Truly impressive whafa. Hope it gets fully functional. In comparison, I feel I fly by the seat of my pants. Many people here would love to go over those reams of data. Could the AI tabulate the correlation between who was right or wrong in some of the approaches? The bear cases? The people who invested in bonds? Or equities? The option traders? Who reported back was slim, but over the course of 25 years, in that mountain of data, things were said, I am sure.
All I can say, whafa, is thank you. I appreciate your work. Please keep going!
Hey! It’s good to hear that from someone with a TMF in their name! I have been wondering if there would be any negative reaction from the Gardeners. If anyone asks, I don’t intend to profit from the archive (and can’t imagine who would pay for access, anyway). Also… In the interest of completeness, is there any way you can get me everything after mid=28685999? That’s where my archive ends. Hey, you don’t get what you don’t ask for, right?
It’s good to hear that from someone with a TMF in their name!
Please don’t get the wrong idea. I am and always have been a contractor and not an employee. I have no authority to speak for the Gardners or for TMF in any business capacity. And alas, I have no access to board post data from the previous or current database. Sorry to disappoint!
Could the AI tabulate the correlation between who was right or wrong in some of the approaches? The bear cases? The people who invested in bonds? Or equities? The option traders? Who reported back was slim, but over the course of 25 years, in that mountain of data, things were said, I am sure.
There are probably infinitely many ways to do this (or practically so), but my plan so far is to calculate quarterly summaries of each author’s posts, to measure how that author’s ideas are changing over time. I could also record their stock picks, etc, with whatever level of detail that lets us check their record. In this way, we could mechanically determine the most prescient TMF author. I’m guessing that correlates very well with accumulated wealth, so we probably have easier ways to determine this.
What I’m interested in is the mood of the TMF authors as a single entity, and how that mood changes over time, and whether those changes presage market changes. That (or something that rhymes) is the question I want to answer.
It’s all good, man! I’m just happy you used it at all. I can see you had some issues, but that’s exactly why I wanted people to try and use it, because I needed to know what normal people would try to do. I’m fixing all the errors I can find in the past week, then releasing v2!
A couple quick things: You can’t use the “collect” panel until you are logged in; also I see Maple tried to open the main TMF dashboard for you, but you failed all the security checks. Curious what you saw, if anything. Logged in users are supposed to be able to see the dashboard, but not use any controls.
Thanks again for being interested! I am definitely still improving it, but all the low-hanging infrastructure stuff that’s always so fun to build in a green field is done, and it’s all hard problems now. So it will be slower progress. You can see the dev log by typing “journal” and Maple will start giving daily processing and code updates soon.
Oh! And probably the most useful thing if you want to see how the agent loop works is typing “mdebug -s” () before you send a prompt. This opens the Maple Debug window.
It wasn’t clear to me that I failed security checks; it allowed me to log in using my Google account but as you saw, most of my attempts to do things failed. I did manage to see an AI-generated summary of a post I submitted in my pre-contractor days as “galagan” without the TMF prefix. Apart from that, not much.
I’ll stay apprised of how you’re doing. Thanks again! best, dan