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 esens's commentsregister

NSO seems to be used by tyrants to go after legitimate opposition. The NSA isn't used by the President to target the party out of power no? But in NSO case in India apparently it was: https://www.theguardian.com/news/2021/jul/19/key-modi-rival-...

NSO is used to keep those with money and access to NSO in power undermine their legitimate rivals. It can be used to plant evidence on their devices as well as monitor everything they do.


To clarify, are you arguing that NSO Group have had a bigger impact on innocent people, suppression of speech, etc. than the NSA?

If so, I'm not sure I buy what you seem to be arguing, that "NSO case in India" and "It can be used to plant evidence" makes it anywhere near as bad as what the NSA has done/does. In my opinion this is exactly how a "poor-man's NSA" would look: What your money can buy from greedy corporations protected by nasty governments.

>legitimate opposition

Who decides what is legitimate though? It sounds like weasel words to me, just like "terrorists" (that get defined by those in power and then maybe later becomes revolutionists and heroes if they actually win). Going after Snowden, torture in Guantanamo, and using three letter agencies for industrial espionage is also "legitimate".


> arguing that NSO Group have had a bigger impact on innocent people, suppression of speech, etc. than the NSA?

I'm not the OP, but maybe a way to put it is that impacts are more variable or chaotic?

Generally speaking, the "impact" of a US government entity is reasonably predictable based on US policy and interests. Something like NSO, where tools are sold on the market to many entities are probably less predictable and thus more impactful. I'd expect a lower level of operational discipline from <random mideast state> than from the US military.

The other factor is who are NSO Group's masters, and what do they know? If <random mideast state 1> compromises <random mideast state 2>, does <third party> get intel?


The people decide what is legitimate opposition by elections.

NSO was used to tap the democratically elected leader of opposition in India. Doesn't get any clearer than that. [1]

I don't know how to compare between hot pan and the fire on who is worse

[1] with Watergate and more recent (unproven) accusations on wiretapping of politicians, the US is no stranger do this type of monitoring either


>The NSA isn't used by the President to target the party out of power no?

No, definitely not.


[flagged]


^This person is wrong. FISA and FBI counter-intel have a low bar to get warrants because that's what congress intended.


Indeed, the FISA court only exists to rubber-stamp warrants.

In 33 years the FISA court granted 33,942 warrants, in that same time only 12 were denied, a rejection rate of 0.03% [0]

[0] https://en.wikipedia.org/wiki/United_States_Foreign_Intellig...


The rejection rate, in a vacuum, isn't evidence that they are rubber stamping warrants.

From a logical perspective, it could mean that those submitting requests are able to avoid sending weak ones and choose to do so.

I'm not saying I believe you/the standard view is wrong, but there must be some other evidence.


On the contrary, a rejection rate that low implies rubber-stamping, prima facie. You would need positive evidence to support your assertion, e.g. that FISA submissions are unusually high-quality. The actual case is, I am sure, that the system was constructed to make allowing the warrant to be easy, rejecting it hard, and the people involved are just responding to incentives. Namely, since it's all secret they are only accountable to each other, so why give each other a hard time?


>a rejection rate that low implies rubber-stamping, prima facie. You would need positive evidence to support your assertion

I've deftly avoided ever taking a class in statistics, but I have gathered there are two schools - Bayesians, who are honest about having priors, and everyone else.


What sorta qanon crap is this?


Anyone notice that this statement from NSO in the article doesn't make sense:

"NSO does not operate its technology, does not collect, nor possesses, nor has any access to any kind of data of its customers."

If this is true, how do we have a singular list of all phone numbers penetrated? If there was this type of "segmentation" or firewall between NSO and its clients, why was there this huge central data leak?

NSO is tracking what its clients are doing. It may not be telling its clients it is also tracking them. I wouldn't be surprised if NSO could also access every one of those penetrated devices as well independently of its clients.


They are trying to claim that the service is so fully automated that it is the client that does the selection of the target. They claim that their system does not require any fine-tuning from their side, etc.

And that's totally bullshit.


“It should be noted that no ethically-trained software engineer would ever consent to write a DestroyBaghdad procedure. Basic professional ethics would instead require him to write a DestroyCity procedure, to which Baghdad could be given as a parameter.” - Nathaniel Borenstein

Quoted at https://blog.codinghorror.com/your-favorite-programming-quot...


So the good old plausible deniability?


It could mean that NSO controls the infrastructure that manages the tool, but that they don't actually collect the data themselves. So what they said could technically be true if all they do is manage the infrastructure that enables their clients to do the collection of data.


But do they have access to the phone numbers that their customers are targeting? That seems by itself to contradict their statement ("nor has any access to any kind of data of its customers") right there.

Something isn't adding up.


> Something isn't adding up

It's bullshit at best.

If we assume they aren't lying, which is generous given their track record, it could be that they provide the tools and infrastructure to collect the data, but don't instruct the software to collect the data. Sort of like if I had a loaded gun and told you I would point and shoot it where you told me to, and then argued that I didn't technically make the decision. It's technically true and complete bullshit.


But then where did the list of numbers come from, if there is no "access"?


They could be lying, or they could just be trying to use weasel words. "Data" could be referring to collected data, and they consider phone numbers "metadata". I haven't been following the story though, so I don't know which is more likely.


Seems more likely they’re lying.


How does that clear with "NSO does not operate its technology" though?


Thank you. I was trying to understand this myself.

NSO seems to be trying to distance themselves from how its software is used by its "clients," but that seems undercut by the plausible supposition that NSO knows exactly who its clients' targets are.


The owner also goes around suing people regularly: https://insight.rpxcorp.com/entity/1308032-wsou-investments-.... Thus you definitely do not want to touch that technology in a commercial product.

Although it does seem like it expires next year. Sweet!


Sounds like a good theme for a Patent Expiration Hackathon!

A great project would be to upgrade the Blender Noise Texture Node, by adding a Perlin/Simplex switch:

https://docs.blender.org/manual/en/latest/render/shader_node...


Has someone come up with a mapping from Perlin to Simplex? It looks to me like Perlin is slightly low resolution and less amplitude scale. Thus it should be possible to write an equation that mostly maps one to the other -- thus simplifying replacing Perlin with Simplex and vice versa while keeping the distribution relatively the same.

With statistics I bet one could derive this mapping equation pretty quickly. Then it would be interesting to compare the resulting residuals or statistical differences at that point. Because it would be comparing the actually differences rather than these more surface differences.


I'm not sure that would be possible. These are both just ways of problem-solving the gradient of diffusion of random bit noise once it's blown up to a larger size. The Simplex method appears a lot more prone to volatility. The peaks and valleys are more extreme given the same parameters. Crushing the peaks and valleys would lead to compression artifacts or else lengthening the frequency to stretch out the band. Maybe I'm wrong, but fundamentally the fastest way would be to reproduce the underlying noise and then re-render it in the other method.


If we have two different random number generators, and yours has various artifacts that make it non-random in some circumstances (and perhaps mine has different artifacts that make it non-random in different circumstances) then there isn't just some simple "mapping" you can do to go from yours to mine.


I was trying to say one could map the std deviation and mean feature size between the two. Currently that isn't done. Thus the 2D scale appears to differ between the two as well as the min/max delta.


You are basically saying do not get upset about this because maybe it is manipulation. But if it turns out it isn't manipulation, you have successfully tamped down outrage over something that was actually deserving it?


The issue is in Egypt and other Arab nations they never meet any Jewish people at all. They just see the stereotypes, etc. When you never meet people you can view them as complete evil because you buy the propaganda.

This leads to the contact hypothesis as a means of reducing prejudice: https://en.wikipedia.org/wiki/Contact_hypothesis


This is a fact of life for Israelis and surrounding Arab countries. I am unsure why we pretend that these negative sentiments about each other do not exist -- they are incredibly widespread in these populations.

Unfortunately in the countries around Israel with whom they have had wars, there is significant anti-Semitism: http://news.bbc.co.uk/2/hi/middle_east/3136059.stm

And in Israel there is significant anti-Arab sentiment, which rose significantly as a result of Palestinian terrorist attacks in the early 2000s: https://www.timesofisrael.com/poll-shows-large-swaths-of-isr...


isn't anti-semitic and anti-arab the same term? historically/technically speaking


No, and it never was. The term 'anti-semitic' always meant prejudiced against jews. This is not changed by the different meaning of 'semitic' in e.g. 'semitic languages'.


Oh, I didn't know that. Thanks for the info!


This makes sense. It is a competitor to Amazon's Windows Desktop environments. Microsoft should beat AWS's offering here because well, they make the OS that they are virtualizing.

I can see this working for a lot of places where you want good information security. You can never get the data out of the cloud except via screenshot.


> Microsoft should beat AWS's offering here because well, they make the OS that they are virtualizing.

Not necessarily. They have already had a similar service. This seems to be a simplification of that service? Maybe it's sort of like Amazon offering Lightsail as an alternative to EC2? After clicking around the site for a half hour trying to figure out pricing for Azure Virtual Desktops, I gave up. I figured if you have to ask how much it is, then I can't afford it. Now they are launching a service seemingly directed to users like me, but they still can't give us prices. I assume they will in time for launch, but I'm getting really sick of hunting for prices on Azure. Every minute I have spent on that site has been wasted.

I doubt they would beat AWS on pricing. The AWS instances appear to be really well priced relative to what you could get if you tried to do the same with EC2. I couldn't see MS being as cheap. From what I remember, Azure is generally more expensive across the board.


Microsoft's secret weapon here is that if you subscribe to Microsoft 365 Business/Enterprise (which is a prereq for this according for what they've said), they include a Windows license that can be ran on Azure for VDI. On AWS, you have to buy CALs and licensing for the instances (Windows VDA licenses? I last looked into that years ago). Or you can buy the licenses from AWS but I'm sure that's not the rate you'd pay getting them directly from Microsoft (and then you often have to use Windows Server which doesn't have things like WSL2).

That's not a gigantic barrier if you're an enterprise, but figuring out licensing for Workspaces if you don't have a couple smart Microsoft admins and money to burn is really hard. This is easy, since all of the license entitlements are enforced through their portal. (if you can start an instance you are licensed)


If you want an AWS instance, it's not much more complicated than pick your instance and go. Of course, we're talking about the AWS console here, which is not so easy if you aren't used to it. If you know your way around, then it's about the same as getting anything else AWS. You don't have to deal with licenses at all.

The pricing is really cheap. Looking at the pricing of instances on EC2, I don't know how they get so cheap. I'm sure the answer is in the details if I were to go digging (different hardware) but it's cheap.

I don't think the Asure alternative will be even close. While AWS makes Workspaces easy and affordable for just about any individual, that person is probably still not the target for MS. I'm guessing this new offering is simply the same service as they already offer, for businesses which want systems for less than 100 people, and are still able to dish out a load of cash.

The benefit to Azure is that it's Microsoft. They aren't going to compete on pricing for anything. If pricing is your main concern, then you don't to Azure (or even Google Compute.) If your boss says you have to use Azure, then that's when Azure has the advantage. ;)


> around the site for a half hour trying to figure out pricing for Azure Virtual Desktops

Pricing for Azure Virtual Desktop starts with a hundred users minimum. (available at https://azure.microsoft.com/en-us/pricing/calculator/)


The service is free (you need Microsoft E3/E5 license), you just pay for the VMs, which you can customize any way you want.


> Microsoft should beat AWS's offering here because well, they make the OS that they are virtualizing.

I don't see the obvious connection here. Software is easy to install. But Azure is not really close in scale of deployment to AWS.


It is the "next closest" right?


Sure - they won't be much disadvantaged. But I meant that if they went Azure only or tried to be the only provider, they'd be shooting themselves in the foot. I completely expect them to want a good experience on AWS hosted desktop. Maybe not the best, but I don't expect them to have a significant advantage. (apart from possible integrations)


And until now at least, workspaces seemed nicer than Microsoft VDI. I presume this changes it. Look forward to testing


> good information security

> You can never get the data out of the cloud except via screenshot.

pick one


100% guaranteed, cast-iron, watertight security is impossible - but VDIs could potentially prevent whole classes of attack. It's about increasing the barrier against realistic threat models.


I completely agree - many attack vectors are more difficult, and some are impossible. However, saying "you can never get the data out of the cloud except via screenshot" is, first of all, untrue, as other means exist, and second, preventing access to data is not the most important security aspect of such a system.


Very neat and well done.

I wonder how hard it would be to make one that takes hand drawn input and converts it to professional like charts. Basically pick the symbol output style and it converts your hand drawn charts to this.


React, Kubernetes both show the big drop at the end of 2020. Strange.


Consider applying for YC's Summer 2026 batch! Applications are open till May 4

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

Search:

HN For You