[ANN][DASH] Dash (dash.org) | First Self-Funding Self-Governing Crypto Currency

[bigrcanada]

dashtalk is up for me...

[semajjames]

Am I the only one who see the dashtalk offline?

can confirm  it's offline for me, too

and me

[Kienbui]

dashtalk is up for me...

It's still offline for me...

[dEBRUYNE]

http://downforeveryoneorjustme.com/dashtalk.org

Says online offline, but down for me as well.

EDIT: Switching back and forth now.

[Minotaur26]

Dashtalk is run independently by community members hopefully the operator will look into any issues he may be having.  I think we should consider creating a forum on the official website to have more alternatives and for the development team to guarantee it is always up, everyone goes there to get the latest wallets anyway so it might make sense to also have a forum.

[AlexGR]

I have a theoretical question, a dev might enlighten me...

Suppose someone has gone to the effort of finding 4-5 equidistant masternodes with approximately the same latency. I mean, let's say I'm in Germany and I select (through a custom script) for broadcast hosts that are around 100-150msec away, like the US, Asia, Africa, etc.

If that guy simultaneously broadcasts an IX through these 4-5 masternodes, in order to do something weird with the IX, or confuse and fragment the network, what happens?

edit: I've pondered a variation of this for bank ATMs... say 2 people have 2 ATM cards on the same bank account that has 1000$ inside. They have both synced their clocks and agreed to press "withdraw 1000$" on 00'00GMT. One is in US east coast, the other is in US west coast. So, in theory, if they do it *simultanously*, the withdrawal can go through and the 2 ATMs will give 2000$ for a 1000$ balance. Then the balance would be negative (-1000$). It could probably be prevented by the software waiting 1-2 seconds before giving the go ahead, to see if other ATMs have an incoming order at the same time. 1-2s should be enough to sync, no matter the distance.

[toknormal]

A very early (2.5 years ago) 'Replace by Fee' discussion.

A trader's perspective.

[A. k. solo]

Dashtalk is run independently by community members hopefully the operator will look into any issues he may be having.  I think we should consider creating a forum on the official website to have more alternatives and for the development team to guarantee it is always up, everyone goes there to get the latest wallets anyway so it might make sense to also have a forum.

Well it looks like they definitely have an issue,  it looks like their running behind Nginx,  and it appears port 443 is either not listening on the server or is being blocked at the firewall level,  failing that they could have null routed some ip's on the firewall and we're just the unlucky few they blocked,  but I think it's the former since a proxy server returns the same results and a simple curl or telnet on 443 returns nothing only port 80 appears to be listening on the server and is redirecting requests to 443 (https) which unfortunately does not seem to be listening. 

Oh and yeah their ssl certificate appears to have run out today,  so I am assuming they are changing it,  but that should only take a few minutes.. Then again maybe their not pointing to the correct server... There's  a whole lot of reasons,  maybe they changed registrars in which case dns would have changed and they forgot to lower the ttl time limit for the new one to kick in...  Hence why we maybe getting directed to the old... just hope they get on top of it.

Do they have an irc chat room?

[moocowmoo]

I have a theoretical question, a dev might enlighten me...

Suppose someone has gone to the effort of finding 4-5 equidistant masternodes with approximately the same latency. I mean, let's say I'm in Germany and I select (through a custom script) for broadcast hosts that are around 100-150msec away, like the US, Asia, Africa, etc.

If that guy simultaneously broadcasts an IX through these 4-5 masternodes, in order to do something weird with the IX, or confuse and fragment the network, what happens?

edit: I've pondered a variation of this for bank ATMs... say 2 people have 2 ATM cards on the same bank account that has 1000$ inside. They have both synced their clocks and agreed to press "withdraw 1000$" on 00'00GMT. One is in US east coast, the other is in US west coast. So, in theory, if they do it *simultanously*, the withdrawal can go through and the 2 ATMs will give 2000$ for a 1000$ balance. Then the balance would be negative (-1000$). It could probably be prevented by the software waiting 1-2 seconds before giving the go ahead, to see if other ATMs have an incoming order at the same time. 1-2s should be enough to sync, no matter the distance.

An instantx consensus lock is determined by top five masternodes of the current block's winners list (masternodeman.cpp CMasternodeMan::GetMasternodeRank).

No matter where the transactions enter the network, the one that reaches one of those five masternodes first will block any other instantx transactions using the same inputs from being confirmed.

Concerning generating a race condition: If the current ix quorum (five masternodes) were all handed different versions of the same transaction, none would attain the required five locks due to contention. Only one would make it into a block by the standard mempool/miner route depending on luck.

[stan.distortion]

EDIT: Moo beat me to it Its covered in vid 5 linked in the OP (one at about 4.40):
https://www.youtube.com/embed/uGh43BQrxK0

[pille]

The Daily Decrypt [Ep.57] - Dash mentioned

And they have submited a proposal: https://www.dashwhale.org/p/Sponsor-DailyDecrypt

[toknormal]

I'm reading this thread and a whole lot of lightbulbs are going off in my head.

That thread is fascinating because it's bringing the stark contrast in priorities between adopters and designers to light.

I've just realised that the 0-conf "sector" is a distinct market. In commercial terms there's a discontinuity between the "0-conf" currency market and the 1+conf market and the pro "RBF"ers are not getting this. This is something I never really thought about before because I generally saw things the way Peter Todd does - i.e. that a 1+ confirmation is just a bit further up the "wedge" of security but the product's the same. The adopters, however, do not seem to agree. They see it as distinct.

Having read that thread I just see things completely differently now. InstantX suddenly makes a whole lot of sense because it independently addresses the concerns of that distinct market while at the same time retaining the "security product" that's being sold (mining confirmations).

Look at this for example...



...well thats precisely because of the disconnect in defining the 0-conf market. The adopters think they have an instant payments service and the bitcoin devs think they don't because they don't see it as an independent market with its own requirements. (Or they do and don't want anything to do with it).

On the other hand, Dash has understood this market and given it proper support (with the complimetary protocol I described here).





I'm currently scratching my head and thinking how did I not appreciate this before. I didn't see the point of InstantX when it first came out. I saw them as the same market. That Reddit thread has made me realise that those adopters have decided for themselves which end of bitcoin's security model they like best and it isn't the one the devs have put any work into. It's the one that Dash has put the work into.

I must've been thinking like a bitcoin dev  

(P.S. In the World Cup of crypto, Dash just went past two defenders from the opposing team and is now staring into an open goal IMO - whether RBF gets universally adopted or not).

[someredguy]

The Daily Decrypt [Ep.57] - Dash mentioned

And they have submited a proposal: https://www.dashwhale.org/p/Sponsor-DailyDecrypt

This is the first time I have voted for my masternodes. I voted YES! Good luck, Amanda.

[toknormal]

This poster seems to "get it":

[TaoOfSaatoshi]

https://www.rebelmouse.com/dashisdigitalcash/tek-tok-rip-1523327979.html

https://twitter.com/taoofsatoshi/status/679809521913565185

[toknormal]

https://twitter.com/taoofsatoshi/status/679809521913565185

I've gone back and sanitised my post for syndication now that I've seen that. (I always forget I need to watch my language when you're around). Perhaps you can oblidge

[Solarminer]

I'm reading this thread and a whole lot of lightbulbs are going of in my head.

That thread is fascinating because it's bringing the stark contrast in priorities between adopters and designers to light.

I've just realised that the 0-conf "sector" is a distinct market. In commercial terms there's a discontinuity between the "0-conf" currency market and the 1+conf market and the pro "RBF"ers are not getting this. This is something I never really thought about before because I generally saw things the way Peter Todd does - i.e. that a 1+ confirmation is just a bit further up the "wedge" of security but the product's the same. The adopters, however, do not seem to agree. They see it as distinct.

Having read that thread I just see things completely differently now. InstantX suddenly makes a whole lot of sense because it independently addresses the concerns of that distinct market while at the same time retaining the "security product" that's being sold (mining confirmations).

This forces bitcoin out of the point of sale market.  It is just not possible to sell a coffee for a $0.25 transaction fee and require a 10+ minute wait until a confirmation.  Adding more layers to gain speed just makes that fee higher.  You need to pay the highest fee to get confirmation in 1 block.  Adding more users again adds to the fee, or even more users at the same time(lunchtime). 

Dash with InstantX is perfect for point of sale.  Fast, private, secure, free, done.  I see this as the future. 

See what you can do with a PI, 3.5" touchscreen, and a bunch of command line graphic design - yep a $69 payment terminal the size of a deck of cards that can store the blockchain and confirm InstantX transactions.

[TaoOfSaatoshi]

https://twitter.com/taoofsatoshi/status/679809521913565185

I've gone back and sanitised my post for syndication now that I've seen that. (I always forget I need to watch my language when you're around). Perhaps you can oblidge

Hey, don't worry, I like that it's raw, straight from your mind type of deal. I'll leave it as is, unless you're specifically asking me to change it...  

[toknormal]

Hey, don't worry, I like that it's raw, straight from your mind type of deal. I'll leave it as is, unless you're specifically asking me to change it...  

Ok. They're your readers so I'll let you be editor as you see fit.

The Dash thread can have the 'polite' version 

[TaoOfSaatoshi]

This forces bitcoin out of the point of sale market.  It is just not possible to sell a coffee for a $0.25 transaction fee and require a 10+ minute wait until a confirmation.  Adding more layers to gain speed just makes that fee higher.  You need to pay the highest fee to get confirmation in 1 block.  Adding more users again adds to the fee, or even more users at the same time(lunchtime). 

Dash with InstantX is perfect for point of sale.  Fast, private, secure, free, done.  I see this as the future. 

See what you can do with a PI, 3.5" touchscreen, and a bunch of command line graphic design - yep a $69 payment terminal the size of a deck of cards that can store the blockchain and confirm InstantX transactions.

Very nice! Looks slick. You guys are keeping me busy...

https://twitter.com/taoofsatoshi/status/679819159249764352