Generally I would expect in france to be able to get a prepaid SIM for €15, which would then last a year. Is that not an option?

I would still object to paying anything, and then being forced to tend to that number which pinpoints a geographical location. If they can get msg to you by sms or voicemail, that likely satisfies any obligation they have to inform you which then creates an obligation on you to monitor the phone (I suspect).

How so? No blow-ups in the decades I’ve been doing it. People are not obligated to be voice-reachable (at least not by any laws I’ve encountered). Creditors need to send you a bill, sure, but that’s their problem. If they can’t handle fax they better be willing to use snail mail.

What’s blowing up in people’s faces is the culture of sharing a mobile number that then takes the role of identification, which then gets exfiltrated by cyber criminals. The abuse of using mobile numbers as an identifier has spread through Europe and only a small segment of privacy advocates currently realise the problem.

Twitter demanded a mobile number from me. Would not take a fax number. So I walked. Shortly after, Twitter had a data breach that leaked everyone’s mobile numbers. Then Twitter was caught abusing the mobile numbers themselves in ways not allowed in the privacy policy.

Americans are extra fucked because there is no privacy safeguard. The bank shares the number with the credit bureau, who then shares it with all members (banks, insurers, etc) and those who will pay for it.

Not even remotely. Ever heard of Efax? You email the phone number and the Efax company sends the fax.

eFax was bought by j2.com, so indeed i’m aware of it. Efax, Jconnect, j2… all the same ownership.

Fax is being ditched by those who think it is no longer used, regardless of whether they have dedicated equipment or a gateway. It’s the same decision. Either they ditch their fax service (i.e. their fax line is virtual), or they ditch their fax hardware. Or they decide to keep the fax number because they see they have customers who still use fax.

More likely they would call it, get the fax tone and mark it as a wrong number until you contact them.

They can suit themselves… that doesn’t matter to me either way if they decide to alternatively pay postage to reach me. Of course they’re going to be waiting a long time for me to reach them if they don’t signal to me that they want to reach me. If I decide to call them from my non-DID SIP line, the caller ID is set to spoof my fax number, which shows them the number is still correct.

There is no hope for people who choose to have a Google account in the first place. They are a lost cause… they cannot be saved.

I was hoping the article would instruct people who send email to a gmail recipient how they can opt their message out of any kind of Google snooping. I envisioned a header that I could add to my msg to opt out. But no, that article is only for Google boot lickers.

BTW, I always do an MX lookup on email recipients and if I see Google or MS, I don’t email them. But it’s not entirely effective because some recipients mask their email provider by using an email firewall service like baracuda. Hence why I would want a header or something to opt-out in case my email inadvertently traverses Google’s servers.

In some cases, snail mail is just too slow for the situation at hand. Speed is the only thing that makes fax irreplaceable in some cases.

In most cases, fax is much cheaper than postage. So I’m always grateful when I can send a fax… so I don’t burn through stamps so quickly.

I’m glad to hear some libraries offer access to broadcast TV signals. I’ve not encountered it but this will help in convincing my local library to do the same.

They have a fee for just about anything I could request that can be done on the web. But I suppose I could come up with something trivial like asking what my balance is. I like the idea of costing them by using the time of their call center as a penalty for trying to push me onto their website.

I’m generally with you on that… but what leg would I have to stand on? The states love their “abandoned” property law which feeds them, and the brokerage is happy to accommodate, to force people onto their website. I would want to show up to court with some kind of evidence that I did not abandon the property… but I’m not sure what that would be.

I’m familiar with the surveillance capitalist streaming svcs (amazon, netflix, etc) but I did not know about Hoopla and Kanopy. They are described as ad-free, so worth a look. But my quick take is that the websites are a bit dodgy/enshitified. Hoopla needs lots of Google JavaScript and after I enable it the page remains blank. Kanopy blocks Tor while playing dumb (“Sorry. An unexpected error occurred.”) I wonder are those US-only services or can a library member outside the US get access?

My local library indeed has DVDs, blu-ray discs, and PCs. I use the DVDs but that’s not really what I mean by broadcast TV. My local libraries seem to have no way to access local broadcast TV. Maybe it’s possible to go on a hunt to work out which networks have local broadcast, then track down their websites to see if they have liberated the content online, which could be enshitified in many ways with ads injected or be a conduit to a shitty place like Youtube. It’s probably not the best experience.

Broadcast TV “just works”. Broadcast TV does not push CAPTCHAs, try to collect data on you, or reject you for not using some proprietary app. It gives a technological guarantee of avoiding most enshitification that offline people expect to avoid.

I was an early adopter of e-mail and was on the web before it was graphical. But commercialisation has ruined them. I have mostly switched back to postal mail and fax. I have unplugged from home Internet service. For me this was an upgrade. In the same way, I think broadcast TV is a better UX than the enshitified net.

This method of Cloudflare would never be used in a site that takes credit card data, for example. That would violate the PCI rules that protect credit card data.

I took a moment to look briefly into this. PCI is not a legal compliance. It’s contractual. Merchants violate their agreement with visa/mc all the time and it tends to go unenforced.

So the next question is whether using Cloudflare’s gratis service (thus the 1st and last diagram in your post) is PCI compliant. Having read the nerdwallet link and this link:

https://listings.pcisecuritystandards.org/pdfs/pci_fs_data_storage.pdf

letting Cloudflare see card № and CVV code seems to be PCI compliant. If the 1st diagram is in play (which is unlikely), that would be non-compliant. But in most cases there will be a CF→origin tunnel (the last diagram which is incorrectly X’d out). The rules are quite loose. E.g.:

Do ensure that third parties who process your customers’ payment cards comply with PCI DSS, PED and/or PA-DSS as applicable. Have clear access and password protection policies

So 3rd parties are allowed to see the data. Those other standards appear to deal with data at rest not in transit, IIUC. From nerdwallet:

4. Encrypt cardholder data when transmitting it across open, public networks. Among other things, don’t send unprotected account numbers via messaging technology. This includes email, instant messaging, text and chat.

When the tunnel terminates at Cloudflare’s server, the supplier just has to treat CF as a 3rd party who complies with PCI DSS, PED and/or PA-DSS.

In the event of disaster, law is out of the picture and all you have is finger pointing between two sides a slippery sloppy worded private contract. PCI does not seem to have any real unambiguous force in the case of Cloudflare’s most common config.

My astonishing claims?

It makes no technical sense that Cloudflare would refuse to proxy a TLS site, which is implied by comparing your 1st diagram to @joepie91@fedi.slightly.tech’s diagram, the only difference of which is the CF←origin segment. Hence why the claim is astonishing.

I failed to support my argument? I read actual Cloudflare documentation, which your sources apparently didn’t.

Cloudflare is a biased source and they have been caught in lies (ref: 3rd article).

I provided screenshots and links to actual facts of the product.

There are no links in your comment. Just pics. You would not likely be able to find a source that supports the claim the CF←origin segment is necessarily in the clear.

You chose to give 3 links and your first two were bad.

You quoted from the first link so obviously it’s a good link.

If you’re actually trying to say the /content/ is bad, this is what you’ve failed to show. You attempted to criticise @joepie91@fedi.slightly.tech’s article which was 2 links deep. You failed because the viability of the 1st diagram does not obviate the joepie’s more accurate reality (most sites use TLS these days).

If your thesis depended upon the 3rd, you should have lead with that.

Indeed it was a non-intuitive sequence. The links were pasted in a hurry.

As it was, your links presented factually incorrect information and further cited factually incorrect information.

This is what you failed to show. You did not even address the 2nd link; in fact said you did not read it. Your 1st response presented bogus misinfo on your part. The last diagram (@joepie91@fedi.slightly.tech’s) is by far the most common configuration.

I trust you that your thesis is built upon your cited works. Therefore, I reject your thesis because your supporting cited words are flawed with bad analysis and incorrect conclusions.

You only read the article about the walled garden. And you actually agreed with the relevant facts that were there, and ultimately concluded that you have no problem with the circumstances that makes CF a walled garden. Your only dispute with the facts were in fact irrelevant. That is, CF is a walled garden regardless of whether there is TLS in the CF←→origin segment. It’s you who has the facts wrong on that (and failed to support your astonishing claim), but either way it does not matter for the walled-garden thesis or for my thesis.

Your provided the supporting documents, which are wrong, and they themselves are citing incorrect works.

As you said, you did not read the 3rd link, so you haven’t even begun to look at the supporting facts for my thesis. The fact that CF is a walled garden (1st article) barely scratches the surface of Cloudflare’s disalignment with EFF principles. That’s mostly covered in the cited works from the 3rd link that you ignored.

Its not my reference, its @JohnnyCash’s.

@JohnnyCash@sopuli.xyz’s reference to malice was different than yours (coming from an entirely different entity in fact). The “twist” was in your misrepresentation of his reference. Hence why your response was a straw man. At 1st I did not regard your strawman as willful malice because it could have been down to very sloppy speed-reading. But now that you have had a chance to revisit his very simple comment, either you’re attempting intellectual dishonesty at this point or English is not your first language.

The sum total of my post was asking @JohnnyCash to expand on his statement for better clarity so we could discuss it.

It was a loaded question. That is, the question itself makes no sense if you comprehend what JC wrote. I don’t believe JC could have been more clear. There was no ambiguity in his reference to malice.

Clearly you misunderstood what you read. @JohnnyCash@sopuli.xyz’s reference to malice is not as you imply.

His fact is correct and his opinion is well supported by it. Specifically, it’s a fact that Cloudflare requires trust. And when over 30% of the (world-wide) web is concentrated in that single walled garden by a single US corporation, it’s obviously sensible to conclude that a lot of trust is required.

Your reference to malice is a straw man. JC did not say CF was itself malicious (but if he were to, it would be a reasonable claim anyway as CF’s harm to legit traffic is deliberate). You must also trust Cloudflare to be competent and not have serious defects (e.g. Cloudbleed). You must trust their diligence with incident response (accidental or malicious). You trust Cloudflare to not suddenly spontaneously hold a website hostage and demand large sums of money (for example).

Finally, JC’s comment that CF is incompatible with an open Internet is an opinion, but it’s spot on if you understand the difference between walled gardens and open resources.

…continued (due to post size limits)…

I can’t tell if the author is being willfully ignorant or if they simply don’t know how technology works. What this comes down to is where in the chain the decryption occurs, if the traffic is ever re-encrytped (and how), or if the traffic is never decrypted to begin with. The article links to secondary another article to explain the technical aspects. However, not only is this secondary linked incomplete, its presents a false scenario which doesn’t actually exist with Cloudflare, but less technically savvy users may not pick this up.

@joepie91@fedi.slightly.tech is an infosec researcher IIRC. I’m not up to speed on any recent CF changes but certainly what you call fiction was in play in 2016. It also make no sense that that would change.

Do you understand the difference between your 1st diagram and your last? The last config (which you call fictional) is actually more secure than the 1st (which has no CF←→origin TLS). The 1st diagram is the most reckless config.

I’m not a CF user, but I am certain admins have a choice whether to use TLS between their host and CF.

However, the author wrongfully assumes this would be commonly used to pass sensitive information. That’s not the use case for this.

What are you saying a gratis (non-paying) subscriber does?

This would be for a non-sensitive site that would improve privacy

No, it does not “improve” privacy (LOL!) to put Cloudflare in the loop, who proxies over 30% of the world’s web traffic all with centralized access in a country without privacy safeguards. Imagine someone in Europe with two ISPs (home+work) and a few VPNs. Cloudflare has an inescapable aggregated view of their activity on ½ dozen different networks.

Separately, Cloudflares exclusion is an assault on privacy. The loss of privacy inherent in CGNAT and Tor is at the hands of CF.

By using this, your ISP loses the ability to see what you’re reading on this site.

Tor is better for that. CF just fucks up privacy.

This method of Cloudflare would never be used in a site that takes credit card data, for example.

Can you cite a source for this claim? The premium (paying) CF subscribers are a tiny minority.

That would violate the PCI rules that protect credit card data.

Well, that’s interesting for sure. Can you link to something about that? I’ve not heard of those rules, but if it’s illegal (in the US, presumably) to let CF see CC data, rightfully so but seems unlikely. I would like to read about that.

BTW, I will be the judge of what is sensitive. A body of law can cover some obvious categories of sensitive data but that’s a very low bar. Each user can do their own threat model which cannot be prescribed by someone else.

The other configurations are end-to-end encryption. There are two configs which I won’t go into here (one avoids another attack vector for bad actors), but for the purposes of this discussion they behave the same.

This means the web traffic is encrypted at the web users side (using an SSL cert) and that data is passed through Cloudflare without ever being decrypted then sent to the web server serving the content.

It also means Cloudflare’s role of bringing the muscle is useless. CF cannot respond to client requests encrypted by another entity’s cert, so the original server bears the full load, thus defeating the top attraction to CF.

This is the config where you’d have your credit card data, name/address, sensitive information, etc. Cloudflare cannot see the data inside this web session.

This simply isn’t a possible config for Cloudflare. The fact that the secondary article’s author completely leaves out the end-to-end encryption options and presents this false narrative as a short coming of the Cloudflare service makes me think they are being malicious.

Can you explain why adding TLS to the CF←→origin segment in a “Universal/Flexible” config scenario would be impossible? If anything, it should be encouraged. It’s malicious to block that possibility.

Unless it wasn’t clear for my assessment of “oppression 3”, I have no issue with this Cloudflare behavior.

I appreciate you sharing your view that Cloudflare is bad or evil here.

You seem to have also missed the thesis of my post. The thesis is important because without it you’re blind about what the facts and arguments are trying to support. To be clear:

• thesis of my post: CF is starkly wholly inconsistent with EFF’s declared and implied values.
• 1st link thesis: CF is a walled garden
• 2nd link thesis: CF’s walled garden is more disempowering than Google or Facebook
• 3rd link thesis: CF is evil. It’s a general smearing with copious dirt on CF’s harm to: privacy, software freedom, netneutrality, vulnerable people, human rights, democracy, censorship, environment, innovation, and their history of poor character and integrity.

Without seeing the 3rd link, you mentally substituted a “CF is evil” thesis when reading my post and when reading the 1st link. So your analysis misses the purposes. I.e. you basically replied to “CF is a walled garden” with “CF is not evil”, and replied to my “CF is not aligned with EFF’s public values” post with “CF is not evil”.

I disagree and hope that some of what I have posted has cleared up some misconceptions and falsehoods being presented as fact that will allow you to make your choice and form a more informed opinion.

Getting the facts right is the most important thing you can do. Opinions, meh, they are useful only to the extent that they put accurate facts into context. But the facts you present are dodgy. Joepie is more convincing. What he says makes sense. And it also concurs with others who have exposed the same problem as Joepie (he was not the 1st). Though you’ve seeded something that could be useful/insightful with the PCI rules.

It makes absolutely no sense that CF’s flexible config would refuse to proxy a TLS-only origin. There is a how-to doc covering how to Cloudflare proxy someone else’s website. I’m not going to dig for the link but that how-to would be fake news if your claim were true (that joepie’s diagram were bogus).

It’s really a tough sell to claim the e2ee configs are common enough to be noteworthy when that config dumps the gratis performance gains that bring CF patrons.

It was interesting to discover that I can see your pics. Lemmy.world is a Cloudflare site (last I checked). Pics are not cached or mirrored, so when pics are uploaded to a CF’d Lemmy node, everyone outside of Cloudflare’s walled garden just see broken links to unreachable images. Yes, CF breaks the fedi. So either LW ditched CF, or LW finally figured out how to whitelist Tor.

(oppression 1)… As a consequence, money-saving shortcuts are taken and Cloudflare uses a cheap blocking criteria based crudely on IP reputation.

Entire subnets or national TLDs are blocked because they come from place or nations that do little to stop bad actors from doing bad acts.

Arbitrary collective punishment has to be seen as arcane and barbaric by 2025, no? I can’t wait until we make enough social progress to collectively see it as zombie-minded as racism.

How many times do you have to get port scans or malware introduction attempts from these subnets,

I was unaware that Cloudflare blocks whole nations. That’s even sloppier than I was aware of. Can you give more details? Which countries? Cloudflare is not transparent about the demographics they exclude.

especially when you have few to zero legitimate users,

People travel. It’s extremely rare that a web admin can block a nation with an expectation of zero collateral damage. The possibility of Cloudflare knowing the web admin’s business is even less likely.

It’s mind-boggling how foolish admins are when they block countries or continents on the basis that residents have no business on their site. So when I travel overseas, there are some affairs I cannot manage in my homeland because of this stupidity.

before the better action is to block the who subnet.

“Better” is a slippery word. If a preemptive DoS attack on legit users is acceptable, you might like to endorse SpamHaus as well. The whole point to fighting spam is to protect the availability of legit traffic. When you directly attack legit traffic under the pretext of anti-spam, you’ve become an obstacle to your own purpose.

As someone that maintains servers, the constant threat and time consumed trying to protect against these is immense.

Pawning your own users to Cloudflare just shifts security problems onto others. You shift a new security problem onto all your users to escape the burden that was rightfully yours. And if you’re like all other CF sites, you also conceal CF’s role and consequences from the users.

Its simply unreasonable to place the burden on server administrators to continuously put their servers in harm’s way

There is no dichotomy of “harm’s way” some magical network that is outside of “harm’s way”. All connected servers are in harm’s way.

It’s simply unreasonable for an unmotivated admin to compromise the security of their users (who lack infosec expertise) in order to have an easier job securing the server.

simply to conform to an ideal when there may even be zero users coming from these places you’re interested in serving.

This place of zero legit users you mention – where is it? It’s certainly not the Tor network. It’s certainly not the CGNAT networks.

I have no issue with this Cloudflare behavior.

Try not to lose sight of the thesis. That behavior is part of what makes CF a walled garden. You may have no issue with walled gardens, but then what would the point be in reading the article?

(oppression 2)… When a website administrator joins the cage by opting to reverse proxy their services via Cloudflare’s walled garden, the visitors of the website have no choice in this decision. The end user is forced into a disempowered take-it-or-leave-it proposition and thus trapped to an essentially absolute extent.

This idea suggests that the mitigation should be the web user should have more power/choice over the web server owner that the owner themselves.

That’s a false conflict. It’s not a competition. A server owner has an independent choice whether to trap their users in a walled garden. Choosing the open-free-world does not elevate the users’ power above the owner. What a bizarre notion. Server owners also have the choice whether to give users choice. E.g. freedom-respecting admins offer onion access as a clearnet alternative, like the privacy international website.

That’s a bizarre notion to me.

What’s bizarre is the idea of competitively comparing admin autonomy to user autonomy. They can (and should) both have autonomy, self-determination, and free choice. How do you make that leap from not trapping users to users have more power than the owner?

A random web user is not automatically entitled to more than what the web server owner is willing to give.

“Entitled” is a slippery word and also awkward in this context. Entitlement can be legal or moral, neither of which is implied by what you quoted. The article covers the meaning of a walled garden, not who is entitled to what.

Though orthogonal to the article, it can still be an interesting discussion. Consider that people are entitled to vote in general elections. Several US states have put online voter registration inside Cloudflare’s walled garden naively¹ using Cloudflare’s default config.

The analysis can get quite complex and messy. Even though /everyone/ is entitled to vote, only demographics of people who Cloudflare Inc. grants access have the privilege of registering online because the website owner is “unwilling²” to serve all those who are entitled to vote. You could say registering is an entitlement but not necessarily online reg, which is fair enough only if there are no eligible voters excluded by that. Not sure that’s a safe stance when all kinds of handicaps and situations might emerge where someone has web access but cannot obtain or complete a paper form. Paper forms are also a problem because of Cloudflare. I do not vote. Kamala lost my vote because even though I can do a paper registration, the data entry worker will still supply the sensitive form data to CF, who I distrust. IOW, trusting Cloudflare has become a pre-condition to voter reg.

¹ I say “naively” under the assumption that the SoS is impartial. Of course if the SoS is republican-leaning, voter suppression serves their party well. ² Unwilling, or in many cases is simply naive about excluded demographics.

I have no issue with this Cloudflare behavior.

Another walled garden feature you are happy with.

(oppression 3) Opacity— to keep people uninformed

The excluded group is wanting more than the web server is willing to give (for whatever reasons).

Of course. This is inherent in being denied access. If the excluded group did not want access, they would not even make the attempt to know they were being excluded. There would be no discussion to be had.

This is the same complaint that the web user should be prioritized of the web server owner. I reject this notion.

It’s not. When an oppressive resource controller marginalises a demographic of people, it is bizarre to frame that scenario as owners vs. users having “priority” over each other. It’s not a competition.

There are lousy owners and admins and there are competent ones. The most competent are skilled at separating spam from ham and not sabotaging copious ham to trash some spam. Fewer legit users are denied service when a competent admin is at the helm and it’s not because the users have more “priority” than the ownership. It’s because the ownership (and who they hire) are more skilled. They are also wise enough to measure detriment to ham (as opposed to the naive measure of just measuring the spam while neglecting collateral damage).

Would you mind saying if you are politically right of center? I’m curious because some recent research found that conservatives have a tendency to view the world as a zero-sum game; that if someone is gaining something then someone else must be losing. It explains xenophobia to some extent (for example) because if immigrants get a better life then it must come at the expense of someone else (per their zero-sum lens). Your tendency to think in terms of a priority between users and owners s.t. when users benefit the owner must be at a loss is analogous to this way of thinking.

And prioritized by WHO? The prioritization comment neglects that every stakeholder has the priviledge to rank for themselves what matters to them personally. Of course from the users’ perspective it’s satisfaction of user needs that matters most. The ownership’s needs only matters to the extent that users needs are served as a consequence. It’s naturally and inherently secondary. And inversely so for the ownership.

Your advocacy for prioritizing ownership above users in line with the enshitification trend that has downgraded all tech we’ve used over the past ~15 years.

Pre-gen-z, suppliers were rightfully expected to serve consumers. That has gotten adversely inverted. So now consumers have been made subservient to suppliers – and they are conforming. It’s fucking shit up. A bathroom remodeling company has an appointment/contact page with CAPTCHA. So customers must dance for the supplier to solve shitty puzzles prior to having the privilege of spending thousands on a new bathroom. I walked, because I don’t bend over backwards to do service for suppliers while feeding those I boycott (Google). Service is their job. My job is to pay them.

In reality the padlock only indicates a secure line to Cloudflare, who sees everything including usernames and unhashed passwords.

The article presents this as objectively true, when in fact its only true in some most cases.

Fixed that for you. It would not make sense for the author to complicate an article about what a walled garden is with rare unverifiable³ corner cases.

³ It’s technologically impossible for web users to prove whether Cloudflare or the server ownership holds the private key associated to the public key that the user’s browser gets from CF. But if you understand business and capitalism, you know the CF e2ee is a rare scenario.

Are you asking how the kill switch is triggered? It’s an autonomous algorithmic kill switch not a remote kill switch. When the machine detects a fault it switches to a broken mode. The switch can only be reset by someone who the manufacturer trusted with the reset procedure (ie. their own repairers who charge more than the machine is worth just to show up).

At that age (27 yrs), I suppose your machines don’t have kill switches. Your next machine will have a kill switch, so even though you can fix it mechanically, the control board will deny you the privilege to start a wash program.

Not to defend garbage business practices, but hand washing REALLY sucks though.

That’s exactly why they get away with it. People’s intolerance for inconvenience is directly proportional to the level of enshitification suppliers can get away with.

I have been washing my clothes by hand for a year now to ensure that I am on the right side of the curve. I wash my much clothes with much less frequency now and do more airing out.

This is a systemic problem and the solution to systemic problems is legislation, not personal responsibility.

We don’t live in the kind of reality where your proposal works. The jurisdiction where legislation is the most viable on the world stage would be Europe. Europe decided against it. “Ecodesign” and right to repair are a shit-show after a 10-year attempt. Have a look at this thread:

https://discuss.tchncs.de/post/46422830

It’s just like the climate problem. You cannot sit back and expect the state to fix it. Hence the existence of Extinction Rebellion. The problem needs both state action and people taking personal responsibility.

Europe has gone as far as to make consumers immune to prosecution for reverse engineering their own property (IIRC). But that’s as far as they go. So effectively, the Polish train hacking approach is allowed but Europe is helpless as far as obligating suppliers to share repair info with amateur repairers (only pros).

People outside of Europe are fucked even more.

I don’t really have time to write a book here and now, but I’ll start with some articles:

• https://thefreeworld.noblogs.org/post/2024/03/18/cloudflare-has-created-the-largest-most-rigidly-exclusive-walled-garden-in-the-world/
• https://thefreeworld.noblogs.org/post/2024/03/20/comparison-of-the-human-disempowerment-severity-of-3-walled-gardens-facebook-google-and-cloudflare/
• https://gitlab.hax0rbana.org/public-repos/liberethos_paradigm/-/blob/master/rap_sheets/cloudflare.md?ref_type=heads