COBALT

Cobalt (symbol Co, atomic number 27) has a standard atomic weight of approximately 58.933 (often rounded to 58.9

[COBALT @COBALT589]

DEVELOPER Supply Secured, please join our poll on TG to vote as a community for plans for the supply
https://t.co/jO7Me8tUGU

[COBALT @COBALT589]

https://t.co/kMq7vScEEm

Good morning the Cobalt Website is now live!

[COBALT @COBALT589]

Website finished, will be live shortly!

[COBALT @COBALT589]

Quietly waiting for our moment
$COBALT

[COBALT @COBALT589]

Follow the $COBALT road

[COBALT @COBALT589]

When $COBALT flips the switch don't say you weren't warned!

[COBALT @COBALT589]

Somes say $COBALT is a GEM worth holding onto

[COBALT @COBALT589]

This is only the beginning of $COBALT
Website should be loaded shortly!

[David 'JoelKatz' Schwartz @JoelKatz]

@witiwest @XRP_stuart @24_18_16_321 @Ripple I think you would need Cobalt, or something similar, to do that. Do you think there's a significant practical benefit to getting the total tx confirmation time down from around 5-7 seconds to around 2-3 seconds?

[David 'JoelKatz' Schwartz @JoelKatz]

@popculture_club @TarinHaig @galgitron @nbougalis The biggest benefit, IMO, is resiliency. The network can keep making safe, reliable forward progress unless both cobalt and the "inner" consensus algorithm fail simultaneously.

[David 'JoelKatz' Schwartz @JoelKatz]

@BitConsultants @ercwl @jratcliff @CharlieShrem Agreed. It causing harm at layer 1 is. People are working to make that even more remote. Look at Ethan's work on cobalt, for example.

[David 'JoelKatz' Schwartz @JoelKatz]

@TheRipplerIAm @EDadoun With Cobalt, you would need both the "inner" and "outer" consensus layers to fail simultaneously for forward progress to halt. If the "inner" mechanism fails, then it does just get very slow as the "outer" one repairs the problem.

[David 'JoelKatz' Schwartz @JoelKatz]

@XRPTrump @emacbrough @RossMacFirdeen @ipinky77 @RippleRiddler @digitalassetbuy Cobalt has a solid, designed in defense to that. So it doesn't need the two second delay and is provably invulnerable to that attack. 2/2

[David 'JoelKatz' Schwartz @JoelKatz]

@XrpCenter @WolfsburgMK1 @StellarOrg @emacbrough Yes. You can use Cobalt to tweak your validator topology in a safe automated way to keep the network's throughput high without risk of getting stuck unless both layers fail concurrently.

[David 'JoelKatz' Schwartz @JoelKatz]

@XRPTrump @h0ezlater @emacbrough @CryptoPennyCO @Ripple Yep. I'd been thinking about both layers often before cobalt was called "cobalt" and when I wrote that tweet (possibly elsewhere as well), I forgot which one was cobalt and which wasn't. Cobalt solves the harder problem of the governance layer and makes the other layer easy-ish.

[David 'JoelKatz' Schwartz @JoelKatz]

@h0ezlater @emacbrough @CryptoPennyCO @Ripple I mixed up the part cobalt does with the part cobalt doesn't do! I literally got cobalt entirely backwards!

[David 'JoelKatz' Schwartz @JoelKatz]

@XRPTrump @h0ezlater @emacbrough @CryptoPennyCO @Ripple By the way, cobalt will make it possible to compensate validators with all, or a portion, of fees should we wish to do that. My gut instinct is no, for three reasons. But others might feel differently.

[David 'JoelKatz' Schwartz @JoelKatz]

@MikeNardolillo @haydentiff @tcherno @EricVailNaylor2 @Ripple Sure. But it's difficult to do that because the network has no idea who the validators are. If you could agree who the validators are, you wouldn't need validators. Cobalt could change this.

[David 'JoelKatz' Schwartz @JoelKatz]

@velardir @haydentiff @MikeNardolillo @tcherno @EricVailNaylor2 @Ripple Cobalt has a two-layer approach where one layer agrees who the validators are for the other layer. This means the network does know who the "inner" validators are.

[David 'JoelKatz' Schwartz @JoelKatz]

@popculture_club @TarinHaig @galgitron @nbougalis A lot of Cobalt's benefits were provided other ways. I still think there are some benefits you couldn't get any other way. But there's a lot of risk with such a change and it turns out to be a bit more complicated to do than initially expected.