SAFE Network Dev Update - January 10, 2019

Great update to start 2019! Thanks team!


Not maybe, some will for sure, depending on the length of time. All we can do is keep going and keep working as hard and SMART as we can. No more and hopefully no less.


And that is no more we can ask from you guys! We will wait patiently on the next update, and the next until we stop waiting and can start using it!

I hope everyone had some great holidays and looking forward to the developments of this year!


Great start to the year, SAFE. The android thing is huge. Access from mobile is obviously critical to wide adoption and this is a major first step. Looking forwards very much to this year’s developments. Keep truckin’!


If you go back and look at what has been accomplished over the last 12 years it’s pretty breathtaking. I’m not an engineer or developer. I can understand things conceptually—at best—most of the time. I know that doesn’t inspire a lot of confidence for some, but I think it’s fairly obvious that this team is not only more than capable but also kicking major, hardcore ass at every level. So excited for 2019.


I could be a smart-ass and say at least we get updates with substance. But we all know this anyhow, including yourself I assume.

Updates and timings are always a major pain in the ass for any large long term project. Just read up on some of the major construction projects that were 10-20+ years long and they only had to give updates to the money suppliers. The media updates were fluff usually.

This was/is a research&development project and a lot of work to be done to bring it to completion in a way that sees it working first time securely without bugs that could destroy the network. This is what makes it more difficult and time consuming, the requirement for whats produced to be secure from hacking when it launches.

The community network shows that we could have a pseudo-safe network almost tomorrow, but honestly it would not be minimally complete and I’m sure quite hackable, especially since it is not even using PARSEC.

Still if you look back over the last 12 months then we can see a lot more progress in areas separate to PARSEC implementation and just as necessary for alpha 3/4 and beyond.

The trimmings are also being made with increase in UX/UI developers and their development needed for the future test/releases. There is marketing staff increases required for the future. And could continue with a list of things done in 2018 in other code developments.

This network is not a software project where the emphasis is on releasing a product in order to get the funds flowing in so they can actually complete the product (like 99% of s/w projects out there). Those products are usually bug ridden and easy to hack. Remember vista, it was an alpha product released by microsoft to keep the money flowing (and full of sh*t).

SAFE cannot be like that, if it was then they could release now and continue. But if it released now then the network would need multiple restarts due to problems (eg no PARSEC) and multiple hacks and spam attacks. Basically more people would leave than will by making people wait until SAFE is actually secure and ready for the hackers/spammers to attack.


Great update for the beginning of the year with all the holidays and so on.

Hoping for alpha 3 & 4 this year seeing as progress seems to be well on track.

And Happy New Year to all the development team, marketing team and any others (Nick etc)


My Friday mornings are back to normal. Just two weeks and it seemed like eternity just to know what is going on. With all hope 2019 will be 'the" year.


Great start to 2019 you guys. Do I detect a new author for the update?

The last time the “around the corner” term was used it came pretty quickly. Anytime this year will be fine.


Another great update, so much going on.
Thank you all for all you have done and continue to do.
As has been mentioned so many times, it really is great getting such quality updates.

Looking forward to another year of groundbreaking achievement.


PS. I thought that the network restarts is to be worked on after Alpha 4.

I’ll choose to ignore the slightly negative tone of the rest of your comment and give you a clarification on this as it may be of interest:
We’re not “working on network restarts” as in: implementing support for network restarts. The phase we are currently at in routing (the subteam that is not focusing on the remaining items in PARSEC) is paving the way for Fleming by having a holistic look at the design of the backend in light of all scenarios that will have to be handled at some point.

So we are looking at all the elements of design we have; and making sure that we won’t tie ourselves into knots once we have delivered Fleming.

It is a really useful time because

  • it allows the entire team to build a common understanding of the complete picture, which is important before getting knee deep in the code
  • it gives us an opportunity to revisit design decisions and see if anything should be tweaked before getting to the code
  • it gives us a possibility to consider some of the “big issues” that we have sometimes postponed as it wasn’t the right time to spend a lot of effort looking into it

So basically, I can understand how this phase is one of the phases where progress looks slow from the outside as there is relative silence from us (as in: github, jira etc. aren’t super active); but it’s a pregnant silence.

Fleming is actually starting to take shape in the team’s mind. You’ll need a bit more of that patience you’re well known for before you get some more measurable outcomes :wink:


Reminds me of Hofstadter’s Law:

It always takes longer than you expect, even when you take into account Hofstadter’s Law.


And Hofstadter was an optimist


Ha! Just re-reading Kahneman’s Thinking Fast and Slow as it happens. Amazing book. I’d also recommend The Undoing Project by Michael Lewis which tells the story of Kahneman and Tversky and how they sparked off each other. Back on topic, software and engineering projects always overrun, both for reasons of over-optimism and also because the need to raise funds means a sales story is required. It’s worth mentioning though that the average time for a scientific research project (for medical science at least) to reach production is 17 years. Since SAFE is breaking new ground in many areas is not reliant on blockchain and is required to be as secure as current techniques allow, I think the analogy with scientific research is fair.


The corner is probably Valve time in the future. And hopefully Murphy isn’t encountered before (or after) reaching the corner :wink:
Edit: speaking about books The Mind’s I of Hofstadter and Dennett was one of the more memorable books I read long time ago (I don’t read many books anymore).


regarding network restart… will it use external state storage? even though I think maidsafe team has best consensus algorithm ATM, its would be hard to recursively store state of network to itself.

It uses the network storage itself - the existing vaults. At least that’s my understanding. Restart means vaults being able to prove to each other who they are, their reputation, that the data they have is valid etc during a reboot.


You would think with all the redundancies happening in the crypto world, there would be a Network Engineer or 2 lying around. Anyone know what the downside is for the project is if one can’t be found?


I just love this comment! Because it boils down to some core fundamental values that our society seems to lack these days:

  • Quality > Quantity
  • Thinking long term instead of short term
  • See the big picture (but also the details!)

It is all about structuring things and reducing complexity, especially in a software project.
I think the following post by @dirvine reflects this pretty nicely, concerning software development:

You can transfer this to so many things in life (especially to problem solving)! It is a designer’s mindset!
The same is true, if someone says to you that something is supposed to be easy. If you really take trouble and dig into the details you will oftentimes see that it is not that easy at all!

@maidsafe Please always stay true to your core values. Not being first is actually an advantage!

I really love the new update! Keep going @maidsafe ! :heart:


While one of my hats is a network engineer and as an actual Engineer I can learn new skills quickly, the requirements for the position are quite detailed and with RUST ability being one of them, this narrows down the field a lot.

Normally a network engineer is about designing networks and keeping them working, it is usually using hardware & wires & router os/config and (lot) less programming. Maidsafe’s role requires that knowledge to be expressed in programming which does not suit a lot of network engineers. Now if I knew RUST (> 1 year experience) and had NAT traversal experience I would even consider applying for the role. But at my age the old gray matter is not as sharp as it used to be and a younger mind would be better because of that and also more importantly the longevity needed for the role. (And I have caring duties at home)

Well the project is slower to implement. The role would be currently filled by multiple other developers applying what excellent skills they have in that area. With the network engineer then the project would be able to progress that much faster and have someone who can be an expert in that area.