@tedchoward
Uncategorized
19
Posts
3
Posters
2
Views
-
@tedchoward @davew I actually never build it by hand, but that is the correct makefile. Develop is pretty broken still at the moment, but should get a lot more stable by the end of today. If you look at CLAUDE.md at the top of the project tree, there's a quick reference with a list of commands for running tests and doing db migrations, but oddly nothing about how to build. Lemme fix that!
@tedchoward @davew I asked Claude to assess the docs gaps and close them. The README.md now has build instructions, and docs/GETTING_STARTED.md is a new "newcomer" guide with: system requirements, prerequisites, step-by-step instructions, UserTalk examples, REPL usage, and troubleshooting guide. Hopefully this is enough to get you started playing around with it. 🙂
-
@tedchoward @davew I asked Claude to assess the docs gaps and close them. The README.md now has build instructions, and docs/GETTING_STARTED.md is a new "newcomer" guide with: system requirements, prerequisites, step-by-step instructions, UserTalk examples, REPL usage, and troubleshooting guide. Hopefully this is enough to get you started playing around with it. 🙂
@tedchoward @davew Just pull the latest develop to get the updates. I also have another PR coming in that fixes some eggregious issues in dot-path resolution. That'll take maybe 20 min, so you'll want to pull again and rebuild in just a bit.
-
@tedchoward @davew Just pull the latest develop to get the updates. I also have another PR coming in that fixes some eggregious issues in dot-path resolution. That'll take maybe 20 min, so you'll want to pull again and rebuild in just a bit.
@tedchoward @davew Well, I got a little ahead of myself. The REPL should work for some basic stuff, but there are still dot-path search issues I have to track down. This one has been a slog, but making slow and steady progress. Either way it's worthwhile to pull again and rebuild. From the top-level of the project you should be able to run:
./frontier-cli/frontier-cli ./databases/Frontier.root
... to start the REPL.
-
@tedchoward @davew Well, I got a little ahead of myself. The REPL should work for some basic stuff, but there are still dot-path search issues I have to track down. This one has been a slog, but making slow and steady progress. Either way it's worthwhile to pull again and rebuild. From the top-level of the project you should be able to run:
./frontier-cli/frontier-cli ./databases/Frontier.root
... to start the REPL.
@jsavin @davew I’m getting a build error when I run `make -C frontier-cli`
```
› make -C frontier-cli
../third_party/cmake-install/bin/cmake -S ../third_party/Paige -B ../third_party/Paige/build-headless -DCMAKE_BUILD_TYPE=Release -DCMAKE_OSX_ARCHITECTURES="arm64;x86_64"
/bin/sh: ../third_party/cmake-install/bin/cmake: Bad CPU type in executable
make: *** [../third_party/Paige/build-headless/libpaige.a] Error 126
```
It looks like it’s trying to use an embedded version of cmake. -
@jsavin @davew I’m getting a build error when I run `make -C frontier-cli`
```
› make -C frontier-cli
../third_party/cmake-install/bin/cmake -S ../third_party/Paige -B ../third_party/Paige/build-headless -DCMAKE_BUILD_TYPE=Release -DCMAKE_OSX_ARCHITECTURES="arm64;x86_64"
/bin/sh: ../third_party/cmake-install/bin/cmake: Bad CPU type in executable
make: *** [../third_party/Paige/build-headless/libpaige.a] Error 126
```
It looks like it’s trying to use an embedded version of cmake. -
-
@tedchoward @davew aah. Good to know. I'll have a look later. I've got an Intel machine I can try it on
-
@tedchoward @davew aah. Good to know. I'll have a look later. I've got an Intel machine I can try it on
-
This is the best thread ever.! 😀
-
This is the best thread ever.! 😀
Feed RSS
Gli ultimi otto messaggi ricevuti dalla Federazione
-
-
This blogpost makes an astoundingly good case about LLMs I hadn't considered before. The collapse of public forums (like Stack Overflow) for programming answers coincides directly with the rise of programmers asking for answers from chatbots *directly*. Those debugging sessions become part of a training set that now *only private LLM corporations have access to*. This is something that "open models" seemingly can't easily fight. https://michiel.buddingh.eu/enclosure-feedback-loop
-
@lritter in this case it was a bad decision made under pressure that I had already refactored out a year prior, I just missed a spot. i like to think i've been able to code most programmers under the table for quite a long time now, but that's not really all that hard a bar to clear tbh.
-
@evan yeah, totally
-
@aeva the most salient moments of this are a few years in the past for me. you *do* get better. most things i find now are code that hasn't been "garbage collected"
-
This is cool
Frontier REPL - Interactive UserTalk Environment
Type /help for commands, /exit to quit
[root]> clock.now ()
1/26/2026; 6:12:54 PM
[root]> date.dayString (clock.now ())
Tuesday
-
This is the best thread ever.! 😀
