Talk:Proof of stake

Explanation of the improvement proposals

The aforementioned additional information is important to keep the article unbiased. Now it presents the attacks without their proposed defences presenting PoS as an experimental insecure technology, while many scientists in the field consider some PoS implementations more secure than PoW. This bias may contribute to the climate change as well, if you take into account the impact of Wikipedia!

If the proposed changes find their way into the article, I will come back to describe more possible attacks on PoS, and why not, describing also the attacks on PoW algorithms (in the relevant article) which are also important, like double spend attacks and selfish mining. Touftoufikas (talk) 02:13, 27 February 2023 (UTC)

I see no bias here. The article states a simple fact that using the PoS opens many new opportunities for adversaries to exploit; this is a trivial consequence of the fact that for PoS regenerating a big piece of blockchain is near-free in computational terms. The list of these new attacks is taken from papers published in peer-reviewd journals, written by researchers who do not have a stake (in any sense) in the projects they discuss. Yes, the known attacks can be mitigated - and have been mitigated in every successful PoS scheme. Yes, the descriptions of these mitigation techniques (based on similar quality publications) are definitely worth presenting here. Викидим (talk) 05:19, 28 February 2023 (UTC)

Wikipedia articles on blockchain topics (and sources about blockchain in general) tend to suffer from a recurring problem. Remember that Wikipedia's mission is (in part) to document and explain major points of view. In practice, this means it is meant to be read and at least partly understood by laypeople. It also means critical content is covered proportionately and according to due weight, instead of being shoved into the corner of a WP:CSECTION talking about what nebulous unnamed "critics have argued".

So the article has problems, and adding this new information will not, by itself, solve these problem.

By all means, expand the article with reliable sources if those sources represent major points of view, but avoid furnishing the attic before the basement has been dug. The best way to explain any technical details, such as how these attacks can be mitigated etc. is to place this information in context. Right now, the article lacks this fundamental context, so this explanation is going to be much, much harder than it needs to be. Grayfell (talk) 11:00, 28 February 2023 (UTC)

Indeed. It is harder work than I thought to have both well documented sources and good explainability in so technical stuff. I admit that my proposals have not the intended quality to be in the article without further edit abd review. Partially this is because I was expecting to see the reactions and feedback before doing this work. By the feedback I understand that there is a potential if I improve the proposals, and now I am working on it. As for the bias, I don't state it is intentional. The information provided is true. But if you reference on attacks on a scheme, without referencing their mitigations, someone who has no relevant knowledge can think that the whole PoS scheme is something insecure. It is all about the impression. And there is a big debate about PoS vs PoW in terms on security. Non technical audience could be affected by unbalanced impressions created by the most reliable knowledge base on earth. And you know the consequences, a single article could affect the adoption of PoW rather than PoS and this adoption contributes in increased energy consumption. Anyway, I don't blame anyone about this, and I understand that the best way to correct this is by doing better work! Thank everyone for your feedback! I also want to clarify that my criticism does not mean that I don't appreciate the work done here. I intend to improve things. I didn't propose to delete something just to extend it! Touftoufikas (talk) 14:41, 28 February 2023 (UTC)

This is not as hard as it sounds: stick to articles in journals published by major bodies like IEEE and ACM and books by major science publishers, like Springer. Avoid works by the authors of the schemes, as they naturally tend to be overly optimistic. Викидим (talk) 02:02, 1 March 2023 (UTC)

improvement proposal v2

I worked on providing suitable sources, I truncated all parts of the text I couldn’t find good sources for, and I added some explanatory for clarification and better understanding for the readers. Please check the final proposal once again.

-

Add a sentence in the abstract, to follow where it says that it is chosen due to its energy efficiency:

Although it is generally considered to be more centralized and less secure, some argue that with the latest advances, PoS has become more safe and more decentralized than PoW, while also offering greater scalability.

-

In the end of the description section, or in the variants section, add the following text:

Byzantine Fault Tolerant (BFT) blockchains require a certain number of nodes, typically more than 2/3 of the total nodes in the network, to agree on a particular block before it is added to the blockchain. This ensures that the network can tolerate a certain number of malicious actors, up to 1/3 of the total nodes in the network, without compromising the network’s security.

The reason for requiring 2/3 or more nodes to agree on a particular block in the BFT consensus mechanism is to ensure that there can only be one block at a given height in the chain. When more than 2/3 of the nodes agree on a particular block, it becomes mathematically impossible for another block at the same height to be signed by another 2/3 or more nodes, unless more than 1/3 of the nodes in the network are faulty and can cooperate with the 1/3 that did not sign the first block to create a second 2/3 block. Thus, BFT chains can reach consensus on the validity of transactions and add them to the blockchain without the risk of forks or double-spending attacks, as long as less than 1/3 of the nodes are faulty.

BFT chains have the advantage of achieving consensus quickly, often in just a few seconds or less, compared to other consensus mechanisms such as longest chain protocols, based on PoW or PoS, which can take minutes or even hours. This fast consensus time allows BFT chains to process transactions more quickly, making them suitable for applications that require low latency. BFT chains are commonly used in permissioned blockchain networks, where participants are known and trusted, but can also be implemented in permissionless blockchain networks.

One of the main challenges with Byzantine Fault Tolerant (BFT) blockchains is ensuring availability, or the ability to continuously make progress and add new blocks to the blockchain. Since BFT consensus requires a certain number of nodes to agree on a particular block before it can be added to the chain, it is possible for the network to become deadlocked if not enough nodes are able to participate in the consensus process. This can occur due to a variety of factors, such as network connectivity issues, node failure, or malicious behavior.

Chain-based approaches for consensus are also used in blockchain systems, relying on proof-of-work or proof-of-stake consensus algorithms to secure the network. In proof-of-work, miners solve cryptographic puzzles to add new blocks to the chain, while in proof-of-stake, validators are chosen based on the amount of cryptocurrency they hold and stake as collateral. These consensus algorithms rely on the concept that the longest chain in the blockchain is the valid chain. Validators compete to create the longest chain, and the longest chain is considered the valid one, while shorter chains are discarded. To incentivize honest behavior, validators must invest their own cryptocurrency as collateral, which they may lose if they are found to be acting maliciously.

Both BFT-based and Chain-based approaches have their own advantages and disadvantages, and each design has been implemented in various blockchain projects. . While BFT-based approaches provide a high degree of security and are well-suited for permissioned blockchains, they can be more complex to implement in permissionless blockchains and require a high degree of coordination among network participants.

Chain-based approaches, on the other hand, are more simple to implement to provide guaranteed liveness for permissionless blockchains, but may not provide the maximum safety.

-

In the end of the long range attacks section, remove the incomplete duplicate about bribery attacks, and add the following text:

One possible solution to mitigate long-range attacks in proof-of-stake systems is through the use of checkpoints. Checkpoints are periodically created as blocks that include a hash of the blockchain up to a certain point. This allows users to verify the state of the blockchain at specific checkpoints and reject any chains that do not match the checkpoint. This checkpoint serves as a reference point for other nodes on the network and ensures that they are all working from the same blockchain history. Consequently long-range attacks are particularly powerful against two types of users: newcomers and disconnected users. Newcomers to the network may not have access to the full blockchain history and may not be able to distinguish between a valid chain and another chain. Similarly, disconnected users who have not been online for an extended period may not have the full blockchain history and may be vulnerable to long-range attacks.

-

However, checkpoints introduce the notion of subjectivity into the system. Different nodes may have different opinions about which chain is valid based on their different blockchain histories. This is known as a subjective view. Additionally, checkpoints can introduce centralization and security trade-offs, as it requires users to trust the checkpoint authority.

In contrast, in the case of PoW, nodes normally choose to follow the chain with the most accumulated work, which is usually the longest chain. However, there are situations where the longest chain is not the one that is considered valid by the network. For example, in the case of the Bitcoin Cash hard fork, the chain split into two, and nodes had to choose which chain to follow. 8 Similarly, in the case of the Ethereum hard fork after the DAO attack, when the network was still based on PoW consensus, the community decided to fork the chain to revert the theft, while others chose to continue on the original chain.9

Weak subjectivity refers to the idea that there is no objective way to determine the true state of the blockchain in the long term, and users must rely on social consensus and subjective judgment to determine which chain to follow. Over time, social consensus tends to converge on a single chain, reducing the risk of conflicting views.

A proposed mechanism to achieve long term objectivity on PoS chains is Cardano’s key evolving scheme in Ouroboros Genesis. By using the key-evolving scheme, Cardano ensures that even if an attacker manages to gain control of a signing key, they will not be able to use it to sign blocks outside of the current epoch. This makes long-range attacks much more difficult to execute, as the attacker would need to continuously control a majority of the stake throughout multiple epochs to carry out an attack.

-

In the end of the Bribery attacks section add the following text:

Bribery attacks are a type of attack that relies on bribing validators or miners to work on specific blocks or forks, with the goal of presenting arbitrary transactions as valid. Bribery attacks are feasible in both Proof-of-Work (PoW) and Proof-of-Stake (PoS) systems, but they can be more effective in PoS due to the lower resource requirements and the possibility of expanding the attack to the “nothing at stake” problem.

In PoS, bribery attacks can be mitigated by enforcing a slashing condition, where validators are penalized for behaving maliciously, or by releasing violators from their position.

References

— Preceding unsigned comment added by Touftoufikas (talk • contribs) 04:08, 2 March 2023 (UTC)

Again, per above, I see several problems

As before, blogs are not reliable. Medium posts are blog posts. While blog posts can be cited on Wikipedia in some very limited cases, this really doesn't appear to be one of those cases. The author of that blog post is neither unambiguously a reliable expert, nor notable enough to be cited for opinion. Please review WP:UGC and WP:SPS.

A second, but related problem, is that the Guardian source doesn't mention "proof of stake" at all, and doesn't even mention "proof of work". Since this source is a separate, unrelated topic, its use here is WP:SYNTH. This is not the correct approach.

So do you understand why this is a problem?

As I have said before, you need to look at what reliable, independent sources are saying about this topic and then summarize those sources. Those sources will tell you what is and is not a useful example for explaining the topic.

Grayfell (talk) 02:47, 3 March 2023 (UTC)

Yes of course I understand, after reading WP:SYNTH. Consequently this paragraph should be removed, since only Ethereum-related blogs and sites discuss about weak subjectivity in a generalized context. A tried to reproduce this explanation relying on sources considered reliable, as guardian, but after reading WP:SYNTH I understood that this is not compatible with Wikipedia rules. How about the rest of the text? Both problems you referenced were about a single paragraph. If the text from "In contrast..." to "...original chain [9]" is removed, along with these 2 references (8, 9), do you see any problem elsewhere? Touftoufikas (talk) 15:46, 3 March 2023 (UTC)

I've taken the liberty of reformatting your proposal to be closer to Wikipedia's style, which will make it much easier for editors to evaluate. Please review and make sure I did not introduce any errors, or revert if you wish (per WP:TPG).

Wikipedia:Citing sources may be helpful in the future. Wikipedia is old software (which especially shows on talk pages) and what is most comfortable for experienced editors may not be the most intuitive or obvious. Strictly from a formatting perspective, the changes I made were not necessarily sufficient for the article itself, but they are closer to what would be expected. This issue separate from the content of the proposed edits. Grayfell (talk) 23:18, 3 March 2023 (UTC)

As for the content of your proposal, I have several concerns.

One issue is the use of subtle editorializing language. As just one example, who, precisely, is saying that "ensuring liveness" is "one of the main challenges" facing Byzantine Fault Tolerant? It doesn't appear to be directly supported by the attached source. This kind of thing is why I stress the importance of looking at what sources are saying and summarizing, instead of adding your own understanding and adding sources which superficially support your own WP:OR. This project isn't the right place to add that kind of thing.

As another of several possible examples, which source is saying that "chain-based approaches for consensus are commonly used in blockchain systems"? The citation is long enough that it might be buried in there somewhere, but it doesn't appear to be the main point that it's making, especially not regarding PoW, which is barely mentioned. That's a red flag all by itself: The entire paragraph is quite long, supported by a single source which barely mentions "proof-of-work" at all.

As just one more example of many, the lone source for the paragraph on "checkpointing" doesn't use the term checkpointing. If sources are not using this term, why are you?

To revisit what I said previously, our goal is to summarize the main points of this topic. Generally, main points are not going to be buried in the middle of long and obscure journal articles. Look at sources which are about "proof-of-stake". Look at the introduction and conclusion of those articles. Is a main point of a source going to be confusing to readers? If so, it's okay to dig into the body to see how the source explains it. If it doesn't explain it well enough, it's sometimes okay to find other sources to clarify this, but the focus should still be on those main points. If some piece of information isn't presented by the source as a main point, and it doesn't clarify specific confusion about a main point, it's probably not going to belong in a Wikipedia article.

Grayfell (talk) 23:51, 3 March 2023 (UTC)

Thanx for reviewing and editing. I provided more references now to the long paragraphs and I deleted some informative sentences that are not directly supported by the sources. Some references were wrong, I assume by my mistake. I corrected them. Touftoufikas (talk) 16:19, 4 March 2023 (UTC)

I want to explain some things also. 1. Liveness and availability are synonymous terms. I substituted the word in my text with the term provided by the source, to make the connection with the source clear, after reading your comcerns. 2. You ask where in the source checkpointing is referenced. Search for the following text in the source: B. Moving Checkpoints Moving checkpoints or simply checkpoints is a mitigation technique used by almost all PoS protocols. Its simplicity and ease in implementation made it one of the first mitigation techniques to be enforced in PoS powered blockchains, after of course the longest-chain rule.

Touftoufikas (talk) 16:23, 4 March 2023 (UTC)

In generic computing source liveliness and availability are not synonymous. For example, while an application is busy processing a long job, it is alive, but unavailable. Typical terminology is availability = liveliness + readiness. Викидим (talk) 20:06, 4 March 2023 (UTC)

We are talking about blockchain consensus here. It is different in the sense that when a blockchain is busy (for example when it reaches the maximum transaction capacity), this does not affect its consensus algorithm iteration. So we often read one of these two terms in blockchain-related literature, and they almost refer to the same thing: the ability of a consensus algorithm to continue producing blocks, or that there are available validators to continue producing and validating blocks. It is the same thing from another perspective. Indeed, by the way, mathematically they are different concepts. I am talking in the context of blockchain and about the cap theorem availability vs finality or in other terms liveness vs safety. Touftoufikas (talk) 21:40, 4 March 2023 (UTC)

Again, our goal is to explain this to readers who don't already know all this. Your interpretation of these sources may be correct, but it's still WP:OR. Summarize what sources actually say, not what they imply.

Relatedly, to quickly clarify one of my points, "checkpoint" is a relatively common word used multiple times by the source. "Checkpointing" is not used by the source, and is rare. This is an encyclopedia, so word usage matters a great deal. It's a subtle thing, but introducing an uncommon word in an idiosyncratic way will only add to the jargon problem. Grayfell (talk) 22:22, 4 March 2023 (UTC)

• To echo comments of Grayfell, we need to find this content in high quality sources before we consider inclusion. Jtbobwaysf (talk) 02:52, 5 March 2023 (UTC)

1. I substituted the term checkpointing with the term checpoint(s), which is used in the sources. 2. The only source which is WP:OR is the paper of Ouroboros which is auxiliary. The rest of the papers are reviewing papers. 3. About sources quality. The most "controversial" topics are covered by 2 papers: 6 and 1. 6 was chosen as a source in the original article also. 1 is signed by IEEE members. I think this is more reliable than anything. And it is not primary work, it is an evaluation of others' work. Anyway, thank you for your attention, I don't know if I can improve to meet your quality criteria. I quit the effort here for this article because I have the sense that I created more issues than those I tried to solve. I hope someone more experienced than me to continue this work and finally complete the article. Touftoufikas (talk) 08:59, 5 March 2023 (UTC)

Extended-confirmed-protected edit request on 15 September 2023

Examples of Proof of Stake crypto currencies are Tezos, Cardano, Solana, and Algorand and Ethereum 2.0

References: - https://www.mckinsey.com/featured-insights/mckinsey-explainers/what-is-proof-of-stake - https://www.forbes.com/advisor/investing/cryptocurrency/proof-of-stake/ - https://www.fool.com/terms/p/proof-of-stake/ Hiraditya (talk) 14:05, 15 September 2023 (UTC)

 Not done Unclear request, format requests in a "change X to Y" manner. WelpThatWorked (talk) 14:32, 16 September 2023 (UTC)

Edit request: Addition of staking explanation to "Description" section

Request: I propose adding a brief explanation of staking to the Description section.

Proposed text:

In Proof of Stake blockchains, participants can engage in staking, where they lock up a portion of their cryptocurrency holdings to support the network's security and operations. In return, these participants, often referred to as validators, receive rewards, typically in the form of additional tokens. This process not only incentivizes honest participation but also discourages malicious behavior through economic penalties, such as slashing.

Justification:

• It seems like the term "staking" is commonly used in discussions about Proof of Stake but is not explicitly explained in the article.
• Not being an expert in the field (and somewhat COI-biased), I'd prefer for another pair of eyes on this. There might be better sources as well, feel free to change.

Thanks! /Urbourbo (talk) 17:25, 8 February 2025 (UTC) Urbourbo (talk) 17:25, 8 February 2025 (UTC)

Done. I have added the text to the article with a minor edit. --MtPenguinMonster (talk) 12:01, 10 March 2025 (UTC)

Sorry, I have reverted this change. Consensys.io is not a reliable source. If you find a reliable source, summarize what that source says, instead. Grayfell (talk) 06:11, 18 March 2025 (UTC)

Hmm good point. How about using CoinDesk instead? Say, In Proof of Stake blockchains, participants can engage in staking, where they lock up a portion of their cryptocurrency holdings to support the network's security and operations. These participants, often referred to as validators, may operate individually or as part of staking pools. By joining staking pools, even those without substantial holdings can participate in the staking process. In return, validators receive rewards, typically in the form of additional tokens, which are distributed among pool members. This process not only incentivizes honest participation but also discourages malicious behavior through economic penalties, such as slashing.

— Preceding unsigned comment added by Urbourbo (talk • contribs) 14:56, 18 March 2025 (UTC)

Coindesk is not reliable either, per WP:COINDESK. To put it bluntly, cryptocurrency websites are dominated by undisclosed conflict-of-interest editing, churnalism, AI slop, and several other issues. This makes the topic difficult to cover to Wikipedia's standards. Even superficially academic sources, such as journals, have more than their fair share of problems. If a detail is not covered by a reliable source (in simple terms this means a source with a positive reputation for accuracy and fact-checking outside of the pro-crypto bubble) the detail probably doesn't belong in a Wikipedia article. Grayfell (talk) 21:03, 18 March 2025 (UTC)

Thanks, how about Britannica then? [1] /Urbourbo (talk) 22:40, 19 March 2025 (UTC)

Realising now that staking is also covered in one of the article's existing main sources: /Urbourbo (talk) 12:26, 20 March 2025 (UTC) Urbourbo (talk) 12:26, 20 March 2025 (UTC)

References

"Security" section needs to be updated

There is now considerable evidence that Proof of Stake is much more secure than Proof of Work. PoS has already been thoroughly battletested.

While numerous Proof of Work networks have been successfully 51% attacked and reorged (including Bitcoin Cash in 2019, Bitcoin Gold in 2018, Bitcoin SV throughout 2021 and 2022, Vertcoin in 2018 and 2019, Expanse in 2019, and Litecoin Cash in 2019), there have been no instances of a successful attack on a Proof of Stake network.

Proof of Stake offers symmetric economic security where the people providing security to the network have a strong economic incentive to protect it because they are also stakeholders of the network. In contrast, Proof of Work miners can often attack the network without taking any significant damage due to its asymmetric security.

In Feb 2024, even though Bitcoin's marketcap was 5x greater than Ether's marketcap, it was estimated that it would cost over $34B to Ethereum while only costing $5B-20B to attack Bitcoin.[2]https://unchainedcrypto.com/51-attacks-on-bitcoin-and-ethereum-impossible-coinmetrics/ In addition, attempting to buy up stake to attack the network would further drive up the price of the security token, making it even harder to attack the network. HSukePup (talk) 06:06, 18 March 2025 (UTC)

"Unchainedcrypto.com" is not a reliable source. Please cite and summarize a reliable, independent source. Please also keep in mind that Wikipedia doesn't publish original research. Thanks. Grayfell (talk) 06:13, 18 March 2025 (UTC)

I'm at a dilemma. The problem is that the current "Security" section references a quote from Bloomberg article where the source is an investment manager and has no other credentials. There's no research backing it; it's pure speculation.

Here's the full quote:

Critics say these alternatives may be less secure than proof of work. The computer code underpinning proof of stake is so complex that there’s a greater risk of undetected software bugs, says Chris Bendiksen, a researcher at digital-asset investment manager CoinShares. And such systems are more susceptible to censorship, because once an entity or group acquires more than half the tokens, “there’s no way for them to be unseated as the controlling entity,” he says. “Bugs and vulnerabilities in system-critical infrastructure like monetary systems can be catastrophic.”

What can we do about this this section?

Unfortunately, it's very hard to find mainstream news about this topic, and only crypto news sites talk about it.

There are other articles that suggest Proof of Stake is more secure or just as secure.

Research Paper 1: Proof-of-Work versus Proof-of-Stake: A Comparative Economic Analysis

There's also a newer 2025 version of this paper

Its abstract keeps it very simple:

We develop an economic model to compare equilibrium security of Proof-of-Work (PoW) versus Proof-of-Stake (PoS) blockchains. We derive general conditions to determine when PoW blockchains are more secure than otherwise equivalent PoS blockchains and vice versa. Applying real-world parameter values to these conditions, we demonstrate that PoS blockchains are more secure than otherwise equivalent PoW blockchains. Furthermore, we demonstrate that PoS’s security advantage over PoW is particularly salient for high-scale blockchains.

Research Paper 2: Breaking BFT: Quantifying the Cost to Attack Bitcoin and Ethereum

In section 6.1.4: Cost to attack Bitcoin: "the attacker’s production costs would surpass 20B USD" using inefficient S9 miners. Would be easier with the newer S21 miners.

In section 6.3: Cost to attack Ethereum: "we estimate it would cost the attacker 34.39B USD to 34% attack the network" for 33% liveness attack.

CoinTelegraph summarizes it as $20B to attack Bitcoin and $34B to attack Ethereum, but there's not a single mainstream news site that reported on this research paper.

How do we resolve a Reliable Source issue where a mainstream site reports on bad speculation while "less reliable" crypto sites report on better research? HSukePup (talk) 21:11, 19 March 2025 (UTC)

You've convinced me that the Bloomberg source is being misused, and "Critics have argued that the proof of stake model is less secure compared to the proof of work model." was far too vague, and far too WP:WEASELish, anyway.

For clarity, I do not think the source itself is inherently bad or anything, I just don't think it is being summarized properly for this one sentence.

The way to expand this would be with WP:SECONDARY sources (but put a pin in that). SSRN papers are not generally useful for Wikipedia, and it's almost always better to cite published papers. The Review of Financial Studies is fine as a journal, but that paper is not actually contradicting the point which was made by the Bloomberg source. One issue is that no such model can account for bugs and errors which haven't yet been identified.

Putting "less reliable" in scare quotes kind of undermines your case here. I could just as easily put "better research" in scare quotes to emphasize my point, which is that these sources aren't all that compelling when placed in their larger context.

At the end of the day, our goal is to summarize a complicated topic for the benefit of disinterested readers. Practically speaking, the best approach for this is to focus on sources outside of the pro-crypto bubble.

The current article over-relies on primary sources, in my opinion, and this is part of a deeper issue, here. Before expanding anything, think about the intended readership of this article. This article has inherited some of the major flaws of other blockchain and cryptocurrency articles on Wikipedia. Would anyone who isn't already deep, deep in the weeds of crypto be able to make sense of the opening paragraph of this article? I doubt it. The article immediately jumps into technical details that presume a large quantity of prior knowledge and context. The body of the article starts out better, but by the third paragraph, the article is lost in the sauce again, and it never finds its way back. Most crypto articles have this issue, and I'm not the first and won't be the last to mention it.

So yes, the section on security could be expanded, but who are you expanding it for, and why? Grayfell (talk) 20:35, 20 March 2025 (UTC)

Thank you.

My main purpose is to fix misleading statements that might be picked up by secondary sources and AI. You're right. I probably don't need to expand on other topics others wouldn't normally look up.

And you make a good point about SSRN. Those papers aren't reviewed yet.

Every blockchain has errors and bugs to fix, and it's not inherently any different for PoW or PoS. Of the top of my head, I can think of the 2010 and 2013 Bitcoin hard fork reorgs to emergency fix critical bugs come to mind, as well as all the Solana outages and the Ethereum finality bug in 2023. Ethereum in particular has high complexity due to having 11+ different teams working on execution clients and consensus clients, but that complexity is due to development decentralization, not a PoS issue. HSukePup (talk) 20:00, 21 March 2025 (UTC)

I suggest looking for reliable, independent sources which give an overview of security issues. Only use primary sources for specifics when necessary to clarify an existing point of confusion. If you can go without citing a primary source at all, even better. I often recommend the essay WP:BACKWARDS as a way to explain this approach. Hopefully, this will make the article easier to understand without over-complicating an already complicated topic and without resorting to jargon and circular explanations. Grayfell (talk) 23:32, 21 March 2025 (UTC)

References