For the best experience on desktop, install the Chrome extension to track your reading on news.ycombinator.com
Hacker Newsnew | past | comments | ask | show | jobs | submit | history | more Jerry2's commentsregister

What's a fair price for a TB of HDD? SSD? Which high-capacity SSDs are worth buying?

I'm not sure if there are any sites tracking this. Anyway, I need to buy 30TB of storage this year so I can upgrade my NAS and make it last few years. Thanks for any replies from anyone who has an opinion!



Thank you.


HDD: $10/TB

SSD: $50-100/TB


Most new drives are sold into consumer/end user retail around $20/TB and only seem to dip a little during sales (with most sellers also quantity limiting any sales, also). Getting to $10/TB is possible but typically involves buying manufacturer recertified, and may involve going smaller (14-16TB drives) than you might prefer.

I managed to find 64x Seagate Exos 20TB for $13/TB new about 2 years ago, on NewEgg of all places, but I’ve never seen that deal repeat. :(

All the new 30TB+ HDDs using HAMR technology from Seagate and WD still feel like expensive unobtainium.


I'm convinvinced that CEO-employee wage gap is due to this not-so-well known legal case from 1919: Dodge v. Ford Motor Co. [1]

Basically, in 1919, Henry Ford sought to reinvest the Ford Motor Company’s profits into raising employee wages and expanding hiring, arguing that sharing success with workers would strengthen the economy and the company’s long-term prospects. However, minority shareholders John and Horace Dodge (who also ran their own competing auto company) sued Ford, claiming that his actions violated the fiduciary duty to maximize profits for shareholders.

The Michigan Supreme Court ruled in favor of the Dodges, declaring that a corporation’s primary obligation was to serve the financial interests of its shareholders and not broader social goals or even the well-being of its employees. This decision established a legal precedent that continues to shape corporate law even today and reinforcing the doctrine of "shareholder primacy" and limiting the ability of companies to prioritize stakeholders (like workers or communities) over investor returns.

It's been downhill for employees since.

[1] https://en.wikipedia.org/wiki/Dodge_v._Ford_Motor_Co.


I think you're off by about 80 years:

> Twenty-five years ago, Bill Clinton campaigned on an idea for limiting excessive pay for American CEOs: Cap the tax deductibility of top executives' compensation at $1 million, and companies, not wanting bigger tax bills, might reel in their pay. In his 1993 budget, advisers suggested a compromise: Companies couldn't deduct CEO pay over $1 million unless it was "performance-based."

> But many believe the loophole had the opposite effect, driving companies instead to pay more in stock options and certain performance-based bonuses, which actually supercharged the growth in CEO pay. In 1989, according to the left-leaning Economic Policy Institute, the median value of annual CEO compensation was $2.7 million. By 1995 it was $6.6 million, and it reached $13 million in 2016.

Ref: https://www.washingtonpost.com/news/on-leadership/wp/2017/11...


Looks like they started measuring during the FDR administration.

https://fortune.com/2025/04/15/ceo-worker-pay-gap-problem-am...


Bill Clinton made irreparable damage to the American workforce. He sold out America to move all manufacturing to China by giving massive incentives to companies to do so. The market boomed and all that money went to investors instead of workers.


While partially true in hindsight, letting China into the WTO, removing tariffs and giving them MFN status was a bipartisan effort beginning in the 80s and 90s. The premise was that China's market would open for US goods and services as well.

It hasn't worked out like that for sure.


It hasn't? If only the US is willing to sell.


>Basically, in 1919, Henry Ford sought to reinvest the Ford Motor Company’s profits into raising employee wages

I don't actually see any reference in the wikipedia article that Ford was saying he wanted to use the dividend money to raise wages:

>...Henry Ford, sought to end special dividends for shareholders in favor of massive investments in new plants that would enable Ford to dramatically increase production, and the number of people employed at his plants, while continuing to cut the costs and prices of his cars.

And as usual in these cases, there are other unstated reasons that might actually be more important to the decision maker:

>...Ford was also motivated by a desire to squeeze out his minority shareholders, especially the Dodge brothers, whom he suspected (correctly) of using their Ford dividends to build a rival car company. By cutting off their dividends, Ford hoped to starve the Dodges of capital to fuel their growth


People always talk about shareholder value like it's some outrageous weird thing. Really, shareholders are just.. owners. And managers are their agents.

Let's say you hire a general contractor to remodel your house. How would you feel about him doing what's good for society without consulting you - e.g. buying sustainable material that is more expensive, or locally sourced material that is less durable or less safe? Or hiring more workers like they do on NYC construction projects cause it's good for labor? Especially if it's something you disagree with, like he's maga and refuses to hire cheaper immigrants, giving preference to disgraced former cops. When the bill comes with all the extra costs, hed just say he's not working for the owner value but for the good of society as he sees it :)


Having people who make enough money to buy stuff is in the best financial interest of all companies in the long term.

Companies and investors need to think further than 90 days ahead.


I'm sympathetic but the problem with this is that you are asking a discrete entity to optimize for the commons. There are more sophisticated solutions.


He wasn't exactly doing it out the goodness of his heart. From the same article:

"Ford was also motivated by a desire to squeeze out his minority shareholders, especially the Dodge brothers, whom he suspected (correctly) of using their Ford dividends to build a rival car company. By cutting off their dividends, Ford hoped to starve the Dodges of capital to fuel their growth."


  > The Michigan Supreme Court ruled in favor of the Dodges, declaring that a corporation’s primary obligation was to serve the financial interests of its shareholders and not broader social goals or even the well-being of its employees.
i am not a legal expert but my laypersons' reaction to this is, how can a court just declare such a thing ... it seems even from the cited criticisms others agree... fta:

  > Dodge is often misread or mistaught as setting a legal rule of shareholder wealth maximization. This was not and is not the law. Shareholder wealth maximization is a standard of conduct for officers and directors, not a legal mandate.


By the same argument shouldn’t CEO pay be reduced?


That's an example of trying to clean up Ford's image.

Ford didn't want to share with the employees. He was very strongly anti-union (granted, not a factor until post-WWII). He was a Nazi supporter and not just because he was a notorious antisemite, Nazis opposed organized labor, too. He is sometimes mistakenly acclaimed for being for civil rights because he hired so many black men (who were not unionized, in an attempt to defeat the unions).

The lawsuit wasn't about shareholders vs broader social goals. It was about shareholders vs the CEO. The article is not about shareholder vs CEO pay. This lawsuit is unrelated.

And before someone claims Ford paid his workers enough to be customers (the reason he still wanted to pay them more in 1918), consider that in the early days after he'd implemented an assembly line the work became incredibly monotonous and workers were leaving for other automakers, so Ford was forced to pay them better to stay with him.


Yep. You can also automatically save them if you use mpv to watch YT: https://github.com/nick-s-b/mpv-transcript discovered this script yesterday.


I switched to Prezto [0] because I found OMZ too slow. Prezto is much faster out of the box and doesn’t have a lot of things enabled by default. Definitely give it a try if you find OMZ too slow on your machine.

[0] https://github.com/sorin-ionescu/prezto


Are there any WMs that are implemented in JS and CSS? Or at least allow you to use JS/CSS? I’d love to be able to use web technologies to manage and decorate windows.


Author didn't disclose if got a reward for his work. Hope he did!


Thank you for your kind words. To respond: 1. I'm not a "he", I would prefer "they". 2. As I mentioned in another comment, I have not received word back yet on any reward.


Maybe they'll put you into their "Hall of Fame"


I think their "Hall of Fame" (or at least whatever people colloquially refer to as that) is their credits for people who found bugs in their web servers, so I don't think that counts here. I did get credited, so I'm happy about that. Now I just have to wait and see if they determine it's worth a reward (and, if so, how much).


It is absolutely worthy of a reward, and it should be worth a few months of your time. This is a nasty security issue, and you showed a ton of restraint not losing patience with Apple.

Honestly, it's bullshit that you don't already know whether or not you're going to get a bounty.


I will definitely admit, it can be a bit of a pain point that Apple sometimes takes a lot of time to determine a bounty. I'm just waiting patiently now to see what they say. I appreciate your kind words and encouragement.


> I typically access the backend UIs provided by each LLM service, which serve as a light wrapper over the API functionality

Hey Max, do you use a custom wrapper to interface with the API or is there some already established client you like to use?

If anyone else has a suggestion please let me know too.


I'm going to plug my own LLM CLI project here: I use it on a daily basis now for coding tasks like this one:

llm -m o4-mini -f github:simonw/llm-hacker-news -s 'write a new plugin called llm_video_frames.py which takes video:path-to-video.mp4 and creates a temporary directory which it then populates with one frame per second of that video using ffmpeg - then it returns a list of [llm.Attachment(path="path-to-frame1.jpg"), ...] - it should also support passing video:video.mp4?fps=2 to increase to two frames per second, and if you pass ?timestamps=1 or &timestamps=1 then it should add a text timestamp to the bottom right conner of each image with the mm:ss timestamp of that frame (or hh:mm:ss if more than one hour in) and the filename of the video without the path as well.' -o reasoning_effort high

Any time I use it like that the prompt and response are logged to a local SQLite database.

More on that example here: https://simonwillison.net/2025/May/5/llm-video-frames/#how-i...


Seconded, everyone should be using CLIs.

Simon: while that example is impressive, it is also complicated and hard to read in an HN comment.


I was developing an open-source library for interfacing with LLMs agnostically (https://github.com/minimaxir/simpleaichat) and although it still works, I haven't had the time to maintain it unfortunately.

Nowadays for writing code to interface with LLMs, I don't use client SDKs unless required, instead just hitting HTTP endpoints with libraries such as requests and httpx. It's also easier to upgrade to async if needed.


most services has a "studio mode" for their models served.

As an alternative you could always use OpenWebUI


I built an open source CLI coding agent for this purpose[1]. It combines Claude/Gemini/OpenAI models in a single agent, using the best/most cost effective model for different steps in the workflow and different context sizes. You might find it interesting.

It uses OpenRouter for the API layer to simplify use of APIs from multiple providers, though I'm also working on direct integration of model provider API keys—should release it this week.

1 - https://github.com/plandex-ai/plandex


>I use Aider. It's awesome.

What do you use for the model? Claude? Gemini? o3?


Currently using Sonnet 3.7, but mostly because I've been too lazy to set up an account with Google.


Get an Openrouter account and you can play with almost all providers, I was burning money on Claude, tried V3 (blocked Deepseek provider for being flaky, let the laypeople mock them) and experimental and GA Gemini models.


Gemini 2.5 pro is my choice


I'm kinda relieved that it doesn't work on an iPhone. I often scan codes posted around to save the time typing URLs and running arbitrary code by just scanning a QR code freaks me out.


Ironically, I actually wrote a blog about how casually we do this and how dangerous it's become lol https://kuberwastaken.github.io/blog/Technology/QR-Codes-and...


The content is good, but fyi the last third or so had a distinct ai padding vibe


AI padding vibe?


I kinda see that too. Basically, the way some authors use to increase content size with redundant words is the default behavior for Ai chats plus all the disclaimers to avoid possible litigations or negative public image.


It runs inside a web browser though. This is no different from visiting an arbitrary link and running whatever arbitrary code in the Javascript sandbox of that link and one already knows a q.r. code an take one to an arbitrary link.


That said, I wouldn't mind an upgrade to the standard of say say if the link be printed above the code in human readable form in some way, the reader would refuse to open it, or at least be configurable to refuse to open it if they not match.


This QR code does. But what about a QR using similar designed by someone less honorable? With QR codes, you have no idea what will happen until you scan it. At that point, it could be too late


As far as I know the only form of code execution they support is by the URL datatype which carries the same risks as wel already mentioned anyway.


How is this different from opening any website through a QR code, that will then run "arbitrary code"?


Wait until you hear about javascript on web sites.


I love Inkscape so much. I use it every other week to make presentations, slides or just simple graphics when I need it. I illustrated my thesis with it.

Another piece of 2D vector software that I use and recommend is Graphite [1]. It too is open source. Graphite has nodes and can be procedural in nature. Have them both in your graphics toolbox.

[1] https://github.com/GraphiteEditor/Graphite


Thank you. This is the first time I've heard of the Graphite editor.

I think it could benefit from availability in package reposirories and looking at the license, it appears the program itself is free under the Apache license but the artwork is non-free. It cannot be modified and commercial use is prohibited.

When combined, Graphite as a whole is non-free so I won't be compiling this until the artwork is gone. I'll look into providing some dummy files and see where it goes.

The other packing problem is that there's already a server monitoring program also named Graphite. But given the above, a change of name and new icons would solve both problems.


For those who are confused I believe you're referring to the application logo and branding. So basically the Firefox > Iceweasel case again. While the art you create with it is still yours.


Of course anything you make yourself is yours. I'm not sure how the license can confuse anyone into thinking it wouldn't be. It likely wouldn't fly with the law anyway. Unless it was all saas.


You'd be surprised. There are a lot of people out there (and here) who think that unilaterally-declared LLM licenses can dictate what they are allowed to do with the output. Hell, a lot of people think that LLM's can be copyrightable. Copyright law is very clear about the conditions required for copyright protection, and LLMs fail to satisfy at least two. Even failing one is sufficient (see the Smiling monkey case).


Another gem in my graphics toolbox is edraw [1], an embedded vector drawing tool for Emacs. I use it to make quick sketches in Orgmode files, and it is highly recommended for any Emacs enthusiast.

[1] https://github.com/misohena/el-easydraw


That is truly impressive. I have a hunch though, that it is not well suitable for proper diagrams, due to all the things like wrapping text that is too long for a line inside a shape, inside the shape onto more lines. Dynamically wrapping text when a node's size changes. Moving edges when nodes are moved. All those things one takes for granted in an actual diagramming tool.


Inkscape has been a staple in my toolbox for decades. The only gripe I have is that it could have been a bit more stable at times.


Is there any open source hybrid vector and bitmap drawing tool?

I used to use Xara Extreme and it's very fast, intuitive and handy mainly due its hybrid features. It's also used to be open source but not anymore:

http://www.xaraxtreme.org/


Xara was never fully open source. They kept the renderer proprietary, so it never really had a chance.


Graphite has raster editing in their roadmap[1], but afaik it's not really there yet.

[1] https://graphite.rs/features/


Krita is the big one.


Krita doesn't have good vector controls, sadly.


Graphite looks pretty cool. Thanks for sharing!


If you've not, please donate x


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

Search:

HN For You