DevUpdate :safe: Transcript Monday 8/12/14

Hi all!

Here is again a rough transcript of Monday’s dev catch up. We are pleased with the progress of the work in routing, account transfer, the network filesystem API and porting the client to Android.

Cheers Ben

Find it here: Transcript 8 december 2014


Stuff like:

porting the client to Android.

should not be sneaked in as an aside. It should be

porting the client to Android

It doesn’t seem to even appear in the development update! You guys. :smile:

This is awesome stuff. Thank you everyone, and these weekly updates are so helpful.


Haha well put @happybeing unfortunately for me I’m waiting for porting to iOS

@BenMS Is it possible to build messaging, and also test it?

Mostly, curious to see how it’s done :wink:

iOS is on the list :wink:


I would love to see this stuff working on iOS and Android, great for storing pics, videos and docs. But what about for politicians? Is it safe enough for them to run on Android or iOS? And for whistle blowers? If a government goes to Google and says it wanna spy on a journalist because “maybe he could be corrupt”, is it possible for Google hack in to someone’s Android and spy them? They probably say they don’t, but what if… Than they could look into the Maidsafeclient, grab some login etc…

No mainstream OS is safe from targeted surveillance, so use of SAFE on them won’t be either, though substantial protection could be gained from a 2FA login (because password stealing is not enough to obtain the login). Still not secure though.

People who believe they may be targeted will need to take extra precautions to ensure the client computer is not compromised, using a specialist OS such as TAILS or Qubes for example.


I’ve not looked at it in detail, but the work done in Qubes can be a very nice start/inspiration for a SAFEbrowser. Compartmentalization of different SAFE apps is an important feature to track, I’d say.


When and where will the installers be available?

1 Like

I was wondering more about what the mobile versions will be like as far as access. When you wake your device from sleep each time will you have to enter all credentials/or just a pin code for your credentials/or just in the background until logged out. Also would it install as an app or a profile? I was also thinking about iOS keychain being able to remember credentials and how that could be bad as well. Prolly no work around for that as far as uneducated users

We will be sure to announce that on this forum without delay. The installers are mostly pending on us finalizing a stable version of next, so that we can merge it into the master branch. Bear in mind that the first versions of the installers are experimental and specifically aimed at creating the first bigger testnet with the help of all you guys ! It will still require mild terminal talents to see anything cool happening - there is no ‘vault manager’ graphical user interface yet.

cheers Ben


I don’t mind using the terminal at all, actually, I prefer it in some cases (like this one). As of now, I can’t get the client to start and connect to the existing network, so having that will be a very welcome addition. Looking forward to it :slight_smile:


I have been trying not to write that post for the past few weeks…
Thank you.

I’m all agog, jist pure agog, so ah um…

It’s a valid question to ask. :smile: especially after an announcement. Also while I don’t especially like the terminal i can use it. Though having some good documentation would be appreciated.

I don’t mind the terminal, I feel “closer” to the code with the terminal.
Good documentation will come, its always hard for devs to write good documentation and the temptation is always to press on and get more stuff working rather than “waste” time writing up whats done.
Although as we all know its not a waste at all - just an awful lot less fun.


Writing code doesn’t help if no one knows how to use the app you created. Such is the nature of language, or of invention in any context. If you create something in order for it to be useful to others you need to explain how it works. That’s why when I code (granted I don’t code on the level or in the language of the devs here) I include annotations about everything I write. And that’s also why take notes as I create things. If you document as you create you don’t have to write mass documentation afterwards. Yeah I know good documentation might seem like a waste of time to the person writing the code but then writing the recepie might seem like a waste of time to a cook that’s creating a new dish on a fly and knows all the incredients and all the proportions in their head but is invaluable to someone who has never heard of it before. But what is obvious to the creator by grace of having created the product must be imparted to the learner via communication and this is what we call documentation and education. The job of the teacher is to make the obvious apparent.

It is an interesting aspect documentation. For code then all you need care about is the public API and that needs documented really effectively. What you see with our docs so far is answering the questions
How can you do X
Y will kill the network

and much much more. Its great as the community get info but tire kickers and experts also get a glimmer and keep telling us how it does not work (even if the testnet shows the part they talk about does work). Its like Tim Berners Lee having to say how httpd + tcp + bgp + atm etc. all works and so on. It is a good thing about dev these days, but at a cost. I think the cost is good as the value I personally get from the open approach is amazing, although way more tiring that you can believe :slight_smile: I think the hassle comes form folk who want a single line description of a deep issue that does not translate into English very well.

In terms of Apps then Gui and docs is good for many, but not all, i.e. physicists in CERN use a system that is command line (interpreted c++ actually). So Gui lets you easily use an app but cannot give as many options for deep interaction. For the general public and further it is way more than enough. For devs and scientists who want deep interrogation then perhaps not.


Yes, consider that all applications have heuristics even if you do not know the mechanisms of the protocols like

Yet you can still use the applications running on http

I’ve run into this same issue in different contexts (In my case web design and game map editing.) so I appreciate what you’re saying about one size not fitting all and the value of having more advanced options available.

I don’t need to know the history of how firefox was coded or the brilliant chronicle of the rise and fall of the HTTP Empire Interwebs in order to browse to a website: I just need to know what field in firefox to enter the url. Likewise when madesafe and it’s subsequent storm of apps are released it’s not that I’m going after more or less info than I need be they gui or command line. I’ve used both kinds of programs. (In point of fact I’ve written a few primitive command line programs.) I just want docs on what all the commands do for said command line program and how to use said given app whether it’s gui or not. David Irvine is quite right, explaining how precisely some of the algorithms work would require increasingly precise technical know how. However if all you want to do is run the app you’re fine. And if you’re looking for more technical documentation then wouldn’t that fall under the development section of the docs anyway? I get the feeling the documentation needs to be seperated into tiers of some kind. Tier 1 for just rough overviews. Tier 2 for more detailed descriptions of usage. Tier 3 for advanced usage. Tier 4 you’d start going under the hood and tier 5 you’d be knee deep in development. So on and so forth. You get the idea. You did say the more detailed questions you had the greater your understanding would need to be but of course the corrolorry of that is also true, the smaller your questions are the less your understanding needs to be. And so your understanding is proportionate to your curiosity. If you want to know more you need to learn more in order to understand more and delve into higher tiers.

1 Like

I cannot stress this point enough. People need to know this or else they might get a false sense of security.

Anyone under targeted surveillance already has no privacy and nothing they do is going to fix that. On the other hand for people who aren’t targeted it would be dramatically more privacy than they previously had.