It took them years ago. From 2015 : An NPR Reporter Raced A Machine To Write A News Story. Who Won?

And its only gotten better: Can ChatGPT write a podcast episode? Can AI take our jobs?

Unless you specifically pay for the rights transfer (and it’s not cheap), the photographer owns the copyright.

https://www.rocketlawyer.com/business-and-contracts/intellectual-property/copyrights/legal-guide/wedding-photos-does-your-photographer-legally-own-them

Under federal law, if there is no agreement to the contrary, your wedding photographer, or any photographer for that matter, owns the pictures that they take. This means that they have the sole right to copy and distribute the photos, including potentially the right to sell the photos, to publish the photos in any form, and to reproduce the photos either electronically or in a printed hardcopy version. And even more importantly, copyrighted material cannot be reproduced or copied without permission from the photographer.

…

Generally, photographers do not like to offer their services to clients through a Work for Hire Agreement. This may be partly related to their desire to require clients to purchase prints and books directly from them. Many photographers, however, do not want to completely relinquish their rights because they may be trying to build or protect their reputation.

Granted, this a US take and may vary by country…

https://www.thecoffeetablebook.com.au/what-do-a-bride-and-groom-need-to-know-about-copyright-when-booking-a-wedding-photographer/

If you are a couple getting married in Australia, the copyright law automatically assigns copyright to you and not the photographer. However, most professional photographers will have their clients sign a contract that reassigns the copyright to the photographer. Now let’s be very clear, this is not the photographer being shady or deceptive in anyway. It’s simply to protect their work, the photographs in this case, that they created.

2- positive supply-chain validation. Not important for the majority of people, but for those who require a little more security, they can be a little more sure that their device isn’t compromised from illegitimate parts. I imagine this to be a fringe benefit for executives and the like

Anecdotally, many years ago I had a Mac laptop that was fairly new but I ran it as what amounted to a portable desktop. It was plugged in 99% of the time. A few months after I got it I had an issue with one of the fans. Instead of taking it to the Apple Store that was 90 miles away (I was living and working in a small town then), I took it to the local computer repair… and they fixed it it.

Afterwards, while I didn’t use batteries much, it seemed off.

Another two years and I had moved to a larger city and took my laptop into the local Apple Store for them to replace a fan that was not operating strongly enough. The repair tech commented that I had a sub par third party and while it should be replaced as its battery health was very low, it wouldn’t be covered under warranty because it clearly wasn’t an OEM Apple part.

I strongly suspect that the repair shop had swapped my new battery out for an old one and hoped I wouldn’t notice because I didn’t use it in that capacity. I can’t prove anything and the shop had gone out of business.

This is anecdotal and batteries for laptops don’t get pairing the same way phones do… but this sort of thing happens.

It’s not outputting ionizing radiation… nor is it outputting radio near the limits that are causing it to get removed from the market except under laboratory conditions where its being forced to operate at maximum power that the transmitter is allowed to (by the software) for extended times… which could get fixed by software… and is 10x under the limits that would theoretically be a problem under the levels that it was observed at… which is being tested at a much closer distance (inverse square getting in there) compared to other accredited laboratories in the EU (EU is up to 25mm, FCC is 15mm, France is doing 0-5mm) without changing the absorption levels to compensate ( https://www.anses.fr/en/content/exposure-mobile-telephones-carried-close-body )

And France has been pushing a “beware of excessive cell phone use” for a long time. https://arstechnica.com/uncategorized/2008/01/france-beware-excessive-cell-phone-usedespite-lack-of-data/

I was just citing the sources in France that was showing that this was taken a bit to the extreme. When you consider all the rest of it, it feels rather silly and Apple is likely doing a “whatever” and pulling a 3 year old phone from the French market.

You don’t need iodine, this isn’t going to give you radiation sickness, but it is a little surprising.

From a WaPo article:

France’s digital minister said the iPhone 12’s radiation levels are still much lower than those that scientific studies say could harm users, and that a software update could solve the compliance issue. The ANFR itself acknowledges that its tests don’t reflect typical phone use.

Which links to:

https://twitter.com/jnbarrot/status/1701848091689693576 (translated)

The @anfr found that the iPhone 12 was emitting a level of waves slightly higher than the authorized threshold.
This level is more than 10 times lower than the level at which there could be a health risk.
But the rule is the rule: Apple must comply.

https://twitter.com/jnbarrot/status/1701852521247965686 (translated)

The wave levels that smartphones emit can vary during a software update. It is undoubtedly because of an update after its release that the iPhone 12 exceeded the authorized threshold. And it’s a simple update that will bring it back into compliance.

https://www.anfr.fr/maitriser/equipements-radioelectriques/le-debit-dabsorption-specifique-das/das-reel-/-das-mesure-maximal (translated)

The classic conditions of use: the real DAS

Laboratory measurements do not reflect what happens during the usual use of the device. Indeed:

• for voice communication, the phone statistically emits only about 50% of the time, the phone does not emit when listening; in addition, the average duration of a communication is less than 3 minutes;
• for data-oriented use (internet or video), the durations of use are certainly longer; but the phone, which generally receives more information (videos, emails…) than it sends, rarely emits more than 10% of the time during the session;
• the value of the DAS displayed in the points of sale corresponds to a precise frequency (e.g. 800 MHz) and technology (e.g. 4G) for which the maximum has been detected; but, in reality, the phone does not work permanently under these conditions: it often changes frequency or technology;
• Finally, for all uses, the phone rarely emits at its maximum power, due to constant interactions with the network to best adapt to its conditions.

The real SAR therefore does not often coincide with the maximum SAR.

As the DAS is proportional to the power emitted by the telephone, it is possible to evaluate the actual DAS by measuring the power put into play by the terminal. This can be evaluated thanks to professional software installed on consumer phones

Apple’s business plan isn’t to sell your data to advertisers but rather to sell you (presumably) high quality items that people are willing to pay a premium for.

To Apple, having your data is a liability rather than an asset and thus they’d like to have as little of it as possible yet still being able to offer you the Apple experience.

Task Warrior ( https://taskwarrior.org ) and syncing it to https://inthe.am

Task Warrior is open source ( MIT license https://github.com/GothenburgBitFactory/taskwarrior.git ) as is InThe AM (AGPL https://github.com/coddingtonbear/inthe.am ) and can be spun up without too much difficulty for self hosted options.

There’s a pair of videos from People Make Games that digs into it:

• Investigation: How Roblox Is Exploiting Young Game Developers - https://youtu.be/_gXlauRB1EQ
• Roblox Pressured Us to Delete Our Video. So We Dug Deeper. - https://youtu.be/vTMF6xEiAaY

Philosophers haven’t come up with a good way to determine if you were the only conscious being in a universe populated with zombies or not. See also https://en.wikipedia.org/wiki/Philosophical_zombie

I’d also suggest going through https://www.3blue1brown.com/topics/neural-networks to understand some of the concepts behind it. The ‘regurgitating data’ isn’t entirely accurate (and that statement is likely to evoke some debate). The amount of source data is compressed far too much for it to be returned largely intact (see also https://en.wikipedia.org/wiki/Kolmogorov_complexity for some bit on the information theory)… though yes, there are some situations where specific passages (or images) get memorized.

AI can be an amazing tool in healthcare, as a double check. For example, assume a doctor thinks you have something. Right now you could have …

Expert systems have been available as part of medical diagnoses for decades. I remember ahem finding one for the Apple ][+ back in the day.

https://pubmed.ncbi.nlm.nih.gov/2663006/ was written back in '89 and you can easily find others going further back.

AI in medical diagnostic capabilities is nothing new or surprising.

ChatGPT works off of a fixed size possible maximum prompt. This was originally about 4000 tokens. A token is about 4 characters or one short word, but its not quite that… https://platform.openai.com/tokenizer

“Tell me a story about a wizard” is 7 tokens. And so ChatGPT generates some text to tell you a story. That story is say, 1000 tokens long. You then ask it to “Tell me more of the story, and make sure you include a dinosaur.” (15 tokens). And you get another 1000 tokens. And repeat this twice more. At this point, the length of the entire chat history is about 4000 tokens.

ChatGPT then internally asks itself to summarize the entire 4000 token history into 500 tokens. Some of the simpler models can do this quite quickly - though they are imperfect. The thing is at point you’ve got 500 tokens which are a summarization of the 4 acts and of the story and the prompts that were used to generate it - but that’s lossy.

As you continue the conversation more and more, the summarizations become more and more lossy and the chat session will “forget” things.

I was not aware there have been leaks.

The big one was when histories (the prompts that other people used) were accidentally made visible to other users.

https://futurism.com/the-byte/chatgpt-bug-chat-histories-email-phone

https://www.theverge.com/2023/3/21/23649806/chatgpt-chat-histories-bug-exposed-disabled-outage

https://openai.com/blog/march-20-chatgpt-outage

Also consider all the ‘ChatGPT extensions’ that people have written for chrome ( https://chrome.google.com/webstore/search/ChatGPT ) and not infrequent occurrence when someone has an extension with a few tens of thousands of users which gets sold and converted into malware or snooping software ( https://www.theregister.com/2023/08/11/chrome_extension_developer_pressure/ ).

I am making sure that if someone else reads it who doesn’t know or wasn’t sure if I used the words properly that the up front clarification would avoid them asking and then waiting a day or so to get a response that clarifies their question. While you and I may be familiar with it, there are also high schoolers who are reading this who are less familiar with accounting terms and their specific meanings.

https://www.axios.com/2023/08/09/disney-earnings-streaming-profits

Disney also managed to narrow its streaming division losses significantly to last quarter to $512 million from $1.06 billion during the same quarter the year prior.

That progress should give investors hope that Disney is on track to meet its stated goal of making its streaming business profitable by 2024.

https://www.businessofapps.com/data/disney-plus-statistics/

Disney Plus generated $7.4 billion revenue in 2022, an 42% increase year-on-year

They’re getting better and have a lot of other revenue that they can use to offset those losses while they figure out how to do streaming more profitably.

All of the following statements can be true simultaneously:

• There is a lot of revenue (note revenue and not profits) associated with streaming content
• Studios are losing money with streaming
• Actors and writers are being paid much less for residuals of streamed content than would be considered fair

Another promise was that we’d be able to basically have access to every tv show or film, and that evaporated very quickly when all the studios decided they wanted to gatekeep everything so they could charge even more for it, and then now we have them disappearing so many things that have been out there just because they don’t want to pay royalties to anyone involved.

That is one part of it. The studios believed that if they did the distribution they’d be able to have a larger slice of the pie. This isn’t working as well as they thought it would in many cases (Disney is likely the exception if there is one with a very large historical catalog and several profitable franchisees) but didn’t realize the corresponding infrastructure and (customer) support costs that would incur reducing the amount that they make.

Furthermore, the residuals (things that are currently at stake in striking) are based on “having it available” even if no one is watching it. This means that studios (and Netflix and Amazon) are strongly incentivized to remove shows that aren’t getting watched sufficiently for the draw of having them there because they’re paying residuals no matter if its being watched or not.

Having a streaming catalog of 10,000 shows (Netflix has about 13,000 world wide by some accounts with about 5000 available in the US) would mean they’re paying small amounts to everyone even if they’re not getting watched. If the company (e.g. studio) isn’t set up around the infrastructure and support needed for streaming, this can easily mean that those small amounts can become more than the amount that the company is making off of streaming.

Ever notice how Amazon Prime Video rotates in lots of B movies that are available for a few weeks and then disappear again? This is to get people to watch them but minimize paying residuals for having a bunch of movies and TV series that no one watches otherwise.

With the combination of race to the cheapest for pricing (and sharing of accounts), and infrastructure / operational costs it is quite possible to be in the situation where studios are losing money on hosting shows and at the same time paying actors and writers far less than is fair… and the easiest solution that studios have to resolve this is to aggressively pull shows.

https://text.npr.org/2023/03/06/1161382179/hbo-max-disappearing-shows-series-streaming-warner (full site for a 26 minute audio version)

(Linking for the pictures - not the site)

https://yesapower.en.alibaba.com/product/60802489471-807064733/14_4v_new_auto_cell_ni_mh_6500mah_car_battery_pack_for_honda_civic_2008_hybrid_battery_ima_replacement_hybrid_car_battery.html

If you look at the battery pack, you’ll see that its a large number of C cells bound together into rods and then into a “brick” of a battery that then looks like:

https://www.sciencephoto.com/media/623772/view/honda-insight-battery-pack

You’ll see a similar setup for a Prius (note this is updated from the older ‘prismatic’ modules): https://electronhybridsolution.com/product/toyota-prius-v-2012-2017-hybrid-battery-rebuild-kit-with-brand-new-cylindrical-cells/

The best site to read about what is actually mandated and to see how they implement it is https://gdpr.eu … which has a privacy pop up on it that shows up each time.

We use cookies to ensure that we give you the best experience on our website. If you continue to use this site we will assume that you are happy with it.

I’m not sure how deliberate it is.

That was one tweet in a tweet thread from a… guy who is a bit of a character and does stuff with AWS. He pokes a fair bit of fun at Amazon and others in the cloud.

The thread reader rollup of it is https://threadreaderapp.com/thread/1173367909369802752.html which is an amusing read by itself.

My favorite is still:

“Why use AWS instead of IBM Cloud?”
“IBM has a cloud?!”
“I’m as puzzled as you, I’m just reading off the notecard here.”

The best part of that is when you find out that IBM’s on prem cloud is called “IBM Cloud Private”.

https://www.ibm.com/docs/en/cloud-private/3.1.1?topic=started-cloud-private-overview

And then, when the sales teams talk about it, IBM Cloud Private is too long to say again and again… so they start calling it by its abbreviations… not IBMCP but rather ICP… and you start picturing the sales team wearing clown makeup. And when they talk about Machine Learning you share Using AI to Find Where Clowns End and Juggalos Begin with the devops guy sitting next to you and get some muffled chuckles.

Not that those events have ever happened… or would be admitted to.

https://proton.me/mail has not unreasonable pricing for just mail.

Compare that to what I believe is the ZERO breaches Google has had in the same time frame.

From earlier this month: Google Cloud Build bug lets hackers launch supply chain attacks

A critical design flaw in the Google Cloud Build service discovered by cloud security firm Orca Security can let attackers escalate privileges, providing them with almost nearly-full and unauthorized access to Google Artifact Registry code repositories.

Dubbed Bad.Build, this flaw could enable the threat actors to impersonate the service account for the Google Cloud Build managed continuous integration and delivery (CI/CD) service to run API calls against the artifact registry and take control over application images.

As to why don’t you hear about more GCP flaws? I refer you to this uncomfortable truth: https://twitter.com/QuinnyPig/status/1173394437298196480

“What does AWS have that GCP doesn’t?”
“A meaningful customer base?”