Baby monitors fundamentally need to be safe and allow parents peace of mind. Knowing they can constantly hear the baby and communicate verbally to soothe, as if you were in the room with them, allows this. It already means you only have to respond and/or get up to tend if they are still unsettled after a period of time. Monitors also have volume control so you can stay alert without being constantly disturbed.
The traditional baby monitor system had three states (ON and functional, ON and non-functional, OFF). It provides a constantly available and instantaneous test for "functional" - as long as you can hear sound from the other unit, it's almost guaranteed to currently be in a safe condition. Monitoring is constant, human-first, and all human. The system is relatively fail-safe.
Your design replaces this with a multi-state system and algorithms (adding ON in listening mode, ON in delay mode with output off, ON and partially-functional etc.). It removes any reliable method to prove an obviously "functional" state at any given time. Monitoring is non-constant (for the human), machine-first, and human as last resort (like corporate customer services). The system is not fail-safe despite being machine-first and at higher risk of error or malfunction due to complexity.
The risks include a few failed notifications or incorrect delay timing leading to early developmental trauma, such as fear of abandonment. It doesn't take much. We are still learning to deal with this human-human, before adding invisible unknowns.
If nothing else, what are babies likely to learn in a pre-verbal state where the days are spent sensing and observing their environment to develop the brain? It'll probably be how to game the baby monitor algorithm.
> Also during the day it's really handy to be notified when baby has woken up from her nap. Let's us be out of the house, or in a distant room, and still keep track of what's going on.
If you are going out of earshot of the baby by doing this, you are fully relying on the technology being functional. That would seem unnecessarily risky, and not nice for the baby (they can sense this stuff). The odds of a catastrophe are low in a singular environment, but still enough to worry. Almost inevitably it would end badly if this were scaled to a mass consumer product.
The idea is interesting from an engineering view, but from a human one it feels dystopian to insert a machine between parent and baby to this extent. It removes/replaces a layer of human-human connection. Where does engineering the natural human experience out of life end? Automated feeding? AI nannies and teachers? Then onto AI therapist?
My suggestion would be to do a lighter version of the features in a system which focuses on safety (always being able to monitor no matter what, for example on the ESP32, you could have the second core independent and direct-output the feed if the first core hangs or crashes). Feature-wise, rather than not alerting for 10 minutes, you could apply a DSP algorithm which reduces harsh frequencies for this period, whilst slowly increasing the output volume of the monitor.
I like the idea of alerts because it expands the base features without risk, and crucially makes the babies life better, as alerts could be sooner for important things that do need attention quickly. As an experiment, you could add an SD card and record snippets of cries which ended up in alerts, tag them with an ordered list of what you did to soothe or what was wrong, and see if AI can make anything of the data. Maybe certain cries can trigger alerts sooner, or cries over an extended period may indicate fever.
Yes, raw PCBs that size are usually closer to 1EUR/unit. DHL shipping on these slightly larger than average boards would be 3EUR/unit.
Are you talking about fully assembled (pcb + components + solder mask + pnp + soldering + test) at 17EUR/unit? If so, that's incredible cheap for a finished product delivered to your door. If not, you're doing something wrong.
Considering the brain functioning as prediction machine, we are constantly correcting the error between sensory perception and our inner reality. This is classic (closed) control loop[1] with self-correcting characteristics updated by adaptation through learning and experience.
At first the process is subconscious, then chaos enters as our conscious awareness develops, morphing the control loop into second or third order states of "correcting corrections" as we perform inner tasks such as ruminating, or external tasks such as group discussion and logical planning.
The perfect prediction machine would be a simulation running an entire up-to-date universe model, but between our limited physical resources and available energy in reality, our evolved aim is efficiency, by creating a state of awareness and reactive patterns with minimal information (lowest entropy). We do this by making assumptions, testing the world, then processing the response and updating our control loop. The tradeoff is lack of precision, as a model without complete information has guaranteed errors.
Children who form a more realistic core worldview through guidance, opportunities and experience are best set up to create solid foundations which are more adaptable to future unexpected situations. Whether this is learning emotional response in social settings or math, the ability to integrate future conscious experience depends on early neuronal structures formed by subconscious expectations of the world. If measurement error is too great from expectations and our current loop/wavelength, our options are to discard this information or learn from it by reflecting on sources of inevitable prediction error through reasoning.
If this problem can actually be solved (requirement for both anonymity and ID in different spaces online without AI infiltration), it appears to be a long road to get there...
The human species. "We" doesn't include everyone and doesn't necessarily imply the process happens through collaboration and planning (conspiracy). The race to automation is happening as expected; outside any group control and bound by competition. Game theory suggests the end result is us being replaced, if we make it that far. "We" as a species are the ones making it happen.
The whole company is like that. If things were as amazing as advertised, they wouldn't even need to advertise. Or to release models to the public at all.
> Being able to be brutally honest with each other about our misunderstandings
Being specific to misunderstandings is an element that's overlooked.
This advice tends to be taken onboard (often to extremes) by those who take it as a free pass to just say whatever comes to their mind, whenever they like, without explaining how they arrived there. Any excuse to avoid putting in effort to be understood or be conscious of the fact that human beings have emotions.
We are not robots.
I'm glad commenters here are aware of this, as HN sentiment is getting close to the point of treating each other as machines, whilst we train bots to have better communication skill such as empathetic reflection, and allow them more creativity and freedom.
Some people are more patient and sympathetic towards computers making mistakes and not following commands perfectly, or being too verbose, than we are with our fellow human beings.
Aside from the poor tone of this style of writing, short declarative statements don't convey the same information and leave a confusing message.
Without knowing how you arrived at "the point", you are pushing all the work onto the recipient (or worse, every reader of your comment on HN) to verify what you say and how much they can trust you. That could involve researching, checking your credentials, or putting in effort to understand/overlook the emotional tone.
"This is the answer. I have the answer" style dumping of information is a poor form of human-human communication, unless you are directly answering a closed-ended question.
Ban reason and the moderator name were public on Something Awful, which allowed the community to respond (actively or passively), and for more senior moderators/admin to take public action against rogue moderators. The transparent audit trail countered the incentive to ban somewhat, but a lot of people also treating getting banned as a game.
First we would run into the spam-filter problem no different to email. Then we have to choose: do we concede to viewing the world through a lens of WhatEverAI, or train it locally on our own thoughts/views on the world, and hope that AI model is never compromised.
We break eggs into the known confines of a pan. We don't spray egg all over the place unless we want to end up with it on our face.
Even if it did make sense to "move fast and break things" inside working critical systems, doing so should surely be within the law and without going against the most basic of known security measures.
The traditional baby monitor system had three states (ON and functional, ON and non-functional, OFF). It provides a constantly available and instantaneous test for "functional" - as long as you can hear sound from the other unit, it's almost guaranteed to currently be in a safe condition. Monitoring is constant, human-first, and all human. The system is relatively fail-safe.
Your design replaces this with a multi-state system and algorithms (adding ON in listening mode, ON in delay mode with output off, ON and partially-functional etc.). It removes any reliable method to prove an obviously "functional" state at any given time. Monitoring is non-constant (for the human), machine-first, and human as last resort (like corporate customer services). The system is not fail-safe despite being machine-first and at higher risk of error or malfunction due to complexity.
The risks include a few failed notifications or incorrect delay timing leading to early developmental trauma, such as fear of abandonment. It doesn't take much. We are still learning to deal with this human-human, before adding invisible unknowns.
If nothing else, what are babies likely to learn in a pre-verbal state where the days are spent sensing and observing their environment to develop the brain? It'll probably be how to game the baby monitor algorithm.
> Also during the day it's really handy to be notified when baby has woken up from her nap. Let's us be out of the house, or in a distant room, and still keep track of what's going on.
If you are going out of earshot of the baby by doing this, you are fully relying on the technology being functional. That would seem unnecessarily risky, and not nice for the baby (they can sense this stuff). The odds of a catastrophe are low in a singular environment, but still enough to worry. Almost inevitably it would end badly if this were scaled to a mass consumer product.
The idea is interesting from an engineering view, but from a human one it feels dystopian to insert a machine between parent and baby to this extent. It removes/replaces a layer of human-human connection. Where does engineering the natural human experience out of life end? Automated feeding? AI nannies and teachers? Then onto AI therapist?
My suggestion would be to do a lighter version of the features in a system which focuses on safety (always being able to monitor no matter what, for example on the ESP32, you could have the second core independent and direct-output the feed if the first core hangs or crashes). Feature-wise, rather than not alerting for 10 minutes, you could apply a DSP algorithm which reduces harsh frequencies for this period, whilst slowly increasing the output volume of the monitor.
I like the idea of alerts because it expands the base features without risk, and crucially makes the babies life better, as alerts could be sooner for important things that do need attention quickly. As an experiment, you could add an SD card and record snippets of cries which ended up in alerts, tag them with an ordered list of what you did to soothe or what was wrong, and see if AI can make anything of the data. Maybe certain cries can trigger alerts sooner, or cries over an extended period may indicate fever.