I think you mean fragment, not defragment.

Satoshi is strongly against other clients because if two clients on the network don't agree on something, there's no central judge to sort it out. It could create problems for both groups of users.

Also, Bitcoin is OSS. Why would we need another implementation?

I agree with dwdollar.

MadHatter, will you do the opposite and send cash in the mail in exchange for Bitcoins?

I'd like to see an option (with an RPC interface as well) to limit the number of connections that a Bitcoin client accepts. My home client is connected to 70 different nodes and my poor wimpy router just can't keep up. It's starting to slow my network down, to the point where I'll need to force Bitcoin to ignore connection requests (with -connect=<a node>) if this keeps up.

A connection limiting option would be a great alternative.

Or you could use my "completely unsupported, don't blame anyone if it doesn't work" patch to rpc.cpp that adds listgenerated as a method. Pass it false (./bitcoind listgenerated false) to get only the unmatured coins. No argument or true gives a list of all coins. The interface is going to change, since a list of strings isn't really the way RPC is supposed to work!

Of course, for that, you need to build from the SVN.

The latest version of the patch can always be found at http://www.alloscomp.com/bitcoin.

True, and it is your project. 1.3 it is!

+1 NLS. Version 1.0 sounds better than 1.3.

However, having a signed certificate has no impact on whether or not somebody's server gets hacked. Also, I think you mean non-Western character sets, since almost everything is a Unicode character.

I think a good community reputation system is essential to trust Bitcoin. Crypto isn't terribly necessary in building such a system, especially if it's forum-based. Of course, then the forums become a single point of a failure and a huge vulnerability.

Jago, do you have all the proper dependencies installed?

The problem with Bitcoin Certify is that it seems a lot like what Verisign does, which is to issue certificates for a fee to anyone who submits correct contact information. The only way something like that might work is if there were already a reputation stat on the forums and you were pledging to contact all involved parties and verify the transactions. And people would have to know and trust your practices as well.

Basically, it would be a whole lot of work for a fairly measly fee.

But I don't want to pay 100BTC for a PGP-signed PNG with my username on it.

What if he's out of the States, though? I thought Paysafe was a UK thing.

Wobber, how much would you like for your drive?

Hello all,

I'm looking to buy a relatively new 4GiB flash drive to use as a root device for a Debian machine that I'm building. Anyone interested?

Yeah, 0.3.0 seems to have a performance increase over svn r84 with Laszlo's performance patch.

Very nice! There have definitely been some major improvements, including the hashmeter, RPC support, and a command-line daemon. It would be nice if the listtransactions RPC method were finished before the next release, though. Also, some code should be added to the JSON-RPC methods to list out dHashesPerSec.

Great work Satoshi!

Yeah, r89 works as expected for me. It doesn't seem to have any problems creating or removing icons, but I'm using XFCE, so that might be the difference.

In addition, there's a difference between standard behaviour for tray icons in Windows and Linux. In most Linux apps, clicking the tray button while the app's window is open will cause it to close again. I personally like that better than the Windows default, where a single click does nothing once the main window is displayed. Is there any chance that feature might get added to the Linux version?

Satoshi, that definitely made it look better for me, but now Bitcoin won't minimize to tray in Ubuntu 10.04. The proper check box is checked, but it just minimizes to my taskbar like always.

The US doesn't have the world's largest army. That distinction goes to China, I believe. They also have the largest air force and I think the largest navy.

El Estados Unidos de América has the most effective army, though. And we have a lot of very big bombs.

You should try one of Laszlo's builds of Bitcoin. They show the number of hashes / second that you're computing. Post that number, and you/we can calculate how often you will find a block (on average, statistical caveats apply as Laszlo suggested).

That's not exactly true. At the moment, Bitcoin only binds RPC to the loopback interface, 127.0.0.1. I would assume that this web interface would be the same. However, there SHOULD be a password to prevent trojans from trivially sending your wallet away. Wallet encryption needs to happen, too.

I think that hiding the complexity of Bitcoin addresses from the casual user is a good thing. Barring that, it should be possible to embed an observable but unalterable message with address transactions. Is there some reason this is technically infeasible?

I think you're misunderstanding the issue. My browser will always be able to go to 127.0.0.1 (barring some strange IE settings or a virus). If I type the address into the URL bar or click a link, it will work fine. However, it isn't possible to use Javascript to complete POST requests between domains (or ports on the same domain).

Try clicking this link:
http://127.0.0.1/
You probably don't see anything (unless you're running a web server on your system), but the browser happily tries to take you there.

XMLHTTPRequest is what we were discussing in that other thread.