narrative (as if Ethereum and other chains don't do research), dodgy deals with third world countries, and they hired McCann, a huge marketing agency. What does a blockchain need marketing for?
Outdated tech that calls itself third gen blockchain but uses UTXO, state channels, it has no priority fee gas model (people need to wait to get txs included at high demand), which is just.. Utterly baffling, nobody uses Haskell, formal verification is possible on solidity too, etc etc etc
It's just a giant narrative of garbage mixed with misleading marketing, stake pools shill the shit out of ADA cause they make like 70% APY when people delegate to them, shady youtubers pump it
It's this cycle's EOS/XRP
Low coin price means noobs pile in on it, Charles is literally a vlogger promising the blue of the sky in his daily videos, other programming languages and even VMs are possible on rollups (eg Rust on Starkware, LLVM on zksync)
Any unique selling point cardano claims to have is literally nonsense
On chain governance is a huge no no (you want retail noobs to decide the gas limit? You're out of your mind), dPoS means staking pools have immense HW requirements -> they all sit in a data center, not at home
EUTxO model is a huge roadblock for dapps, you literally need off chain tx ordering to be able to build a DEX like Uniswap, this would be posted back to the chain via a centralized relayer, defeating the entire purpose of a DEX
Hydra is outdated tech and still won't release for another two years the way its going, there's only one or two papers on it, no spec, nothing. State channels can't support composable complex, ownerless smart contracts, they have liveness assumptions, and rely on stake pools."
Very interesting is the statement that UTxO is outdated. In short, it is totally meaningless and is also belied by the choices made by several projects.
Some points are not totally off. The academic research narrative is a bit overused. Others do research, too. You can write solid tech in other languages. You can do verification on other languages. And the papers by IOHK do not prove flawlessness, cannot do it in principle. But condemning everything completely for that, is also very much non sequitur.
In fact, there is no mathematical formalization for Casper, if we want to take a concrete example. If anything, I read about a formalization of Tendermint done recently. If this is considered an overuse of research, then we need to go back to basics and ask why it is needed.
It's not totally wrong that it makes DEXes' job more complicated. Over at the Eternl channel, there regularly *are* people having problems with them, because they do not choose the right UTxOs etc. pp. I tend to think that their advantages outweigh that, though.
Not overuse of research, overuse of using it as a marketing narrative. Who really looks at the papers? Who really checks if what is in the papers is really implemented in the code? Documentation of Cardano's core components is enervingly scarce. …
Can you pls elaborate on the last part? As I have read all the core components’s docs. networ, ledger consensus etc. though once I had to directly asks from iOG from ch itself to get one (the network one which was in very early stage at that time) As I wanted to compare the new Shelley stack with the broken kademlia based one
To make a fair comparison, one should also consider the time it took to develop the first dex on EVM since the launch of Ethereum. Not to mention the early experiments with colored coins on Bitcoin and other blockchains.
Are you really sure what you are talking about? I have found very few projects with engineering specifications as clear as Cardano's. Not to mention that it is a bit contradictory to say that scientific research is used for marketing purposes if no one reads it. That would be a very bad choice for publicity.
Both have advantages disadvantages as almost everything, y
I am quite sure. I spent the last months trying to find out how exactly this all works. All too often, I have to wade through half-outdated documentation from three different sources to maybe find an answer.
And there is no contradiction. “We have SCIENTIFIC papers!!!” can be and is used as marketing for Cardano by a lot of people who haven't and probably couldn't read them.
Concrete example: I wanted to know, who exactly is allowed to spend from the treasury how. Couldn't be found explicitly anywhere.
Could you provide a specific example? Having read the documentation, I did not notice this outdated information (assuming I found the latest available documentation). Obviously, it is not meant to be read by a general audience.
Yeah that is true, and makes sense as it is constantly evolving, changing.
But this is not the fault of those who wrote the articles. At most the fanboys who use it uncritically.
You should ask @disasm they use a few federated keys atm until onchain governance i.e. couple of more years now
Okay, but if the documentation was really that good, I should be able to find the answer immediately, shouldn't I?
https://cardanocataly.st/en/faq/#funding https://docs.google.com/spreadsheets/d/1xny1W7HhHANLNeQcnbr8El5rakJ80VIM7fgDwJ-uqys/edit#gid=0
I feel your frustration, it is not perfect but much much more better than anything from competitors where they can say whatever thy want without any formal proofs to fool ppl. A lot of examples from the past with very tough consequences.
1.) How and where are transaction fees calculated exactly. Had this post in the forum. https://forum.cardano.org/t/variable-fees-on-cardano/95517?u=heptasean The theoretical maximum transaction fee should be 0.876 ADA, but Daedalus automatically used 1.19 ADA (although he was probably far from the maximum transaction size of 16 KiB). Where do I find that bug? Daedalus source? Or does it delegate to cardano-node and there is something off there? 2.) cardano-cli uses a socket to communicate with cardano-node. What protocol is spoken on that socket? How can I write my own client in a language more accessible (to me)?
Fee calculation is tricky as it is like chicken and egg situation, as it is based from the size of tax but a tax must include the fee. That is why Daedalus just overestimated but it was very long ago iirc.
That was not the question. The treasury is somewhere on the chain. What happens on the chain? Who executes on behalf of “the community” what has been (kind of intransparently) decided by Catalyst? I was told that there are a few keys distributed among IOG, Emurgo, and CF and a majority of them can sign governance transactions spending the treasury. Can I see such governance transactions on Cardanoscan? Where in the code are they verified?
The one, where he paid more than I think is the theoretical maximum, was three months ago. … And, yes, if I need an additional input to cover the fee, it could be a chicken and egg problem, if not, not so much. The number of bytes for the fee does not change if just the number changes.
Sorry autocorrect and also I have fat fingers and typing on tablet from bed.
I know how it works thx.
1) The calculation of fees is described here and is essentially unchanged: https://iohk.io/en/blog/posts/2017/10/19/how-cardanos-transaction-fees-work/ You also need to add the mechanism for calculating the ada to be added to the token bundle, if any: https://cardano-ledger.readthedocs.io/en/latest/explanations/min-utxo-alonzo.html 2) Depending on the OS used cardano-node and cardano-cli communicate via unix sockets or via a named pipe. The details of the communication protocol are described here: https://docs.cardano.org/explore-cardano/cardano-network/networking-protocol You should also understand the serialization format used, CBOR.
This has already been answered by Pal. As the Voltaire phase progresses, there will be a gradual transition from the current system where spending keys are in the hands of the federation of entities developing the protocol to a system based on on-chain voting by all stakeholders.
The meta question was, why I think the documentation is severely lacking. I can find some answers somewhere, too. I do it often. Point is and was that it's much harder to find them than I would like it to be.
Yeah, but it is much better than only have some max 10-page white papers that claims they solved everything. As I said it is far from perfect but much much better then others’.
Haven't worked with others. There are quite a few Ethereum developers who claim that it's less frustrating. But they are, of course, also biased.
We are all biased in one way or other.
Functional programming is very hard for those who have grown up on declarative programming, and I feel their frustration. In ethereum they can use declarative where states of objects variables can be changed but that causes inevitably indeterministic behaviours, while in Cardano it cannot happen as the states (as form of UTXO, variables and ledger states) are immutable, which gives high confidence that the program does what it meant to do.
But, hey these declarative programmers slowly moving to the functional world, by using sound type programming languages and final variables in their code.:):) but it will take at least 20 rs at least to have it as common practise.:):)
Good prorgrammers can write bad code in any language.
Indeed, even bad programmers can write, sometime, good code too. But heeeey, there is no such a bad programmer, as - almost - every programmers believe he is the best of the bests (in general). To simply test, just ask any "Are you a bad programmer?". Human psychology is very fascinating.:)
Dear Sofa, You are absolutely correct in your opinions, since they are your opinions. Facts, deliverables and extent products do not fit your opinions though. If you are seeking a debate or just wanted to throw down the gauntlets you may get it - if you could work with facts. Or, are you in search of what Andy Warhol termed 15 minutes of fame. Either way this is a pretty diverse community that is confident in the technology and the principled founder. Maybe why there is the amount of staked #ADA, which btw does not impose an unstaking penalty. Why do you think that may be? I obviously, am not the genius you are since I just don’t understand your opinions or your soapbox or your purported positions. I am so dense apparently, too since I choose to educate myself and deal with facts and keep my investment staked across a number of independent stakepools to help maintain the Decentralized nature, that is so important. Oh, and I trust this was not a vindictive knee jerk reaction because Charles invited an old friend back out of the rain. That, after all is what friendship is about. Regards P.S. the community still welcomes your investment into any of the stakepools.
not my opinions offered the critique up to the room in quote marks you
incidentally @Headelf prefer this respinse https://www.reddit.com/r/CryptoCurrency/comments/ut5i61/comment/i9859qu/?utm_source=share&utm_medium=web2x&context=3
Was originally a bit skeptical of the need for your tone earlier, but you have actually raised some good points. Another area where documentation is failing is initiate coding documentation for beginners (Plutus pioneers) there's a good post in Cardano's education subforum about this. Perhaps when things slow down, they will give that some serious attention. I don't know how hard it is to keep documentation for something that changes so much up to date...
Too much shilling – “Oh, no, they are totally wrong. They must be stupid or evil. Our project is perfect. It is proven to be perfect.” – tends to make my comments more blunt. I will give anybody that it's not easy to keep things up to date, let alone improve their didactics. I am searching the answers and references for peoples' questions every day. I just know that not all is shiny. And I think it would be good to concede that more often.
Обсуждают сегодня