51
11
submitted 3 weeks ago* (last edited 3 weeks ago) by brucethemoose@lemmy.world to c/technology@piefed.social

"I love solutions that teeter on appearing almost naive in their simplicity," Ive said. "I also love incredibly intelligent, sophisticated products that you want to touch — and you feel no intimidation, and you want to use almost carelessly, that you use them almost without thought, that they're just tools."

Altman, elaborating on Ive's simplicity mindset, said that AI "can do so much for you that so much can fall away. And the degree to which Jony has chipped away at every little thing that this doesn't need to do or doesn't need to be in there is remarkable."

"We just started talking about: What does it mean that this thing is going to be able to know everything you've ever thought about, read, said? ... And finally, we have the first prototypes."

Altman recalled that Ive once said they'd know they had the design right when the user wants "to lick it or take a bite out of it, or something like that."

"There was an earlier prototype that we were quite excited about, but I did not have any feeling of: 'I want to pick up that thing and take a bite out of it.' And then finally we got there all of a sudden."

I dunno about that

52
6

Regions include (in order of video):

  • US
  • Australia
  • Germany
  • UK
  • Canada
  • India
  • Philippines
  • Brazil
  • Poland
  • Netherlands
53
4
54
16
submitted 4 weeks ago* (last edited 4 weeks ago) by misk@piefed.social to c/technology@piefed.social
55
72
56
8
57
11
submitted 1 month ago by misk@sopuli.xyz to c/technology@piefed.social
58
31
59
12

WhatsApp is close to rolling out third-party chat support across the European Union, as part its compliance with the bloc's Digital Markets Act (DMA)...

60
7
61
10
62
51
63
22
64
6
65
4
66
4
67
41
68
23
69
3
submitted 1 month ago* (last edited 1 month ago) by misk@piefed.social to c/technology@piefed.social
70
23

Social media giant Meta is about to start reading users’ conversations, including direct messages (DMs) and chats with its AI, with no option to opt out other than not to use their platforms – which include Facebook, Instagram and WhatsApp, which provides no option to turn its AI off. This data access will start in most of the world from 16 December though in the EU and UK, which have stricter data protection laws, it will start later – apparently from 4 March next year.

PC World Magazine reported:

The initiative will begin starting December 16th, 2025, initially outside the EU and UK where stricter data protection laws will force a later introduction. The data will be used to further personalize advertising and content, and it won’t be possible to opt out.

Meta spying

Industry media analyses have tended so far to focus on the issue of AI chats but Facebook, in a privacy update titled “Your activity and information that you provide” includes DMs in the data it can access, gather and use:

Meta has been known to provide ‘near real-time’ data on its users to the authorities since at least 2021, though previously this has not usually – at least officially – included the content of DMs. A report in Israel’s 972 Magazine and analysis by Tech for Palestine last year revealed that Israel’s ‘Lavender’ AI targeting system was using WhatsApp data to target Palestinians for murder, often based on as little as a ‘target’ being in a WhatsApp group with someone else who had been targeted and killed. One of 972’s sources told the magazine that after Lavender identified a victim, Israel:

bombed them in homes without hesitation, as a first option. It’s much easier to bomb a family’s home. The system is built to look for them in these situations.

Far beyond overreach

Journalist Jamal Khashoggi was also murdered in a Saudi embassy after his family’s WhatsApp messages were hacked by Israel’s ‘Pegasus’ spyware, which has also been used to spy on human rights activists, journalists, political opponents and Western government ministers.

Meta denies that its products contain backdoors and that it is (currently) reading messages. However, the company is now being sued in the US by its former head of security, who alleges that it allows thousands of its engineers to access sensitive user data and has not adequately tackled issues allowing the hacking of over 100,000 accounts a day. Complainant Attaullah Baig claims that the company ignored his warnings and sacked him for raising concerns. Meta denies the allegations.

The Meta issue comes on top of wider concerns over online security after digital rights group SMEX revealed that all Samsung mid-range handsets in large parts of the world come pre-installed with ‘unremovable’ Israeli spyware.

71
27
72
50
73
39
submitted 1 month ago* (last edited 1 month ago) by rimu@piefed.social to c/technology@piefed.social

In June, President Gabriel Boric announced that Microsoft’s hyperscale data center cluster would generate over 81,000 jobs.

In reality:

those projects would add only 909 permanent positions during the operational phase, which lasts about 30 years.

74
12
75
8

A lot!

One of the most common reactions or questions I get about curl when I show up at conferences somewhere and do presentations:

is curl still being actively developed?

How many more protocols can there be? This of course being asked by people without very close proximity or insight into the curl project and probably neither into the internet protocol world – which frankly probably is most of the civilized world. Still, these questions keep surprising me. Can projects actually ever get done?

(And do people really believe that adding protocols is the only thing that is left to do?)

Everything changes

There are new car models being made every year in spite of the roads being mostly the same for the last decades and there are new browser versions shipped every few weeks even though the web to most casual observers look roughly the same now as it did a few years ago. Etc etc. Even things such as shoes or bicycles are developed and shipped in new versions every year.

In spite of how it may appear to casual distant observers, very few things remain the same over time in this world. This certainly is also true for internet, the web and how to do data transfers over them. Just five years ago we did internet transfers differently than how we (want to) do them today. New tweaks and proposals are brought up at least on a monthly basis.

Not evolving implies stagnation and eventually… death.

As standards, browsers and users update their expectations, curl does as well. curl needs to adapt and keep up to stay relevant. We want to keep improving it so that it can match and go beyond what people want from it. We want to help drive and push internet transfer technologies to help users to do better, more efficient and more secure operations. We like carrying the world’s infrastructure on our shoulders.

It might evolve for decades to come

One of the things that actually have occurred to me, after having worked on this project for some decades by now – and this is something I did not at all consider in the past, is that there is a chance that the project will remain alive and in use the next few decades as well. Because of exactly this nothing-ever-stops characteristic of the world around us, but also of course because of the existing amount of users and usage.

Current development should be done with care, a sense of responsibility and with the anticipation that we will carry everything we merge today with us for several more decades – at least. At the latest curl up meeting, I had session I called 100 year curl where I brought up thoughts for us as a project that we might need to work on and keep in mind if indeed we believe the curl project will and should be able to celebrate its 100th birthday in a future. It is a slightly overwhelming (terrifying even?) thought but in my opinion not entirely unrealistic. And when you think about it, we have already traveled almost 30% of the way towards that goalpost.

But it looks the same

— I used curl the first time decades ago and it still looks the same.

This is a common follow-up statement. What have we actually done during all this time that the users can’t spot?

A related question that to me also is a little amusing is then:

— You say you worked on curl full time since 2019, but what do you actually do all days?

We work hard at maintaining backwards compatibility and not breaking existing use cases. If you cannot spot any changes and your command lines just keep working, it confirms that we do things right. curl is meant to do its job and stay out of the way. To mostly be boring. A dull stack is a good stack.

We have refactored and rearranged the internal architecture of curl and libcurl several times in the past and we keep doing it at regular intervals as we improve and adapt to new concepts, new ideas and the ever-evolving world. But we never let that impact the API, the ABI or by breaking any previously working curl tool command lines.

I personally think that this is curl’s secret super power. The one thing we truly have accomplished and managed to stick to: stability. In several aspects of the word.

curl offers stability in an unstable world.

Now more than ever

Counting commit frequency or any other metric of project activity, the curl project is actually doing more development now and at a higher pace than ever before during its entire lifetime.

We do this to offer you and everyone else the best, the most reliable, the fastest, the most feature rich, the best documented and the most secure internet transfer library on the planet.

view more: ‹ prev next ›

Technology

1334 readers
191 users here now

Tech related news and discussion. Link to anything, it doesn't need to be a news article.

Let's keep the politics and business side of things to a minimum.

Rules

No memes

founded 6 months ago
MODERATORS