|
nimda
|
 |
April 19, 2014, 01:01:50 AM |
|
- where is my (linux) userstuff? I'd expect something like ~/.bitmessage, but can't find anything. The program folder is 6mb, which can't contain the "blockchain". It's in ~/.config, where it's supposed to be. - I believe it would help adoption *a lot* to have plugins for the common chatclients. Like kopete, for example. No matter if it's the whole bitmessage implementation, with huge size, as long as it visually fully integrates into the common clients. What do you people think?
Bitmessage provides a daemon mode, so it would just take someone with the time and motivation to write the plugins. Bitmessage isn't exactly instant though, I'm not sure how well it would work with clients focused around a primarily XMPP based ecosystem. |
|
|
|
|
|
|
|
|
|
TalkImg was created especially for hosting images on bitcointalk.org: try it next time you want to post an image
|
|
|
Advertised sites are not endorsed by the Bitcoin Forum. They may be unsafe, untrustworthy, or illegal in your jurisdiction.
|
Ente
Legendary
 Offline
Activity: 2126
Merit: 1001
|
|
April 19, 2014, 11:27:10 AM |
|
- where is my (linux) userstuff? I'd expect something like ~/.bitmessage, but can't find anything. The program folder is 6mb, which can't contain the "blockchain". It's in ~/.config, where it's supposed to be. - I believe it would help adoption *a lot* to have plugins for the common chatclients. Like kopete, for example. No matter if it's the whole bitmessage implementation, with huge size, as long as it visually fully integrates into the common clients. What do you people think?
Bitmessage provides a daemon mode, so it would just take someone with the time and motivation to write the plugins. Bitmessage isn't exactly instant though, I'm not sure how well it would work with clients focused around a primarily XMPP based ecosystem. Whoops, thank you for pointing out. No idea how I could miss this. Most programs have their config in ~/.program, was my impression.. It doesn't help that bitcoin-core now has a config in ~/.bitcoin and ~/.config/Bitcoin. At least doesn't help my overview! :-) Daemon mode sounds good! Sure, it wouldn't be exactly the way we are used to from instant messengers. But it would make it so much more convenient for BitMessage users, and it would make it less of a hassle for new users to start with BitMessage too. I want to start a bounty: "Program a Kopete plugin which integrates BitMessage into its GUI".Where should I do this, for maximum exposure? Any other messengers with high user counts and easy plugin-usability? Is there any other encrypted p2p messenger, fully anonymous (without knowing which IP sent a message)? Ente |
|
|
|
|
dewdeded
Legendary
Offline
Activity: 1232
Merit: 1011
Monero Evangelist
|
|
April 20, 2014, 08:37:33 AM
Last edit: April 20, 2014, 09:10:09 AM by dewdeded |
|
XMPP with PGP Encryption over TOR or I2P (other anon net access like VPN, ...)
(OTR encryption is also very common for XMPP)
there are some semi-dead encrypt. p2p messenger projects like:
- TorChat
- Crypto.cat
- RetroShare
which had/have some community/users in the past
Tor Project is developing a new Instant Messenger to be bundled with their Browser/TOR Client downloads (will be an hardened fork of Instantbird with OTR encryption, that only uses TOR obvious)
I guess this will be the "market leader" / anon messenger with the most users medium term.
|
|
|
|
|
Ente
Legendary
Offline
Activity: 2126
Merit: 1001
|
|
April 21, 2014, 07:39:47 PM |
|
XMPP with PGP Encryption over TOR or I2P (other anon net access like VPN, ...)
(OTR encryption is also very common for XMPP)
there are some semi-dead encrypt. p2p messenger projects like:
- TorChat
- Crypto.cat
- RetroShare
which had/have some community/users in the past
Tor Project is developing a new Instant Messenger to be bundled with their Browser/TOR Client downloads (will be an hardened fork of Instantbird with OTR encryption, that only uses TOR obvious)
I guess this will be the "market leader" / anon messenger with the most users medium term.
Thanks for the extensive list. Honestly, all of this sounds awful. i2p/TOR won't gain enough momentum for a messenger any time soon. OTR is no good solution for a messenger neither. Well, cool, I'll focus my mental energy on bitmessage then! :-) edit: Maybe it all doesn't matter, as 90% of all IM happens on mobiles anyway? So the top priorities would be to support Android and iOS (and desktop OS), and to have low hardware/storage/traffic needs. Which would mean BitMessage would be a total niche product.. Ente |
|
|
|
|
|
nimda
|
|
April 23, 2014, 12:08:19 AM |
|
So the top priorities would be to support Android and iOS (and desktop OS), and to have low hardware/storage/traffic needs. Which would mean BitMessage would be a total niche product..
Here's the problem I see: - Almost nobody runs an email server on their mobile device or even tablet. They connect to an email server. - Unfortunately, a third party server would have to either compromise anonymity, or serve every message to the phone (not feasible due to mobile network limits). - Running your own server requires a server (not free, in contrast to most people's email), or having a home computer reliably on for a good portion of time at least every two days--more if you want to have email-like speed. - Even if the home computer or server was being run, it's nontrivial to securely connect to it (required if the server isn't serving every message on the network to the client). |
|
|
|
|
|
Atheros (OP)
|
|
April 23, 2014, 06:07:10 AM |
|
Do people these days still keep a computer on in their house? Assuming yes..
IPv6 should make it easier for mobile and home devices to connect. Then devices could pair by requiring that the user take a picture of a qr code on the computer screen using the phone. That's all it takes to authenticate the computer and the phone to each other when they connect forever after. Perfect forward secrecy could even easily be used for this connection. We can use all this to guarantee security but unfortunately not anonymity because a local passive attacker would be able to see when the user gets a message: it would be immediately* relayed to the phone.
*One might argue that you could introduce a randomized delay. But it would still be discernible after a number of trials and the number of trials wouldn't need to be very large. The only solution I'm aware of is to send data of definite size to the phone every x minutes thus an attacker wouldn't know what is real data and what is just filler. But I don't like this solution.
|
BM-GteJMPqvHRUdUHHa1u7dtYnfDaH5ogeY Bitmessage.org - Decentralized, trustless, encrypted, authenticated messaging protocol and client. |
|
|
|
nimda
|
|
April 29, 2014, 03:50:10 PM |
|
Some tradeoffs have to happen. If we send a fixed amount of data on fixed intervals, we can completely preserve anonymity provided the message data stays under that bandwidth limit. Unfortunately, mobile data caps suck, and would make that limit rather small.
|
|
|
|
|
phelix
Legendary
Offline
Activity: 1708
Merit: 1019
|
|
June 17, 2014, 01:33:38 PM
Last edit: June 17, 2014, 02:55:22 PM by phelix |
|
Is it possible to sign text with a Bitmessage address that can be verified externally by a Python script?
I wonder how I could securely link from a Bitmessage address to a namecoin ID e.g. by signing the ID with a Bitmessage address.
edit: nevermind, found private keys in "keys.dat" and some info in the wiki, should be able to figure out the rest from the source.
|
|
|
|
|
|
R4v37
|
|
June 23, 2014, 11:56:51 AM |
|
is it the same as "troll box" or "chat box"  maybe the difference is in the encryption? wont it delay the message? |
|
|
|
|
|
nimda
|
|
June 24, 2014, 04:59:54 AM |
|
is it the same as "troll box" or "chat box" maybe the difference is in the encryption? wont it delay the message? Bitmessage is more of an email replacement than an IM one. |
|
|
|
|
|
|
|
RVienz
|
|
July 09, 2014, 10:27:40 AM |
|
what Bitmessage ?? Desentralize Chat
Good Invention will Thumb up for this project
|
|
|
|
bitXbay
Newbie
Offline
Activity: 38
Merit: 0
|
|
July 10, 2014, 11:17:32 AM |
|
Hi, I create decentralized marketplace with bitmessage and bitcoin. And it can be used like exchange, more about BitXBay here. |
|
|
|
|
|
|
Ente
Legendary
Offline
Activity: 2126
Merit: 1001
|
|
July 13, 2014, 07:34:31 PM |
|
Awesome!
I felt the next step for Bitmessage would be a broader userbase, via "popular chat client addon". I didn't think much about mobile integration, as it seemed too impossible to me.
Two thoughts on this:
- did you check out bitcoinj? That's what "Bitcoin Wallet" on Android is using, and it has an efficient "lite" modus.
- whatever the way to go, I would really prefer that all BM messages are all the same. As soon as some of them have a tag and some not, plausible deniability is gone, more information is added via that distinction, people need a different backchannel to say "hey, please tag messages sent to me", which again leaks info, and finally there will be many people not receiving messages because of this and dropping it all together in frustration.
Sorry for that long sentence.
Please make all BM messages the same :-)
You are doing wonderful work, this will be a big step forward!
Ente
|
|
|
|
|
jeezy
Legendary
Offline
Activity: 1237
Merit: 1010
|
|
July 13, 2014, 07:57:04 PM |
|
Agreed. This sounds like some awesome changes for Bitmessage. Looking forward to the new features and using it on my mobile!
|
|
|
|
|
Koshi
Newbie
Offline
Activity: 20
Merit: 0
|
|
July 14, 2014, 11:37:02 AM |
|
Awesome!
I felt the next step for Bitmessage would be a broader userbase, via "popular chat client addon". I didn't think much about mobile integration, as it seemed too impossible to me.
Two thoughts on this:
- did you check out bitcoinj? That's what "Bitcoin Wallet" on Android is using, and it has an efficient "lite" modus.
- whatever the way to go, I would really prefer that all BM messages are all the same. As soon as some of them have a tag and some not, plausible deniability is gone, more information is added via that distinction, people need a different backchannel to say "hey, please tag messages sent to me", which again leaks info, and finally there will be many people not receiving messages because of this and dropping it all together in frustration.
Sorry for that long sentence.
Please make all BM messages the same :-)
You are doing wonderful work, this will be a big step forward!
Ente
Thanks! We're very excited to be bringing Bitmessage to Android. To answer your questions: 1) There's a small amount of code from BitcoinJ in the Android client, so yes that's been helpful for us. All credit to Mike Hearn and the others for producing it. 2) I agree that having all Bitmessage messages be structurally identical is definitely desirable. If we can preserve that while still finding a way for lite clients to identify messages addressed to them, that would be brilliant. If you follow the Bitmessage forum thread there is a suggestion from Thomas that may allow us to do that, so hopefully it will be possible. The current message tagging proposal specifies that all messages will have a tag, but if the destination address is not a lite client then it is just 32 bytes of random data, so we're trying to keep as close as possible to messages being unidentifiable. Suggestions and ideas are very welcome. Agreed. This sounds like some awesome changes for Bitmessage. Looking forward to the new features and using it on my mobile!
Cheers  If people are prepared to get involved and contribute to Bitmessage then I think we can make some really good progress in the near future. |
|
|
|
|
HostFat
Staff
Legendary
Offline
Activity: 4214
Merit: 1203
I support freedom of choice
|
|
August 06, 2014, 10:00:56 PM |
|
Nuova versione di Bitmessage! 0.4.3
Support pyelliptic's updated HMAC algorithm. We'll remove support for the old method after an upgrade period.
Improved version check
Refactored decodeBase58 function
Ignore duplicate messages
Added bytes received/sent counts and rate on the network information tab
Fix unicode handling in 'View HTML code as formatted text'
Refactor handling of packet headers
Use pointMult function instead of arithmetic.privtopub since it is faster
Fixed issue where client wasn't waiting for a verack before continuing on with the conversation
Fixed CPU hogging by implementing tab-based refresh improvements
Added curses interface
Added support for IPv6
Added a 'trustedpeer' option to keys.dat
Limit maximum object size to 20 MB
Support email-like > quote characters and reply-below-quote
Added Japanese and Dutch language files; updated Norwegian and Russian languages files |
|
|
|
Rassah
Legendary
Offline
Activity: 1680
Merit: 1035
|
|
August 21, 2014, 04:35:40 AM |
|
Why can't light clients just request message headers that all messages will have, and if they don't decrypt using the address, just reject them? Similar to how SPV clients just request block headers, and discard blocks or addresses not relevant to them.
|
|
|
|
|
|
|
|
