Hacker Newsnew | past | comments | ask | show | jobs | submit | jxf's commentslogin

In the mathematical sense, all functions are relations, but not all relations are functions.

What are relations, if not an indicator function of a cartesian product?

"Personal blogs are back", says article, without providing any data or evidence to substantiate this claim.


As a native English speaker, "I'm sorry for how you feel" is patronizing. Genuine apologies acknowledge mistakes rather than invalidate feelings.


Agree, containers are exactly what I want and I use them all the time. Profiles are too much isolation, like installing separate VMs to run processes in a completely isolated way when something like Docker is fine.


The key element here is switching windows across profiles. Safari almost gets this right but fails in the UI imho. Edge has "workspaces" that could almost work but they don't provide isolation.


Zen [0] can bridge that gap. I have it set up so that each workspace in practice maps to a specific container.

[0] https://zen-browser.app/


Yes, Zen is extremely good at that. Alas, as it is using Firefox Sync it doesn't sync your sidebar states (yet).


I tried this for over 20 minutes and made almost zero progress because I thought the expressions also had to be words, like a crossword puzzle. Oops. It might be worth clarifying that somewhere.


The idea with really cheap batteries is that they don't need good energy density. You just swap them every so often and put the one you aren't using in the charging rack. You could even carry your own reserve energy with you!


We used to have swappable batteries in virtually all of portable electronics. You could even get them in a rechargeable accumulator format. Virtually all of portable electronics has integrated batteries.


Chinese devices have standardized on taking 18650 "3.7V" cells for this purpose.

We might eventually get back there; maybe the EU will do for e.g. hand tool batteries what they have done for phone chargers and mandate an interchangeable standard.


Yeah but AAs suck, and the newer more advanced batteries all have different voltages and require different charge circuitries so it’s hard to create a new standard for them.


It’s all depends on device size and required capacity. AA is not a bad choice for many cases. And there are other replaceable batteries with higher capacity e.g. 18650.

Most modern devices have an integrated 3.7v Lithium battery so standardisation should be possible but I see no market forces for this - devices with short lifespan (limited by a non-replaceable battery) are more profitable.


The lifespan of the devices I own is generally limited by security patches not by batteries.


How often do you receive security patches for your Bluetooth speaker? A cordless drill? Cordless vacuum cleaner? Cordless shaver? Sex toys...


I don't recall throwing one of those out because of the battery either.


That's true for smartphones (I had bad experience with Android from 2 big vendors which stopped updates around end of sales date) but many other battery powered devices don't need regular security patches.


For me, this limitation applies only to my phone. I have plenty of other wireless devices and deliberately prefer devices that can use AA batteries. One reason is that I don't have to manage the multitudes of internal batteries as much and I need only 1 battery charger with batteries always ready to go. And obviously, battery going bad won't make my device useless (My DS4 controller's internal battery went bad in about a year. So I'm sticking with xbox controllers.)


My mother bought a flash light with non-rechargeable batteries. That type of product is basically destined to be thrown away on day one.


18650 are everywhere from hand tools, drones and vapes to Teslas and scooters. High energy per cell (up to 3600 mAh * 3.7V ~= 13Wh) and fairly cheap


Almost nothing lets you replace the cells individually though. They seal them in a pack which requires buying a replacement proprietary pack.


A circuit that accepts the voltage range of lithium ion is probably 90-100% of the way to accepting a range of cell chemistries. And you can put in a half cent identifier chip to say what the charging voltage is.

I don't think the technical difficulties are the problem here.


It is always the case that custom configurations have advantages over other configurations. But standards give a good trade off between performance vs having a large ecosystem. Integrated batteries just add to the e-waste problem.


Li-ion cells are available in AA- and AAA-scaled sizes, e.g. 14500, 10440, plus fractional sizes that can be shorter or longer for the same diameter. If we wanted to not glue batteries into devices, we very well could, but that would make it harder to force purchase of a new device when the consumable component inside it fails.


I agree that we could use standardized lithium-or-similar batteries in a ton of devices.

But please don't exactly match AA/AAA sizes. That will cause much more harm than good.


I think very few manufacturers are optimizing for that. The move to integrated batteries for most portable electronics happened when the price of the battery plus charging ICs became lower than putting in a battery holder. Doing battery holder is currently simply more expensive, design is more complex putting it together is more complex. The cost are not intuitive, you can get 10+ microcontrollers for a price of a single physical on-off switch.


That and there's an incentive to try to lock customers into your particular battery ecosystem.


Why do you think AA suck ? It’s the chemistry, not the standard size, voltage or swappability right? 18650 and 21700 also have those assets. Some modern devices let you swap 18650.


Every chemistry outputs a different voltage and requires different charge controllers. So sure we could have created a standard lithium size but it would have just locked us in to one chemistry again which will eventually be obsolete. 18650s are also too bulky for most applications. Usually you want flat rectangles. Another benefit of the proprietary batteries is they can completely fill the space available rather than being constrained by the standard.


Swappable proprietaries are still better than not swappable at all. For my previous phone I managed to order an external charger and several replacement batteries.


My understanding is that it's a poor form factor for lithium ion - which operates at higher voltages, and thus needs an extra voltage regulator to step them down to 1.5V if you're packing them into the AA format (adding cost, reducing capacity, & introducing conversion losses.)

https://www.youtube.com/watch?v=uKYF1CXZPng


Did you also use Claude, and you like Codex better, or are you making a more general observation about the leapfrog in creative power agents are bringing to engineering?


I’ll tell you something. I love working with Claude. It’s enthusiastic, it’s nice, it’ll give you suggestions. It’s an all around pleasant experience.

I hate working with codex. It feels like a machine. You tell it to do something, and it just does it. No pretension at being human, or enthusiastic, or anything really.

But codex almost always does it right. And the comments are right, I never run into random usage limits. Codex doesn’t arbitrarily decide to shrink the context window, or start compacting again after 3 messages.

The codex client sucks, claude code is much better. But the codex client is consistent, which is much more important. Claude was amazing 3 months ago. The model is still fine, but the quality of the experience has degraded so far it’s hard to consider using it.


This is my experience as well. Codex is very verbose which is annoying considering the limits. My work flow tends to be have Claude code describe the problem (succinct as it can) based on my mashing of the keyboard description of what I want done then send that to codex. I've tried it the other way around doesn't work nearly as good. Disclaimer: not using the 5 prompts per week opus.


> The codex client sucks, claude code is much better.

Are you using in in VSCode?

I use the web based Codex (which I love) and the VSCode clients of both. I don't think there is a huge difference in the VSCode plugins.

I've tried the cli versions but don't have enough experience to have a strong opinion.


Personally I prefer Codex's less-chatty nature nice. I prefer to save my human emotions for humans.


I'm totally on board with some people preferring that. I don't. However I do prefer my AI assistant to work.


I've used Claude too and I prefer Codex. I generally have confidence in both to tackle large scale problems I won't tackle with Gemini 2.5

I've had a few small bug that Codex has fixed where Claude hasn't.


IMO a certain amount of youthful indiscretion that takes the form of challenging systems and structures feels like it's both tolerable and important. Agitation prevents calcification.


Wait till that guy figures out what the hacker in hacker news means


This is a good reason why even cursory checking for whether dependencies actually get used is valuable.


On the other hand, if you find a way to make it easier (I believe the general case is O(n²) in gates), then you've improved a very hard computer science problem!


Yeah, the problem is in a sense solved asymptotically by the optimal construction in https://arxiv.org/abs/quant-ph/0302002, but that one tends to lead to long solutions in practice, so there's plenty of room to try to come up with solutions that give shorter solutions for concrete instances.


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: