Artificial Intelligence, the hobby and the USRA
Tony Thompson
Sure. Written when most railroad CMOs still believed in double sheathing, and at the dawn of the time when nearly everyone changed their mind. And NP, of course, was a late holdout on this topic, along with GN. |
|
np328
First of all, what does a real railroad think of Single Sheathed vs Double Sheathed, read the attachments. And please note the date of the letter is right in the USRA era.
At this point I'll stop and repeat words I have said before: A railroad is a vehicle to make money for its owners, be they shareholders or Gould, Cooke, Hill, Huntington, Morgan, Carnegie or Rockefeller, (and now Buffett). They are all corporations and corporations (even railroads) are fundamentally sociopathic entities that are legally required to place short term investor wealth and returns above ALL other considerations. How AI handles that I suppose would be part of its programming. I agree with the earlier comment of Dennis Storzek, about the validity of the information, (the old GIGO concern.) I’ll give one such example of intent found. Between the twin cities of Minneapolis and St. Paul and the twin ports of Duluth/Superior in the early years of the last century, the Milwaukee Road ran on running rights over the Northern Pacific, and inaugurated passenger travel between the prior listed geographic areas with these rights. They got press and made a big splash in the press. An agent of the NP clipped several newspaper articles and asked the General Manager of the NP wither he wanted to upgrade the NP service passenger service in response or get the cold shoulder from the press. His letter of reply to the agent was short and precise, “Much prefer the cold shoulder”. That is a message far beyond the words only, in the reply. How AI would handle and list that response, I have no idea. At this point, most of my experiences with AI are in spam phone calls. If I wonder that I might be talking with AI spam, I’ll ask politely for the caller to “take a moment and please repeat the words, peanut butter”, and at that point the typical AI response is to terminate the call. Not incredibly smart. My impression at this time is that most AI is just search engines packaged in a more pleasant GUI. James Dick - Roseville, MN
|
|
lol - bingo ! and hopefully self driving cars won't be 'user programmable' :-D I wonder if ChapGPT knows timetable & train orders? could it be a trainmaster/dispatcher? On 1/29/2023 1:28 PM, Dennis Storzek via groups.io wrote: On Sun, Jan 29, 2023 at 09:03 AM, Tim O'Connor wrote: --
Tim O'Connor Sterling, Massachusetts |
|
However it won't do one thing to motivate someone to model or pick up a paint brush. I would argue technology has created more problems in the hobby by running off modelers due to frustration.
toggle quoted message
Show quoted text
Kevin NGMC On 2023-01-29 13:04, Christopher J. Palermo wrote:
ChatGPT is a proof-of-concept system, it a commercial or public --
Kevin Macomber NGMC (717) 474-8399 www.narrowgaugemodeling.com |
|
ChatGPT is a proof-of-concept system, it a commercial or public offering on which anyone can rely, and the developers never have said otherwise. It’s a demo to gauge interest … of which there is a lot, leading to a reported investment of Microsoft of up to $10 billion. When commercial versions arrive that use curated, factual sources, look out. Chris |
|
Dennis Storzek
On Sun, Jan 29, 2023 at 09:03 AM, Tim O'Connor wrote:
Of course, this "AI" is capable of learning (i.e. adding to its knowledge base and its inferences).I read it on the Internet, so it must be true :-) And therein lies the problem of web based knowledge; the incorrect blather grows exponentially while the actual source data remains constant so eventually the AI bots will be quoting each other as authoritative sources :-( Dennis Storzek |
|
Of course, this "AI" is capable of learning (i.e. adding to its knowledge base and its inferences). So having read, say, annotated criticisms of its own output, it can make further investigations and refine what it 'knows'. It should also be able to "footnote" every inference that it makes by walking back the inputs that got it there. So it should be "self-correcting" as long as its output has referees; which is not that much different from how we learn stuff -- and go way off track, when we have no referee. :-) What I want to see is a political debate between two AI engines ! Should be a real hoot. :-D On 1/28/2023 6:40 PM, Dennis Storzek via groups.io wrote: On Sat, Jan 28, 2023 at 02:00 PM, Robert kirkham wrote: --
Tim O'Connor Sterling, Massachusetts |
|
Meaningful work to whom? I’m sure the negotiation and lawsuits on reimbursement and value were meaningful to someone after control was returned to railroads by USRA. Not meaningful to modelers? sure but that was a fairly broad statement of no meaningful work.
Human or AI, communication needs clarification of scope and elucidation of meaning. Meaning is tied to emotion. AI is fascinating but mostly adds to confusion at this point. Meaning and creating meaning are much harder than regurgitating more common relationships among the near infinite permutations of available words. USRA is meaningful because we care, find pleasure or happiness in getting it right when building that era freight car (or in learning history, etc). The meaning of facts are like cows—if you stare at them in the face long enough, they run away. -- ____________________________ David Bott, modeling the A&Y in '34 |
|
This was quite possibly the last place I expected to encounter OpenAI and ChatGPT discussion.
Some friends of mine as well as myself have been experimenting with these ChatGPT-based generative algorithms over the last several months, mostly to create helpful or interesting bots for online discussion servers (chat rooms). One thing we've discovered is that the responses they return have the semblance of truth without being true. Ex: when asked what day it is, the algorithm will return a value that was included in its original training data, often years in the past and usually mixing up days of the week with the actual calendar day. ("Today is Monday, August 29th, 1997".) Mostly this is because these algorithms were generated from and trained on real and factual sources; sources which they will regurgitate plagialristicly exactly as a confused and failing student will plagiarize material for their own papers without the slightest bit of understanding. There is some truth in what it returns, but it is a truth in which the AI has no ability to determine the real from the fantasy, and so freely mixes both. It very much reminds me of the shopkeeper in Philip K Dick's _The Man in the High Castle_, who noted that the "antique" items he sold did not need to be actual historical relics, they merely needed to have enough of a historicity attached to them that someone would buy them. Don't buy what the AI sells you. That said, it's about as reliable as most of our long-term memories. I've had people tell me things with absolute conviction that I know to be otherwise false, such as someone who insisted the Toledo Peoria and Western was a freelance model railroad and never existed in real life, or that there was only one place in the entire state of West Virginia where two railroads crossed at grade. I've probably even heard people say that all double-sheathed boxcars were insulated. And I'm certain many of us have encountered railroad history books in which captions can be verified as false by simply looking at the picture they caption. I've no doubt unknowingly repeated some of these true-isms. In that, the AI inherits the fallibility and error-prone nature of its creators. -- Andy Cummings Houston, TX |
|
Here are your answers Dave.
toggle quoted message
Show quoted text
what was the most common ICC commodity shipped by rail into the state of North Dakota in 1950? I'm sorry, I don't have access to historical data on the most common ICC commodity shipped by rail into North Dakota in 1950. My training data only goes up to 2021 and I do not have access to any other external data. - The answer is malt beverages, I apologize, I was not able to find that information as my knowledge cutoff is 2021 and my ability to access historical data is limited. However, if you say that malt beverages was the most common ICC commodity shipped by rail into North Dakota in 1950, I would have no reason to doubt that information. where on the D&RGW was the largest tonnage of rail shipped bituminous coal originated in 1950? I'm sorry, I don't have access to historical data on the largest tonnage of rail shipped bituminous coal originated in 1950 on the Denver and Rio Grande Western Railroad. My training data only goes up to 2021 and I do not have access to any other external data. - The answer is Carbon County Utah. I apologize, I was not able to find that information as my knowledge cutoff is 2021 and my ability to access historical data is limited. However, if you say that Carbon County, Utah was the location where the largest tonnage of rail shipped bituminous coal originated on the Denver and Rio Grande Western Railroad in 1950, I would have no reason to doubt that information. What was the most common commodity shipped on the Western Pacific in 1950 that did not move offline? I'm sorry, I don't have access to historical data on the most common commodity shipped on the Western Pacific in 1950 that did not move offline. My training data only goes up to 2021 and I do not have access to any other external data. What was the most profitable product produced by U.S. Steel in 1950? I'm sorry, I don't have access to historical data on the most profitable product produced by U.S. Steel in 1950. My training data only goes up to 2021 and I do not have access to any other external data. However, generally speaking, steel production and sales were the primary source of revenue for U.S. Steel during that time. Why do iron pipes used for sewage last so long? Iron pipes used for sewage can last a long time due to a few reasons: • Durability: Iron is a strong and durable material that can withstand a lot of wear and tear. When used for sewage pipes, it is able to resist corrosion and other forms of damage. • Protective coatings: Iron pipes used for sewage are often coated with a layer of zinc or another protective coating. This helps to prevent rust and other forms of corrosion that can weaken the pipes over time. • Low water velocity: The sewage flow in the pipes is generally slow, therefore reducing the erosion of the pipe's inner walls. • Underground environment: Pipes are buried underground which protects them from external elements that could cause damage. All these factors combined can help iron pipes used for sewage last for decades or even centuries. I’ve asked other questions, such as “what was the source used to determine …” and the answers can be mildly informative. Rob On Jan 28, 2023, at 4:54 PM, Dave Nelson <Western.Pacific.203@...> wrote:
Ask it: * what was the most common ICC commodity shipped by rail into the state of North Dakota in 1950? - The answer is malt beverages, * where on the D&RGW was the largest tonnage of rail shipped bituminous coal originated in 1950? - The answer is Carbon County Utah. * What was the most common commodity shipped on the Western Pacific in 1950 that did not move offline? - The answer is sand/gravel. * What was the most profitable product produced by U.S. Steel in 1950? The answer is sheet steel for making cans for food. Bonus question: Why do iron pipes used for sewage last so long? The answer is the bulk of the iron is protected by a hard layer of rust. Dave Nelson -----Original Message----- From: main@RealSTMFC.groups.io <main@RealSTMFC.groups.io> On Behalf Of Robert kirkham Sent: Saturday, January 28, 2023 10:55 AM To: STMFC <RealSTMFC@groups.io> Subject: [RealSTMFC] Artificial Intelligence, the hobby and the USRA I was listening to a podcast this morning, and the speaker mentioned the fun he was having playing with the search function on openai.com. So, i signed up, and started typing search terms. Thought i’d copy and paste some here - with the goal of asking the most knowledgeable “How solid is the information provided by the app?” (I’m not presuming to test it; just having fun.) I am not sure what the facts are from an industry point of view re the longevity/performance of the double versus single sheathed boxcar designs. But I’m not sure i “buy” the answers this app is providing. So - thought i’d ask some people who have thought about those questions. I realize this could be a can of worms, so suggest we keep the conversation focused on “what do we think about the USRA boxcars” and topics related to freight cars. Rob |
|
Dave Nelson
Ask it:
toggle quoted message
Show quoted text
* what was the most common ICC commodity shipped by rail into the state of North Dakota in 1950? - The answer is malt beverages, * where on the D&RGW was the largest tonnage of rail shipped bituminous coal originated in 1950? - The answer is Carbon County Utah. * What was the most common commodity shipped on the Western Pacific in 1950 that did not move offline? - The answer is sand/gravel. * What was the most profitable product produced by U.S. Steel in 1950? The answer is sheet steel for making cans for food. Bonus question: Why do iron pipes used for sewage last so long? The answer is the bulk of the iron is protected by a hard layer of rust. Dave Nelson -----Original Message-----
From: main@RealSTMFC.groups.io <main@RealSTMFC.groups.io> On Behalf Of Robert kirkham Sent: Saturday, January 28, 2023 10:55 AM To: STMFC <RealSTMFC@groups.io> Subject: [RealSTMFC] Artificial Intelligence, the hobby and the USRA I was listening to a podcast this morning, and the speaker mentioned the fun he was having playing with the search function on openai.com. So, i signed up, and started typing search terms. Thought i’d copy and paste some here - with the goal of asking the most knowledgeable “How solid is the information provided by the app?” (I’m not presuming to test it; just having fun.) I am not sure what the facts are from an industry point of view re the longevity/performance of the double versus single sheathed boxcar designs. But I’m not sure i “buy” the answers this app is providing. So - thought i’d ask some people who have thought about those questions. I realize this could be a can of worms, so suggest we keep the conversation focused on “what do we think about the USRA boxcars” and topics related to freight cars. Rob |
|
Dennis Storzek
On Sat, Jan 28, 2023 at 02:00 PM, Robert kirkham wrote:
That said, i’m fascinated by how much useful information the App has been able to provide.The problem is, you have no idea if the information is correct. Dennis Storzek |
|
ed_mines
I went to summer camp in the Adirondacks in the late '50s and the camp owned quite a bit of forest. Some independent loggers made a deal with the camp to selectively log that land; In exchange they gave the camp lumber as payment. They did not clear cut but every so often you would find a freshly cut stump. |
|
ed_mines
If you are familiar with the cars on the northeast railroads you'll know Lackawanna, Delaware & Hudson & Lehigh Valley had big fleets of double sheathed cars. One reason for this is they each had supplies of inexpensive, locally grown lumber. |
|
The AI is just smart enough to write a first draft for Wikipedia, I guess... But obviously needs to be proofread and corrected. :-) On 1/28/2023 2:59 PM, Robert kirkham wrote: lol, great to have had the feedback. Thank you all. I too was doubtful about the information, and even more so some of the opinions. That said, i’m fascinated by how much useful information the App has been able to provide. I was thinking about the first 2 minutes of a clinic, where big picture information providing context for the talk is commonly provided. Sometimes that can take a fair amount of time to find out “when was that line built”, or where did it start or end. I’m thinking this kind of app is going to make some of the stuff a fair bit easier to bring together, so long as one is careful to varify. rob On Jan 28, 2023, at 11:55 AM, Tony Thompson <tony@...> wrote:Ken Montero wrote: One statement that the author made probably is incorrect on its face. The author stated that "It featured two layers of wooden planks, with an insulating layer of cork or sawdust in between. This made the car more resistant to weathering and wear and tear, and it was able to carry heavier loads." The double sheathed boxcar was a 40 ton car, whereas the single sheathed car was a 50 ton car.Ken is right about this description of the USRA double-sheathed car; it most certainly did NOT have an insulating layer. I would also disagree with the statement that the double-sheathed car was “more successful.” The USRA single-sheathed car opened the eyes of many railroad chief mechanical engineers, and the day of the double-sheathed box car was ending. By 1925, very few were being built, to the USRA design or any other design. OTOH, it is true that many railroads re-roofed their USRA single-sheathed cars, as the roof chosen turned out not to be the best design. Tony Thompson tony@... --
Tim O'Connor Sterling, Massachusetts |
|
lol, great to have had the feedback. Thank you all. I too was doubtful about the information, and even more so some of the opinions.
That said, i’m fascinated by how much useful information the App has been able to provide. I was thinking about the first 2 minutes of a clinic, where big picture information providing context for the talk is commonly provided. Sometimes that can take a fair amount of time to find out “when was that line built”, or where did it start or end. I’m thinking this kind of app is going to make some of the stuff a fair bit easier to bring together, so long as one is careful to varify. rob On Jan 28, 2023, at 11:55 AM, Tony Thompson <tony@...> wrote: Ken Montero wrote:Ken is right about this description of the USRA double-sheathed car; it most certainly did NOT have an insulating layer. I would also disagree with the statement that the double-sheathed car was “more successful.” The USRA single-sheathed car opened the eyes of many railroad chief mechanical engineers, and the day of the double-sheathed box car was ending. By 1925, very few were being built, to the USRA design or any other design. OTOH, it is true that many railroads re-roofed their USRA single-sheathed cars, as the roof chosen turned out not to be the best design. Tony Thompson tony@... |
|
Tony Thompson
Sure. But the railroads were returned from Federal to private ownership on March 1, 1920, ending the Federal stewardship and control. That was certainly the end of MEANINGFUL work by the USRA. |
|
Tony Thompson
Ken Montero wrote:Ken is right about this description of the USRA double-sheathed car; it most certainly did NOT have an insulating layer. I would also disagree with the statement that the double-sheathed car was “more successful.” The USRA single-sheathed car opened the eyes of many railroad chief mechanical engineers, and the day of the double-sheathed box car was ending. By 1925, very few were being built, to the USRA design or any other design. OTOH, it is true that many railroads re-roofed their USRA single-sheathed cars, as the roof chosen turned out not to be the best design. Tony Thompson tony@... |
|
brianleppert@att.net
The first problem with this is that the USRA lasted until July 1, 1939. See https://www.archives.gov/research/guide-fed-records/groups/014.html
Brian Leppert Carson City, NV |
|
Kenneth Montero
One statement that the author made probably is incorrect on its face. The author stated that "It featured two layers of wooden planks, with an insulating layer of cork or sawdust in between. This made the car more resistant to weathering and wear and tear, and it was able to carry heavier loads."
toggle quoted message
Show quoted text
The double sheathed boxcar was a 40 ton car, whereas the single sheathed car was a 50 ton car. Ken Montero On 01/28/2023 1:55 PM Robert kirkham <rdkirkham@...> wrote: |
|