[Proposal] Implement DT1 algorithm for DT2 members

[dragonvslinux]

Taken from trust changes announcement, January 2019:

As a special exception to the normal algorithm for determining a user's trust network, if you are on the default trust list ("DT1") but more other DT1 members distrust you than explicitly trust you, then it is as if you are distrusted by the default trust list for all purposes except for this very DT1-composition determination.

So if someone on DT1 is doing something stupid, you can ask other DT1 members to distrust them.

Therefore the following suggestion is to make DT2 as accountable as DT1 using the same algorithm with the same reasoning:

• If you are on the second default trust list ("DT2") but more other DT2 members distrust you than explicitly trust you, then it is as if you are distrusted by the default trust list.
• So if someone on DT2 is doing something stupid, you can ask other DT2 members to distrust them.

Apologies if this has already been referenced elsewhere, or is already the case.

Edit, suggested reqreuiments for DT2 exclusion:

• Must be distrusted more than trusted by DT2 members
• Must have distrust greater than 2 DT2 members
• Distrusting DT2 members must be trusted by different DT1 members(*)

(*) This would resolve the sockpuppet problem

[1714045557]

1714045557

[1714045557]

1714045557

[1714045557]

1714045557

[suchmoon]

Apologies if this has already been referenced elsewhere, or is already the case.

This might be prone to abuse. A malicious DT1 member right now can't unilaterally kick someone out from DT2, they need one or more other DT1 members to exclude a DT2 member. In your proposed scenario a DT1 member could keep adding sockpuppets (or just random users who happen to exclude the targeted DT2 member) until they reach their goal.

[dragonvslinux]

Apologies if this has already been referenced elsewhere, or is already the case.

This might be prone to abuse. A malicious DT1 member right now can't unilaterally kick someone out from DT2, they need one or more other DT1 members to exclude a DT2 member. In your proposed scenario a DT1 member could keep adding sockpuppets (or just random users who happen to exclude the targeted DT2 member) until they reach their goal.

Fair point, though I think the issue you've described here would be the problem with untrustworthy DT1 members rather than anything else. For this reason and your argument, I believe this would be good to additionally "root out" any untrustworthy DT1 members who are abusing their power at the same time. You do have a point though, I'm not denying that.

A solution to this could be an addition to this algorithm that DT2 members (who are as a "majority" excluding a DT2 member) must be trusted from different DT1 members. This would then require a conspiracy within DT1 to exclude a DT2 member, which would be less of a conspiracy and more of a co-ordinated effort.

[o_e_l_e_o]

I'm not sure how this would work, in practice. Let's say you are included by 3 DT1 members and 5 DT2 members, but you are also excluded by 10 DT2 members. At DT1 you have +3 and so are included. But at DT2 you would have -5 and so would be excluded. Overall between DT1 and DT2 you have -2, so although you are net included by DT1 (+3), are DT2 now able to overrule DT1?

I think this over-complicates things. If someone on DT2 is doing something stupid, simply appeal to their DT1 "sponsor(s)", and if that doesn't work, open a thread to appeal to other DT1 members to exclude them.

A solution to this could be an addition to this algorithm that DT2 members (who are as a "majority" excluding a DT2 member) must be trusted from different DT1 members.

I am a big proponent of DT2 members requiring inclusions from 2 different DT1 members prior to becoming DT2 at all.

[DiamondCardz]

There is nothing wrong with how DefaultTrust works at the moment in my opinion. The fact that DT1 is now accountable means that someone who is adding members to DT2 who should not be added can be veto'ed out of DT1 by users with sufficient merit. I think that system works completely fine.

Thanks for mentioning DefaultTrust though, it reminded me to go through and fix up my trust list

re: what o_e_l_e_o said: I like that idea but that would massively shrink the size of DT2. I guess that's not necessarily a bad thing?

[o_e_l_e_o]

re: what o_e_l_e_o said: I like that idea but that would massively shrink the size of DT2. I guess that's not necessarily a bad thing?

Loyce crunched some numbers on that a few months ago. At the time, it would have shrunk the list of DT2 from 229 to 63.

63 was probably a bit on the low side, considering we have 100 DT1s (or usually around 95 if you exclude the handful who are excluded). However, given the DT2 list has now almost doubled to 434, if the ratios were to stay the same (I'm not saying they would, but as a ballpark) we would be looking at around 110-120 DT2 users, which I think is entirely reasonable.

I think you are far less likely to have someone who is trusted by 2 DT1s "doing something stupid" than someone who has a single inclusion, as per OP's concerns. It also gives you two users rather than one you can appeal to if they do - they only have to lose one of their inclusions to drop off of DT2.

[DiamondCardz]

re: what o_e_l_e_o said: I like that idea but that would massively shrink the size of DT2. I guess that's not necessarily a bad thing?

Loyce crunched some numbers on that a few months ago. At the time, it would have shrunk the list of DT2 from 229 to 63.

63 was probably a bit on the low side, considering we have 100 DT1s (or usually around 95 if you exclude the handful who are excluded). However, given the DT2 list has now almost doubled at 434, if the ratios were to stay the same (I'm not saying they would, but as a ballpark) we would be looking at around 110-120 DT2 users, which I think is entirely reasonable.

I think you are far less likely to have someone who is trusted by 2 DT1s "doing something stupid" than someone who has a single inclusion, as per OP's concerns. It also gives you two users rather than one you can appeal to if they do - they only have to lose one of their inclusions to drop off of DT2.

I think that's a reasonable enough idea then. Rather than having it just apply to DT2, maybe add in an option to change the minimum number of trusts required for someone a depth below to appear on your trust list, and set the default to 2 - so by default you would trust 2 depths below your trust list, and for someone to appear a depth lower on your trust list they need to be trusted by two people a depth higher. Could be called 'trickle-down factor' or something similar.

I have less of a problem with DefaultTrust right now however than I do with trust ratings being used for disagreements and petty things. There are a few people on DT1/2 who use negative trust against people just because they spouted some opinions they didn't like, and pad their reasoning with some words about how "this makes them untrustworthy". It's very blatant mis-use of the system IMO.

[Macadonian]

I have less of a problem with DefaultTrust right now however than I do with trust ratings being used for disagreements and petty things. There are a few people on DT1/2 who use negative trust against people just because they spouted some opinions they didn't like, and pad their reasoning with some words about how "this makes them untrustworthy". It's very blatant mis-use of the system IMO.

Whatever trust system is in place this will always be the down fall of it. Trust and petty disputes go hand in hand and will the trust system will always be misused. The only solution I see to this is a trust system is not accepted until its been approved by a majority of people. So if you were to put a trust rating on me "This dude is cool" then similar to how flags work members could either support it or oppose it for being a good reason to leave trust. If the majority of people oppose it then it would go under untrusted feedback and if enough members support it will go in the trusted feedback. The support/oppose weighting would be relative where that member is in the DefaultTrust list. So DT1 would be higher weighting than that of DT3

[LoyceV]

I don't think this is ever going to happen. The entire basis of the DefaultTrust system is that each depth is chosen by a higher depth. As theymos wrote: it's a "special exception" that was introduced instead of having theymos hand-pick all DT1-members, and I don't think this should be introduced on all DT-depths.

Do you have a particular user in mind for this suggestion? And if so, have you checked how your suggestion would influence the DT-status for that user? If it's Lauda (I'm assuming this based on your recent negative feedback and Flag), you may be interested in reading this and this.

re: what o_e_l_e_o said: I like that idea but that would massively shrink the size of DT2. I guess that's not necessarily a bad thing?

Loyce crunched some numbers on that a few months ago. At the time, it would have shrunk the list of DT2 from 229 to 63.

Actually, it would have gone from 378 to 149, based on the link you posted. It's still a significant reduction and would mean more concensus and less individual power.

I've made suggestions before, to improve DT:

I've posted a few suggestions before, it won't fix everything, but I think it would make DT better if:
1. Each DT2-member needs at least 2 inclusions from DT1 (this would cut the number of DT2-members in half and make "Selfscratching" impossible without collusion)
2. Each DT2-member must have been active and posting in the past 6 months (this gets rid of abandoned accounts on DT2)
3. Getting banned should instantly remove a user from DT1/2, and their DT1-voting-power should be removed.

[dragonvslinux]

Let's say you are included by 3 DT1 members and 5 DT2 members, but you are also excluded by 10 DT2 members. At DT1 you have +3 and so are included. But at DT2 you would have -5 and so would be excluded. Overall between DT1 and DT2 you have -2, so although you are net included by DT1 (+3), are DT2 now able to overrule DT1?

Technically it would be DT2 overuling DT2, DT2 would still not have any influence over DT1 in the direct ruling sense.
I otherwise do agree that if you are overall distrusted by DT1 & DT2, you shouldn't be either.

I don't think this is ever going to happen. The entire basis of the DefaultTrust system is that each depth is chosen by a higher depth.

With this proposal, DT2 depth would still be chosen by DT1. The difference here would be DT2 members ability to exclude their own. I am not suggesting to change the selection process directly, but the exclusion process, even if they are interlinked. Similar to how Theymos overules DT1, but ultimately DT1 have the ability to exclude their own. The proposal is a trickle down theory of the current DT1 selection and exclusion process, allowing certain power to continue to roll down hill: that of exclusion.

Do you have a particular user in mind for this suggestion? And if so, have you checked how your suggestion would influence the DT-status for that user?

Nope nobody in mind, but you're right to check how it would influence DT2 members. Admittedly the original theory would be completely flawed, as I'd overlooked that many DT2 members aren't trusted by other DT2 members, therefore a single DT2 member would be able to kick out another which would be no good.

At minimum a DT2 exclusion algorithm would need to be something like distrusted>trusted with distrusted>2 or higher. So it would require a minimum of two DT2 members (or more) to exclude another, as opposed to a single member. From the sample of 200 of 463 DT2 members this wouldn't exclude anyone I can see, putting 0/no scored users closest to exclusion (by default) than any others. This was mainly what I was thinking of, another form of horizontal accountability that wouldn't immediately change anyone's trust depth, but would instead help to solve the possibility of DT1 using sockpuppets to exclude DT2 members, or the theoretical problem of DT1 including DT2 sockpuppets, that I hadn't even considered properly. Hence why the factor of 2 members required for inclusion/exclusion generally makes sense, in order to increase trust accountability.

I can otherwise see why you came to the following conclusion:

I've posted a few suggestions before, it won't fix everything, but I think it would make DT better if:
1. Each DT2-member needs at least 2 inclusions from DT1 (this would cut the number of DT2-members in half and make "Selfscratching" impossible without collusion)
2. Each DT2-member must have been active and posting in the past 6 months (this gets rid of abandoned accounts on DT2)
3. Getting banned should instantly remove a user from DT1/2, and their DT1-voting-power should be removed.

Although somewhat unrelated as I was discussing exclusion of DT2 as opposed to the process of inclusion of DT2...
The main suggestion I would have is if you wanted to cut the size of DT2 without doing so as drastically:

1.Each DT2-member needs at least 2 inclusions from DT1 or a combination of DT1 and DT2.

[The Cryptovator]

If implement your proposals then what will difference between DT1 and DT2? If a DT2 can't include you DT2 then how he will exclude you. And if DT2 voting is enough to become DT2 then what is the necessary of DT1 system. DT2 controlled by DT1, so if anyone from there make some stupid then ask DT1 to exclude him. Is it not simple? On current system you will no longer in DT network if you made something stupid. If you give exclusion/inclusion power to DT2 then who will prevent abuse. Even you will find some DT1 who have used sockpuppets to become DT1.  Perhaps a war will start if you DT2 network when they will get exclusion/inclusion power. Let it run as usual same like current system.

[o_e_l_e_o]

Technically it would be DT2 overuling DT2, DT2 would still not have any influence over DT1 in the direct ruling sense.

What about the scenario where someone has a higher net DT1 inclusion than their net DT2 exclusions? Would we just ignore the DT2 exclusions in that case? Otherwise if we exclude them, then DT2 have overruled DT1.

but would instead help to solve the possibility of DT1 using sockpuppets to exclude DT2 members, or the theoretical problem of DT1 including DT2 sockpuppets, that I hadn't even considered properly. Hence why the factor of 2 members required for inclusion/exclusion generally makes sense, in order to increase trust accountability.

Requiring two or more DT2 members does little to nothing to address the problem of sockpuppets. A DT1 user can make as many DT2 users as he likes. One malicious DT1 user could quite easily create/find multiple accounts to add to DT2 to achieve their desired result.

Requiring two or more DT1 members to become DT2 in the first place is a better solution, as it is far harder to game the DT1 selection process.

[TECSHARE]

Apologies if this has already been referenced elsewhere, or is already the case.

This might be prone to abuse. A malicious DT1 member right now can't unilaterally kick someone out from DT2, they need one or more other DT1 members to exclude a DT2 member. In your proposed scenario a DT1 member could keep adding sockpuppets (or just random users who happen to exclude the targeted DT2 member) until they reach their goal.

That sounds familiar...

[dragonvslinux]

Requiring two or more DT2 members does little to nothing to address the problem of sockpuppets. A DT1 user can make as many DT2 users as he likes. One malicious DT1 user could quite easily create/find multiple accounts to add to DT2 to achieve their desired result.

Thanks for pointing that out, I was meant to include the following requirement for DT2 exclusion too but forgot:

• Must be distrusted more than trusted by DT2 members
• Must have distrust greater than 2 DT2 members
• Distrusting DT2 members must be trusted by different DT1 members

This would resolve the sockpuppet problem unless I'm agian mistaken. OP updated again.

If a DT2 can't include you DT2 then how he will exclude you.

Same as a DT1 member can't include another DT1 member, but can (collectively) exclude them.

And if DT2 voting is enough to become DT2 [...]

Again, this is not what I said. To avoid regurgitation:

With this proposal, DT2 depth would still be chosen by DT1.