3 ms·

Humans have limited RAM, so we have to put our ideas into an external medium that can then be refined. I've been finding AI's suggestions -- even when rather w

Humans have limited RAM, so we have to put our ideas into an external medium that can then be refined. I've been finding AI's suggestions -- even when rather wrong -- help me do that initial step faster. Which, I think, jives with their findings here.

nomel 2y ago
First we extended the reach of our perception with language. Then we extended the energy in our calories reserves with crops/livestock. Then we extended the length of our memories with writing. Then we extended the breadth of our thinking with AI?
- lifty 2y ago
  Then we extend the vastness of our consciousness with the Borg.
  - longerd2 2y ago
    please let me leave my flesh prison
- czl 2y ago
  ... We extend our perception with remote sensors (video, sound). We extend our muscles with machines. ...
gerdesj 2y ago
"Humans have limited RAM" I would suggest we have flexible RAM. Also, we have an awful lot of it. The analogy breaks down as soon as you look at it too seriously! In IT we largely deal with compute, persistent storage and non-persistent storage. Roughly speaking: CPU, RAM, HDD. In humans we might be considered to have similar "abilities" but unlike IT there is a mostly a single thing that performs all of those functions - the brain. That organ is both compute and storage. LLMs can be surprisingly useful but they are a tool. As with all tools they can be abused and no doubt you have spotted all those tech blogs that spout the same old thing and often with subtle failings (hallucinations). Keep your tools sharp and know how to safely use sharp tools.
- gbnwl 2y ago
  Just because the brain is a single “thing” doesn’t mean it doesn’t have distinct types of memory. Consider looking up “working memory” as it’s probably the best analogue to RAM here.
  - lanstin 2y ago
    Not really, it's more like CPU registers. Very limited and stuff has to be in their to be computed on (consciously) (lots of unconscious computation as well of course; a lot of efficiency to be found in moving computation from conscious to unconscious).
- ffsm8 2y ago
  These kinds of similies make less and less sense nowadays because we've got nvme storage nowadays, and that can be as fast as 7GByte/s. That's a lot faster then the RAM in most devices today. And with less latency too. RAMs differentiating factor is increasingly just that it can handle a lot of read/write cycles, not it's speed. And that doesn't map to anything in biology
  - andersa 2y ago
    Wat? NVMe storage is much slower than RAM. And orders of magnitude slower than VRAM.
  - vonmoltke 2y ago
    > These kinds of similies make less and less sense nowadays because we've got nvme storage nowadays, and that can be as fast as 7GByte/s. That's a lot faster then the RAM in most devices today. And with less latency too. 7 GB/s is the low end of the DDR3 performance range; DDR3 is 17 years old. Meanwhile, DDR5 performance ranges from about 33.5 GB/s to about 69 GB/s. RAM latency, even on DDR3, is measured in nanoseconds; NVMe latency is measured in microseconds, making it about three orders of magnitude higher.
    - ffsm8 2y ago
darby_nine 2y ago
> I've been finding AI's suggestions -- even when rather wrong -- help me do that initial step faster. I have no idea how I could even integrate AI into my workflow so that it's useful. It's even less reliable than search is for basic research and can't even cite its sources.... This argument held a lot more weight when it was a search engine playing the role of our memory.
- runlevel1 2y ago
  That comment was solely about AI code suggestions. Generative AI still has a ways to go for other forms of research, and it will never fully replace the utility of a search engine. They're two different tools for different but overlapping tasks.
  - mrslave 2y ago
    Just to be charitable to GP and not to enter the debate, many of my colleagues have replaced Google with ChatGPT as their first port of call.
  - darby_nine 2y ago
    > That comment was solely about AI code suggestions. Even AI code suggestions seem to be only a minor improvement over basic LSP integration. One major exception is tedious formatting of the text—say you want to copy over a table by hand to a domain value, copilot is really good at recognizing values and situating them appropriately in the parent l-value. If chatbots could serve as my RAM, surely they'd be able to generate code relevant to the rest of the codebase or at the very least not require deep scrutiny to ensure their RAM matches mine (it most often does not).
3abiton 2y ago
> I've been finding AI's suggestions -- even when rather wrong -- help me do that initial step faster. Which, I think, jives with their findings here. What's your process? Can you give an example? So far for me, I found them to be most useful using LLMs as code copilot.
- runlevel1 2y ago
  For example, I asked it for a database schema given some parameters. What it gave me wasn't quite what I wanted, but seeing it written out helped me realize some flaws in the original design I had in mind. It wasn't that what it gave me was better. The act of evaluating its suggestions helped me clarify my own ideas. It's sort of like Cunningham's Law with a party of one. Giving me the wrong answer helps me clarify what the correct answer should look like. Or, perhaps a better way to put it: It's easier to criticize than to create. It gives me something to criticize and tinker with. Doing so helps me hone in on the solution I want. (Provided its suggestion was at least in the right universe, of course.)

nomel 2y ago

First we extended the reach of our perception with language. Then we extended the energy in our calories reserves with crops/livestock. Then we extended the length of our memories with writing. Then we extended the breadth of our thinking with AI?

lifty 2y ago
Then we extend the vastness of our consciousness with the Borg.
- longerd2 2y ago
  please let me leave my flesh prison
czl 2y ago
... We extend our perception with remote sensors (video, sound). We extend our muscles with machines. ...

gerdesj 2y ago

"Humans have limited RAM" I would suggest we have flexible RAM. Also, we have an awful lot of it. The analogy breaks down as soon as you look at it too seriously! In IT we largely deal with compute, persistent storage and non-persistent storage. Roughly speaking: CPU, RAM, HDD. In humans we might be considered to have similar "abilities" but unlike IT there is a mostly a single thing that performs all of those functions - the brain. That organ is both compute and storage. LLMs can be surprisingly useful but they are a tool. As with all tools they can be abused and no doubt you have spotted all those tech blogs that spout the same old thing and often with subtle failings (hallucinations). Keep your tools sharp and know how to safely use sharp tools.

gbnwl 2y ago
Just because the brain is a single “thing” doesn’t mean it doesn’t have distinct types of memory. Consider looking up “working memory” as it’s probably the best analogue to RAM here.
- lanstin 2y ago
  Not really, it's more like CPU registers. Very limited and stuff has to be in their to be computed on (consciously) (lots of unconscious computation as well of course; a lot of efficiency to be found in moving computation from conscious to unconscious).
ffsm8 2y ago
These kinds of similies make less and less sense nowadays because we've got nvme storage nowadays, and that can be as fast as 7GByte/s. That's a lot faster then the RAM in most devices today. And with less latency too. RAMs differentiating factor is increasingly just that it can handle a lot of read/write cycles, not it's speed. And that doesn't map to anything in biology
- andersa 2y ago
  Wat? NVMe storage is much slower than RAM. And orders of magnitude slower than VRAM.
- vonmoltke 2y ago
  > These kinds of similies make less and less sense nowadays because we've got nvme storage nowadays, and that can be as fast as 7GByte/s. That's a lot faster then the RAM in most devices today. And with less latency too. 7 GB/s is the low end of the DDR3 performance range; DDR3 is 17 years old. Meanwhile, DDR5 performance ranges from about 33.5 GB/s to about 69 GB/s. RAM latency, even on DDR3, is measured in nanoseconds; NVMe latency is measured in microseconds, making it about three orders of magnitude higher.
  - ffsm8 2y ago

darby_nine 2y ago

> I've been finding AI's suggestions -- even when rather wrong -- help me do that initial step faster. I have no idea how I could even integrate AI into my workflow so that it's useful. It's even less reliable than search is for basic research and can't even cite its sources.... This argument held a lot more weight when it was a search engine playing the role of our memory.

runlevel1 2y ago
That comment was solely about AI code suggestions. Generative AI still has a ways to go for other forms of research, and it will never fully replace the utility of a search engine. They're two different tools for different but overlapping tasks.
- mrslave 2y ago
  Just to be charitable to GP and not to enter the debate, many of my colleagues have replaced Google with ChatGPT as their first port of call.
- darby_nine 2y ago
  > That comment was solely about AI code suggestions. Even AI code suggestions seem to be only a minor improvement over basic LSP integration. One major exception is tedious formatting of the text—say you want to copy over a table by hand to a domain value, copilot is really good at recognizing values and situating them appropriately in the parent l-value. If chatbots could serve as my RAM, surely they'd be able to generate code relevant to the rest of the codebase or at the very least not require deep scrutiny to ensure their RAM matches mine (it most often does not).

3abiton 2y ago

> I've been finding AI's suggestions -- even when rather wrong -- help me do that initial step faster. Which, I think, jives with their findings here. What's your process? Can you give an example? So far for me, I found them to be most useful using LLMs as code copilot.

runlevel1 2y ago
For example, I asked it for a database schema given some parameters. What it gave me wasn't quite what I wanted, but seeing it written out helped me realize some flaws in the original design I had in mind. It wasn't that what it gave me was better. The act of evaluating its suggestions helped me clarify my own ideas. It's sort of like Cunningham's Law with a party of one. Giving me the wrong answer helps me clarify what the correct answer should look like. Or, perhaps a better way to put it: It's easier to criticize than to create. It gives me something to criticize and tinker with. Doing so helps me hone in on the solution I want. (Provided its suggestion was at least in the right universe, of course.)