SAFE Network Dev Update - July 26, 2018


#22

Great Update as usual, plenty in it this week.

Also Great effort and job there @Sotros25. And always great to see the face behind the forum’s userID. Sounds like it was a roaring success.


#23

I really appreciate all the data about the website design and logic. I felt the impact looking at the site, and it is neat to see the background on how it was derived.


#24

Any synopsis of what the paper was really lacking from their perspective?

Also neat to know the Chicago meeting went down! Glad we have awesome community members spreading the good word. I may do some tech presentations in SC/MN when the network is available and I have given enough technical study to the project as a whole to be able to answer the hard q’s to some degree(or understand it enough to know how to formulate questions back to MaidSafe team to get clear answers :smiley: ) . Could even have people from MaidSafe team on some of the big community meetups in future via skype or something to actually tackle challenging questions on the spot(which are sure to come when talking to a bunch of computer scientists).

But only that last bit after network is done because until then the race is on, which “crypto” team can arrive at the moon(a quality offering that meets all original goals/check boxes and solves real world problems, coin price will reflect this too anyways to all you impatient investors :stuck_out_tongue: ) the quickest to market.


#25

Really outstanding work as always!


#26

Actually felt like reading this was less of a chore/more fun than normal… yet is twice as long! Maybe people forgot Thursday is update day; …I was expecting at least 70 likes by this time (barely over 60). Or maybe that’d have been reserved for PARSEC paper not having been ‘rejected’ (by 1 group…out of 3?) for the CESC, whatever that means! Luckily SAFE Network is so profoundly positive that no hurdle matters!


#27

Thanks as always Maidsafe devs for your hard work, really liked the design mindset and Information Architecture.

Shoutout to @Sotros25 Really amazing work, can’t wait for the video :+1:

:stuck_out_tongue:


#28

Please dont, the existing site is excellent, modern and to the point…unlike that new monstrosity that looks like it was made with a windows 95 web template.

If this is the quality coming out of the Indian operation ( and talked up by Marketing) something has gone badly awry.

I’m thinking Shona is the only person at Maidsafe that has any Web / Video/ Audio Design skills…time for an audit, who actually knows what their doing and are people in the right positions?

There’s standards in companies called ‘style templates’ I would consider the current Maidsafe site as the ‘style template’ that should be cloned.

Get it back on track guys.


#29

PARSEC paper not having been ‘rejected’ (by 1 group…out of 3?)

Well you can’t argue with the 1/3 or more, necessary to reject/disrupt, if you are for PARSEC :wink:


#30

Any synopsis of what the paper was really lacking from their perspective?

Here is the entire review comment from the reviewer who rejected the paper:

Authors provide no background/related work. It merely mentions work it depends on which may make it difficult for a non-familiar reader to understand its content. There is also no empirical evaluation of their protocol. There is no high level intuition of how their protocol works - they perform a deep dive into the construction itself.

The above makes it difficult to evaluate the protocol (i.e. due to the way it is written). Basic premise is that nodes come to agreement on the ordering of blocks via meta-elections. A “block” appears to simply be a network event; that anyone can propose. If there is a supermajority for a “block”, then it is accepted. At the start of the paper, it mentions this protocol has no leaders, but later discusses how to handle a leader timing out. It isn’t really clear what role a “leader” plays and why it is important we care they time out.

Perhaps the protocol is novel, but its writeup doesn’t make it easy to evaluate. So I’d propose to reject it.


#31

Well, good thing is that all of those things can be fixed.

Background/related work I’m sure can be gathered.

Empirical evaluation… Not sure exactly what they’d expect from it, but mav posted something like it right after parsec code release. Can surely be expanded on.

High level intuition of how it works. (Partly same as above).

And it seems they think the word “block” is used for something that don’t quite merit the label/classification/distinction (based on other parts in the protocol, or maybe the prevalent use of block becoming standard).

And then some other naming/concept inconsistencies like “leader”, must be some typo.


#32

For the leader, we have a “gradient leadership” and a “most leader node” as part of the concrete coin implementation. The point is that the “leader” being malicious is handled by the protocol, which distinguishes it from “leader-based” systems that can’t work in permissionless settings.

Like you say, these points can and will be addressed before the deadline for our next submission: it would be too bad to be rejected again based on the form since the protocol itself we truly feel is novel and quite groundbreaking.


#33

The term “block” is too vague and doesn’t convey any meaning. I have mentionned this several times in the topic about data chains in the dev forum (for example in this post).

A more meaningful term should be choosen.

Additionally, this is an unfortunate choice because the safe network is not blockchain based.


#34

Thank you for explaining the rationale behind the new website. I understand better now. It doesn’t make me like the aesthetics on my 24" screen any more, but that is subjective, I guess.
Here is a link to possible coding errors:
https://validator.w3.org/nu/?doc=https%3A%2F%2Fsafenetwork.tech%2F
I think it’s important to get those sorted, especially since Sir Tim Berners-Lee is involved with the W3C.


#35

Appreciate the transparency, keep up the great work! Excited for all things MaidSafe/SafeNetwork for the next year as I think many pieces will start coming together. MaidSafe will wow the crypto-community as a team that they should not have overlooked. Steady goes the ship.


#36

Apologies for the slow response. We have done this by putting some sell orders on the exchange and maybe some buy orders as well, not sure if buying our own coin could be viewed negatively? If the community wanted to support that would be great. Openledger seems easy to use and the support team have responded rapidly to issues thus far.


#37

Please add a link to the correct website when referring to e.g. Openledger. There are both openledger.io and openledger.info.


#38

is the milestone 2 work on hold this week as some of you go to san francisco?


#39

Isn’t “block” the perfect word for describing SAFE as a giant block storage device?


#40

That spaghetti is dangling above your desk @Fraser


#41

no (20 characters)… …