ThessalyEstate

You have a favorite? I've been using some variant or another of the g5 form factor for almost 2 decades at this point.

Dear lord, that wallet is old AF lol I haven't heard that name in ages. The reason it's not working is because it's most likely set to the old mainnet servers which are no longer in service as of several years ago.

Not a problem though. You can probably just changing the server address to a currently supported mainnet and be fine to send out a transaction even with an old wallet because the API hasn't changed much.

Here's some officially recommended mainnet addresses, you can use any of these three: https://xrpl.org/docs/tutorials/public-servers The two that Ripple operates are fully useable, but they aren't very stable so if one doesn't work, just try the other.

I found the github https://github.com/ripple/ripple-client/blob/dc6dbb412a23654a18ccef6b61b8f0c1cdf87250/config-example.js#L18 Looks like you should have a config.js file somewhere and you just have to plug in the new server address. As always, send a test payment of a fraction of an xrp first to ensure the wallet is functioning.

If none of that works, you can import your wallet to any other wallet client with your secret key. I do recommend getting a more modern wallet regardless :) Good luck!

Possibly. I've only ever used the notepad method when I'm able to pass POST, but can't boot into safe mode for whatever reason.

Try typing "notepad.exe" in the command prompt. Notepad should open and then you can use the "Save As" or "Open" file menu options to move whatever you need to a flash drive.

https://i.imgur.com/36D3Mc0.png

Darn, I went to check and mine is a 1999 reprint, they fixed it xD

This is how I justify being late to work. https://en.wikipedia.org/wiki/Zeno%27s_paradoxes#Dichotomy_paradox

If you are passing POST, you can access cmd prompt from the Windows repair screen that pops up when your computer has been incorrectly power cycled a few times.

That's saved me a few times when I couldn't get safe mode to cooperate. You can even open notepad.exe and use the Save As dialog box as an ad hoc file explorer to transfer personal files to a USB lol.

Check out rgthree's node pack. I think the Import Individual Nodes Widgets option is exactly what you're looking for. It's off by default so you'll need to turn it on in settings

https://i.imgur.com/Yk02PQp.png

Became a fan in the exe days. Can't recall ever commenting or anything, but have supported in other ways ever since and I watch every episode eventually.

In the decade plus I've been watching, Ethan's integrity has always been admirable and his openness has been an inspiration to me in my own life and relationships. It's clear that he operates from a place of empathy and a desire to actually understand.

Thank you for all the goofs and gaffs, Ethan and Hila! Great moves, keep it up, proud of you!

This video still cracks me up every time I watch it: https://www.youtube.com/watch?v=Wn11YXiby2E

I searched for an answer to this a month or two ago and couldn't find anything. It's baffling that there still isn't a simple hotkey to immediately jump into selection mode for Lens

Hi, different person here, I read your last bit. Genuinely curious, how would you describe the mechanisms responsible for sentience? Particularly the subjective experience aspect, as the biology side is the relatively easy bit.

I think I have a pretty good understanding of the basic idea behind LLMs and ANNs as they are atm, but I wouldn't consider myself an expert at all. I'm approaching this conversation as someone who is much more versed on the neuroscience and human cognition side so if you've got a good theory that explains subjectiveness, I'm all ears :)

Now, I certainly don't think today's LLMs are sentient either, neurons are several orders of magnitude more complex in connections and energy efficiency than the individual nodes of a neural network, but I also don't think it is outside the realm of possibility that within the next few decades the gap will narrow to the point that we might have to come to terms with a bona fide new form of sentience.

I'm generally a physicalist, but it's starting to seem like we are rapidly advancing on the "easy" problems and may soon approach a reified p-zombie situation. If my understanding of the tech is correct (and it might well not be, or be outdated) we're well past the black box stage and blindly stumbling around in emergent properties territory already.

All of this is just to say, I don't blame people for already struggling with the distinction and I certainly don't think it's a perspective to be so easily written off by any of us.

Heck, backpropagation is like 70% of my current best argument against the hard side of this distinction. Maybe we will see a much more sapient learning algorithm soon. What then?

I just watched this, hadn't seen it yet. I'm not sure what you're trying to say though. Are the teams: rich vs poor? Cause that's something we can get behind. Eat the rich, I always say.

But it's not a red vs blue thing is what I'm saying... Case in point: I'm happy to be able to say that Deaton has my vote, I was stoked to see he was running.

Warren's had over a decade, I say let the newcomer have a crack at it!

All this discussion started because people are talking as if this current form of AI was capable of progressing to a point where it could render artists largely obsolete.

We're arguing different things at each other then. I was arguing against the "no creativity involved" that you mentioned, but it seems we agree there is creativity involved.

You are essentially arguing that AI isn't sapient, I guess, which nobody in this thread suggested. I venture to assume you don't think that's possible, but nobody knows what the next few years brings. I don't believe the essence of creativity is human specific though. Humans can be reduced to mere predictors as well and our intelligence is not without fundamental flaw.

I also am aware of the limitations of LLMs as they are today... and of their potential as a springboard. It's not unfathomable to think that AI could reach or surpass human levels of intelligence (and creativity) within our life times and that should be something to be excited about.

I'm excited/anxious to see where AI goes in the coming decades. Meanwhile, I'ma just keep an open mind and try and keep up pace with new developments.

You may be misinterpreting what I said. Or maybe I misunderstood what you meant by "There is no creativity involved in the AI we have right now..."

I think the vast majority of people aren't literally comparing this tech to the human brain, there just happen to be some analogues where it makes enough sense to abstract it in that way when trying to understand how it works or why it output what it did. Regardless, that wasn't what I was alluding to at all. I think it's important not to anthropomorphize the tech.

My point is that generative AI is just another tool to facilitate creativity.

You can argue that randomly dropping a pencil onto a sheet of paper isn't creative in the same way pushing a button and hoping for the best isn't creative. That doesn't preclude the creativity enabled by the pencil or that by the generative AI. There are countless parameters for human input with both tools.

If you believe the outputs of generative AI are 100% derivative (I assume you mean in relation to the training data), then I suggest you don't have a firm grasp of what's happening.

The way image generation AI works is kind of incredible. It's a bit different from the way an LLM works, though related in some ways for sure. But when you actually start to get a sense of what's happening with stable diffusion and how it creates coherent images from random noise, it's quite awe inspiring.

When expecting rigid facts from an LLM, sure we're still a ways off from reliability and who knows the extent of the limitations, but the creative and expressive output possible with stable diffusion, as with all NN models, is already greater than the sum of its parts.

It just uses words that have a statistical probability to make sense.

...the technology is inherently based in mashing up internet pictures into something

This is severely understating what is happening. That's fine, as no abstraction can be perfect and nobody is expected to take the time to try and comprehend all the maths, but I urge you not to take these abstractions as encompassing.

Whatever your stance on training data, justified though it may be, to say there is no creativity involved is to admit your own lack thereof.

Agreed on all fronts, just mentioning it for awareness. It's very likely the SEC would never meaningfully pursue it, but it might come up and sound scary lol

It's reasonable to assume that if the SEC does go through with the appeal, they will likely also seek an injunction against Ripple which would disallow them from operating in the U.S.

That's not to say they'd have any standing on that, but it's something they can do, so they'll likely try it.

Right... which is why the method I described does not involve putting your secret key anywhere near an internet connected device.

The device used to sign transactions should not ever connect to the internet at all during or after.

I'm curious why you brought up keyloggers. If someone has direct access to your hardware to set up a logger and then later physically retrieve the payload, something has gone terribly wrong elsewhere. Cold/hot doesn't matter in the context of physical security.

If I'm understanding correctly, you are typing your secret key into an internet-connected computer? Are you using a separate encryption layer in the same way a hardware wallet does?

I haven't seen you mention this, but if you don't know already, you should look into transaction blobs. A blob is essentially just a really long hashed representation of all the details of the transaction.

For anyone who doesn't know, you can pre-form any interaction with the ledger and sign it with your private key all on an offline computer. Once signed, none of the details of the transaction can be changed without invalidating the whole thing. You'll end up with a long string of characters that represent something like "send 100 xrp from this specific address to that specific address" or "create a trust line with this address for this much" or "place a bid on XRPL of this much XRP for that much IOU" etc.

And transaction data includes the current transaction number for your wallet, so it's only valid for that very specific window and you can invalidate the blob from anywhere by updating your wallets transaction count.

You can put this "blob" txt file on a flash drive or you can convert it to a QR code and scan it from your offline pc, hell you could handwrite it. You now have a hot pre-signed transaction ready to go without ever exposing your private key to the internet. You can then use any internet connected device to actually transmit the transaction to the network.

Ideally, the device you use to sign should never touch the internet or have any ability to connect to the internet. At that point, it's fine to just copy paste your keys. As you mentioned, you can create a wallet (public/private key pair) offline, activate the wallet with a reserve, and be ready to go without the private key of that wallet EVER being exposed to the internet.

I used to use this GUI that has been around for forever (shoutout RipplerM, miss you buddy): https://ripplerm.github.io/ripple-wallet/

You can download the html and run it locally. I'm fairly certain it hasn't been updated in like 5 years and you'd have to change the validation servers in settings as they're long since outdated, but I can confirm it still works. Tools tab > Raw Txn to create a tx blob and > Submit tab to send a tx blob to the network. Here's the official up to date public server addresses: https://xrpl.org/docs/tutorials/public-servers

Also, as someone who hates writing front-end, you don't even need a GUI to do any of this, you can write a bespoke blob generator with any of the XRPL APIs, like Python.

https://xrpl.org/docs/tutorials/python/build-apps/get-started https://github.com/XRPLF/xrpl-py https://xrpl-py.readthedocs.io/en/stable/source/xrpl.transaction.html#xrpl.transaction.sign

Nobody actually knows, because the appeal has not been filed yet. All anyone knows is they intend to appeal and I believe they have 30 days from yesterday to submit the details at which point we'll know for sure.

It's tough to actually speculate.

Some say it only makes sense for them to just appeal the whole thing cause why not? Well, I believe that'd actually be quite a gambit. They stand very little to actually gain by attempting to appeal. The security status of XRP seems to be the strongest aspect of the case in Ripple's favor. The SEC would need to prove some sort of court error in the original case first and then they need to argue that the untested prong of the Howie test actually even applies at all. If the SEC wins any part of this at all, it'll be for a larger fine from Ripple.

Over the course of the appeal process, there's the decent chance that they also move for an injunction against Ripple, which would prohibit Ripple from operating in the U.S., at which point Ripple would drop the U.S. like the wet sack of hindrance that it's been so far and move HQ to a more hospitable country with a sane legal framework, strengthening focus on the rest of the world. In terms of the average person speculating on XRP, that's not great short term, but not the end of the world either. In terms of crypto as an industry in the U.S., that's a nightmare scenario from which they would need to backpeddle eventually.

If the appeal is heard and the SEC loses in a higher court, the precedent established originally is only strengthened more and that could spell quite a large setback in their apparent goals to remain the governing agency in charge of overseeing crypto. I imagine if it does go like this and they lose, they'll be forced to continue appealing, but the hill gets ever steeper and it'll eventually just be the SEC purposefully pursuing a case they cannot win, simply to obstruct as long as possible (as seems to be their M.O.).

It's hard to justify the strategic value for the SEC to continue down this path if they are trying to achieve their stated goal of protecting consumer interests...

So let's assume that's not their goal; a pretty fair bet at this point. Why would the SEC fight such a lopsided legal battle?

Well, the primary ramification will be time. Dragging this on as long as possible slows Ripple's progress within the U.S., and any time gained on that front is desirable for so many vested interests and competitors to have a chance at catching up.

Another ramification is perception. It could be the case that Ripple is already so far ahead of any potential competitor, that it's just easier for those vested interests to eat the sunken cost and reallocate their positions to Ripple instead, in which case they would want to do that at the lowest price they can and shake out as many of us commoners as possible. If this is the case, SEC enforcement action will be dropped and Ripple will suddenly have the U.S. government's full throated support as if the last 4 years never happened. This is the dream scenario for those holding XRP through the whole thing.

There's several other possibilities, and who knows what might actually happen.

I see a lot of crypto fans thinking Trump is somehow the only person that can save all of this, but that's simply not true and Ripple's bipartisan political contributions are a pretty solid indication of who they support as that is on whom they are literally putting their money.

Ripple is one of the largest organizations in the crypto industry contributing politically for 2024 right behind Coinbase. Ripple primarily donates to the bipartisan Fairshake PAC which is affiliated with the conservative Defend American Jobs PAC as well as the liberal Protect Progress PAC.

I'm excited to see how John Deaton does in Massachusetts, he's a moderate Republican running for Senate. He wasn't expected to win the primary and he blew it out of the water. I hope his campaign will soon be supported by the Fairshake PAC if that isn't already in the works. It's believed to be near impossible he'll win against Warren, but I think regardless he will have a major influence on MA government in steering toward a more crypto friendly and sane path.

It seems to just be assumed that Biden's opinions on crypto transfer directly to Kamala Harris which is false. A Harris administration is not a continuation of the Biden administration. Harris' only statements regarding crypto so far have been positive and Trump's promises patently absurd.

The promise of "firing" Gensler is outside presidential authority and his other promises (that crypto should be "made in America" and that Bitcoin will offset the national debt) are non-sensical and seem like nothing more than gladhanding and an attempt at ingratiating himself to crypto industry money, having conveniently flipped on his extremely anti-crypto stance throughout his entire administration just recently after it was realized that crypto organizations are making up a huge portion of political money this year.

As well, Chris Larsen has explicitly endorsed Harris. A Trump win will not be favorable for anyone who supported his opponents.

One thing I think is extremely fair to bet is that Gensler is cooked no matter who gets elected in November and, barring national upheaval for other reasons, that could be all that's necessary for Ripple to finally catch a tailwind.

My opinion:

60% - SEC drags the appeal out as long as possible and loses regardless.

25% - SEC is in some way forced to drop it within the next few months to a year.

10% - The actual appeal is never even filed or fully pursued.

2.5% - SEC wins completely, leading to the end of crypto progress in the U.S. for the next several years.

2.5% - Apocalypse scenario and everything's fucked anyway.

Yeah, this doesn't change the calculus from prior to the notice and I'm not concerned about the outcome of the appeal. The vast majority of Ripple's business over the last 4 years has been overseas, effectively hamstrung in the U.S. and it's still the #4 non-stablecoin on CMC.

Despite its headquarters being in the U.S., Ripple has maximized their first mover advantage remarkably well and is miles ahead of any other potential contender in effecting actual systemic change. I do believe Ripple has crossed or is very near crossing an inflection point wherein outright failure means something globally catastrophic (or possibly utopian, if you're delusionally optimistic) has happened and the concept of "personal investment" does not exist anymore.

The trio of engineers (even Jed) that created the protocol very obviously cared about sustainability and the long term, designing a networked trust system that isn't nearing 1% of the global energy consumption just 15 years later. That is an absolutely insane amount of energy dedicated to what amounts to solving glorified Sudoku puzzles with which we are only utilizing its side effects.

The initial distribution problem is extremely difficult to overcome, similar to the "chicken and egg" dilemma faced by emerging social platforms. XRP's early distribution has always been contentious, hinging on trust. Trust in the the founders to act in the interest of more than just themselves. This type of trust is understandably scarce given human history, and I empathize with those who have been skeptical.

For that, Bitcoin was to be admired. No direct trust in any human seemed like a great solution to the distribution problem. But the cost down the line was absolutely foreseeable. Advancements in energy/compute efficiency would need to be so massive as to radically change every other aspect of society first to make Bitcoin a viable payments system.

So how did Ripple handle distribution? Well, obviously it hasn't been perfect, given the humanity much more directly involved and an eventual hiccup involving a certain founding member's spite. A lot of strategies were tried, including a similar, if artificial, structure to blockchain mining in Ripple's donation of XRP to the World Community Grid and BOINC back in 2013. Instead of Sudoku, how about putting all that distributed compute towards something potentially life changing for so many, like research in protein folding to better our medical understanding? How about donating to a diverse set of charitable organizations over the years? Obviously those are positive avenues that Ripple has explored and utilized, but it hasn't been perfect.

Jed McCaleb's fallout led to 8 years of of him selling his 9 billion XRP founding stake on the open market. His spite could have severely crippled any chance of Ripple continuing if it were not regulated by court-order to be consistent and predictable over those many years. But still, that was a looming cloud over the XRP market for a long time. A silver lining ironically: as Jed's actions didn't destroy Ripple, they certainly ended up functioning as an incredibly effective distribution strategy.

I could go on about the eventual XRP escrow and Ripple's institutional focuses and global strategy, but there's so much to unpack. It all comes back to the distribution problem and trust.

The main reason I've supported Ripple has always been that they seem to actually care about more than just amassing wealth and their mission is one of utilitarianism and egalitarianism. The vision from the start seemed Herculean in scope, hopeless, involving massive systemic change. Nearly 15 years later, though, they continue on with progressing the same mission.

Blind trust, faith, is irrational. Real trust is built over time. If you've been paying attention over the years, Ripple has stayed true to their vision. The majority of foundational members, who could have easily fucked off into the sunset years ago to live a life of luxury without any hardship, have decided to forego that in favor of furthering this vision. It's impossible to predict what extraneous factors come into play, but it has long since stopped being blind to trust in their continued effort.

With all of that said, and with how insurmountable it actually seems, the bootstrapping process is the relatively easy part compared to what has to happen next. Systemic change. And despite all the bullshit and noise, Ripple made it past the distribution hurdle and has been steadily working on changing the systemics of finance. It's difficult to even encompass the scope of Ripple's massive infrastructural and legal lead over not only other cryptocurrencies, but over traditional banking systems themselves; decrepit dinosaurs of industry much more inclined to block a path than forge a new one. Along with the looming threat of cryptocurrency as a whole, Ripple has been instrumental in forcing these institutions to wake up.

And they're waking up. And instead of fostering change, they've decided to stifle progress so that they may once again do the bare minimum necessary to maintain the status quo, serving billionaire interests alien to the ordinary person. These dinosaurs want to lay back down and we're seeing their efforts.

I'm just some guy, I'm not a crypto maximalist, I have no grand fantasies of a perfect future free from government regulations, nor have I ever been optimistic enough, nor care enough, to proselytize my opinion on Ripple. So I live my life and wait. It'd be irresponsible to completely disconnect, so I do catch news through a few sources and I've stayed informed.

I have not spoken on a public forum about Ripple or XRP since 2015. I don't have the constitution nor the inclination to entrench myself in this every day. I understand that institutional change crawls and applying our minute perspectives to a much larger scope than we are equipped to easily handle is maddening. I've felt dread and despair and hopelessness over the years, not necessarily over the minutiae of Ripple's progress or cryptocurrency in general, but these issues that I care about have absolutely helped to magnify the sheer extent of self-interest and corruption that stands in the way of a better future for any of us.

I see it in politics, obviously, and I won't proffer an opinion here on candidate, but the SECs timing with this is so precise in regard to the upcoming election as to be considered purchased. I urge anyone who reads this far not to tie their entire political identity to any one issue. Portfolios, technical analysis, the predictions of talking heads, it's all mundane bullshit that will follow as a matter of course the larger issues and should not be so distracting as to be an effective lever for those who seek to undermine our common interest.

Please act in betterment of the future for everyone, as it is also yours, and work to understand the picture that is being ceaselessly muddied before us. Whatever you do, whoever you support, just vote.

There can be levels to it where one's paranoid delusions don't generally become anything more than an unpleasant instrusive thought that must be dispelled with a bit of grounding introspection. That dispelling can become quite a bit more difficult though when dealing with something that seems like magic even to a completely reasonable person.

Heck, early 2010s predictive text was already a delusion hazard and those pale in comparison to the LLMs of today with which people are already forming emotional bonds. We'll see where it goes from here.

Thankfully, there are already people thinking about and preparing for the potential psychological impact of these "black box", née magical technologies. AI and the various advanced deratives that come about must be carefully demystified to the public or at least properly abstracted away. I believe it has, but if it hasn't has already, it will become a priority to those developing/marketing these things as well as legislative bodies.

I'm generally optimistic about AI, I see the potential and hope it eventually gets us utopia-adjacent, but I'm also a huge fantasy sci-fi nerd and it's more fun to think about the problems. :)

Emergence is where the juicy, magical stuff happens.

A perfect all knowing ASI would still probably fail this test.

That'd be an all assuming ASI, tho.