MaidSafe Dev Update :safe: 2nd February 2016

In the core libraries the wrapping up of the refactoring work, code modularisation and integration between libraries continued last week, while in the upper layers work continued on Launcher and the Drive application. Steady progress is being made toward a Minimal Viable Product (MVP) release.

##Crust - uTP / API : Vinicius / Andrew / Spandan / David

The rust-utp library can be considered to be almost stable with only one intermittent issue left to resolve; this has freed up @vinipsmaker to work more closely with the other guys in Crust.

The refactor is nearing completion and now focus is on stabilisation and improving test code coverage by migrating pre-refactor tests. New crates were created to make libraries even more easily maintainable. These include the socket_addr crate for easy serialisation and deserialisation of socket addresses and the nat_traversal crate which will contain the implementation of the SAFE network’s socket mapping and hole punching techniques.

Preliminary support for tcp hole punching has been implemented. The Crust examples have been updated to make use of the new APIs.

##Routing - Vault Integration : Andreas / Brian / Adam / David

We discussed and agreed the RFC and implemented most of the changes to the Routing mechanism, to align it with the model that guarantees its functionality. This week, we will finish adding the last few bits to that and write additional tests to make sure that it works.

We will also work towards migrating Routing to the new Crust version, which provides a friendlier API and improved modularisation.

##Vaults : Fraser

Other than the work on Messaging (see below), the main thrust in Vaults this week has been laying out the framework to allow for unit tests (using a mock Routing interface to avoid the cost of setting up a testnet for each unit test) and for larger integration tests which will use a testnet.

The initial work on both of these aspects has been completed, paving the way to allow for more complete testing to be carried out in the coming weeks.

##Messaging : Qi / Brian

Most of the JIRA tasks across the relevant libraries affected by the Mpid-Messaging RFC have been completed. Having secure messaging as an integral part of the core libraries should ease the development of applications wishing to take advantage of such a feature. An initial command line example will be forthcoming to allow thorough testing of the code.

The definition of the mpid_messaging crate, according to the latest agreed RFCs, has been successfully implemented. The work on the Vault side supporting the messaging flow is ongoing and is making good progress. Planned tasks on the Client side need to be updated to reflect the latest changes in RFCs and the mpid_messaging crate.

##Client - Launcher : Krishna / Spandan / Shankar

The implementation of the new Launcher approach has begun, based on the agreed RFC and is very much the focus for the Client team this week. The specified tasks for Launcher are expected to be completed by next week.
The tasks required to implement the Drive application shall be added to JIRA in the next couple of days. The Client guys are then planning to start coding these JIRA tasks for the Drive application early next week.
@ustulation will be away from next week till March getting married. @Krishna_Kumar and @ustulation will take time this week to plan out tasks that need to be carried out during his absence on the Client side. Best wishes to Spandan from the entire MaidSafe team :tada:.

##UI Design : Scott

Design work for investor materials is still ongoing from last week and is nearly complete. Some time was also taken at the start of this week to work with @nicklambert on the EU funding application document and to get that ready for submission.

While we haven’t been able to put any software in your hands this week, please note that there has been some excellent progress since our last update. The networking libraries (Crust and Routing) are now much more efficient and are coming together. With Messaging being implemented into Vaults and Client, the Client team moving forward with the Virtual File System Storage App and the Launcher API now well progressed (we will start documenting that this week), there is a lot to be really optimistic about.

Thanks again for your time and support, here is a link to the transcript.


Another great update. Thanks for keeping everyone informed.


I just read it and look closer than ever. I hope you will solve all the bugs soon and give to the community a testnet to prove it.


EDIT: Thanks for the update and your response @nicklambert.


We’re all just trying to stay positive. What’s happening is unfortunate but understandable. I will admit that my enthusiasm has waned a bit. I hope to be rejuvenated soon. I think I speak for many here…:sleeping:


We are acutely aware of the need to release working software and we are all very aware of the importance of proof points with revolutionary technology. The MVP that @ross refers to in the update is a effectively a beta (the first step) and from there we will start to release apps, VFS in the first instance, before adding more features and stabilising and optimising what we have. We totally understand and get the frustration, the community have been unbelievably patient and have shown remarkable good humour and faith. We are getting close guys, just that final couple of steps to get that MVP and from there…well that’s the really exciting bit.


Agreed although I believe they truly are close and the pressure is on to reach a minimum viable product so the wheels aren’t still spinning in place. Providing any kind of security and making that claim is a tall order so to me this kind of slow progression is understood. But I am very very anxious as well

Edit: well said @nicklambert!


Keep up the great work. Don’t release anything until you are totally ready.

And happy groundhog day everybody. My groundhog Fuzzy emerged from his hole today and he saw his shadow. That means six more weeks until a testnet release.


PS Fuzzy’s predictions are not completely official.

1 Like

Great work and thanks to everybody involved in this project!


I really appreciate the weekly updates, and of course with updates being weekly, there isn’t going to be major progress every update - this is a complex project.

Excellent work & I’m greatly looking forward to the MVP!


I noticed you deleted your comment in respect of the team. That’s something I applaud and cherish but at the same time I share your frustration because it’s been almost 2 years since I started watching the project, participated in the crowd sale and every week the updates are similar. It feels like the product will be finished any time now but still it’s not.
I still want to see the network up an running at least a testnet just like all of the people here but I won’t lie my enthusiasm has vanished a bit.
Other project are releasing and capturing the attention of many but Maidsafe still after many many years not even a testnet.
Again, I believe in the team and I am fully invested financially and emotionally, so I’m with you guys but please understand all of us who are waiting for so many years that we need to see something this year.
Good luck!


I think it’s all about perspective. For us who have been watching this project for that past two years, waiting a couple more weeks or months to an MVP seems like an eternity. But to David and others who have been working on this project for the past ten years, a couple more weeks or months is nothing. After all, in a 100 years, this time we have waited will seem like a drop in the bucket so let’s hang in there a little while longer.


Bro I’m always OPTIMISTIC reading these Dev Updates, Thanks for all your hard work & another great update

This made my whole week, I can go on for another week :stuck_out_tongue:

[quote=“Tonda, post:5, topic:6906”]
I hope to be rejuvenated soon
[/quote]Please do you got the BEST THING, that ever happened to humanity heading your way. I’m supercharged about the SAFE Network, but I can’t be too overexited or else I’ll get taser (Taser taser taser)


Keep grinding, good things will come

1 Like

I have patiently waited since before the crowdsale. My enthusiasm has not waned. I did expect it would be at least MVP a while ago. But if it takes just a little bit longer then when augur REP is traded then I might just benefit from the timing.



You might think that, but at least for me, it’s not the case. If you don’t get why it’s late, then you haven’t built software before, and if you haven’t built software before, you can easily be forgiven for not understanding the difficulty-level of what the SAFE team is trying to do. Allow me to summarize in layman’s terms:

Double black diamond of software development.


How is Augur coming along?


I hear you. Keep pushing, personally I can’t wait to work on hardware integrations.

Yes that will be good. A lot of the work is making things easier on hardware and Vinicius recently found an issue stopping arm, so no problems exist there AFAIK. It is a long slog and many early mornings late nights right now, but then again I seen recently the last 90% of a software project is harder than the first 90% :smiley: and I truly appreciate what that sentiment means. However, it is in good hands now but incredible focus on a lib at a time helps. So routing (main lib) and now crust is a huge deal. Vaults are easy and client parts are well handled. So we complete crust and then run through vaults (which is just some logic now and much simpler) and we are there. It is frantic and the guys are pushing hard. It will happen as soon as we can make it.

Thanks for all the understanding and yes weekly updates mean we see slow weeks and fast weeks, but it probably is for the best. I bet when we launch folk say, look at them right place right time, lucky beggars :wink: we won’t mind a jot though.


"luck" indeed…

Glad to hear you won’t mind; you shouldn’t. I’m still chuckling at the crossed-out SAFE on the Ethereum Home page-- these really aren’t projects that should compete, you and your team seem to get that, and have wisely said nothing about the “so… WTH does ETH do, anyway” factor… I mean, you could use it to build something like maidsafe, but the fact that no one is doing that, says a great deal, if you ask me…