|
Originally Posted by indyfabz
(Post 23202249)
I’ve been happy seeing the ISS pass overhead three times.
Not blinking like an aircraft, hauling keester across the sky, and I thought I could see the two enlongated lit up solar panels. |
I watch it often. It's easy to identify if you are familiar with the night sky, and satellites. The ISS is very bright compared to other satellites. If you want to know when it is going over, you can use this website. https://www.heavens-above.com
|
thanks for that p
|
Originally Posted by phughes
(Post 23202926)
I watch it often. It's easy to identify if you are familiar with the night sky, and satellites. The ISS is very bright compared to other satellites. If you want to know when it is going over, you can use this website. https://www.heavens-above.com
|
Originally Posted by Tourist in MSN
(Post 23200130)
Maybe you can hitch a ride with the space aliens for part of the distance? Or, do I have the wrong Roswell?
|
Originally Posted by mev
(Post 23197019)
For fun, I did an experiment to see how well a local chatbot could plan an upcoming bicycle trip. I was using a local copy of "LM Studio" on my PC and picked the Mistral Instruct 7B model. Overall less powerful than many large language models out there, but also something I could run local and portable on my PC.
Here is the prompt I provided: > Develop an itinerary and guide for a bicycle trip from Alamogordo New Mexico to Lubbock Texas. Include daily distances and places to stay. Here is the output it generated: Title: Bicycle Trip Itinerary from Alamogordo, NM to Lubbock, TX **Day 1:** * Distance: 48 miles (77 km) * Start the day in Alamogordo, New Mexico. * Head north on US-54 towards Tularosa, NM (20 miles/32 km). Stay overnight at one of the local motels or campgrounds. * Continue your journey to Carrizozo, NM (another 28 miles/45 km) and rest for the day. **Day 2:** * Distance: 69 miles (111 km) * Depart from Carrizozo early in the morning. * Bike through Capitan and Lincoln, New Mexico. * Cross the state border into Texas and reach Roswell (34 miles/55 km). Spend the night here. **Day 3:** * Distance: 61 miles (98 km) * Begin your day in Roswell and pedal towards Lovington, NM (27 miles/43 km). * Then continue to Midland, Texas, (34 miles/55 km), where you can spend the night. **Day 4:** * Distance: 58 miles (93 km) * Leave Midland and head towards Greenwood, Texas (28 miles/45 km). * Then ride to Big Spring, Texas, (20 miles/32 km), where you can spend the night. **Day 5:** * Distance: 61 miles (98 km) * Depart from Big Spring and head towards Slaton, Texas (25 miles/40 km). * Continue to Littlefield, Texas (another 30 miles/48 km), and spend the night here. **Day 6:** * Distance: 51 miles (82 km) * Leave Littlefield in the morning and head towards Lubbock, Texas. * Stop in Levelland for lunch (19 miles/31 km). * Finish your journey in Lubbock, where you can spend the night. **Places to Stay:** Throughout this bicycle trip, you will find various accommodation options such as motels, campgrounds, and local lodges. Make sure to plan and book ahead for a comfortable stay during your rest days. Please note that distances may vary slightly depending on the terrain and route conditions. This itinerary is meant to be a guideline and can be adjusted according to your personal preferences and abilities. Enjoy your bike trip from Alamogordo, NM, to Lubbock, TX! ************* In total it created a six day trip and told me the mileage was 348 miles. Doesn't look too bad as far as averages go. Below is the route if I ask Google Maps to plot a route (Automobile with "avoid highways" set) of everything except last part to Lubbock. In general it does seem to have picked stopping cities with places to stay though I am slightly confused with a "stay overnight" instruction in the middle of day one. I am also unclear on where I cross the border on day two. The route has some similarities with an upcoming trip I am planning though I prefer to go directly from Tatum to Levelland. Definitely needs some polish but eventually use of techniques like retrieval augmented generation can probably also improve things. https://cimg0.ibsrv.net/gimg/bikefor...85e243add6.png As a further exercise, I then asked the chatbot to generate a blog posting that described the ride from the first day... > Now generate a blog post that describes the cycle ride on the first day of the journey from Alamogordo to Lubbock Title: Day One - Pedaling Through Beautiful New Mexico: A Cyclist's Journal As I embarked on my bicycle adventure from Alamogordo, New Mexico, to Lubbock, Texas, I could feel the excitement and anticipation building within me. The first day of this journey promised to be a memorable one as I set out to discover new landscapes, meet friendly locals, and challenge myself both physically and mentally. The morning sun greeted me with its gentle warmth as I prepared my bicycle for the day ahead. After a quick breakfast, I said goodbye to Alamogordo's charming small-town charm and mounted my bike on US-54, heading north towards Tularosa, New Mexico. As I pedaled along, I was struck by the stunning vistas surrounding me – rolling hills dotted with sagebrush and cacti, crystal-clear skies filled with vibrant clouds, and the occasional grazing herd of goats or horses on sprawling ranches. The serene silence of the open road gave way to a symphony of bird calls and gentle breezes that seemed to whisper encouragement as I continued my journey. Passing through Tularosa, I took a moment to rest and refuel before pressing onward towards Carrizozo, New Mexico. This quaint little town welcomed me with open arms and offered a chance to stretch my legs while exploring its unique shops, galleries, and local eateries. As the sun began to set, I found a cozy campground on the outskirts of Carrizozo where I pitched my tent for the night. Surrounded by nature's beauty and the sounds of crickets chirping in the distance, I couldn't help but reflect on the day's adventures – a perfect start to this incredible bicycle trip from Alamogordo, New Mexico, to Lubbock, Texas. Stay tuned for more updates as I continue my journey through the Lone Star State! |
Originally Posted by mev
(Post 23197019)
...
Throughout this bicycle trip, you will find various accommodation options such as motels, campgrounds, and local lodges. Make sure to plan and book ahead for a comfortable stay during your rest days. ... |
Originally Posted by WaveyGravey
(Post 23207752)
Thanks for this tip. Asked Co-Pilot on Bing to plan a itinerary for riding the Natchez Trace including Motel/Hotel and restaurant accommodations. And it laid it out for me.
https://www.bikeforums.net/touring/1...p-summary.html |
Originally Posted by mev
(Post 23197050)
For those more familiar with the east coast, here is what I got when I asked
|
Originally Posted by ericoseveins
(Post 23216392)
Amazing, just a jumble of loosely associated words bordering on total gibberish. As a non-data scientist it's hard for me to conceive of any original source in which those words could've had actual relevant meaning. I'll have to ask my data scientist wife how to understand this. My unlearned hypothesis is that very little of the source data entered into the language model was actually contextually relevant to the question. Is that maybe the right interpretation?
1. This is an example of a text generation (chatbot) task. Essentially providing a small amount of tokens as input and having the model create probability guesses of what comes next and using that to generate text. A different task would be summarization where we take a large amount of tokens as input and then create a summary. The model was trained on a variety of inputs - essentially encapsulating that knowledge as built in. This isn't the largest model (only 7B) and not highest rated so there could be improvements - though likely also shows limitations. 2. Text generation is essentially a guessing what to give next. Arguably, this example does better at "structure" than "content". I asked it for mileages, stopping places and things to see and the output contains those elements. 3. This model essentially has very little sense of geography or spatial awareness. So I expect the actual content to be garbled - distances to be incorrect and place locations to be mis-matched. Only if some of the inputs it remembered had some of them together would it at least put them together in output. So this also means it will do much worse with small details than generic larger places. 4. There are probably some areas this can improve over time. - One example is using more of a mixture of experts (MOE) approach. Rather than making the models larger and larger - recognize that some models are set up and trained and set up for different tasks. So essentially have one switchboard deciding the type of task and then dispatching to a sub-model. - Another example is retrieval augmented generation. For some tasks like riding the C&O, there could be a larger source of existing information, e.g. trip reports or advice columns. Essentially set up the question model with this specialized input and have it handle queries more specific to this data. The last comment is my own general perception on some of this: A. There are problem areas that are really hard - and arguably todays large language models are no closer to getting there than 60 years ago. The example that comes to mind was when I in college years ago, I took an AI course. We studied things that came many years before us including some of the first chatbots (Eliza in the 1960s). Those first chatbots essentially worked by taking some input words you provided and then reformatted them back as part of the response. Some people got excited since you could sort-of have a conversation. At that level how long before we reached general intelligence... B. Needless to say that was a much more difficult task than was believed in the 1960s. AI had a pattern of going into over-enthusiasm and then dis-illusionment and an "AI-winter" when research largely stalled. I think we are perhaps at a similar over-enthusiasm on some aspects and don't think we are any closer to general intelligence than we were when I was in college. C. I do think there are two aspects that have significantly improved as part of this latest boom: = We have gotten much better at finding and exploiting patterns. There are some tasks that work well with patterns (e.g. spam detection, some summarization) and some that don't as well (e.g. arithmetic, maps/locations). Some of those hard tasks are generically hard and I don't expect us to solve them anytime soon. Some of the pattern-based tasks can be useful in their own sense. As an example of the difference - I wouldn't ever ask AI to write a letter of recommendation -- however, if I needed to write one, I might ask for an example so I could use it as input on the structure of what one typically looks like. I would use that structure with my own content to create an actual letter and not use any of the words. = We have also gotten much better at scaling to pull in more and more data. For example the model I used was essentially an encoding of much of the internet captured in a way that the information can fit on a thumb drive. So I am both excited on some possibilities and deeply skeptical/cynical of some of the results. I've watched some things progress pretty far in the last ten years - while simultaneously seeing limitations that we had when I took my AI class 40 years ago and won't be solved in the next 40 years. |
Originally Posted by mev
(Post 23216494)
Here is my description of the results.
So interesting, thank you. A human (who presumably wanted another human to successfully execute this bike ride) would perceive this task very differently, prioritizing descriptions of places that actually exist and meaningful content over structure of the response. In this case the only relevant information in the output was essentially provided in the question, i.e., that there would be a bike and the C&O Canal was somehow involved. As I think I understand you, using a relatively small and undifferentiated model decreased the probability that the guesses would be at all relevant. A smaller but selected model (say, incorporating a hundred thousand actual conversations about biking on the C&O) might more likely produce relevant output. The analogous switchboard in the human brain would be the frontal cortex, filtering and interpreting the question to understand the correct priorities in context. |
My two cents :
I'm an old psychiatrist and psychoanalyst, I'm sorry to see how many of my younger fellows works with all those scales : MADRS, PANSS, PSYRATS, BAVQ-R, Y-BOCS, CGI and so on, these tools are indeed extremely AI-adaptable. Their reliability is unfortunately poor, compared to a non or semi structured interview. As for bike touring, I plan vaguely and mostly interact with local humans for the real "POI". I'm patiently waiting for the next AI winter. |
| All times are GMT -6. The time now is 06:01 AM. |
Copyright © 2026 MH Sub I, LLC dba Internet Brands. All rights reserved. Use of this site indicates your consent to the Terms of Use.