Our informal group has been making replicas of ancient computers for years.
Of course, the learning curve for these machines is steep. The Agent-PDP1 github is our hope of now 'adding a replicant PDP-1 Hacker to sit with you on the replica machine'.
This has been fun, but there is a long way to go. Still, it really does feel like you are sitting next to the PDP-1 from Stephen Levy's "Hackers" with an old hacker. A little fuzzy in his old age, but with a phenomenal knowledge and skill set...
I do feel like such Hacker Replicants can add to our goal ("Making fully functional replicas, their historical software archive built in. We think of them as computer time capsules")
Whether a Hermes agent with PDP-1 skills is the best setup, I am frankly not sure. We arrived at that over the past year, but it won't be the end.
In fact: the end goal is to have a replica Arpanet network of replica IMPs, with the MIT/Stanford/UCLA/BBN nodes attached to it:
https://github.com/obsolescence/arpanet
And THEN have such replicants roam the Arpanet, free to finger/chat for advice.
We'll see :-) But advice on making the next steps with these 'replicants' is welcome. We need to make them for the PDP-8/11/10 plus the Arpanet IMP as well.
ozymandiax•35m ago
Of course, the learning curve for these machines is steep. The Agent-PDP1 github is our hope of now 'adding a replicant PDP-1 Hacker to sit with you on the replica machine'.
This has been fun, but there is a long way to go. Still, it really does feel like you are sitting next to the PDP-1 from Stephen Levy's "Hackers" with an old hacker. A little fuzzy in his old age, but with a phenomenal knowledge and skill set...
I do feel like such Hacker Replicants can add to our goal ("Making fully functional replicas, their historical software archive built in. We think of them as computer time capsules")
Whether a Hermes agent with PDP-1 skills is the best setup, I am frankly not sure. We arrived at that over the past year, but it won't be the end.
In fact: the end goal is to have a replica Arpanet network of replica IMPs, with the MIT/Stanford/UCLA/BBN nodes attached to it: https://github.com/obsolescence/arpanet
And THEN have such replicants roam the Arpanet, free to finger/chat for advice.
We'll see :-) But advice on making the next steps with these 'replicants' is welcome. We need to make them for the PDP-8/11/10 plus the Arpanet IMP as well.