GitHub - heliaxdev/namada-shielded-exped...

[icy spindle]

When I run this, will it show the scores that nebb is currently showing? @wild tartan

[grizzled glacier]

What command did you use?

[icy spindle]

cargo run --release -- and flags

I don't know the upgrade epochs etc, so I'll wait for the instructions

[scenic stratus]

Score extractor code is opensource, it compiles and everything, but it requires a connection to a database with the indexer data to recompute the scores

so, we opensourced the code for people to see how their are computed

people had questions on "how is uptime computed" and so

now you can check

[wild tartan]

@wicked nymph

[icy spindle]

does nebb currently contain the final a/b/c class results? @wild tartan

can we start the celebrations?

[wicked nymph]

it's far too complicated to overlook in any coherent fashion imo. no disrespect but without documentation. a view of the actually compiled sql queries used for scoring would be much more useful

the point is the format is completely inaccessible

[wicked nymph]

a few concrete questions: does it handle the relative block numbers, or does it need only indexed up to right block in db? does it run on an out-of-box Namadexer install or does it need any custom configuration?

[violet sundial]

My knowledge of SQL is like an infants 👶

[wicked nymph]

mine is pretty good and I did all my scorings in sql, in this app it's abstracted in a query format that makes it kinda hard to see exactly which data is pulled from one place only

[violet sundial]

Got ya.

Its basically just for transparency then atm the code. Not for us to compete anything since we don't have access to this database from my understanding huh?

Which kinda is a semi transparent approach since it depends on the data.

[icy spindle]

brothers, let's make sure of this first 😭

[wicked nymph]

hence my questions if it runs on a Namadexer out of box. but doubt I will get it set up. A scoring file would be better at this time. Or maybe the plainly compiled SQL statements the app generates

[icy spindle]

as far as I understand it; the 'database' is a synchronized indexer. to calculate the scores we need to set up an indexer with memo decoding capability (newer versions of namadexer)

[violet sundial]

Oh yeah true. Gavin is this already included? Also asked this in validator circle but got read over. Or I didn't pay enough attention.

[violet sundial]

Aah got it, got it.

[wicked nymph]

I have this running already

[violet sundial]

I'll get behind the pc shortly. See if I can make sense out of it. Stupid of me btw, two days ago I removed my indexer.

[wicked nymph]

used that to calculate my unofficial scoring files

[violet sundial]

Think it needs a local db though, but will check the code.

[icy spindle]

cargo run -- --database-url postgresql://myuser:pass@localhost:5432/mydb --cometbft-url https://rpc.namascan.com --upgrade-proposer tnam1qqsa3segx4jfe4zdkkj0q5md5e9jffv0hu6kknc0 --v0-to-v1-upgrade-epoch 15 --v1-to-v2-upgrade-epoch 20 --sleep-duration 0qs

[wicked nymph]

could provide a db dump or someth if anyone needs

[scenic stratus]

the indexer is not the namada indexer

[icy spindle]

then this command should run the score calculator

[violet sundial]

You see how genius this guy is?

[scenic stratus]

we planned to used namadaxer but it didn't support all the queries we had to do

[violet sundial]

Ah damn.

[icy spindle]

does nebb currently contain the final a/b/c class results?

[wicked nymph]

ok well how do we run this then?

[scenic stratus]

why do u want to run it?

[wicked nymph]

btw I did all queries I needed to calculate scoring on namadexer

[scenic stratus]

also I have some sql knowledge

[wicked nymph]

was not the point of providing the source that we could check the scorings?

[scenic stratus]

and I made most of the queries

thats the code used to compute the scoring

[wicked nymph]

ultimately, a scoring file is sufficient. but that would be great in terms of checking against my own data tables

[scenic stratus]

if you need to recompute them its different

we have several microservices

to index stuff from the chain

[icy spindle]

so does nebb currently contain the final a/b/c class results?

[scenic stratus]

what does this mean?

if the nebb is up to date?

[icy spindle]

yes

[wicked nymph]

my concern rn is as best I can tell the scoring data on nebb must be inaccurate. but happy to dialogue on this. I have some unofficial scoring files I created on my own from the dexer

[scenic stratus]

it is

[icy spindle]

OMG

[wicked nymph]

just to confirm, does nebb include any s-class atm?

[scenic stratus]

do u have this data anywhere?

[scenic stratus]

yes

[wicked nymph]

yes, can I dm?

[scenic stratus]

put them here

[wicked nymph]

https://testnet.one/scoring

https://testnet.one/uptime

it's partial but some of them seem off re the numbers on nebb, particularly re governance for instance

can provide the SQL queries used to generate as well if requested

but fairly sure they are accurate

[scenic stratus]

for the governance voting

lets say a proposal was sent with no memo

but you vote for that proposal

and you add a memo

do u count it?

like, do u count the vote?

[wicked nymph]

I count all votes that have memo in the vote yes

[scenic stratus]

we dont

[violet sundial]

Aah that's an interesting edge case

[wicked nymph]

and only those with valid pubkey corresponding

[wicked nymph]

why not?

[scenic stratus]

because the proposal was not part of the competition

[violet sundial]

So an init proposal that has no memo is an invalid proposal

[scenic stratus]

if the proposal has no memo -> not part of the competition

[wicked nymph]

that doesn't make any sense to me tbh

[scenic stratus]

well we can argue about that

[wicked nymph]

you have a list of which proposals you consider valid?

[scenic stratus]

not on this laptop

[icy spindle]

when do you plan to add s class tasks to nebb?

[wicked nymph]

but again, if scoring files were provided, it would be easier to get an overview of how many got xyz category and such

[icy spindle]

[ #1 ] tpknam1qz36mzdvdmxvvcpv3c36zzs5v369jauws0tzrxesuaexy0p25r5sy0jsuac | 30780630243.73121
[ #2 ] tpknam1qrflwggnn9g4wp6p8u63nvte6y7nf9xr2cwqsvltgvauxl9s4nmw7476fnl | 23703793716.73121
[ #3 ] tpknam1qr37yj7d9al5yerzn68kn7gtttfh25fqk45wwwvmt8snwgp5zwrzvjfujne | 20111709827.73121
[ #4 ] tpknam1qpall8pt60ek5zgeyxstdkhrp34rrqj67a2jrklrl9sddln0grq87nluum5 | 19384139813.54309```

[scenic stratus]

we did already

[scenic stratus]

which scoring files???

[icy spindle]

what?

[wicked nymph]

nebb hasn't really changed for a week or two?

so how is that possible?

[scenic stratus]

nebb was updated some time ago with the S-class submission and the latest data from the chain

[wicked nymph]

a complete file in some formate showing which categories were awarded to which players

[supple drift]

New spreadsheet after reviewing by Spork they mean?

[wicked nymph]

I understood from gavin the idea was such would be provided in some format?

[scenic stratus]

we are waiting to remove some players

[wicked nymph]

at that time the s-class submits weren't finished though?

[scenic stratus]

as they are part of some sybil cluster

and then we will provide the data

[wicked nymph]

sounds good

[icy spindle]

Is it possible to see the results of only A/B/C class tasks?

[supple drift]

Can we have final results today?

[wicked nymph]

meanwhile I should get working on a revised sql query for governance apparently

however I really do think it's strange to only count memo-generated proposals

[scenic stratus]

i have no idea what u mean with A/B/C

[wicked nymph]

as per scoring tables in shielded expedition. A-class, B-class and so on

[violet sundial]

So everything except S classes.

[scenic stratus]

I think I can manage to get them

[violet sundial]

But this still needs calculations.

Probably excluding uptime and gov voting then?

[wicked nymph]

what would be great is a way to look up on player basis "what did this pubkey/player get" and also per category "who scored in this category"

[icy spindle]

it will be more useful data for us, so we can merge it with the S class task table in the spreadsheet and see the final results

[wicked nymph]

if scoring data is stored in a db, I assume a db extract in csv would be a very decent way to have the data. we could build it into some lookup-tool or something if permitted

[wicked nymph]

btw if you exclude players from competition does that also then remove from competition any transactions they done, any proposals they made and so on, in terms of scoring?

[scenic stratus]

exatly

thats why we were waiting on releasing the data

[wicked nymph]

so their proposals would then be voided too in terms of gov participation?

(afk for a bit)

[scenic stratus]

yes

no no

actually I think so

[violet sundial]

That is lowkey messed up tho. Sybils could have just butt fucked us for fun then XD.

I'm a sybil and post a ton of transactions for you to vote onnnn wieeeeehhhh

We do the work, it in the end gets voided.

[icy spindle]

when will you share this data?

[violet sundial]

I don't mind though. Just saying XD.

[icy spindle]

looking forward to it

[supple drift]

Do you have the sybil list now?

[wicked nymph]

I'm confused

[wild tartan]

you're correct, and there are other reasons why imo various gov stuff shouldn't have been in the competition

but that's a side convo, not for here

[wicked nymph]

I feel quite strongly all votes (ie all proposals) should count. but point taken with not the place

[violet sundial]

So means you can't either do the calculations now.

Wait, was it a yes or no to the question whether sybils' proposals get voided or not?

[wicked nymph]

there was not any way participants could know the proposals were not valid and part of competition...

[violet sundial]

Aren't we a bit overthinking this sybil issue?

[wild tartan]

yeah i think it's being over-thought

the players are players until they are removed, which hasn't even happened yet

[icy spindle]

hello gavin

[violet sundial]

I see. So they become non existent basically in the current structure of data. Including all transactions and whatnot.

[wild tartan]

🤷 idk
at least they'll be removed from S Class, but idk about how the transactions side works

[violet sundial]

Got it, thank you Gav.

[icy spindle]

can we see the results of a/b/c class tasks Gavin, now I have to choose between celebration and suicide rope

[violet sundial]

Yeah I think we're just unsure how they're doing it exactly.

Haha no biggy.

[violet sundial]

Oh noooooo

[supple drift]

Why suicide what was wrong about these tasks?

[scenic stratus]

anybody knows how roids are converted to nam?

[icy spindle]

There is no 1:1 conversion, there is a scoreboard

https://blog.namada.net/content/images/size/w1000/2024/01/08-SE-PRIZES-1.jpg

[iron hull]

No, the last S task db update is from 29th March @last wyvern in current nebb for example none shielded app tasks is yet included, and many explorer tasks, etc. missing, current nebb for S tasks reflects db update from 29th march which is very outdated

Also @scenic stratus @wild tartan in 29th March db for S tasks only 3 crew had vulnerability approved and 1 protocol improvement approved. So in current nebb rigorous should have over 66B roid, and the 3 crew with vulnerability approved over 22B roid, but they have around 6B roid, this means in your code there is a bug and you are using the pilots pool of 62.5B roid for both pilots and crew, so this needs to be fixed

[wicked nymph]

honestly, knowing the brief bit of info we got here from fraccaman today was really valuable. I wish team would have provided it several weeks ago when asked about my scoring queries and such.

[violet sundial]

Compensate this dude!!

💵 💵

[wicked nymph]

thanks but I didn't really do anything useful for the community it turns out. However that would have been nice to know before spending days digging into queries. which is what it is, but the information we got here now would have been so nice to have at an earlier time

[violet sundial]

Lol, sorry I just blurted that out.

Lol, that's literally what hkey explained right?

[icy spindle]

What does 'There is no 1:1 conversion, there is a scoreboard' mean? Are you having problems?

[wild tartan]

part of the issue is that i don't know how to ask these Qs correctly and i don't know when i'm not asking them correctly

[icy spindle]

hello gavin,
when will we see the results of a/b/c class tasks?

[violet sundial]

True man, I'm sure they won't take this for granted right?

Your validator public key to tendermint addresses mapping is invaluable.

[wild tartan]

wdym?

[supple drift]

Any hope to have S results with removed sybils today?

[wicked nymph]

gavin you're a decent guy and you have done so much to keep this expedition hanging together. (and I think team could have engaged with this earlier if they wished to)

[icy spindle]

Currently NEBB shows the results of some s class tasks and A/B/C class tasks. If you only give the results of A/B/C class tasks, we can merge them with the s class results in spreadsheed and see the final results

[scenic stratus]

ill be back guys with more info

[grizzled glacier]

Are we going to see result today?

[iron hull]

I already have this info, in current nebb there are 12 s5 approved for pilots and crew and pool of 62.5B shared for this so s5 gives 5.2B roid, for s6 it gives 4.46B roid with similar calculation. I substracted s5 and s6 in current nebb and that gives A/B/C and s1, s2, s3. I also adjusted for some pilots who didn't yet have s3 and added s4 given the amount approved, also s5 and s6 after audit (but this unreliable). @scenic stratus could you please clarify if in the current nebb only for vulnerabilities and protocol improvements the 62.5B roid pool of pilots if shared for pilot and crew, or if for the rpc, relayer and explorer it is also shared? Since many rpc, relayers and explorers hard to tell for these categories, but for vulnerabilities and improvements easy to tell looking at the top 4 crew roids, this info is critically important since could mean the top 10 or not for several pilots

[iron hull]

Also @last wyvern by the last S tasks db update on 29th march, how many explorer/tools tasks were already manually reviewed, marked as yes on the excel and added to the nebb on the 29th march? On the 19th march you announced that all submissions up to the 19th march were added on the excel, however there was big backlog and delay of reviews, in 8th march announcement you mentioned you were finishing reviewing submissions around 22nd February, so I assume by 29th march you had only finished reviewing submissions until around 10th march or so?

[pine locust]

well this is interesting, I maybe missed 10-15 votes, almost always due to it not letting me for some reason or another and it says im missing 200+ . I used a script so I never missed memo

[wicked nymph]

I very much doubt that's the case, but if you send me your tnam and public key, I'm happy to check. this is all based on namadexer db

[wicked nymph]

I'm looking at all votes cast (successfully) with your public key in memo now, and there are chunks missing randomly up throughout SE. will try and check if you have a bunch of failed votes too

[grizzled glacier]

Maybe we should stop thinking and just hold team in the neck to gib result today😅

[wicked nymph]

you can look at my tool on governance.testnet.one and see which votes you cast successfully. it's not memo-based. you can see there are big chunks of votes missing. not sure if that's due to a lot of them failing, but can check that. in any case the data have been fairly consistent (edit: checked now for any failed vote and there is a single one. so for some reason, a bunch of the votes you thought you made aren't registered anywhere..)

we have no idea if that's the case or not. we can't know anything in fact when team operates a custom indexer where we can't inspect the data nor data models (nothing wrong with that, but we just don't know)

[static minnow]

Hello, Gavin, I want to ask if there will be any updates today because it's already Friday. If there are no updates today, it means we'll have to wait another two days.

[wicked nymph]

the more I think about not counting all proposals, the more wrong it seems tbh. (I assume this is only proposals that were possible to vote on before end of epoch 91 ie end of SE)

[iron hull]

On Monday you mean next week?

[supple drift]

He didn't say on Monday!

[iron hull]

He is based in EU I think and it is 20h Friday evening, I doubt any updates until Monday

[supple drift]

🥲🥲

[grizzled glacier]

I'm so tired of waiting

[iron hull]

Maybe this is a hidden A tasks, those who survive the long waiting are able to receive the prize😅

[sharp elk]

Hi all. I read your messages, as I understand it, @scenic stratus wrote that data about s classes was added to nebb. But nebb is not updated, or did I understand something wrong?

[iron hull]

The last S tasks db update was on 29th march, so while A/B/C points are up to date on the current nebb, the S tasks are very outdated from 29th march, which includes submissions reviewed until around mid March

[grizzled glacier]

Good things come to those who wait right?

[sharp elk]

@iron hull I understand this, but read what he wrote, that some time ago they added the latest data to the nebb #1235664670011035721 message

[iron hull]

Well, this seems a bad communication between @scenic stratus and @last wyvern I guess. If you look at the excel of S tasks, second tab 'latest refresh' you will see 29th march

[neat ember]

To see my points for SE, I need to compile this repository? Is nebb explorer showing up-to-date results now?
...Although, there is nothing relevant in the explorer. I looked now. My validator uptime is 0%

[gleaming turret]

it has been almost a month since the end of the Shielded Expedition, the speed of preparation of the final results is "amazing"

[dense locust]

@scenic stratus There are some cases where multiple Players have the same ranking position. Or a Player has a ranking position that is out of order relative to the amount of ROIDs.

Looking at the source code, I am not sure how that could happen

let affected_rows = diesel::sql_query(
        r#"
        INSERT INTO player_ranks ( ranking, player_id )
        SELECT
          ROW_NUMBER() OVER (ORDER BY score DESC, internal_id)
            AS rank_index,
          id
            AS player_id
        FROM players WHERE players.id in
            (SELECT id FROM players WHERE kind = $1)
        "#,
    )
    .bind::<schema::sql_types::PlayerKind, _>(&player_kind)
    .execute(conn)
    .with_context(|| format!("Failed to set new player rankings for {player_kind}"))?;```

It is like some rows failed to be updated.

[last wyvern]

yeah i think just some minor miscommunication about how up to date the db was

but i've updated the db a few minutes ago with the remainder of the s-class data, like before it will probably take some hours to show up in the front end

the sybils have not been removed yet -- i'm not directly involved in that part so i'm not sure of the eta. i was hoping to wait until everything could be updated in one go, but i'll be away for a bit next week and didn't want to hold everything up until i got back. if because of this the scores jump around a little in the meantime, let's all try and keep a cool head about it 😬 🤞 .

[violet sundial]

Hmm the screenshots (the data) look like they're correct, it might be in the front end that the sorting isn't performed properly?

Aah wait I see, nvm

The first one is ofcourse not correct

[icy spindle]

now all s classes have been changed to the updated versions after the latest rules, right? so one security vulnerability for pilots equals about 12B points. Sooo the final results will be published in a few hours. I want to be sure and never have a heart attack again 🤣

[violet sundial]

Hey Spork, could you before you go away check my DM real quick?

Only if you want to cause I know your DM's are likely a nightmare

[last wyvern]

assuming everything is working the way it should, yeah the scores should be essentially final. A/B/C + the S class. the only thing left would be to remove sybils (so i think that would mean people could only move up, not down)

[violet sundial]

Ah right! Sounds great!

[last wyvern]

(A/B/C tasks were already up to date right?)

[icy spindle]

yes, but it was mixed with some s-classes

[last wyvern]

ok perfect, now the s-class should also be up to date so that should be everything

[thick jasper]

Perfect news. Can you quickly check DM? A question about human error. Sorry for disturbing

[iron hull]

but what about the pools of points for S tasks for pilots and crew being shared @last wyvern? if you look currently in nebb at top 4 crew, why they have around 6B roid with only around 4 vulnerabilities approved for crew on 29th march? If just crew pool they should have over 20B roid not 6B? In your latest update will this be fixed also or will still pending to fix the shared pools for crew and pilots?

Also, do you recall in the 29th march update until which day of march you had manually reviewed explorers/tools tasks? If you recalll more or less would be very helpful, thanks

[regal arrow]

please check my dm

[last wyvern]

i think i recall some sort of glitch in how the front-end was displaying crew vs pilots pool, but i don't know the details. fraccaman or gavin would know

sorry but i can't remember what day the march 29th update went to. i think it probably included up to the 29th, but not sure (also some tasks might have been evaluated out of order)

[iron hull]

In our case we submitted our tool task on 26th March, but only on the 4th April you confirmed via dm that it was approved after we asked for confirmation, so I assume many other explorers/tools around the 26th march you hadn't yet manually checked neither and approved. But definitely up to 19th march only because on 19th march you made an announcement saying that you included 1800 submissions on the excel up to 19th march, but after this no further announcements about adding more submissions to the excel, so when the 29th march db update happened I assume in the excel there were only submissions up to 19th march, but mostly due to the backlog and need to manually check explorers/tool not reviewed all up to 19th march I think but mostly until around 10-15th march. But when you updated the db on the 29th March you didn't have to add how many explorers were approved in total, etc. for each of the S task categories, is this data saved somewhere?

@scenic stratus and @wild tartan should definitely identify the bug in the code that is sharing the same pilot S class point pools for crew also. For example, for the improvements tasks for crew there is around 66B roid, and in the current nebb only Rigorous has this task approved from the 29th march db update, so he should have over 66B roids. Instead he only has 6B because the correct crew pool of points is not being used but Rigorous is considered as a pilot in the code and given a share of roids from the pilot pool of roids incorrectly

[merry viper]

I wonder how you identify sybil?

[quiet aurora]

Yes bro, I think so

[final galleon]

I have two concern in this code:
1.
ProvidePublicRpcEndpoint => Some(relative(match player_kind { Crew => 66_666_666_667.0, Pilot => 62_500_000_000.0, })), OperateNamadaIndexer => Some(relative(match player_kind { Crew => 66_666_666_667.0, Pilot => 62_500_000_000.0, })), OperateNamadaInterface => Some(relative(match player_kind { Crew => 66_666_666_667.0, Pilot => 62_500_000_000.0, })), OperateCosmosTestnetRelayer => Some(relative(match player_kind { Crew => 66_666_666_667.0, Pilot => 62_500_000_000.0, })),
=> With this code who tag Yes many times on sheet maybe get more points. Addition, The Score of player done RPC vs Deploy Interface vs Deploy Namadexer will different

2. The code below shows that Crew's 2 C tasks have not been removed
   `
   UnshieldNaan => Some(relative(42_857_142_857.14)),
   ShieldToShielded => Some(relative(42_857_142_857.14)),

`

Sorry for tagging: @wild tartan @scenic stratus

[merry viper]

when nebb update?

[ornate thicket]

According to the table, namely the "last DB refresh" page, then the rating should be updated in 3 hours

[merry viper]

LAST UPDATED: 2024-04-19 22:08

[ornate thicket]

Table with S class. Page - "last DB refresh"

[merry viper]

table is not correct

[still iron]

[ornate thicket]

🙃

[grizzled glacier]

It was updated yesterday for backend. Front end is coming soon

Let's say in 3 hours

[devout bronze]

is it? 😳🤔

[grizzled glacier]

Yeah

[sharp elk]

why 3 hours?

[grizzled glacier]

Can't even tell. Might be more

[merry viper]

is it include sybil?

[grizzled glacier]

Yes

[merry viper]

i dont see confirmation on the ann, where I can find it mate?

[worthy nexus]

Hello @spork | Knowable . Why not replay my DM? I ask about fairness of the competition

[merry viper]

they are still not public sybil list

[devout yoke]

nice job

[grizzled glacier]

Check the dB update section of the sheet

[supple drift]

What algorithm it uses that needs 24 hours to update!?

[merry viper]

Algorithms come from space

[violet sundial]

I think it periodically updated or was done manually by fraccaman at some point in the past.

But I got a feeling it won't do anything in the weekend 😂...I hope it will though, but yeah...best to keep expectations low, cause it's tiresome to keep on waiting.

[grizzled glacier]

But it's really not fair for them to keep us waiting like this

[supple drift]

Refreshing every minute really tired

[violet sundial]

Yeah I feel ya :(.

[grizzled glacier]

Even when I sleep. My spirit keep refreshing the Nebb

[grizzled glacier]

I don't feel myself

[violet sundial]

Oh nooo, just a lil bit longer then this chapter in your life will be done

This vibe is hitting me atm.

https://youtu.be/2SUwOgmvzK4?feature=shared

[violet sundial]

Damn lol, Astral projecting huh

[supple drift]

Press that button please

[velvet ledge]

By the way, but if he manually updates the data, what is the chance that transactions made after April 11 will not be counted?

[velvet ledge]

@scenic stratus I hope you will be careful when updating the data, otherwise you don’t want to think about it again later.

[violet sundial]

I didn't mean manually like having to do it one by one. Likely it was more like a press 'sync' button in the past. Though I dunno if that was how it went + if that's how it's being done now.

[velvet ledge]

I understand what you mean. I just don’t know if they take into account transactions after the 11th when updating, so, in my understanding, by clicking on the update button you synchronize the data with the last block + with the database.

[dense locust]

Did they not tell you? Every time someone presses the refresh button, the process starts over.

[grizzled glacier]

Huh? Seriously?

[scarlet night]

😜

[wicked nymph]

strongly suggest an actual grace/appeals period after final results are published

[wicked nymph]

this should be looked into

[iron hull]

Hi @last wyvern it has been more than 24h since you updated the db for the nebb but still all the same. Does @scenic stratus need to approve or add the update which means no updated nebb until monday or tuesday? Sorry but there are thousands of people exhausted waiting for many weeks already, we would all appreciate a bit more of clarity in terms of communication, thanks

[wicked nymph]

when you look at the proposals, do you grab their ids given on-chain (as they exist post submission), or the ids they were submitted with in init-proposal (which are not the same)?

[wicked nymph]

any chance we can get a list of which proposals you call as acceptable? as best I can tell those are impossible to find off chain?

[astral olive]

Why does a ROIDS update take so long? I don't understand why the team's progress is so slow.

[wicked nymph]

I have a few questions:

• are postgen validators that have inited with correct memo counted in the uptime / signing blocks tasks?
• is the case where people change consensus keys and thus change tm addresses accounted for?

with the level of complication of the score extractor tool, as well as it running against a custom indexer which data model we have no idea of, the code at hand is nearly unusable for any sort of verification on how scores are tallied imo

[unique orbit]

Or why trying to get them to communicate with us is like pulling teeth

[merry viper]

poor communication guys

[grizzled glacier]

Lol they don't communicate to us on week days. How much more weekend .

[placid wasp]

Please dev do something

[gleaming turret]

Please dev do something! 😁

A month to prepare the results, it's strong

[placid wasp]

And the final results in next quarter..🥳

[astral olive]

Is it really that difficult to update ROIDS?

[supple drift]

Hi @scenic stratus Spork updated the S tasks two days ago. I think you should sync it. it would be great if you sync results. Many are waiting for this thanks in advance

[grizzled glacier]

Should we max tag @scenic stratus ?

[velvet ledge]

Today weekend) chill guys, he only tomorrow start working

[astral olive]

SE ended on 4.11, with an announcement stating that the results would be out a week after 4.22. However, the final results of ROIDS have not been updated till now. The team has never worked according to the expected schedule.

[supple drift]

Just trying our chance😅

[peak jolt]

Hey @scenic stratus can you check dm?

[thick jasper]

Let's keep calm and wait 🙂
I think the team doesn't just wasting time. If we are waiting, then there is an important reason.

[wicked nymph]

agree waiting is worthwhile but would really be good if team was working alongside community here

[elder gate]

Nobody minds waiting. It’s just that now the problem is the same as at the beginning of the expedition, there are no comments from the team about how things are. If there are any delays, they can just go into the chat and write one message.

[thick jasper]

Yes, I agree with that. It would be cool if we were kept up to date.

[grizzled glacier]

What if they've changed about the whole SE thing😂. This abandonment is hitting hard

[unique orbit]

I don't think they'd just rug us
They might pull a Safe where the token isn't tradeable for 1-2 yrs
But at very least we'll get distribution even though the hard work in SE isn't paid off immediately (which is sorta fked up tbh)

[placid wasp]

They didn't anything about vesting

[cobalt cliff]

need big chunk of the rewards to cover therapy expenses

[unique orbit]

We also don't know how long each phase will take
SE participants specifically the top users will most likely have the most NAM aside from team
If turning on tradeability is up to governance and not just default like other projects, that just sounds like another way to keep trading turned off

Recent comments on this matter haven't bolstered any confidence in this matter either
It seems like participating in SE may end up being a big mistake where time committed isn't offset by reward

[placid wasp]

I hear helix said that the each phase will not last more then a week if governance want

[coral cliff]

Wasn't something to do with the hard fork vetoed? lol

[quiet aurora]

The important thing now is that the rankings have not been updated yet 😆

[iron hull]

Hi @scenic stratus @wild tartan I would like to report a tx for A tasks that should be removed from the calculations since several pilots have unfairly cheated this and got an advantage over other pilots. Several pilots with over 99% uptime that were never jailed created another validator, and then jailed and unjailed this other validator and added the pubkey of their main validator in the memo of the unjail tx of the other validator, thus cheating the competition and getting both unjail tx roids and over 99% uptime roids (this uptime couldn't be achieved if their main validator was really jailed and unjailed later their main validator). There are many pilots who didn't cheat and since they were never jailed didn't get this unjail tx and lost an advantage vs the pilots cheating this, so I think this A tx task should be removed from the calculations?

What I am reporting is even worse than this, some validators with perfect uptime never got jailed actually and they got the uptime 99% roids points. But to also get the A task points for the unjail tx, they found a way to cheat the SE system: they created a new validator, they jailed and then unjailed this validator and in the unjail tx they added the memo pubkey of their validator with perfect uptime to cheat and pretend that their main validator was unjailed and they got points for for over 99% uptime and unjail tx, but these two can't be possible, if you have perfect uptime you can't have unjail tx, and if you have unjail tx you can't have perfect uptime

[elder gate]

Regarding the hidden task of Unjail for the crew members. I think everything is fair. In the description for category A - class A asteroids are unknown, the only way to determine what they are is by continuous searching and experimenting with the protocol.

[iron hull]

Also @scenic stratus for the hidden A tasks txs, the pool of points for crew is 300 billion roid, and for pilots 250 billion roids. I suspect also here the same pool of 250 billion roid is incorrectly being used for both pilots and crew and this could significantly impact the final rankings, so I ask you please to check the code to make sure 100% pilots are getting points for all categories from their pool and crew from their pool, not both pilots and crew taking roids from the pilots pools

[wicked nymph]

I think this is cluttering the chat from much more important observations. Can we please not do this?

[iron hull]

I'm not reporting this for crew, there is no uptime for crew. This is reported for the pilots only, since some pilots cheating got perfect uptime and unjail tx which is not possible unless cheating is done with another validator created

[steel basin]

Why would this be cheating? anyone could make as many validators as they wanted and do whatever they wanted with them, distributing their wallet among them. There was a problem that PostGenesis did not consider uptime at all and there was advice to create a new validator.

[wicked nymph]

creating a new validator wouldn't have helped. there's something really wrong with how they (don't) count postgen uptime imo

[steel basin]

at least the new validator began to display uptime, but it was already too late

[wicked nymph]

the information they need to calculate the uptime correctly is possible to find from various sources, and I compiled a nice file, but not sure anyone is listening.

[iron hull]

it wouldn't help with uptime that's why main validator never jailed to get uptime points, but creating new validator and doing jail and unjail with this other validator with pubkey in memo does help to get the unjail tx points

[wicked nymph]

mate we're not talking about that

[supple drift]

Did parameters in a command had roids too? for example changing gas-payer? or the roids were all about the main command?

[wicked nymph]

from a brief look of some of the sources people have shared it's mainly about main commands (with some exceptions). I really wish the calculation methods would have been shared in a less obscure format.

[supple drift]

Thanks, could you figure out what is ibc envelop task?

[steel basin]

I don't understand your indignation. For me everything is within the rules. Nowhere is it stated that only the main validator should go to jail. this contradicts uptime in meaning.

[supple drift]

I think we have (not perfect) but good transparency untill now compared to other teams that many things are somehow a blackbox.

[wicked nymph]

the format needs to be different. the published source says very little about what's actually going on unless you get a complete overview of the entire application. and even if you perfectly understand the source, the underlying data is proprietary and something we don't have access too since they use a custom indexer. So no, atm I don't feel we have what we need to go through the scoring results. (prefered would be a file breaking down who got points for what and such)

[gleaming turret]

@wild tartan @last wyvern any updates about results?

[sharp elk]

Why doesn’t the team communicate more with SE participants?) I would suggest a good outcome for everyone: post the results, some metrics, how transactions were taken into account, a list of who was credited and what was credited. Then I would hold a meeting in zoom, where there would be developers who were responsible for counting transactions, spork, who checked the task, and gavin. The community has too many questions that only create new questions. I think it’s definitely possible to find 2 hours of time, we’ve already been waiting for at least a month, we deserve a little attention)

[astral olive]

The seminar causes more confusion, every competitor has different demands, this is a terrible competition

[rigid glen]

I agree. A meeting is unnecessary. Just finalize the results. People will just need to accept it. The announcement also touches on this. The grace period is simply a means to catch human errors not for appeals.

If we even entertain appeals, it will be endless. It's impossible to satisfy everyone. Someone's always going to have a grievance.

[sharp elk]

But here people have several questions every day, some find inconsistencies in the scoring algorithms, others don’t understand why this task was removed or left. Also, let me remind you, we have a file with 5-10 controversial questions from SE participants for the Anoma Foundation. Answers to them have not yet been given, but there are such moments as participation in the expedition as a team, participation in two statuses at once - pilot and crew, sybil and so on. These questions can also be answered through a meeting or an announcement.

[unique orbit]

We can't even get a reply in text and you want a coordinated scheduled voice meeting 😂

[rigid glen]

We definitely need answers. In general. That we can agree on.

[sharp elk]

thats true😂

[grizzled glacier]

Let's see the final result today?

[wicked nymph]

please, no

the results are nowhere even close to being in a state of ready to be finalized. for one I'm almost sure there are big errors in governance participation calculations

and that's just one out of many issues

but we can't have a constructive dialogue with team about this when they communicate once every 30 days. (sorry I don't mean to be harsh but this is actually the impression we are getting on the ground)

I suggest data is released in a tabularized (csv) format like this: public key, player, class, task 1, task 2, task 3 etc up to and including all s-class tasks, (preferably including a-class broken down by specific task, if not then total A-class), total roid, rank, where each field lists the number of roid gained. (maybe separate file for pilot and genesis). a format like this would make it easy to analyze the data and check if results are correct. also should be straightforward to extract from any db holding the data atm

I also suggest (though a bit more tentatively) the data in team's custom indexer is made available for inspection (sql dump maybe) so any data errors or gaps can be identified or so trust in data can be established (this would also make it easy to compare with namadexer data to see if we are working from the same understanding)

all of this is basically for the purpose of making sure we have faith in the calculations being made, reason for which being that atm the number don't seem to add up.

[astral olive]

Please update the final results of ROIDS, do not keep the community waiting.

[wicked nymph]

I saw you liking my post above

we do need that

[supple drift]

S-tasks should be updated then we can review A tasks

[wicked nymph]

it's not just A-tasks. it's also B in particular

[supple drift]

I know, I'm talking about S-Tasks they should have been updated as Spork said.

[violet sundial]

Trying to make contact

[sharp elk]

Trip to Switzerland?😄

[still iron]

🥶

[violet sundial]

Hehehe yes, I'll be like:

Ola chicos, chicas,

With all due respect, we want our ROIDs 🔫😊.

https://tenor.com/view/the-mask-guns-ammo-jim-carrey-gif-12371085

Nah but jokes aside, we should have an update today. No one is a fan of radio silence.

I know that Spork isn't available this week, I got a feeling the Nebb's auto update functionality has been turned off for a while. So if the Nebb isn't completely broken, we'd need one of the devs to manually perform this sync.

[sharp elk]

Spork 😢

[violet sundial]

Again if it's still functional, else we'd have to come to terms with getting raw data.

Yeah he said this in this chat. Let me scroll up to make sure.

[wicked nymph]

would rather have raw data tbh (broken down). raw data can be analyzed, nebb can't (edit: just to be clear, the raw data I mention here is not a full db dump, but rather a csv approximately in the format above - scores broken down on player and task/subclass)

[sharp elk]

thats true

[violet sundial]

#1235664670011035721 message

[violet sundial]

True we'll likely get this as well. Though would have been great to already get some sort of indication via the Nebb.

[sharp elk]

I'm not sure, but can we tag heliax so that they invite @scenic stratus to our chat?

[supple drift]

Tbh I think it would take 2 weeks with our slow progress

[violet sundial]

Fraccaman was in here at some point above as well. Also left a message in a private chat I had, but no response.

[wicked nymph]

extracting a csv should not be a lot of work

[dense locust]

From which source would the CSV come from in such a way that you trust it?

[wicked nymph]

it's not about trust initially. it's about seeing how the scores have been calculated

[supple drift]

Updating the nebb also was not a lot of work!

[wicked nymph]

and only team can provide that as the rest of us don't have that data

[wicked nymph]

this is different

[dense locust]

https://tenor.com/view/simple-math-gif-25174149

Somehow the text makes it less funny 🤔

[grizzled glacier]

The 5 mainnet phases. When is it starting

[velvet ledge]

Do you have proof this point or its just your opinion?

Can you send link, pls?

[supple drift]

It should be parallel to the mainnet I think

[sterile hound]

So tge testnet hasn't started?

[wicked nymph]

Could we keep this thread on-topic please?

#shielded-expedition

[dense locust]

https://tenor.com/view/eating-wink-hot-dog-yum-gif-17878001

[steel basin]

Hi girl! next question wen listing. 🤣

[cobalt cliff]

nice idea for Yakuza members

[still iron]

I'm thinking about what to cut next tonight 🤔

[still iron]

I would have xxxxxxxx if it had been delayed another week

[bright wadi]

please relax your words a bit . no need to use violence in your sentences . Thanks

[placid wasp]

Are team on holidays 🤔

[thick jasper]

Victory Day)

[placid wasp]

For them

[icy spindle]

Hello,
Spork said the NEBB would be updated in a few hours to show the final results, but there's been no update for 5 days. Can you check it?

@scenic stratus

[wicked nymph]

just want to highlight this again. it's in my opinion the main thing we need in order to process results and make sense of them.

[unique orbit]

Simple time conversation error, ser.

[icy spindle]

day 10 🙂

[wicked nymph]

We had the start of a good conversation here with fraccaman. Who promised to get back with info. Now it's been one and a half week.

I still think that reckoning governance participation only on votes that were initialized with correct memo is the wrong approach

all proposals created on testnet should naturally count in governance participation

I also still suspect scoring is off on some items. Not sure, for instance, if team has accounted for changing tendermint keys and such when counting uptime (again, I have provided files that would make this possible if not happening already)

I think fraccaman also promised to send a list of which proposals are deemed "in the contest", and that would be great to have

but even better to use a correct approach to tally votes on all testnet proposals (votable up to end of SE naturally)

[scarlet night]

Any kind of update/progress feedback would be nice.

[icy spindle]

@last wyvern @wild tartan @scenic stratus

[astral olive]

Soooon

There is no feedback, only waiting

[last wyvern]

i don't want to get everyone's hopes up prematurely... but my understanding from talking to the team is that they fixed the nebb issues yesterday, so the updated scores were/are expected to show up on the nebb page sometime yesterday or today. i guess there's some sort of calculation running on the backend and as soon as that's done running the scores should update automatically

[supple drift]

I wish there were a progress bar🥲

[scarlet night]

will those updated scores include the removal of sybils?

[last wyvern]

yes they will

[storm thicket]

SE1 points are preserved in the calculation, right? I mean, only uptime and governance participation were resetted during the switch from SE1 to SE2

[last wyvern]

yes, everything should be otherwise the same except that all s-class are now included and sybil accounts are removed

[velvet ledge]

Will there be any more summaries or results on SE, or maybe you can tell us what other stages there will be before the final results?

[last wyvern]

assuming the calculation completes without errors, that should essentially be the final results!
there's plans to do a little extra reward recognition afterwards for a few of the best submissions but that's outside of the ROIDS system so it won't change the rankings

[supple drift]

Is it possible to know when the calculation of results would be available approximately?

[last wyvern]

i don't think so unfortunately 😦

[icy spindle]

I guess fraccaman can tell us up to which block the calculation was completed

[supple drift]

As my understanding only @last wyvern can have access to him

It will save us from refreshing😅

[bright wadi]

How.many times have you refreshed it today :)))

[velvet ledge]

As the great rapper 21 Savage sang - A lot

[bright wadi]

Consider it like the Chrome Dinosaur Game :)))))

[scarlet night]

https://tenor.com/view/keyboard-hyperx-rgb-hyperx-family-hyperx-gaming-gif-17743647

[bright wadi]

Ranking for top refreshers
:))))

[grizzled glacier]

54 here

[bright wadi]

Per min?

:))

[grizzled glacier]

/secs😂

[floral copper]

I build the telegram bot auto send the request to the nebb api. if it get the result, it will send the notification to me that is my dream. Actually I working hard today with the real life job 🤣

[bright wadi]

🤣

[wicked nymph]

are results in yet?

[devout bronze]

nope

[velvet ledge]

My intuition tells me that the calculations on the servers cannot produce a 404 error, it sounds so-so.

[sharp elk]

Not today again

[iron hull]

If the nebb was fixed yesterday, then as before we need to wait 24h for the updated S tasks to appear on the nebb I assume. Depending on the timezone when they said fixed yesterday it may not appear today in some other timezones, but definitely if the nebb was fixed on the 14th May for the S class db update by tomorrow 16th the 24h passed in all timezones so the results should appear on the nebb

[quiet hornet]

I hope you are top 1 bro 😎

[thick jasper]

I Hope you are top 25 or higher bro😎

[iron hollow]

[grizzled glacier]

I just hope I'm top 100😅. Even if I'm 100. Cool

[scarlet night]

same here, but I dont think it'll happen

unless there were a lot of sybils

[supple drift]

Me too

[velvet ledge]

@last wyvern Can you talk with @scenic stratus and find out what's wrong with Nebb? I think there are no server-side calculations here. More than 24 hours have already passed.

[placid wasp]

Never ending calculation

[velvet ledge]

This is just ridiculous) If you can’t fix it, just upload the backend data and that’s it, people check it themselves, some craftsmen will even make a frontend display for this data. We've been waiting for 2 weeks already, let's respect each other.

[grizzled glacier]

Se closed 11th of last month. Today is 16th of May

[sterile hound]

Very not cool. Keeping someone waiting this long. We are all humans.

[unique orbit]

They clearly view us as subhuman
Otherwise communication wouldve been better last several month 💀

[supple drift]

Quantum Calculations

[bright wadi]

depend on how many Qubits 😉

[wicked nymph]

it's the backend data that's the issue I think

[astral olive]

😂

[sharp elk]

I think you are right, I also think that this will not be corrected automatically

[astral olive]

Can the backend data be made public so that the community members can assist in its processing? I don't want to keep waiting endlessly.

[sharp elk]

@last wyvern @scenic stratus Sorry for tagging you a lot, but tired of waiting endlessly, can the developers confirm that the calculations are happening on the server and that’s why the Nebb error?

[astral olive]

I don't understand why it takes over a month to get the results after SE. I know there are many difficulties, but it shouldn't take this long.

[wicked nymph]

already asked this 1-2 weeks ago

and yes that would be a dream scenario

[scarlet night]

I would argue we're somewhat beyond the dream scenario atm 😅

[final galleon]

Another friday is coming.

[unique orbit]

See u next friday 😫

[grizzled glacier]

Imagination

[floral copper]

today is today

[supple drift]

@last wyvern

[wild tartan]

hi @here, i've got hundreds of unread msgs in this thread, what are the key things that need to be addressed?

[wicked nymph]

let me link you the most pressing sections, one sec..

[wicked nymph]

at the risk of amplifying my own stuff, I think this section and shortly after is the most pertinent that needs answering. also above this for context

[gaunt marsh]

can you share the sybil list?

[quiet hornet]

afraid you are there? 😄

[wicked nymph]

My personal feel about this is that there are important objections or concerns about the calculations that should be heard, and at least I have tried getting those concerns to team for a month now, as well as provided a pretty solid base for comparison with what comes out of namadexer, as well as some comprehensively calculations around uptime and address maps (which I have spent a lot of time putting together, at this point maybe irrelevant but it's solid work at least in my own opin :))

[wicked nymph]

seeing as they joined the server today...

[gaunt marsh]

Yes I'm 😀

[wicked nymph]

some objections/questions to/on scoring:

• is uptime being tracked for postgen (this influences more subclasses than just uptime)
• are validators who changed their TM address still tracked
• can we reconsider the very roundabout way of calculating voting percentage?

not an exhaustive list..

[floral copper]

I'm afraid too 🤖

[wild tartan]

1. do you mean Pilots that weren't in the genesis set? here are a couple examples of non-genesis operators that hit 95% uptime: tpknam1qq2lhj5mllyl9tdgcavaqvwfk4yn5lqcrvp62lewrxffvk9yuphcve9e0j8
   tpknam1qr3u70ng268u5fw4wsuzwrnksexlgc4w47dtq22lzgyqhhj3ec0tq892se4

2. do we have a reason to think otherwise?

3. i asked Spork about this--are you talking about how some gov proposals aren't included in the gov participation scoring? eg. if PropX didn't have a tpknam in the memo when launched, then it isn't counted in gov participation tally, so voting / failing to vote for it has no effect on gov participation

[wild tartan]

can you help me to understand a big gov participation error?

[wild tartan]

i totally get the issue with team communication. two main issues: 1) noisy Discord server isn't well set up for eng team, who work in Slack/Github and have been sprinting a marathon and 2) eng team is prioritizing mainnet

there's a pretty hard trade-off between trying to perfect the Shielded Expedition flaws and launching Namada (and Heliax has been prioritizing the latter)

[wild tartan]

i don't think that this is going to happen, because the scoring system (that i'm not familiar with) wasn't designed to be replicable, and it would take a lot of eng time to make it replicable, i'm told

@wicked nymph if there is a specific thing to investigate (like is there something that is most clearly incorrect?), that would help

[cursive brook]

Can you answer this? #shielded-expedition message

[wild tartan]

btw @wicked nymph i ++appreciate the time, attn, effort you've put into helping to maintain the integrity of the Shielded Expedition ❤️

[iron hull]

@wild tartan Why major rule changes of the SE are not only still being made now, but also without any announcements but quietly in a forum post? According to the forum post the uptime category of the B class has been entirely removed because 'few participants were able to complete these tasks' how is this a reason to remove the uptime category? You cannot do this, this is not serious. If you remove uptime entirely then you have to do the same for governance also, because also not many completed the governance tasks. This is incredibly unfair. People who were in top 10 you are now putting them out of top 10 because uptime was giving a significant amount of points. Is this why the nebb was not updating, because you were picking which categories to remove until the desired top 10 is obtained? This is an insult for all of us who worked so hard since 1st february for the SE. If you remove uptime like this you have to remove governance as well, you can't cherry pick like this which categories to remove or not, especially because 'few completed a task', then remove also vulnerabilities and protocol improvements since even fewer completed them. I trust this will be fixed asap, I'm amazed about such huge decision to quietly remove uptime entirely in a brief paragraph in a forum: https://forum.namada.net/t/nominations-outstanding-shielded-expedition-submissions-contributions/644

The SE ended on April 11th, before the end you mentioned and announced some C tasks from crew not possible to do and after discussing with Anoma it was announced that they would be removed but nothing else. After SE ended you announced the auditing. But how is it ok to totally change the rules of the SE more than a month after it ended, especially removing categories that would have huge impact on the nebb ranking? I advise to rectify promptly this, either keep uptime or remove governance also, but you can't penalize those who got uptime and not governance, and now put above in the ranking those who got governance but not uptime. Reflect carefully about this and the long-term Namada community

A bounty for uptime? Then make also a bounty for governance, those who have 99% uptime we dont want any bounty, we want the well earned roid for uptime that could put us in the top 10 or top 25, you remove uptime now, put us much lower in the ranking with much lower prize and want to give a little bounty? Terrible decisions

[quiet hornet]

“These two uptime missions had to be removed from the Shielded Expedition because such a limited set of participants were able to compete for these missions”. Compete, not compLete, which is a fair point, if we take in consideration all the troubles with uptime calculations, jailing bugs (silent chain restart, where only se-100 knew the guide to fix their node) and etc.
P.s. I’m one of a few post-gen participants who managed to reach 95% uptime

[iron hull]

We as a community cannot allow this aberration. SE ended 11th april, today is 16th May, and quietly in a forum paragraph without any announcement they are changing the rules of the SE and removing a category critical and giving a lot of roids, destroying all our work and giving our prizes to those with governance. We cannot allow this, there are 2 options here: keep uptime category, or also remove governance there is no in between

[icy spindle]

Hey @hkey there have been ongoing discussions for a long time and the consensus is to respect what it has been decided by the team evaluating rather than continue repeating and insisting about topics extensively discussed before and already decided

🫵🏻 🤣

[iron hull]

I'm sure you have been strongly involved in pushing to remove uptime right as well as trying to invalidate other s6 submissions all to benefit yourself

[wicked nymph]

3. yes this is what I am talking about. it seems almost self-evident that all proposals on-chain (possible to vote during SE of course) should count towards governance percentage from the voting perspective

[wicked nymph]

1. the previous official uptime scorings from team has given almost all postgen zero uptime with a few outliers. those you mention may be outliers. compare with my scoring tables where I believe more postgen figure though this is from memory atm
2. I don't have a full overview of the score extractor because it's too complicated for me in terms of the actual queries that are abstracted away, but I'm not sure I have seen any logic there beyond simple mapping of one TM address. I'm happy to be corrected on this, and this is also part of my questions towards team about 4-5 weeks back when I supplied my own scoring tables (as well as mappings of TM addresses for all validators)

[wicked nymph]

my issue here if I may be perfectly frank is that a lot could have been solved early if team displayed any want to communicate with us and involve us in the process. as it stands now we have a finished scoring product that just is "the way it is" whether that be good or bad (not saying it's either)

[wicked nymph]

I would suggest that if scores are not even broken down internally in the db but come out as a single figure from a black box, there is a very low degree of auditability and a very high possibility of any errors not being caught. A simple breakdown is needed to verify the different subclasses are assigned correctly. I was hoping they were broken down internally in team's db.

[wicked nymph]

of specific things I would like to investigate are questions like:

• are postgen validator signatures for first post-fork block counted?
• how many more people got governance participation 99 and 90 from the roundabout scoring method?
• and similar questions
  these questions can't be answered with source code only. one needs to observe the provable results

but again it's frustrating to me that at this point asking for these things are admittedly a big ask. early on when I first raised the issues it was in my view a less big ask. sorry if I'm a bit stern here, it's not intended as such

[wicked nymph]

as we know for all of SE, you are one of the few outliers that had their uptime tracked for post-gen. we don't know why unfortunately

ok these were my comments for now 🙂

sorry about the essays

[cyan chasm]

Why would some proposal be removed from the counting?
I don't get the logic 🤔

[wicked nymph]

Me neither

but the logic used is that only proposals inited with a player pubkey are "in the competition" and thus counts. Which to me is wrong. that's the explanation however

[iron hull]

Governance category should be removed before uptime for many reasons, giving roids to people creating themselves dozens of spam proposals and voting on spam proposals nonstop, while entirely removing the roids from the best pilots who got 99% despite all the chaos during the SE? When the namada mainnet launches, please call the spam proposals voters to maintain the namada, the network will run very smoothly, and push away the best validators who achieved 99% uptime even in the chaos of the SE for 2.5 months

[wicked nymph]

none of what you describe there happened

[iron hull]

It did happen, some pilots who missed proposals and were not going to get 99% governance then created dozens of spam proposals to increase their governance %. So then governance should be removed, because of naan asymmetry also, some pilots had more naan than others allowing them to create many more spam proposals and vote to increase their governance %. Furthermore, many had votings bots automatically voting the same for all proposals without even reading proposals, including official or critical proposals @wild tartan

[wicked nymph]

if it helps you, Hector, I did all my votes one by one

it's true one pilot created a swarm of spam proposals and that was annoying, but also a vast minority of gov proposals total

[iron hull]

This is not about you, you are actually saying that you dont agree to remove uptime like that, this is about behind doors decisions done and announced quietly while making us think the nebb was not updating for some bugs, when actually it was because of these quiet decisions that will now totally change the nebb ranking. I mean, according to Spork the nebb was fixed by fraccaman days ago and by yesterday it should have shown the final results but still there was no update nebb. Today it was quielty and subtly mentioned in the forum about entirely removing uptime category, so if the nebb was actually updated yesterday or two days ago since it was confirmed that it was fixed would uptime had appeared still there? Or they didn't release the final nebb in purpose just to give them time to changes now these rules about the uptime and maybe others? Maybe the final top 10 was still not the expected one, so some additional rules/categories had to be changed or removed to edit the top 10

[wicked nymph]

I can see that

[wicked nymph]

I'm also not saying that, I'm saying I don't really mind either way. I don't mind if it stays. It can stay for all I care

[wild tartan]

@iron hull pls don't write me a book, give me your position in a few sentences

[wild tartan]

agreed, but big picture--this seems pretty minor, no?

[wicked nymph]

not really. it greatly helps those who did not maintain the full vote percentage

I can't calculate the diff for technical reasons (ie I can't tell exactly which proposals are "in" and tbh have my doubts if this is even possible also for team's solution), but there is no doubt it will cause a lot more to qualify without having actually fulfilled the most reasonable objective criterion

[wild tartan]

i can ask @scenic stratus to check in with you

[wicked nymph]

there is also the very real lingering doubt here if it's calculated correctly in teams version. again, this is a doubt for almost all categories if all we have is a final number.

[iron hull]

Sure, here are the sentences @wild tartan:

Potential reasons presented to remove uptime:

‘naan asymmetry’: this was not a bug, the naan allocations were decided before the SE started by the Anoma foundation, also for governance there is naan asymmetry too since many naan were required to create proposals

‘unjail bug’: assuming there was no unjail bug, this wouldn’t have changed anything because those who got jailed and down for several epochs during the jail period cannot already get the 99% even if normal unjailing was possible

‘no announcement for postgen for upgrades vs genesis’: this is a lie, we were genesis and some upgrades and restarts happened suddenly without announcements for anyone, many other genesis can confirm this too that they missed several restarts and complained about the lack of announcements, we even missed some blocks because of this in some of the sudden and no warning restarts. The claim that this lack of information was only for postgen is an absolute lie

Potential reasons to remove governance:

‘spam proposals’: most of the governance proposals were spam, many validators assumed this wouldn’t of course be part of the competition and focused on the official and important proposals. Some pilots also even used spam proposals just to increase their governance % to achieve 99% governance

‘spam voting’: many created scripts to automatically vote the same in all proposals, without reading or anything, potentially negatively affecting important hardfork and upgrades proposals some of which failed because their default script vote was likely nay and had to be implemented anyway even rejected mentioning this could never happen on mainnet

‘unclear governance rules’ only the governance proposals with the correct memo are considered but this was never mentioned before and we also don’t know the number of these proposals

[wild tartan]

you've historically engaged in good faith, but accusing me of lying isn't a good faith discussion

[wild tartan]

it's a good point (and i'm frankly surprised that it wasn't designed this way..)

just that i'm being told that the resources to provide verification are directly competing with the resources needed to launch mainnet

i'll still pass this along

[wicked nymph]

thanks. the reason for my concern is that as best I can tell, the numbers just aren't adding up. (though maybe this in part can be due to assumptions about method)

[iron hull]

Not you of course, I meant people who made those claims about '‘no announcement for postgen for upgrades vs genesis’

[wicked nymph]

it's true. it's pointless to discuss in a postfactual environment

it literally happened

[iron hull]

It also literally happened for genesis too, some restarts happened suddenly without any announcement or timeline, genesis that were reading discord around the time of some restarts saw the discussion and joined, many others only learnt about the restart many hours later, so was there lack of information for postgen? Yes, was there also lack of information and announcements for genesis? Also yes, the same for both, and this can be verified easily asking genesis validators, many complained during SE about the lack of announcements and info for restart, so dont try to put it as only a 'postgen issue'

[wicked nymph]

(this was during the first halt before gavin assumed communications. this was also the direct cause why @violet sundial was included in the shielded 100 group, to prevent it from happening again)

this part of the discussion is going in circles though and I'm clocking out of this particular section with Hector.

[iron hull]

You are talking about 1 event, in the SE there were many restarts, halts, etc. not only this one, your argument would hold if there was only that restart, which is far from the case, and the same as you described happened for genesis also in many restarts and halts afterwards

[wild tartan]

i am the one making this claim

#🛡︱se-100 and #se-100-announcements you can see, but anyone not tagged Shielded-100 cannot. during the numerous chain halts, there were days between restarts, and coordination only happened using these channels, which are private

this is one of the main reasons we took over comms

but this is not the primary reason for removing the uptime missions. it is because a client bug caused some nodes for some reason to restart over and over, leading to jailing. and then for some reason they could not unjail for a month. dozens of validators

[iron hull]

most of the announcements were in the se-announcements, can you point me to which restarts/halts announcements were shared in se-100-announcements only and not se-announcements, if you dont have time I'll do this research myself as evidence

[wild tartan]

well i made the claim, so it's on me to support it

[iron hull]

Gavin, can you please explain what you don't understand in the following: imagine that jailed validator could unjail normally and there were not issue for unjailing. Still, even they unjailed they still couldn't get 99% uptime due to the downtime during the jail period, hence it is irrelevant because even if they unjail when they could they couldn't already get the 99%, meaning that not getting 99% uptime was due to the jailing event, not the inability to unjail. Then about that client bug that 'caused some nodes to restart and jail' why this bug didnt equally affect all pilots?

[iron hull]

In the 2nd february Awa made an announcement in the general se-announcements channel about the halt, and in the se-100-announcement she said 'posting here for visibility' and linking her post in the main se-announcements channel. In fact I dont find any restart/halt announcement in the se-100-announcement that is not also written first in the general se-announcements. Even before the 1st of february when the SE launched, all the announcements were already being shared in the general se-announcements channel

I would like your feedback also @wild tartan about why not removing governance. Potential reasons to remove governance:

‘spam proposals’: most of the governance proposals were spam, many validators assumed this wouldn’t of course be part of the competition and focused on the official and important proposals. Some pilots also even used spam proposals just to increase their governance % to achieve 99% governance

‘spam voting’: many created scripts to automatically vote the same in all proposals, without reading or anything, potentially negatively affecting important hardfork and upgrades proposals some of which failed because their default script vote was likely nay and had to be implemented anyway even rejected mentioning this could never happen on mainnet

‘unclear governance rules’ only the governance proposals with the correct memo are considered but this was never mentioned before and we also don’t know the number of these proposals

[wild tartan]

thought we already discussed this

some nodes kept restarting over and over again, operator couldn't stop it from happening. this led to jailing

[iron hull]

But why some nodes and not all? if it is a bug why affecting only some nodes?

[wild tartan]

before we took over, i advised removing points for submitting gov proposals early in the game bc it leads to spam and makes validators need to keep voting on spam, was denied

[wild tartan]

i don't know, but i posted the thread if you wanna read through it

[iron hull]

Before you took over, but now you took over right? you are making the decisions now to remove categories of the SE or is it anoma foundation making these decisions?

[wild tartan]

took over comms

[iron hull]

We don't need to know actually, if all nodes were running the same software and some had issues and got jailed and other not, this hints it is issue with operators and how pilots and managing nodes rather than a general bug affecting all pilots?

[static minnow]

Let's give up the argument and move on to build another project. I personally feel that the team doesn't care about our feelings.

[iron hull]

So anoma foundation is making decisions about categories to remove or who is making these decisions? This is important to know for transparency, it is serious to remove major categories like this in the last minute over a month since the SE ended

[wild tartan]

Feb 7 example

public channel: https://www.photopea.com/#iAvb7yyT5

private channels:
https://www.photopea.com/#iEA8a6Pmz - https://hackmd.io/JFTcbzDmRi2DCxVA98eg3w?view
https://www.photopea.com/#i7-hQ0PXn
https://www.photopea.com/#ixKpQTGEv

public channel: https://www.photopea.com/#icdEHkQ3Q

[sharp elk]

@wild tartan While you're here. Will Anoma Foundation respond to questions that have been raised for them? Regarding the number of prizes, about participation within the team, participation as pilots and crew both?

[iron hull]

Here in one of the links you share, @hardy ivy clearly stated there 'the goal is to get the network up and running with the ~100 genesis validators' so this especific link you shared is invalid, it was in the se-100-announcements because as Adrian said it was directed only to the genesis pilots. The hack file you show for se-100 is also invalid since the same exact link is also shared in the se-announcement channel

Who is deciding to remove major categories of the SE over one month after SE ended if not you Gavin?

[wild tartan]

to be clear, i'm showing screencaps where the revival of the chain is only coordinated with the SE100 in private channels, and 15 mins after the chain is making blocks, the fix is posted

[wild tartan]

i've been clear that we've proposed ways of dealing with the Shielded Expedition issues to the Anoma Foundation. you've been a part of these conversations

we being Knowable

[iron hull]

Well, this was done in purpose because as Adrian said the goal was to restart the chain with the genesis pilots. But even if as you suggest the brief delay make some postgenesis miss some blocks, this same situation also happened to the genesis pilots when later halts and restarts happened without any announcements and this led genesis pilots including ourselves to also miss blocks in some restart, therefore claiming this only happened to postgenesis is not correct, if needed I can go throw all the discord history chats to find evidence of this

[iron hull]

Well, so you are effectively deciding, since you present what you want to the anoma foundation right? did you suggest to remove governance or only uptime?

[wild tartan]

#1235664670011035721 message

[iron hull]

We thought the nebb would be finally updated few days ago and there was total lack of communication from you, and fraccaman only spork giving some updates. We thought you were fixing the nebb to show final results, but instead you were choosing categories to fully remove now a month after SE ended and asking approval to anoma foundation to remove those categories. Do you understand now why I asked a lot for the results before the SE ended for transparency?

[wild tartan]

i'll ask someone from Anoma Foundation to join the chat

[iron hull]

Yes and also to present our arguments directly to them about uptime and governance, sorry but seems strong influence and biases here to ask anoma foundation to remove uptime and say nothing about spam governance. This is not nice Gavin, the bounties will be orders of magnitude smaller than a top 10 or top 25 prize difference, and removing uptime unfairly while leaving spam proposal voting will make huge changes in the nebb ranking because most pilots apart from those with vulnerabilities have the same S tasks, the same C tasks and the same A tasks, so the difference is from governance or uptime

[sharp elk]

It would be great to invite someone to Circle. Well, or you have a list of these questions, let them answer, and you publish the answers. These answers are supposed to be the final truth.

[wild tartan]

that's a great idea, let me see what i can do

[wild tartan]

Hector, i have already told you that i engaged about spam governance and was denied

[iron hull]

but you mentioned this was initially before you took over comms, not now when you present major categories like uptime to be remove? can you now please present also your concerns about spam governance so anoma foundation remove also this category?

[wild tartan]

nothing has changed since when i first gave my concerns

[iron hull]

you gave these concerns: 'i advised removing points for submitting gov proposals early in the game bc it leads to spam and makes validators need to keep voting on spam, was denied', you were mentioning removing some points for early proposals, this is very different from explaining all the reasons I outlined above of why the governance category should be removed entirely and asking their approval to remove the governance category, exactly as you did for the uptime category

[wild tartan]

i made several recommendations that all involved ending the need to vote on spam proposals

i understand why you're concerned about this, but i don't have unlimited attention to go in circles about this

there's a balance between perfection and "good enough to move on to mainnet"

and i think we've probably gone a little too far in the perfection direction already

the resources put into trying to land this in a reasonably safe way have set back mainnet and a number of other initiatives substantially

[iron hull]

Gavin, you took the decision to ask the anoma foundation to approve the removal entirely of uptime and this now affects many validators and our ranking in the nebb, this is very serious matter. Why remove only uptime and not governance? This is super unfair, and gives advantage for governance while widly penalising those who got 99% uptime, this is not fair. Either both uptime or governance are removed or both kept, otherwise this is a total unfair advantage for spam proposal voting, which is ridiculous to rewards spam proposal voters and penalized the top performing validators in the SE which will be critical for the successful launch of the namada mainnet🤦‍♂️

[iron hull]

This is not at all about perfecting the SE excuse me, this is about CHANGING the rules of the SE over a month after the SE ended. You say 'the resources put into trying to land this in a reasonably safe way have set back mainnet and a number of other initiatives substantially', but why? the SE ended 11th april, why not just fix the nebb with the final results a few days after 11th April as initially announced? You decided about the auditing and all the other things, if the rules of the SE hadn't been changed this wouldn't be still ongoing now

[unique orbit]

Favoritism toward Shielded 100 when they alr had every possible edge 💀

[wicked nymph]

gotta say, guys, the discussion you are having is going in circles and hector's position is wild (also, it's detracting from a number of the other issues raised here). simply trying to leverage governance where there were never real issues, because unhappy with uptime removal. (which to some degree I understand, but advocating for removing governance is extremely opportunistic).

as @river harbor correctly mentioned in #shielded-expedition , there were never any uncertainty in governance, except some people weren't happy with "spam proposals". but all proposals were available for voting by all, everyone could have attained a 100% governance participation. (though many did not bother)(unlike the uptime issues that were very real, especially for postgen, but also for some unlucky genesis validators)

[wicked nymph]

finding a solution for uptime might be doable if one was willing to admit the real issues there have been, but it's hard to have a conversation where basic tenets of events that happened beyond reasonable doubt are being challenged constantly. it's also extremely tiring and tbh I don't think it's fair to be debating in that manner. it just drains.

[iron hull]

this is total hipocrisy, opportunistically wanting to count all governance proposals not only those with memo to get more roid from spam proposals voting while happy to remove uptime since dont have this category, then you criticize others doing the same as you. You keep insisting all was clear for governace, it wasn't, even Gavin was already at the beginning asking to remove and not count spam proposals so this was an issue being discussed also since the beginning and many assumed spam proposals would of course not be counted for the governance % participation. And the uptime issues you have been repeating nonstop are just some made up arguments to try removing uptime. 1) naan asymmetry is not a valid argument since there is also naan asymmetry for governance, 2) saying only postgen were not aware of some restarts is not a valid reason because the same happened also for genesis pilots 3) the unjail bug is also not valid because if you get jailed then already cant get 99% uptime and if some got jailed and others not this is due to operational skills and not a general bug affecting all. The only valid reason you say is about not counting correctly uptime for all postgen, so this is easy to solve, just fix this and count the uptime for the rest of postgen correctly as for some who achieved over 95% uptime

@scenic stratus did you already find the bug of why for some postgen like @quiet hornet the uptime is counted correctly and they got over 95% while for other postgen the uptime due to some bug is not correctly counted?

[cobalt cliff]

Governance issue is also seems very troublesome and unfair since only few user had access to enough tokens to create proposal, and also even vote for them. I faced shortage of token during the competition just to make tx and asked Gavin for token and he refused it. said I better ask other users on support channel. Spork was kind enough to send me 10k token on my request to continue using my script for Hermes updates . . .let alone keep up with spam proposals. now I also see uptime is not reliable based on the data you released on Github. I really don't know. tried to avoid any debate here and trusted the team since the beginning but I start to think some areas are more hazy than it needed to be

[iron hull]

Exactly the faucet was broken for a long time and this created huge naan asymmetry also for governance, those with more naan could vote and create dozens of spam proposals to increase their governance % voting, and in purpose putting proposals during night or similar in major timezones to make many miss these votings

[cobalt cliff]

I didnt have tokens to make shielded tx for two or three days in a row. If I'm not msitaken Spork was sick during that time and couldnt reply dms. but he did after he was online and sent me tokens. also I faced frozen client just cuz my script stopped working and had to re-do that for my app again.

you can see here I aso asked in channels for token

#🫂︱se-community-support message
#shielded-expedition message

[wicked nymph]

irrelevant for governance participation (voting). I already said I'm pulling out of the discussion with hector and I'm not reading or engaging more with his comments if it continues in the same path

[cobalt cliff]

I dont talk on behalf of him. my question is. is voting for a proposal requires making a tx or not?

[wicked nymph]

I know, sorry, mixed two things in one comment. Could you clarify your question, I don't understand it

[cobalt cliff]

my question s. voting on a proposal needs a tx and gas fee ?

[wicked nymph]

yes, but minimal. for pilots with their endowment it's not really an issue

it's not 5000 naan for voting on a proposal 🙂

(and only pilots have gov % rewards category)

[cobalt cliff]

then I rest my case, I didnt have single token for few days

[wicked nymph]

fair. but that's a matter of prioritizing

I deliberately didn't bond all my tokens as I knew I would need them for other stuff

[cobalt cliff]

I see. I just pointed out my concerns. I really dont want this competition turn into something bitter for all of us.

[wicked nymph]

the reasons I'm coming off so strong on this is the super bad behavior of some in here taking this as an opportunity rather than wanting to find solutions.. (not refering to yourself)

[wicked nymph]

hi, spiderhero 🙂

[iron hull]

Here Spork said on the 3rd May that there was a minor miscommunication issue with you Fraccaman since you thought the nebb at that time included S tasks correctly and Spork informed you that no, it was from an old db update of S tasks from 29th march. Now the same S tasks from 29th march still appear exactly the same as in the nebb on the 3rd May, but you still suggest they are the correct S tasks

[supple drift]

No one says this is the correct S tasks. it will change

They are checking this

[iron hull]

Fraccaman on the 3rd May said the S tasks were correct in the nebb, Spork said that not, the S tasks were from an old db update from 29th march. Today the same S tasks appear, and Fraccaman still said they are the correct S tasks, so Spork will need to again fix this miscommunication issue

[bright wadi]

current Nebb ranking is wrong . and i am sure team is working on it . we just need to wait .

[buoyant flax]

@bright wadi hey qucik question, do we know if the nebb right now (which is wrong) includes a tasks though?

[bright wadi]

Seems to be it include all tasks except the latest S class rewards

A significant change will happen to results when it get fixed later. So lets wait

[buoyant flax]

@bright wadi sorry, just so Im clear, by "latest s class rewrrds" you mean the last calculation of the spreadsheet correct? That smy last quesiton 🙂

[bright wadi]

Yes the latest pilot/crew results in the xls sheet tab

[buoyant flax]

okay thanks

[wicked nymph]

just bumping this thread to keep it active

[wicked nymph]

reminder that it would be really great to know how team gets the tally of which proposals to include (how they index them with memo in the first place) as well as the list of which proposals are deemed "in the contest"

(while still very much thinking all proposals voteable during SE should count wrt governance percentage)