MaidSafe Dev Update - October 12, 2017

[19eddyjohn75]

Thanks Maidsafe devs for another great update,

@maidsafe

but fortunately we have been able to minimise the disruption to the engineering team

Nobody can disrupt this engineering team, because they are disrupting everything in the galaxy

Always good to see new devs coming on board, so welcome @Rachit

[Stark]

@dirvine

The debates almost finished me off to be honest.

Wow! This almost happening to someone with such fortitude is scary. Having invested so much of your time and energy into something so earth shaking just to have thrown in the towel would have torn my
morale asunder.

You’ve never even hinted at the possibility of wavering in all the years I observed the evolution of this project. Your steadfast push to get us here and blinding determination have been a source of strength for me. Reminding me constantly of the power generated from one’s resolve.

Never again stare into the abyss my freind. For in there only lies emptiness and regret. Soak your sores in creativity and patch your wounds with logic. Your tempered heart will do the rest. Stay solid David. Victory against every target you’ve aquired is the only option. You will never again buckle under any weight. Strong calm creative intelligent and passionate are your only parameters. Let your desired results be the searing hot blade that clears a path through any colossal obstruction in your path. Always remember that you have somewhere to be and no f@!king time waste. To hell with traffic!

Out the way people!!! David has a world to save!

[happybeing]

Let’s not put so much on David. I can’t image what it is to carry this weight. It is his project, his vision, but there’s a cost and we can all fail and he must be allowed to too, if it comes to that. I hope @dirvine that you know how to cater for yourself - I think you do but obviously you’ve been tested again. Take care man, and the same goes for everyone who is working on this inside and outside of MaidSafe. We all have our limits, our capabilities, our little part to play and we can lose sight of ourselves in that. Take care all

- YouTube

[Josh]

Perhaps I’m not in touch with my sensitive side, but I simply took it as a figure of speech.

Nice new digs MaidSafe, looks like it should do just fine for first leg of the launch party

[VaCrunch]

I don’t see any difference.

[VaCrunch]

@draw

Up to 5 of these same block-shaped buildings, maybe also side by side. Are you sure this is a good idea …

Just paint them a different color. You’ll be good to go.

[draw]

Visual hashing? Hmm, that’s only improving my lame try to a certain comparison

[nice]

@maidsafe

Another issue that we were working on during the last few days was related to the URI scheme registration. . . . . . . we encountered different types of issues in each of the platforms.

Great news to see you are on this ! Satysfying all the platforms and WMs must be a real puzzle.

@maidsafe

reviving the Java safe_app API soon

Woot !! Yet another entry to the network More developpers develloping more great stuff !

Thank you for the inspiring news once again, glad to see you are in your new offices. In fact, the support on the dev forum and github didn’t seem to stop during the move. Did you put the guys in a truck with mobile enabled laptops ?

[Viv]

@nice

Did you put the guys in a truck with mobile enabled laptops ?

lol how did we not think of that

orrrrrrrrr just maybe be boring and get them to work from home while the move was underway lol

[happybeing]

That nice new building looks like it would make a good github activity board.

[rupert]

nice modern and easy…

well done chaps!!

rup

[bullrun]

Doh, now Prism knows where you’re located. Please tell me you put faraday cages around everything.

/me crawls back in his hole, dons the tin foil cap and mumbles muaahhaha a few times…

Seriously though, congrats on the upgrade digs.

[kemal1132]

@dirvine

Data chains part 1 - Secure the autonomous network .

This is the options listed in the update recently, option A and option B. We did a lot of work on A and now are testing B via simulations and better definitions of the “rules” there. This part is producing resuls and we are happy enough there, but until it’s completed to our satisfaction, we will all be nervous

@maidsafe

start discussions related to the implementation for the first part of Data Chains, which we are calling the “Section Graph”.

So basically you guys found a good enough solution and started discussion on how to write it down as code?HYPE

[Nigel]

Excited to see the comments plugin return. Good luck and welcome @srini and @Rachit as well!

[abittner]

Congrats to new office and space. Infrastructure fan here, how broad are the pipes in the new office? Cheers.

[TylerAbeoJordan]

Excellent. Hope the new space is warm and comfortable for everyone.

@maidsafe

@joshuef is looking into options to scale the safe_browser to improve efficiency. This is still in the early stages and we will share a detailed plan once we have one in place.

Scale the browser? Multi-threading? --what does it mean?

Great update. Thanks

[davidpbrown]

4AM and I should be asleep!.. forgive the rambling then

@dirvine

debates about absolute knowledge in an eventually consistent network

So, I wondered earlier, what this is - “an eventually consistent network”… as it sounded unnatural.

I read now that acknowledged:

@dirvine

Bottom line, any system that tries to have all peers in complete agreement at all times before acting on anything will fail 100%, its unnatural and cannot be found anywhere in nature.

So, yeah.

Disagreement is natural.

I’m guessing then at what is debateable…

In a network that wants to serve a definitive copy, there is a problem then of resolving what is the definitive and what is the error.

Perhaps the solution, if there is a problem, is to allow disagreement and instead look to a higher authority.

For example, if the network knows the file.sha256sum and keeps that from the moment it was uploaded then how it does that, is irrelevant - the sha256sum proves it integrity. If then some event or action sees another element of the network is corrupted and it provides instead file.zzz256sum, what to do?

The problem of having connected consensus when an event my disconnect elements:
So, network goes down in real world in such a way there is a split; electric cables between countries are severed for a period and rejoin - Puerto Rico event. Then, Network1 still knows what it knows; Network2 still knows what it knows. Network1, at least, will have spawned multiple other copies of the file; Network2 might have done the same or exotically might have be paused and rebooted. Still, at the point of rejoining, network 2 can suggest it knows file.sha256sum and network1 acknowledges that file.sha256sum is the same based solely on the sha256sum - not on the crypto signatures or any multiple of them… because matching those signatures perhaps impossible where a link really was lost, which would be to confirm consensus was connected at all times. Equally, even in a network that is not disconnected entirely, perhaps disagreement might arise and then the same path to resolution can be taken… “the sha256sum is correct, therefore we do have the same copy” - without stressing about the trivia of whether signatures match.

In SAFE files are fragments and the route to recombining those, becomes a different perspective.

Perhaps it necessarily becomes more complex, where datachains are blockchain like - tempting that the crypto is definitive rather than the content??

[Krishna]

@TylerAbeoJordan

Scale the browser? Multi-threading? --what does it mean?

Scaling the browser with features for end users, devs (headless mode, etc) and also in computation/multi-threading is what we would like to focus and deliver. But the question is, whether the current beaker browser would help us scale in all planes.

Since beaker browser is becoming more DAT centric and also the support for plugins are being deprecated, the problem with the SAFE Browser is keeping it up-to-date with the beaker browser upstream. @joshuef, can even expand a lot more on the issues/limitations we have right now. We are also forking components to get features baked in and make the APIs dev friendly. For example, we would like to keep the DOM API to be same as the nodejs API. The browser RPC calls doesn’t allow complex objects to be passed between DOM. We have to fork the RPC plugin and tweak them and we must also keep updating to the upstream of all these forks.

Now, we are thinking about all possible options before we put forward something which is going to be future proof.

1. Use current set up itself and keep expanding the source. No more sync ups with upstream. We will have to maintain and manage in and out. The speed of adding features might get slower depending on in house competency, compared to a pulling updates from upstream project.

• Pros
  – Well known tool set for development (Electron and Nodejs)
• Cons
  – No support from upstream fork

2. Firefox fork just like TOR. Web extensions is suggested approach with js-ctypes for FFI integration.

• Pros
  – Can support multi threading
  – Tried and tested over the years
  – Features like headless browser support are already available

• cons
  – New ecosystem for us to get used (Shouldn’t be a problem )
  – Users can install other extensions, which can lead to privacy leak. Is it possible to block extension installations? Should be possible. We will have to check/explore more.

3. Brave browser is one more option which we are considering. Brave is providing extensions through chrome extensions. For FFI integration we will have to rely on web-assembly or emscripten.

• Pros
  – Brave puts Privacy first
  – Know toolset (Nodejs, Electron)

• cons
  – Extension support
  – Custom Protocol Registration

Please do keep in mind that am spilling the information while we are still exploring. At the same time, am sure, we will get some useful inputs from the community. We will have to pick an approach after making sure we have a POC in place to ensure that what ever we require can be implemented. Also, please share other options that you may consider would be useful.

[Nigel]

I think functionality/ease of development for newcomers and privacy are the main goals. If those can’t be achieved with beaker, that’s a shame but be it sad it doesn’t mean that it isn’t true. So if Brave browser can better suit SAFE then I’m sure we could adjust. I like Brave quite a bit, am not a fan of Firefox personally.

[happybeing]

The mini-mammoth in the room… mobile browser support