It would be nice if there were some sort of “proof of ingress” or “signed chain of calls” at the protocol layer that a service could use to find the originator and verify the chain of calls is expected.
I suppose these could get very long if you have circular awaits amongst canisters doing messaging. Still…I can see many use cases for having access to this info.
I guess basically it would be cool to have access to the current call stack at the protocol level.
I think it is time to build some libraries to simplify this ping-pong match and abstract it from the user. Adding it to the list for bounties.
I don’t know what a system level solution would take.
In the meantime, it might be possible to build a workaround in canister space without modifying the system by introducing a dedicated (trusted, non-upgradeable) canister whose job is to track “capabilities”. Then callees would have to make an asynchronous inter-canister call to find out on which principal’s behalf the current call is being made. This adds latency whereas a system-level solution could be faster. So the workaround does impair speed. But it’s doable. Everyone would have to agree on what this dedicated “capabilities” canister is. But also everyone right now agrees on what the ICP ledger is and that doesn’t seem to be a problem. So why not also agree on a capabilities canister.
Won’t their CDN also be useful/necessary for IC apps?
[quote=“lei, post:2, topic:17178”]
this attitude is quite irresponsible.
[/quote] exactly! i appreciate your write up you put into words what I have been thinking exactly!
Read all through. This is great. I hope the dfinity core team work on all this issues, so the giant in ICP can rise and fulfill its potential.
This is 1000% spot on. We also invested millions dollars in engineering/design/marketing resources into supporting the IC to build Earth Wallet as a solution to some of the problems we saw with IC user experience (from real experience launching multiple top 100 cryptocurrencies and a decade of experience in the space). We had incorrectly assumed that Dfinity would operate logically and invest in growing their community projects, and in turn, would help make Earth Wallet the Phantom wallet of the IC (and no Phantom wallet is not a conspiracy theory to destroy the IC, it’s just a good product that users have a good experience with). Instead all of our requests just went in one ear and out the other, and we were forced fed Dfinity failures with everyone else.
I am not the least bit surprised at the response from Dfinity to this thread. Once again, the solution is not “we hear you and let’s build together next time so this failure does not happen again”, it is instead, “let’s ignore what these inferior idiots in our community are saying because they don’t know what’s best for themselves, we need to double down on what us geniuses at Dfinity create in our echo chamber and launch People Parties ASAP!!!”.
The problem is you can only force feed people what they don’t want to eat for so long before they all leave to go eat somewhere more enjoyable. I hope Dfinity realizes this someday so that the community can actually grow.
Given the boundary node setup and the distributed nature of the IC, I’m not so sure that the IC can’t turn into a pretty fantastic CDN itself. It may take some incentive structure, but imagine adding a response header to your http_request response that asks the boundary nodes to distribute your content to be cached for X amount of time and they charge you some cycles for doing so.
The great thing about an IC that does this is that you’d have control over both static content and adaptive content from the same system.
And they are still quibbling!! This is a sad thing.
What are they doing when the community is doubting their capabilities, the user experience is so poor. Let’s do a last kill mechanism.
By those clowns, yeah the ICPunks* sale.
ICP will rise to the top, thanks to its dedicated and loyal developer community.
Suddenly the amount of negative posts in this forum increases significantly after the main-net launch of bitcoin integration.
Suddenly a few people keep posting the same negative contents across different topics.
In crypto and signaling theory, I guess we are in the right direction.
Loyal to ICP or Loyal to DFINITY? – is the question.
Tech without decentralization or web3 values is a perfect catalyst for Dystopian Nightmare
I don’t get it, how can you guys spend millions on developing a wallet in crypto?
Nothing but do not want this post down.
It’s called psych ops in my field.
This is not the same problem. The punk was an NFT launch, which has been over 100 times already. The SNS 1 launch was something new which meant to reveal issues before real projects start using the SNS. If the actions taken after icpunk launch is any indication, following SNS launches will be much better like it did after icpunk. The UI/UX for the NNS is extremely important. The community has criticized the foundation a ton for not making the NNS user experience better. I was one of them who criticized them a lot. The NNS is probably the top place where new investors go to. So the user experience need to be up to par.
It’s funny. As soon as things start getting better. You got a ton of people start posting bunch of negative crap.
lol, im sure you dont even know what youre talking about.
you spent millions of usd on your app/software, but you want Dfinity to do the marketing for you?, i really dont get your point, its obvious they have a roadmap, so they should set that aside cause you choose to build on their network? lol.