Hello all -

I've put together some custom package repositories for RHEL/CentOS 7 - only x86_64 available

http://awel.domblogger.net/7/php/

That has php 5.6.0 - which isn't something I would suggest putting on a server (I would but I like pain) as it is fresh and does break some webapps, but if you develop php webapps, you should probably have 5.6.x on a dev server.

http://awel.domblogger.net/7/media/

That has FFmpeg and dependencies, I had a need for an FFmpeg build capable of eating almost anything and spitting out HTML5 formats - I think that one does it.

http://awel.domblogger.net/7/gstreamer/

That depends upon the media repository for dependencies. The GStreamer in RHEL/CentOS 7 is old, too old to decode VP9 - so that GStreamer replaces the stock GStreamer with current.
It also provides many plugins that RHEL/CentOS + EPEL do not.

http://awel.domblogger.net/7/misc/

Misc. crap - basically rebuilds of Fedora 20 src.rpms that for whatever reason are not in EPEL

http://awel.domblogger.net/7/crypto/

My crypto-currency repo

Right now the only real thing of interest there is namecoin - the bitcoin RPMs are just rebuild of http://www.ringingliberty.com/bitcoin/ so just use his if bitcoin is all you are interested in.

The namecoin right now is only daemon (that's all I ever use anyway) but I will build the qt in a week or so.
And I will then add quarkcoin and then start work on getting armory packaged.

-=-

Anyway - for CentOS/RHEL 7 users - I thought those might be of interest to some of you.

CentOS 7 really is a decent Desktop OS once you replace the Gnome 3 shit with Mate and update the GStreamer.

All of the repos use EPEL for dependencies, but no other external repos are needed.
All packages are built with mock and signed.

Setting up bitcoin-qt on a new laptop.

Need to tell it to use my desktop as a block source so it goes faster.

There use to be a ~/.bitcoin/bitcoin.conf file where I could do that, but I don't see that file.

What is the current method for telling it to use the "caught up" client I have running on another machine on my gigabit wired LAN ??

Looking via google seems to all point to old documentation for 0.8 client - which used ~/.bitcoin/bitcoin.conf

Hello,

I have an idea for a utility coin.

A utility coin is a coin whose function is to provide a useful de-centralized distributed feature rather than to act as a monetary currency. Such coins are probably a poor investment if you are looking to make a profit, since their value should be tied to the utility provided and what demand there is for it.

Namecoin is an example of a utility coin.

Namecoin actually is what I would choose to fork to make my idea happen, as the utility I want to provide is similar but different.

What I want to provide is namespaced key=>value on a blockchain (much like namecoin does) but where you can purchase exclusive use of a namespace. With namecoin, you can't buy exclusive use of a namespace.

Some potential uses:

* OpenNIC is a centralized alternative to ICANN. Even though they are centralized, they could register namespaces within this coin and manage domains the same way namecoin manages the .bit TLD - except since they own the namespace, only they can create key=>value pairs within the namespaces they manage.

The benefit to OpenNIC would be that those wanting to run a DNS server including their domains could do so using the blockchain as a source. This de-centralized distribution of their domains would eliminate single points of failure.

* GPG Keyring - a company could run its own GPG keyring so that it is easy for their partners to get public keys for people within the company they need to contact, either via encrypted messages or to check a signature.

If I know acme corporation owns the namespace acme/ and I need to check the GPG signature on an e-mail sent to me by john@acme.tld then I can look up acme/john to get his GPG public key.

* Age Verification

One of the things slowing down acceptance of crypto-currencies in the adult content industry is a lack of age verification.

A company could set up an age verification service where they give a user a username and then when they have verified the users age - make an entry in their database containing an sha256 has of the username + e-mail address.

So when John Wanker wants to buy webcam time using bitcoin or whatever, her signs up at the webcam site and provides his username with the age verification service and the e-mail address used for age verification. The webcam site can check to see if the sha256 sume of the username + e-mail address matches the checksum associated with the username in the blockchain and if it does, send a confirmation e-mail to that address he has to reply to in order to verify his age.

* Altcoin Trade Trust

A namespace could be set up where users of bitcointalk can register that includes their bitcointalk username and a GPG key. From that, strangers doing business can use GPG to make sure they are talking to the actual person and not a hacked account, and then look at their ratings on bitcointalk to see if it is a trustworthy person to trade with or to use as an escrow agent for a trade.

-=-

There are lots of uses for this kind of a service where a namespace is restricted to the buyer of the namespace.

When a user _has_ purchased a namespace, there would be special key=>value pairs.

* First would be an optional api key that points to a web page describing how they use the json values in the key=>value pairs. This is for third party developers who want to use the data in the namespace.

For example, an age verification service may want to specify the age verification method (e.g. credit card, in person ID verification, etc.) so that websites that have certain policies about how an age was verified can still confirm an age and know how it was verified. Having an API to look at will tell them exactly what all the data means.

* Second would be an optional authorized key=>value pair with a specific format, containing the public address of people authorized to make changes to the namespace and what kind of changes. That would allow they private key used to create a namespace to be put into cold storage.

-=-

My idea is that buying a namespace would be rather expensive compared to making entries in the namespace, say 20 coins. Adding an entry would be say 0.01 coins.

Block time would be about every 15 minutes, since it isn't intended to be used as a currency fast confirms are not needed and this would reduce number of blocks where only transaction is the block reward.

Initial two weeks, mining reward would be 3 coins.

Following initial week, mining reward jumps to 48 coins to be halved about every year until it drops back down to 3 coins. Never drops below 3 coins. Finding 7 blocks is thus enough to buy a namespace when reward is at minimum, and have a coin leftover to make some entries.

What would be kind of cool is if coins used to purchase namespaces and entries in a namespace rather than being destroyed were split up between the last 5 addresses that received a block reward.

For block hashes I would want to use the Quark 6+ model because I want coins to mostly be mined by people, not professional miners.

-=-

No qt GUI - this isn't intended as a currency to spend, and is targeted as a data store for those providing a service, not the individual. So maintaining a GUI codebase is not necessary. Let third parties provide that if necessary. I just want the daemon.

Thoughts?

If say pool X has 40% of hashing power, then 51% of the remaining 60% is (rounded) 31%

So could 31% of hashing power combined with a successful DoS on pool X pull off a 51% attack?

Granted the block time would be increased during the attack but this seems like a danger of the ASIC pools that have taken over bitcoin mining.

Trying to build quarkcoin headless daemon in OS X

OS X 10.9.4
xcode 5.1.1
build dependencies installed via MacPorts

It appears there are some issues resulting from OS X native build tools being more pedantic than GNU build tools.

I’m not a c++ guy but a lot of the style actually doesn’t look that different from php. But anyway… I need some help.

I already figured out the following fix:

Code:

--- blake.c.orig	2014-07-18 15:06:24.000000000 -0700
+++ blake.c	2014-07-18 17:26:55.000000000 -0700
@@ -883,7 +883,7 @@
 		sph_enc32be_aligned(u.buf + 60, tl);
 		blake32(sc, u.buf, 64);
 	}
-	out = dst;
+	out = (unsigned char *)dst;
 	for (k = 0; k < out_size_w32; k ++)
 		sph_enc32be(out + (k << 2), sc->H[k]);
 }
@@ -988,7 +988,7 @@
 		sph_enc64be_aligned(u.buf + 120, tl);
 		blake64(sc, u.buf, 128);
 	}
-	out = dst;
+	out = (unsigned char *)dst;
 	for (k = 0; k < out_size_w64; k ++)
 		sph_enc64be(out + (k << 3), sc->H[k]);
 }

However I am still getting a different type of build error I need help with.

Code:

blake.c:1002:2: error: no matching function for call to 'blake32_init'
        blake32_init(cc, IV224, salt_zero_small);
        ^~~~~~~~~~~~
blake.c:791:1: note: candidate function not viable: cannot convert argument of incomplete type 'void *' to 'sph_blake_small_context *'
blake32_init(sph_blake_small_context *sc,

That type of error is repeated a dozen times but with different functions.
Here is what blake32_init looks like - which happened to be where line 791 is, I suspect that’s the real error:

Code:

static void
blake32_init(sph_blake_small_context *sc,
        const sph_u32 *iv, const sph_u32 *salt)
{
        memcpy(sc->H, iv, 8 * sizeof(sph_u32));
        memcpy(sc->S, salt, 4 * sizeof(sph_u32));
        sc->T0 = sc->T1 = 0;
        sc->ptr = 0;
}

Can anyone suggest how I need to fix that? I suspect it is something similar to first patch where the code just isn’t explicit for OS X but…

Thanks

I'm not making a coin.

Just an idea to float out to those who are making them (and please don't without good reason)

10% pre-mine but the pre-mine is locked, blockchain won't allow them to be spent.

Every X blocks, where X is the typical number that would be generated in 3 months, 1/12 of the pre-mined coins are released and can be spent by the devs.

This would prevent a pump and dump as it would take 3 years for all the pre-mined coins to be made available and would incentivize the devs to keep the project on track - if the coin goes down the dumper, then the value of the coins in the pre-mine go down the dumper too before they can be sold.

But for the first 3 years of development it could inject much needed dev funds to the dev team if the coin does well.

Thoughts?

Code:

[alice@localhost genKeyPairs]$ nslookup www.qrk.cc
Server:		192.168.0.1
Address:	192.168.0.1#53

Non-authoritative answer:
Name:	www.qrk.cc
Address: 141.101.116.112
Name:	www.qrk.cc
Address: 141.101.117.112

[alice@localhost genKeyPairs]$ nslookup www.quarkcoins.com
Server:		192.168.0.1
Address:	192.168.0.1#53

Non-authoritative answer:
www.quarkcoins.com	canonical name = quarkcoins.com.
Name:	quarkcoins.com
Address: 209.59.158.135

Different IP address, are they the same or is one a phish?

Let's say US Government wanted to get into alt-coin business

We'll call it CUSD - Crypto USD

Blockchain is open for inspection but blocks are created by government, not miners.

Government decides how many new coins get created - just like they do with paper money.

A user can create a "bond" transaction - where he specifies that X coins will not be spent for Y years.

A reward for doing this is a share of the new coins created, proportional to his share of the total bond transactions.

Creating a bond transaction allows users to hoard while at the same time since the user can not spend then, the value in the bond transaction can be made available for loans.

So in this way, hoarding coins does not reduce the coins in circulation.

If a user who creates a bond transaction finds he needs money, he can sell the bond transaction to someone else.

-=-

Totally crazy or might it work?

I posted this in development but I think this might be better suited since it is not code related. So I locked that one. Discuss here.

Hello,

I'm designing a few fonts and I would like to add currency glyphs for bitcoin and namecoin.

Until there is an official currency symbol with an official unicode assignment, any bitcoin symbol in a font (that is not an existing glyph anyway) will need to be assigned to the private area.

I would suggest that we pick a number in the private area that font creators can use for bitcoin until such time that an official unicode assignment is made.

More on private use areas: http://en.wikipedia.org/wiki/Private_Use_Areas

I would suggest that we attempt to pick a number that is not already in public use. For example, Apple Computer uses U+F8FF and http://www.evertype.com/standards/csur/ has un-official assignment for many scripts not in unicode, such as Science/Fantasy fiction languages.

Having an agreed upon Private Use number for the Bitcoin currency symbol would allow us have consistency when changing fonts, the character could easily be used in PDF and ePub documents by embedding the font, and could probably even be used on web pages by specifying webfont that use that number for bitcoin.

My suggestion is that we only agree upon a number, allow the font authors to decide which proposal they want to draw in their font. I like the one by Pander and that is what I will add to my fonts, but by using a consistent number in the unicode private range, until there is an official symbol agreed upon, let the font typographers pick what they like.

When we do pick a number, https://en.bitcoin.it/wiki/Bitcoin_symbol should be updated to reflect it, so that font creators who don't come by the forum but google the currency symbol can find it.

Thoughts?

Hello,

I'm designing a few fonts and I would like to add currency glyphs for bitcoin and namecoin.

Until there is an official currency symbol with an official unicode assignment, any bitcoin symbol in a font (that is not an existing glyph anyway) will need to be assigned to the private area.

I would suggest that we pick a number in the private area that font creators can use for bitcoin until such time that an official unicode assignment is made.

More on private use areas: http://en.wikipedia.org/wiki/Private_Use_Areas

I would suggest that we attempt to pick a number that is not already in public use. For example, Apple Computer uses U+F8FF and http://www.evertype.com/standards/csur/ has un-official assignment for many scripts not in unicode, such as Science/Fantasy fiction languages.

Having an agreed upon Private Use number for the Bitcoin currency symbol would allow us have consistency when changing fonts, the character could easily be used in PDF and ePub documents by embedding the font, and could probably even be used on web pages by specifying webfont that use that number for bitcoin.

My suggestion is that we only agree upon a number, allow the font authors to decide which proposal they want to draw in their font. I like the one by Pander and that is what I will add to my fonts, but by using a consistent number in the unicode private range, until there is an official symbol agreed upon, let the font typographers pick what they like.

Thoughts?

If I'm having trouble in Linux getting /dev/random to generate data, is it safe to use the haveged daemon to help generate entropy or is that not a really good idea? With that daemon running, /dev/random works just fine, I just want to make sure I'm not defeating the purpose.

With a watching-only wallet, is the seed needed to generate private keys still stored in armory?

I've read, though I have not really looked at the process but I've read you can use a public key to generate new addresses which can only be spent using the private key that originally produced the public key.

So the thought it - create a paper wallet with private key and import public key into a wallet that generates addresses from it, allowing you to deposit to a new address each time you add money to the paper wallet.

We know someone is sending small amounts of money to random addresses and that got me thinking.

It wouldn't be too hard to look at the blockchain and find public addresses that are quite like a paper wallet. Lots of transactions going in spread over time, no transactions going out.

So once a criminal identifies sizeable paper wallets, criminal then starts doing micropayments to addresses that deposited looking for how they are then spent in hopes of identifying who owns the paper wallet.

Then when they identify who owns the paper wallet, thugs come and extract the private key.

But if the address used to deposit changes every time, it will difficult if not impossible to determine it is a paper wallet or how much is in it.

The issue isn't just random thugs. Maybe John bought something from me and thus already knows who I am, followed the transaction and saw the bitcoins end up in an address that has large quantity of BTC. John comes to my house with a hammer...

Anyway, to work the smoothest it would need client intergration so I could import a public key into client and label it, and anytime I want to deposit I ask to send to the label and it generates new public address based that doesn't have a private key.

Any clients do this?

I searched for lite client SSL and got no results.

The lite clients, as I understand it, query external servers for the data they need. Is this done over SSL?

Reason I ask is if it is not done over SSL, then couldn't someone sniffing the wire identify public addresses to IP address thus making it easier to identify owners of specific addresses?

Issue I am having is not bitcoin specific but it is related to a bitcoin article I am writing.
It is with using foreign characters in a PDF bookmark, specifically mu.

I posted at

http://latex-community.org/forum/viewtopic.php?f=48&t=23567

Lot of geeks here, lot of math geeks here, so hoping maybe some LaTeX users who have come across this and know a solution.

I'm not a guru when it comes to economic terms and what they mean in economic theory.

However I was looking at http://en.wikipedia.org/wiki/Creative_destruction

It occurred to me that bitcoin and/or crypto-currencies may be a form of creative destruction in the banking industry.

As people become less reliant upon tradition banking in favor of crypto-currencies, traditional banking may be in for a world of hurt, an old financial order being replaced by a new.

Is this a valid look at what bitcoin is doing?

I took a class on macro-economics my freshman year of college when I was 17, and I do not want to admit how long ago that was.
Let me put it to you this way, we still used 5.25 floppies and the `stoned' and `jerusalem' boot sector viruses were a big problem.
Lotus 1-2-3 was the app you had to know for business. The science classes however preferred Quatro Pro.

I got an A but I can't name a single concept I learned that semester.

I'd like to hear the hear the opinion of people who have studied economics more recently.

Thank you.

I did a search and the only results don't look like they apply.

Running Armory in Fedora 18, built an RPM. The spec file needs work but anyway -

First - Fedora does python bytecompile when building python packages. They check for syntax during that.
If you have SELinux enabled (as it should be on any Linux box running bitcoin) you have to bytecompile at package time or SELinux pukes when python tries to when run.

To get it to bytecompile I had to apply the following patch:

Code:

diff -ur BitcoinArmory-0.87.2/googlecode_upload_release.py BitcoinArmory-0.87.2-patched/googlecode_upload_release.py
--- BitcoinArmory-0.87.2/googlecode_upload_release.py	2013-07-12 01:50:56.000000000 -0700
+++ BitcoinArmory-0.87.2-patched/googlecode_upload_release.py	2013-07-12 03:46:56.956465040 -0700
@@ -288,10 +288,10 @@
              ['_win32.msi',         'Version %s for Windows 32- and 64-bit', ['WindowsMSI',    'OS-Win32']], \
              ['_OSX.dmg',           'Version %s for Mac/OSX 10.8+',          ['MacosxDMG',     'OS-MacOSX']], \
              ['_sha256sum.txt.asc', 'Version %s SHA256 hashes of installers',['HashesSHA256']]]
-             ['_OfflineBundle_Ubuntu-10.04-32bit.tar.gz',     'Version %s Offline Bundle for Ubuntu 10.04-32bit', \
-                                                                             ['DebianPackage', 'OS-Linux32', 'Offline']], \
-             ['_OfflineBundle_Ubuntu-10.04-64bit.tar.gz',     'Version %s Offline Bundle for Ubuntu 10.04-64bit', \
-                                                                             ['DebianPackage', 'OS-Linux64', 'Offline']], \
+             #['_OfflineBundle_Ubuntu-10.04-32bit.tar.gz',     'Version %s Offline Bundle for Ubuntu 10.04-32bit', \
+             #                                                                ['DebianPackage', 'OS-Linux32', 'Offline']], \
+             #['_OfflineBundle_Ubuntu-10.04-64bit.tar.gz',     'Version %s Offline Bundle for Ubuntu 10.04-64bit', \
+             #                                                                ['DebianPackage', 'OS-Linux64', 'Offline']], \
              # Actually, we can just include the hashes of the offline bundles in the sha256sums file
              #['_OfflineBundle_Ubuntu_10.04_32bit.tar.gz.sig', 'Detached Signature for %s 32-bit offline bundle', \
                                                                              #['DetachedSig', 'OS-Linux32', 'Offline']], \

The complaint was indentation error. I figured (hoped) the _OfflineBundle_Ubuntu stuff really didn't matter to Fedora but if it does, someone familiar with python might want to look at the indentation of the lines I commented out.

-=-

Got it installed and launched, and got an error reported in the client:

Code:

/sbin/bitcoind: line 21: /etc/bitcoin/bitcoin.conf: Permission denied

bitcoind is owned by the bitcoin-server package installed, and I'm not sure what armory wants with that file, but I do not have the server enabled so it looks like armory is trying to start the server but doesn't have permission to. Is that case?

What does armory do with the server? Would setting the server to start at system boot give armory what it needs? I've been using bitcoin-qt which does not use the server, it seems that the bitcoin-qt rpm I have actually runs its own server within the user directory to avoid issues of permissions.

Is that something I am going to need to do with armory?

I do not want to have un-install the bitcoin-server rpm.

Thanks for suggestions. I'll look at this more tomorrow.
I really like the creation of paper backup at the start, that's awesome.

Since I joined this forum, I've seen more than one thread where someone forgot their passphrase to their encrypted wallet and wanted help cracking it.

What if as an option, a user could enter the address of, say, a paper wallet beforehand.

Then whenever the user enters the passphrase to decrypt the wallet, a transaction is created and signed but not sent - that would send all inputs to that address. This transaction is then stored in the data directory.

In the event that the user can not recall their passphrase, that transaction could then be sent so at least the user has not lost all their coins.

Obviously it would not be able to include inputs created after the last time the user decrypted the wallet.dat but it could include all inputs that could be spent the last time wallet.dat was decrypted.

It might soften the financial blow to some people when they forget their passphrase.

Speaking as someone who sometimes has memory problems due to head injuries, I think it would be a welcome addition, and might even convince me to not keep my passphrase on a piece of paper in my desk drawer (There are days I can not even remember my own phone number).

I use Android but heh, other phones this would be cool too.

What I want is a bitcoin client that is locked. Numeric keyword required to open it. Say 52864 - obviously set by the user.

Three consectutive failed attempts and all funds in it are sent to my `oh shit my phone was stolen' address.

It wouldn't stop the person who takes the phone apart to read the data card, but it may prevent my coins from vanishing if my phone fell out of my pocket on the bus and a kid picks up.

2-step verification to spend. Optional of course.

User opt in. Phone number tied to sending address sent to 2-step verification server. Maybe even google's could be used.

When client wants to spend, text sent to phone. User enters code. Miners have to verify it is valid before adding the transaction to the block.

It couldn't be that hard, would be neat to see it as an option in bitcoin, but alt-currencies are a good place to test how practical such changes are first.

So what alt currency wants to try to implement it?