MaidSafe Dev Update - 26th July 2016

[frabrunelle]

New versions of Launcher and Demo App are almost ready to be released

We are currently working on a few blockers all within the demo app:

• The app is unable to complete the upload of certain file formats, such as JSON and PDF.

• Content streamed from the demo app to Launcher sometimes gets corrupted and therefore it’s not able to be stored on the network correctly.

We have fixes in place for both of these issues, we just need to do a bit more testing to confirm that they are fully resolved.

Although we are making a few UI changes to the demo app, we want to be upfront in stating that we don’t want the demo app to get bloated with too many features. There are a lot of useful features we could add, however, we don’t intend on adding more as the demo app shouldn’t do too much. It is also worth noting that the demo app’s primary functions are to showcase the network’s functionality, while also serving as a tutorial for developers. Developers can learn how to use the Launcher API by looking at its source code.

We are currently discussing internally the best approach to apps. One approach would be that we start to split apps, providing very rough yet functional applications for developer tutorials, while producing more polished ones for end users. This is still very much under consideration and we’ll advise in future updates whichever path we decide to go down.

RFP Update

The application period for the SAFE Browser RFP ended last Thursday. Previously we had two proposals, but now we only have one because @MrAnderson has decided to cancel his proposal (SAFEbeaker Browser) and to support @joshuef’s proposal (SAFEr Browser(s)) instead.

We intend to now give a two week period for this proposal to be discussed and then onto the funding round. At the end of the discussion period, we will publish a donation address and community members will be able to show their support by donating to the project.

Congratulations @joshuef on what looks like a solid achievable proposal and for taking the first steps with us on this process. We are currently under pressure in house as you can imagine and hope we can accelerate this process in the coming weeks.

Crust, Core: Spandan (tl) & Vinicius

Core has added support for timeouts. So if messages are dropped by the network and responses don’t make it back to Core, it will timeout (currently after 1 min) and inform the front end.

self_encryption has also undergone some changes. Previously that crate used to keep the data in a memory map as one kept writing to it and flush it to the network only when close() was called on it. This lead to difficulty in propagating the progress indication to the front end. The current modification means that each write() will try to write data immediately to the network if it can, which means more accurate progress indicator for the front end. Core has been modified accordingly to use this new facility where required.

Apart from that there was a minor fix in the DNS module of Core which was exposed during our testing as well as general code improvements. Core now has much better logs.

Routing, Vaults: Adam, Andreas (tl), Fraser & Qi

While some code reorganisation is still going on, a lot of discussions are taking place about what is still missing to get to a secure and stable network. The central piece of the answer will be data chains, and we have come up with two RFCs which are prerequisites for that and are ready to be implemented:

• We will simplify the notion of a close group of nodes, so that every node is only a member of one group.

• We will remove the distinction between node name and public ID: the name will be the public signing key of the node.

Client: Krishna (tl), Scott & Shankar

The bugs identified from the previous release are being fixed and tested aggressively by the team. The demo app has also undergone some significant changes in order to improve the user experience. Though Launcher and the demo app seem to be stable under normal network speed, we are doing some regression testing with throttled nodes and churns. There are few blocker issues that are identified as a result of the regression testing. We hope to release the binaries later this week, once these blockers are resolved.

[davidpbrown]

@frabrunelle

We are currently discussing internally the best approach to apps. One approach would be that we start to split apps, providing very rough yet functional applications for developer tutorials

Seeding app development in this way, would be very interesting to see.

[happybeing]

@frabrunelle

We are currently discussing internally the best approach to apps. One approach would be that we start to split apps, providing very rough yet functional applications for developer tutorials, while producing more polished ones for end users. This is still very much under consideration and we’ll advise in future updates whichever path we decide to go down.

Both seem to be vital so I’m not sure choosing one is sensible. My first thought is that you are concerned about resources, in which case why not create one template developer tutorial with App, and then offer bounties for people to create the remaining ones required to illustrate use of the API to other developers? I suspect there are enough developers around who can do this, and create the path for others to follow.

I’d really like to see MaidSafe create some useful Apps for end users, because you have the resources and skills to create the professional experience we need to get good take-up and favourable review coverage in the early days. Independent developers can sometimes do this, but in general the quality and usability is likely to be mixed so I think it is good for us to have some “flagship” products that stand-out and show the network at its best. They would also provide a source of revenue for MaidSafe which is also important!

[polpolrene]

@frabrunelle

Congratulations @joshuef on what looks like a solid achievable proposal and for taking the first steps with us on this process. We are currently under pressure in house as you can imagine and hope we can accelerate this process in the coming weeks.

Yes, congratulations. Also thanks to @MrAnderson for putting up time and energy. It’s appreciated

@frabrunelle

We are currently discussing internally the best approach to apps. One approach would be that we start to split apps, providing very rough yet functional applications for developer tutorials, while producing more polished ones for end users. This is still very much under consideration and we’ll advise in future updates whichever path we decide to go down.

My 2 cents here:

• The Demo App should focus on everything around identity. These are: Messaging, Personal blog/website, Public ID, Safecoin wallet. We should find a great name for that one. It’s like the killer app provided by Maidsafe, one that everyone uses.
• Upload of private data has nothing to do with identity. Great as we have it now, as people can test. But a better option might be a dropbox like map on your desktop? Don’t know how much work that is, but people would understand it directly.

[MrAnderson]

@polpolrene

My 2 cents here:

The Demo App should focus on everything around identity. These are: Messaging, Personal blog/website, Public ID, Safecoin wallet. We should find a great name for that one. It’s like the killer app provided by Maidsafe, one that everyone uses.
Upload of private data has nothing to do with identity. Great as we have it now, as people can test. But a better option might be a dropbox like map on your desktop? Don’t know how much work that is, but people would understand it directly.

I think you hit the bullseye on the target. I would call the core killer application that establishes end-user identity on the SAFEnetwork…MaidSafe.

[dirvine]

@polpolrene

Yes, congratulations. Also thanks to @MrAnderson for putting up time and energy. It’s appreciated

Yes, thanks @polpolrene I take the hit for this one, I typed that part of the update and totally missed the opportunity to recognise @MrAnderson and his contribution and heart to have done so much work on this one.

I apologise @MrAnderson for not bringing this up specifically in the update, you deserved to have been recognised. I hope this small after the fact message though at least helps a little. We all seem to be prepared to work for the greater good at times and it does re-affirm our humanity. So again nice one and thanks again.

I hope this is one of many RFP’s that we do and I hope they increase in volume very soon. I have a feeling they will.

[waynenilsen]

Maybe I missed it but any update on structured data and safecoin?

[MrAnderson]

That is quite thoughtful, thank you @polpolrene and @dirvine . This community has given me so much and I’m happy to help where I see opportunity. Recognition isn’t required (Probably not why most of us are here), but greatly appreciated.

[19eddyjohn75]

Fun update, good to hear that new versions of the L & DA are coming, can’t wait.

@happybeing

Independent developers can sometimes do this, but in general the quality and usability is likely to be mixed so I think it is good for us to have some “flagship” products that stand-out and show the network at its best. They would also provide a source of revenue for MaidSafe which is also important!

Yeah if only I could hire the Maidsafe team to create my app, ideally have this running on my mobile it would showcase that you don’t need a playstore or Ios platform.

With all this API talk isn’t it also helpfull/spreading the word, if devs put up a few examples on Codecademy?

,

[kcyy002]

It’s very nice to see a new update, thanks for maidsafe development team, you did a great job

[joshuef]

Thanks! I’m excited to get going .

I’ll look to refining and defining a few goals for network access over the next weeks. A lot has already been put forward in the proposal thread, so there’s some interesting things to consider!

The more we can define a ‘pattern’ of what to expect from a SAFE browser, the better.

Feedback will be key before starting, so please fire into the thread with ideas and discussion points. I’ll be punting some suggestions soon for consideration also.

[m3data]

Great work as always from the team!

I had a play with the recent release over the last week and impressed with the progress. I recently started contract work with a company and am drawing some attention to the capability of the SAFE network to the new CTO and current CEO. Working on some great use cases at the moment around personal data ownership and control, MaidSafe is really such a nice fit for what they do. Updates to architecture are being discussed so I am looking to plant some seeds. In time I shall nurture them and see them grow

Again top work guys and respect to all of those who have persevered to bring it to where it is!

[Nigel]

@m3data how does meeco work? Are they open source and do they have a github?

[m3data]

I don’t want to hijack this dev update thread with how the platform works in detail. Worth checking out some material on the meeco website. I’m hoping to engage this community as I get more influence within the business. Unfortunately it is proprietary software and from my understanding I don’t see the scope for it moving to open source. Hard to make that move for a whole platform once you take the proprietary route, maybe someone can correct me on this though. API documentation and an SDK will be available soon. Probably start a thread on here in the near future.

[happybeing]

@m3data meeco will be an ideal for for SAFEnetwork so just keeping them informed of progress and benefits will be suitable. See what happens! Good luck with the job.

[Nigel]

Okay good idea and thanks, had checked your original link but couldn’t find and source etc.

[m3data]

Yeah there is clear synergy. I’ll do my best to get “enterprise” focused CTO interested and aware of progress and benefits, lead dev is already keen, so that’s a start