Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Already on GitHub? Sign in to your account

Is mSIGNA a full validation wallet? #645

Closed
gurnec opened this Issue Nov 13, 2014 · 21 comments

Comments

Projects
None yet
3 participants
Contributor

gurnec commented Nov 13, 2014

I believe that mSIGNA is an SPV client, although it's listed (on the choose-your-wallet page) as a full validation wallet. Maybe I'm wrong, though....

Perhaps @CodeShark could clarify?

Contributor

saivann commented Nov 13, 2014

Like Armory, both require a local full node to work, and inherit the security of a full node. mSIGNA also lets the user configure the wallet to use a remote full node, but that isn't the default and no list is pre-populated. As far as I'm concerned, I would also be fine with giving both Armory and mSIGNA a specific score text instead of the current one to better document this.

Contributor

gurnec commented Nov 20, 2014

(Apologies for the delayed response.)

It seems counterintuitive that adding a small feature to mSIGNA (the ability to connect to multiple random nodes as opposed to a single statically-defined node) would actually decrease its score.

However I'm convinced that your line of reasoning is correct.

If there were any interest in a slightly more accurate score text, I'd suggest replacing the first sentence which currently reads:

This wallet is a full node that validates and relays transactions on the Bitcoin network.

With this first sentence, which would result in (the other sentences are unchanged):

This wallet requires you to install full node software alongside it that validates and relays transactions on the Bitcoin network. This means no trust in a third party is required when verifying payments. Full nodes provide the highest level of security and are essential to protecting the network. However, they require more space (over 20GB), bandwidth, and a longer initial synchronization time.

It's unfortunately a bit on the long side... I don't have any strong feelings here, so I've no problem if anyone thinks it's simply not worthwhile.

Contributor

saivann commented Nov 20, 2014

@gurnec Sounds good to me, agreed on keeping a good score for Armory and mSIGNA here.

How about you drop "alongside it" in your suggested text? (It fits better in the bubble this way).

This wallet requires you to install full node software that validates and relays
transactions on the Bitcoin network. This means no trust in a third party is required
when verifying payments. Full nodes provide the highest level of security and are
essential to protecting the network. However, they require more space (over 20GB),
bandwidth, and a longer initial synchronization time.
Contributor

gurnec commented Nov 20, 2014

@saivann I was debating how to emphasize that additional software is required, e.g. "to install additional full node software" or "requires you to also install full node software", or the first version I mentioned.

I have no strong opinions on any of them -- if you think the shorter version is best, that's fine with me.

(FWIW, all of these versions overflow to 7 lines in my Chrome/Win7, and only the shortest (your) version squeezes into 6 lines in IE11.)

Contributor

saivann commented Nov 20, 2014

@gurnec Isn't it mentioned in your text? "requires you to install full node software". My small suggestion (removing "alongside it") is just meant to prevent the design issue of having one word on an empty line, which isn't so good looking / readable (see below)

capture du 2014-11-20 14 27 35

Contributor

gurnec commented Nov 20, 2014

Hmm... I'm confused now, sorry.

Were you thinking of changing checkGoodValidationFullNode? I was thinking of adding checkGoodValidationFullNodeRequired for just Armory and mSIGNA with different wording, which is why I was also trying to emphasize that those wallets require additional software to be installed along with Armory or mSIGNA.

Regardless, I'm perfectly happy with the shorter text since it seems less likely to wrap (although in my Chrome/Win7 it still does wrap):

image

Contributor

saivann commented Nov 20, 2014

I was thinking of adding checkGoodValidationFullNodeRequired for just Armory and mSIGNA with different wording.

Yes, that's also what I was understanding, that sounds good to me.

Contributor

gurnec commented Nov 20, 2014

I was thinking of adding checkGoodValidationFullNodeRequired for just Armory and mSIGNA with different wording.

Yes, that's also what I was understanding, that sounds good to me.

OK, thanks for clarifying. (You're screenshot threw me a bit, it looked like Bitcoin Core in the background).

Contributor

saivann commented Nov 20, 2014

@gurnec Ah sorry yes, I was editing the text live to create a screenshot :) I tested the short version on Chrome Linux, IE9 Win 7. Anyway, that's just a suggestion :) . Your screenshot with Chrome Win 7 doesn't look too bad to me compared to the one with just "time" on the last line.

Contributor

gurnec commented Nov 20, 2014

@saivann Ah, I understand now, thanks (and I agree that they both look fine).

Contributor

saivann commented Nov 21, 2014

@gurnec Did you plan to submit a pull request? Otherwise I can push a direct commit in the next days.

Contributor

gurnec commented Nov 21, 2014

@saivann For this small a change it didn't matter to me one way or the other; I have a commit ready to go so I'll go ahead and submit it.

Contributor

CodeShark commented Nov 21, 2014

mSIGNA is no less a full validation node than Armory - but mSIGNA also allows you to share a single full node installation across multiple devices to get full validation…and also supports SPV from an untrusted node. So mSIGNA doesn’t really require you to install full node software alongside itself…it allows you to, and assumes this as the default option.

-Eric Lombrozo

On Nov 20, 2014, at 4:41 PM, Christopher Gurnee notifications@github.com wrote:

@saivann https://github.com/saivann For this small a change it didn't matter to me one way or the other; I have a commit ready to go so I'll go ahead and submit it.


Reply to this email directly or view it on GitHub bitcoin#645 (comment).

Contributor

saivann commented Nov 21, 2014

@CodeShark Does the current version sound good enough to you? Since mSIGNA requires the user to take extra steps either ways (intalling a full node or configuring a remote node), I felt this text was OK, taking into account that we wish it to be used both on both Armory and mSIGNA while remaining short as much as possible.

Contributor

saivann commented Nov 22, 2014

Given the absence of response from @CodeShark, I will merge the pull request.

@CodeShark if you want to submit a change, please open a pull request in the following days otherwise current text will start being translated. Thanks!

@saivann saivann closed this in 9cb90ab Nov 22, 2014

saivann added a commit that referenced this issue Nov 22, 2014

Merge pull request #654 from gurnec/full-node-reqd
Modify validation text for wallets requiring additional full node software (closes #645)

gurnec added a commit to gurnec/bitcoin.org that referenced this issue Nov 22, 2014

Contributor

gurnec commented Nov 22, 2014

@saivann My sincere apologies, however I just noticed that the commit in my repo which you just merged was the earlier version (with the extraneous text "alongside it") which we agreed was too long. I just pushed the correct text to my repo. Please take whichever action is easiest to correct my faux pas....

Contributor

saivann commented Nov 23, 2014

@gurnec No problem, I just pushed your last commit. I wasn't sure if you prefered the other way and thought I had commented enough on this :) . Thanks for the update!

Contributor

CodeShark commented Nov 23, 2014

@saivann Technically speaking, you could use someone else's node - mSIGNA still does SPV. It would be vulnerable to SPV attacks (and particularly block withholding attacks) - and there are ways to mitigate these risks...but this is not the default use mode.

Contributor

saivann commented Nov 23, 2014

@CodeShark Yes, I think I understand that. I was just wondering if you had any issues with current text, which doesn't explain all details or optional features. And if so, do you have a better sugestion that would be reasonably short and would work for Armory too?

Contributor

CodeShark commented Nov 23, 2014

@saivann If it is to work with both mSIGNA and Armory, it's a bit tricky to find a wording that's both accurate and short since mSIGNA and Armory work so differently when it comes to network synchronization. Having said that, it doesn't really matter all that much - the typical user of mSIGNA is expected to have at least a passing familiarity with Bitcoin Core.

Contributor

saivann commented Nov 23, 2014

@CodeShark By default both use and inherit the security of a full node, right? This is the only outlined information here (previously it said that your wallet was a full node, I think current text at the very least is more accurate, maybe it can be improved a bit, or maybe we can consider it good enough).

@gurnec gurnec referenced this issue Jun 7, 2015

Closed

mSIGNA full node? #878

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment