Hivemind API being updated to allow following mute lists

in Hive Improvement4 years ago (edited)

In my previous post on decentralized blacklists, several people mentioned they would also like the ability to follow another account’s mute list, something I was planning to do eventually.

A user currently has the ability to “mute” accounts, so posts and comments for that account don’t show up when the user is browsing. But they don’t have the ability to follow an existing mute list that someone else has created.

But it turns out we need this feature sooner rather than later, as I later learned that there already exists a global mute list for handling accounts that do massive comment spamming, and that list should also be replaced by an opt-in mute lists using the “follow API” method, to create a level playing field.

This means that instead of one group of people controlling who gets muted across the Hive web sites, users will be able to make their own choices about who has the best mute lists (or if they just want to use their own list).

Also, another important benefit of this change is it will improve performance of many Hive API calls, since it will speed up the processing of any API calls which are affected by muting.

The rest of this post is aimed at programmers who want to make use of the new API functions, so non-programmers probably don’t want to read the post past this point...but please feel free to comment on the functionality itself below the post.

Updates to the “follow” API method to support following a mute list

Note: There’s already a post that describes our initial changes to the follow API, and that’s probably worth a review before reading the latest changes below.

To support following a mute list, we updated the allowed values for the “what” parameter:

blog ignore blacklist unblacklist follow_blacklist unfollow_blacklist follow_muted unfollow_muted

The last two values allow for following and unfollowing another account’s mute list.

Follow API method updated to allow adding multiple accounts using array syntax.

Previously the following parameter only took a single parameter, now you can optionally pass an array of account names instead. For example,

["follow",{"follower":"jes2850","following":["netuoso","blocktrades"],"what":["unfollow_muted"]}]

would make jes2850 stop following the mute lists of both netuoso and blocktrades, instead of requiring two separate calls.

This functionality is particularly useful for user interfaces that want to enable a user to do batch following, muting, or blacklisting, as it can dramatically reduce the number of API calls required.

New API call to get paginated account names from a follow list

When a UI wants to display the list of accounts for a user, it can use the new API call:

bridge.get_follow_list(observer, follow_type, starting_account, limit)

where

  • observer is the account to get the list for
  • follow_type is one of blacklisted follow_blacklist muted follow_muted
  • starting_account and limit are used for pagination in the standard way

Condenser (e.g. https://hive.blog) being updated to support the new follow API-level

We’re also doing a major overhaul to condenser, both to allow for advanced display and filtering of follow lists and to allow for batch entry of accounts to follow/blacklist/mute. One of the goals of the update is to unify the way all the different types of follow lists are displayed/edited as much as possible, to keep usage as simple as possible.

Sort:  

This is a critical asset in phishing mitigation considering we keep our phishing list in mutes on Hive.

very true

says the phishing scammer.

This sounds great!

Just a question here, how do individuals stop the bot promotions in the discord and wallets?

@kryptokaden can the @telokanda effort do this?

We are a legitimate charity and I am not interested to know that I have won a 30btc whatever if I reply to this or that in a limited time.

Within the hive wallet, you can mute such accounts. Within discord, it's more difficult as they can easily create free accounts to spam you, so it becomes difficult to block such messages. @gtg has recently updated the chat software to enforce the chat accounts to be tied to Hive accounts, so using that may be a solution to the problem for you (discussed at the bottom of this post: https://hive.blog/hive-160391/@gtg/witness-update-brand-new-hive-eclipse-is-coming).

Thank you kindly for this advice.
It clearly reflects your care for the people and the things called Hive 😂

very good update... you guys are all doing a great job at running this block chain. the mute list is a wonderful idea but what if a person decides to change his ways after being muted?

The current global mute lists are only for people doing really bad things: phishing links that do things like try to steal people's keys and spammers that are continuously posting comments using a bot. These aren't people that are likely to change.

But if they do, they can always try to get removed from a mute list, or even try to convince people to stop using the mute list that has their account on it. And of course, there's always the option to create a new account and behave better on that new account.

That's nice...Good job on keeping the community sane

This is optional right? I don't plan on using any mute list at all. It rings of censorship to me and lately I've been extremely sensitive and wary about what has been going on with the liberal media and social sites.

Yes, it is completely optional.

But it's worth mentioning that the phishing mute lists will mainly be used to hide posts from accounts that put dangerous links on their posts (e.g. links to web sites with malware).

So unless you are a careful person that never follows external links, it will probably be worth following at least a phishing mute list. Such lists are not political, they are just aimed at protecting inexperienced users from real criminals.

As long as it's not political, i.e. anti-conservative bias, then I might follow a list.

Will you be able to subscribe to multiple mute lists? I believe this will be important.

Will the current irredeemables be a default list? This list is a very fine tuned list to extreme cases of abuse and is used delicately.

I assume since this is HiveMind it will carry over regardless of the front end used?

Yes, you'll be able to subscribe to as many mute lists as you want. It works just like following blacklists.

When a user logs into hive.blog, the UI will prompt them to make choices about blacklists and mute lists they want to use. If they don't want to use any of them, they can follow the null account's lists (we plan to use that to avoid prompting them again).

Each condenser instance will be configurable, so that the host's operator can make suggestions about which lists a user may want to follow. For hive.blog, we will have the irredeemables as one of the choices for a mute list to follow.

List following is a hivemind-level feature, so list following effects apply across all front ends.

A huge hug 🤗 and a little bit of !BEER 🍻 from @amico!


Un caro abbraccio 🤗 e un po' di BEER 🍻 da @amico!

Thank you for the information @blocktrades. I want everyone to know about this information.

There is a new and very unique dapp on hive, which got my attention and I think I should bring this to your attention also
It is called @telokanda

When I read their whitepaper it was full of insights and they are building an ad revenue system that rewards people with KANDA tokens for clicking on ads.
https://nitrous.telokanda.com/hive-182425/@telokanda/whitepaper-for-the-telokanda-network

I think if thy have acess a huge delegation that would make them stay, they would build most of the projects on their WhitePaper on HIVE
Think about reaching out to them to help, they need it. I am currently suporting them in a way that I can at https://nitrous.telokanda.com


Hey @blocktrades, here is a little bit of BEER from @amico for you. Enjoy it!

Learn how to earn FREE BEER each day by staking your BEER.

Congratulations @blocktrades! You have completed the following achievement on the Hive blockchain and have been rewarded with new badge(s) :

You published more than 150 posts. Your next target is to reach 200 posts.

You can view your badges on your board And compare to others on the Ranking
If you no longer want to receive notifications, reply to this comment with the word STOP

Do not miss the last post from @hivebuzz:

The Hive community is in mourning. Farewell @lizziesworld!
Support the HiveBuzz project. Vote for our proposal!

Everyone gat different choices to made but we all hope for the best.

Thanks for the information @blocktrades, you are doing well

Excellent work @blocktrades... Interesting topic I will look for some information about that. Thank you.

Excelente trabajo @blocktrades… Interesante tema buscare algo de información sobre eso. Gracias.

Nice ....
Hi

MUTE??? How big is the effect if ever happened? But what if the person decides to change? It made me think a little bit. (^_^)

Curated for #informationwar (by @aagabriel)

  • Our purpose is to encourage posts discussing Information War, Propaganda, Disinformation, and Liberty. We are a peaceful and non-violent movement that sees information as being held back by corrupt forces in the private sector and government. Our Mission.
  • Discord, website, youtube channel links here.

Delegate to the @informationwar! project and get rewarded

Hello @blocktrades, as far as I can see I have a lot to learn from hive.

This will be a useful feature, however there is likely a lot of overlap between people who are vulnerable to phishing and those who will not know about or not bother to opt-in to community curated mute lists.

Hi! I'm new to the community, I think you're right in what you're saying and you speak the truth!

Hi @blocktrades - I don't know if you remember, but we spoke a little about a web of trust system back in Poland in 2018, and it is something that I think about from time to time.

With following blacklists and mute lists, would it also be useful to have a "follow whitelist" option, where users could whitelist the accounts they follow in some way. Would then this create a matrix of trusted and untrusted accounts that will provide a granular account confidence level of increasing sophistication and precision (of some kind). Trusted accounts would have a greater impact on the trust of other accounts but they can also be labelled untrstworthy in a type of social consensus algorithm and drop down the rankings, pulling their "recommendations" down also, unless their recommendations are propped up by other trustworthy accounts.

I don't know if I am explaining this well, but I think you will get the idea at least and process it fast enough as viable or useless :)

Thanks for the recent updates and I am looking forward to hearing more on HF24 - Enjoy your week mate.

I remember it well, and my decision to work on Hive was in large part because I decided that Hive was the perfect blockchain to build a web-of-trust system of the type we talked about back then. I believe the applications for such a "trust computation" system are nearly endless, and will have radical implications for productive organization of human endeavors.

And you're also not the first to see an analog between what we're doing with decentralized lists and the trust computation system I've described in the past. One of the guys here actually suspected that this was my stealth way of introducing the trust computation system on Hive :-)

But in truth, the decentralized list implementation was just introduced to solve immediate problems with Hivemind performance and centralization issues. Nonetheless, the nature of the solution was no doubt influenced by my thoughts on web-of-trust based systems and we may gain some useful real world information related to web-of-trust systems from its deployment on Hive.

And your idea for whitelists could certainly extend the abilities of the system to start serving as an early web-of-trust prototype. We could even potentially go further and allow for users to create new types of lists. For example, someone could create a "good programmers I know" list and others users could then make "good programmers I know" lists of their own. The complete trust computation system I envision eventually is much more nuanced than this, but it might be worth deploying a very early prototype just to get ideas for the future.

Nonetheless, the nature of the solution was no doubt influenced by my thoughts on web-of-trust based systems and we may gain some useful real world information related to web-of-trust systems from its deployment on Hive.

I am looking forward to seeing what kind of data can be leveraged to add to the trust conversation, as it is a very important one for Hive with much larger ramifications.

We could even potentially go further and allow for users to create new types of lists.

From just an enduser perspective, it would be a nice way to build "playlists" that could be used to enhance all kinds of interactions and relationships.

The complete trust computation system I envision eventually is much more nuanced than this, but it might be worth deploying a very early prototype just to get ideas for the future.

No doubt, but a few little steps to see what can be learned may prove invaluable in the larger steps to follow.

Thanks for the info