Lenard Flören, a Germany-based art director at an advertising agency, said he quickly realized that trying to create his dream fitness app with one lengthy prompt would lead to a plethora of bugs that “neither ChatGPT nor my clueless self had any chance of solving.”

If everyone can create programs, and everyone fails, maybe it’ll bring increased appreciation to development and good development and products? One could hope. I guess the worst offenders won’t even try themselves either way. The services are not that accessible.

I’ve aired my frustration about the terminology previously; anyway, I’m trying to accept the terminology in the interpretation it could make some sense:

You tell the AI the “vibe” of what you want the result to have, and it does that - but of course it’s not necessarily that simple. You may end up doing prompt engineering, multiple iterations, trial and error, etc

When we tried a product at my workplace generating a web app prototype in react seemed viable and reasonable, possibly good for prototyping and demonstrating. We also tried a Blazor app, and it utterly failed. I suspect because of less training on it and much more complex mixture of technologies.

, but it works reliably well. It takes a second or two to be redirected to the site you’re visiting.

Do you mean it works reliably well in letting users through, or in blocking AI?

Do you have sources or more information about the effectiveness of it in blocking AI? What else it blocks as collateral damage would also be interesting.

/edit: Clicking through some links (specifically canine.tools) I have to say - it may also be effective in annoying me personally, and eventually exiting those websites. Similar to consent dialogs you could go into settings for and save with opt-outs. But it’s a barrier and user-opposing functionality.

I certainly don’t see it as a simply or only good and effective thing.

It doesn’t open with a summary or overview but dives right in to exploration, but I think the point comes across:

The copy and paste key codes, which have no physical keys anymore, are - to a degree - supported in software. Their claim is that those key codes are the tool for universal copy and paste, and then it’s the input interpretations job (key and combination mapping) to offer bindings to those key codes.

GTK added support the copy and paste keyboards in January 2025. QT also added support for copy and paste key codes the same month. I’m not sure of the first released version of the GTK toolkit that will contain the fix. For QT, it will be QT 6.10, scheduled for release in September 2025. Together, this will cover many apps built for Gnome and KDE as well as others that use the same toolkits.

… followed by some more “current state of support for those key codes”.

merge

You linked a tutorial to sh. Note that nobody ends up shell scripting in sh. People will use bash, which is an alternative shell and shell language, and almost universally available where sh is available. sh is very old and limited. bash is much more common.

There’s many other kinds of shells as well though. And you such an automation task you could use any number of scripting languages. The part that makes it a shell, which is interactive use, is not necessary for a scripting task like this of automating an operation. Shell languages can be used as scripting languages too though. I just want to point out alternatives and context.

Personally, I use Nushell as my daily shell and for scripts and am very satisfied with it. It’s not universally available as in pre-installed, but is multi-platform and easy to install through an exe or package. Because it’s a newer project, there’s not that many resources yet, and still occasionally makes changes to its language with new releases. But, for me, the upsides to other shells are obvious and significant. I posted my Nushell solution in a separate comment (separating concise solution from this general prose exploration).

My preferred shell is Nushell. I would write:

glob **/*.mp3 | wrap mp3 | insert txt { $in.mp3 | path parse | update extension 'txt' | path join } | each { ^mp3splt -A $in.txt $in.mp3 }

or with line breaks for readability

glob **/*.mp3
  | wrap mp3
  | insert txt { $in.mp3 | path parse | update extension 'txt' | path join }
  | each { ^mp3splt -A $in.txt $in.mp3 }

1. glob to find the files (according to pattern from current dir)
2. wrap list values in a named column
3. add column txt with extension replaced by txt
4. => now I have a table with mp3 and txt columns with respective full paths
5. call mp3splt for each

The HackerOne report that does not even apply has 44 upvotes.

What do upvotes mean on HackerOne?

I guess, at least here, they’re mindless “looks interesting” or “looks well worded” or something?

The license doesn’t get revoked. It does not apply to things it does not allow in the first place.

Some kind of restrictions are easier to describe and assess than others.

I doubt someone that generates AI slob reports would care about the restrictions anyway.

3 hours, no chapters? :(

When you’re working for Heinz, is it a mustard bottle?

At my work we explored a low-code platform. It was not low on code at all. Beyond the simplest demos you had to code everything in javascript, but in a convoluted, intransparend, undocumented environment with a horrendous editing UI. Of course their marketing was something different than that.

That was not the early days of low-code mind you. It was rather recently; maybe three or four years ago.

I expect an update on 999

I’d love to read a list of those instances/claims/tech

I imagine one of them was low-code/no-code?

/edit: I see such a list is what the posted link is about.

I’m surprised there’s not low-code/no-code in that list.

Correlation does not equal causation.

Are you saying Google paid for them to stop implementing pwa?

Start as in developing or as in hosting?

You’re asking in a community called devs, implying you want to develop, but it seems like you want to only host?

I looked at Wikipedia; Taler then Blind signature, then looked for docs on GNU Taler where I didn’t immediately find any technical overview of how that works. Phind gave me a seemingly reasonable and understandable answer. (Surely sourced from somewhere.)

When we multiply the original message by rere, sign it, and then multiply by r−1r−1, the blinding factors cancel out while preserving the signature.

The success of this process relies on two critical properties:

• The blinding factor must be relatively prime to N
• The RSA keypair must satisfy the congruence relation red≡r(modN)red≡r(modN)

The magic is that you can

1. Apply a mathematical operation on your data
2. Sign that data
3. Revert/Invert the mathematical operation

and the signature remains valid.

It does sound like magic. But isn’t most of cryptography like that?

There’s a python example in there as well, with such a calculation. I didn’t go through it though.

I just wanted to add an additional resource related to the topic.

There’s a lot more variance in the specifics, but I think for an overview like this it’s certainly missing dual-licensing and “business-open” licenses like “readable but limited now, but free software two years from now”.

But I guess with the specific target audience of this post the reduction for simplicity is a good thing.

https://choosealicense.com/