HN user
The default: https://code.claude.com/docs/en/sandboxing#filesystem-isolat... already restricts writes to only the current folder. I can understand adding the "denyRead" for the home folder for additional security, but the other three seems redundant considering the default behavior.
Why does it seem like many parents are unaware that a hand me down iPhone can be heavily locked down with screen time settings? A list of things you can do:
- allowed list of apps, can reduce it to just phone, imessage, and utilities like weather app
- effectively permanent downtime, just set the end time less than start time such as 3:00 am to 2:59 am (technically 1 minute of non downtime). This blocks apps except for the allowed apps
- disable installing apps from app store
- disable adding new contacts and block calls and messages not in contact list. This allows parent to control who the phone can be used to contact
- none of these settings can be changed without the screen time pin
- also configure the phone with a minor apple account and add to your family group so you can monitor and control screen time settings from your phone.
So start with a super locked down phone that can only be used to communicate with parents. This is very helpful when they start after school sports. And the phone is so locked down they don't really have any interest in it.
Later when they're older start allowing communication with friends from school. But still only phone and imessage, no other apps. This reinforces that it's a communication device, not for endless scrolling and watching videos.
This whole discussion reminds me of the beautiful design of UTF-8. They used the lower bits to be ASCII which made backwards compatibility so much easier. It also reminds me of the failure of Intels Itanium and the success of AMD x64. Engineers often want to abandon backwards compatibility to make a new "beautiful" design, but it's the design that has full backwards compatibility that's actual impressive.
When you hear about unions in software, stop thinking about auto-workers and think about NBA players instead:
Lebron James and Stephen Curry are in a union and they don't seem to be having any issues making a lot of money.
Or maybe something more similar to software development, the screenwriters guild:
Again, there are many rich screenwriters, Google for a list of the top paid and it's obvious being in a union hasn't stopped high compensation.
Amazon already openly stated that there is no data to support RTO:
https://www.businessinsider.com/amazon-andy-jassy-no-data-re...
https://fortune.com/2023/08/03/amazon-svp-mike-hopkins-offic...
These are also hard to measure:
* Employee happiness
* Less sick employees since they don't spread their germs in an office
* Much lower attrition and retention of institutional knowledge
* Lower rent costs or possibly zero rent costs for office (actually this one is very easy to measure)
* Able to hire from outside local metro area
None of these was enough to move companies even an inch towards WFH pre-COVID. And yet now vague issues due to lack of water cooler conversations is enough to shift everything back to in-office?
This line from the article "...in the number of calls handled per hour by employees of an online retailer that had shifted from offices to homes.." shows that this study is about a call center. A few points on this:
* The discussion here is about tech workers, not call center employees
* Here's another article from 2014 that showed a 13.5% increase at a different call center https://hbr.org/2014/01/to-raise-productivity-let-more-emplo... So study vs study, which one is correct or better? This isn't good evidence either way for software engineers.
The data and evidence we need is from the loud RTO companies (Google, Amazon, etc.) in the software industry pushing for RTO. These supposedly heavily data and metrics driven organizations have NO DATA supporting their RTO efforts. Some random study about call centers is irrelevant here.
Where's the data? Before COVID, there were plenty of anecdotes from remote companies about how it helped their hiring and productivity, but that wasn't enough to convince the vast majority of companies to try WFH. They stubbornly said the status quo of in-office was enough and no further discussion was allowed.
Now where's the data to change the status quo from WFH to the office? Amazon admits they have none. If the other companies forcing in-office had data they would be shouting it as much as they could, but when asked for data, it's just silence. Companies have had record profits and quarters with WFH, so clearly the financial data shows no issues with WFH.
Again where's the data? All we hear are anecdotes, that wasn't good enough to change the status quo before COVID, why should it be enough now to change the status quo away from WFH?
What was the #1 argument against WFH before COVID? It was fear of productivity drop, that the company simply can't function with WFH. Then COVID happened and companies worked fine for three years with WFH. At this point, it shouldn't be called RTO, it should be called STO (Switch To Office), because WFH is the default existing state. And the companies that want to STO, they admit there's NO DATA to support this:
https://www.businessinsider.com/amazon-andy-jassy-no-data-re...
https://fortune.com/2023/08/03/amazon-svp-mike-hopkins-offic...
The hypocrisy is obvious, they were all so against WFH before COVID, demanding data that it would work. Now it's working fine for three years, and yet they switch to office with no data, a simple "gut feeling" argument. It's indeed bullshit.
We don't need to guess at this, companies like Amazon already admitted there's no data to support RTO:
https://www.businessinsider.com/amazon-andy-jassy-no-data-re...
https://fortune.com/2023/08/03/amazon-svp-mike-hopkins-offic...
USB 2.0 is a missed opportunity for bringing Stage Manager/screen mirroring feature from iPadOS to iPhone. This is already possible with iPad and a USB A + HDMI -> USB C adapter (plug keyboard/mouse into USB A, monitor into HDMI, and USB C into iPad).
Main reason why I like this change: single cable in the car for both Android Auto and CarPlay. It was so annoying having both in the car.
California has CCPA https://en.wikipedia.org/wiki/California_Consumer_Privacy_Ac... Could this be used to force these companies to delete all personal data? It would have to be done periodically since after deletion the data would accumulate again. It seems like there's a potential business idea here of automatically sending out CCPA deletion notices to companies on a schedule. While this wouldn't stop the collection of data, regularly interrupting them with deletion requests could make storing personal data costly enough to at least reduce whatever profits they would get from it.
EDIT: Looked at a few privacy policies and the CCPA link is often hard to find. Keywords to look for: "CCPA", "California Privacy", some examples of links I found:
https://www.honda.com/privacy/your-privacy-choices
https://www.tesla.com/legal/privacy#data-sharing
https://www.ford.com/help/privacy/ccpa/
https://ksupport.kiausa.com/ConsumerAffairs/PrivacyManagemen...
Something interesting I found is also this: https://www.honda.com/privacy/CCPA-Metrics which shows how many requests Honda received. It seems not many are aware of CCPA rights and this number of requests is not enough to deter companies from gathering personal information. These metrics need to be orders of magnitude higher to make a difference in company behavior. It seems like an automated service to send these requests and more public awareness of CCPA could help here.
EDIT2: A lot of these forms ask whether you're submitting the request for yourself or you're an authorized agent doing it for someone else. I found more details on "authorized agents" on the CCPA FAQ: https://oag.ca.gov/privacy/ccpa. Maybe an organization like Mozilla or EFF could setup a service where you can authorize them to do this for you? Then you could just select a checkbox of companies that you want CCPA deletion requests for and it would be sent on a regular schedule (quarterly? yearly?). If such a service became popular, it could really disrupt the personal data gathering of companies.
I was thinking about a similar idea while reading the HN discussion on NASA's Deep Space Network issues. For the Artemis moon mission, instead of wasting precious DSN bandwidth, wouldn't it be better to simply store video on a hard drive and send it back to earth? That would even allow recording in full 4k. Same for things like the James Webb telescope. The latency is certainly much higher (a few days for hard drives to return to Earth and get sent back on rockets), but it seems the orders of magnitude higher bandwidth would be worth it
People in this thread are really talking past each other. I've been to the nice Asian mega cities with great and clean subways and buses. And I've lived in the American suburbs. You can't make the American suburbs like the mega cities by just making them walkable.
Everything in a mega city works together to make transit work. Those tall buildings? They provide great shade no matter how sunny it is which is critical for walking to bus stops and subway stations. Also, the walk itself is so much more interesting, random stores to stop at and places to eat and go to. Density makes transit work.
You can't just put random stores in a suburb and make it "walkable" and expect the same thing. Just as everything in a mega city works together to make transit work, everything in a suburb works together to make cars work.
We need to give up on the mass transit solutions that work for dense cities (subways and buses) for suburbs. It's a waste of money and completely the wrong solution. It hasn't worked for decades and never will.
Shut down bus systems for suburbs and use the government funds to give out ride sharing (either Uber or government run) credits for everyone to use (low income can get more credits). That's what a suburb is designed for, point-to-point travel such as cars. And invest massively in real protected, useful bike lanes and stop trying to kill e-bikes with regulations (which a lot of cities are trying to do). e-bikes are finally a real alternative to cars in suburbs, it has just the right amount of travel speed and ease to challenge the car, but it's already under attack. Ride sharing credits and e-bikes, these are the solutions for suburbs. Stop trying to fit a square peg (buses and subways) into a round hole.
We need to stop thinking carbon chemical fuels are the problem. FOSSIL fuels are the problem, not carbon fuels itself.
We should be blanketing every inch of desert with solar. And pair it to use excess energy for carbon fuels synthesis. Reuse all the existing natural gas power plants to run on synthetic carbon fuel. Batteries are not the solution for this. It only fixes the day/night imbalance, but not the seasonal summer/winter imbalance for solar production.
But most importantly, we don't have time. We don't have time to wait for the beautiful, elegant solution of all cars EV, all power storage in batteries, all planes flying on electricity. Perfect is the enemy of good. Look at the arctic and ocean temps, we do not have time. The developing countries will not wait for the perfect nice solar and battery solution. We need to reuse as much of what we have now in a way that will make a difference for carbon output. Again, we do don't have time for the most efficient solution.
What is industry good at? Mass producing a lot of stuff. We can do that now with solar. Stop worrying about matching it to daily power usage. Just pump out those panels and get it installed everywhere. Get the excess into synthetic carbon fuel and we can quickly make a difference in carbon output.
This changes my perspective on the Android phone support situation. I always thought it was Qualcomm's fault because they didn't update proprietary binary blobs and so it was impossible without Qualcomm's help. But according to what Fairphone is doing, Qualcomm only stops making it easy, it's still possible for others to keep it updated. So Google and Samsung and others which have massive resources could easily keep a phone supported if they wanted to.
This is why I think it's ridiculous when car rental is recommended as a solution to the range issues of EVs. There is so much friction, have to shop around to get a good rental price, pickup and drop off the car and ask someone to pick you up (or uber), very unpleasant waiting at the counter, mileage restrictions, driver restrictions (can't ask your friend to help with taking turns driving on a road trip). Compare all of that to just getting in your own car and going. If deciding on whether to go on a road trip to somewhere like a national park, this can mean the difference between going and just giving up because of all the hassle in renting cars.
EDIT: I'm not against EVs, just against dismissing the EV issues with ideas like "go rent a car" as if that's a realistic solution considering how inconvenient it is.
That's a great article and shows exactly the issue. The bus system has become the transportation of last resort. It's not meant for this and isn't good at this. It's only good at highly used routes that can maximize bus capacity. Let rideshare handle these social safety net transportation issues. Ask a low income rider using the bus system, would they prefer to continue using it or get a free rideshare card with a few hundred dollars in it every month? Most would probably very enthusiastically go for the rideshare card.
And I know rideshare has a bad reputation because of Uber/Lyft. But government can create rules they must follow to accept these funds. They can either accept those rules or some other company will. Or maybe a government rideshare service is a better option. This is a classic free market vs government service question and there's many options here. But the important point is allowing everyone to use rideshare as a transportation option.
This whole thread exists because the public bus system is a failure in the US. How many threads on car dependence has there been on HN? The bus system is used only by the lowest income members of society who can't afford a car. They suffer long transit times, lack of point-to-point mobility, and delays. Sure that "works" to some extent for those who have no other options. And this leads to other effects like decreased health and social mobility. Want to go to college after work to improve your life? Can't because the bus routes take 2x or 3x the time it would in a car to get there. Want to get a checkup for that cough? Again riding the bus takes too much time.
You can't force a top-down solution for public transit with the road system in the US. The great strength of the US road system is point-to-point transportation. Let everyone benefit from that instead of running buses that only the poorest use. Publicly funded rideshare is the way to do this. After a while, the bus routes will naturally appear in the data. This is the bottom-up way to build a bus system.
Here's a radical idea: disband and shutdown the public bus system. Before you reply with an angry post, read the rest of the plan. These systems take hundreds of millions of public funds and are completely ineffective in suburban areas (most of the country). Take that money and give a "rideshare card" with funds automatically filled every month (lower income will get more free funds). Either work with Uber/Lyft or start a similar government rideshare service. Something like this will actually get people to consider giving up their cars.
After a while, certain high usage routes will be noticed in the rideshare data. It will become obvious which streets and destinations could be optimally served with high capacity buses. Now is the time to bring in bus routes. Setup these bus routes and offer a discount for using them.
The current system isn't working, we need to try something different.
Does the California CCPA apply here? I've only seen it discussed for websites, but does it work for any company doing business in CA? Wondering if California residents could send a delete request to Toyota (and other companies like Samsung for data gathered by smart tvs).
If a lot of people start regularly sending CCPA delete requests to these companies maybe they will stop gathering this data.
We have let the perfect become the enemy of the good. With the urgency of climate change, current grid infrastructure, and battery technology and production, the solution we needed to push hard was PHEV. For the battery in a single EV, it can build four PHEVs which means four households reducing most of their gas miles versus a single household. A common strawman is that a PHEV is more complex and harder to maintain. And yet that theoretical argument is directly countered by the existence of the Toyota Prius, one of the best for reliability, so good that it's used in taxi fleets. A PHEV is basically a Prius with a much larger battery. Another strawman is that there wouldn't be demand. And yet there is incredible demand for the Toyota RAV4 Prime. A PHEV with 40-50 miles EV range can be charged overnight with a simple 120V outlet, no expensive electrical upgrade required.
There are two versions of the S22, Exynos (Europe) and Snapdragon (everywhere else). Is it correct to assume that the Snapdragon S22 is not affected by this?
Why is this still true? I can understand in the past, but after the rise of all the tech companies and obvious important software they use everyday (Android and iOS) how can anyone at this point think software is a joke and lesser than hardware?
They're focusing on transportation fuel (cars and airplanes), but another area of great potential is power generation. The current trend is to build solar/wind and replace coal with natural gas plants as a stop-gap until some grid-scale energy storage is ready. Everyone assumes that energy storage will be batteries.
But what if the natural gas plants don't have to be a stop gap? Just keep building more and more solar/wind, as much as the land can handle (imagine most of the desert in California converted to solar). Who cares if generation greatly exceeds daily demand. Use all the excess solar/wind to create fuel for the natural gas plants. There's already a vast infrastructure and experienced workforce to do this. Use the fuel during the evening and put any excess fuel into storage, there's so much existing ways to store fuel. Then use that during winter when solar generation decreases.
We need to stop thinking carbon fuel = fossil fuel and so carbon fuel = bad. Carbon fuel is simply a form of energy storage, a kind of "battery".
The hype for Netflix stock was based on them becoming the Spotify of video and a tech advantage in streaming. It became clear a while ago that the studios wouldn't allow this and they no longer have a tech advantage. The market is finally adjusting their stock to match reality.
Now they are fighting a battle they can't win. HBO, Disney, etc. have huge back catalogs that immediately makes their streaming services more valuable, in addition to all the new content they make.
They shouldn't have given up so quickly on physical media. Physical media does not have the licensing issues. They could've invested heavily like the way Amazon did on lowering delivery time. Maybe they could've gotten physical media delivery down to one hour or same day. This would counter the studios back catalog because while Disney can instantly stream their back catalog, it's only Disney. Netflix would have access to back catalogs of all studios (at the tradeoff of waiting for the physical media) That's a tradeoff that many customers would make (access to more back catalog diversity in exchange for the delay of waiting for disks).
At the same time, they could produce streaming content with a different business model. How many times do customers complain about favorite shows getting canceled after a season? What if there was a way for customers to select which shows they want supported? Like pick your top five every month and your subscription money will go to that show.
Instead, they went the route of just spending a lot of money making new content as if they're a studio. But they can't beat the studios at being a studio, Disney and the others are the best at that.
The current release is 102.0.1, 104 is Beta: https://wiki.mozilla.org/Release_Management/Calendar
At this point, probably, but in the past when FF had 30% market share? It would've at least forced a serious discussion in the industry on what are acceptable ads. Or maybe it would've been the impetus for a working microtransaction system built into the browser. Instead, they did nothing with their influence.
Firefox had the chance to remain relevant when it still had some significant market share and if they had gone all in on ad blocking (basically something like merge ublock origin directly into Firefox). Unfortunately, since most of their revenue is from Google, it could never happen due to conflict of interest.