Measuring network density around my vaults and averaging the results is exactly what I do. I also found out empirically that subtracting a fraction of standard deviation improved the result. See my post on the subject.
Yes, I think we can do better than that. But I am not specialist on the subject, you and people in your team are.
Finally the guilt of not running a vault consistently has gotten the better of me and I now have one running on a Windows 7 machine. The chunks I found placed in my vault was pretty immediate and of varying sizes.I need to install the vault software on a Raspberry Pi I have had for a while.
Will Test Net 4 be closed or could it be transitioned into being the stable developer network?
Thanks for everybodyâs hard work!
Awesome! Will do and thanks for the info. I just received an email the $5 Raspberry Piâs are back in stock and wanted one of those also. 
If I understand correctly Testnet 4 will be wiped at some point fairly soon so Iâd keep local backups of all your proposed uploads, public and private.
Someone will correct me if I have this wrong, Iâm sure.
Strange⌠after failing to register half-a-dozen times on a Linux desktop, it worked 2nd time on Linux laptop. Same Distro on both.
I donât think it was officially announced but this limit was already raised from 100MB to 500MB in SAFE Network - TEST 3.
See commit 716de96:
const DEFAULT_MAX_CAPACITY: u64 = 500 * 1024 * 1024
What is the difference between these two errors?
ERROR 22:27:32.080706681 [safe_core::ffi mod.rs:330]
--------------------------------------------------
| FfiError::DnsError -> DnsError::CoreError -> No such dataCoreError::GetFailure::{ reason: NoSuchData, request: Structured(b914cc.., 5)}
--------------------------------------------------
ERROR 22:27:32.213595727 [safe_core::ffi mod.rs:330]
--------------------------------------------------
| FfiError::DnsError -> DnsError::ServiceNotFound
--------------------------------------------------
The second is more common when requesting a subdomain that doesnât exist but the first request was of a similar form and I think that the PublicID existed for both, so was expected two lots of the second error not mention of Core.
You will get the second error when you havenât registered your service for www.
You can remove this error by posting data to /dns.
{ longName: 'hob', serviceName: 'www', serviceHomeDirPath: '/www', isPathShared: true }
This links the url www.hob.safenet to the folder /www on the Safe network.
If you then try to access this service through a browser (e.g www.hob.safenet) it attempts to load www.hob.safenet/index.html which points to /www/index.html. If it fails to load this file then you will get the first error returned.
You can remove this error by posting to /nfs/file:
{ filePath: /www/index.html, isPrivate: false, metadata: null, isVersioned: false, isPathShared: true }
This creates an empty file at /www/index.html. This means the url www.hob.safenet will now return a completely blank page.
You can add data to the file, and hence the webpage, by PUTing to /nfs/file/:filePath/:isPathShared.
This is all handled by the demo app but this is roughly what itâs doing behind the scenes.
Can I upload a video streaming?
Sorry, you will need to link to the video.
You can upload images, but not videos.
The vault now does not work with video streaming? OkâŚ
Sorry I thought you were asking about the forum. My Bad
Current limitation is 1mb file size limităăăăstill valid??? how to creat a complicate web for streaming video for nowâŚ
I imagine that this test might be ending right around the time of tonightâs weekly dev update.
At that point Iâll ârebootâ community1 as âThe Bleeding Edge Community Networkâ (donât take that too literally 
Its distinction will be to host the very latest builds of the SAFE network software.
The configs are here: http://91.121.173.204/configs/
I have nightly builds of the binary at the links below (they include the libsodium library):
http://91.121.173.204/amd64/safe_vault
http://91.121.173.204/armv7/safe_vault
Those are fresh builds from the repo, automatically generated late every night.
âŚjust AMD64 and ARMV7 for now, but other platforms will soon be supported.
âŚor you can use the release from Testnet4:
https://github.com/maidsafe/safe_vault/releases/tag/0.9.0
Note that community1 is not running right now; please wait for the announcement.
When is the next test network being released? I cannot say i was not disappointed that the release did not happen on tuesday⌠One would think a weekâs worth of effort was testable by the wider community.
As soon as itâs ready and if you thought a weekâs worth of effort was testable by the wider community, you were wrong. Maybe you should not decide to create an imaginary ETA yourself to not be disappointed next time. Everybody likes to have the next step today or tomorrow but I think you can be assured that they donât hold anything back when itâs ready to be tested by the community.
Why did you decide to repeat much of what i said? It is not like these questions are unwarranted considering how the numerous estimates for completion have already failed, and no new estimates are given.
Is it too much to ask for an estimate every week when they expect the MVP (or even next TEST) to happen?
Because[quote=âreivanen, post:324, topic:9829â]
numerous estimates for completion have already failed
[/quote]
now
You can always ask and like you said yourself before, theyâve always (?) been wrong so why are you asking in the first place if you know this . Donât take this as me saying you canât ask or anything but I think you need to accept that they donât know. It is step by step, piece by piece right now.