[ANN][XCP] Counterparty - Pioneering Peer-to-Peer Finance - Official Thread

[panonym]

I still haven't read anything satisfactory about wagers centralisation/trust-issue with the implemented betting system.
As I understand it, I will never use it.
A single feeder guy arbitrarily choose the result... deciding who win or lose. No guarantee feeder result = actual result.
Big trust issue there.

Feed operators have a strong incentive to be honest, because those betting can easily see which feeds are and which are not. Honest operators will obviously make more in the long run.

Operator is anonymous...
Sure, let's be honest. Slowly more people will make higher bet.
Until big money is tempting enough, he bet on his own feed the maximum he can, feed what he need to win against all odd.
Hop 'one rich guy and hundreads fucked.
Let's start again!

It's not a possibility, it WILL happen.

[PhantomPhreak]

I still haven't read anything satisfactory about wagers centralisation/trust-issue with the implemented betting system.
As I understand it, I will never use it.
A single feeder guy arbitrarily choose the result... deciding who win or lose. No guarantee feeder result = actual result.
Big trust issue there.

Feed operators have a strong incentive to be honest, because those betting can easily see which feeds are and which are not. Honest operators will obviously make more in the long run.

Operator is anonymous...
Sure, let's be honest. Slowly more people will make higher bet.
Until big money is tempting enough, he bet on his own feed the maximum he can, feed what he need to win against all odd.
Hop 'one rich guy and hundreads fucked.
Let's start again!

The bigger the bet is, the higher is the fee paid to the operator.

The question isn't whether or not a feed operator could screw people over for money. The question is whether or not he would make more money by being honest. A good reputation will be very valuable, as it is on the Silk Road.

[panonym]

The bigger the bet is, the higher is the fee paid to the operator.

The question isn't whether or not a feed operator could screw people over for money. The question is whether or not he would make more money by being honest. A good reputation will be very valuable, as it is on the Silk Road.

I understand what you say. But it's not convincing enough for me to use your betting system.
'certainly not the only one.
Sure an honest feeder will make more money as time pass, reputation increase, ++money as bet rise higher.
So thousands of people will want to become feeder.
Random guy must choose randomly which feed chose, or after 2 month, choosing the one with the best reputation (discovered through hard research, nothing easy).

Might work for let's say 5 honest guy.
Creating a tiny feeder elite.
Yet if bet start to be REALLY huge, at one moment one of the top 5 have a choice:
A: continue to be honest, I already made x$, if I continue in 2 month I might have 2x!
B: months of time spent to build a reputation (hard to find), made y$ in 6 month. Do I wait 4 month more, playing nice, to perhaps reach 2y$ ? or do I screw them all and take 16y$ now?
Supposedly 16y$ = 1'000'000$, temptation is huge

He could make more, but 'need +1yr and result is uncertain. As is BTC&XCP value as time pass.
Big Big now? or moderately great over very long time? 'arrive one point for many...
Tony had a very good reputation on SR I heard...

[Winslow Strong]

Community participation should be better with Counterparty considering that you have to literally burn BTC to receive XCP and there will not be issues of enriching the developer because no one has access to the burnt coins. This "sounds" less scammy. In addition we have an actual working code base from the start.

Also I expect the initial distribution of XCP to be slightly more balanced given the additional exposure of additional protocols running on top of bitcoin compared to a few months back

The downside though is without reservation of any bounties or incentives, moving forward it's highly likely that future  developements will need to be driven by the community

This seems like an unfortunate tradeoff to (?have to?) make.  Is the scam risk really perceived to be so high, given that so much code is actually already finished, open source, and demonstrably working?  Then again, mastercoin has had tons of funds available for bounties, and it hasn't progressed very quickly.  Does anyone understand what is taking them so long?

[panonym]

Counting money they made

[xnova]

Counting money they made

Well, there's Counterparty business idea #2,329, which would be a system that indexes, ranks and orders bookies (i.e. feed-for-bet creators) on trustworthiness. Without this, one could simply look at the history of the bet-on broadcasts from that address to get an idea of the reputation/honesty of the feed operator. Yes, none of it is perfect, but neither is buying something on eBay or craigslist. Centralized organizations are welcome to (and I believe will) spring up around Counterparty to offer such reputations/warranty/insurance services, and if thee is a way we can extend or better the core protocol while maintaining decentralization and overall simplicity, then we'll definitely investigate it.

[jimhsu]

I get the following error when trying to make a test order:

C:\Program Files\counterpartyd>COUNTERPARTYD order --from [redacted] --get-quantity 0.1 --get-asset BTC --give-quantity 1 --give-asset XC
P --expiration 65535
Traceback (most recent call last):
  File "C:\Python33\lib\site-packages\cx_Freeze\initscripts\Console3.py", line 2
7, in <module>
  File "C:\counterpartyd_build\dist\counterpartyd\counterpartyd.py", line 485, i
n <module>
  File "C:\counterpartyd_build\dist\counterpartyd\lib\order.py", line 31, in cre
ate
  File "C:\counterpartyd_build\dist\counterpartyd\lib\bitcoin.py", line 292, in
transaction
  File "C:\counterpartyd_build\dist\counterpartyd\lib\bitcoin.py", line 170, in
serialise
  File "C:\Python\64-bit\3.3\lib\importlib\_bootstrap.py", line 1567, in _find_a
nd_load
  File "C:\Python\64-bit\3.3\lib\importlib\_bootstrap.py", line 1514, in _find_a
nd_load_unlocked
  File "C:\Python\64-bit\3.3\lib\importlib\_bootstrap.py", line 313, in _call_wi
th_frames_removed
  File "C:\Python\64-bit\3.3\lib\importlib\_bootstrap.py", line 1567, in _find_a
nd_load
  File "C:\Python\64-bit\3.3\lib\importlib\_bootstrap.py", line 1531, in _find_a
nd_load_unlocked
ImportError: No module named 'pycoin'

Win7, x64, latest windows installer (0.1.0.0)
Thanks

I did an update on an existing counterpartyd install. Is this causing the problem?

Searching further, it appears that the windows install is missing the pycoin module. How do I "embed" the module into the exe (installing it into my Python directory didn't seem to work)?

Confirmed, this works with the manual python install. The windows installer should be updated.

[xnova]

I get the following error when trying to make a test order:

ImportError: No module named 'pycoin'

Win7, x64, latest windows installer (0.1.0.0)
Thanks

I did an update on an existing counterpartyd install. Is this causing the problem?

Searching further, it appears that the windows install is missing the pycoin module. How do I "embed" the module into the exe (installing it into my Python directory didn't seem to work)?

I just updated the build system to include pycoin. Also created new installer. Please grab the newest installer .exe from github (https://github.com/xnova/counterpartyd_binaries) and try it again, and let me know if you still have issues with it. Thanks!

[jimhsu]

The transaction seemed to go through, but I am stuck with Status: 0/unconfirmed without any broadcasts for about 20 minutes so far. Bitcoin-qt is well connected (10 active connections). How long does it take typically?
It did actually go through, though not for 30 minutes.


Also, more error reports:

1. rpc-user and rpc-password should be prefixed with "bitcoind-" in the autogenerated config from the installer. Or at least make that part of it clear.
2. I still get the u2026 unicode errors in windows when doing "counterpartyd market".

Open Orders
Traceback (most recent call last):
  File "counterpartyd.py", line 608, in <module>
    market(args.give_asset, args.get_asset)
  File "counterpartyd.py", line 41, in market
    print(str(table.get_string(sortby='Price')))
  File "c:\Python33\lib\encodings\cp437.py", line 19, in encode
    return codecs.charmap_encode(input,self.errors,encoding_map)[0]
UnicodeEncodeError: 'charmap' codec can't encode character '\u2026' in position
483: character maps to <undefined>

This worked before (albeit being blank), so obviously some of the new transactions that have come up in the last day are causing this error.

[SyRenity]

I wonder if it possible to know the coins distribution, so the whale sale leading to crash (aka what happened to MSC, and apparently NXT) can be detected (and ignored)?

Or due 1 BTC limit, this won't be possible?

[peled1986]

Or due 1 BTC limit, this won't be possible?

Its 1 BTC per address meaning  1 person can "burn" as many BTC as he wants - it would just take him more time.

[jimhsu]

Agreed, the database rebuild should only take 20-30 minutes at most (that's around what it took for me). Try looking at your processes and make sure nothing is hogging the CPU or I/O.

I'd love it to take only 30min.
I'm at block 279148... So it does take hours and hours and hours on my end.

Any help trying to figure why (from anyone) is appreciated.
From speedtest.net I got: Ping 7ms, 25.71Mbps on DL, 2.41 Mbps on UP

I only know the command top to check CPU/mem usage. Here is the result:

Code:

top - 12:23:52 up 2 days, 45 min,  4 users,  load average: 2,11, 1,97, 1,89
Tasks: 214 total,   1 running, 212 sleeping,   0 stopped,   1 zombie
%Cpu(s): 10,0 us,  2,5 sy,  0,3 ni, 41,6 id, 45,6 wa,  0,0 hi,  0,0 si,  0,0 st
KiB Mem:   2005716 total,  1762632 used,   243084 free,     3100 buffers
KiB Swap:  2050044 total,  1274540 used,   775504 free,   443924 cached

  PID USER      PR  NI  VIRT  RES  SHR S  %CPU %MEM    TIME+  COMMAND          
 2199 panonym   20   0 1828m 618m  23m S  11,3 31,6 180:13.59 firefox          
  987 root      20   0  568m  43m  17m S   2,7  2,2  47:02.65 Xorg              
19342 panonym   20   0  155m  17m 2288 S   2,7  0,9  18:28.92 python            
 2194 panonym   20   0 1693m 152m 9368 S   2,0  7,8  35:46.45 btc              
 2003 panonym   20   0 1255m  25m 7904 S   1,7  1,3  42:11.56 compiz            
 7274 panonym   20   0 1004m  55m 9524 S   1,7  2,9  21:22.09 plugin-containe  
 8413 panonym   20   0 1335m  18m 4520 S   1,3  0,9  12:06.88 transmission-gt  
 2107 panonym   20   0  641m  12m 5144 S   0,7  0,6   0:19.76 gnome-terminal    
   17 root      20   0     0    0    0 S   0,3  0,0   0:55.67 rcu_sched        
 7442 panonym   20   0 1543m 145m 2092 S   0,3  7,4   4:39.35 ltc              
10778 panonym   20   0 29080 1508 1016 R   0,3  0,1   0:01.64 top

Of course it fluctuate (%CPU firefox 2.3 only after checking 2min later)

I'm short on RAM, 2 GB. Could it be the matter? (2GB wasn't that low a few years ago)

Code:

free -m
             total       used       free     shared    buffers     cached
Mem:          1958       1883         75          0          5        528
-/+ buffers/cache:       1349        609
Swap:         2001       1241        760

public upload of ~/counterpartyd_build & ~/.config/counterpartyd would still be appreciated.

Your firefox instance seems to be pretty greedy. Kill it with fire?

In other news, 16GB dedicated servers can be had for not much more than $20/month these days (albeit a very special deal on lowendbox). (Yes, a real dedicated, KVM and power management and all).

[kdrop22]

The bigger the bet is, the higher is the fee paid to the operator.

The question isn't whether or not a feed operator could screw people over for money. The question is whether or not he would make more money by being honest. A good reputation will be very valuable, as it is on the Silk Road.

I understand what you say. But it's not convincing enough for me to use your betting system.
'certainly not the only one.
Sure an honest feeder will make more money as time pass, reputation increase, ++money as bet rise higher.
So thousands of people will want to become feeder.
Random guy must choose randomly which feed chose, or after 2 month, choosing the one with the best reputation (discovered through hard research, nothing easy).

Might work for let's say 5 honest guy.
Creating a tiny feeder elite.
Yet if bet start to be REALLY huge, at one moment one of the top 5 have a choice:
A: continue to be honest, I already made x$, if I continue in 2 month I might have 2x!
B: months of time spent to build a reputation (hard to find), made y$ in 6 month. Do I wait 4 month more, playing nice, to perhaps reach 2y$ ? or do I screw them all and take 16y$ now?
Supposedly 16y$ = 1'000'000$, temptation is huge

He could make more, but 'need +1yr and result is uncertain. As is BTC&XCP value as time pass.
Big Big now? or moderately great over very long time? 'arrive one point for many...
Tony had a very good reputation on SR I heard...

Agree with your point.
However, centralized feeds seem to be the only workable option right now. Protoshares has proposed an alternative model, based on bid/ask prices for their BitUSD, however that model seems a little bit naive to me. That just because we call something BitUSD, the price of that will track the USD.
If you have an alternative model, please feel free to propose it.

[SyRenity]

Its 1 BTC per address meaning  1 person can "burn" as many BTC as he wants - it would just take him more time.

Yep, this what I meant.

Btw, congrats to devs for 700 BTC invested (burnt) XCP seems on it's way to 1K.

(If someone would say me a month ago people will willingly throw away 1M$+, I would say he needs to get his head checked! ).

[mtbitcoin]

The transaction seemed to go through, but I am stuck with Status: 0/unconfirmed without any broadcasts for about 20 minutes so far. Bitcoin-qt is well connected (10 active connections). How long does it take typically?
It did actually go through, though not for 30 minutes.


Also, more error reports:

1. rpc-user and rpc-password should be prefixed with "bitcoind-" in the autogenerated config from the installer. Or at least make that part of it clear.
2. I still get the u2026 unicode errors in windows when doing "counterpartyd market".

Open Orders
Traceback (most recent call last):
  File "counterpartyd.py", line 608, in <module>
    market(args.give_asset, args.get_asset)
  File "counterpartyd.py", line 41, in market
    print(str(table.get_string(sortby='Price')))
  File "c:\Python33\lib\encodings\cp437.py", line 19, in encode
    return codecs.charmap_encode(input,self.errors,encoding_map)[0]
UnicodeEncodeError: 'charmap' codec can't encode character '\u2026' in position
483: character maps to <undefined>

This worked before (albeit being blank), so obviously some of the new transactions that have come up in the last day are causing this error.

Yes, running into the same issues. Even the order are not working and throwing errors. However I suspect this is only for the Windows builds because if you observe the transactions log there are buy/sell orders being placed



Here is another one for "order"

  File "C:\counterpartyd_build\dist\counterpartyd\lib\bitcoin.py", line 170, in serialise
    from pycoin.ecdsa import generator_secp256k1, public_pair_for_secret_exponent
ImportError: No module named 'pycoin'

[jimhsu]

The transaction seemed to go through, but I am stuck with Status: 0/unconfirmed without any broadcasts for about 20 minutes so far. Bitcoin-qt is well connected (10 active connections). How long does it take typically?
It did actually go through, though not for 30 minutes.


Also, more error reports:

1. rpc-user and rpc-password should be prefixed with "bitcoind-" in the autogenerated config from the installer. Or at least make that part of it clear.
2. I still get the u2026 unicode errors in windows when doing "counterpartyd market".

Open Orders
Traceback (most recent call last):
  File "counterpartyd.py", line 608, in <module>
    market(args.give_asset, args.get_asset)
  File "counterpartyd.py", line 41, in market
    print(str(table.get_string(sortby='Price')))
  File "c:\Python33\lib\encodings\cp437.py", line 19, in encode
    return codecs.charmap_encode(input,self.errors,encoding_map)[0]
UnicodeEncodeError: 'charmap' codec can't encode character '\u2026' in position
483: character maps to <undefined>

This worked before (albeit being blank), so obviously some of the new transactions that have come up in the last day are causing this error.

Yes, running into the same issues. Even the order are not working and throwing errors. However I suspect this is only for the Windows builds because if you observe the transactions log there are buy/sell orders being placed



Here is another one for "order"

  File "C:\counterpartyd_build\dist\counterpartyd\lib\bitcoin.py", line 170, in serialise
    from pycoin.ecdsa import generator_secp256k1, public_pair_for_secret_exponent
ImportError: No module named 'pycoin'

Try the most recent update that phantom pushed.

So far, I've tried orders (both XCP > BTC and BTC > XCP) and issuance and they both seem to work.

[mtbitcoin]

Try the most recent update that phantom pushed.

So far, I've tried orders (both XCP > BTC and BTC > XCP) and issuance and they both seem to work.

Thanks. Just noticed that too.. and it looks like the orders are indeed working

[jimhsu]

If people want to provide their addresses, I want to test transactions by sending them some TEST (lol). Until my bitcoins run out from all the fees ... lol.

BTC/XCP tips appreciated.

[mtbitcoin]

If people want to provide their addresses, I want to test transactions by sending them some TEST (lol). Until my bitcoins run out from all the fees ... lol.

BTC/XCP tips appreciated.

If you want to send some test coins, you are most welcomed   

1L5dKm5ftLfiMBjq3yZDJMgr365J5SHkUC

[becherovka]

Announcement: We've added to Counterparty the ability to imbed and parse data in multi-signature outputs, in addition to OP_RETURNs. This means that all message types (not just burns) now work on mainnet, at regular speed. Counterparty is fully operational (like the second Death Star), without Eligius and without Bitcoin 0.9.

When OP_RETURN goes live, the multi-sig method will probably be retired, just because it's not officially supported and is slightly less elegant. However, the Bitcoin developers have (counter-productively) decided to allow only one OP_RETURN output per tx, so without multi-sig we are limited to 80 bytes per tx. With multi-sig, the limit is much higher. (N.B. Both methods may be combined in a single transaction.)

counterpartyd will continue to function as it always has, except now everything will actually work. As always, be careful: this software is still largely untested.

Did everyone see this announcement?

Are there questions about what was launched on yesterday?