionality between the portable device and the infrastructure.
Key Results
As this is a conceptual paper outlining a future system, it does not present quantitative results from experiments or prototypes. Instead, it offers a vision and a set of technical specifications and requirements for the proposed Personal Information Portal. The "results" are the detailed exploration of the technical feasibility and design considerations for such a device and its supporting infrastructure.
Limitations and Future Work
The authors acknowledge several limitations and areas for future work:
- Infrastructure Dependency: The PIP's functionality is heavily reliant on the existence of a robust and pervasive information infrastructure, which was still nascent in 1990.
- Technological Maturity: While many technologies were emerging, achieving the proposed specifications (e.g., high-resolution displays, low-power components, high-bandwidth wireless) required further development.
- User Adoption: The paper recognizes that non-technical users might be hesitant to adopt new technologies and emphasizes the need for intuitive interfaces.
- Dynamic IP Addressing and Handoffs: A significant challenge identified is handling network and IP address changes when moving between different network segments or locations, requiring new protocols.
- Research Directions: The paper explicitly calls for research in areas such as higher-density, lower-power devices; advanced display technologies (e.g., scanning-mirror displays); systems software for prefetching, data integrity, and naming/binding; and the systematic division of functionality between the portable device and the infrastructure.
Summary updated