New Mars Forums

tahanson43206

Moderator

Registered: 2018-04-27

Posts: 19,139

Aristotle Euclid and Computer Science

SpaceNut .... we did not have a topic with Aristotle in the title...

https://getpocket.com/explore/item/how- … ket-newtab

This article contains a brief history of logical thought, starting with Aristotle and Euclid, and concluding with today's computer science.

Today’s most promising machine learning techniques use neural networks, which were first invented in 1940s by Warren McCulloch and Walter Pitts, whose idea was to develop a calculus for neurons that could, like Boolean logic, be used to construct computer circuits. Neural networks remained esoteric until decades later when they were combined with statistical techniques, which allowed them to improve as they were fed more data. Recently, as computers have become increasingly adept at handling large data sets, these techniques have produced remarkable results. Programming in the future will likely mean exposing neural networks to the world and letting them learn.

This would be a fitting second act to the story of computers. Logic began as a way to understand the laws of thought. It then helped create machines that could reason according to the rules of deductive logic. Today, deductive and inductive logic are being combined to create machines that both reason and learn. What began, in Boole’s words, with an investigation “concerning the nature and constitution of the human mind,” could result in the creation of new minds—artificial minds—that might someday match or even exceed our own.

Chris Dixon is a general partner at Andreessen Horowitz.

The article by Chris Dixon is fairly meaty ... I'm hoping there is/are forum member(s) who are willing to add their insights to this topic.

(th)

tahanson43206

Moderator

Registered: 2018-04-27

Posts: 19,139

Re: Aristotle Euclid and Computer Science

This topic was sitting out in the dim recesses of the NewMars archive.

It deserves another moment in the sun, before it falls back into oblivion.

(th)

tahanson43206

Moderator

Registered: 2018-04-27

Posts: 19,139

Re: Aristotle Euclid and Computer Science

Microsoft just announced a research result that reminds me of Deja Vu All Over Again ...

In the early years of computing, Analog Computers had a long and successful run.  i recall seeing labs set up with devices in this category.

The great success of digital computers seems (to me at least) to have caused analog computing to fall by the wayside.

Microsoft's research teams appear to have brought it back in a BIG way!  Here is a link to the story:

https://www.msn.com/en-us/lifestyle/sho … da02&ei=14

The key for success, aside from the use of photons, appears to me to be the integration of this analog computing technology with the digital one.  The first major customers are (apparently) large financial institutions, where huge numbers are routine. 

Here's a snippet from the closing of the article:

That's where AIM comes in. This "analog optical computer" can do more, much much faster… at the speed of light, in fact.

For a little context, photons don't interact with one another, but they do leave imprints on the matter through which they pass. AIM researchers have been able to use this fact to their advantage. It throws light, which passes through several layers, making impressions on each part of what's known as a "modular array". It's this process of projecting light through the array that replaces the function of a standard transistor.

Since light-based technologies can perform powerful linear operations, and electronic components can work on non-linear operations, the researchers have been able to combine both optical and electronic analog technologies, to form a computer that "sidesteps the diminishing growth of computing capacity per dollar in digital chips."

Essentially, AIM is making Moore's so-called law look more like guidelines.

This article may contain affiliate links that Microsoft and/or the publisher may receive a commission from if you buy a product or service through those links.

Wikipedia offers a history of Analog computers:
https://en.wikipedia.org/wiki/Analog_computer

It would appear the topic will be able to continue on into the future, as the Microsoft announcement is brought into the document.

(th)

tahanson43206

Moderator

Registered: 2018-04-27

Posts: 19,139

Re: Aristotle Euclid and Computer Science

The article at the link below is about the ancient hardware and software (and even a few people) who are keeping probes going/talking despite decades of Solar System and even Interstellar travel.

https://www.bbc.com/future/article/2023 … ions-alive

I thought the article was well written and researched.  it is for a general audience, but NewMars members may find some of the technical details interesting.

(th)

tahanson43206

Moderator

Registered: 2018-04-27

Posts: 19,139

Re: Aristotle Euclid and Computer Science

For SpaceNut ... we have only one topic that contains the word "Euclid"..... This post is about the Euclid Space Telescope that was launched in 2023 and started in production in February of 2024.

The article reports results of the first 1% sky map, and related background information.

https://www.cnn.com/2024/10/23/science/ … id=ios_app

(th)

kbd512

Administrator

Registered: 2015-01-02

Posts: 7,802

Re: Aristotle Euclid and Computer Science

I would like to know what the anticipated or empirical service life of various modern computer components are in a high radiation environment, either deep space or the surface of Mars.  The one way in which a mechanical or photonic computer may be superior to current semiconductor-based computer technology is in the area of resistance to upset, malfunction, or destruction from relativistic heavy ions, aka "galactic cosmic rays".

For certain kinds of computing, the speed and computational capabilities of a general purpose CPU are superfluous.  A simplistic life support computer is required to read various sensor inputs to determines the CO2 content of cabin air, air flow rate, etc.  That does not require incredibly precise control.  So long as the control responses remain within acceptable ranges, a computer that opens and closes solenoids to push more (consuming more power) or less (consuming less power) cabin air through a CO2 scrubber has done its job.  Infinitely fine control is not required.  Robustness / tolerance to extreme temperatures and radiation inducing voltages in the circuitry are required.  That is the sort of "gross control" which must be utterly reliable and not readily subject to the kinds of total failures that cosmic rays might induce in a highly integrated microchip that contains logic gates so small that they can be fused together from one or more ion impacts.

Fine control, of the type useful and necessary for scientific observation, trace contaminant elimination, or general computing, is the sort of tasking you can only realistically accomplish with a microchip under the size / weight / power constraints imposed by the operating environment.

Menlo Micro Solid State Solid State Switches

The switches and relays from the link above are capable of carrying very large voltages and currents without damage or excessive signal losses, parasitic power draw, and heat generation typical of mechanical relays or semiconductor switches.  They should be suitable for the computing tasks necessary for reliable and long-term durable life support and communications that a Mars colony will require.

Input and output are also very important to computing.

Here are the anticipated service lives for various display technologies in a rather benign radiation environment:
CRT: 10,000 - 30,000 hours
LCD: 30,000 - 60,000 hours (there are TN, IPS, VA, QLED, and other types)
LED: 50,000 hours (some will last up to 100,000hrs, but most won't)
QLED: 60,000 - 100,000hrs ("LED" might be a misnomer- I think these are actually LCDs with Quantum dots layers added)
OLED: 100,000 hours (a somewhat dubious claim, but many manufacturers make similar claims)
Plasma: 100,000 hours (newest models only)

Mechanical keyboards can last 10 to 15 years.  There are models made from plastics, toughened glass, various light alloys, and stainless steel.  Some are sealed to prevent ingress of dust and water, while others are not.  There are industrial track balls and mice as well.

One of several manufacturers of environmentally protected compute devices:
Hardbox Corporation

Getting data into and out of a computer are fairly important tasks associated with any analysis work.  Since the resupply warehouse (Earth) is so far away from the colony, everything shipped to Mars needs to be made to last- something that is especially true of functionally non-repairable compute devices.  A tool that is sheared or bent after being used as a pry bar may be repurposed or it could be remanufactured, but compute devices are nowhere near as easy to replace.  We have globe-spanning supply chains here on Earth to deliver the minerals, refined metals, plastics, ceramics, individual components, final assembly, and testing of finished computers.  That's going to take time to establish on Mars, likely years to decades.  Thus, the first applications should be control of critical life support, communications, and consumables logistics accounting computers (computers that track consumption of O2 and N2, food, water, fuel, filters, clothing, kitchenware, furniture, tools, spare parts, construction materials, etc).

I'm sure it's possible to design a highly-integrated computer that can perform all of those functions, but the probability of such a device being damaged or destroyed by cosmic ray impacts or solar particle events is very high.  While perhaps not the last word in absolute efficiency, simpler purpose-built devices are also much more likely to survive long enough that periodic resupply is sufficient.