The biggest feature in this change is the addition of a 32-bit windows wallet. The 32bit version can also be used if your CPU doesn't support SSE4 instructions.

GUI wallets can also enable RPC control in the config file.

RPCs added:

'block_account' to list the account to which this block belongs
'history' to print an account's transaction history.

Am I the only one who hates the phrase “off-chain scaling”? I guess my reaction when I hear that phrase is, well yeah, of course Bitcoin can "scale" off-chain. If you want to scale to “VISA levels,” and you're willing to do it off-chain, it can be pretty simple. Take the existing VISA infrastructure (or recreate it) and, instead of using it to process fiat-denominated IOUs, use it to process Bitcoin-denominated IOUs. Now obviously the “Bitcoin as settlement network” camp will say: “hey, that’s not fair! We’re not talking about simply recreating the custodial, trust-based credit and banking models of the fiat world in Bitcoin. We're talking about the Lightning Network which uses crypto-magic to enable off-chain transactions while still avoiding centralization and custodial risk. Lightning transactions are Bitcoin transactions.”

Well, no. They’re not. I’m far from an expert on the Lightning Network, but as I understand it, the proposal involves the repeated exchange of unconfirmed, unbroadcast Bitcoin transactions, i.e. potential Bitcoin transactions. Now there are some people who are very excited about the LN’s potential, and there are others who are very skeptical. But for purposes of the point I’m making it doesn’t really matter who’s right. I have no doubt that Bitcoin, by its very nature as an open, extensible network, enables the creation of some really novel, really useful “layer two” solutions (whether or not the LN proves to be one of them). But the fact remains: when you move transactions to a “layer two” solution, you have – by definition – added a layer of risk. Your layer two solution might be really, really great and a huge improvement over the traditional banking model such that the added layer of risk is relatively thin. And that’s awesome… but it’s still there. So you can’t just say: “who cares if Bitcoin’s layer one is artificially crippled? We’ll just move everything to layer two.” To the extent that on-chain scaling is artificially restricted (rather than being constrained only by technological limits), there is going to be an unavoidable deadweight loss. And all that can do is open the door for Bitcoin’s competitors.

Also, it seems to me that if there were no downside to using a “layer two” solution to make a particular payment, if it were really true that “Lightning transactions have the full security of on chain transactions” (as I saw one redditor claim), that would be a very dangerous state of affairs. If everyone could get all of the benefits of an on-blockchain transaction without actually using the blockchain (and thus paying the fees to secure it), that would seem to create a tragedy of the commons.

Since there hasn't been a new release for a while, I am making a new one. Since there isn't much work left to do before the 0.6.0 milestone, this will likely be the last 0.5.x release. If you want to make sure that 0.6 works smoothly, please try out this release and report back issues that you're having. Also, I would appreciate if people helped with the translations, I set it up on an online service Transifex so that you can do that very easily from your browser:

https://www.transifex.com/bitmessage-project/pybitmessage/

This release includes bugfixes, security improvements and some refactoring (thanks to @mirrorwish). In particular due to the security improvements I recommend that 0.4.4 users update too, as some denonymisation attacks have been addressed.

This is the list of changes:

newly arrived messages were sometimes missing or were not displayed correctly
IPv6 thread names can be correctly parsed by some logging tools
private IP range checks improved
lock file is cleaned up on exit
blacklist/whitelist labels can now be edited
don't unnecessarily send ACKs (this could have been misused for deanonymisation attacks)
don't allow getdata spamming (mitigates deanonymisation attacks)
search has been improved, now it can search without having to press return, and applies to newly selected folders and newly arrived messages
email gateway status query and settings change can be done from a menu
email gateway registration rejection messages now popup a request for changing the email address
always try all decryption keys (mitigates timing attacks)
translation system changes, fixes and updates to some translations
nodes with port 0 are rejected
HTML parser has security and rendering fixes
links for http(s) and bitcoin URIs allow an external handler in text mode also (subject to manual confirmation)
email links are handled internally (by clicking an email link, it will start composing a message to that address)
close to tray is now available
simple keybindings for quick navigation (N)ext, (P)revious, (R)eply, (C)ompose, (F)ind.
copying unicode to clipboard fixed
"Sent" folder in chans now displays both messages you sent to the channel as well as messages you sent from that channel address
multiple unicode fixes
"All identities" now has a "sent" and "trash" folder too
improvements in Tor support (reduced number of parallel connections, and PyBitmessage can now connect to hidden service nodes, but still cannot fully host a hidden service itself)
UPnP fixes
some UI refactoring (thanks @mirrorwish)
build scripts for OSX and Windows updated (I have been using them just haven't published the updates until now)

Since there hasn't been a new release for a while, I am making a new one. Since there isn't much work left to do before the 0.6.0 milestone, this will likely be the last 0.5.x release. If you want to make sure that 0.6 works smoothly, please try out this release and report back issues that you're having. Also, I would appreciate if people helped with the translations, I set it up on an online service Transifex so that you can do that very easily from your browser:

https://www.transifex.com/bitmessage-project/pybitmessage/

This release includes bugfixes, security improvements and some refactoring (thanks to @mirrorwish). In particular due to the security improvements I recommend that 0.4.4 users update too, as some denonymisation attacks have been addressed.

This is the list of changes:

newly arrived messages were sometimes missing or were not displayed correctly
IPv6 thread names can be correctly parsed by some logging tools
private IP range checks improved
lock file is cleaned up on exit
blacklist/whitelist labels can now be edited
don't unnecessarily send ACKs (this could have been misused for deanonymisation attacks)
don't allow getdata spamming (mitigates deanonymisation attacks)
search has been improved, now it can search without having to press return, and applies to newly selected folders and newly arrived messages
email gateway status query and settings change can be done from a menu
email gateway registration rejection messages now popup a request for changing the email address
always try all decryption keys (mitigates timing attacks)
translation system changes, fixes and updates to some translations
nodes with port 0 are rejected
HTML parser has security and rendering fixes
links for http(s) and bitcoin URIs allow an external handler in text mode also (subject to manual confirmation)
email links are handled internally (by clicking an email link, it will start composing a message to that address)
close to tray is now available
simple keybindings for quick navigation (N)ext, (P)revious, (R)eply, (C)ompose, (F)ind.
copying unicode to clipboard fixed
"Sent" folder in chans now displays both messages you sent to the channel as well as messages you sent from that channel address
multiple unicode fixes
"All identities" now has a "sent" and "trash" folder too
improvements in Tor support (reduced number of parallel connections, and PyBitmessage can now connect to hidden service nodes, but still cannot fully host a hidden service itself)
UPnP fixes
some UI refactoring (thanks @mirrorwish)
build scripts for OSX and Windows updated (I have been using them just haven't published the updates until now)