Crust has arrived at version 0.14.0
4 Likes
Isn’t there a dev update this week?
2 Likes
Yes - later today (once I have pulled it together) 
12 Likes
Ah, great thanks 
1 Like
No, you’re wrong. You need to be involved in testing, regardless of its current flaws. Otherwise you’re asking for everything to be handed to you on a plate; that would be boorish and inconsiderate, and I’m certain you don’t want to be that.
You have said that you aspire to be a programmer, but that culture is one of can-do and solving problems; only by doing so can you get to the point where you help others and to be an asset rather than a burden (people in general, not you specifically).
6 Likes
Okay so how would you suggest I perform a test with constantly vanishing accounts?
1 Like
Are you trying to test your own app against the test binaries or just the test binaries??
2 Likes
No I’m just trying to run the bloody apps that have been released. And was responding to @bluebird statement that I shouldn’t give up just because my accounts keep dropping out. So I was saying how would one perform such tests in such a circumstance? I can’t even upload data when the system won’t remember my login credentials.
1 Like
@Blindsite2k with testnet3 the main testing was/is for people to provide vaults rather than run the apps (because as we know data was expected to be dropped all over the place).
If you want to learn there are other things you can do without the live testnet though. So it depends what your aim is. For example, I have not taken part in testnet3 because I’m working on something that needs a functional API. So instead I built a version of safe_launcher that uses mock routing and is therefore independent of any testnet.
So step one is decide what you want to do and then ask for suggestions or help as you need it.
2 Likes
I think the fact you are participating and maid safe can visualize the network is good enough. Seeing the problems etc will help them build a stable release, you’re doing your part in my opinion
3 Likes
Yes it’s just that he had a very unfortunate start, getting off on the wrong footing—by not being able to get Test 2 to work because of his very specific EXE-not-working troubleshooting issue (or he just didn’t see Test 2 in the tiny time frame that it existed; I forget), & having the same EXE problem with Test 3, & then once he got that to work, the network was failing (as predicted and expected). So it’s a large series of unfortunate events. I can see the frustration, but just hoping I can provide my take on it so that he can realize that he will likely have a really good time with Test 4—and frustration will be no more, unless focusing on the scars (instead of the stars…). Everything starts to click at a certain point, and happens at different times/rate for every person.
2 Likes
Whatever you feel most able to do that will find out something interesting or useful, even if only to yourself.
The data loss is not “constantly vanishing” but follows a time profile, and there are some unanswered questions.
But one question I have a tentative answer to is how long it takes for the data to disappear and what can stop it: I have found that accessing the data (all the way from logging, and running the demo app to accessing ones uploaded files in the demo app) on a cycle of an hour will preserve it indefinitely. Unless the network completely collapses!
I have now sussed what is required to avoid having the network collapse (touch wood).
My enquiries into automating the access to the data now has me poking around in the launcher source code. Remember that the API for this system is RESTful which means it talks like web clients and servers. These enquiries will yield learning benefits even if the transience of the data is itself a transient problem that completely disappears next week (which I doubt, although it will improve, so maybe on a tiny network you might only need to access the data once a day.)
1 Like
did you notice? 
@bluebird
we are getting our “official” safenet back! =) (and probably a more stable community net if it persists aside)
3 Likes
@dirvine 's comment made no mention of a network, but only of client stability. So we’ll have better binaries to run, meaning: more stable data storage, which has been mentioned by @ross as a feature of testnet4.
I didn’t take it to mean there will be an official permanent network in the near future.
5 Likes
oh - then i misunderstood him :-\ shit - thanks for the clarification @bluebird … everybody hears what he wants to hear … >.<
edit: but no real problem more stable client is perfect too and my vault is&will be running for our community 
4 Likes
Also, a subtle point, and one of my motivations:
I don’t believe one can accurately judge the progress of this (very important) project without getting ones hands dirty with the tech. So, at the very least, if you are participating in test networks you get to find out in an immediate way what the issues are, and with each iteration you confirm that technical progress has indeed occurred. I don’t see how one can safely delegate this. Every person (on the planet) has an agenda, and even ones own understandings have cognitive biases built-in that have to be be double-checked. With something so important it is hazardous to resort to exclusively listening to experts, because the question of which experts is unsolvable. Only by immediate sense experience can one gain a fairly firm ground to make cautious assessments.
The trolls who show up here, yammering about investment, have no idea of any of this, and to them (speculator drop-ins) progress must seem glacial.
If you’re a hands-on SAFE-geek then you will indeed know when the speculative investment timing is imminent, if that’s your interest.
10 Likes
So, 512 bits was considered overkill. Makes sense.
4 Likes