Update 21st November, 2024

[maidsafe]

Well, the network is behaving itself quite nicely, so no updates there this week. Instead, for those that missed the chat with FYEO on Discord on Tuesday, Tammy and Brian did an excellent job on explaining why Autonomi is such a great fit for KryptPass, their decentralised password manager, and how they have been participating in the Beta testing, launching nodes of their own. The first use case is for FYEO Dark, which is the API that shows users whether their passwords have been breached and leaked on the dark web, to alert them that they need to shore up their security. FYEO already has a thriving security audits and consultancy business, and this is a new venture for them. Ultimately the client will be open-sourced and the community will be invited to beta test it. More on that discussion here.

Also worth pointing out is the Week 3 (November 12-19) leaderboard for node runners testers, which you can find here, and @bux’s clarification of the situation for token holders, shareholders and network supporters.

Thanks to @happybeing for the PRs about the APIs and registers (1, 2 and 3).

General progress

@anselme has been reviewing storage proofs, particularly the viability of checking storage proofs client side, as well as reviewing our documentation. He’s also looking at a new concept of Node Connection Score, which is a score that each node assigns to each of its peers based on its own observations. The higher the score, the more likely the node is to be honest, thus the connection is trusted.

@bzee has been delving into nodes and chunk verification failures. He also implemented a retry algorithm on the client for retrying PUTs when they fail.

Out back, @chriso and @shu have been investigating the effect of changing the number of uploaders and also, with @bzee, investigating chunk proof errors. We’re still seeing some CPU and memory spikes, but these do not seem too serious. They are also working on tweaks to the ELK monitoring stack.

And in the labs @dirvine has updated his self_encryption python module with full doc strings and examples.

Ermine worked on improving the traceability of autonomi and evmlib.

Meanwhile @rusty.spork has been manning the Discords.

On transactions, @mick.vandijke worked on reducing the amount of MetaMask approvals when uploading files through the webapp and updated the CLI documentation. He also raised a PR to add CLI wallet documentation. He also worked on optimising the upload process in the web app so that encryption occurs just once.
Finally, @qi_ma has been working on documentation and also investigated an issue reported by @aatonnomicc where a 6 Gb file Ubuntu upload is reported as successful but fails to download. Current investigations are whether archive/metadata is inconsistent across repeated uploads.

[Secretariat415]

Thanks so much to the entire Autonomi team for all of your hard work!

And good luck to everyone with your Beta Rewards!

[aatonnomicc]

Second

Well done then all the team and really looking to meeting Dave hopefully soon™

Keep up the great work

[19eddyjohn75]

Thx 4 the update and all your hard work Maidsafe devs

Welcome to the SAFE universe FYEO

@happybeing @aatonnomicc

Hopefully the devs are already feeding AIs the documentations

keep coding/hacking/testing super ants

[Michael_Hills]

@maidsafe

network is behaving itself quite nicely

This is no small achievement in my eyes. I know we get spoiled with all the great progress, and we can’t wait to get more, but a stable network at the moment is the most important thing. Next, beyond DAVE, comes a stable economy, which will guarantee a network for the future. Looking forward to the new white paper. Great work team, beta testers and all!

[peca]

@maidsafe

…also looking at a new concept of Node Connection Score, which is a score that each node assigns to each of its peers based on its own observations.

That sounds very interesting. I may be also useful to inform the node about the score so the node owner could see how well are his nodes doing in terms of being well integrated in the network.

[Southside]

I’d love to post my usual “well done and thanks to all involved” as usual - BUT

@maidsafe

Well, the network is behaving itself quite nicely,

Only because it is being tickled gently under the chin.
It will most likely be a different story when it gets tested for real.

There is ZERO to be gained by kidding ourselves on here.
No proper testing has taken place and it doesn’t look like it will. This looks ominously like a “launch” that is mandated for financial purposes and not the final rollout of a project that has been developed for many years. And if it fails will be a slap in the face for those devs who have worked so hard over the years. And I havent even mentioned the betrayal of @dirvine. Add to that a possibly a reputation-trashing event for the project and a disappointment for the new partners.
Not to mention the money invested by many of us over the years for what could well turn out to be just another shitcoin.

Communication recently has been rubbish. Plenty of fine words but little actual substance and important points from the community have simply been ignored.

Its time to get real, its time to face up to realities - There is no way this network is ready for “launch” or “TGE” or whatever form of words you want to pick.

I am not doing myself or anyone else any favours by remaining silent on this. I was quite prepared to cut the team some slack over the past few months but the lack of action and honest communication recently is a severe disappointment and a total negation of the spirit of the project so many of us have supported faithfully for over a decade in many cases.

I truly hope I will have to apologise for this post within a few weeks but somehow I doubt it.

[scottefc86]

@Southside

Communication recently has been rubbish.

Have to agree with @southside here. Another week has passed with no mention of the api, white paper, Dave app in the stages or the update. Some events were visited by the team in the last week or two and not a word about any of them. Either somethings not right or we can expect a big bang sort of launch in January but we were told it wouldn’t be that so I’m confused???

[rreive]

I’m an old developer and QA guy, so I generally agree, the hardening testing for load i/o across multiple aspects of this system (uploads, connections, relays, chunk copy and redistribute, big swaths of safenodes joining and leaving etc,) for multiple use cases in this regard has been scant, or at least not revealed much to the community. I think its time to clone the network into a Community test platform for performance first, and to issue dummy tokens to fuel such testing by the community in an organized fashion.

There needs to be a Community Test Plan and Community Acceptance Plan with some serious pass/fail metrics set for each test case mapped to primary (daily network, node and client testing individually and collectively in different sizes), secondary(upgrades, controlled rollbacks) and exception(crashes, power outages etc…) use cases,

Both documents need to be open and collaborative, so they are open and transparent with direct community involvement. I’ll help with this, any other takers?

The node-launchpad mechanism can be altered in this regard to actually trigger the running of a series of tests outlined in the Community Test Plan with the results mapped to the pass/fail metrics Acceptance Plan. Plan builds and test result reporting against established pass/fail metrics can happen in gsheets.

My 2 cents for what its worth. Happy to help out.

We need to as the Autonomi community pound the heck out of a test platform clone in a collectively collaborative and organized fashion, with the existing tools transformed and instrumented, where the instrumentation of those tools need to be re-rigged to generate test results mapped to pass/fail metrics listed in the Acceptance Plan both for primary/secondary and exception testing of individual node test sets and for a system fleet of nodes test sets , and a network of system node fleets test sets.

That said we need Maidsafe’s help to make that happen.

Rinse and Repeat the above for security testing.

That and using Github to record the Plans and the updates to both and of course report the bugs, open and close the cases, etc.

Time for the Community to get organized and roll up its collective sleeves, this is going to be an ongoing journey with milestones, but no finish line, because its continuous improvement @ scale.

Really the above is where the rewards should be aimed moving forward to support the above proposed Community Test Network effort.

[Atom]

I’ve been watching for 10 years now, never doubted, but still…
What an achievement !!
Congrats to you rockstars.
Can’t wait for next milestones ! End of January will come so fast…

[Southside]

@rreive

Really the above is where the rewards should be aimed moving forward to support the above proposed Community Test Network effort.

exactly - in my naivety I thought this is what was to happen after the Waves had ended. Though it should have happened before the Waves anyway.

There is a definite whiff of the community being ignored here and this needs to change

[Southside]

@Atom

End of January will come so fast…

Aye - too fast

[Atom]

It’s true it’s difficult to tell the network is behaving that well by ourselves for now.
But I guess Dave will help speeding things up on that front ! There might be some bad surprises.
Edit: If team maidsafe tell us a community initiative would definitely help then count me in. I still trust the team first and foremost. We have to.

[Toivo]

@Southside

There is a definite whiff of the community being ignored

Whiff? What are you talking about man, a “whiff?”
It’s not a whiff, it’s a track, a definite track.

---

I don’t know, I think the team was caught pants down when the LibP2P didn’t get the AutoNAT holepunching working in the timeframe we hoped for. I haven’t been following that for a long time now, but I guess it still doesn’t work the way it should. The basic premise of making the network run from the stuff people have at home is not yet ready technically. But there is this weird over commitment to “Launching” in certain timeframe, that everything is being turned inside out to fit into that mold.

For a network project, there has always been suspiciously little developement in the networking side of things.

[riddim]

Edit:

=) I noticed twitter posts that could be seen as communication outside this community and possibly something that counts as marketing

[Southside]

Can you do that image again without the poxy Union Jack please?
I was having my tea…

[riddim]

Hahaha - oh wow that really is an interesting flag =D

[Traktion]

I don’t really understand the extreme negativity.

I run 6 nodes on a 15yo laptop. I could run more, but I’d run out of disk space. It is very much commodity hardware.

We also don’t know what testing the team is doing with its partners. They are clearly uploading a lot of data and I suspect it’s in all shapes, sizes and much of it at the same time. Moreover, they’re likely validating it too with regular downloads.

I do understand that some in the community feel jilted though. Things are happening much more behind doors now and partners seem to be being wooed more than the community. It is a change, for sure.

However, the network does seem more stable then ever and resource usage has gone down and down for me. Devs seem to be working hard in the background to make the launch happen.

I’d like to see more openness, but I don’t see cause for panic.

[riddim]