My best guess is worth nothing as I don’t know anything about software development & don’t know the complexities of integrating Data Chains to the Safe network code, but yeah, definitely longer than ‘soon’
To put some rough times on this, my unqualified guess is more toward 6 months than 6 weeks to Alpha 3, with tests coming before Alpha 3. Of course, if testing throws up issues that require significant re-working of other parts of the network, it could take much longer.
Thankfully with these great weekly updates we get front row seats to see how the work is going, and it’s brilliant to see!
We’ve walked through doors Alpha 1-2, we are now knocking on the door of Alpha 3.
That’s great progress. Won’t be long before we are knocking on door Alpha 4.
It grows so fast that you will forget what it was like just a few years ago. And before you know it the baby is an adult and living down the street with the nieghbours kid, in a happy little node, enjoying a mutual vault.
Great work Safe parents: you’re raising a prodigy! Cheers
I foresee us moving into alpha 5 around September of next year. Unless something catastrophic occurs we will likely not end 2018 without a beta launch. Here’s to hoping .
I strongly doubt we’ll move directly into beta after alpha 4. The audit will likely discover some issues however small that will warrant an another alpha. Data republish and network failure recover will each probably require an alpha to highlight and test their functionality. But only time will tell.
When it all suddenly click together progress will be exponential.
‘Behind it all is surely an idea so simple, so beautiful, that when we grasp it we will all say to each other, how could it have been otherwise?’ - JA Wheeler (1986)
This phrase always comes to mind when I think of Maidsafe…
From what I’ve read, it went down something like this
David proposes Data Chains, Devs (being empowered to do so at maidsafe) propose Section Graphs…heated discussions ensue for 2 weeks, which almost exhaust David, Devs fly too close to the sun…David is vindicated.
So only David and Viv (after 2 years min) understand the network…and I distinctly remember a couple of years ago David saying most if it was still in his head.
Hopefully episodes like this make the team stronger in the long run, fail fast, fail often etc…or if in doubt just say ‘yes David’
I might be misreading your post and I mean no disrespect but I like that the team is able to hash out similar or even opposing designs together. All these devs have their own strengths, this is not to take anything from David but who’s to say who might come up with the next great missing piece of the puzzle or spark inspiration in someone else? In the end there is more reason to go one way knowing both ways and more options on the table in the event of some unknown mishap. This network is happening but it won’t get done on the backs of solely two people that is certain.
In the first moment, two possibilities were posed, one, let’s say, a classic consensus (basically tangaroa) and another, more Safe style, based on rules designed mainly by Andreas (Sections Graphs).
After choosing the second option, two new options were proposed. One call Vanilla Datachain and the other Sections Graphs.
Vanilla Datachain has been chosen but until we have some more information we will not know the real difference between them.
To be clear, I never presented 100% complete data chains, but 80-90% complete for the team to finalise. The design team then looked at every option they could for alternatives/inspiration. That took way way longer than any of us wanted, but came round to section graph. It is so similar to the original data chains we looked again at it. Using group size again in a section seemed to resolve an concerns the team had about the data chain security etc. So here we are.
I know 100% that without the team looking at every angle we would not be in such a strong position now, but yea it does mean taking time and time for us right now is something we cannot afford. So the time expense of the searching for alternatives and inspiration was for me difficult. To be clear though many of these guys, probably all of them could code circles round me and they do as well as dissect stuff ot the tiniest detail way better than me. So its a strong team and such a strong team makes non traditional approaches difficult to agree on. Overall it works, but sometimes takes too long for me and others like Viv. It seems to be the speed we can go though as Viv pushes the team really hard and efficiently to deliver “anything” every day.
So bottom line is, this stuff and the whole network challenges traditional / classic computer Engineering and doing so means we need to have the pain of proving its worth, even to our own team members at times. The better they are the harder that job is, but thank (any deity) that they do this, regardless how hard/tiring I personally find it. I just hate the fact the team seem to need to sleep and eat as that is so very time consuming and could be done after launch IMHO (joking, seriously joking)
PS Andreas though is probably the smartest dude any of us have ever worked with, plus a nice bloke, which is quite a combination. Disjoint sections and then section graph were huge advances for the team, I doubt we would be close to here now, had he not done that
Watch out with saying things like “I’ve done 90% of the work”
The last 10% is the most difficult.
This reminds me of the Belgian football coach George Leekens, who suddenly left the national team for another opportunity, saying he had done 90% of the work. He was ridiculed a lot for that, but he had already his opportunistic reputation going for him.
And of course thank you for the insight!
I still believe it should be done like that. if there is bug or efficiency fixes , they dont need to be named as another alpha. That would only create confusion
.
