Talk:Hashgraph

One RS

There is literally one RS in the whole article. It's a good RS! ... but the rest is entirely primary sourced.

I just tagged the bad sources. Are there any other RSes? Can we base an article on literally a single RS? - David Gerard (talk) 18:59, 10 April 2018 (UTC)

Suitable other sources: TechCrunch, Forbes. That Forbes link is particularly good as it gives an opinion from a relevant expert critic, Emin Gün Sirer - David Gerard (talk) 19:04, 10 April 2018 (UTC)

Per the above, I've cut the primary-source material ... which was >90% of the text - David Gerard (talk) 10:12, 15 April 2018 (UTC)

Restored initial content. Capankajsmilyo (talk) 10:21, 15 April 2018 (UTC)

+1 - and I added Sirer's substantive critique - David Gerard (talk) 10:50, 15 April 2018 (UTC)

Primary-sourced version was restored, and I just removed it because it's as unacceptably badly sourced now as it was then. Dbal99 do we have third-party Reliable Sources on any of this stuff? (There's no way Squawker counts as an RS, particularly for mathematics.) - David Gerard (talk) 09:52, 25 June 2018 (UTC)

I don't see the problem with it being the primarily sourced, in this particular case. My goal was to actually summarize what is unique about the algorithm. All you have done by requiring technical secondary sources is made the page useless and deletable. - User:Dbal99 01:02 PM, 25 June 2018 (UTC)

If that makes it deletable, then it was already deletable - David Gerard (talk) 18:34, 25 June 2018 (UTC)

I have two more RSes here: Businesswire ZDNet Blearn (talk) 18:32, 7 September 2018 (UTC)

BusinessWire is literally a press release service - that's a primary source, not an RS. ZDnet source is reporting a funding round for Swirlds, nothing about Hashgraph itself - David Gerard (talk) 09:54, 8 September 2018 (UTC)

Looking for some context

Hashgraph has been spoken about left and right and Wikipedia community keeps deleting new material all the time. I understand there is a huge discrimination towards cryptocurrency in general, but surely in some form, a certain flexibility needs to be accepted. (some projects in the crypto world actually tries to resolve real world issues)

It doesn't make any sense that everybody's Wikis got more info on Hashgraph than Wikipedia themselves. This is the point where you have to ask yourself how far are you willing to censor data that might be seen questionable but still 3rd party reliable?

I use to rely on Wikipedia for valuable info, but I found out in the recent days that a lot is filtered out. (It makes it hard to consider the next donation to the website when the community itself stop considering curious people like myself)

And before you tell me you are following the community guidelines, I will have to strongly disagree: Tons of article on different types of crypto currency has been submitted in different languages without this level of censorship and using poor or lacking citations. It gets to a point that the editing is not consistent across articles and then feels more like dictatorship.

Please do not delete this comment until a panel of your peers had time to revise it. No more dictatorship editing.

I speak as a community member, I expect some resolution in a near future.

If you delete someone's addendum, at least have the curtesy to explain and help, straight out deletion with a single line reason is plainly aggressive behavior. Some of us put in the hard work to find the articles, it hurts to see it deleted in seconds.

P.S.: This technology has yet to be fully release to the public, this article is just going to get more and more traction and it is better people like myself who cares to the quality of the edit than the masses flocking to Wikipedia overloading this page with edits or trying to create new articles about it. Don't wait too long to start allowing more flexible changes, I doubt the community would wait; just a matter of time until more people like me keeps asking why, instead of having 1 every week, you'll eventually have 10 edits a day. 66.46.127.94 (talk) 23:10, 9 August 2018 (UTC)

The context is that cryptocurrency articles are magnets for advocacy and financially-interested spam. This is why the general sanctions on cryptocurrency articles, which you've been notified of, exist. WP:RS, WP:N, WP:OR and WP:V apply, and there's no reason cryptocurrency articles somehow constitute a special case. Also, WP:OTHERSTUFFEXISTS has never been an excuse to add rubbish to an article - perhaps you should be removing it from those other articles - David Gerard (talk) 16:03, 10 August 2018 (UTC)

Moreover hashgraph is not that special. It's a certain incarnation of a broader family called "virtual voting, total order algorithms". See the last paragraph. It would therefore be much more interesting to write an article on that broader class then to focus on one (proprietary) incarnation. This would also solve the concerns, that this article is just a business trick from the Hedera-cult. So maybe delete th hole page and write a page regarding virtual voting itself?

Can a ledger raise money?

Regarding:

"Hedera Hashgraph is a public distributed ledger based on the Hashgraph algorithm which in 2018 raised $100 million at a $6 billion market cap."

How can a ledger raise money?

--Mortense (talk) 09:31, 17 July 2019 (UTC)

Lol. That part is probabliy written by someone closely affiliated to Hedera. For those people, the algorithm is like a holy grail and from that mindset one can understand that they think the algorithm raised the money. Its almost cult-like and ingenious marketing.

Semi-protected edit request on 27 July 2019

Please change: https://www.hedera.com/hh-whitepaper-v1.4-181017.pdf

to: https://www.hedera.com/hh-whitepaper-v1.5-190219.pdf

because this is an updated Whitepaper. 2A02:C7F:184C:9700:5923:3753:23B1:BC07 (talk) 21:44, 27 July 2019 (UTC)

 Done After confirming the quotes are in the new version. --Trialpears (talk) 21:27, 13 August 2019 (UTC)

Hashgraph is a clone from a 1993 family of algorithms

The original idea can be found in Moser & Melliar-Smith [Louise E. Moser and P. M. Melliar-Smith: 1999, Byzantine-Resistant Total Ordering Algorithms. Inf. Comput.. 150. 75-111.]

In that orginial paper virtual voting on "hashgraphs" [business term for Lamports causality graphs] is explained in detail and four algorithms are presented to establish linear order on these graphs. However Moser's algorithms are faster. Hashgraph has local computational complexety of O(n^3). Moreover Mosers algorithm respect the natural partial order of causality inherent to any Lamport clock, or Lamport timestamp. Hashgraph does not respect that order.

If you look at Swirlds paper, Moser & Melliar-Smith are never mentioned. Also nowhere in all of Hederas business talk. This so awkward, I get red flushes seeing this Lemon Baird guy talking about in videos how he had "invented" virtual voting.

That entire Hedera company is a blueprint of the imperalistic US way, stolen ideas, sold as originals, protected by "laws" most of the world rejects (patanting of mathematical ideas, e.g. abstract algorithms).

I think it is important for a neutral PoV to mention this in the wiki. Mostly to make people aware of the unscientific approach this "former" mathematician took. However I'm too opinionated to write it.

Besides that, it is at least necessary to put hashgraph into perspective, as a certain type of offshot of Moser & Melliar-Smith's idea of virtual voting in causality graphs. Other offshots are given by Avalanche and blockmania. They are, too, based on virtual voting in causality graphs. — Preceding unsigned comment added by 79.235.125.68 (talk • contribs)

yeah, I've been told about this. The problem is to find a Wikipedia-quality "Reliable Source" that makes the comparison - and for a crypto-related article, that means either mainstream news coverage (not just a crypto site), or a peer-reviewed academic source that makes the comparison. We can't just say it because we think it's the case, that's called "original research" or "synthesis" in local jargon - David Gerard (talk) 20:46, 30 July 2019 (UTC)

Moser and Smith explicitly say: "”votes are not contained explicitly in the messages, but are deduced from the causal relationships between messages" [page 84]; Thats also Bairds definition of virtual voting, but Mosers is from 1993, published in 1999 in the mentioned paper "Byzantine resistant total order". One can at least reference that line from Mosers paper in the article as a comparison to similar/alternative and former approaches, saying that virtual voting is from 1993, which is proofable as its in the Moser paper. The latter of which is as reliable as it gets, since Melliah Smith is one of the godfathers of byzantine fault tolerance. This is also relevant for everyone who want to implement a virtual voting based algorithm, but is currently scared due to obscure US patents on mathematics. In that case wikipedia might suggest Mosers algorithm as a free and more efficient alternative, rendering hashgraph as somewhat anachronistic.

One just has to do a google image search for "lamport clock" to see the "hashgraph" appearing.

Besides, why is there a wiki about hashgraph at all? That algorithm is not that special. The cult like company Hedera invests a lot of money to make it look like its special. Instead, there should be a wiki on "virtual voting based total order algorithms" because that is a superclass, where hashgraph is an example. In that article, one might give a paragraph to hashgraph, but not an entire wiki-page. Thats much more neutral, because currently, it looks like hashgraph gets unjustified credits, while literally no other virtual voting based total order algorithm gets even a single line. Hedera and Swirlds make it even look like, that this Leemond Baird had invented virtual voting. Thats a bit strange giving the fact the original inventors Moser and Melliah-Smith are credited NOWHERE on wikipedia. Same for the entire category.

So shoulden't the burden of proof be on Hederas side? They should give proper references to their claim of originality not the other way around. Virtual voting is Moser & Melliah-Smith thats the common knowledge and tought in every graduate curse on total order in distributed systems

Concept

There are also many experts who see fundamental flaws in the hashgraph design, once a transition to a permissionless setting is made:

https://ethresear.ch/t/hashgraph-consensus-timing-vulnerability/2120/45

But the article appears as to picture this algorithms much more favorable then it actually is. This gives a very wrong picture, especially for investors and those previously mentioned "cult-like" believers. It needs an update soon — Preceding unsigned comment added by 141.45.12.194 (talk) 13:12, 21 August 2019 (UTC)

Yahoo! Finance Revert

@David Gerard: Would you mind sharing a link to the "no-account crypto blog" the Yahoo! Finance story was copied from? I did some Googling and couldn't find it, it seems to have been buried by re-shares of the story. Also, while we're at it, I'm not familiar with the "fidelitous reprint" doctrine. Am I missing that on WP:RS? It struck me as odd and I'd like to look into it more. Thanks in advance.  White Whirlwind  03:37, 10 June 2021 (UTC)

The Yahoo article literally says at the top that it's from BeinCrypto. Yahoo Finance original articles are RS, but it also reprints a lot of trash. WP:RSN has always considered this sort of straight-up reprint to have the reliability of the source. NASDAQ News has the same problem. If you don't see why that would be, RSN would be the place to ask - David Gerard (talk) 08:54, 10 June 2021 (UTC)

We are only using top-shelf sources for blockchain articles, if the source is even remotely questionable it isn't kosher. Yahoo finance is generally all junk and reprints, I dont think they have their own editorial department. Jtbobwaysf (talk) 13:07, 10 June 2021 (UTC)

oh, they do! And their own stuff is WP:NEWSORG. But I haven't seen anything original from them on crypto - David Gerard (talk) 23:28, 10 June 2021 (UTC)

@David Gerard: Ah, I didn't know that little logo meant anything. And thanks for the info on the RS situation. I was pretty sure there was nothing official on that on the guideline, but I wanted to check.  White Whirlwind  02:45, 11 June 2021 (UTC)

Update: Present-Day Governing Council Members

Hi all, I'm A at Chainlink and I've been paid by Chainlink Labs. I won't make any edits to this page myself to remain compliant with COI rules and have disclosed my COI above. I wanted to suggest an update to content regarding Hedera Hashgraph's governing council, which was last updated in May of this year. Since that time, a number of organizations that now have Wikipedia pages have joined the council, so my suggestion is as follows (edits bolded):

Hedera is owned and managed by a "governing council" of global companies and entities that have invested in it. The council's members include Swirlds, as well as Google, Boeing, IBM, Deutsche Telekom, LG, Tata Communications, Électricité de France, FIS, University College London, the London School of Economics, DLA Piper, Shinhan Bank, Standard Bank, Avery Dennison, Chainlink, Dentons, ServiceNow, Tata Communications and several others.

Let me know what you think. Thank you! A at Chainlink (talk) 19:47, 10 December 2021 (UTC)

References

Critism section Requested Edit

Professor Emin Gün Sirer who is being quoted in this section, is the Founder and CEO of Avalanche, which is a direct competitor to Hedera.

He should not be quoted, because he is clearly running down and talking negatively about his competition.

This reflects badly on Wikipedia, in my opinion.

I had edited and removed this whole "criticism" section completely, but the edit was reverted.

Can someone else take a look and see if they can take a stab at editing the Criticism section.

My thought on removing the whole section is If you remove the quote by Sirer, the rest of the section is basically not needed, because it is a rebuttal to the quote by Sirer. Vicwd (talk) 14:09, 22 February 2022 (UTC)

Semi-protected edit request on 25 February 2022

Hashgraph is now an opensource technology, At the top of the page it says hashgraph is patented Batteagle (talk) 22:32, 25 February 2022 (UTC)

 Not done: please provide reliable sources that support the change you want to be made.  melecie  t - 05:01, 26 February 2022 (UTC)

Source: https://hedera.com/learning/hedera-hashgraph/open-source-at-hedera 2A02:8109:8613:7900:9754:D990:F366:704 (talk) 09:39, 28 August 2024 (UTC)

It's still covered by a host of patents: https://patents.justia.com/assignee/hedera-hashgraph-llc, indeed the Apache 2.0 license only license patents with respect to the work in question. Any other implementation (especially those not covered by Apache 2.0*) would still be infringing. It is important to note that Hedara has applied for additional patents after the license change, which would be pointless if they had no plans to exercise them.**

*One could implement an otherwise infringing Apache 2.0 licensed version Ship of Theseus style from the released code

**Publication is sufficient to establish prior art, there is no need to obtain a patent simply to prevent another party from obtaining a similar patent SnoFoot (talk) 01:42, 30 August 2024 (UTC)

Hashgraph is a Rebranding of an Algorithm from the 90th

The idea of how to achieve consensus in directed acyclic graphs of messages acknowledging other messages is from the paper "Byzantine-Resistant Total Ordering Algorithms" ... Hashgraph rebrands and specializes this but without any reference to that paper. 2A00:20:6082:B92F:CCEC:E21B:C545:8C5E (talk) 21:48, 30 July 2024 (UTC)