@cdew, you might be confusing days with some weeks and weeks with some months; Just watched the videos of that event
I love the optimism; certainly; all good things come to those who are patient and work very hard;
MaidSafe is 130 million years in the making after all
you might be confusing days with some weeks and weeks with some months; Just watched the videos of that event
where is the video posted? they repeated several times that testnet3 would launch that week and beta several weeks later. it could be that they have edited those comments out of the video, perhaps due to further delays, but I have no reason to be optimistic one way or the other, all i know is what was said at the event and am simply relaying that message here.
assuredly no (about editing that comment), expanding the API this taking place this week and testing of the work before test net 3 is out definitely; This week SAFEX specifications and implementations have taken place;
I think we talked about trading algorithms and also how to build a simulator for that ;
Even though Ethereum is not much farther along; casually downgrade MaidSafe remember don’t bet the whole farm on any ONE trade; so good to know of the diversification man;
I understand you guys are creating a new internet and that things will creep up in development but I really think there needs to be a firmer development shedual. 3 to 4 weeks compared to 8 months? That’s over an 8x increase! Either be more pragmatic in your release date estimates or more efficient in how you organize things. If a company said “We’re going to release our software at such and such a date.” and then didn’t they’d lose creedibility. And while I realize that maidsafe is voluntary and working hard the same principle applies because it’s about creating expectation and fullfilling that expectation. So either don’t create the expectatoin that you’ll be done at x date or get it done by x date. Still that all being said it would be good to see a clear progress bar to when you guys would be done.
The Devs were “trying” to finish TestNet2 by Dec 2014.
But recent code rewrites delayed it… for good reason. This thread and @Al_Kafir’s telephone analogy explains it nicely. See below…
Could we have pushed through with the way the code was written? Probably.
Gavin Anderson said collaborators of Bitcoin thought the code was genius but also outright messy in other areas. It’s still being developed.
I realize none of this matters to some people, and that’s okay. I too thought we were going to hit TestNet3 by Jan 2015. But I’m glad the Devs are trying to simplify the code, and make it easier for future developers to understand and build on top of it.
It is easy to say, but hard to do though.
The only way which seems to work well with software is to define small release cycles, with small, defined tasks. This has resulted in an explosion of companies adopting the agile scrum methodology/process.
The idea is that you build additional functionality incrementally, so that each phase adds a small amount, but with plenty of communication to ensure the right thing is delivered within predictable time scales.
Software is malleable and it doesn’t need everything planned out at the start. However, it needs to work correctly at the end. This is why engineering processes designed for less malleable technologies don’t fit so well.
It is a change of thinking though. It means thinking about delivery at every sprint. It means breaking things down so that this can happen. In short, you deliver what you can by a release date, rather than slipping the release date to fit in what you originally thought was achievable.
Failure to deliver anything in time is much worse than delivering something less feature rich thanks hoped.
This is spot on and a reason we switched to Jira. It will help a lot (already has) as we move to sprint based deliverables.
The part folk find hard though is we cannot deliver partially an autonomous network. So that part needs to be 100%. This is why we are using the refactor we are completing just now to deliver very basic but solid functionality.
I have been like an evil axeman and cutting away parts to be delivered as the network and vaults are very stable and very easy to upgrade. This is what the last few weeks has been about.
Then you will see sprints for messaging (again) safecoin, client API parts etc. (Viv and the team there have been hidden by us in the core side and that must frustrate the life out of them) and these will involve core devs app devs marketing admin deign community etc. basically everyone.
So we need testnett3 started again and then we can deliver smaller chunks on top of the network. We have a ton of stuff already done, so sprints will be very clean and clear for all to see. Its another reason for looking at rust, a sprint in c++ is very hard to have a deliverable in a week or two weeks (max sprint really should be two weeks).
This part has been very hard in house with many folks very agitated and much more fighting than usual as nerves are exposed with the refactor. This is hard but it needed done. I was very demanding there as we had too much code and had found some very basic but powerful algorithms. So the next few weeks will see the testnet3 up, but hopefully you will see more than that, bus weekly progress again and not refactoring (which his always an alarm bell situation, many times it means rewrite, we got it wrong).
So until the testnet3 is up its almost impossible to state dates and we try very hard no to for good reason. After than though sprints will happen and crucially as you say we deliver, but perhaps reduced functionality, but deliver is the key. Everyone benefits then community and the devs who get very little thanks for this effort they have put in since November, it has been herculean and also tough as I have never been tough but have had to be and that is new and hopefully over I am sure if it were not the guys would slit my throat
Ha ha, refactoring in this sense is the equivalent of pivoting in the business world.
I too am a bit impatient about the MaidSafe launch but the SAFE network seems like impressive technology to me, so I can wait. Besides, as a developer I can keep developing for the Web and switch to SAFE later. It can be a more anxious situation for those who own a lot of MaidSafeCoins.
I hear what you are saying. Even if you break everything into tasks which fit within a sprint, it doesn’t mean that you will have something releasable at the end of 1 or more sprints.
However, I do think it gets the team thinking about breaking things down into digestible chunks. Breaking these down further each sprint then really helps to define with some accuracy what can actually be delivered in a sprint.
I am sure I’m preaching to the converted, but it took us getting agile scrum training to really ‘get it’ where I work. It is easy to adopt daily scrums, sprint cycles etc, but somehow miss the context. We were very much going through the motions, but missing some core areas which were essential to it working.
Saying that, we are still settling into full agile scrum ourselves, but each cycle is getting better as we understand more completely how it all works out.
Anyway, I have absolute confidence in the technical capabilities of the team. I am totally confident that you will deliver. I just get the impression that the release cycle needs a bit of greasing!
I believe the exchanges are open 24hrs if you have lost faith
I’m very excited to hear this!
Or learn how to code in any language and start your own little simple project. This will give anyone a realistic perspective of what it takes to create something huge and complex as maidsafe is. What the maidsafe team is doing is unprecedented and is impossible to foresee every potential problem and situation that can arise. Anything this new and innovative will certainly have plenty of bumps and setbacks along the way. Let’s get beyond using maidsafe as a quick-rich scheme and embrace the incredible journey this project is undergoing. We are witnessing history mate!
After that, the community would grow exponential (just like it will after launch is my guess), and they would create support for mobile applications, build browser-plugins etc. create a Java-version etc.
But that’s my point of view. Maybe I’m wrong. Really excited to see Testnet3. I have no problems with waiting some extra weeks. Ethereum was supposed to be live already. They will “only” launch a blockchain this month, no GUI etc. which you can only run using the command-line.
This is the frontend team. They are pushing ahead with many different app helpers and examples. Helped by Lee and Fraser from the core team (nearly all Lee though in the Posix and Rest like API (amazon type s3 interface) Fraser as usual is everywhere really). This is where all the visibility will go when us lot in core have done what we are supposed to. We are really annoyed to have taken so long, but this is nice actually, when the core is stable again there is a lot to show on the frontend (the part users will see).
That team have really been chapping at the bit to get moving and we keep pushing them back, so they have diverted into many more areas in the hope that we can show significant progress on launch as far as user app development goes.
There is no point making this team core devs they are too important for that and their skills are not aligned (well except Lee and Fraser) with [edit core ] code.
FYI This team is growing fast and nobody has seen their output apart from some web interface stuff/visualiser and some designs and videos. I think this is our wee secret mission force who will surprise everyone with a whole new raft of additions showing the capability of the network. They deserve it for sure, but you will see them all soon. This team are as large (or small == perspective ) as the core dev team so imagine what they may have up their sleeves. I really do not fully know, but I see parts and its impressive. A way to go and a ton of work but these folks will make sure the front-end progresses fast whilst core implements safecoin and compute and so on in core.
I should say this team is Shona, Viv, Spandan, Krishna, Scot, Gildas, Shankar with help from Lee and Fraser as I said, its a great team and they are really great at working closely on slack.
OK, this makes great sense! Guess that’s the part I didn’t see while following the project. The moment the core team decides to implement something new which takes some extra time, they divert into different platforms. That’s actually a great thing than Again, I can wait for some more weeks… I know that you’re all are working very hard to make this thing happen. keep up the good work. You will blow a lot of people away with this project
Well if we were no closer than 6 months ago, it would be infinity
Seriously we have put in a ton of work and not in a non SMART way either. You may have missed lots of the updates etc. but there has been a pretty tremendous and I think brave effort to make a huge improvement before launch that we will all benefit from from many years. It was brave and it frightened the living daylights out of us doing it but the outcome is a much simpler and should be significantly more efficient and certainly secure system.
You won’t find people more dedicated and desperate to launch, we have incredible focus on that, but I have also a razor focus on doing that right as well as fast as possible. So we won’t be perfect, but we will be much closer to that than we will be to sloppy.
Hope that explains a wee bit.
That guy in your profile pic, wasn’t he the ‘wax on’ ‘wax off’ guy…breathing, relaxing, focussing…he seemed to display great patience
He was! He also caught a fly using his two little sticks while he was eating.
@dirvine, thanks for the reply. So you say TestNet 3 will be out soon, is 2-months a valid timeframe for TestNet 3 to launch?
Also with TestNet 3, will we have public access to the BETA version of the client?