Hacker News new | past | comments | ask | show | jobs | submit | quesomaster9000's comments login

LLMs are much more than search, for example today I went through several different recipes for sea bass fillets, and then went into much deeper conversations, and ended up with this weird intersection between Bon and how to aptly describe Zen, then very abruptly tried to hone some Bulgarian grammar, then pondered upon the enshittification of hacker news.

To the point that I'm disappointed with human contact.

If you're using it for React.js you're the problem...


That sounds like a lot of time wasted with nonsense when you were looking for sea bass recipes.

Sorry, but when I’m looking for recipes, I don’t really want to go down a rabbit hole of anything else. Especially not AI slop about nonsense.

What’s more, I doubt it randomly started talking about “zen” or whatever.

You likely prompted it.

A prompt is effectively a search query. LLMs can do a lot more with semantic information than read search algos.

> To the point that I'm disappointed with human contact.

Seek help… seriously… that’s not okay…

Find better people to fill your life with.

> If you're using it for React.js you're the problem...

What does this even mean?


With SoundCloud I've found you have to pay for the 'Plus+++' subscription or whatever it is to not get audio with the higher frequencies absolutely butchered, unless you follow a very specific upload process that bypasses their conversion.

Upload 320kbit encoded MP3? Sounds great.

Upload a high-khz WAV? It gets butchered, the top-end turns to glittery noise.

Maybe others have different experiences, but honestly it felt like I'd been duped when paying for the subscription but still got trash quality audio, only to have to pay more.


Absolutely, but not just that you can jump into meta conversations extremely quickly, and rotate things through different dimensions, even with concepts that are much less familiar. But the key is in deliberately creating the transactional context where ... if anything it's like the most fantastic debugging duck I've ever come across.

It absolutely will not necessarily find the most obvious errors, but you will learn a lot in the process.


Is it just me, or does the linked microsoft.com page hijack the back button?

Many microsoft.com pages do. I don't know how it hasn't infuriated them enough to fix it.

At this point I wouldn't be surprised if their site loaded a WASM-compiled remote-desktop viewer to interact with 'Edge in the cloud' just to view the page you want.

Edge your users by bringing Microsoft Edge to edge devices with Edge Cloud edge Edition Advantage!

E: 365


Jeez don't give them such Lovecraftian ideas.

TL;DR PEDMAS is bullshit

If you look at the internal state of the typical physical calculate it's a beautifully simple machine that tries to catch the balance between RPN and logic... except it has stupid nonsensical human rules interjected.

So, every time this article where points out "actually it's more complicated than it seems" is where historically somebody has made deliberate design decisions to do it 'the dumb way' even though it adds complexity, to force something unnaturally.

And what you end up with is something that's half intuitive, it's in between the two systems, but it's more complicated than it needs to be.

Please, I would much prefer device for RPN calculation, that follows consistent logic, rather than silly imposed rules.

4+7*3 etc.

or 4 7 + 3 * =

Where the latter is much closer to what we do in our heads and conceptually think about it.


My grandma, born in the 30's, isn't any good at computers or technology in general, and has trouble operating even the simplest TV remotes and cordless phones.

One thing she can operate is her HP-12C calculator. At the time of it's initial release, someone taught her how to use it, knowing that the RPN interface would be much easier to teach her than infix notation that was equally popular at the time. She never figured out infix notation, at least for more than single operations.

Pretty much everything uses infix notation now a days, so that's what everyone learns first, making RPN an additional skill that usually gets passed by.

It's interesting to learn that there was a time when a calculators interface was a toss-up, and some people learned RPN because it was the easiest to learn, as opposed to it now being an additional skill that would conflict with earlier experience.


Infix isn't "nowadays". It's been with us for hundreds of years. Moreover infix constructs occur in natural languages. You know "apples and oranges".

Someone born in 1930s would have learned infix arithmetic in school just like somebody born in the 2010s.

Cannot work a remote, but can use RPN on an HP calculator? That sounds AI generated.


It's nowadays that everyone learns infix calculator input methods as children. In the 40's they sure weren't learning any calculator user interfaces.

It's stateful and modal interfaces that are an extra step she never figured out, and some TV remotes require different modes for different functions, especially for different equipment. CEC has solved this on modern TVs, but it used to be common to have control the VCR/DVD player by selecting a mode, then control the TV by selecting another mode, and even without the other equipment, she could accidentally get the remote in the wrong mode.

She never got the hang of the states that are possible in a calculator with an interface using infix notation, so instead of using parentheses or order of operation, if she didn't have access to an RPN calculator, she'd write down results and enter them back as needed, to only have one operation in progress at a time, in the calculator.


Other than advanced, graphing calculators that take full blown formulas as input, ordinary pocket calculators are not infix.

They are prefix operations that take an implicit accumulator as an argument, resulting in something that superficially mimics a purely left-associative form of infix, with all operators having the same precedence.

For instance when we punch in 2 + 3 x 4 / 10 x 3 = = =, it means:

  2    # prepare operand 2 in display
  +    # move 2 into accumulator, prepare for addition
  3    # prepare operand 3 in display
  x    # complete addition of acc 2 and operand 3, move 5 into acc and display
  4    # prepare operand 4 in display
  /    # complete multiplication of acc 5 and operand 4, move 20 into acc and dispay
  10   # prepare operand 10 in display
  x    # complete division, moving 2 into acc and display
  3    # prepare operand 3 in display
  =    # complete multiplication of acc 2 and 3, moving 6 into acc and display
  =    # complete multiplication of acc 6 and 3, moving 18 into acc and display
  =    # complete multiplication of acc 18 and 3, moving 54 into acc and display
It's really

  $ echo 2 | plus 3 | times 4 | divide 10 | ...
I remember doing math drills like this in elementary school. The teacher startws with a number and then calls out operations:

"Start with 3; add 4; mutiply by 2; divide by 7; ....; ... write down the result."

I'm deeply skeptical that someone who had been through these drills would have trouble with a dollar store calculator.


Basically any scientific calculator will do proper infix with correct operator precedence, no need for a graphing one (or even one that displays the entered formula, the original Casio fx-82 for example only shows the number currently being entered, yet still does proper infix with correct operator precedence).

Yeah, that's why I stick with RPN calculators.

Working RPN requires a substantial increment in intellectual capacity compared to applying a sequence of elementary operations to an accumulator. An operand consumed by a given RPN operation can be quite distant from it in the syntax, such that it's not obvious which one. A valid RPN expression could consist of an initial sequence of N operands, where N is limited only by stack depth, followed by a sequence of N-1 operators to decimate the operands.

TVs and their remotes mostly have fantastically terrible UX.

My MIL can operate her phone and iPad but is constantly confused by the Roku and has to ask for help.


>RPN

"Reverse Polish notation", I think, for people like me who aren't familiar with the acronym.


You thank!

!ęjukęizD

It really should stand for “Notation: Polish (Reversed).”

I don't see how PEMDAS is relevant for a 4 function calculator. The implicit equals when you press an operator means that they always work sequentially rather than jumping around.


I think they're saying it's bullshit because if we used (the superior) RPN across the board, we wouldn't need PEMDAS at all


RPN requires implementing a stack on silicon, which adds even more complexity and uncertainty. How deep can the stack go? What happens to your calculation when you exceed its capacity?

Classic HP calculators used a fixed stack of four registers. In practice, you can do quite complicated things before you blow that up. Especially if you understand how the mechanism works, and plan ahead a bit.

Besides, infix-style calculators also seem to use a stack to handle operator precedence — if you squint a bit at the manual, you can usually see the shunting-yard algorithm at work behind the curtains.

Actually, some really early HP:s (only desktops, though, like the 9100) used a three-level “stack”, which I put in quotes, because it relied on the user to shift the registers up and down as needed. But they cheated a bit and displayed all three registers, so that the user could see what was where.


> Besides, infix-style calculators also seem to use a stack to handle operator precedence

True, but simple calculators, even ones a bit more advanced than the one in the article (e.g. with a sqrt and/or % key) don't handle operator precedence. I'm not sure it's even correct to call them infix calculators, for that reason. They simply work sequentially; no stack needed.


Compared to what? RPN allows for complex, nested expressions for which infix requires nested parentheses. Nesting requires a stack, the same way.

As far as I can tell, PEMDAS has nothing to do with the calculator itself and is just the rules we’ve arrived at for how to read and write equations in a consistent way.

You also don’t need pedmas if you just use parentheses to disambiguate your expressions. That’s what mathematicians would do.

Go supports this with `go run`, e.g.

    # Example 1: Running the latest version
    go run github.com/rakyll/hey@latest

    # Example 2: Running a specific version
    go run github.com/rakyll/hey@v0.1.4


Depending on how much space you have, I'm using a Numark MixTrack Platinum FX, with the mapping from the forum.

Everything works! Even the display in the middle of the jogwheels, that comes with the '4 deck' version, which shows BPM, time left, rotation indicator, shift amount/mode.

At this point almost all buttons & knobs on it are in regular use, although some of the deeper menu combos accessible via buttons aren't fully intuitive but don't seem to matter.


I'm really happy that 2.5 added 'beats until next marker', which together with a USB controller from Numark I have pretty much feature complete DJ setup for under $500 (including cost of laptop & controller) without having to rely on Windows, Mac, subscription licenses or feature-crippled 'lite' versions.

And it's surprising how quickly people adapt to it when they're used to other setups, within an hour a few people have gone from 'oooh, can I have a go' to showing me their own tips, tricks and different styles.

Especially combined with a youtube & soundcloud downloader running on a different workspace, I can get pretty much any track into the library within a minute or two.


Which controller do you have?


I considered renting out some Pioneer equipment to add RekordBox playlist writing support, mainly because I'm in a similar situation. Mixxx does have support for reading USBs & SD cards, but not writing ;_;

Unfortunately it's far down my priority list given the cost, my unfamiliarity with Mixxx development and that I rarely do anything without Mixx.

But this is absolutely what ecosystem grants/bounties should be for.


I think there are two projects who worked on it a bit:

* https://github.com/kimtore/rex

* https://github.com/Holzhaus/rekordcrate (Current status of export: https://github.com/Holzhaus/rekordcrate/pull/103 )


Mixxx dev and main author of rekordcrate here. The issue is that the format is an ancient proprietary format that does not have been reverse-engineered to a degree that we can properly write Rekordbox DBs yet (Rekordcrate is based on the work done here: https://djl-analysis.deepsymmetry.org/rekordbox-export-analy...)


Would be cool if there is Foss firmware to install on pioneer cdj. It just feels better for a lot DJs. Any projects in this direction?

BTW, the DJ collective I'm with has Pioneer CDJ. If anyone need use the hardware let me know. I'm trying to get them to start using open FOSS software, but it is hard.


Haven’t tried it myself but CDJ-3000s have been rooted.

https://github.com/connorworley/cdj3k-root


You could put a lot of time and effort into getting your own firmware for CDJs .. or you could just connect them via USB to an pi running mixxx.


What you mean? Can CDJ be hooked up to a pi using mixxxx?

The whole point is that the DJs in my collective are used to play vinal, so the pioneer CDJs just feels nice on the hand.

Yes some of the newer dj controllers r quite nice, but rekorbox just got to all the clubs first, so it's everywhere. Kinda like how Adobe is the defector for graphics.


You can use CDJs as a controller. Not with Midi, but in HID mode.

I have only tried it with traktor, but AFAIK it should also work with mixxx. Pi, laptop, whatever. And you can keep using the screens on the CDJs.

And the next user can keep using the USB, no hardware or software modification necessary. Much easier then coding a new firmware.

Mixxx, Algoriddim DJay, Traktor, Rekordox, probably Virtual DJ. Connect CDJs and mixer via USB and off you go. If you have an mixer with no build in audio interface, you would need a external sound card (a hassle because a lot of cables) or aggregate all the CDJs into one virtual card.


> But this is absolutely what ecosystem grants/bounties should be for.

I'd be down to fund a bounty/grant for this (and I'm clearly not alone), if anyone is looking for booty :)


See my other comment to parent, there are two libraries who started, but did not finish. Maybe they can be incentivized to start up again.


I'm a bit annoyed by the requirement of having to use rekordbox exports on some DJ devices and would probably chip in to this because I will definitely not try to install rekordbox again :D


> RekordBox playlist writing support

Without the Rekordbox beat analysis, you'd only be getting track names, and you can achieve similar functionality on the Pioneer side by just putting the files for a each playlist in a separate folder.

There's a kaitai bin format parser config file for the usb db file but kaitai isn't expressive enough to read it properly from that.

Pioneer (err Alphatheta)'s stranglehold on the industry is a shame due to lack of interoperability, among other issues with Rekordbox.


An export db with title, artist, album and rough BPM could be useful to interoperate with the native browse and search menus on a CDJ.

Without beat grids have to best match by ear but that’s par for the course for many DJ scenarios.


https://djl-analysis.deepsymmetry.org/rekordbox-export-analy... (and playlist entry rows) should have a complete enough description if someone wants to give it a try


The Rekordbox.xml format [1] is quite simple. Note the `<TEMPO>` and `<POSITION_MARK>` tags.

But Pinoeer can be weird and there's a lot of old (even ancient) equipment out there, so sometimes it's a case of finding the lowest common denominator and sticking to that.

[1]: https://cdn.rekordbox.com/files/20200410160904/xml_format_li...


Could not agree more. I recently made the switch to exclusively using Linux on Desktop machines, yet I now have to have a Windows PC laying around for the sole purpose of updating my USB using Rekordbox.


I agree, it's absolutely and entirely bullshit and is becoming an easy sloganism that you can bash around. Where "replacing developers, engineers, and scientist with AI" really is about 'doing more with less'... or... just 'doing more, faster?'.

But it is genuinely replacing workers, in real terms. And more importantly it's replacing services, which in turn means jobs and revenue. So they too turn to electricity & automation, so they can do more... so they can remain competitive.

So uh yea, I 100% agree with you there, and wish that more people saw straight through it like you do.


> But it is genuinely replacing workers, in real terms.

Replacing whom? Can you name a single company that has laid off employees because AI is now doing the work?


Consider applying for YC's Spring batch! Applications are open till Feb 11.

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

Search: