Hacker News new | past | comments | ask | show | jobs | submit login

That Apple documents that 'normal' VPNs are broken on iOS doesn't change the fact that they're broken.



VPNs were always meant to carry internal traffic to a private network, not the public internet (hence the name Virtual Private Network).

The fact that a VPN server can send you a route for 0.0.0.0/0 always was and always will be a happy accident.


> VPNs were always meant to carry internal traffic to a private network, not the public internet

This. And the idea that these so called ‘VPN’ services somehow improve your security and privacy on the internet is laughable. All they do is let you get onto the public, untrusted, internet through a different on-ramp. There is no point to them. The internet is just as untrustworthy through a VPN service as it is through any other internet connection.


> All they do is let you get onto the public, untrusted, internet through a different on-ramp. There is no point to them.

Not true, at all. There are several good reasons to use VPNs to get a different on-ramp to the otherwise untrusted internet.

  - Avoid ISP tracking: Your ISP should see only traffic to and from the VPN. 

  - Access content intended for those in other regions: Many sites and services only show certain content to people who enter the internet from specific places. 

  - Limit the amount of activity linked by trackers: Visiting certain sites only from different IPs/browsers will help keep logs of that traffic isolated from the logs of your other browsing.

  - Allows you to connect to sites and services that cannot connect back to you once you've disconnected: A lot of people, even those with dynamic IPs, keep their address for months or years at a time. VPNs provide a great way to cycle through IPs. 
VPNs don't solve every problem, but they're a powerful tool to keep in your tookbox.

There are many many very valuable uses for VPNs some that offer privacy/security benefits and some that are just plain useful. It's wild to hear anyone say that "There is no point to them."


Thank you for saving me the typing.

No it's not gonna make you an invisible unhackable ghost, but at least I don't have to worry about my ISP screwing me over.


Why are you worried about your ISP tracking you but not the VPN company? In effect you've simply added another ISP on top.


Most people can switch VPN companies more easily than they can switch ISPs.


This is the case for me. I _know_ that my ISP is untrustworthy, I've read the ToS.It's also bundled with my apartment and I can't switch.

What I can switch however is what VPS company I use, or what commercial VPN I connect to.

Plus, the harm from being banned by a VPN is a lot lower than an ISP, as low as the chances of either are.


Wouldn't a VPN also prevent traffic sniffing if you're connected via a public wi-fi point?


It would, but that’s already generally impossible thanks to everything being on HTTPS (including DNS).


HTTPS typically still have the site domain in clear due to SNI.


> HTTPS typically still have the site domain in clear due to SNI.

Can you eli5 that please?


To support multiple sites per IP the browser has to send DNS name of the site to the web server. Moreover, since certificates that the server uses for encryption depends on the site name, the name cannot be encrypted within HTTPS. So the browser sends the name in clear when initiating the connection. This is called SNI, server-name-identification.

It is possible to encrypt SNI, but most sites do not support that as the setup is non-trivial and error-prone.


Huh

So HTTPS doesn't really stop tracking, it only prevents people from snooping on what data you are sending to and receiving from the website


There has been some work on encrypted server name indication: https://www.cloudflare.com/learning/ssl/what-is-encrypted-sn...


I use CloudFlare's 1.1.1.1 DNS to mitigate my ISP tracking the domains I visit. They have an iPhone app too. I don't know enough about the space to say whether it's 100% effective.


Again, due to SNI tracking sites that one visits is possible even if DNS traffic is encrypted. At the very least one needs for the site to support ESNI or encrypted-site-name-identification.

But then in practice for ESNI to be effective one needs for the site to use CDN or similar solutions to ensure that there are a lot of sites for the given IP.



You just replaced ISP tracking with VPN tracking

And your "VPN in YouTube ads" surely does that


I rather have my VPN track me than ISP does. For example, connection to Signal and Kakaotalk servers is used as an evidence of being a member of a terror organisation (!) in Turkey. Evidence is gathered by Ministry of Communications from the ISPs, whom they have to provide real-time connection data to the State including CGNAT records. With a VPN, it is almost impossible (depending on setup, i.e. NAT) to obtain such an evidence - especially for political purposes, even if the VPN provider willingly give logs, no way to prove that info is legit.


I guess it depends who your adversary is. If it's the Turkish government, then you are probably safe to trust a foreign mainstream VPN. If you're trying to avoid scrutiny by a five-eyes western government, you need to be somewhat more careful.


But that’s a general trueism in security. Whether your changes are effective depends on what and whom you’re trying to protect against.


> For example, connection to Signal and Kakaotalk servers is used as an evidence of being a member of a terror organisation (!) in Turkey.

And you are sure your connection to a VPN service won't get you flagged ?


I'm not paranoid, and I have no evidence of anything at all, but it does occur to me that if I ran the US CIA/FBI, I would totally start a company like NordVPN and give them enough money to advertise widely and get sponsor spots on popular YouTube channels.

Sure, you'll probably end up facilitating a few petabytes of copyright infringing torrent traffic every day, but you'll also have a direct connection to countless people who think they're totally anonymous.


> Sure, you'll probably end up facilitating a few petabytes of copyright infringing torrent traffic every day, but you'll also have a direct connection to countless people who think they're totally anonymous.

This is the best part really. No three letter agency is going to spoil their honeypot in order to bust some kid for downloading movies and video games. They've got bigger fish to fry. The more successful they are at keeping their users safe from the copyright goons the more popular and "trustworthy" their little honeypot will seem. Your downloads will still probably earn you another black mark in your dossier, one more thing they could use against you if you ever became a problem for them, but still, free movies, music, and games! Might as well make oppression work for you.


Most people who use VPNs for privacy will readily admit that their use is based on relative trust. They know that ISPs are notorious for harvesting and granting access to browsing data, and they trust that their reputable VPN provider of choice is truthful about their logging and internal data handling policies.


> You just replaced ISP tracking with VPN tracking.

Yeah, that's the point. People have infinitely more choices for VPN providers than for ISPs. The VPN provider could be in another country. You can even run your own VPN on hosted infra.


> You can even run your own VPN on hosted infra.

This comes with its own risks and you lose some advantages of a commercial service, such as being "lost among the crowd" of other VPN users.


What if you provision a new IP address every other week/month? i.e Assuming the provider allows it and the cost is negligible


If your hosting provider makes similar privacy guarantees to a similar VPN provider, then perhaps. For bonus points, use a hosted VM and commercial VPN in serial.

Again though, if the US Government is a potential adversary, there's still risk here. Intelligence services could have a 0-day on your linux distro, or an operative within the hosting company. If they get access, they can turn your hosted VM against you.


I think you just have to assume there's a possibility that the infrastracture can be compromised at any moment. For both of these two options. Guarantees from providers are valid until the day they are not.

Though a point has to be made about using a hosted VM for VPN purposes, you have some level of control about which specific software, configuration and encryption schemes etc. you want to use for your stack. Only downside is that currently there isn't sufficient hypervisor protection from the host kernel afaik.

However I do agree with you that either way, the risk threshold is too high if you are concerned about high level state actors. Neither commercial VPNs nor hosting providers can solve the "anonymity" problem for you.


> You can even run your own VPN on hosted infra

True. But we know very few people do that


So now you trust the infra hoster AND their own ISP.


You will always have to trust someone. Having more options when choosing who you trust is a good thing. Does that make sense to you?


You can choose your VPN. Your ISP, not so much. It's also possible to run your own VPN on a public cloud or VPS. You might still need to worry about your provider then, especially if a government might request who is renting the server that is sending certain traffic.


There are levels to trustworthiness. I mostly trust my ISP enough to not bother with VPN at home, but when I'm traveling I'll take my VPN provider over a hotel WiFi any day.


> You just replaced ISP tracking with VPN tracking

Huge benefit in my opinion.


I see a lot more ads for VPNs that I do for ISPs.


> There is no point to them.

What about this?

"Under the provisions of the Investigatory Powers (IP) Act, it is now possible for the Law Enforcement Agency (LEA) community to lawfully obtain Internet Connection Records (ICR) in support of their investigations. Following the completion of some initial trial activities, work is now underway to provision a national ICR service."

https://www.digitalmarketplace.service.gov.uk/digital-outcom...


And what makes you trust that the VPN provider wouldn’t share these records if asked?


I’ve heard the idea of “competing jurisdictions” advanced here: run your own Wireguard VM in China or Russia and route traffic through that box, with the goal of essentially introducing unfriendly international borders in your traffic. Ideally your Russian VM provider would be uninterested in responding to an American subpoena.

The other concept I’ve heard put out is to add layers - your traffic hits Vultr, then Hetzner, then etc etc. Conceivably if you cross enough jurisdictions you can make it very difficult for a legal adversary to attack your traffic.

Would I bet on this for activities I needed to remain exceptionally private? No - but if I needed relatively consistently low latency traffic and a decent baseline of privacy it might just work.


In the end you have to trust someone or you might as well just stay offline forever. In the US, very few people have much choice in their ISP. If you are lucky you might have more than two options. Many ISPs are known to be untrustworthy. They outright state they will mine your internet history and sell it. If I have the choice between an ISP who I know will sell my browsing history and a VPN which might, but claims not to? Well, I know which one I'm going to pick.


They also state they can and will shut down your service for copyright infringement.


If ISPs don't shut down people's accounts they'll get sued for billions in fines. There are still open court cases against multiple ISPs for not cutting enough people off from the internet. So far, courts have agreed with the RIAA/MPA who paid good money for the laws we have.


Yes, the media industry is certainly more to blame here. Although, the current ISP oligopoly means that there are less companies to sue or otherwise bully. And it doesn't help that some major ISPs are owned by companies that also produce copyrighted content they want to protect.


Moreover, why do you imagine that the VPN is not being operated by the NSA? THere are really a few different kinds of VPN operators: intelligence services, organized criminals, and legit businesses. You have no way of determining which is which.


Pick one in a suitable jurisdiction with a track record of sending such records and requests for them to the bit bucket?


use VPN provider that does not store these ICR records and is physically unable to fulfill such request


Thing is, if you ping experts in the privacy field (like Mike Bazzell, the former FBI OSINT guy who billionaires and celebs hire to keep their personal info off the internet), they will all say the VPNs are a very important tool to create a layer of privacy between you and the site you are visiting and also a good tool to prevent said sites from easily profiling you. No, they are not a panacea -- much like the security universe, the privacy universe requires lots of other active and passive behavioral and technological changes to properly lock things down to whatever standard you require for your threat model. But they are very much an important tool. For a small subset of people it's literally a tool that protects their life.

So yeah, it's a kinda a big deal if it leaks. (Which is why most privacy experts, were you to tell them you were sufficiently paranoid, would have you fire up a pfSense and link it permanently to a VPN service and then run a separate brand of VPN software on whatever device, so that you have two layers going through two companies.)


> who billionaires and celebs hire to keep their personal info off the internet

I also do this type of work on the side. When it matters a device level VPN is never the correct option, because every OS leaks to some extent. They get a device where the cellular components have been disabled and it can only connect to a fixed wifi AP carried by one of their EP guys that tunnels the traffic back to a datacenter.


In the UK if you download a TV show or other copyrighted material without a VPN, you can get a letter from your ISP as a warning together with a list of the content you've downloaded, and they may terminate your service if you continue.

A VPN solves this, and does protect your privacy, so your comment is just needless hyperbole.


> And the idea that these so called ‘VPN’ services somehow improve your security and privacy on the internet is laughable

I trust Mullvad more than I trust Optimum.


But the point is you shouldn't trust either of them.


You're paying to insert a potential man-in-the-middle attack. That's got to be worth something.


That would be true if we didn't have regional content blocking, location tracking and general surveillance by supposedly democratic states.

I recommend VPN services in regions where legislation of your home country might difficulties getting data.


> This. And the idea that these so called ‘VPN’ services somehow improve your security and privacy on the internet is laughable.

Actually, they do: Neither your ISP nor the government (assuming the VPN provider is in a "hostile" jurisdiction) can intercept, analyze or modify your Internet traffic when you are using a VPN to mask your Internet access. There have been multiple instances of this in the past [1][2] and ongoing (e.g. DNS [3]), and ISP "middleboxes" have been historically the biggest impediment in rolling out new features.

Ubiquitous HTTPS has shut down a lot of that shit, but until DNS-over-HTTPS becomes actual mainstream DNS (and SSL SNI!) will still leak a lot of information to entities that have a direct financial interest in collecting, packaging and selling this data to advertisers - there is a reason why ISPs oppose any legislation that turns them into "dumb pipes" after all.

Security-wise, at least if you are using any kind of untrusted network (e.g. university campus, public hotspots) a decent VPN software that uses the OS-provided firewall to completely drop any incoming and outgoing packets except for the VPN tunnel connection is also a massive benefit.

The downside of course is that you are now forced to trust the VPN provider instead of the ISP - but at least the VPN provider market is healthy and extremely competitive, which means any sort of shady bullshit would be a virtual death sentence, unlike the ISP market where you are in many cases stuck with one or two options.

Not to forget, VPNs also provide privacy on the "other end": as many providers don't cycle through IP addresses sometimes for months, advertising providers can track your movement across the Internet simply by collecting your origin IP. A good VPN provider regularly changes the origin IP visible to sites you access.

[1] https://www.privateinternetaccess.com/blog/comcast-still-use...

[2] https://labs.ripe.net/author/babak_farrokhi/is-your-isp-hija...

[3] https://www.csoonline.com/article/2953718/t-mobile-caught-in...


> Actually, they do: Neither your ISP nor the government (assuming the VPN provider is in a "hostile" jurisdiction) can intercept, analyze or modify your Internet traffic

No, you're just delegating those capabilities to some completely unregulated random actors instead.

> which means any sort of shady bullshit would be a virtual death sentence

This assumes that their shady bullshit is discovered by someone. I would bet good money that the vast majority of it isn't. They could be sampling traffic and selling it to other companies without modifying it and users would never be any the wiser.

Honestly, I wish we could get past this broken narrative that VPNs are a panacea.


> No, you're just delegating those capabilities to some completely unregulated random actors instead.

It's a question of trust in the end. Telco providers not just in the US but across the Western world have shown time and time again that they cannot be trusted: traffic manipulation, DNS hijacking, selling data to the highest bidder [1], engaging in open corruption to prevent competition, predatory sales tactics, fee scams, peering extortion [2], misappropriating government funds [3] - name the act and you'll find a dominant ISP having done or still doing that practice.

VPN providers generally don't have that baggage attached.

> This assumes that their shady bullshit is discovered by someone. I would bet good money that the vast majority of it isn't.

I agree, but at least the incentives are aligned completely different than with ISPs. The large ISPs can do whatever they want, even breaking the law, because their consumers have no other choice - rural ISPs will get competition from Starlink soon enough, but people in condos? They're stuck with whatever the landlord offers, and the landlord won't care even if there is competition as long as the monopoly ISP pays higher kickbacks.

[1] https://www.ftc.gov/news-events/news/press-releases/2021/10/...

[2] https://www.heise.de/hintergrund/Missing-Link-Regulierer-vs-...

[3] https://eu.jsonline.com/in-depth/news/2021/07/14/weve-spent-...


> Telco providers not just in the US but across the Western world have shown time and time again that they cannot be trusted

This is true, although there's still a very large risk element here. The average person is not likely to be able to safely determine which VPN providers are trustworthy or not. They also aren't likely to understand their limitations, i.e. they don't grant you perfect anonymity, they don't grant you perfect immunity, they may or may not capture the traffic that you intended or thought.

In that case, is it really a good thing for VPN internet provider usage to be on the rise? All we're seemingly doing is handing people more guns to potentially shoot themselves in the foot with.

ISPs, for all of their transgressions, tend to be registered and regulated companies and that makes it much easier to at least find someone to target with legal action if needs be. The bar might be low but there are some standards to which they have to adhere to. Many VPN providers are nameless and faceless "organisations" with little-to-no regulation or responsibility. It's difficult to know if they take your privacy seriously, whether they are taking adequate precautions not to log, to not leak data or to adequately secure their systems.

Hell, it's entirely possible that your VPN provider is actually just an FBI honeypot on the lookout for people who are only actually using a VPN service because they have something to hide. How would you tell if they were?

I just don't really buy the argument that having different incentives means they are any less likely to be nefarious or negligent.


> Neither your ISP nor the government (assuming the VPN provider is in a "hostile" jurisdiction) can intercept, analyze or modify your Internet traffic when you are using a VPN to mask your Internet access.

This is exactly the problem with VPNs, they give you a false sense of security. When your traffic goes over the public internet, you should assume everyone and their grandmother can track it. So the traffic cannot be intercepted at your ISP, that only leaves a billion other places where it can be intercepted.

> The downside of course is that you are now forced to trust the VPN provider instead of the ISP

No. The point is neither should be trusted.


> There is no point to them.

There might be no point to their security and privacy, but they are still good for getting foreign-country Netflix.


I'm not sure what your point is here?

To be clear, is what you're saying that it is ok for VPNs to be broken (or at least less bad) because their most popular usage isn't what they were originally intended for?

If that wasn't your point, what was?


i will have agree with your definition, but the issue here isn't with the nomenclature, it's with the bill of goods being sold as a service.

what people expect, and what is being sold, is an encrypted tunnel that all traffic goes through, to an endpoint. That this is called "VPN" is irrelevant.

I have a GL-iNet Mango that i have setup to provide "always on wireguard" to a computer in a datacenter i control the public IP for. I haven't tested, but i expect all data sent to and from any devices connected to that Device's SSID to be tunneled via wireguard to the computer in the DC, and therefore, to all outside observers the DC is where my device is. Obviously the ISP can see the session, but since they have no say over the DC endpoint, they have no way of knowing what the traffic is or where it's going. It could just be me doing SSH or video streaming or backups to and from the datacenter, or i could be watching netflix or youtube.

In that circumstance, an iOS device shouldn't be able to leak my local network's ostensible "public IP", since the actual transport layer is outside of the iOS device's control.

With all of this being said, i don't think there's any way to guarantee that leaks are impossible without literally air-gapping your devices and forcing all traffic through something that cannot communicate with anything but the remote endpoint - that is, if the wireguard connection fails, all pings fail, all TCP/UDP/etc traffic times out, and so on. In this manner, probably all things sold as "secure VPN" or as a service that does that are scams. This is the issue that TFA is complaining about.

in a situation where it's life and death - i would find an open wifi access point and connect a wireless bridge device (e.g. tp link TL-WR802N), with an STP ethernet cable to something similar to the gl-iNET mango, with 100% forced wireguard connectivity. I'd only consider this viable after doing tshark or tcpdump on the server i control log access to, to verify that my (local) MAC address and/or stuff like webrtc or whatever are blocked/dropped.

sorry for the length, but i didn't want to make multiple comments all over the threads.


I don't like how the word "proxy" got replaced with "VPN" either, but I think it was to create a distinction between the per-connection, often single-protocol nature of proxies (HTTP/HTTPS/SOCKS) and something that acts like a whole (virtual) network interface.


Maybe that wasn't the original intent, but it's a designed and supported configuration. Many places don't want a fleet of Windows laptops to become network bridges to their intranet. The default configuration for most VPN setups is to route all traffic. If you want to selectively route traffic, you have to specifically set up a split tunnel.

While you could only route client traffic to an intranet endpoint and prevent access to any external services, that wouldn't be very practical in most deployments so a proxy is added on top. This type of deployment is common and has been used for decades.


'VPN' is just a nickname for secure tunnels. These can be used for many different purposes.


We should start calling them vISP or something.


[flagged]


> You are wrong on this, the private part indicates the privacy it provides not the destination.

I remember using VPNs long before, to my knowledge, people were using them in the way you describe, and I was always under the impression that the "P" in VPN meant "connecting private networks" together over the Internet.

This document from 2001 agrees with me: https://docs.microsoft.com/en-us/previous-versions/windows/i... "From the user’s perspective, the VPN connection is a point-to-point connection between the user’s computer and a corporate server. The nature of the intermediate internetwork is irrelevant to the user because it appears as if the data is being sent over a dedicated private link."


VPNs were used primarily by companies to allow secure access to their network from the outside. It doesn't surprise me that documents aimed at businesses running Windows servers would describe them in the context of that use case. It doesn't mean that was the entire point, or purpose. It's just one thing they were commonly used for.


> VPNs were used primarily by companies to allow secure access to their network from the outside.

Yeah, that was the entire argument.


At the same time that I was working at an ISP on a product that would let employees from various companies connect via dial up and VPN into their corporate networks (whose gateways were also on our network) several of my co-workers were using a VPN of their own to connect to their home networks, but not to access the resources on those home networks (although some did that too). They wanted to use the internet from their machines at home in order to hide their internet activity from our IT department.

This is basically the same thing people do with VPNs now, only instead of hiding their internet activity from IT, they hide it from their ISP.

VPNs were always used for things other than connecting someone to a corporate network, it's just that most of the general internet population at that time (and I'm guessing you're old enough to remember this) were not aware of the technology and not tech savvy enough to set it up. This is true even for the employees of the companies we had as customers. We had to build entire software products that did nothing but hand hold people through setting up a dial up networking connection. It's not surprising that corporations were the majority users of VPN technology until the rest of the public (who don't have IT staff) caught up, at which point it became increasingly more common for people to use it to hide their internet traffic.


I don't doubt that, I was replying to this:

> You are wrong on this, the private part indicates the privacy it provides not the destination.

In your own recollection, what do you think "Virtual Private Network" stood for? Connecting private networks together, or privacy?


I took it to mean a private network connecting two end points, not necessarily connecting two private networks, just that the tunnel (the network between them) was virtual and secure so the traffic exchanged couldn't be eavesdropped on, or modified, by every random node as it passed though the untrusted internet.

I might have been influenced by the product we were selling though. These were dial up users on workstations looking to access their company's LAN so the idea of connecting two discrete private networks wouldn't have fit as well. There was also a lot of focus on the insecurity of passing traffic (even encrypted traffic) over the internet. We had companies paying us a premium to sign up for the service and host their gateway on our network so that the traffic between the users who dialed in and the company's own network never left the ISPs network (never reached the internet at large). I knew at the time it was marketing and that with a well encrypted connection it shouldn't matter if the traffic ever left our "cloud", but it could have helped to shape my view of the technology.

Clients ate that up too. The internet was scary to them. Being able to say that their employee will dial directly into our equipment, and that no packet would pass through a device we didn't operate until the moment it hits your company's gateway made a lot of companies feel better about letting workers remote in.


Fair, it does seem that "privacy" VPNs are a lot older than I thought and possibly as old as the "VPN" moniker. (Assuming that your project was in the 90s, which it sounds like.)

At this point it just seems like arguing for arguing's sake, but I was rejecting the notion that VPNs were always intended for privacy (along with saying others are wrong for suggesting otherwise). It still seems to me that VPNs did not always imply "privacy", and I think in my sibling comment to this, an RFC from 1999 seems to support that (focusing on "intranets" and "extranets" in the definition of a VPN, and only mentioning encryption once as an optional component, with possibly only authentication instead, or even none).


I agree, VPNs were absolutely not always used to hide internet activity, but sometimes they were. Early on they were certainly most often used by companies to connect networks or to connect to resources on their intranet, the need (and the money) was primarily there, but I'm not surprised that using VPNs for privacy reasons got more popular as time went on. Even back then I thought it was pretty cool/useful tech and I had no LAN to speak of.


I agree with all of that, too! And yeah, I think by now the meaning of "VPN" has well shifted, likely because of the privacy enhancement getting to popular (and I think I've also noticed that terms like "Intranet" and "Extranet" have somewhat fallen out of favor, too, but maybe that's just in my environments).


It's not a question of recollection. Read the acronym carefully. It is virtual and it is a network. Not the destination but the tunnel itself is the network that is private. It was described as such from the start and in no networking context have I ever heard otherwise (correct me if wrong please).

Let's say you have an IPSec tunnel between a branch location and HQ site. The typical solution was GRE where you encapsulate it inside another IP packet that has public IPs only for the destination to decapsulate it. When VPNs came along they added privacy hence the name.

In networking you are not connecting two networks. You are interconnecting three networks! the branch would have its own subnet so would HQ but the VPN also would have its subnet all routed as separate networks. The tunnel network getting privacy because it traverses untrusted networks (back in the day it wasn't typically the interent but ATM, frame relay,T1,etc... "directly" between sites), that's where term cloud comes from FYI the untrusted magic ISP network in the sky.


This might be the most pedantic thread (on my part, too) I was ever part of. :)

The earliest reference to VPN I can find in the RFCs, RFC2547, seems to call the "destination" (the network spanned by the tunnel) the VPN, not the tunnel itself:

    "If all the sites in a VPN are owned by the same enterprise, the VPN
    is a corporate "intranet".  If the various sites in a VPN are owned
    by different enterprises, the VPN is an "extranet".  A site can be in
    more than one VPN; e.g., in an intranet and several extranets.  We
    regard both intranets and extranets as VPNs."
That same RFC has only one mention of encryption at all, in passing, and as being optional (note the "and/or"):

    A security-conscious VPN user might want to ensure that some or all
    of the packets which traverse the backbone are authenticated and/or
    encrypted.
It does not seem to me that privacy was implied.

I still think that VPNs were invented to connect smaller private networks to a larger private network together, where private != privacy. (But rather related to authorities, such as using "private IP addresses" in e.g. 10.0.0.0/8, instead of publicly routable ones.)

Privacy was a (good, likely popular) option, but just not part of the strict definition of what a VPN is (much unlike today).


If you show me GRE being called a VPN I will concede, and yeah very pedantic but it is good to learn through discourse as well I think.


Well companies used VPNs that were not site to site, since the first ASAs and probably before then.

And even then it isn't like individuals did not VPN in the 90s at all.


Yes, too bad about the downvotes but let me add on that and ask those who disagree what P stands for in WEP and WPA? lol. Wep is wired equivalent privacy.

A connection that does not provide privacy like a GRE tunnel for example is called a tunnel never a VPN or more and GRE specifically connects networks which are typically private.

You can also have VPN between two ASes on on the internet which are public networks. Wrong is wrong. Give me another argument to shoot down against VPNs lol.

The correct term for both private and non-private network tunnels is an overlay network (includes stuff like 6-in-4).


The P in WPA stands for Protection, the P in WEP stands for Privacy, the P in VPN stands for Private. Private and privacy don't have the same meaning, and one does not imply the other. I've cited an RFC from 1999 (RFC2547) in another reply to you that strongly suggests (to my reading, at least) that privacy was not necessarily implied in the notion of "private" (although of course a VPN could provide privacy). "Private IP addresses" in the form of "not publicly routable", but not necessarily with any privacy-providing encryption in the mix, seem closer in meaning in this case, and were often (but not always) part of it.

The same RFC also pretty clearly calls the network spanned by the tunnel, not the tunnel itself, a VPN.

By now, the meaning has shifted.

That being said, I think this is my last message on the topic, since, well... it's quite a lot of wasted time on pedantry (which is totally my fault).


> Private and privacy don't have the same meaning, and one does not imply the other.

These two words have the same root and etymology. The adjective "private" is transformed into a noun using the abstract noun suffix, "cy," to become "privacy." These two words have the same word root dressed as different parts of speech.


Of course they have the same root and etymology. Of course they do not have the same meaning, look them up in the dictionary. A private parking spot is not a privacy parking spot.


They are strongly associated, and the definitions are related. One can not have a private conversation without privacy, and when seeking privacy it is exclusively to do something in private. Contrary to previous claims, one implies the other.

The argument that privacy doesn't mean private, and vice versa, in regards to the meaning any of the letters in acronyms is specious, such as, the word private in VPN does not mean that you will have privacy, because, in fact, any transfer of data between the VPN nodes will be kept private from the Internet at large, thus the transfer is in privacy.

The fallacy you and OP committed (if you are not OP, didn't check) is known as the appeal to definition.


> One can not have a private conversation without privacy, and when seeking privacy it is exclusively to do something in private.

One can have a private parking spot without privacy, though. Or a private pilot license.

> any transfer of data between the VPN nodes will be kept private from the Internet at large, thus the transfer is in privacy

Not if your VPN is not encrypted, which exists, although it isn't very common anymore, for obvious reasons.

The IP address "10.1.1.1" is still part of IANA's private IP address space, no matter whether it is transported in a way (say using an unecrypted tunnel over the public Internet) that provides privacy or not.


> One can have a private parking spot without privacy, though. Or a private pilot license.

This is equivocating between two distinct and separate definitions of "private." You are mixing these homonyms.

In the sense you're using in the quote I pulled from your comment, it means intended for or restricted to the use of a particular person, group, or class, but in the sense that it is used in Virtual Private Network, it means something else, namely not known or intended to be known publicly; secret.


Settling that question was the entire discussion you butted into. I quoted an RFC from the late 90s that shows what meaning of "private" was originally intended, while you seem to a priori assume what "private" means here.

> in the sense that it is used in Virtual Private Network, it means something else, namely not known or intended to be known publicly; secret.

Great. Present some evidence or stop wasting time. My evidence that it actually does not mean that is RFC2547.


> My evidence that it actually does not mean that is RFC2547.

RFC2547 does not support your claim... anywhere.

I'll use Webster's definitions for my evidence. Also, every explanation of what VPN is everywhere on the Internet speaks of anonymity and privacy. This means the P in VPN could only mean free from public attention, secret and NOT for the use of a particular person or group, as in private parking.


RFC2547 supports my claim that a VPN does not imply privacy. There is only one mention of "encryption", and it's called out as optional:

   A security-conscious VPN user might want to ensure that some or all
   of the packets which traverse the backbone are authenticated and/or
   encrypted.
Webster's definition of a VPN is (https://www.merriam-webster.com/dictionary/VPN):

    a private computer network that functions over a public network
A private network at the time was defined according to RFC1918. It also provides the motivation for private networks/the private IP address space, which was growth, not privacy/secrecy.

It's a happy coincidence that you can sell a "Virtual Private Network" to endusers as a privacy-enhancing machination, given that it already contains the word "Private", even if that was meant as in private parking space, not private conversation.


You've turned me around. I am convinced. Well done.


This reminds me of the people who think public companies (traded on the stock market) are publicly owned (part of the government) and so want the First Amendment to apply to getting banned from Facebook.


> You are wrong on this, the private part indicates the privacy it provides not the destination.

Sorry - that's just not the case, you're retconning VPN terminology. VPNs were originally implemented to replace dedicated WANs and dial-in access to private networks - they were not originally designed to provide privacy for individual access to the Internet. Heck, even the RFC for VPN terminology makes that clear (RFC 2764 is over 20 years old).

Oh, and this: > There have always been client-access VPNs

If your first exposure to VPNs was from shady privacy-snake-oil salesmen, I can see how you'd think this, but take it from the people who were there before client based VPN access was even a thing: Site-to-Site VPN was the original use case for VPN, and you didn't waste tunnel bandwidth (encryption chips were slow and expensive back in the day) routing general internet traffic over your tunnel...


> Heck, even the RFC for VPN terminology makes that clear (RFC 2764 is over 20 years old).

What it actually says is:

The widespread deployment of VPNs has been hampered, however, by the lack of interoperable implementations, which, in turn, derives from the lack of general agreement on the definition and scope of VPNs and confusion over the wide variety of solutions that are all described by the term VPN.

Some people's ideas of the definition and scope of "VPN" might have involved the requirement that it terminate on a private network, but others would not.


No one ever called GRE or 6-in-4 a VPN, why is that!? Let me copy paste my other comment:

Read the acronym carefully. It is virtual and it is a network. Not the destination but the tunnel itself is the network that is private. It was described as such from the start and in no networking context have I ever heard otherwise (correct me if wrong please). Let's say you have an IPSec tunnel between a branch location and HQ site. The typical solution was GRE where you encapsulate it inside another IP packet that has public IPs only for the destination to decapsulate it. When VPNs came along they added privacy hence the name. In networking you are not connecting two networks. You are interconnecting three networks! the branch would have its own subnet so would HQ but the VPN also would have its subnet all routed as separate networks. The tunnel network getting privacy because it traverses untrusted networks (back in the day it wasn't typically the interent but ATM, frame relay,T1,etc... "directly" between sites), that's where term cloud comes from FYI the untrusted magic ISP network in the sky.


> the private part indicates the privacy it provides not the destination.

I wouldn't entirely agree with that (although out of context I do agree).

It's a Virtual Private Network connection, you create a Virtual (not physical) Private Network (between your device and another device/server) there's no real difference between a site-to-site VPN and a client-access VPN other than if the devices at each end route more than just the partner traffic over the private network.

If I connect a "site-to-site VPN" between my computer and your computer, if I add a route to send all traffic for a particular network to your computer as the next hop, that makes it "client-access" for that particular network, if I add a default route it then sends any internet request I make to your device as the next hop.

If your device decides to forward the packets (and probably NAT them) then I now have some privacy for my internet traffic.

VPNs were originally designed to replace dial-up modem connections since the internet was becoming more ubiquitous and it would be far cheaper for someone to connect to their local ISP then use a VPN to connect to their remote network (either personally or usually between sites), than dial directly to their other site (also it was usually far cheaper for one internet connection than a bank of modems and ISDN lines (if you wanted 56.6Kbps or 64Kbps connections)


> It's a Virtual Private Network connection, you create a Virtual (not physical) Private Network (between your device and another device/server) there's no real difference between a site-to-site VPN and a client-access VPN other than if the devices at each end route more than just the partner traffic over the private network.

The difference is one end is not a network but an endpoint part of a network. Multiple client access VPNs can be part of the same subnet.

> If I connect a "site-to-site VPN" between my computer and your computer, if I add a route to send all traffic for a particular network to your computer as the next hop, that makes it "client-access" for that particular network, if I add a default route it then sends any internet request I make to your device as the next hop.

In site-to-site VPN, your computer would need to route a separate site network as would the remote end. With client access only the remote end routes a sparate network. Windows for example cannot be used (unless server versions) to provide site-to-site connectivity because it does not route between NICs. Your tunnel IP is used for connectivity with client access but with site-to-site the remote end expects you to adverise a route or have a separate config for a static route back to some other network on your end which is what it will route (won't work otherwise). Hope that is more clear. You can turn your nix box to a s2s vpn terminator but in every VPN type this requires different config which is why the different terms exist.


You are technically correct that we have a bad naming thing going on here, but I"m reminded of legal principles similar if not identical to "implied warranty of merchantability."

What are VPN services SOLD AS? If they promise something and do not do that thing, then this is a problem. They should be sued or regulated or similar, and they shouldn't be able to get around that, even based on techinical definitions.


Which VPNs? Nord or Azure Express route?


Nord et. al. Yes, they should have been calling themselves "anonymous proxies," but either way, that's the service they promise.


They are not proxies because they don't well... proxy your request as in make a second connection on your behalf which is what proxy means. Those VPN services are the equivalent of moving your ISP gateway to someone's datacenter somewhere . They are internet gateway VPNs.

I see no deception or nothing misleading about the service they provide. The fact that you need to trust them more than your ISP and country's network is not unclear to anyone, they even advertise "no log" because obviously they can log if they choose to.

It is preposterous that you call them snake oil when there are so many discoveries about ISPs doing mitm on their own behalf or governments and even in the US/west it is a default that they will sale your location and address along other things as a service. You have no choice in the matter because they are monopolies. It is a simple risk calculation that if I trusr some rando vpn provider over my last-mile or country/locality network I can pay to use them to move my trust boundary to their servers. Now marketers and other hostile parties can buy my metadata from a myriad of VPN providers who all risk to lose their whole business if that deal was ever exposed (unlike ISPs).


I fully understand everything you're saying and you have very good technical points, but (and I would draw this out to a LOT of things in IT) these aren't taken away by MOST USERS. Most users, who are novices in tech, read the "label" of the VPN and presume that they are covered, when we know they are not.

SOMEONE should be punished for this because it's effectively a lie. This is the only way we will actually improve the bad situation you're talking about.


That's the Apple Way though isn't it? Apple says "Do it exactly how we tell you" and you shouldn't expect anything to work if you deviate from the One True Path that Apple has laid out for you. "Think Different, Do As You're Told"


Oh, come on. This is an utterly lazy argument and it adds absolutely nothing useful to the discussion.


Is Apple responsible for the false advertising of other companies?




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

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

Search: