[ 3 / biz / cgl / ck / diy / fa / ic / jp / lit / sci / vr / vt ] [ index / top / reports ] [ become a patron ] [ status ]
2023-11: Warosu is now out of extended maintenance.

/biz/ - Business & Finance


View post   

File: 170 KB, 750x452, api3_dimi_1.png [View same] [iqdb] [saucenao] [google]
25001272 No.25001272 [Reply] [Original]

>> No.25001899
File: 29 KB, 772x492, api3_dimi_2.png [View same] [iqdb] [saucenao] [google]
25001899

>>25001272

>> No.25001926

>>25001899
what else did he have to say

>> No.25002044
File: 64 KB, 990x168, api3_dimi_3.png [View same] [iqdb] [saucenao] [google]
25002044

>>25001926

>> No.25002096

dimitri left chainlink in like 2019 right?

>> No.25002131
File: 75 KB, 988x204, api3_dimi_5.png [View same] [iqdb] [saucenao] [google]
25002131

>>25002044
>while competitors are giving away price info to callers for free, albeit unsustainably
hmm i wonder which project he's referring to ...

>> No.25002165
File: 155 KB, 978x444, api3_dimi_6.png [View same] [iqdb] [saucenao] [google]
25002165

>>25002131

>> No.25002200
File: 470 KB, 994x1218, api3_dimi_7.png [View same] [iqdb] [saucenao] [google]
25002200

>>25002165
and the response

>> No.25002264
File: 243 KB, 998x720, api3_dimi_8.png [View same] [iqdb] [saucenao] [google]
25002264

>>25002200
i'm averaging 50% checks in here

>> No.25002293
File: 38 KB, 750x132, api3_dimi_9.png [View same] [iqdb] [saucenao] [google]
25002293

>>25002096
yes. maybe he's looking for a new job? pic related.

>> No.25002610

>>25002200
>beacons aggregate/calculate inputs to minimize on chain costs
why not just t sig

>> No.25002850

Interesting, thanks.
He had his disagreements with CL but he also seems bitter about being let go early.
The biz model question is the single most important question wrt competitors and no one in their tg had asked about that.

>> No.25002964

>>25002610
I don't think this is referring to threshold signatures, it's referring to each oracle doing aggregation separately (as well as maintaining and updating their own data points) and then that aggregation being verified on-chain.

E.g. median is computationally expensive because you have to sort all data points -- O(n log n) for CSfags.

To save costs, you can get each oracle to sort the data points off-chain and then report them on-chain. The on-chain aggregator then only needs to check if the data is sorted -- O(n).

None of this requires threshold signatures.

>> No.25002991

>>25002850
>He had his disagreements with CL but he also seems bitter about being let go early.
Are you an insider? o.0

>> No.25003149

>>25002850
True. But the thing is, Chainlink subsidizes users of their data feeds by giving them tokens. This is clearly unsustainable (like what Dimi said). At some point, people are going to need to start paying. Then it's a matter of whoever has the cheapest data feed (w.r.t data quality etc). The API3 team make good arguments in their whitepaper re: why their data feeds will be cheaper. But I guess only time will tell.

>> No.25003216
File: 730 KB, 1000x750, 1592918206038.png [View same] [iqdb] [saucenao] [google]
25003216

>>25003149
as demand for offchain data onchain grows, there will be more parties willing to pay, so each datafeed becomes a community resource. You'll have 300 parties all paying for X data, meaning the price for each party is 1/300th of the price of producing that data. the more use the cheaper it gets porportionally

>> No.25003232

>>25002991
He left a core position in a leading blockchain startup for a mid level position at IG. He was the second biggest code contributor after Steve.
Either got bored or had issues with management.
If you had looked into his resume then, it was also obvious he was doing a lot of small projects just to ass on his resume; I think his game plan was always work for a niche startups for a couple years so it looks good on a resume, just to get hired by a big tech company. A common strategy for young devs, since most startups fail anyway.
For reference the price was crabbing at two dollars range when he left and maybe he thought this was an unsustainable top.
He made a bad call and now likely wants to invest in newer projects.
He didn't even understand the potential of his own work, many such cases.

>> No.25003329

>>25003216
I don't see why this business model can't also be used by other oracle projects?

>> No.25003401

>>25003232

Oof, mfw i had more faith than a lead dev.

Give me 2021 Link price prediction Mr Based Insider.

I've been furloughed most of the year and need to know if my 50k Link will set me free and let me start my own business or if I will have to go back to some crappy desk job.

>> No.25003409

>>25003232
idk, man. this sounds like linkie cope. how do you even know he's mid-level? faang engineers don't advertise their levels internally or externally, i know because i worked at one.

>> No.25003419

>>25001272
Hi can you stop posting about API3 thanks.

>> No.25003472

This project is probably funded by chainlink lol.. isn't his just clcg? or someshit.

clcg.io the site redirects to chainapi.

>> No.25003474

>>25003401
$1000

>> No.25003497

>>25002964
Yes, he was not referring to threshhold signatures but my question was, why use dedicated "beacon" nodes to aggregate and publish results when you can have your favorite number of nodes reach consensus off chain and publish that via a single multisig tx, i.e. the competition's tsigs scheme? what is gained by burak'a solution?

>> No.25003533

yes please keep posting these. it makes my API3 bags seem lighter and healthier

>> No.25003546

>>25003419
u 'coomulating?

>> No.25003611

>>25003497
I'm pretty sure he addresses that: everything is more transparent with the beacons, because individual points are on-chain, so you know what the data was pre-aggregation. This also doesn't involve the complex/nuanced decision of selecting your favourite nodes, or however you put it.

>> No.25004084

FUD = buy. that simple sometimes.

>> No.25004321

>>25001272
bumperino

>> No.25004341

>>25002096
Right around when they quietly started to realize LINK staking needed either complete rework or abandoning. Bullish af for API3 if it got Dimitri interested again

>> No.25004356

Interesting. I've always wondered why he left so early, especially when he was thought of as this whiz kid dev with a top-notch background in AI.

>> No.25004365

holy fuck you fucks really can't stop posting about this shit for TWO days. Can you atleast wait until we get near the cex listing ?

>> No.25004532

>>25004365
lol no

>> No.25004538
File: 94 KB, 800x800, 1598314057941.jpg [View same] [iqdb] [saucenao] [google]
25004538

>>25002165
>They get sponsored offline
Comfy

>> No.25004731

>>25001272
delet

>> No.25004808
File: 54 KB, 604x360, asdsa.png [View same] [iqdb] [saucenao] [google]
25004808

>>25003329
this was the beauty of chainlink's ico and distribution. the 65% is there to incubate the network and solve the chicken and egg problen. remember that sergey is ruthless as fuck
in terms of dimtri, yeah he got pushed out for the doing side projects and not showing sufficent commitment.
https://twitter.com/mobealo/status/1340697540094640129

>> No.25004842
File: 78 KB, 344x497, gent.jpg [View same] [iqdb] [saucenao] [google]
25004842

oracles are dead meme, this is like people backing "revolutionary L1s" when Chainlink was already the next big thing.
Nobody cares about getting data on chain any more. We indexers now.

>> No.25004928

>>25004808
>this was the beauty of chainlink's ico and distribution
>sergey holding 65% of the tokens
>beauty
>your brain on chainlink

>> No.25004939
File: 4 KB, 811x133, dimitri github.png [View same] [iqdb] [saucenao] [google]
25004939

I wonder if he signed a non-compete. Been about a year now, right?

>> No.25004977

>>25004842
retarded argument. indexers are only interesting if you have interesting on-chain data, which requires interesting dApps, which require off-chain data ... it's a positive feedback loop

>> No.25005065

>>25004808
Mat Beale retweets Joe Biden.

>> No.25005112

>>25005065
not only that, he talks about drug use on twitter, completely unprofessional
an idiot

>> No.25005116

bump good thread.

>> No.25005489

>>25001272
DLEETE
DIS
PAWST!!!!!!!!

REEEEEEE

>> No.25006105

>>25004842

Indexers - enjoy the wagecage

my god, GRT vc capitalist that are 500x up truely did a number on these guys.

Talk about pump then dump on normies.

>> No.25006144

comfy hold

>> No.25006294

just sold all my link and went all in API3

>> No.25006674 [DELETED] 

>>25001272
andre too

>> No.25006716 [DELETED] 

>>25001272
andre too

>> No.25006717

>>25006294
proof or larp.

>> No.25006745

>>25001272
andre too

>> No.25007026

>>25006717
sorry just a larp

>> No.25007514

>>25007026
F

>> No.25007700

>>25001272
what is it capable of reaching?

>> No.25008180

now imagine not holding at least 1000 API3 for the upcoming atlszn and binance listings

>> No.25008422

>>25008180
10,000 is the make it stack.

>> No.25008453

>>25008180
yep, api3 is useless but definitely worth having for the coming oracle hype cycle

>> No.25008577

>>25008453
uselless for now, but they are pushing this hard, i bet many announcements will come this month

>> No.25009500

>>25001272
Be D. Roche
>high iq dev wishing to change the world for the best
>hangs out with ETHfags, learns solidly and is captivated by CL promises
>Quickly Climbs the CL ranks
>D. Roche, CL second in command dev
>spends his nights eating pasta and not developing
>time passes and becomes tired of rhetoric
>Fights with Steve over CL direction
>D. Roche storms out promising himself to never work for snake oil sales man, and turns his focus where his thoughts would be respected, IG
>tired of IG Stacy, he begins day dream of ETHfag promises
>finds API3 and re kindles his passion for bringing real world data on-chain
>begins to wonder if API3 can restore his broken heart
>after years he finally goes to sleep with a smile on his face
Where were you anon when the oracle world shook.

>> No.25009759

comfyness level: all time high

>> No.25009810

fuck off timo

>> No.25010136

>>25009810
few understand

>> No.25010157
File: 15 KB, 451x75, dimi.png [View same] [iqdb] [saucenao] [google]
25010157

>> No.25010208

>>25003232
left or let go?

>> No.25010674

>>25010208
talents leave

>> No.25010703

>>25010208
>>25010674
Im assuming these guy just do contract work. So they are just hopping project to project