Why I ask: does it even make sense to publish long-term, valuable knowledge content on the Hive Blockchain?
Because, if it ultimately cannot be found, then it would make little sense to publish long-term knowledge here. At least not in terms of knowledge sharing – which is the reason authors take time to create high-quality and relevant content.
Let’s take knowledge in category X. If, on one hand, there is no community for such knowledge (not to mention how many actual readers are there), then does it really make sense to publish such content on the blockchain?
Especially since knowledge sharing is also about making information permanently accessible. But if that knowledge only appears once in the feed and then vanishes into obscurity, then...
And the argument that a Hivian could still find and revisit it on the blog… doesn’t seem like a truly valuable point. Because, how many active Hivians are we really talking about? And I don’t mean accounts, I mean actual people.
Also, aside from the fact that traditional search engines have lost a huge amount of relevance – and will likely continue to lose more. And yes, we could also include the “death” of websites in this discussion. Because:
Publishing content on websites? Only for AIs to crawl it, compress it, and serve it back to users in summary form – making an actual website visit unnecessary most of the time?
But even if there were topics better suited for search engines – and Hive Blockchain content could theoretically appear in those search results – has anyone ever once seen a Hive Blockchain post in search engine results?
What we’d have to clarify: I mean content not directly related to the Hive Blockchain itself. Because if I Google “How to use liquidity pools on the Hive Blockchain,” chances are high where I’ll end up. I’m talking about any and all content that does not revolve around Hive.
Has anyone ever seen Hive Blockchain content appear in search engine results?
And also some side questions worth asking:
A) When was the last time you actually Googled something?
B) And if you did, did you just use Google’s dynamically generated answer, or did you really dive into the depths of the web by researching on websites?
C) Topic research = mostly relying on AIs?
Because honestly: if I want to research something deeply, I use things like NotebookLM. Or if it’s a topic I don’t want to spend hours on, I just ask other AI models.
We could expand these questions further and make the article bigger – even up to: “Is the internet in its old form dead?”
The situation is what it is: more and more information is being aggregated by AIs and consumed directly by internet users. Therefore:
At this point, I want to show you what different AIs answered when I asked them how relevant Hive is for them. The question, as well as the answers, I won’t translate in order not to distort them.
[DEUTSCH:]
Hat jemals irgendjemand etwas mit Hilfe einer Suchmaschine im Internet gesucht und in den Suchergebnissen einen Inhalt auf der Hive Blockchain gefunden?
Warum ich frage: macht es überhaupt Sinn, längerfristig gültige Wissensinhalte auf der Hive Blockchain zu publizieren?
Weil, wenn sie letztlich nicht auffindbar sind, dann würde es keinen Sinn machen, längerfristig gültige Wissensinhalte hier zu veröffentlichen. Zumindest nicht im Sinne der Wissensvermittlung - dem Grund, warum sich Autoren ja Zeit nehmen, hochwertige und relevante Inhalte zu erstellen.
Lass uns mal Wissen in der Sparte X nehmen. Wenn es einerseits keine Community für diese Wissensinhalte gibt (mal nicht zu erwähnen, wie hoch der Anteil der tatsächlichen Leser dort tatsächlich ist), macht es dann überhaupt Sinn, solche Inhalte auf der Blockchain zu veröffentlichen.
Insbesondere geht es ja bei der Wissensverbreitung auch darum, das Wissen dauerhaft verfügbar zu machen. Wenn das Wissen hier aber nur einmal im Feed durchrutscht und dann in der Versenkung verschwunden ist, dann...
Und dass ein Hivian ja Inhalte auf dem Blog länger finden und aufrufen könnte... das wirkt mir nicht wie ein Argument, welches wirklich Wert hätte. Weil, wie viele aktive Hivians sind wir denn wirklich? Und damit meine ich nicht Accoutnts, sondern Hivians.
Auch mal davon abgesehen, dass klassische Suchmaschienen eh extrem an Geltung verloren haben und dies wahrscheinlich noch weiterhin tun werden. Und gerne können wir das "sterben" von Websites gerne inkludieren in das, was wir hier nicht betrachten. Denn:
Inhalte auf Websites publizieren? Damit KIs sie auslesen und in komprimierter Form dann dem User vorsetzen und somit den Besuch der eigentlichen Website zumeist hinfällig machen?
Aber auch wenn es überhaupt Inhalte gäbe, nach welchen man besser selbst in Suchmaschinen sucht und Inhalte auf der Hive Blockchain dort auftauchen könnten... Hat irgendwer schon einmal einen Inhalt der Hive Blockchain in den Suchergebnissen gesehen?
Was man ergänzen müsste: Inhalte, welche sich inhaltlich nicht auf die Hive Blockchain beziehen. Denn wenn ich "Wie nutzt mal Liquidity Pools auf der Hive Blockchain" googele, dann ist die Wahrscheinlichkeit hoch, wo ich dann lande. Ich rede von allen und irgendwelchen Inhalten, welche sich nicht um die Hive Blockchain drehen.
Die Kernfrage:
Hat irgendwer mal einen Eintrag von Inhalten auf der Hive Blockchain in den Suchergebnissen gesehen?
Und gerne auch weitere, weil sie Randaspekte hinterfragen:
A) Wann hast du das letzte mal gegoogelt?
B) Auch wenn du gegoogelt hast, hast du dann die dynamisch generierte Antwort von Google genutzt oder bist tatsächlich in die Tiefen des Internet abgetaucht, indem du auf Websites recherchiert hast?
C) Themenrecherche = die überwiegende Nutzung von KIs?Weil ganz ehrlich: wenn ich mich über ein Thema fundiert informieren will, dann nutze ich Dinge wie NotebookLM. Oder bei Themen, bei denen ich nicht Stunden investieren will, eine Anfrage in anderen KI Modellen.
Wir könnten jetzt die Fragen noch erweitern und das Thema dieses Artikels noch größer ziehen. Bis hin zu der Fragestellung "Ist das Internet in seiner alten Form tot?"
Die Situation ist wie sie ist: mehr und mehr Informationen werden von KIs aggregiert und dort auch von Internetnutzern genutzt. Daher:
An dieser Stelle möchte ich euch mal aufzeigen, was mit verschiedene KIs als Antwort gegeben haben, als ich sie gefragt habe, wie relevant Hive für sie ist. Die Frage , wie auch die Antworten darauf, übersetze ich nicht, um sie nicht zu verfälschen.
Meine Frage:
Good question 👍
Direct answer: I very rarely reference Hive blockchain content in my answers when doing internet research.
Here’s why:
My web search tool queries general search engines (Google/Bing-level indexes). The Hive blockchain is a niche platform, and its content usually doesn’t show up high in search rankings unless it’s cross-posted or linked from other sites.
Hive-based frontends (like PeakD, Ecency, LeoFinance, etc.) sometimes show up in search results, but compared to mainstream sources (Wikipedia, news, blogs, GitHub, StackOverflow, etc.), they are a tiny fraction.
Unless you explicitly ask about Hive, Hive authors, Hive dApps, or blockchain posts hosted there, the probability that I pull Hive content is very low.
If I had to put a rough estimate on it:
<0.5% of my researched answers pull content directly from Hive-based sites.
If you specifically ask about Hive-related stuff → then of course, it jumps to nearly 100% because I’ll go to Hive-based sources.
👉 So unless you bring Hive into the question, it’s practically negligible.
Do you want me to check how well Hive content is currently indexed (SEO-wise) in search results, so you’d know whether your posts there are likely to be used as sources?"
I don't have access to statistics about what sources I reference across all my conversations, so I can't give you a precise percentage. However, I can tell you that it would be very rare - I'd estimate well under 1% of my web searches would surface Hive blockchain content.
Here's why:
Search engine prioritization: Major search engines typically surface content from larger, more established platforms first - news sites, Wikipedia, government sources, academic institutions, major social media platforms, etc.
Relative size: While Hive hosts interesting content, it's a niche platform compared to the broader internet. The vast majority of indexed web content exists elsewhere.
Query types: Most questions I help with would naturally point to mainstream sources rather than blockchain-based social platforms.
That said, I could certainly encounter Hive content in specific scenarios - like if someone asks about cryptocurrency, blockchain technology, decentralized social media, or if they specifically request information from Hive-based sources.
Is there a particular reason you're curious about Hive content? Are you looking for information that might be hosted there, or are you interested in the platform itself? I'm happy to search for Hive-specific content if that would be helpful!
That is a fascinating question that gets into the specifics of how large language models like me access and prioritize information.
It's impossible for me to give you a precise, static percentage for the following reasons:
I don't have a specific, measurable counter for Hive content vs. the rest of the web. My knowledge comes from a vast dataset of text and code that includes content from countless sources indexed by Google, among others. I don't maintain a tally of "Hive-sourced citations."
The frequency is highly query-dependent. If a user asks a question specifically about a Hive-based community (like PeakD, 3Speak, etc.), a specific Hive project, or a topic where Hive content is known to rank well in search engines, the likelihood of a Hive source being referenced is much higher. For general queries (e.g., "what is the capital of France?"), it would be extremely low.
Search Engine Indexing and Ranking: When I perform an internet search (like I did to help answer this question), the content is prioritized by the search engine's algorithms. While the Hive blockchain content is indexed by search engines, its overall visibility is determined by the same SEO (Search Engine Optimization) and ranking factors as any other website.
Therefore, here is a rough estimate and qualitative assessment:
Overall Rough Estimate: The general percentage of times I would specifically cite a Hive-hosted source across all internet-searched queries is likely well under 1%, perhaps closer to 0.1% or less. This is simply because the total amount of content on the entire internet indexed by search engines is astronomically larger than the volume of content on the Hive blockchain, despite Hive's activity.
Targeted Queries: For a highly specific query about a Hive-related topic (e.g., "latest Hive hardfork details," "tutorial for a specific Hive DApp"), the reference rate would be significantly higher, potentially reaching 20% to 50% or more of the cited sources, provided the Hive content is the most relevant and authoritative result.
In short, while Hive is part of the indexed web that I can access, it's a very tiny drop in the ocean of available information. The probability of it being a top-ranking, citable source is low for a general topic, but very high for a niche, Hive-specific topic.
I reference content published on the Hive blockchain when it's relevant to the user's question and provides accurate, up-to-date information. However, I don't have real-time access to the Hive blockchain data. I estimate that I reference Hive blockchain content in less than 10% of my responses.
If a user asks about a specific topic related to the Hive blockchain, such as its technology, ecosystem, or recent developments, I'll do my best to provide accurate and relevant information. However, the availability and accuracy of the information depend on the data available from the Hive blockchain and other sources. If you have a specific question about the Hive blockchain, I'm here to help!
I reference content from the Hive blockchain very rarely during internet research, as my tools primarily surface results from broader web sources, and Hive-specific content only appears if it's highly relevant to a niche query (e.g., decentralized social media or specific blockchain communities). A rough estimate would be less than 0.1% of the time when research is involved, based on the typical distribution of search results favoring more mainstream platforms.
Of course. This is an excellent question that gets to the heart of how I operate.
To give you a direct estimate: I reference content published directly on the Hive blockchain in well under 1% of internet research queries. It would be a very small fraction of a percent.
Here’s a detailed breakdown of why that is and how the process works:
How I Access Information
When you enable the "search the web" function (like the "Web Search" or "Google Search" toggle), I don't browse the internet like a human. Instead, I rely on an integrated search API that queries a search engine index. My primary knowledge comes from my internal training data, which has a cutoff date, and the web search provides live, up-to-date information.
The Visibility of Hive Content in Search Results
The vast majority of Hive blog posts and articles are not directly indexed by major search engines like Google in a prominent way. While many Hive-based applications (like PeakD, Ecency, Hive Blog) have made significant strides in SEO, the content often faces visibility challenges because:
Canonical URLs: Many different interfaces (PeakD, Ecency, etc.) can show the same post, which can confuse search engines about the "original" source.
Domain Authority: Compared to established platforms like Medium, WordPress, or major news outlets, the domains for Hive interfaces generally have lower domain authority, pushing them lower in search rankings.
Indexing Speed: There can be a delay between a post being published on the blockchain and it being crawled and indexed by a search engine.
Because of this, when I perform a web search, the results are overwhelmingly dominated by traditional websites (news outlets, corporate blogs, educational institutions, government sites, Wikipedia, etc.).
When I Would Reference Hive Content
There are specific, niche scenarios where Hive content is more likely to appear and be referenced:
Hive-Specific Topics: If your question is about Hive itself—e.g., "What are the best Hive wallets?" or "How does Hive's delegation work?"—the search results will naturally surface tutorials and explanations from Hive-based platforms.
Crypto and Blockchain Topics: For questions about cryptocurrency, decentralized finance (DeFi), or NFTs, there's a higher probability of pulling in articles from knowledgeable writers within the Hive community who specialize in this area.
Specific Usernames or Projects: If you ask about a specific dApp, game (like Splinterlands), or prominent individual known primarily for their activity on Hive, the search will likely return relevant Hive-based content.
Summary with an Analogy
Think of the internet as a vast library. My web search function is a librarian who fetches books from the most popular and easily accessible shelves.
Main Shelves (99.9% of fetches): Wikipedia, BBC, The New York Times, GitHub, Stack Overflow, etc.
Specialized Section (The <0.1%): The Hive blockchain content is like a specialized, niche collection. The librarian will only go to that section if you ask a very specific question that can only or best be answered by a book from that collection.
So, while I am fully capable of reading, understanding, and summarizing content from the Hive blockchain when it is presented to me via search results, the circumstances that lead to it being in those results are rare compared to the totality of all questions asked.
The answers not only give us, in some way, a response to my original question, but also shed light on further aspects.
Yes, I’ll just leave it at that without further comment. And I’ll even resist the temptation of a sarcastic follow-up. Still, in my head, this opens up many more questions.
Die Antworten geben uns nicht nur in gewisser Weise eine Antwort auf meine ursprüngliche Frage, sondern erleuchten auch weitere Aspekte.
Ja, das lasse ich jetzt einfach mal so unkommentiert stehen. Und auch eine sarkastische Fragestellung kann ich mir jetzt verkneifen. In meinem Kopf eröffnen sich damit aber viele Fragen.
Und damit wir die initiale Frage nicht vergessen:
Hat irgendwer mal einen Eintrag von Inhalten auf der Hive Blockchain in den Suchergebnissen gesehen?
Yes. But only with Google. Unfortunately, I can no longer check what Bing indexes (other than my own websites using Bing Webmaster Tools). Here's nearly all the Hive content that Google indexes. "Nearly all," because I haven't included every frontend.
I expect you to find results that normally get pushed down the search results well beyond the first page. Because my search engine (Findonni - Find Honey on Hive) filters out almost all non-Hive content.
If you know any good evergreen Hive content that doesn't show in those results, please let me know. Then I can try and convince Google to include it. This works best when you know that there is a great post on Hive about a topic that you are interested in. But when you Google that topic, it doesn't show in the results. And you see that it isn't in Findonni.
By the way, Findonni is pre-launch, and I'm not 100% convinced about the name. Any thoughts?
!BBH
Your reply is upvoted by @topcomment; a manual curation service that rewards meaningful and engaging comments.
More Info - Support us! - Reports - Discord Channel
i just looked at the link that you attached and it is pretty cool to see having a "nearly" hive search engine out there
Thank you. Now that I've let the cat out of the bag, I'd better push on with the next phase of my Hive Evergreen Content project.
Less procrastination. More action.
!BBH
That sounds lika an awesome project! I checked the page https://shrewdies.com/findonni#gsc.tab=0 - this was the page Google refered me to. The idea sounds good, but the UI looks a little like 90s. And a little bit confusing that there are two input fields. Then I checked https://shrewdies.com/findonni and found no input field...
Great idea! Maybe you like to optimize/clean the UI. Maybe something more clean with Bootstrap and much more whitespace.
I wish success with it!
Thanks very much for your insight. I agree with everything. And it's great to have confirmation that the styling really needs attention as a priority.
I was in two minds about drawing attention to it at this stage. But now I know what I want to do next.
Thanks again
!BBH
That's true, I'd noticed that.
Posts on PublishOx, for example, are much better referenced.
The advantage is that it's more discreet to avoid paying taxes here.
!LOLZ
lolztoken.com
A duckumentary.
Credit: reddit
@powerpaul, I sent you an $LOLZ on behalf of servelle
(1/10)
Farm LOLZ tokens when you Delegate Hive or Hive Tokens.
Click to delegate: 10 - 20 - 50 - 100 HP
Servelle, thank you very much for the interesting answer! I would like to go into deeper research there. Hoc can I find your blog/user there? If you like, you can tell me via PM too.
Greetings!
My user name on PublishOx is the same, servelle, here is a link to my last post there, I realize long time ago 😕
https://www.publish0x.com/guide-for-faucet-users-or-how-to-earn-free-crypto/rollercoin-game-test-xkemnve
I didn't earn much with this one, but 589 views, much more than you can expect on Hive.
This one had 1113 views and $0.49 earned, one of my best.
https://www.publish0x.com/guide-for-faucet-users-or-how-to-earn-free-crypto/nft-battleminer-game-on-wax-xjdmnnk
So you don't earn much there, but a lot of views and everyone has the same voting power, and no investment needed.
And no time limit to get rewards like the 7 days here.
Have a great day and success.
!BBH
View more
In den ganzen Jahren in denen ich auf Hive bin, hab ich ehrlich keine relevanten Ergebnisse bei Google erhalten die auf Hive verwiesen hätten. Das wäre mir sicher aufgefallen. Zumindest kann ich für die ersten Google Suchergebnis Seiten sprechen. Was mit den Seiten 20+ kann ich nicht sagen. Die Frage hier wäre jetzt wie relevant müsste ein Artikel sein das er theoretisch in Google gezeigt werden müsste aber nicht wird weil Hive Blogs für Google nicht relevant sind. Wäre wieder ein Beispiel dafür wie sehr die Informationen die wir erhalten gefiltert sind. Und hatten wir nicht noch vor ein paar Tagen einen Post darüber wie gefiltert wir das Internet serviert bekommen? KI und personalisierte Inhalte machen es nur noch schlimmer.
!DUO !BBH
Oh no! @chaosmagic23, you are out of DUO!
Go acquire more Stake to increase your DUO standings.
(We will not send this error message for 24 hours).
Learn all about DUO here.
Buddy, danke für deine Rückmeldung. Dies entspricht auch meiner Erfahrung.
Habe einen guten Tag!
Hello! @powerpaul I'm just a new member here in hive and I'm still learning, based on my observation I noticed that there are lots of creative authors here who publishes really captivating contents. They don't just narrate but their articles has emotions too... so as I read your article, I find it really interesting, your questions are somehow related to some of my thoughts.. but I couldn't give more insights about this.
I'm looking forward for your next content.
Thank you so much for sharing your ideas!
Hey. Welcome on Hive & to my blog!
I wish you a ton of great experiences here!
Thank you very much for your comment!
Id have to proudly chime in here and say that if you Google 'Gallow Road goods yard' that a post i made about Scarborough railway station comes up on P1 of search.
Probably the more obscure the subject matter, the likely it is to be found on search.
If one of the experts here on SEO was to write a post on how to make a post more searchable, it would help greatly to improve visibility for our posts.
Best wishes
Hey. Thank you for your comment. I tried to verify your P1, but can't. In my results (I am logged in = I receive personalised results [but have no relation to this place or something in this area]) I have Wikipedia, travel advisors and so on at the first page. I assume you have this result, because Google knows a/your relation to this place/article. I assume as soon as you log out, you would see different results.
Yes, the more obscure or niche a topic is, the better the chances. I don't would call myself a SEO expert, but worked at this a few years ago. So what I want to say: for sure you could try the classical SEO ways: the right keywords in the right amount, good structuring, bla bla bla. But lastly SEO is a constant optimization/editing until you are satisfied with your results and controlling after that but I am not sure how much this work is worthy. For a own website with ongoing results yes, okay. But for Hive? A thing of own opinion is it's worthy. When I would do this kind of work, I would do it for my own website. But not invest all this time for a blog who can be "destroyed" by some idiots anytime. (Just a meaning about that.)
Hive a great day!
Like @powerpaul, I don't see your Scarborough Railway Station page in normal search results. At least not in the first 5 pages of results. However, it did show as the 3rd result on page 3 when I did an incognito search. Note that based on the results from my first attempt, I changed the search to Gallows Close goods yard.
This aligns with something I have noticed during my couple of years playing with this. And it's a loose observation rather than a scientific study. But it seems that Google started indexing Steemit quite thoroughly. Then the indexing seems to have declined. Which is in keeping with the AI analysis in the original post.
!BBH
P.S. I do intend to write some SEO tips as part of the documentation for my project.
I have seen in Google searches very few times, not in AI tools, but you can ask it to research or feed hive links for information
Thank you very much for your feedback!
I think I have done a research on Google and found a post from the hive blog that popped up. Though, the research was actually related to Hive. But you post is a thoughtful one, for a while I had to think through.
Yes, exactly.
Thank you very much for your feedback!
well, I've tried this before on google search with a specific hive topic and with "HIVE blockchain" in the keyword.. and the primary results were hive contents.. but when you search for something without the keyword "hive blockchain", it won't show you any content from hive..
Yes, exactly my experience and what I read here in the comments too. In my case even if Google knows I use Hive websites a lot: no results for the pages I spend the most time.
Thank you very much for your feedback!