Hacker News Reader

Comment by treesknees | Hacker News Reader

I wish people would stop using the term “bricked” for fully recoverable failure conditions.

Jeep has already confirmed they’ve pushed out a fix. That is not bricked.

If it's not user recoverable at the time, and it renders the product as useless as a brick, then it seems like the most accurate word to use, from the customer perspective. Some people will prefer stricter semantics, sure. It was later still able to download and apply updates over the air to undo the problem, so it was a milder form of bricking.

I've had some pretty nasty brickings of devices, like overwriting the bootloader, that I've been able to recover from by getting it into some barely documented system on chip mode with a special cable, booting a new bootloader into RAM via the cable, and reflashing that way. One could go to the extreme and say any flash storage chip where all software bits are directly writable by a factory tool is technically unbrickable. But the customers won't see it that way.

I would love to read the essay that proves the word "bricked" has a highly specific technical meaning that excludes recoverable failure.

The entire premise is that you’ve turned the device into a brick. If the failure is recoverable, it’s not a brick.

Hard to imagine unrecoverable device. Maybe physically melting it into the brick will do the job. In any other case it is recoverable: you can replace whole memory with a bootloader, other corrupted modules and recover device.

I think this may have more to do with a combination of insufficient imagination and fault-tolerant manufacturing.

There are plenty of devices that can be rendered inoperable via non-physical destruction. There used to be more of them, but manufacturers try to make it impossible because it’s a support nightmare.

I think of it as generally meaning "unrecoverable for the current owner" or "unrecoverable for the average consumer", not "even the manufacturer, with hardware access, couldn't find a way to factory reset this."

It's also not literally a brick, regardless of future functionality. The ability to metaphorically compare it to a brick doesn't seem to hinge critically on whether the metaphorical brick is a permanent metaphorical brick or a temporary metaphorical brick.

Sure. I’m not going to nitpick exactly how long or how severely something has been rendered inoperable. If somebody wants to refer to their phone as a brick because they’re camping and forgot the charger, that doesn’t bother me.

I’m just pushing back on the idea that “bricked” is some random word with no meaning whatsoever.

This is truly one of the most HNey threads I've read in a minute.

we have a saying for that, soft brick.

That wording manages to avoid answering the question entirely. It just shifted the ambiguity from defining the word “brick” to defining the word “recoverable”. Recoverable by your grandmother with dementia, before the next Matlock episode airs? Recoverable by one person on Earth for whom unbricking this one thing is their special interest? Recoverable by a hypothetical future technotopian society that we predict will be capable of vaporizing atoms and rearranging them into any desired form? Do you draw a hard line in the sand for everyone to share one definition? Or is it entirely subjective, a tautology, a self-fulfilling prophecy - meaning anyone’s feeling of brickiness and recoverability is valid by definition?

If you can desolder and replace a few ball grid array ICs and then get Linux running on it, it was never True Scotsman's fired clay brick bricked. It was only Lego brick bricked.

I've "bricked" many automotive systems where they weren't truly unrecoverable, but doing so involved another team disassembling them. The parts were cheaper to throw out instead.

Being strict about the word "bricked" and limiting it to the truly unrecoverable situations just makes it nigh-on useless.

Very few things can make a modern system truly unrecoverable if one is willing to pour unreasonable resources into them. It's incredibly common to be in a situation where a system is unrecoverable by you though. There's no practical difference between these two except that one depends on the surrounding context.

I think the reasonable extent of "bricked" is: can you plug in a USB cord and use publicly available software to fix it? Or wireless equivalent.

Most automotive systems would be bricked by this definition. Very little of the tooling is open source/publicly available, reprogramming is usually a specific, non-default mode gated by passwords or cryptography and inaccessible to end users.

There are a thousand ways to describe this without misusing and ruining a word like “bricked”.

Being strict about a word makes it more useful, not useless. A useless word is one with no identifiable meaning, one which requires copious clarifications, or one which invites confusion and debate instead of delivering meaning.

I actually think your first sentence is a spot on definition for 'bricked'. However, this specific scenario does not meet the criteria you've defined. Nobody is throwing out their car because it was only temporarily disabled. Another OTA update fixed it minutes later.

The first sentence is an example that also conflicts with a strict definition of "bricked", not something comparable to the situation in the article.

The definition I was offering just appends "by you" to the strict definition and encompasses both in some contexts.

Apparently for some people the update makes it worse.

https://www.jlwranglerforums.com/forum/threads/2024-4xe-loss...

Not even two weeks after going all-in on enterprise vibe coding including for "engineering workflows".

> [Stellantis'] determination to apply AI across every part of the enterprise

https://www.stellantis.com/en/news/press-releases/2025/octob...

Wikipedia says

> A brick (or bricked device) is an electronic device, specially consumer electronics (such as a mobile device, game console, computer, etc.) that is no longer functional.

These jeeps are no longer functional.

Speaking of terminology, though, "crashed" really takes on an ominous meaning. I am really glad not to write software for safety-critical systems.

I agree, but I can't think of another term that would convey the severity of this offending update.

How about "catastrophic"? Or "total failure"? Or "we can't find the word to convey the severity"?

Anything else than words that already have existing meanings. With that motivation, they could have said "... update that exploded all ..." since it's a really severe situation, but obviously we/they should use words that has the right meaning instead.

"Jeep just pushed an update that was catastrophic to all 2024..."

"Jeep just pushed an update that was a total failure to all 2024..."

Idk... Doesn't have a very good ring, because "catastrophic" and "total failure" in the realm of tech usually means something that if you try again it could possibly work.

As I said, I agree that "brick" is a good word, I just don't think any of the alternatives are any better.

> As I said, I agree that "brick" is a good word, I just don't think any of the alternatives are any better.

As I said, "brick" is as good of a choice as "explosive", but you do you.

It's not even close. We're talking about soft Vs hard bricking, and how "to brick" usually means hard bricking instead of soft, whereas if you said the update exploded the devices you'd be outright lying.

But you do you

> 're talking about soft Vs hard bricking

You might be, but the submission and the submission title doesn't, hence people calling it incorrect. As it stands right now, it's as much of a outright lie as saying they're exploding.

"Jeep pushed an update that made moving vehicles uncontrollable"

If we are allowed to move the goal posts anywhere we want then nothing is ever bricked unless it is smashed to a powder.

ffs. really? the clickbait headlines need to stop - I'm for full banishment of people who post them and the publications they came in on.

Reply on news.ycombinator.com