Talk:Nervos Network

Feedback from New Page Review process

I left the following feedback for the creator/future reviewers while reviewing this article: Thanks for creating the article!.

✠ SunDawn ✠ (contact) 19:30, 7 October 2022 (UTC)

Changes

I have tried to construct the lead in light of the sources provided. The sources are of reasonable quality for the content discussed. Terms like decentralized, permissionless, are factual, informative and present in the provided sources. Layers which are linked to share particular attributes is a theme that's ubiquitous through the sources and the passage regarding trust-anchor is specifically mentioned in the cited source. I trust we aren't going down the road of expecting exceptionally high quality sources for content which is uncontroversial. In any case, we do have cite tags if further sourcing is requested. That seems more collaborative and collegial than just deleting passages outright. RiskAficionado (talk) 17:38, 7 November 2022 (UTC)

Wikipedia is a general audience encyclopedia. Our goal is to explain things so that people who do not already understand all of this jargon can understand it, at least broadly. Your version of the article did not add true clarity, it added technical details that implied significance without explaining it. The inclusion of details supported only by the project's own website, without neutrally summarized context from a WP:IS, reads like jargon. But worse, again, it implies this project is significant without properly explaining why it is significant, which makes it read like promotional content. As a reminder, Wikipedia is not a platform for promotion or advocacy.

In order for this seemingly promotional content to be included in the lead, reliable, independent sources need to contextualize why this is encyclopedically significant. Ideally they need to do so constantly, and this is first used to clean up the body of the article. Then, once this has been added to the body of the article, it can be briefly and neutrally summarized in the lead. Grayfell (talk) 20:39, 7 November 2022 (UTC)

Oh, also, the Yahoo! source is just republished from CoinDesk, which is not a reliable source, per Wikipedia:Reliable sources/Perennial sources and many past discussions on many noticeboards. As a softball interview, the Newsweek source is not as weak as CoinDesk, but it is still pretty weak and should likely be treated as WP:PRIMARY, meaning every claim supported by it should be attributed to the interviewee. Grayfell (talk) 20:51, 7 November 2022 (UTC)

Just coming back to this (wow it's been almost a year..), a few of the academic sources I added introduce it as public and permissionless. The latter is slightly jargony but this article is a sub-niche of an already highly niche topic - common jargon can be easily addressed by linking to the appropriate article (Blockchain does this in its first sentence with cryptographic hash). It would be somewhat unnecessary to unpack every technical phrase each time it appears in a new article. My second point is that these descriptions help to clarify what type of blockchain it is, as some blockchain protocols by design are private or permissioned. RiskAficionado (talk) 22:57, 22 October 2023 (UTC)

Citation requests

As per WP:PRIMARY, relatively straightforward information about the design of the network can be sourced to primary sources as long as care is given to weighting. Could you explain where you disagree Grayfell as there have been requests made for secondary sources RiskAficionado (talk) 22:38, 22 October 2023 (UTC)

For simplicity, I will address this in one section, since my original concerns remain undressed.

I don't know what requests you are referring to.

Regarding this edit, MDPI is a poor-quality publisher. If I had noticed this source before, I would've removed it sooner. I dispute that the International Conference on Software Engineering & Knowledge Engineering or its host, KSI Research are reliable in this context. The term "trust-anchor" is cryptography jargon, and per trust anchor, this refers to "an authoritative entity for which trust is assumed and not derived" which raises many more questions than it answers. Using jargon in this way is not informative and therefor not appropriate for an encyclopedia article. So, we would need reliable source to explain this, and further, we would need reliable, independent sources to indicate why this detail is important to this article. Simply presenting this as an isolated fact without commentary reduces it to a factoid, which is, as before, too promotional.

As a minor note, I do not necessarily accept that the Romanian Academy has relevant expertise here, since its primary focus appears to be Romanian language, literature, and culture. Grayfell (talk) 23:38, 22 October 2023 (UTC)

I mean, you can "say" your concerns remain unaddressed, but I see no attempt to respond to my answer further above.[1] Additionally, you tagged multiple areas requesting non-primary sources for what is basic information, my comment above means to demonstrate why that request that is unnecessary.[2]

I believe you are conflating "source-based explanation" with "promotional content," which based on your edits elsewhere stem from taking an overall cynical view of the subject matter (blockchain). That is the only way I can explain your oversensitivity to seemingly basic, innocuous information such as describing a blockchain to be public, permissionless, or indeed how a multilayer network derives some security from its base layer. Or arbitrarily raising the bar for reliable sources to dismiss publications outright. Reliable sources are not a monolith and should be weighed against the nature of the information conveyed. As a general rule, I would say that papers in independent academic journals with established peer-review procedures are quite suitable here.

• Mathematics (journal) appears to meet those standards and is also well cited. Regarding MDPI, it seems clear that there is no consensus about the publisher's overall reliability and that some journals in its stable (mostly related to healthcare) were labelled by the Norwegian Scientific Index as predatory, of which Mathematics was not one. Again, in the context of basic information about CKB, and the fact a series of related papers are published across multiple journals, the grounds for dismissing this source is fairly weak.
• Regarding this source: (Feng, Yichun; Lu, Yuteng; Sun, Meng (July 2021). "Modeling and Verification of CKB Consensus Protocol in UPPAAL" (PDF). Proceedings of the 33rd International Conference on Software Engineering & Knowledge Engineering (SEKE 2021): 1–10. doi:10.18293/SEKE2021-072.), I'm afraid saying you dispute something is simply not enough. The researchers are independent of the blockchain, are experts in their field, and have presented this research to an established conference. KSI Research is also the publishing house behind the Journal of Visual Language and Computing which is an independent, peer-reviewed academic journal with relevant expertise.[3]
• Lastly, whilst I try to always assume good faith, you wholesale deleted a bunch of unrelated text about the NC-MAX protocol in addition to deleting the source. What you haven't realised is that the other NDSS source discusses all of those elements in detail and is cited later in that passage. This illustrates to me that you haven't read the sources and are just deleting what you personally and erroneously view as promotional.

RiskAficionado (talk) 14:00, 23 October 2023 (UTC)

...whilst I try to always assume good faith... I think you should try a bit harder. Perhaps you are a bit rusty. Since you have decided to try and personalize this discussion, for some reason, and since you have functionally been a WP:SPA for this article for all of its existence, I will take this opportunity to remind you that editing articles with which you have a conflict of interest is strongly discouraged. If you have such a conflict, I would strongly encourage you to disclose it.

I make no secret of my skepticism towards blockchain hype, but Wikipedia as a community increasingly shares this skepticism. This is in large part due to the overwhelming volume of poor-quality sources which drown-out better-quality ones and make evaluating WP:DUE for these kinds of details a hassle. While some of this problem is more obvious and unrelated to these specific issues (such as WP:COINDESK) a related cause is the glut of academic paper-mining, conference proceedings, pre-prints that never go anywhere, journals that are too tiny to properly evaluate, journals with extremely lax submission standards, predatory journals, superficially-legit appearing self-published books, and similar academic and pseudo-academic noise. There are an absolutely mind-boggling number of journals just in the blockchain sphere alone. Clearly, not all of them are broadly reliable, much less reliable for these kinds of niche contexts.

For this article, specifically, whether or not KSI etc. are reliable is not based on my opinion nor on yours. It is based on the outlet's reputation for accuracy and fact checking. That a publisher also publishes a separate journal, the Journal of Visual Language and Computing (not to be confused with the Journal of Visual Languages & Computing which continued as the Journal of Computer Languages), really doesn't matter much for this. All sources are evaluated in context, and you do not have consensus that these sources are reliable in this context. The more niche the claim, and yes, the more flattering the claim, the more strictly these sources should be judged. This is how WP:NPOV is maintained.

To put it another way, our goal isn't to find a way to justify including these details, it's to figure out which details will help readers understand this topic as separate from all the other blockchain article. We decide this via reliable, independent sources. Adding information which you know to be true and then back-filling with flimsy sources is not the best approach. Grayfell (talk) 20:23, 23 October 2023 (UTC)

Oh, also, KSIresearch.org doesn't link to the JVLC journal on their main page for whatever reason, but they do prominently link to "The TDR System for Chi", which is WP:FRINGE and concerning for multiple reasons. This is enough to cast doubt on the outlet itself. Grayfell (talk) 21:24, 23 October 2023 (UTC)

No, I am not involved with this project. I have an interest in this article because I started it. As you might have noticed, time doesn't permit me to edit more blockchain articles although I have been following the topic for a number of years.

Secondly, I appreciate you explaining your perspective but I don't agree at all with your scepticism. Not all published content in academia is of good quality, that's fairly obvious, although you seem to be demanding sources above and beyond the standard expected considering the context of the information, which in this article does little more than outline the architecture concisely. As editors our goal is to appraise what the reliable sources say and present that content neutrally and with a weighting proportional to its coverage in the sources. There are multiple academic, independent reliable sources that explain NC-MAX for example, and the passage you deleted faithfully and factually summarises those papers in a few short sentences. I am still awaiting an explanation as to why you deleted it. And there has been no response about Mathematics which as I showed has flimsy grounds for dismissal. As I have reiterated and as has been linked to, the appropriateness of a source in part depends on what it is being used to support. For basic facts, the sources provided suffice, as do primary sources in a limited context.

Having looked into KSIresearch, I do find it somewhat odd that a fringe study should be linked on its main page, although it seems that the paper relates to technology applied to an ancient tradition. However, I can agree that this calls the usability of the publisher into question and should be avoided.

I will move to restore some of the better sourced content that was deleted (if your concern is tone perhaps you could rephrase it instead of deleting). I also note a {notability} tag has been added: based on the number and types of sources available I do believe this topic meets notability guidelines. RiskAficionado (talk) 22:15, 4 December 2023 (UTC)

I hold cryptocurrency articles and their sources to the same standards I hold other articles. Specifically, I try to remove pseudo-journals and predatory journals from articles related to other areas of pseudoscience whenever I notice them. Since much of blockchain-based cryptocurrency is based on pseudo-economics, this applies here, as well. As with any pseudoscience, there is a mountain of sources, sometimes even reliable sources, which might lend credibility to a fringe position or even superficially support that position in some contexts. Since this is an encyclopedia, Wikipedia should not accept this at face value. It is not enough for us to include information which is technically accurate based on our own evaluation of due weight. We also have to provide context to readers, and the way to do that is with reliable, independent sources. Sources which do not start from a position of skepticism, or which merely pass-along bland technical details without context, are inherently less reliable and less useful for writing neutral articles.

I would request that if you insist on restoring any of this, you reevaluate the tone and level of detail you have been using so far. If these specific details are important, it should be possible to indicate why they are important as explained by reliable, independent sources. If you understand why KSIResearch has a lot of red flags, than you should understand these concerns for the rest of the article as well. Having to dive into obscure conference proceeding, preprints, predatory publishers, and similar borderline (or worse) sources is a sign that the article is not being written with disinterested readers in mind. All sources must be judged in context, and several of the sources you cited are not reliable.

As an additional note, per Wikipedia:NEWSWEEK I don't think this source is ideal. It's a puff-piece interview, and the author's last article for Newsweek is credulous churnalism promoting an unrelated crypto scheme. Grayfell (talk) 22:56, 4 December 2023 (UTC)

You are stringing together a lot of incorrect assumptions. First and foremost, the basis of blockchain is cryptography securing a distributed ledger. That Blockchain wiki article certainly does not paint it as a pseudoscience (because it isn't). Crypto-economics on the other hand is more speculative, and there is sometimes overlap of the two. Yet the journals cited here (and I will share them below again for convenience) deal with the topic in respect of the blockchain technology, the consensus mechanism, the overall architecture. For which, their coverage in journals that relate to mathematics and computer science is fitting.

• Sun, Meng; Lu, Yuteng; Feng, Yichun; Zhang, Qi; Liu, Shaoying (November 2021). "Modeling and Verifying the CKB Blockchain Consensus Protocol". Mathematics. 9 (22): 2954. doi:10.3390/math9222954. ISSN 2227-7390.
• Bu, Hao; Sun, Meng (March 2020). "Towards modeling and verification of the CKB block synchronization protocol in Coq". Formal Methods and Software Engineering: 22nd International Conference on Formal Engineering Methods, ICFEM 2020, Singapore: 287–296. doi:10.1007/978-3-030-63406-3_17.
• Zhang, Ren; Zhang, Dingwei; Wang, Quake; Wu, Sichen; Xie, Jan; Preneel, Bart (2019-04-01). "NC-Max: Breaking the Security-Performance Tradeoff in Nakamoto Consensus" (PDF). Network and Distributed Systems Security (NDSS) Symposium 2022. ISBN 978-1-891562-74-7.

Again, your stated opposition to the subject matter taints your judgement, resulting in vague criticisms such as "obscure conference proceeding, preprints, predatory publishers" which do not stand up to inquiry.

• I note, you did not respond to my findings about Mathematics (journal) that it is not among those listed by the Norwegian Scientific Index as predatory (whereas some others within the MDPI stable are, particularly healthcare related). As per its article, it appears to be reasonably well cited by other journals. If there is no overall consensus about MDPI (as per RS/N), I see no justification in dismissing Mathematics through mere guilt by association.
• ICFEM, according to Complutense University of Madrid, "is an international leading conference series in formal methods and software engineering."[4] Furthermore, its conference proceedings are published in Springer. To call ICFEM obscure is misguided.
• NDSS Symposium is a global security conference hosted by the Internet Society. According to the University of Oxford, NDSS is "one of the top 4 academic security conferences in the world." [5] A cursory search shows that its events are sponsored by Microsoft, Google, IBM, etc. Again, not obscure in the slightest. RiskAficionado (talk) 00:33, 6 December 2023 (UTC)

I don't think you understood what I said, so calling this "incoherent" suggests that further discussion will be difficult. I did not say that blockchain was pseudoscience. I said, and maintain, that much of blockchain-based cryptocurrency is based on pseudo-economics. A classic sign of pseudoscience is the use of related sources about a valid science to imply legitimacy to a neighboring pseudoscientific field. A water ionizer is based on pseudoscience, so any use of sources about electrolysis and water (of which countless exist spanning hundreds of years) should be treated very skeptically, even if they directly mention water ionization. This doesn't mean the sources themselves are pseudoscience.

Likewise, any claims made about this particular cryptocurrency would need reliable sources to be evaluated in context. Lengthy details about technical specs which very, very few readers understand or even care about is not proportionate and not helpful. It is not enough for a detail to be technically accurate, it has to be contextualized by reliable sources. This is not only how Wikipedia handles WP:FRINGE topics like pseudoscience and pseudo-economics, it is how most articles should be written.

As for MDPI, I did not respond to you because I've said plenty already. It's a bad publisher which shovels out a lot of unusable crap. If this particular source is the only one you can find for any particular detail, do not assume that particular detail belongs in this or any other article.

As for NDSS being a "top 4", I do not find a perfunctory promotional press release for an unrelated team working on an unrelated project several years prior to be a compelling indication that this specific source is useful. Being sponsored by large corporations is absolutely not an indication of reliability either. That specific NDSS source is both primary and says comparatively little about Nervos itself. Not every source which mentions Nervos needs to be cited. This applies to all sources. Grayfell (talk) 01:12, 7 December 2023 (UTC)

Where did I say incoherent?

I'm glad that you agree blockchain is not pseudoscience. On that basis, you should be able to identify its distinction from cryptocurrency or cryptoeconomics. The latter does contain economic ideas one can describe as experimental. As I said earlier, there is overlap, but they are not one and the same. For instance, China takes a harsh view on cryptocurrencies but not blockchain technology. In this article, the sources used are talking about the blockchain architecture of Nervos, and that is the intended focus of the article. So the arguments built on this flawed assumption of pseudoscience do not carry weight.

About MDPI, I'm really trying to glean from you any clarification about why Mathematics (journal) is an unreliable source. If you personally don't like MDPI, fair enough, but there is no RS/N consensus about the publisher (i.e. it is case specific) and this particular journal is not included amongst the ones singled out for dubious practices. It's an academic journal that gets cited by other academic journals. I'm not sure what else needs to be said.

NDSS - you claimed it was obscure - in the global security industry it definitely is not (I knew this before citing the Oxford university comment). It's a well-respected conference for experts and I'm happy to find more quotes to confirm that notion but you and I both know how that will end up. The source itself is primary albeit published through a reliable independent medium. The content is NC-MAX (which is the consensus mechanism that is bespoke to Nervos) and how it works.

Lastly, if your concern is lengthy details, I can summarise a few of those subheadings into one paragraph if that helps to alleviate your concerns. But I'm hoping we can move past the objection of not having enough sources (the news items are less reliable but not unreliable) and concentrate more on presenting factual content in an encyclopaedic manner. RiskAficionado (talk) 23:37, 10 December 2023 (UTC)

Sorry, I meant "incorrect". Whether or not my assumptions are assumptions, and whether or not they are incorrect, you still do not seem to understand what I have been saying.

I'm not saying that blockchain is a pseudoscience, but I'm also not saying it isn't a pseudoscience. My point was that these details in isolation are pseudoscientific even if they are accurate in other contexts. Calling these fringe economic ideas "experimental" is a bit ahistorical. Many of these "experiments" have already been conducted in past centuries, which has lead to the now-common observation that cryptocurrency is what happens which engineers try to use their knowledge of cryptography and computer science to reinvent 6,000 of economics from first principles. Just as someone can ionize water without it being pseudoscientific, someone can create a superficially decentralized append-only ledger. Whether or not this is pseudoscience depends on why they are doing this.

To put it another way, what, exactly, does this blockchain do that justifies an encyclopedia article? The lead is extremely vague about this (it is "designed for different functions"), and the body lacks sufficient explanation as well. The source for "different functions" is quite old and is entirely about funding. It mainly parrots the project's founder and investors without adding any real explanation beyond out-of-date hype.

So why are any of these sources (reliable or not) even talking about this specific blockchain? Apparently, it's because of the amount of money that was poured into it. This is the reason general-audience sources are giving an overview of this specific blockchain. If the substance of the body is about arcane technical details, but not what is being done with that money, the article is using technically-reliable sources to mislead readers. Curious and attentive readers are going to be frustrated by this approach, and rightly so.

How do we summarize this topic in a way that is informative to disinterested readers of an encyclopedia? Expanding technical details in the body of the article isn't the correct approach even if we agreed on the reliability of these sources. (To emphasize: we do not yet agree on the reliability of these sources).

If you are a member of the global security industry, as your comment suggests, than perhaps it will be helpful to remember that this is a general audience encyclopedia. It doesn't matter that you personally know the symposium is well-regarded if it doesn't help us explain the topic to readers in a proportionate, neutral, and WP:DUE way.

Incidentally, if you are confident you have better sources for the reputation of the NDSS, by all means add it to Internet Society (or create a standalone article if it meets WP:GNG). Grayfell (talk) 00:48, 11 December 2023 (UTC)

Regarding NDSS, I've created a separate topic for that. Regarding other sources like Mathematics (journal) - I haven't seen any evidence it should be treated differently to other respected scientific journals. I think it is quite appropriate as a source - there is no consensus about MDPI on certain topics, but this topic is not one of them nor is this journal identified as predatory as mentioned previously. It is well-cited by other scientific journals and has a decent impact factor. On this basis I don't think it's justified to remove this source from the article.

The sources on this topic are largely of two types:

1) News sources covering its launch or its capital raised. That's understandable from a business and web-tech startup perspective. Companies or projects that are able to raise such capital do so because of the solutions they claim to offer. So I don't really see a problem with such articles in establishing both notability and their use as sources for information. Clearly, if multiple outlets decide to cover it then I'd argue it certainly does belong in a modern encyclopedia that doesn't exclude articles on novel technologies.

2) Academic texts covering the architecture of the blockchain. These are by and large reliable, with some exceptions that we've talked about. and provide further information about the key features of the blockchain. If such features weren't novel or research-worthy, they would probably not justify articles in the first place nor presentations at expert conferences.

I would also argue that, after the above, supplementary information from primary sources to fill in gaps is acceptable and that is established in policy.

About the lead, it's not necessary to fill it with fine details that should normally occupy the body. Admittedly it reads worse in its current state because much of the other subsequent sentences that describe said functions, despite being attributable to sources, were deleted somewhat haphazardly because you saw it as "too promotional" even though you could have reworded it if that was your concern.

I don't have a strong view on cryptoeconomic theory as I think that's a distinct matter to blockchains, cryptography and their functions. But I note your scepticism of the overall genre, which I reiterate shouldn't colour your appraisal of the sources thereby creating higher barriers for inclusion, or taking a default-negative view of things as evidenced by NDSS and Mathematics. RiskAficionado (talk) 01:28, 15 January 2024 (UTC)

NDSS

Starting a new topic for this. NDSS, based on a cursory survey, according to multiple experts in the security field, is a "top tier" conference.

• According to Temple University Computer and Information Science Department, NDSS is " one of the top 4 security conferences" [6]
• According to University of Oxford Department of Computer Science, NDSS is "one of the top 4 academic security conferences in the world." [7]
• According to Jianying Zhou, Professor of Cyber Security at Singapore University of Technology and Design, NDSS ranks #2 out of the top 20 cyber security conferences.[8][9]
• According to Guofei Gu, Eppright Professor in Engineering in the Department of Computer Science & Engineering at Texas A&M University, NDSS is a Tier 1 computer security conference.[10][11]

There's more I can find but honestly I don't have the time for an exhaustive search. But I trust this clarifies that NDSS is a highly respected - and not obscure - global security conference by academics. RiskAficionado (talk) 23:59, 14 January 2024 (UTC)