TEDxQueensU - Selim Akl - Unconventional Computing
## Speaker Context
- Role: Computer scientist.
- Setting: A talk/lecture where the audience was present.
- Framing: To question the conventional wisdom associated with computers and computation in general.
## People
- Sophia: Speaker's daughter, who came from Montreal to listen to the talk.
## Concepts & Definitions
- Suum ergo computo: The Latin motto of the school of computing.
- Computation: The central concept being debated; defined by how the speaker uses it throughout the talk.
- Recursive functions: Referential self-referential processes (like the pictures of Maurice Asper).
- Computational Paradigm: A very powerful but very simple explanation of the processes of Nature.
- Universal computer: A theoretical computer that can perform any computation, regardless of time or memory constraints (as debated in the end).
## Organizations
- None.
## Places
- Montreal: Location from where the speaker's daughter traveled.
## Tools, Tech & Products
- Laptop: Device used by the speaker to illustrate concepts.
- Supercomputer: A type of conventional computer.
- Toaster: Device housing a chip used as an example of a conventional computer.
- Three-dimensional model of the patient: Used in surgery simulation, requiring computational power.
- DNA computer: A proposed technology for computation; can encode data, solving problems like breaking a cipher.
- Qubit: The quantum equivalent of a bit, capable of being zero and one at the same time.
- Accelerating machine: A theoretical machine that doubles its speed at every step.
- Black hole: A physical object that can be harnessed for computation.
- Time machine: A theoretical device related to computation.
- Clocks hanging on the wall: Used in a challenge demonstrating information loss over time.
## Numbers & Data
- 2 + 2 = 4: Simple arithmetic example.
- 32 bits / 64 bits: Word sizes mentioned for current computers.
- 1 trillion compact discs: Amount of information one gram of DNA can hold, equivalent to the information density of the auditorium filled with CDs.
- 8 possible states: The number of states possible for a register of three qubits.
- Two to the 221 years: Estimated time for current computers to factor a 1024-bit number.
- One second: Estimated time for a quantum computer to factor a 1024-bit number.
- 27%: Efficiency of a solar cell, if engineered very well.
- 99%: Efficiency of leaves, in absorbing solar energy.
- Number of atoms: Used to show information density (one atom can be either a zero or a one).
- 1 mm to the side: Size of the cube in which the entire knowledge of humanity can be encoded.
## Claims & Theses
- The motto "suum ergo computo" speaks at different levels.
- Being is computing at the deeper level.
- Computation is ubiquitous, meaning "to be is to compute."
- Computing permeates the universe and drives it.
- Every atom, every molecule, every cell everywhere at every moment is performing a computation.
- Nature does not need to know that it is performing a computation for it to be true.
- The computational Paradigm is a very powerful but very simple explanation of the processes of Nature.
- Nature's action is acquiring, manipulating, and transforming and transmitting information.
- Conventional computers are limited when processing huge amounts of data and performing huge numbers of number crunching in a reasonable amount of time.
- Conventional computers are "hopeless for solving unconventional problems."
- Unconventional computations occur when data changes over time, when deadlines must be met, or when natural laws interfere.
- Parallel computing can speed up processes by using multiple devices.
- Nature performs computation using a "distributed parallel algorithm" (example: stomata).
- Nature does not use infinite Precision real numbers in its computation.
- Information transfer is the secret/meaning of life.
- Information is fundamental to all life on Earth.
- The cell's instruction is to "make a copy of yourself."
- A DNA computer "wins hands down" against a conventional computer.
- Quantum computers can be in a superposition of zero and one simultaneously.
- The efficiency of leaves (99%) surpasses engineered solar cells (27%).
- The ability to predict or detect if a program will crash or run forever is theoretically impossible.
- No finite computer can be universal.
- The universe itself is the only thing that can be universal (one that can do an infinite number of operations per time unit).
## Mechanisms & Processes
- Reading temperature via a thermostat: Described as input reading from the outside world (is this a computation?).
- Thermostat signaling the furnace: Described as affecting the outside world (is this a computation?).
- Recursive functions: Used as a model that some people claim is what computers do.
- Photosynthesis: Mentioned as a process involving computation.
- Butterfly migration: Mentioned as a process involving computation.
- DNA replication: Mentioned as a biological process involving computation.
- Stomata opening/closing: Process governed by neighbors' states, described as a distributed parallel algorithm.
- Quantum computing: Utilizing qubits that exist in a superposition of states (0 and 1 simultaneously).
- Solar energy absorption by leaves: Process involving the FMO bacteria allowing photons to travel quickly to the reaction center by exploring all possible paths.
- Accelerating machine function: Executing steps sequentially in decreasing time intervals (1 second, half a second, quarter of a second, etc.).
- Slime mold operation: Replicating the highway system of Canada on a map.
## Timeline & Events
- Billions of years: Time period over which Nature has been continually computing the next state of the universe (according to Tomaso Tool).
- 1962: Year the Nobel Prize was given to three fellows for discoveries concerning the molecular structure of nucleic acids.
## Examples & Cases
- Pictures of Maurice Asper: Used as an example of self-referential processes.
- Thirst in the house reading temperature: Example of input from the outside world.
- Thermostat telling the furnace: Example of affecting the outside world.
- Pictures inside pictures inside the picture: Description associated with Maurice Asper's work.
- Spring flower opening: Example of a physical process questioned regarding computation.
- Chemical reactions: Example of a physical process questioned regarding computation.
- Cell multiplication / DNA replication: Examples of biological processes questioned regarding computation.
- Electrons spinning: Example from the deepest level of matter questioned regarding computation.
- The neighbor clearing the driveway with four kids: Example contrasting manual labor speed with natural coordination.
- Stomata movement: Example of a leaf mechanism demonstrating a distributed parallel algorithm.
- The smooth incline with the ball: Visualizing traversing an infinite sequence in finite time.
- Nobel Prize citation: Specifically mentioning "discoveries concerning the molecular structure of nucleic acids and its significance for information transfer."
- The process of biological replication: Described as the instruction "make a copy of yourself."
- Solving a cipher using a DNA computer in a test tube: Proof of principle case.
- Factoring a 1024-bit number: Comparison example showing quantum speedup vs. classical time (seconds vs. $> 2^{221}$ years).
- Queens game: Example where each piece is in a superposition of two pieces until touched.
- Solar cell efficiency: Comparison to the 99% efficiency of leaves.
- Slime mold on a map of Canada: Case study where the mold replicated the highway system.
- The riddle sequence (Bob's machines): Escalating challenge from $n$ to $n+1$ to $n+2$ to $n+3$.
## Trade-offs & Alternatives
- Computational Limitation: Conventional computers (e.g., on a laptop) vs. Unconventional/Future Computers (e.g., quantum, analog, biological, etc.).
- Computational Power: Conventional vs. DNA computer (DNA computer wins hands down).
- Computational Speed: Classical algorithm vs. Quantum algorithm for factoring (years vs. seconds).
- Computational Energy Source: Solar cells (27% efficiency) vs. Leaves (99% efficiency).
- Computational Limits: Finite computer capacity vs. Universal computation (Universe itself).
## Counterarguments & Caveats
- "If you tell me some of them are not [computations] you'll have to tell me why they are not computations." (Challenging the limits of computation).
- "We don't care whether the atom or the molecule or the cell or any of the constituents of the Universe knows that it is performing a computation is no more relevant than your computer on your laptop knowing that it is Computing." (Dismissing the requirement of awareness).
- "No finite computer can be Universal." (Addressing the concept of computational limits).
- "Simulation simply does not make sense" (Regarding simulating the clock ticking/information loss).
## Methodology
- Comparison/Contrast: Comparing conventional computation (laptops, word processing) to other systems (biology, quantum).
- Analogy: Using the leaf/stomata mechanism to explain distributed parallel algorithms.
- Thought Experiment: The challenge involving the randomly ticking clocks to demonstrate information loss.
- Iterative Challenge: The escalating riddle posed by Alice to Bob regarding the required number of operations ($n+1, n+2, n+3$).
## References Cited
- Maurice Asper: Mentioned in relation to pictures demonstrating self-referential processes.
- Tomaso Tool: Cited for the claim that Nature has been continually Computing the next state of the universe for billions of years.
- David Deutsch: Cited for the concept that the whole process of life/culture is executing a "self-motivated self-generating computer program."
## Conclusions & Recommendations
- The core argument: "I would like to argue that computation is ubiquitous to be is to compute."
- Best understanding of Nature: "Nature does this by virtue of the quantumness it explores all possible paths that lead to the central processing unit and finds the shortest one and goes there directly."
- Final Universal Conclusion: Only the universe itself can be truly universal.
## Implications & Consequences
- If computation is universal, then "everything you do today is conventional... it's all conventional computations."
- If the computational paradigm is the key, then "we can better understand the world that surrounds us."
- If information is lost (like the ticking clocks), "too bad it's gone," applicable to all models of computation.
- If a universal computer were possible, it would supersede all current and future models.
## Open Questions
- Whether neurons in our brains are computing or doing something magical.
- Whether physical phenomena like a flower opening or chemical reactions are computations.
- If there are other processes that can be called computations beyond the listed examples.
- How to build a device that reliably warns if a program will crash or run forever.
## Verbatim Moments
- "I am therefore compute."
- "suum ergo computo."
- "being is computing."
- "Computation is ubiquitous to be is to compute."
- "Being is Computing."
- "Nature has been continually Computing the next state of the universe for billions of years."
- "The whole thing is executing a self-motivated self-generating computer program."
- "Conventional computers are hopeless for solving unconventional problems."
- "I can explain Nature by saying acquiring manipulating and transforming and transmitting information is all what nature does."
- "Information transfer that's the secret that's the meaning of life."
- "I am therefore I compute."
- "The important thing is that the computational Paradigm is a very powerful but very simple explanation of the processes of Nature."
- "No finite computer can be Universal."
- "The clocks are ticking the information is lost."