> Monty avoids the cost, latency, complexity and general faff of using full container based sandbox for running LLM generated code.
> Instead, it let's you safely run Python code written by an LLM embedded in your agent, with startup times measured in single digit microseconds not hundreds of milliseconds.
Oh I did read the README, but still have the question: while it does save on cost, latency and complexity, the tradeoff is that the agents can't run whatever they want in a sandbox, which would make them less capable too.
I'm with you. It scares me how quickly some of my peers' critical thinking and architectural understanding have noticeably atrophied over the last year and a half.
I'd argue that employers shouldn't have access to employee's home addresses either, outside of situations where it's needed (e.g., employee chooses to get paycheck by mail instead of direct deposit). Most employers keep access to personal employee information (PII) restricted to HR/timekeeping/payroll departments anyway.
Why would my direct supervisor need my home address?
My point is that if they’re talking specifically about SVR4, then it was popular. And if they’re not talking specifically about SVR4, then it’s not “early Unix”.
As I said, I’m not trying to claim that they’re “wrong”. Just that the whole thing is phrased poorly because it’s really not clear what their context is. And that’s easily demonstrated by the fact that we’re arguing over said context here.
That's incredibly reductive. I'm sure some people's depression can boil down to a matter of perspective, but it's naive to extrapolate that to everyone with depression.
I'm incredibly optimistic and am content with my position in life. My default state is being mindful of the present and I don't think about things too far into the future. I very rarely ever feel stressed out over things in life.
However, none of that changes the fact that I feel completely empty and find no joy in things. Interests are nearly non-existent, emotions dialed to 1, and the only thing I'm motivated to do is lay in bed staring at the ceiling... unless I'm on sertraline.
Admittedly that's just anecdotal, but I worked in a clinical neuroscience lab researching treatments for severe treatment-resistant depression (read: people who tried so many options including CBT that they even tried electroshock therapy). The only thing that helped those subjects was a regimen of personalized neuroimaging-guided transcranial magnetic stimulation for 10 minutes every hour for 10 hours every day for a week. Even then, it wasn't permanent. Some saw improvement for months, others only weeks.
For some people, it's not just a matter of "perspective".
I'm not telling it's a matter of perspective, my point is that I see no objective metrics to tell apart if the situation is bad so it's one expected to be depressed and when the situation is good (so only medication / therapy would help). And it makes discussions around this topic harder.
If its not just a matter of perspective and only medication can help, etc, then why do we call depression a "psychological" or "mental health" concern? Why isn't it just considered a neurological disease?
Depression is increasingly starting to be seen as a neurobiological disorder as we learn more.
In my own opinion, we need to stop viewing "mental health" as a separate class of conditions from general/physical health. A mental illness is a health/medical condition just like any other and shifting our views and diagnostic criteria in that direction would do a lot to remove the stigma associated with mental illness.
Someone with depression has a chronic illness, not a temporary "it's just in your head" condition.
It's not as liberating as you might think. A joyless existence is either suffering or nothingness. A life without meaning, either internal or external, is one where nothing is meaningful with no motivation thus one of crippling catatonia til death.
All I can say is just that it doesn't feel good and if you can't feel good about anything, your calculus of your life inevitably leads to the conclusion that existence isn't worth it.
Yeah, most folk get the causation backwards. They think having meaning in life will pull you out of depression. It's the other way around. You have to get pulled out of depression to be able to find meaning in your life.
Realizing that some "chemical reactions" can change your destiny then further learning that when you want something, all the universe is bound to conspires in helping you to achieve it.
There is only one cure/hack for Nihilism or similar...
Go somewhere where you need to work physically you az off to afford daily food. You will be so exhausted eventually that:
1 you will not have any energy left for thoughts.
2 If you have any energy left, you will give it to angriness which will lead to other circumstances which are none related to find the meaning of life.
I think a lot of people are conflating depression with "bad thoughts". That's just one possible symptom, usually as a result of a combination of both anxiety and depression.
I didn't have anxiety, just depression. I rarely thought. I existed on autopilot. I was physically exhausted on a daily basis as a division 1 athlete in college. Often went days without eating either because I simply forgot to eat or forgot to make time for it between classes and training. Didn't change anything.
I think something people are forgetting is that motivation is either driving you toward something you want or driving you away from things you specifically don't want. A complete lack of motivation means I wasn't motivated to do anything to get something, but also I wasn't motivated to anything to avoid something either. I wasn't motivated to eat to avoid hunger pangs. I wasn't motivated to quit my sport to avoid routine physical exhaustion. Instead, my empty autopilot existence just freely acted on the expectations of those around me as a proxy for motivation.
I don't really like serif fonts, but the two that immediately come to mind are Noto Serif and IBM Plex Serif. Both are open source. I know Noto Serif is variable, but not sure about IBM Plex.
1. It'll become open-source with 1.0, as the article mentions.
2. Just requires someone (or you) to write a kernel for your GPU, which is done in Mojo itself. I'd double check the supported GPUs or if someone else has already done it.
I understand. Until it is open source it's still a blocker for me though.
I watched a community video for the roadmap and it sounds like hardware is not the focus until sometime after 1.0 release. So I think I can assume it'll be a while (if ever) before I can even think about using it.
I think there's a middle ground between trying to follow a path you set for yourself and thoughtlessly wandering by gut feeling: set a direction you want to go, not a destination.
You'll find that it doesn't require much thought at decision points to choose the options (in aggregate) that push you in that direction. As they say, it's about the journey, not the destination.
With that said, it's still difficult because you have to learn to forego long term expectations and/or acquire discipline not to just "stay put" lest you fall back into the habit of stressing over end goals or the comfort of a stress coping loop (anime, video games, etc), respectively.
> Monty avoids the cost, latency, complexity and general faff of using full container based sandbox for running LLM generated code.
> Instead, it let's you safely run Python code written by an LLM embedded in your agent, with startup times measured in single digit microseconds not hundreds of milliseconds.
reply