Total tokens received do not match amount reported on grant page #4362
Comments
|
Hey @lrettig I believe this is related to an ongoing issue with the current API we're using as discussed in #4186, sorry about the issue with conversion. The thing is, because all projects were affected the same way by the conversion issues we'll likely leave it this way for Round 2 (feel free to chime in if we want to change this policy @owocki) and make sure we have the proper price conversions in place for future rounds. |
|
@ceresstation thanks for weighing in, and just to be clear, you're suggesting that we didn't actually raise as much as we thought we did? But as I indicated in the original message here, nearly all of the funds were raised in DAI itself, so it doesn't seem to be a conversion issue and I'm pretty sure the math still does not work out! |
|
@lrettig so after talking with @danlipert in this case I think it was actually due to the fluctuation of DAI's value over the contribution period, not due to the API conversion issue. We use the values of assets at the time they're committed to calculate the CLR matching amount which is why the amount on etherscan now is different. The reason for doing it this way is that the value the donor intended to give (i.e. the signal) is best represented by the value at the time of donation. |
|
Okay. Is there a way to know how much DAI we actually raised? Thanks! |
|
Actually is there a way to get a dump of the raw data for the transactions? |
|
Ping here - would it be possible to get access to this data? Thanks! |
|
thx for the bump! heres the output of the following query |
|
Thanks - what's the mapping of fields to columns here? I can't work it out
from the raw data.
…On Tue, 25 Jun 2019 at 16:54, Kevin Owocki ***@***.***> wrote:
thx for the bump!
heres the output of the following query
select
grants_contribution.created_on,
grants_contribution.tx_id,
grants_contribution.success,
grants_subscription.amount_per_period,
grants_subscription.token_symbol
from
grants_grant
inner join grants_subscription on grants_subscription.grant_id = grants_grant.id
inner join grants_contribution on grants_contribution.subscription_id = grants_subscription.id
where grants_grant.id = 82;
http://bits.owocki.com/f5cbf7bb26e6/foo.csv
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#4362?email_source=notifications&email_token=AAZJWNB6YWP4E62X2SI3OQ3P4IWQ5A5CNFSM4HKXNJUKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGODYQQPBY#issuecomment-505481095>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAZJWNECOCAJCCPTRWF6ISDP4IWQ5ANCNFSM4HKXNJUA>
.
|
|
I don't think this is the right data. For instance, there's no True/False column for success, nothing that resembles a tx hash, either, and no token symbol. Can you check this again please? |
|
sorry guys, uploaded the wrong file. this is the correct data: http://bits.owocki.com/19800a5b618e/Untitled.csv columns are: |
|
I’ve been looking into this hoping to figure things out. Before I spend too much time on this, I wanted to give an update of what I am seeing so far. I’m assuming this is the ECH multi-sig account with the discrepancies: And I’m looking at the csv data dump of contributions from Kevin Owocki The txid entry for that 13,548 DAI csv deposit is: That 9K DAI went to a mysterious address I don’t understand: There is an entry in that mysterious 0x620a account for 13,548 DAI going into the ECH 0x6808 account on Jun 9. So, other than the dates, it seems there is almost a connection from the 0x3705 csv txid to the 13,548 deposited into the ECH 0x6808 account, through this mysterious 0x620a account. But the 40 ETH -> 9,377.59 DAI going into the mysterious 0x620a account from the txid in the csv file doesn’t match the 13,548 DAI listed in the same csv file line, and coming from that mysterious 0x620a into the ECH 0x6808 account. So, it doesn’t seem like I’ve yet figured out everything necessary to start reconciling things. |
|
@owocki, could you explain the stuff I'm seeing in bold? |
|
thanks brent. confirming this is on my TODO list, but i have a few things ahead of it (and meetings, ugh). let me see if i can knock this out by end of w1 August. |
|
I’m assuming the grants.contribution and grants.subscription tables are tables in the gitcoin db. Perhaps you could tell me more about these tables, in particular the “amount_per_period” values. Could this represent more than the amount in the corresponding transaction id in the same column? (i.e. more than one txid “per period”?) If so, what other txids might be involved? |
yes
that number should only include successful tx's
Yes ! here it is - i added tab with etherscan data and one with gitcoin data too. There seems to be more rows from etherscan than with gitcoin though -- is this account used for something else too? |
it should only be one tx per period |
|
Oh, thanks for sending this google sheet, that really helps. @owocki aked:
If you are talking about the 0x68085e7f88e90fa9247489c83fab30b177ebccb3 account, yes my understanding is that this account is used for all Cat Herder funds. There has been talk of splitting the ProgPOW audit funding to its own account. The grants page Total seems incorrect. When I total up the amounts on the grants page it doesn’t match the “Total Funded” field, today (Aug 3, 2019) 29,198.41 DAI Manually calculated (matches spreadsheet + 5 additional contributed by @owocki) Is this the problem? @lrettig said:
But this balance never existed, that I can see, no matter how you run the balance from all contributions in order. |
|
I just noticed that is about a 3% discrepancy. Maybe this is a gitcoin fee? |
no, no gitcoin fee |
|
If it isn't a get coin fee. Where is the money going? |
|
talking to @BrentAllsop on twitter today; we need to figure out why the 3% difference in the numbers. |
|
hmm ; looking at the calcs... i wonder if the ZRX/LPT contributions (both of which have had some price fluctuations) have something to do with the 3% discrepancy..
maybe if i add a USD equivilent column to the 'activity' tab on the grants activity page; that will help us reconcile this.. having both the gitcoin DB values and your values on one spreadsheet will make it easier to zero in on the root issue. also.. hindsight is 20/20 but next time we should use one account per grant |
|
I don't know what you mean by "USD" equivalent. My understanding is that Dai are Dai. If X + Y + Z dai are donated, you should have a total balance of dai contributed which equals X+Y+Z DAI, not 3% less than that total amount of DAI? The exchange rate between USD and DAI or the "USD" equivalent has nothing to do with these DAI totals, right? |
|
Not all of the contributions were in DAI.
On Tue, Sep 10, 2019 at 9:35 PM Brent Allsop ***@***.***> wrote:
I don't know what you mean by "USD" equivalent. My understanding is that
Dai are Dai. If X + Y + Z dai are donated, you should have a total balance
of dai contributed which equals X+Y+Z DAI, not 3% less than that total
amount of DAI?
—
You are receiving this because you were mentioned.
Reply to this email directly, view it on GitHub
<#4362?email_source=notifications&email_token=AAD5PCJ6VQWSBR4G6NCOFDTQJBRQTA5CNFSM4HKXNJUKYY3PNVWWK3TUL52HS4DFVREXG43VMVBW63LNMVXHJKTDN5WW2ZLOORPWSZGOD6NEWGI#issuecomment-530205465>,
or mute the thread
<https://github.com/notifications/unsubscribe-auth/AAD5PCILAYXZZUCLBLGU6DTQJBRQTANCNFSM4HKXNJUA>
.
…_________________________________________
check out what i'm building on github <http://github.com/owocki>
or what i'm shooting on photoshelter <http://photography.owocki.com/>
or find me on twitter <http://www.twitter.com/owocki>, facebook
<http://www.facebook.com/kevinowocki>, instagram
<http://instagram.com/owocki>, linkedin <http://www.linkedin.com/in/owocki>,
and owocki.com <http://www.owocki.com/?utm_source=emailsig>.
see you around teh internets!
PS -- Come to the next Boulder Blockchain
<https://www.meetup.com/Boulder-Blockchain/> meetup. Be there or be ^ 2.
PPS -- Have I mentioned that *G**itcoin is live?* More @ https://gitcoin.co
|
|
here i wrote a script that will show everyone exactly how the contributions we've got tracked add up to the total reported on the grant page (the original intent of the ticket).. hopefully this clears things up; i wish i had just done this in the beginning: |
|
Hi Kevin @owocki, You did provide some data in the Untitled.csv file, above, as of Jun 25. As I indicated, that data matched the individual amounts donated, as listed on the activity tab on the grants page. When I remove the non DAI entries from this list you just provided, and then total only the DAI, I get 28,003.41 total DAI. The total DAI donated, on the grant page, as of today says: 28,196 DAI. This is 192 less DAI. This is only about 0.6% difference, as of today. As I indicated, above, the difference on Jun 25 was 909.41 DAI, when doing this same summation. That was a much larger 3% difference I guess our question is, what is accounting for these differences? |
|
hmmm... let me restore a backup on my local from june 25 (when i posted that csv) so ican look at the diff |
|
@owocki, |
|
I provided the totals, with the 3% deficit, in my August 3rd post, above. |
|
I agree with @lrettig initial statement. There is some mismatch in numbers. I took a dump of ECH multisig account (Date Oct 29, 2019) and matched with transactions available on Gitcoin page. Converted value of ERC tokens in DAI and USD (Rate: CoinmarketCap) and it shows a difference of 50DAI.
|
|
can i have access to the spreadsheet? what does the difference look like if you use the conversion rates that the gitcoin system uses ? which i noted in this doc... #4362 (comment) note: we sync our price data from etherdelta/poloniex |
Attached.
That may be a probable reason for the difference. |
|
Kevin, just FYI, I sent Pooja a copy of this spreadsheet, I think originated from you. |
|
Revisited the sheet and figured an error. Now the total fund is only a DAI off. Unsure, but that could be because I used coinmarketcap rates and Gitcoin uses etherdelta/poloniex.
Updated sheet attached |
Hello,
Our grant page (https://gitcoin.co/grants/82/official-progpow-technical-audit-funding) says we received 15,799.70 DAI but we didn't actually receive that many (https://etherscan.io/address/0x68085e7f88e90fa9247489c83fab30b177ebccb3#tokentxns). We are short by a few hundred DAI. How was this 15,799.70 DAI figure calculated? Does it include the CLR match amount? Maybe this hasn't actually been paid out yet? We also received ~3 contributions in other tokens but they were small and even adding them, it's still short.
Thanks!
The text was updated successfully, but these errors were encountered: