@teilweise@layer8.space

Der Teilweise @teilweise@layer8.space · Jul 20, 2026

Replying to @HackyScientress@chaos.social

@HackyScientress@chaos.social Laut https://www.elektro.net/media/file/Rjgn4k4OM3/5edd304c/DE_2020_6_67-69_MvM1_LOW.pdf könnte es Wellenlänge (B=850nm) und Dispersion sein. Ansonsten: https://www.corning.com/emea/de/contact-us.html#corp-comm (Kommunikationsnetze)

0

Der Teilweise @teilweise@layer8.space · Jun 25, 2026

Replying to @spaetzlesmasher@sueden.social

@spaetzlesmasher@sueden.social @tonfolge@zug.network Ich hätte jetzt angenommen, dass am Dienstag Abend mehrere Stunden später daheim anzukommen für die meisten schlimmer ist, als am Samstag Abend. Weil mehr der Betroffenen am Mittwoch trotzdem zur gleichen Zeit aufstehen müssen. Von der Anzahl der Betroffenen her kann es aber natürlich sein.

0

Der Teilweise @teilweise@layer8.space · Jun 25, 2026

Replying to @tonfolge@zug.network

@tonfolge@zug.network Die Frage, die ich mir schnell gestellt habe: Ist ein Dienstag Abend wirklich der Zeitraum, in dem man einen *planmäßigen* Tausch einer KRITIS Komponente macht? Wenn Dich die Redundanz rettet, hattest Du keine Redundanz mehr. Entweder war der Tausch nicht so planmäßig, wie das dargestellt wird (sondern da war etwas kurz vor dem Abrauchen) oder da wurde bei der Planung ein Fehler gemacht. Da hätten mögliche Worst-Case-Szenarien einbezogen werden und Werktage damit rausfallen müssen.

0

1

0

Der Teilweise @teilweise@layer8.space · May 29, 2026

Replying to @nSonic@troet.cafe

@nSonic@troet.cafe LibreWolf (Firefox Clone): DevTools → Network → "Disable Cache" sollte helfen. Gibt es, glaube ich, bei Safari auch.

0

1

0

Der Teilweise @teilweise@layer8.space · May 27, 2026

Replying to @heliographe_studio@mastodon.social

@heliographe_studio@mastodon.social “Ugly” isn’t the problem there. It could use screaming purple, green, yellow, magenta, you-name-it all over it. Like Windows 3.1’s “Hot Dog Stand” theme. THAT (capital letters for emphasis) was ugly. (See https://www.windowscentral.com/microsoft/windows/windows-3-1-hotdog-stand-garish-choice-with-catchy-name) BUT (see above): It did not break usability (at all). The screen you showed is not ugly. It’s way worse: It’s UNUSABLE (read this with your most powerful voice in your mind).

1

0

Der Teilweise @teilweise@layer8.space · May 25, 2026

Replying to @danyow@norden.social

@danyow@norden.social @cocoafrog@hachyderm.io Zum Befestigen der Kabel an der Außenwand um das Auto zu laden (also nur für Deinen Fall, nicht Joachims) habe ich heute gelesen, dass §554 BGB u.a. Ladetechnik explizit privilegiert (§554 wurde 2020 dahingehend geändert). Da könntest Du ggf. nochmal nachfragen, ob bzw. wie Dein Vermieter das erlauben würde … https://www.gesetze-im-internet.de/bgb/__554.html

0

1

0

Der Teilweise @teilweise@layer8.space · May 22, 2026

Replying to @colinmford@typo.social

@colinmford@typo.social This is doomed to suffer from Tourette syndrome …

0

Der Teilweise @teilweise@layer8.space · May 12, 2026

Replying to @noheger@wien.rocks

@noheger@wien.rocks They surely did test it with all default accent colors … did they?

1

0

Der Teilweise @teilweise@layer8.space · May 07, 2026

Schade, dass das hinter einer Pay-Wall ist.

https://www.sueddeutsche.de/kultur/afd-rechtsradikale-meinungsfreiheit-li.3472726

Aber Aufmacher auf der Startseite ist für sich schon eine Ansage.

Genauso Sätze wie „Natürlich hätte Deutschland die AfD schon vor Jahren verbieten müssen – und müsste es spätestens heute tun.“ oder knapper „Ein Verbot dieser Partei ist zwingend”.

Was mich (wie so oft) stört: Wir können die Partei nicht *verbieten*, aber wir könnten prüfen lassen, ob sie verboten *ist*. Optional ist nur die Prüfung, nicht die Folgen.

0

Der Teilweise @teilweise@layer8.space · May 07, 2026

Test von USB-Netzteilen.

Negativer Punkt bei einem: „In vielen Farben erhältlich“.

Dabei hatte ich gar nicht gefragt, ob die Redakteure dumm sind, aber dass der Test nichts taugt, das erkennt man nicht nur an dem Punkt.

0

Der Teilweise @teilweise@layer8.space · May 06, 2026

#Apple is all things half finished nowadays.

The Gregorian Calendar skipped Oct 5. – Oct 14. 1582.

Apple did implement this in the month view of the iOS calendar app.

They failed to implement it in the year view.

If you open the day view of Oct 4. 1582, try to add a new event, then tap the start date selector, the calendar app crashes.

layer8.space

Layer8 in Space

0

2

1

0

Der Teilweise @teilweise@layer8.space · May 03, 2026

Replying to @oliverandrich@social.tchncs.de

@oliverandrich @chrismarquardt *) gemäß EU-Gesetz. Wenn man einen Strommesser zwischenschaltet, dann sieht man, dass es sehr wohl Strom zieht. Wenig genug, dass es nach EU-Vorgabe als „Aus-Zustand” gilt, aber Strom fließt. (Zumindest bei dem Gerät, bei dem ich den Strommesser zwischengeschaltet habe, war es so.)

1

0

Der Teilweise @teilweise@layer8.space · May 02, 2026

“IF THIS IS UNEXPECTED, PLEASE FILE A BUG REPORT.”

If that’s not what a dev screaming for help against management looks like, I don’t know what is.

Yes, that was unexpected.

What was not unexpected is that it deleted the half finished SDK download and that it does not automatically resume it.

No, I will not file a bug report.

I just hope @mjtsai@mastodon.social et al. pick it up and blog about it.

Those blogs are easily worth 10 bug reports.
100× or more if by a major printed publication.

#Apple #Xcode

2

1

0

Der Teilweise @teilweise@layer8.space · Apr 29, 2026

Replying to @tempelorg@iosdev.space

@tempelorg@iosdev.space For me as a user of apps on TestFlight, the switch from TestFlight to AppStore is not the big problem. Manual step but it works. The bigger problem is after that: I was on TestFlight, the beta expired, I switched to AppStore version. Now the app is no longer visible in TestFlight (!). Even if a new build becomes available, I no longer get notifications because the installed version is the AppStore one. If possible: Upload new (unchanged) builds to TF regularly.

1

0

Der Teilweise @teilweise@layer8.space · Apr 25, 2026

Replying to @JennyFluff@chitter.xyz

@JennyFluff Wild idea: Did you ask them directly? E.g. by mail leserservice@stiftung-warentest.de I can imagine one of them would have a good laugh, walk into the archive and scan it for you.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 24, 2026

Replying to @below@mastodon.social

@below@mastodon.social Are you sure about four?

1

0

Der Teilweise @teilweise@layer8.space · Apr 24, 2026

Replying to @leberwurstsaft@mastodon.social

@leberwurstsaft @chbeer … which in turn explains why the main video stutters …

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 23, 2026

iOS 18.7.8 was released.

It’s also available for iPhones/iPads that are supported by iOS 26.
The update is still well hidden but it is available. (For a video see @cosmic040@chaos.social)

Could it be that #Apple accepted that those that remained on iOS 18 are not going to update to iOS 26?

chaos.social

cosmic040: "Wenn Apple hide & seek mit uns spielt - bei Secur…" - chaos.social

1

0

Der Teilweise @teilweise@layer8.space · Apr 22, 2026

Replying to @nico42@mastodon.social

@nico42 Style felt strange, contained error. Great progress. I do not consider that blog article “very good on the first shot”. Even the published version contained incorrect information. My bias was and is that AI generated texts often contain errors that are not caught by proofreading or that their users skip proofreading altogether. That bias got confirmation. No yelling at clouds, if it works for you, go for it. Just do not expect me to praise the result.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 22, 2026

Replying to @nico42@mastodon.social

@nico42 I suspected it was written by AI. Confirms my bias.

0

3

0

Der Teilweise @teilweise@layer8.space · Apr 22, 2026

Replying to @nico42@mastodon.social

@nico42 Minor correction to your blog post: “Accumulated brush time”. The example “00:00:02:00” does not mean 2s. The meaning of the “02:00” is not documented right now. (see Künzi: “Only the first two bytes at address 0x24 are used for timekeeping. Once the counter reaches FF:FF:02:00 it stops going up (18 hours of continuous brushing).”) (Using “address” for the “page” or “block”: Not wrong in the NFC context but confusing if you are not used to NFC …)

1

6

0

Der Teilweise @teilweise@layer8.space · Apr 22, 2026

Replying to @nico42@mastodon.social

@nico42 Bonus points to Philips for the “Page not found” on https://www.philips.com/cfcbrushheadtap

0

Der Teilweise @teilweise@layer8.space · Apr 21, 2026

Between 1954 and 1990 (36 years), the number of active nuclear power plant reactors world wide changed from 1 to 416.
Between 1990 and 2026 (36 years), the number of active nuclear power plant reactors world wide changed from 416 to 415.

That’s what “every country but ours uses nuclear power to become carbon neutral” looks like. (If you think the total net electrical capacity changed a lot: Nope. +2% from 2005 to 2024)

https://de.wikipedia.org/wiki/Kernkraftwerk#Anzahl_der_Kernkraftwerke
https://pris.iaea.org/PRIS/WorldStatistics/OperationalReactorsByCountry.aspx
https://pris.iaea.org/PRIS/WorldStatistics/WorldTrendNuclearPowerCapacity.aspx

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom Do you know the difference between 'const char *andchar * constin C++? I thinkletis a more likechar * constthan likeconst char *`.

The "\("string literal")" case probably gets special treatment when optimizing. I vaguely remember it from when I implemented https://github.com/dhoepfl/DHLocalizedString/ because it was concatenated even when I wanted to get every interpolation but could not.

Not to forget: Interpolation is used for more than just plain old strings.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom

import Foundation

print("Cannot \("action") because \("reason")")

gets optimized when compiled with swiftc -O as you assumed.

BUT

import Foundation

let action = "action"
let reason = "reason"
print("Cannot \(action) because \(reason)")

is not optimized to "Cannot action because reason".

For localized strings, it cannot be optimized, the key format is documented, it will be “Cannot %@ because %@”.

0

3

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom Because of the way string substitutions work in the Swift/Obj-C/SwiftUI/UIKit/AppKit world.

If you write "Cannot \(action) because \(reason)", the compiler generates code similar to StringInterpolation(…).appendLiteral("Cannot ").appendInterpolation(action).appendLiteral(" because ").appendInterpolation(reason). This would show up as "Cannot " and " because " in the binary.

For localizable strings, something like "Cannot %@ because %@" would show up.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom I am not attributing to malice (see: Hanlon’s razor). I assume that the people at Apple us the normal tools. And if they do so, it is most likely that the error message shows up in the binary as is because someone wrote it exactly like that in the source code. Is it possible that the string shows up like that due to compile time substitution? Sure. Is it likely? IMHO: No. That being said, I do not see this thread leading to anywhere, I think it is time to end it.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom OK, maybe it is generated code. Or someone wrote a pre-compiler to generate that string. Or someone trained an AI to write that very specific piece of code. Surely possible. However, in my opinion, the most simple explanation, given the fact that the very string appears in the executable, is, that someone wrote that very error message exactly as shown. Unless I see evidence for anything different, I claim Occam's razor.

0

9

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom It reveals that someone did write “Cannot delete because already installed.” (No more, no less.) How substitution strings are constructed is well documented, so I am very confident that “Cannot delete because already installed.” does not contain any substitution marker. I used the strings (that does things on executables on macOS) command to simplify the command/output for the post but I did verify it without strings.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom That’s the search pattern, I should have given the result of the grep, too: It is the exact error message.

I did a different grep to find the location of the string, the grep I gave in the post was just reduced to produce the single one line that matches the error message 1:1.

Initially I used grep -a -r Cannot' dele' /Applications/Xcode.app/ | less but that contains lots of (binary) output and makes it hard to see the interesting part.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom Maybe I misunderstand something here. The point is not that it only had one match. The point is that it is an exact match. That makes it a separate error string. Other errors when deleting components use other error messages. Of course I cannot tell if they use generated error messages (e.g. “Failed to due to: ”) or also separate messages but for the case I reported, I am convinced that someone came up with a unique message.

0

13

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @leeloo@c.im

@leeloo @Sarnau @mayo @dasdom I don’t know that. All I know is that said someone did add a separate error string for that very case.

I assume that it was not expected that it is possible to even try to delete the installed component but whoever wrote that error message had to implement a protocol where delete() has to be implemented. Instead of going the extra mile to make sure the delete button is disabled, that someone called it a day.

0

15

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @electret@merveilles.town

@electret @Sarnau @mayo The result matched my expectations, unfortunately. (Had it been substitutions, I would have expected double white spaces.)

1

0

Der Teilweise @teilweise@layer8.space · Apr 20, 2026

Replying to @Sarnau@mastodon.social

@Sarnau @electret @mayo @leeloo @dasdom

All of you got me curious:

strings /Applications/Xcode.app/Contents/Frameworks/IDEKit.framework/Versions/A/IDEKit | grep 'Cannot delete'

Result: One string, no substitution markers.

3

22

0

Der Teilweise @teilweise@layer8.space · Apr 19, 2026

Replying to @dasdom@chaos.social

@dasdom That’s why I never have generic error handlers. I try to handle every case separately. It’s a whole list of problems: • I would not have tried to uninstall it if there wasn’t a download notification every now and then at the start of Xcode. • If you cannot uninstall it, there should not be a button to delete it. • If it is an optional component, uninstall should work. • The alert should not be “inactive”. • The details are useless. • The transition to show details is … great. • […]

1

0

Der Teilweise @teilweise@layer8.space · Apr 19, 2026

Someone at #Apple sat at their keyboard and wrote this error message for #Xcode.

Maybe replacing devs with an AI tool is a step forward in some cases.

I’m not even starting to talk about the fact that the alert looks like it is disabled/in the background but is not.

layer8.space

Layer8 in Space

71

30

19

0

Der Teilweise @teilweise@layer8.space · Apr 17, 2026

Replying to @arroz@mastodon.social

@arroz Yes, please! It’s very annoying that there is no way to tell if pressing Return, Shift-Return, or Option-Return will send the message or insert a newline. If you use Shift- or Option-Return, PLEASE, render the send button disabled as long as the modifier is pressed. (And this settles that you should not use Shift-Return because you do not want to blink the button for each upper-case letter.) Actually, plain Return should insert a newline and send should be triggered by Command-Return.

2

0

Der Teilweise @teilweise@layer8.space · Apr 09, 2026

Replying to @tnt@chaos.social

@tnt Just a guess: Did you check bit order? Big/Little Endian? Maybe you want to read address 0x4800 but the IC responds with the content of 0x0012? The IC seems to support several conventions, see 0x480A, the default (0x00) being … undocumented.

0

1

0

Der Teilweise @teilweise@layer8.space · Apr 09, 2026

Replying to @tnt@chaos.social

@tnt I cannot help you but just to clarify: You are talking about the OV5647 sensor this https://cdn.sparkfun.com/datasheets/Dev/RaspberryPi/ov5647_full.pdf datasheet is about? According to https://docs.arducam.com/Raspberry-Pi-Camera/Native-camera/5MP-OV5647/#image-sensor there might be two versions: V1 (probably what the data sheet is about) and V2 which uses some kind of encryption (your problem?). Did you try to contact OVT? Sometimes asking companies for information yields surprising results.

0

3

0

Der Teilweise @teilweise@layer8.space · Apr 07, 2026

Replying to @jenni@appdot.net

@jenni@appdot.net @below@mastodon.social I didn’t know that. For those that want to read more about the problem, here’s a link I found useful: https://mathiasbynens.be/notes/mysql-utf8mb4

1

0

Der Teilweise @teilweise@layer8.space · Apr 02, 2026

Replying to @jxrxme@metalhead.club

@jxrxme „Viewers will see a blue screen if there is a loss of signal, or if the bandwidth is needed for mission activities.“ Video-Streaming ist nicht deren primäres Missionsziel.

1

0

Der Teilweise @teilweise@layer8.space · Mar 23, 2026

Bei courtlistener, z.B. https://www.courtlistener.com/docket/67239065/cornet-v-twitter-inc/ Eintrag vom 19. März, steht oft „SO ORDERED AND ~Util“.

Ich wollte wissen, ob meine Interpretation, dass „~Util“ bedeutet, dass die Anordnung sofort in Kraft tritt („direkt angewandt“), korrekt ist. Google siehe Bild.

Können wir das mit den KIs jetzt bitte wieder lassen? Das hat in den 70ern nicht funktioniert, in den 80ern nicht, in den 90ern nicht, und auch die letzten 25 Jahre ist es nicht so zuverlässig geworden, dass man es haben möchte.

1

0

1

0

Der Teilweise @teilweise@layer8.space · Mar 22, 2026

Replying to @nikitonsky@mastodon.online

@nikitonsky@mastodon.online @gklka@mastodon.social @nickheer@c.im I am 100% sure it did work. I do not know what version of macOS/iOS broke it but I know that it did work as one would expect it to in the past.

0

Der Teilweise @teilweise@layer8.space · Feb 13, 2026

Replying to @lapcatsoftware@mastodon.social

@lapcatsoftware@mastodon.social Sorry, my fault, I thought you noticed the post I replied to when I added you. I assume there are technical or UX reasons for “loop videos” depending on the generic native controls setting?

0

1

0

Der Teilweise @teilweise@layer8.space · Feb 13, 2026

Replying to @lapcatsoftware@mastodon.social

@lapcatsoftware@mastodon.social Normal YouTube videos. With native controls enabled, I loose all YouTube specific controls (chapters, most viewed chart, settings, closed captions, …). That’s why I do not want native controls all the time. Where I want them is like in the blog post in the post I added you (https://noheger.at/blog/2026/02/12/resizing-windows-on-macos-tahoe-the-saga-continues/) Currently, to get the native controls in the blog post, I need to turn them on for normal videos to be able to turn them on for looping videos. I only want them for looping videos.

0

1

0

Der Teilweise @teilweise@layer8.space · Feb 13, 2026

Replying to @lapcatsoftware@mastodon.social

@lapcatsoftware@mastodon.social I could probably use that … it’s just that for many sites (e.g. YouTube), I do not want native controls because I like the chapter/most viewed info those provide. Having “Native controls on Loop Videos” instead of “Include Loop Videos” would probably work for me.

0

5

0

Der Teilweise @teilweise@layer8.space · Feb 13, 2026

Replying to @teilweise@layer8.space

@lapcatsoftware@mastodon.social Feature request for StopTheMadness: If “stop autoplaying audio and video” is enabled, overlay stopped animations with a play button.

0

1

0

Der Teilweise @teilweise@layer8.space · Feb 13, 2026

Replying to @can@haz.pink

@can@haz.pink @noheger@wien.rocks Aaaaah! The 26.2 image is animated! I have auto starting animations turned off and there is no hint that’s a video. I even expected a change of the image when hovering over it but that did not happen. That’s what confused me. I thought the 26.3 image was supposed to be the pre release state.

0

1

0

Der Teilweise @teilweise@layer8.space · Feb 13, 2026

Replying to @noheger@wien.rocks

@noheger@wien.rocks Did you use the wrong image in your blog post for 26.3? Looks square to me:

0

1

0

Der Teilweise @teilweise@layer8.space · Feb 05, 2026

Replying to on mastodon.social

@chbeer@mastodon.social Non AI-generated projects are no longer supported. (The same happened quite regularly for me when I switched between Xcode beta versions and the release version.)

0

1

0

Der Teilweise

Posts

Layer8 in Space

cosmic040: "Wenn Apple hide & seek mit uns spielt - bei Secur…" - chaos.social

Layer8 in Space

Remote instance

Media