I'm really enjoying using Armory and have expressed my appreciation in a small donation (which I intend to repeat whenever I can).  Thank you for all the work you've put into Armory. 

I only find myself missing 2 features:

1.  The main thing I'm missing is the ability to filter the wallet view.  Say I'm holding a wallet for someone else and I don't want to see it.  It's possible to save it offline and mark it as someone else's watching only wallet, which removes it from the total balance, but I still can't filter the view so that this wallet does not appear.  Is this a feature you plan to implement?  If so, if you'd tell me your planned approach, I'd be happy to give it and try and submit a pull request for you to review.  If you're not interested in it, I'll probably keep a private fork with this feature since it's so important to me.

2.  I'm surprised you don't generate a qrcode for offline transactions (for both the online transaction data  and offline signed tx).  Some of us have qr code scanners which would make the whole offline signing process a lot quicker.  Any plans to implement this?

Thanks again for all your work on this -- I'm very impressed.  I've been a happy user of the Satoshi client for ~2 years but am switching to Armory for the offline transactions and other features.  I'm recommending the same to my friends and family.  Well done!

Thanks for the tips.  Those types of UI additions are exactly what I'll be working on tomorrow.  However, this being a web-based project, there are limits to what kind of I/O monitoring I'll be able to do.  In any case, if you're dealing with large sums of btc, I'd really recommend disabling bluetooth and wifi altogether on your offline computer you use for signing, and keeping them off.  I physically removed the wifi receiver from my laptop that I turned into an offline transaction signer.

I'm using JSON as the format of the paste generated by the online page.  As such, the JSON.parse() function provides significant protection against what you are describing.  JSON.parse() was created specifically as a way to prevent execution of arbitrary code embedded in JSON and is available on all modern browsers.  Plus, as you specified, I have a confirmation modal popup on the offline app displaying the values as they will be inserted into the raw tx.

Nonetheless, after creating an initial working prototype, I'm feeling more and more that the appropriate place to enter the destination address and transaction amount may be in the offline component.  Is convenience the main reason you want to enter the address and amount in the online component?

I've gone ahead and pushed my initial version as a way to kind of pre-claim the bounty, but I'm not finished yet.  In fact, please don't use the app yet for any transactions, even though I've already used it to send btc on the blockchain [1].  It's not been fully tested.

I plan on doing a lot of work on the UI tomorrow, doing some testing, and also producing a GPG-signed release of the MD5 hashes.  In fact, I may throw significant energy into this project, including hosting the online portion with SSL, hosting my own full Bitcoin node for querying the blockchain and broadcasting tx's, instead of farming it out to blockchain.info (which as done now requires a third-party intermediary since it's a cross-domain request), and perhaps some other fun stuff, all while keeping it extremely simple to use.  I'll probably put up some tasteful ads on the hosted online app to try to recoup the development and hosting costs, but the all the code needed to conduct these transactions will be open source so no one will have to use my hosted version if they don't want to.

I'm open sourcing the code under MIT, and will continue to open source the offline secure signing app and online component as described in xDan's spec above.  If I do code up some of the stuff I mention above, I may keep some of the server-side code closed.  We'll see.  But all the security-critical code will be open-source.

I've used significant code from both Brainwallet and BitcoinJS, which are Public Domain and MIT-licensed respectively.  So a big thanks to those guys, it's incredibly cool that all this stuff is being done with JS.  I've also used AngularJS and a few other standard liberally-licensed JS libraries.

Anyway, once xDan and crazy_rabbit are satisfied, the bounty can go to: 1KpN5iePG1czLnBvJLzSQWXz9cerBWuBo2  As a said, it's not ready yet, I still need to pretty-up the UI and build these apps into single-page HTML files, and issue GPG-signed hashes of the code.

GitHub repo: https://github.com/esbullington/bitcoin-secured
Transaction on blockchain created using this app: https://blockchain.info/tx/6549d360e1493865bebe5b90649de8f0d068e109655a2b618fa6d1f7648d8892  As you can see, the change goes back to the sending address, as requested.  Do note that you give up some anonymity in exchange for this convenience, since all your transactions will be tied to a single address.

One final note: if you're interested in offline transactions and don't mind maintaining a full node with bitcoind or bitcoin-qt, I'd strongly recommend taking a look at Bitcoin Armory.  Etotheipi has done a really great job.  That said, I think there's room for a real lightweight implementation of this that doesn't require the blockchain, thus my interest in xDan's proposal.

I've bought VPN services, dog tags (http://www.mydogtag.com/), the services of a talented web designer (http://www.blindmikey.com).  And I'd buy a lot using Bitcoin if more vendors offered that option.  And I wish to god I were paid in Bitcoin for all the overseas freelance jobs I do for which the banking fees are over $30 for a transaction of a few hundred dollars.

Out of curiosity, I started playing around with this to see how it would be done, and I've ended up finishing most of Part 1.  I'll see if I can finish up Part 1 and do Part 2 this weekend and release them both on Monday.  It's a good idea.

It would be absolutely awesome if blockchain.info could support CORS:

Access-Control-Allow-Origin: *

It would allow client-side projects to access the blockchain.info API without having to resort to hacks like using YQL.

What kind of Bitcoin mobile app would you like to see?  I've got some time to take up a new side project and would like to build a Bitcoin-related app.  Any app you'd like to see?

I thought that was the idea, too, but reading over the proposal, I'm not 100% sure:

"Somewhat more care must be taken regarding extended keys, as these correspond to an entire (sub)tree of keys. One weakness that may not be immediately obvious, is that knowledge of the extended public key + a private key descending from it is equivalent to knowing the extended private key (i.e., every private and public key). This means that extended public keys must be treated more carefully than regular public keys."

From: https://en.bitcoin.it/wiki/BIP_0032

Question: there's no relation between the seeds of different wallets, through, correct?  If someone gets the paper backup for one wallet, is the other wallet still safe?

OK, thanks.  I just read a post of yours from a few months ago mentioning it being planned for an upcoming version, so I was just curious.

Thanks again.  Just sent donation to your btc address in your signature block.

qtqr is a nice program for reading qr codes on Ubuntu: https://apps.ubuntu.com/cat/applications/qtqr/

However, I've had trouble with a lot of webcams reading qr codes, probably because of the resolution.  Your mileage may vary.

Perfect.  Thanks for that tip.  I see it now at the bottom of the page, but it might be helpful to link to that from around the deb file download link.

I'm very impressed with Armory -- finally tried it after 1+ years of hearing about it.  You've done a great job.

One question: does Armory 0.87 beta implement BIP 32 for its deterministic wallet?

Are these available anywhere for the latest releases?  I see the windows and osx md5 hashes, and the hashes for building from source, but nothing for the deb files.

Thanks!

First, and most important, the original Bitcoin client is MIT-licensed, which makes it very unlikely to be Stallman.  Next, the client is written in C+, which is not a favored language of old school hackers of the Richard Stallman school (C is).  On a related note, C++ is very much the language of academic and researcher types, which makes me think that Satoshi is probably a professor or some other academic or researcher.

If you've got good HTML5 experience and are good with JavaScript, I *highly* recommend you look at trigger.io.  Unlike many "non-native" frameworks like Titanium and even the Lua mobile frameworks and Kivy (which I actually like), trigger.io has no delay in starting up -- it's as fast as native apps on both Android and iOS.  It also hooks into a lot of native UI components, which results in an app that is indistinguishable from a native one that runs on both Android and iOS.  If you find you need some native functionality that doesn't exist, it's easy to build native plugins using Java and/or Objective C.

It's definitely worth a look; I'm totally impressed.  Combined with parse.com, and it makes building mobile apps very simple as long as you know JavaScript, CSS, and HTML5.

Only if it's a government-guaranteed student loan.  If it's a private student loan (and that's what it sounds like), it may be dis-chargeable like any other private loan (but check your terms first).

Agreed.  I think you'd have to give up access to your private keys for a knowledgeable lender to lend fiat against btc collateral.

But I like the OPs creative approach to traditional financial transactions.  Keep 'em coming.

Jr. Member?  You mean the number of posts he has on the forum?  Look, I don't know the gentleman in question at all and have no idea whether or not he is qualified for such a position, but to disqualify him on the basis of how many times he's posted to the forum is shortsighted at best.  I'm technically a "junior member" here (I think am still, I don't really pay attention), even though I joined in Summer 2011 after monitoring the boards since late 2010.  For the past 3+ years, I've mined Bitcions, spent Bitcions, and have bought and sold Bitcoins since then in no less than 5 different exchanges, not counting bitcoin-otc.  I've programmed multiple applications using the bitciond API, and have informed quite a few people about the project, some of whom went on to buy, sell and spend Bitcoins.  I paid for VPNs, VPSs, and developer and designer services using Bitcoin.  On a personal note, these forums, and most importantly the Bitcoin project overall, have had a huge impact on my political and philosophical outlook on life, for which I am grateful.

Nonetheless, I take care not to deluge the forums with my thoughts on Bitcoin, and try to limit them to when I have something to add to the conversation based on my personal experiences.

On the other hand, I've noticed quite a few people who have joined only months ago have spammed the forums with utter nonsense and "+1" posts who are accordingly no longer "jr members".  Frankly, I could care less, but the fact that some people use this as an actual metric for experience in the Bitcoin community is a major concern.  I actually think this perspective was one contributing factor to a lot of the fraud that has occurred in previous years.  People would see someone was a "hero member" because he had spammed the forums for the past month or two with useless posts, used that as a proxy for trust, and entrusted him with more money than was wise.

Folks, instead of the number of posts, if you're really interested in someone's experience with the community, look at the year when they joined, what type of projects they've been involved in, and most importantly, the content of their posts.  There's no substitute for quality!

Well-done, bravo!

Thanks for helping bring Bitcoin to the mainstream.

You didn't record your wallet generation seed mnemonic anywhere?  The 12-word passphrase you would have been shown at the beginning?  Did you record it and delete the file, or not record it?

You said the password may be up to 12 chars long and you remember specific parts of the password.  Do you know the order of those parts, like if they are beginning, end, etc.?