hmm, this chart is good but it doesn't really show the best part about the curl man page over time, which is how its organization quality, sectioning, and cross-referencing have improved.
Look, Jeff Atwood, it is difficult to take you seriously when you write authoritatively on a subject you clearly don’t understand.
GDPR doesn’t mandate cookie notices.
Cookie notices are *malicious compliance* by the surveillance-driven adtech industry.
If you’re not tracking people, you do not need a cookie notice, period.
If you’re only using first-party cookies for functional reasons, you do not need a cookie notice, period.
If you’re using third-party cookies to track people – i.e., if you’re sharing their data with others – then *you must have their consent to do so*. Because, otherwise, you are violating their privacy. Even then, the law doesn’t mandate a cookie notice.
How would you conform to EU law without a cookie notice if your aim wasn’t malicious compliance?
You would not track people by default and you would make it so they have to go your site’s settings to turn on third-party tracking if, for some inexplicable reason, they wanted that “feature”.
Boom!
No cookie notice necessary.
What’s that?
But that would destroy your business because your business is founded on the fundamental mechanic of violating people’s privacy?
Good.
Your business doesn’t deserve to exist.
Because the real bullshit here isn’t EU legislation that protects the human right to privacy, it’s the toxic Silicon Valley/Big Tech business model of farming people for data that violates everyone’s privacy and opens the door to technofascism.
Look, EU, it is difficult to take you seriously when you forced all this cookie notification bullshit on us. That feature a) should not exist and b) if it did, should be a BROWSER feature not "every website in the entire world now has to bother everyone forever about this stupid thing" blog.codinghorror.com/breaking…
The Firefox add-in Firesheep caused quite an uproar a few weeks ago, and justifiably so. Here’s how it works:
* Connect to a public, unencrypted WiFi network.
1. Every browser needs to, by default, be set to allow "strictly necessary cookies" only. 2. Every site that wants to serve EU users must honour this setting. 3. Impose massive fines on sites that don't do this or that choose to interpret "strictly necessary only" in "creative" ways.
So that anybody who does not want other cookies has to do exactly nothing to achieve that.
@rozeboosje and invest the money from the fines in privacy respecting FOSS solutions to replace the non-sovereign software the EU runs for its own services. Win-win(-win, for www) 😉 @aral
Look, EU, it is difficult to take you seriously when you forced all this cookie notification bullshit on us. That feature a) should not exist and b) if it did, should be a BROWSER feature not "every website in the entire world now has to bother everyone forever about this stupid thing" blog.codinghorror.com/breaking…
The Firefox add-in Firesheep caused quite an uproar a few weeks ago, and justifiably so. Here’s how it works:
* Connect to a public, unencrypted WiFi network.
That's a myth perpetrated by adtech industry. There is no EU obligation to spam cookie notices. There's an obligation not to track without explicit consent, and everyone illegally uses the cookie nag popups as a basis for claiming consent (which it's not). A legitimate, non malicious site has no need for cookie nags. Ever.
Moreover there *was* a browser feature to set it globally and all the assholes running websites refused to honor it and instead used your setting as an additional fingerprinting bit to track you.
True, but my point remains. This shitty experience we're collectively having here this isn't "the EU forcing cookie notification on people", it's "the malicious compliance of companies that profit from user tracking."
Every company that shows you an cookie popup has made the choice to put a few fractions of pennies of possible future profit ahead of your experience.
Cookies can give businesses insight into their users’ online activity. Unforunately they are subject to both the GDPR and the ePrivacy Directive, making compliance difficult.
@javier Websites that don't use cookies are not involved. Neither are websites that only use cookies that are _required_ for the website to function, e.g. session tokens.
It's only when you'd like to use cookies to track users and deliver personalized ads that you have to deal with this stuff.
It's a choice.
Most websites simply don't choose the privacy-friendly option.
@scy @javier one of the big problems nobody talks about: tech is largely only explained by entities who have no incentive to explain it *well*.
Google, Meta, large ad networks are all like "stupid EU makes us do Cookie banner".
While the actual regulation is actually pretty good. The regulation is basically "don't fuck around with user data. But if you do, you at least need to tell the user".
for those who just want to browse the catalog: until fixed, you can use apt.izzysoft.de/fdroid/?repo=m… (should images not load, it's because they would be loaded from f-droid.org. If you reload the page, another mirror should be picked for those, and images should show up again. A little lottery: you might hit a slow mirror then 🙈 But you can browse the catalog.)
keďže mám stále prístup do EduPage by to mohlo znamenať že ma zatiaľ nevyhodili a zajtra by som mal ísť do školy.. ešte stále ma môžu vyhodiť keď tam prídem
škola sa nám ozvala že prax bude započítaná iba tomu kto absolvoval prax v ich zazmluvnených servisoch... viete koľko z nás bolo v ich servise? nula. ja mám pocit že si to vymysleli/nám to povedali až teraz.
Day 1284. I am in the heaviest battle of Pokrovsk, Donetsk, East Ukraine.
This is my combat gear and my weapon. It does not shoot, but it delivers A LOT of pain to Russian invaders. With this weapon, they have nowhere to hide.
I am Ukrainian Soldier. That’s what I go through 🧵
Greetings all. I was really excited about the new music mix feature in Spotify for premium user’s. Unfortunately my excitement was short lived. This feature is not accessible and as Spotify have demonstrated time and time again, they do not care about accessibility. What really frustrates me with Spotify is when you try and lodge an accessibility bug with Spotify support, they have no idea what you are referring to and they always tell you to go and write on their community forums which also are not that accessible. I have given up on Spotify as a company and will no longer pay for Spotify premium. Accessibility has been a problem for years with Spotify so I will stick with Apple Music. Their support for those interested is now chat based via a website and although it is accessible, it’s a pain to get started with support. Very disappointed with Spotify as a company.
@darrell73 As far as I know, Spotify has no equivalent of Apple Music's Classical app, or of competitors such as Idagio. That rules out Spotify for me.
Is there any ARM64 SoC that can boot to Linux without either requiring a DRAM controller initialization/training blob or punting the whole boot process to a second processor running wholly closed firmware? That second one rules out Raspberry Pi (for personal projects). I'd grudgingly accept ARM Trusted Firmware persistently running in the background if the SoC's branch/fork of that is fully open-source. But I still don't like how much code has to run before you get to the OS kernel.
I'd say, SoC yes, SBC definitely not. And SoC is really just Qualcomm as they're the ones MS partnered with. For a bit of time folks did try to get WOA on R-pi 4 and 5 working, however you lack a lot of drivers for basic things such as Wi-FI, HDMI Audio, so virtualized as sad as it is for now on most 3rd-party boards will be needed. There's a lot based on the RK3588 which are getting popular but they would also have the same faits as the Pi in terms of hardware drivers needed to get it native boot.
I found what seems to be the ideal 32-bit Linux-capable ARM platform in terms of boot transparency and simplicity: the Microchip (formerly Atmel) SAMA5 platform, based on a Cortex-A5 core (I think it's always a single core) and LPDDR2 DRAM (up to 512 MB in the boards I've seen). The official boot loader, at91bootstrap3, is MIT-licensed, including DRAM initialization code. It can chain to u-boot, load a Linux kernel directly, or load an arbitrary bare-metal program.
The SAMA5 platform really does have the least bad boot process I've found in any recent-ish (though only 32-bit) Linux-capable ARM SoC. According to the SAMA5D2 series datasheet (ww1.microchip.com/downloads/ae…), it has a 64 KB boot ROM. In standard (non-secure) mode, that ROM reads the second-stage boot loader from a boot.bin file on an SD/MMC device's FAT partition. That's practically perfect IMO. And the boot.bin in the buildroot config I tried is 14 KB. That loads u-boot which is ~400 KB.
And like I said, that second-stage boot loader (boot.bin) can apparently be configured to skip u-boot and go straight to Linux or some other bare-metal program.
The datasheet does mention a secure boot mode, for which documentation is only available under NDA. Not ideal, but I don't want secure boot anyway.
BTW, I ended up not ordering a Raspberry Pi Pico-based microcontroller board like I was thinking about doing last weekend. A SAMA5 board, if I can get one with an audio DAC and headphone jack built-in or as an add-on, would split the difference between a microcontroller and an application processor, as the SAMA5 platform explicitly advertises bare-metal support with sample code.
Curious, would you mind sharing links with good material on how the Raspberry Pi boot process works? I am working with RPI 4B specifically, so I suppose if the resources are specific, then that's what I'd be looking for. Thanks much. I hope your adventure goes well.
What I dislike about the Pi's boot process is that it runs on a whole other processor (the VideoCore) that's running its own embedded OS. In the Pi 4B, that processor's firmware/OS is split between the built-in EEPROM and a file you have to put on your SD card. In the Pi 5, it's all in EEPROM, so probably out of sight, out of mind for most people, but I know it's still there.
Thanks for the link and explanation of the other coprocessor with its own OS on it. I knew the RPI had an EEPROM, but didn't look into it hard enough to find out that it was so advanced. I guess my thinking was it was something to address and control some of the specialized stuff on the board. Good to know stuff for sure. Thanks.
I wish the creators of the Raspberry Pi had been working at Atmel rather than Broadcom. A wildly popular hobbyist SBC with built-in audio DAC and headphone jack based on the SAMA5 SoC family would have been perfect!
Frustratingly, there _was_ a SAMA5 evaluation kit with built-in audio DAC and headphone jack, but it's now discontinued. Presumably not enough demand for that particular combination.
The bare-metal SDK for the SAMA5 family and some other SoC's (github.com/atmelcorp/atmel-sof…) even include audio examples, with configurations for the discontinued eval kit.
I'm talking about this eval kit: microchip.com/en-us/developmen… It basically had everything except WiFi, including audio codec, USB, Ethernet, and even HDMI. And apparently the SAMA5D4 SoC does have video decode (probably a proprietary driver for that). DigiKey knows about that eval kit, but lists it as obsolete, and provides a coipy of this document explaining why: mm.digikey.com/Volume0/opasdat… But the newer, lower-cost SAMA5D4-Xplained board doesn't have audio!
@thejpster Looks promising. The non-OS BSP you linked even includes an MP3 player example, apparently running on bare metal. On the Linux side, they have a buildroot BSP. Looking at that now.
@thejpster Hmm, looking at the number and size of the .bin files in the output from the MA35 buildroot build, it's aesthetically not any more satisfying than any of the other non-Raspberry Pi ARM64 boards I've seen. The SAMA5 platform I mentioned elsewhere on the thread is much more appealing, especially if I can cut out u-boot entirely, as it seems the Atmel boot loader allows. I just have to figure out how to add a DAC and headphone jack to one of the off-the-shelf eval boards.
Note that #Quicksy users won’t be able to opt out of TLS 1.3.
If you are sending files to Quicksy users (via HTTP Upload), make sure your HTTP server supports 1.3; otherwise Quicksy users won’t be able to receive the file.
The #Conversations_im update that requires TLS 1.3 is currently rolling out.
Apparently, judging by the bug reports coming in, a good number of servers do not support TLS 1.3 yet.
Please upgrade your servers and/or check their configurations. 🙏
Users can opt out via a setting in the Security section of the app, but this setting will likely go away in 1-2 years. The only correct move here is to fix your servers. Also, check the HTTP servers used by HTTP Upload.
glad I saw this toot. It made me able to use my self-hosted @prosodyim again. Before, I had just turned off the server connection to fix it at a later time, especially since I didn't know what caused it. I just checked the validity of the cert and didn't find any issues.
It would be great if the error message in the app would indicate the cause. "TLS negotiation failed (TLS v1.3 not supported by server)"
@SimJoSt @prosodyim I agree that this would be a better error message. I originally planned to do that but then somehow wasn't able to differentiate between 'general certificate problem' from 'protocol too old'. I'll take another look to see if this can be improved.
I may have underestimated the number of servers effected by this.
unfortunate, that you cannot discern the specific error easily, but good that you are already thinking about it. I'm grateful :) It surprised me that my @prosodyim setup didn't support TLS v1.3 already. Checking the docs it is not mentioned at all and I will have to figure out how to enable it, if it is supported at all.
well, unfortunatelly, there is not only "good number of servers do not support TLS 1.3 yet", but also there are some countries that has gorernment-ruled DPI hardware deployed country-wide and that blocks TLS 1.3. If you're lucky, it can be "time to time" or "some specific things", but if you're not - it can be total breakage of any TLS1.3 conenctions.
So, users from such coutires won't be able to have secure communications at all: from the one side there is government pushing, from the other — such a decisions 🤷
Ayer estuve con mi hija en el acuario de Veracruz, y por contestar correctamente una pregunta sobre manatíes, me dejaron darle de comer a un manatí.
Entre tres manatíes, se empacaron una hielera completa de verduritas y manzanas en unos minutos. Para darles, había manzana, pepino, calabacita y jitomate. Según yo, lo que más les gustó es el jitomate.
Pueden mover los pliegues que tienen debajo de las fosas nasales; ahí se les da de comer y ellos llevan la verdura a su boca. Son adorables.
por dios, nunca vi um manati então imaginava que eram menores... Mas en portugues lo llamamos también "peixe-boi", entonces deve ser mismo grande como un buey!
Started using an electric fan to blow the soldering fumes away when I'm working after seeing @feld 's post not too long ago and there's no way I'm going back.
Last day enjoying this calendar as wallpaper on my second monitor in the office. It's one of my fave shots. Good memory of a nice evening spent with mates.
If you compare _pixman_setup_combiner_functions_32 and _pixman_setup_combiner_functions_float, you'll see that there is no integer version of PIXMAN_OP_SATURATE and if falls back to floats. We found this...
my next few months working on Plasma accessibility will involve:
- Touch gestures for desktop zoom - Individual zoom level and tracking for each display on multi-display systems (if you enable it, some users may prefer it acting like one giant screen) - Shape-aware upscaling for extreme zoom levels. Makes text appear far less blurry. I'll turn it off by default.
Paperback 0.2 is out! What's new: • Added markdown document support! • Added PDF document support, including the ability to navigate between pages! • Added keystrokes for navigating by headings in HTML content, including epub books and markdown documents. These keystrokes were designed to work similar to a screen reader. • Fixed loading epubs with URL-encoded filenames in their manifests. • Fixed loading epub 3 books with XHTML embedded inside of them. • A message is now spoken if the document doesn’t support a table of contents or sections, as opposed to the menu items being disabled. • Added a recent documents menu! It currently stores your last 10 opened documents, and pressing enter on one will open it for reading. • Completely rewrote the Find dialog, making it much simpler to use, while also adding a history of your last 25 searches and regular expression support! • Previously opened documents are now remembered across application restarts. This is configurable through the new options item in the tools menu. • Added shift+f1 to open the readme directly in Paperback itself. Download: github.com/trypsynth/paperback… Enjoy!
Edit: sorry about that, Enafore decided that I meant to edit my post when I pressed C, not create a new post.
Bro thank you so much for this. I'm going back to school in a few days and definitely plan on taking full advantage of this utility to read my stuff from Bookshare. I'm assuming yes, but is this compatible with Mac as well as Windows? I haven't downloaded yet. Thanks again!
I upgraded to a Pixel 10 Pro because the promos made it super cheap for me—and honestly, Pixelsnap (Qi2/MagSafe) without a case. I disable all the AI features I can on my Google Account, so none of that new stuff interests me, but there *are* a handful of new things I’ve seen literally nobody mention in “reviews” or sales pages. I’ll compile them here in this thread for anyone interested.
Planning to buy a P10 Pro, too, so looking forward to your findings. Not really concerned about integrated AI since I will instantly reflash it with GrapheneOS.
Fan ChannelOriginal Recorded on11/23/2022Official DougDoug: @DougDoug Full VOD: https://www.youtube.com/watch?v=PHuLF7p42yo&t=6441s&ab_channel=DougDougDougDo...
I keep meeting people who tell me they read and appreciate my writing - both the weekly emails and my blog. That is what inspires me and gives me motivation and energy. Knowing that I'm not always just shouting into the void makes a difference!
@Zykino @raspbeguy the graph shows number of names in the THANKS file. In 2005 I went back and populated the file with all the names that had participated but that hadn't been added to the file properly. Hence the bump.
* FlorisBoard: the stable version of the popular customizable keyboard 🛡️ * VMPK for Android: the Virtual MIDI Piano Keyboard is a MIDI events generator simulating a MIDI controller 🛡️
RB status: 699 apps (53.6%) (4 failed builds still pending)
This is a repository of apps to be used with your F-Droid client. Applications in this repository are official binaries built by the original application developers, taken from their resp. repositories (mostly Github, GitLab, Codeberg).
Před pár roky jsem u nás při přerovnávání dřeva objevil kudlanku. Měl jsem z toho vánoce, protože jsem do té doby naživo asi žádnou neviděl, nebo možná jo a zapomněl :)
Dnes jsem ničil s kosou trávu, kterou se nechce jíst ovcím a i přesto, že nebylo moc času na koukání jsem zahlídl kudlanek hned šest, nádhera.
to je vliv globálního oteplování. Mají rády sucho a teplo. Dřív se vyskytovaly jen na jižní Moravě. Dnes už jsou na většině území ČR. Nepamatuji, že bych je jako děcko někdy potkal. Teď ji najdeme na zahradě několikrát do roka. Šest naráz je ale pořád hodně velký nadstandard. 🙂
Niri is a scrollable-tiling Wayland compositor. Windows are arranged in columns on an infinite strip going to the right. Opening a new window never causes existing windows to resize.
A month ago, o...
So we can use the new desktop? Considering turning an old computer in to a linux one. Was thinking Debeon, but don't want o use gnome unless that's better.
@vol4life8657 I haven't tried it yet, but yes, in theory, the niri compositor should be usable with Orca, along with accessible applications, of course. And I think I would be happier with niri than with GNOME.
So, I'm not trying to be stupid here, because I haven't really used linux since ffadora 16, but how would I add this to a distro, since it just came out?
Ok, so if the correct definition of anarchy is that everyone is equal and everyone can be heard -- how is this functionally different than a direct democracy?
And more realistically -- how can everyone's voices be heard in a meaningful way that scales with the population?
it's not really about the voting processes. direct democracy doesn't necessarily seek to establish a classless/stateless society. direct ballot initiatives are a form of direct democracy, and several states use them, but you would never call the states anarchist because they still have governments.
@me it doesn't make any sense because people still want rules and regulations and laws about what is permissible. So there has to be a government of some form to enforce them. The people can't enforce on their own
actually a really clever strategy for blocking bots: if someone is only requesting html content then block them after a certain number of requests, but reset it after they request css or js files blog.frehi.be/2025/08/26/block…
Does that really work? Don't most modern bots actually use a headless chromium/firefox so it can render the page and get more useful data out? Especially for being able to digest websites that are just SPAs and require executing the javascript to see the real content?
daniel:// stenberg://
in reply to daniel:// stenberg:// • • •Jonathan Yu
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to daniel:// stenberg:// • • •CatSalad🐈🥗 (D.Burch)
in reply to daniel:// stenberg:// • • •Demiguise 🇮🇱
in reply to daniel:// stenberg:// • • •Ben Zanin
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to Ben Zanin • • •wertzui
in reply to daniel:// stenberg:// • • •synlogic
in reply to daniel:// stenberg:// • • •daniel:// stenberg://
in reply to synlogic • • •daniel:// stenberg://
in reply to daniel:// stenberg:// • • •Bojidar Marinov
in reply to daniel:// stenberg:// • • •Yaksh Bariya
in reply to daniel:// stenberg:// • • •Perry
in reply to daniel:// stenberg:// • • •