<stellar-slack> <monsieurnicolas> If anybody has feedback on https://github.com/stellar/stellar-core/issues/719 please feel free to chime in. We do not have complex processors on top of the data generated by core yet but I'd rather future proof things than be stuck down the line :)
<stellar-slack> <lab> i won't support this proposal
<stellar-slack> <jed> can you explain why in the issue?
<stellar-slack> <dzham> Just switched over to live.stellar.org.. same issue.. sequence numbers don’t get incremented
<stellar-slack> <jed> dzham: that seems really strange? Does the client still work?
<stellar-slack> <jed> How are you seeing that the # isn't going up?
<stellar-slack> <dzham> stellar-lib doesn’t work, unless I close and reopen the remote
<stellar-slack> <dzham> I’m using my own library, and just checking account_tx
<stellar-slack> <dzham> I’ve got four consecutive tx’s from the same account, with seq# 7
<stellar-slack> <dzham> gUSPmLR8WKgrrnhPAmLKxoz2WuTvkCTREa
<stellar-slack> <sacarlson> what time frame was it that you performed the operations on seq#7
<stellar-slack> <sacarlson> were they all performed within 20 sec?
<stellar-slack> <dzham> no, these are spread out over say half an hour. all of them are validated, in the ledger
<stellar-slack> <jed> idk the seq# is incrementing for me
<stellar-slack> <jed> are you sure the 3 txs succeeded?
<stellar-slack> <jed> maybe 2 failed?
<stellar-slack> <dzham> tesSUCCESS
<stellar-slack> <lab> @monsieurnicolas: i posted my feedback.
<stellar-slack> <jed> gUSPmLR8WKgrrnhPAmLKxoz2WuTvkCTREa is on the live network?
<stellar-slack> <dzham> both on live and test. have the same problem on test
<stellar-slack> <jed> you are saying that it has 3 txs all with the same seq # that succeeded?
<stellar-slack> <dzham> yup
<stellar-slack> <jed> what seq # on test net?
<stellar-slack> <dzham> for that address I think it’s stuck on #16
<stellar-slack> <dzham> alright, now it looks ok on live :S
sv-logger has quit [Remote host closed the connection]
sv-logger has joined #stellar-dev
<stellar-slack> <lab> i'm afraid SDF is compromising more than i thought they would.
<stellar-slack> <sacarlson> what does that mean?
<stellar-slack> <lab> it's despairing trend
<stellar-slack> <sacarlson> funny how some see the bad and others see the good. I see the light at the end of the tunnel
<stellar-slack> <lab> there is the light at the end.
<stellar-slack> <buhrmi> RE that proposal: i actually like the fact that stellar core doesn't really care much about keeping history itself and leaves it to the network observers
<stellar-slack> <lab> it's a portal to another trivial world which i regarded as the new world
<stellar-slack> <buhrmi> as long as there's beer and sausages in the new world
<stellar-slack> <sacarlson> here here
<stellar-slack> <sacarlson> I did read the above https://github.com/stellar/stellar-core/issues/719 but parts are above my paygrade. I had a better understanding of 712 and 713 I guess they were that were clear as to what the problem is
<stellar-slack> <lab> from my observation, SDF supposed an active community initialy
<stellar-slack> <lab> SDF take charge of core develepment, and documents and few samples of surrounding application.
<stellar-slack> <sacarlson> @lab I'm now writing documents that I publish so they can't control that
<stellar-slack> <sacarlson> we can also branch at any time we want
<stellar-slack> <sacarlson> I was running fredolafritte branches and still am on parts
<stellar-slack> <sacarlson> it's a simple world, if you don't like it rewrite it. if they don't like it branch
<stellar-slack> <lab> federations by community members, wallets by community members, block explorer by community members, exotic apps by community members...
<stellar-slack> <sacarlson> are those requests @lab?
<stellar-slack> <lab> it's the new world i believe.
<stellar-slack> <lab> the first bad feeling is from issue of "federation field" https://github.com/stellar/stellar-core/issues/461
<stellar-slack> <sacarlson> what wrong with that? I created my own private federation with a website. I see this already exists in the stellar-core database as I recall. It's opitional if you want to use it
<stellar-slack> <lab> it's a trend of being "monolithic solution provider"
<stellar-slack> <sacarlson> it's just too hard to remember a big number easy to remember zipperhead
<stellar-slack> <dzham> @jed: my bad, “ga1VLQ7ya1NgTHyZbbC77rxs1L8w7ZUoMi” and “gpStBpZYVAW6HaxHX1FDedYBTd4dtUFf9P” on test. AccountRoot updates seq#, but stellarlib uses the old one, and stellard accepts the tx
<stellar-slack> <dzham> It seems to be working correctly on live, so I’ll just stick with that for the mean time
<stellar-slack> <sacarlson> cool so your happy now then
<stellar-slack> <dzham> Legacy code. *sigh*
<stellar-slack> <lab> i'm not saying we need no federation. it was just arguing keep more room for community as initial goal.
<stellar-slack> <dzham> +1
<stellar-slack> <buhrmi> +1
<stellar-slack> <lab> even the community is not grow as expected
<stellar-slack> <buhrmi> can't force it
<stellar-slack> <buhrmi> personally i don't like segregation into stellar-public and (probably) stellar-private?
<stellar-slack> <buhrmi> stellar-public should just be stellar...
<stellar-slack> <lab> the reparation should be more efforts on community growth than lowing the positioning
<stellar-slack> <buhrmi> ah, http://stellar.slack.com|stellar.slack.com is already used
<stellar-slack> <lab> because most of the residual community members, few and hardcore, is attracted by positioning.
<stellar-slack> <lab> there may be no stellar-private.
<stellar-slack> <buhrmi> yeah
<stellar-slack> <lab> even so, i don't think SDF running is transparent as expected
<stellar-slack> <sacarlson> don't worry we will have buhrmi setup private-core for us
<stellar-slack> <lab> There must be something happened turns SDF from initial idealism.
<stellar-slack> <lab> we have no unveiled information on the process, but guess on the consequences.
TheSeven has quit [Disconnected by services]
[7] has joined #stellar-dev
<stellar-slack> <lab> the uncertainty gave more negative factors to community growth.
<stellar-slack> <lab> regarding me, i has to make more efforts to *dig* out what SDF is doing and planing to do to adjust my business to match.
<stellar-slack> <jed> lab > i'm afraid SDF is compromising more than i thought they would.
<stellar-slack> <jed> in what way?
de_henne has joined #stellar-dev
<stellar-slack> <buhrmi> in the cute stellar galaxy comic, the cute aliens say there is no central authority...
<stellar-slack> <buhrmi> but isn't the SDF the central authority by controlling the stellar / lumen give-away?
<stellar-slack> <eno> network is not equal to these coins IMO
stellar-slack has quit [Remote host closed the connection]
stellar-slack has joined #stellar-dev
de_henne_ has joined #stellar-dev
de_henne has quit [Ping timeout: 264 seconds]
<stellar-slack> <lab> @jed: the responsibility between core team and community.
<stellar-slack> <lab> do you think SDF need provide a whole stack, from stellar-core to horizon and client? every thing should be state-of-art?
<stellar-slack> <lab> i know community can't absorb new end-users without these. doing so is what i called compromising.
<stellar-slack> <lab> i expect SDF put most resources, if not whole, to produce a minimal essential version of stellar-core. abandon end user experience before production.
<stellar-slack> <lab> the above argument is based on assumption of lack of resources guess by observation of public activities.
<stellar-slack> <lab> the process of make a whole stack with such limited resources is dangerous.
<stellar-slack> <lab> the deliver time or quality has to be compromised. these consequences will drive last residual community members away. it will be true disaster of project.
de_henne_ has quit [Ping timeout: 260 seconds]
de_henne has joined #stellar-dev
<stellar-slack> <lab> actually, i intended to setup a team for client web/app. but i'm hesitate to invest more because of the uncertainty of process of core team. i think many potential entrepreneurs around stellar ecosystem were facing the same situation in the last months.
<stellar-slack> <lab> our stand instead of work let core team need do more. it's vicious circle. we need break it.
de_henne_ has joined #stellar-dev
de_henne has quit [Ping timeout: 264 seconds]
<stellar-slack> <buhrmi> what is "pricen" and "priced" in an stellar core offer?
<stellar-slack> <fredolafritte> numerator - denominator
<stellar-slack> <buhrmi> clever
<stellar-slack> <dzham> hooray for descriptive variable naming! :)
Kwelstr has joined #stellar-dev
<stellar-slack> <buhrmi> @sacarlson: your offer is showing up ^_^ http://open-core.meteor.com
<stellar-slack> <sacarlson> it looks to have a decimal in the wrong place as I offered units at 1.25
<stellar-slack> <sacarlson> the quantitiy is correct at 24
<stellar-slack> <buhrmi> i think you issued it incorrectly
<stellar-slack> <sacarlson> opps wrong one
<stellar-slack> <sacarlson> that's what I ran
<stellar-slack> <buhrmi> hmm
<stellar-slack> <buhrmi> whats the offer() function do?
<stellar-slack> <sacarlson> so someone else will have to create an offer to verify where the fault might be
<stellar-slack> <buhrmi> actually i don't think there should be any math applied to asset offerings
<stellar-slack> <sacarlson> I would have to look into stellar_utilities to know but that mostly just send it to ruby-stellar-base my stuf does very little other than find the sequence number
<stellar-slack> <buhrmi> anyway i now display assets without dividing by millions
<stellar-slack> <buhrmi> so now you offer 1 beer for 12,5 million usd
<stellar-slack> <buhrmi> sacarlson-usd, that is ^^
<stellar-slack> <sacarlson> it accepted a real number. it's not interger
<stellar-slack> <sacarlson> 2's complement?
de_henne_ has quit [Ping timeout: 260 seconds]
de_henne has joined #stellar-dev
<stellar-slack> <sacarlson> this is the example seen in ruby-stellar-base that he uses a decimal number in the price. maybe they convert it to integer like native when they put it into the database?
<stellar-slack> <sacarlson> but this is then number I see in the stellar-core database for my order 12500000 it's not shifted 6 dec it's shifted 7
<stellar-slack> <sacarlson> does the js-stellar-lib and js-stellar-base orders end up looking the same? I guess as long as they all do the same thing shift of 7 doesn't really mater
<stellar-slack> <fredolafritte> the js lib doesn't perform any shifting
<stellar-slack> <fredolafritte> the information about the decimal place in not and will not be in the ledger according to the stellar dev
<stellar-slack> <sacarlson> oh so it's up to the market to decide what decimal place they want it to be?
<stellar-slack> <sacarlson> so that should make them all in the same place and that's the important part
<stellar-slack> <fredolafritte> so we need a centralized place to get this info
<stellar-slack> <buhrmi> working on it
de_henne_ has joined #stellar-dev
de_henne has quit [Ping timeout: 264 seconds]
de_henne_ has quit [Remote host closed the connection]
sacarlson has quit [Ping timeout: 244 seconds]
de_henne has joined #stellar-dev
<stellar-slack> <buhrmi> initial layout stuffs ^_^ http://open-core.meteor.com
sacarlson has joined #stellar-dev
<stellar-slack> <lab> :+1:
<stellar-slack> <lab> we have 13 nodes already
<stellar-slack> <matschaffer> nice buhrmi !
<stellar-slack> <scott> @sacarlson @fredolafritte I haven’t read back through the whole thread but it looks like you’re trying to figure out how we convert a floating point price into the fractional price. We use a continued fraction to approximate the decimal price up to the maximum accuracy provided by the space allowed. You can see where that happens in ruby here: https://github.com/stellar/ruby-stellar-base/blob/master/lib
<stellar-slack> <fredolafritte> I'm trying to figure out why a float is used
<stellar-slack> <fredolafritte> float and financial data doesn't play well togother
<stellar-slack> <scott> it’s not used for any math. It’s used because that’s how most people will write a price.
<stellar-slack> <buhrmi> yeah i think stellar core does not save it as float
<stellar-slack> <buhrmi> its a bigint64 i think
<stellar-slack> <fredolafritte> the price in float is not used for any math ?
<stellar-slack> <fredolafritte> how can we be sure of that assertion
<stellar-slack> <buhrmi> look at the ledger
<stellar-slack> <buhrmi> the price is an integer
<stellar-slack> <scott> I think the only way you could be sure is to either trust something or to audit the code paths that go from user input to ledger entry...
<stellar-slack> <scott> The transaction helpers in the client libraries don’t really do anything with the decimal other than convert it into its rational form
<stellar-slack> <scott> When the need arises for client side math, I’ll probably add it to the `Price` objects that each client library has
<stellar-slack> <scott> What I found was that if you look at almost every trading interface in any exchange (I’d love a counter example, if you have one) is that they all ask for decimal input, whereas stellar-core uses rational numbers, and so that conversion needs to happen somewhere. My choice was to put it into the transaction builders, but there is certainly nothing stopping someone from building a `ManageOfferOp` xdr struct
<stellar-slack> attribute on it.
<stellar-slack> <fredolafritte> the price in float is manipulated: https://github.com/stellar/js-stellar-base/blob/master/src/operation.js#L262
<stellar-slack> <fredolafritte> why not directly use pricen and priced instead ?
<stellar-slack> <fredolafritte> trading interface usually uses string to send numbers and let the server handles the convertion
<stellar-slack> <buhrmi> in video games you dont trade with floats
<stellar-slack> <fredolafritte> and in any other financial app
<stellar-slack> <scott> I don’t understand why you think we can use pricen and priced directly. Line 261 is where we convert from a js number into the pricen/priced form
<stellar-slack> <scott> I think I get what you’re saying though… we should be using a bigdecimal library from string -> bigdecimal -> rational, correct?
<stellar-slack> <scott> @buhrmi: could you give an example?
<stellar-slack> <buhrmi> an example of what :O
<stellar-slack> <scott> > in video games you dont trade with floats
<stellar-slack> <buhrmi> counter strike
<stellar-slack> <fredolafritte> bigdecimal is a solution
<stellar-slack> <buhrmi> any game that has currency
<stellar-slack> <buhrmi> mhh not true
<stellar-slack> <buhrmi> i was thinking about RPGs
<stellar-slack> <buhrmi> where you have gold coins
<stellar-slack> <buhrmi> you dont trade 0.3 gold coins
<stellar-slack> <buhrmi> you trade 100 silver for 1 gold
<stellar-slack> <buhrmi> or 100 cents for 1 dollar
<stellar-slack> <buhrmi> but not 0.01 dollars
<stellar-slack> <buhrmi> but it's okay
<stellar-slack> <scott> and how do you price it when they don’t line up on base 10 digit boundaries
<stellar-slack> <buhrmi> the ledger does not contain floats. that's all that matters
<stellar-slack> <buhrmi> bigint comes out of the ledger, bigint goes back in to ledger
<stellar-slack> <fredolafritte> https://bitcointalk.org/index.php?topic=13837.0
<stellar-slack> <fredolafritte> "Please learn to use fixed-point arithmetic, people... this is not a pet project anymore, it's a financial application"
<stellar-slack> <buhrmi> if people are concerned that the stellar libs are doing something wrong, they should develop with the ledger data directly
<stellar-slack> <buhrmi> @scott: i don't know ^^
<stellar-slack> <scott> I think we’re on the same page buhrmi, you’re essentially arguing for using rational math, which is what stellar-core uses. I’ll fix the client libraries such that we don’t interpret user input as float prior to converting the input to rational numbers
<stellar-slack> <buhrmi> hmmm
<stellar-slack> <buhrmi> i think i am mistaken
<stellar-slack> <buhrmi> when using int you can't express "2 for 3"
<stellar-slack> <buhrmi> rational numbers all the way lol
<stellar-slack> <buhrmi> rational prices
<stellar-slack> <fredolafritte> +1
<stellar-slack> <buhrmi> @scott: yeah i think accepting prices as rational numbers instead float would make sense
<stellar-slack> <scott> It’ll be an either/or sort of thing… I must still accept decimal numbers encoded in a string (which will then be approximated into rational form), since that is going to be the normal usecase, but I’ll also accept Price structs directly (which are rational).
<stellar-slack> <jed> lab: I think we agree. we are only focusing on stellar-core and horizon right now. It will be awhile before we create a client for end users. The hope is that one is created by the community.
<stellar-slack> <buhrmi> weed is now legal in the us right?
sacarlson has quit [Ping timeout: 246 seconds]
de_henne has quit [Remote host closed the connection]
<stellar-slack> <scott> @buhrmi: it’s still illegal at the federal level, but most states have medical exemptions and a handful of states have it legal for recreational use
sacarlson has joined #stellar-dev
<stellar-slack> <jed> dzham: can you see if you are still having issues on the testnet?
sacarlson has quit [Ping timeout: 246 seconds]
sacarlson has joined #stellar-dev
<stellar-slack> <jed> actually old testnet is down for a bit
<stellar-slack> <jed> dzham: ok now can you let me know if you are still having issues
<stellar-slack> <jed> I wiped it so you may have to remake accounts
<stellar-slack> <jed> we are changing the name of `js-stellar-lib` to `js-stellar-sdk`
<stellar-slack> <jed> same with ruby one
<stellar-slack> <scott> note: the ruby library will still be named `stellar` from the context of ruby… the name will be the same in rubygems and you’ll still get at it using `require ‘stellar’`… jed is just changing the github repo names
<stellar-slack> <scott> Also, if anyone wants to review some ruby: https://github.com/stellar/horizon/pull/43
Vinnie_win has quit []
<stellar-slack> <scott> FYI: I’m updating horizon (and go-horizon) on testnet. You may see some errors if interacting with them
<stellar-slack> <scott> updated, back up and running