4,318 karma · joined April 27, 2013
But yes, iBooks did start with the same design: https://www.redmondpie.com/ibooks-3.0-all-set-to-be-announce...
If there's a blindness in this article, it's not so much the trucking industry response as it is the question of whether or not this has been implemented anywhere else in the world? https://www.reifflawfirm.com/canadian-and-european-truck-und... suggests that, "in Europe since 1994, front underride guards and side underride guards have been required on large trucks," and that these were strengthened in 2007.
That same page suggests front and side (of truck) collisions with a passenger vehicle are more likely than rear collisions, and that existing US and Canadian protections aren't strong enough.
If you opt in, a simple page like this ends up looking pretty modern anyway, but plain, while if you opt out, your site is often rendered as if made for an older desktop monitor, like this one.
Another way of putting the problem, we preserved old desktop styles rather than invent new default responsive styles for touch-based browsers.
Without a standard like grey backgrounds and blue links to fall back on, websites had to start making decisions, especially when sites go responsive. At that point it is easier to invent new styles… there aren’t that many useful defaults to fall back on.
This is particularly true for HTML5 form controls, which rarely look good in any browser without styling…
But perhaps there’s hope for the future as new design language becomes HTML tags. E.g. details has an arrow, datalist builds a native autocomplete, and so on. It’s a shame that dialog doesn’t have native formatting tho, or that the idea of creating new, modern default styles hasn’t caught on yet in a cross-browser way.
Normally my favourite method to switch apps is Spotlight, accessible with ctrl+space or cmd+space, I forget which. Anyway, Spotlight is a great app switcher, you can disable most everything else and just type a letter or two to switch. I use Powertoys on Windows to get a similar switcher there.
But again, Spotify breaks this, and switching using Spotlight does not recreate the Spotify window, if I remember correctly. Once again reaching for the dock…
Contents
The Hunttinge of the Hare.
A Mock Sermon.
...
From there we can flip forward to the actual content and ... yep, can't read it. Honestly the only reason I know it says "huntinge of the hare" centred at the top is because I read it in the table of contents.I can tell it's English, but that's about all I can tell, and that only because I can read some of the words, but not all of them. :) Maybe I can send the image to GPT-4 for transcription and translation. (lol)
FYI, Atlassian has a non-profits license. While the cloud version only has a discount so you still have to pay per user, the on-prem version is free. Best part about running on-prem under this license is that a lot of on-prem extensions have opted in to atlassian marketplace licensing and offer free plans for non-profits. While this would be expensive to do in the cloud, because the plugins run on your server, it’s basically free for everyone involved. The downside is you still have to manage and renew licenses, and don’t expect too much support given the free price.
> Atlassian also offers free Data Center (self-managed) licenses for registered charitable non-profit organizations that are non-government, non-academic, non-commercial in nature, non-political, and have no religious affiliation.
More at https://www.atlassian.com/licensing/purchase-licensing#trial... under the faq question about non-profit pricing…
- extreme corporate profits
- better wages
- US jobs
- lower prices
You can’t optimize both for extreme shareholder returns and extreme wealth at the top and also pay first world good wages/living standards and have lower prices. Unfortunately, in this race, shareholder returns is often still the last thing compromised on, it seems.That said, I also find it somewhat unfair, as likely does anyone not in the US :)
For comparison, here’s an article from a fan: https://twitter.com/ejdickson/status/1656425063321677825 - or for the direct link, https://www.rollingstone.com/culture/culture-features/heathe... Not that the article is any less frank about it, but at least it addresses some other points.
And it’s not like there aren’t other posts from contemporaries to quote. There are, just search for them: https://twitter.com/anildash/status/1656344993995411456 for example.
This is more what I expected to read: https://www.washingtonpost.com/opinions/2023/05/10/heather-a... or https://www.washingtonpost.com/technology/2023/05/10/heather...
But I wanted to clarify about video recording and viewing 100% of clips:
there are two service levels from Google: either you’re grandfathered on older pricing plans per camera and can see a 24/7 recording or you upgrade to new plans per household that they sell now where the camera only records and sends motion or sounds, not 24/7 feed recordings like the old Dropcam used to.
You can’t buy the old way of recording 24/7 from every camera as far as I know, and the only reason I can think of that they switched recording modes is because it saves power and is more efficient so hardware might not break as often via overheating. I mean, they might claim it is more efficient for storage, but they have stationary cameras - compression has to be good, right?
Anyway, just pointing out that generating clips from video recordings is kind of the old way, and the new way is that it somehow automatically makes clips for you as motion alerts.
But… outside of that I agree 100% that Nest has been basically dead since the rebranding. That Ecobee still has any unique features Google doesn’t have largely stems from this lack of innovation.
Ecobee is now owned by a company that makes generators, the kind that runs on gas or diesel. It’s hard to imagine that Google can’t ship new features, apps or software products faster than a company that makes industrial appliances and yet here we are.
Google makes too much money from ads to care about any other way to make money, long term. I’m shocked they don’t just give away these smart devices and wire them up to play YouTube ads with facial recognition at this point.
Maybe they’re waiting for Amazon to try it first, or want to experiment with Google TV ads like Fire TV and Samsung already do. If my Apple TV ever starts auto-playing trailers and ads, I’m going to permanently switch to using AirPlay, I think.
As for the Dropcam, maybe I’ll replace it with an iCloud video camera when the time comes…
The theoretical best approach would be to integrate lots of GPU cores on the motherboard alongside very fast memory/storage combos such as Octane, but reality is very different because we also want portable, replaceable parts and need to worry about silly things like cooling trade offs between placing things closer for data efficiency vs keeping things spaced apart enough so the metal doesn’t melt from the power demands in such a small space. And whenever someone says “this is the best graphics card,” someone inevitably comes up with a newer arrangement of transistors that is even faster.
You need to load the data so the graphics cards - where the compute is - can use it to answer queries. But you don’t need a separate copy of the data for each GPU core, and though slower, cards can share RAM. And yet even with parallel cores, your server can only answer or process so many queries at a time before it runs out of compute resources. Each query isn’t instant either given how the GPT4 answers stream in real-time yet still take a minute or so. Plus the way the cores work, it likely takes more than one core to answer a given question, likely hundreds of cores computing probabilities in parallel or something.
I don’t actually know any of the details myself, but I did do some CUDA programming back in the day. The expensive part is often because the GPU doesn’t share memory with the CPU, and to get any value at all from the GPU to process data at speed you have to transfer all the data to GPU RAM before doing anything with the GPU cores…
Things probably change quite a bit with a system on a chip design, where memory and CPU/GPU cores are closer, of course. The slow part for basic replacement of CPU with GPU always seemed to be transferring data to the GPU, hence why some have suggested the GPU be embedded directly on the motherboard, replacing it, and just put the CPU and USB on the graphics card directly.
Come to think of it, an easier answer is how much work can you do in parallel on your laptop before you need another computer to scale the workload? It’s probably like that. It’s likely that requests take different amounts of computation - some words might be easier to compute than others, maybe data is local and faster to access or the probability is 100% or something. I bet it’s been easier to use the cloud to toss more machines at the problem than to work on how it might scale more efficiently too.
Original post follows: I haven’t read any white papers on how this works, but I’m going to call “slight BS” anyway. Technically, Apple knows the location of the AirTag, that’s how their network works. The payload of airtag data might be encrypted, but Apple also keeps the keys for your AirTags so that you don’t have to associate new AirTags with each Apple device you want to use Find My on. While I don’t doubt that Apple strictly reduces what information they log or identify about AirTags to literally the bare minimum that keeps the service running, it’s a bit ingenuous to suggest Apple doesn’t know. What I mean by this is Apple isn’t lying when they say they don’t know the location of your AirTag (until you use their app to look for it) but it is true they know the location of all AirTags synced to their network, they just don’t know which one is yours until you go looking for it. Encryption probably plays a role in that asymmetric encryption would allow Apple to not know any AirTag details until your devices provide the decryption key. A handshake could be involved too so that your device can’t decrypt the location of the tag without proving they own the associated encryption key. But presumably Apple has a way of issuing or remotely authorizing against such keys. It is likely the case that your password for your phone protects Apple from gaining access to the keys thus justifying the position that Apple doesn’t have access because they don’t have your keys. Apple would have to act maliciously with their software into order to break this system, and it’s not worth it for them to do so. I’m not saying they’re lying but they are simplifying things a bit.
0.24 L | 23.66 cL | 236.59 mL | 236.59 cm^3 | 0.0625 gal | 0.5 pt | 0.25 qt | 8 fl oz | 16 tbsp | 48 tsp | 0.00836 ft^3 | 14.44 in^3
My “measurement system” under language and region in settings is set to “Metric”. Perhaps yours is incorrectly set to “UK”?
I’m also running iOS 16.5 beta, if it helps.
If you want an easy all-purpose login flow, sign in by clicking a link in email is still an easy fallback.
OAuth from multiple providers also has complexity - Auth0 makes it so you have to maintain separate databases for passwords if you want to support login via OAuth and login from password. You have to link accounts to login from multiple providers.
Logins are simply hard work no matter how you slice it. Eventually they will be easy but… I’m not holding my breath. :)
Plus there was a time, albeit almost a decade ago now when the dollar was on par. Even about a year ago the dollar was still aiming for 0.8 rather than the 0.74 it’s flirting with right now.
Lastly, if you want an Apple Watch with cellular, you end up needing a more expensive plan from the big three rather than their sub-brands. And the Canadian government is threatening regulation to encourage carriers to offer more data for less. There was a time within the decade when $60 at Fido would get you 2GB and that was during a “double data” promo. This was 2018. https://forums.redflagdeals.com/fido-60-unlimited-canada-wid...
> In summary, you should use read_message when working with packed messages and deserialize_message when working with unpacked messages. Make sure you choose the appropriate serialization and deserialization functions based on the format in which your messages are stored or transmitted.
Google was surprisingly little help on the topic. At best it pointed me to https://capnproto.org/encoding.html#packing which covers the general idea but glosses over parts.
The problem with ChatGPT is you can never be certain or confident that its answers are correct. It’s as useful as rolling dice in guessing a number sometimes. With the training from the internet, the dice are loaded, but the answer is still likely to be wrong because it just assembles words together.
If you ask for something impossible in a library it will also frequently make up functions or application settings. If you ask for something obscure but hard to do, it might reply that it's impossible but it is possible if you know how and teach it.
I sort of compare prompt engineering to Googling - you sometimes have to search for exactly the right terms that you want to appear in the result in order to get the answer you're looking for. It's just that the flexibility of ChatGPT in writing a direct response sometimes means it will completely make up an answer.
There's also a limitation that the web interface doesn't actually let you upload files and has a length limit for inputs. For Copilot, I'm looking forward to Copilot X: https://www.youtube.com/watch?v=3surPGP7_4o