windexh8er

Can you change a newspaper ad on the fly?

Because ads will no longer look like ads. Model bias from the highest bidder, what fun.

This is exactly why OAI and Anthropic enjoy spreading FUD about open weight models.

One of the best ZWave mmWave sensors on the market is the Everything Presence Pro [0]. It has 3 motion sensors on board: DFRobot SEN0609 mmWave, HiLink LD2450 mmWave and Panasonic EKMC1603111 PIR. It has way more than this but the package is truly one of the best combinations of sensors and software. The sensor can do multi-target tracking and the HomeAssistant tuning software to define zones is next level compared to more popular sensors from places like Apollo. And I do like Apollo gear, I own a lot of it, but the quality and execution of the EST products is on a different tier of build quality for the price. Apollo just feels cheap in comparison and the reliability and accuracy of the sensors just isn't quite there, and to their credit they keep getting better. That being said they have a long way to go to catch up to EST. For true pro level installs the Everything Presence Pro is awesome, especially because of the 802.3af support. For battery installs the Everything Presence Lite [1] is also amazing. It only has the LD2450 is still really good and has multi-target tracking as well, but definitely won't support large spaces like the Pro.

I hate to even recommend EST because they're always hard to get based on lack of availability. But, you can't go wrong with those two sensors.

And, yes, most of the options listed in the article are garbage unless you're looking for one off options.

I'm not a fan of the Aqara gear as they've had some funky limitations on the software side and I've not had good results even compared to the Apollo [2] gear.

[0] https://shop.everythingsmart.io/products/everything-presence... [1] https://shop.everythingsmart.io/products/everything-presence... [2] https://apolloautomation.com/collections/sensors

Kind of amazing that given the state of the world money is still focused on remixes of different forms of social engagement. Is this because it's an obvious place to display ads? Or is it something else?

I'd like to think if I were JD that I would be focusing on something a bit more tangible for the world. A more useful aspect of tech to emerge or be invested in.

But, apparently AI agents in chat that sit adjacent to code is The Next Big Thing?

At least my original comment started from a point of reference. There's nothing in theshrike79's response that is of any value to improving the conversation, so why should they not be called out for it? Isn't that a part of HN's approach? I've been called out for it (rightfully) many times.

The fact that you don't think I realize what I did is also rather entertaining. Thanks for mansplaining, though! I was lost until you saved me.

First, those are all verbatim quotes from the educators. They were taken directly from letters of reference (which I recently had to compile for a quite intense Technical Due Diligence for a district). They're real school leaders. They're listed plainly on the website. You can look them up online. Whether they used AI in the creation or editing of their letters, I do not know.

This feels off to me as they don't feel like real quotes given they're all very similar.

The platform has been designed to minimize PII required. We have a very standard third-party vendor stack. Basically just OpenAI and Anthropic as frontier labs for the models, but neither is permitted to train on our data (we have Zero Data Retention agreements with those organizations). The other vendors are standard for operating the app (e.g., Supabase, Vercel). Nothing crazy here.

There are many sections in your "Privacy Snaphot" section where you don't explicitly say there is data retention. This to me feels like a convenient workaround to the other training callouts. You listed 4 vendors above yet none of that stack is called out transparently in any of your publicly facing documentation. Why? Do kids educations not deserve for parents and staff to understand where the data actually resides and how it's processed? That's the thing with technology in schools - there's always a loophole way for any company to collect and store data on kids. Their behavior, their persona, their intentions, their thoughts... We all know how egregiously Flock is being used against taxpayers. I suspect these types of learning platforms will also be used against kids in the near future and startups like Bloomy don't actually have the improvement of kids as their goal - the goal is $$$. And when you sell off Bloomy to Google or Meta, then what? You state no long term goals and you make no commitments to student, staff or parents. Those districts who trust you have no clue in how they're handing data over to you.

Not only can users request deletion of their data -- it's standard for any user who cancels their relationship, within 30 days. You can read it right in our privacy policy.

I've read a lot of ToS and Privacy policy for companies like this. They all fall apart when you ask for proof. So... How do you prove the data is gone? I'm a parent who requests my kids account do be deleted along with the data. How do you showcase all of this? I know that, from experience, nobody puts time or effort into this because it's not a money maker. But, yes, I can read your policy. Your policy actually says nothing to the extent of giving me any sort of confidence in how you handle student data. By all means please explain.

Finally, you don't address (at all) the impact study I referenced. So where is your data coming from? Any long term studies? Or are these metrics one offs? You don't share # of students, nothing. I don't understand how you build trust with kids data that is very proximal to AI systems that have not been proven safe with respect to young children. What is Bloomy's response?

Edit: I think it would be rather interesting for Bloomy to provide free accounts to some red-teamers to really understand if what you claim is true with respect to your pipeline of protections. I have zero confidence an educator building something like this has actually thought through data protection, model protection or any of the base level considerations you need when building a tool that handles user data, let alone data of minors.

100%

This looks very poorly thought out by the site alone. The quotes are all concerning as well because all three of them have emdashes. Maybe they're paraphrased, but then those aren't the actual quotes from those educators.

Parents and teachers who are willing to put middle schoolers in front of screens more are doing our future a disservice. There are real studies (none that I can find on Bloomy) showcasing how this negatively impacts students [0]. I doubt you'll find a correlation that more screen / compute interaction at younger ages improves long term cognition as it's not how kids develop.

Beyond this every parent should run after a glance at the privacy policy and the sheer number of different backend services required to run this. This is a privacy nightmare, long-term, for kids who have no choice. School districts, and a lot of parents, don't understand any of this. I don't even see how this abides by laws in some states simply because there are no protections of model training by some of the examples provided in the privacy policy. This does not abide by our state laws requiring a provider to be able to remove student data permanently within 90 days. In fact no LLM that doesn't have a "no train" clause abides by this simple requirement and it's unfortunate that districts will get sued over using tools like this in the near future.

My advice to this is: if you want to build a tutor then target parents. When you try to shove this down the throat of every student in a district you're making a lot of assumptions and impacting many negatively long term. This may be great for disadvantaged areas, but this is not how kids need to learn. They need to learn, explore and make mistakes on their own and not be preconditioned to the easy answer.

The iPad generation of parenting is about to get a whole lot darker as more and more of these types of companies are taking shots at easy targets, like school districts, with little to no studies behind impacts.

Finally, very expensive! I ran a simple 5 student cost calculation and it's over $2000 a year for those students. And that cost does not include math which is currently not available which would be another $1400.

[0] https://jamanetwork.com/journals/jamapediatrics/fullarticle/...

Ahh, yes another garbage comment that adds no value. Thanks for your contributions!

Except the transition was never necessary. Yet another Anthropic marketing effort so they could claim a shallow victory [0].

[0] https://andrewkelley.me/post/my-thoughts-bun-rust-rewrite.ht...

You can. Both pfSense [0] and OPNsense [1] have good API coverage.

[0] https://pfrest.org/api-docs/ [1] https://docs.opnsense.org/development/api.html

It's almost like Anduril has little to no background in building military soft/hardware [0][1]. The modern day charlatans are alive and living very well.

[0] https://www.techtimes.com/articles/320165/20260711/army-ar-g... [1] https://www.techbuzz.ai/articles/anduril-s-autonomous-weapon...

That feature experience often crops up as a choose-your-own-adventure game! Looks like their users are loving it [0].

[0] https://www.neowin.net/news/it-admins-feel-overwhelmingly-si...

Then use batteries as an example. Going on 200+ years now most of the major changes have been an easy swap: chemistry. Lead acid --> nickel cadium --> nickel metal hydride --> lithium ion. Improvements in batteries are hard and the surrounding ecosystem has gotten better. Better BMS and better motors have also made gains to increase the efficiency of the battery. Compounding gains within the ecosystem doesn't truly improve the battery directly, but improves them for implementation and use cases.

All the remaining problems with batteries are a hard solve because it's tradeoff after tradeoff. Lithium sulfur, for example, could be amazing but die after a couple hundred cycles at most [0]. So... If LLMs are so novel, as many continue to claim, why hasn't this been solved? Probably because LLMs are an interpretation of our current understanding of everything. An LLM currently only does what any human can do, albeit in a compressed timeframe, mostly.

I think it follows. LLMs have made some considerable improvements, but use "reasoning" (really just a mechanical, autoregressive, loop: generate tokens --> append tokens to context --> feed the expanded context back into the model --> generate the next token --> repeat) as an example. You can do this by hand as well in a traditional chat volley, but it's slower. So we improved LLMs by putting them in an automatic loop (oversimplification - but at the end of the day holds water).

Until LLMs showcase true external breakthroughs that aren't driven by human guidance (not happening anytime soon) we are in this loop of hacks being used to improve the 80% (the LLM itself). Notable jumps? "Reasoning" and with Mythos-like models we now have "Advanced Reasoning" (by leveraging a more capable looping framework such as an agentic harness).

[0] https://advanced.onlinelibrary.wiley.com/doi/10.1002/adfm.20...

It's a "big deal" to who? People in a bubble on HN? For sure. But the rest of the masses? Do they think it's a big deal, do they care? Really they shouldn't. AI has very little to show for benefits at this point for the everyday, average human being.

Not everything is best served by code.

This. I would never allow my kids onto a platform like Ello. My kids are starting to appreciate being shielded from social media and technology in their lives as they watch their friends go down dark paths and stuck in feeling like they have to participate there.

I don't think kids, of any age, should be in front of LLMs unless by complete agreement of their parents and only after their parents are made very aware of how this technology came to be, how it can be used and how it cannot be guaranteed to be safe. Ello leading with that is disingenuous from the start and that should tell parents all they need to know up front.

My kids read hundreds of books a year. They love reading, we read with them every single day as they learned. They learned to read the right way and that doesn't involve a screen.

Keep kids out of this garbage.

This is wildly dangerous comment and I'm glad to see it downvoted.

I have 2 kids. Both of them went through early grades during COVID. The change in learning and dislike of school was immense during that time-frame. My eldest missed out on the most and it's very clear that their age and time with which they should have been in-class and in-person impacted them.

Not only the screen-time aspect of this ignorance, but the sheer inconsiderate approach to what a model even is. It's not a human, it can't cope or understand a child's current state of being. And if you think it can then stay away from K12 tech, please.

Teachers make mistakes, yes. But can you swap a teacher with another teacher without a student knowing? No. You can with a model and that is dangerous. And not only the company. If Ello wants to they can move students to a Chinese model and nobody would know the difference. Or the OAI model they've chosen can be diluted or changed without Ello even knowing.

Kids deserve better. I truly hope Ello fails in its mission because companies shoving models into K12 are truly going to be the stain we all have to deal with decades from now. I am beyond grateful that my kids learned reading the "old fashioned" way and it speaks volumes that both of them read constantly. Their peers and friend's kids who are addicted to YouTube, TikTok and other socials struggle reading. It's so very obvious to anyone who has kids at this age these days and I think a lot of parents are now realizing their mistakes and are often times embarrassed when they see these kids side-by-side in public displays of their work.

I keep seeing this over and over. I am on a parent technology group at my kids school and it is a monumental effort to keep these "technology directors" in check. They are some of the easiest to sell to given they go into conversations with companies like Ello with nothing but trust. They don't consider data, privacy and security of these students or very rarely. They often don't understand relationships with respect to technology and the areas where issues can surface or be present.

I hate everything about this blog post and, again, I truly hope Ello fails before they contribute to the continued degradation of these kids early lives.

How are models improving your process? And how do you grapple with model inconsistencies?

Not saying you aren't good at your job but this statement gives me pause. Given what I see on the daily as output from models I hope there's still a lot of human interrupt behind the wheel with respect to the design of drugs.

And GMKtec just released an EVO-X3 [0] based on the same board but with Oculink. When AMD launched this I thought it was the new chipset that was going to have 192GB of unified memory, but they look to be capitalizing on the inflated premium these "dev" focused desktop solutions.

Regretting not buying the Framework back when it was around $2k.

[0] https://www.gmktec.com/products/gmktec-evo-x3-ai-mini-pc-amd...

It was national news [0].

[0] https://www.reuters.com/business/openai-proposes-handing-tru...

If neither model+1 or model-1 are providing tangible value to the business does anyone really care, though? At a certain point nobody believes Chicken Little.

I get it. These models can be powerful. But will they be useful is a different question.

They already are. Altman is basically begging the US to buy into OAI, that's just the start. Both OAI and Anthropic are going to have to go down this path or their financials will never work out. Open local models are where the enterprise will need to go for any of this to be cost feasible, but we can almost guarantee this will be a battle nobody using AI will have asked for. You can thank Dario and Sam for the dystopian future that will pad their bottom line!

We live in a non-deterministic world. Anything "deterministic" in it is a castle built on quicksand.

Except the Enterprise, and a lot of what people want compute for, is built on deterministic systems or processes. I'm not saying the non-deterministic nature of LLMs isn't useful. However I've worked with a lot of organizations on SOAR projects, for example. When you can weave the deterministic and non-deterministic together you get a relatively efficient system. A workflow that will stay on the rails and will come to a conclusion as expected. And the "as expected" part is critical in these types of systems. The reality of, using SOAR as an example, is also that most enterprise would be much better served by fast SLMs. Parse an email and validate if it's SPAM / Phishing or read a chunk of firewall logs and look for outliers / indications for escalation - those things can get messy in a deterministic system because of potentially unstructured data.

I don't believe it's either / or. And I believe that LLMs just aren't efficient, fast or reliable in the sense that deterministic are. It seems, at least to me, a better together story.

Yes, DDoS was definitely their entry point. I remember recommending them to a friend about a year or so after they had launched with the free tier. He was managing a small school district that was dealing with DDoS issues intermittently. What he needed was just outside of free at the time and I believe Cloudflare was still small enough where he had a call with Mr. Prince.

I was a strong proponent of Cloudflare for years, but looking back should have known better. I felt like others in the space would have tracked along how they went to market but that didn't play out as I would have suspected. I still use Cloudflare for DNS on domains that I use sparingly (mostly just for mail records), but no longer recommend anyone let Cloudflare terminate TLS unless they need it.

It's pretty amazing what you can get for a server host (bare metal) these days at the price point. I don't run any of those behind Cloudflare and haven't had any issues as of yet.

Yeah, I definitely remember Jennicam now that you bring it up. There were a lot of people vying for cam attention back then, I think it's basically the inception of the influencers we see today. Then again I don't know that anyone thought it was really something of a way to make a living. Maybe a very small few, but if any of those early day cammers/streamers had tried to get a discount at a restaurant for a positive review that would have been met with confusion for sure. I guess when one calls themselves an "influencer" they know what they're after.

Mine was also around '91. I lived in a small town and so while it seemed as though people started buying more PCs a few years later I was definitely one of the few who had access to the Internet early.

Forums and chat were captivating at the time. I remember timing my after school routine to be able to hit up a "chat room" of people I had found through a random forum. And then we found IRC which changed the game.

I also got a check pretty early on the Internet for banner ads I had on my site. That was around '95 or '96, I believe. I was amazed that someone would send me money for that. The site back then was probably popular because I had an early web cam and would often have it on while I was talking in public chats or on IRC. I feel like the Internet was friendlier back then, definitely not something I'd be comfortable doing anymore. But I remember continuing to collect those checks all the way through early college as the site changed, I ran a small forum, and started to write small how-to posts as I had gotten more intrigued with BSD & Linux around '98.

I'm surprised the timing of connection for the author, though. We had dial up first, obviously. But I got a cable modem around '96 or '97. 1Mb/s down (no idea what it was up)! Game changing for sure. Today I have symmetrical fiber to the house, yet it's not fun like it used to be. It's turned into a commodity, a utility you just require as the author points out.

I think the Internet for me changed around the time the first iPhone came out. Prior to that I feel like the Internet still had character and most generally didn't have access to the Internet from their phone, or if so it was very limited. The mobile web back then was still pretty bad, especially with all of the heavy browser components mobile devices definitely couldn't handle. Flash, Silverlight, Java, etc.

I've spent time with my kids to show them things on the Internet but for them it's very different. Access is assumed and it's generally looked at like I looked at FM radio or broadcast TV. It's hard to get excited for them when my main concern is making sure they know about data, privacy and general security. Very different indeed and feel lucky to have experienced the early Internet.

I often wonder what those people are like IRL. I'd surmise they're the people that are easy to hate. Greedy and intolerable yet want to be the focus.

The point is Anthropic has advertised their models in this way. There are plenty of models that can be used in long running situations that have proven to be more capable. Opus 4.8 is not that, and ironic given it's their top public model.

Except for the fact that Opus 4.8 is not good. Constant hallucinations, doesn't use the web very intentionally until you explicitly ask it to and it nopes out rather quick on benign items. Anthropic has been very disappointing as of late. All of the gatekeeping is taking a toll on what should be some of the better models out there, but you can't trust 4.8 to go off on its own. It will burn down tokens doing what it deems correct as per its guidance. Truly painful to use.