Dec 21, 16 / Cap 20, 00 12:48 UTC

Data security, authenticity and integrity  

Hetzner is a more sane choice than amazon's cloud - but as it is claimed this is taken seriously, then I must seriously ask: what steps have been taken to ensure the security of my data whilst it exists on these systems? It's still third party so largely hinging on a vast measure of trust for a large number of topics - which in the realms of "safety and security" isn't considered the best available strategem and could be considered a loose digital equivilent

On that subject, what measures have been taken to ensure the data you've already collected using "the cloud" have been sanitised? how can you be sure you've nuked out the data mailchimp hold? Even worse, companies with proven track record for moral and ethical violations with regards to use of the data they hold, like facebook - A company famed for it's large scale storage, multiple offline copies stored indefinitely, and abusive use of this stored data - How are you going to secure and/or protect the data of the citizens simple enough to actually use something so detrimental?

Authenticity is another issue, especially as it appears I've accessed this forum, and made posts without logging in.

Dec 21, 16 / Cap 20, 00 12:49 UTC

Seems it does know who I am - I'll assume that's something on my side, browser cookies or similar. It thinks I'm logged in, didn't indicate this when I made the post.

  Updated  on Dec 21, 16 / Cap 20, 00 12:49 UTC, Total number of edits: 1 time
Reason: typo

Dec 21, 16 / Cap 20, 00 16:17 UTC

[Mod Post] Thanks for sharing. We are gathering a list of suggestions and ideas from this forum in order to see what options our users would like to see. This is just a basic forum in order to get things running for now. We will be making improvements according to the input on this board. I will be moving this post to the Feedback forum to help us in keeping track of these suggestions and concerns.

Dec 21, 16 / Cap 20, 00 17:53 UTC

As it's kind of security pertainent question, and not really any measure of feedback, I'd assumed that it was suitably placed already.

Where it was was prime for finding out what security polices and procedures are already in place, and leading to discussions of what I and others can do to assist you in improving this. I sense this will be a vast list, if for no other reason initially attempting to willingly use FB as anything other than a point of contact to direct people to trustable services, the fact that registrations initially used mailchimp to track users, the way the site was initially hosted by amazon's cloud, the way was initially devel sans HTTPS, and countless other glaring security failures suggests that this is a topic that hasn't been considered. At all, let alone wisely.

  Updated  on Dec 21, 16 / Cap 20, 00 17:54 UTC, Total number of edits: 1 time
Reason: typo

Dec 22, 16 / Cap 21, 00 12:19 UTC

Hetzner is a bad choice, they do lack of customer support and they aren't able to professionally mitigate attacks against their networks.

I can not make any statement regarding the data security on But all i can say is the integrity of posted content here is destroyed ( at the current state of the forums ). If i would interpolate the conceptional constraints of the actual software, i would assume that the security level of the website is not where it should be. But enjoy this statement carefully, i am no official and i am not part of the development team. I just analyze what i do experience based on decades of experience and a stable set of knowledge.

Dec 23, 16 / Cap 22, 00 15:50 UTC

Hetzner isn't that poor of a choice, from a trustability point of view - Short of a full scale invasion of Germany the only "security services" they're likely to let in is the .de which in honesty I've little concern from overall and generaly Hetzner have a good track record for keeping their facilities secured, as well as operable, and in terms of their DataCenter(s) they are some of the better I've experienced -=- But third party services to hold "sensitive" data is never clever, and ultimately, inviting issues.

With that in mind, I would of expected by now the folks responsible for maintaining this to of offloaded a loose overview of procedures followed to ensure data security whilst it's in a third parties posession, or at least a link to the PCI DSS report and or the ISO/IEC 27002 audit.

I'm now unsure if these haven't been presented because they have no clue what they are and thusly are insane enough to deploy services unto public access without first auditing, or if thanks to an overeager mod simply this post has been buried under piles of posts that have nothing to do with security in an aera likely to remain unveiwed by the people with the capacity to answer.

I could audit this for you - but from a legal perspective, I would require permission to do so.

  Updated  on Dec 23, 16 / Cap 22, 00 15:56 UTC, Total number of edits: 1 time
Reason: Additional data

Dec 24, 16 / Cap 23, 00 02:40 UTC

Well i don't doom hetzner. I just wouldn't use it for a project that may be used / visited by - in the worst case - hundreds of thousands of people, each day. From my point of view, Companies like RackSpace are actually more experienced in serving environments that may be in danger of attacks or may experience heavy performance spikes when they grow. Regardless if you need a ( probably managed ) cloud or a single server. The services of RackSpace do not stop you.

Well at the end, its a matter of money, for sure. But if money stops us in serving uninterrupted services in the future, something in scale of a nation went entirely wrong.

I am also a bit puzzled why we had experienced offline times at this forums or why specific features did failed. All i could assume is that there is no strict staging process with unit tests that prevents failures within continuous integration and automated deploy. All i know is, that whatever asgardia officials do, may be in the best case professional. Not using fail-safe mechanisms is away from that orientation.

I personally would like to know if we generally agree to security standards ( BCRYPT Password Hashing / Argon2 Password Hashing ) and do we utilize them ? If not, what do we use and who is responsible for downgrading the most important aspect of software programming. Which language is used and which version of that language ? I am open to sign a NDA for such informations.

Dec 24, 16 / Cap 23, 00 03:36 UTC

Trustability is a far superiour facet, and a "cloud" operated by other nations companies cannot ever fullfil that definition. With that variable firmly established, it's then a case of least trustable. There are just too many variables you have no control over. Companies like rackspace are are more experienced in lots of things, like making customer data available to third parties at the behest of anyone with the right peices of paper - which are not difficult to obtain, in the slightest. And physical access will enable to defeat almost anything, given enough time. They also tend to host for a resonable niumber of botnet nodes, and as they seem willing to rent services repeatedly to various criminal groups it would suggest that multiple nearby nodes will be potentially hostile, and linked by a really high speed connection - in the case of cloud based, potentially even on the same hardware.

There's no way money should prevent anything - I could do that on my own(maybe I do already), and I'm poor. I'm not sure if you've noticed total population numbers, but if only 10% of that was prepared to make a contribution towards, then it's take an incredibly minor individual contribution indeed - even at 10% - to secure decades of services. For a less minor, but not significant contribution we could build an entire datacenter.

I've not noted much offline time for forums, since this incarnation came up I've noted one that lasted a few mins and as that was 404 responses I'd suggest the hardware wasn't anywhere near overloaded and was responding perfectly. The initial forum that appeared then absented I'd put down to insufficient testing before deployment to production servers - which raises questions about multiple procedures. I've seen nothing in the way of failsafe mechinisms indicated.

Password hashing is an important topic, and would be addressed in an audit, but I'm far more concerned with how that data is stored, and what controls are over it's access.

For an uneducated guess as to "language" - then I'd suggest the CMS is possibly Drupal which if memory serves is composed in PHP, to assume Drupal8 then PHP >5.5.9.

Dec 24, 16 / Cap 23, 00 03:53 UTC

Well, i suggest a CMF ( not CMS ), based on PHP 7.0+ and swoole. From my point of view a CMS is not suitable in case of Anyway, i agree with your statement.

Dec 24, 16 / Cap 23, 00 07:40 UTC

Security is a tough beast to tame. It's a fight that's never ending.

Dec 24, 16 / Cap 23, 00 14:18 UTC

It's not that tough, overall. If attacked correctly.

Ignored however, is fatal.

Dec 26, 16 / Cap 25, 00 01:40 UTC

After nearly a week I'm finding it quite disturbing this thread still remains without a single post of viable information on the topic in hand.

It's not as if I have posed incredibly difficult questions, really.

I'll give it another 24hrs, then I'll open it back up where it was, and hope someone in the Ministry of Safety and Security would actually know something about the security that's in place.

Dec 26, 16 / Cap 25, 00 10:52 UTC

I'm not sure the staff would tell you exactly how they secure the data. The thing about security is that it's often more secure when people don't have any idea how it works. And so far the Asgardian dev team has not really published anything as open-source so I guess they like it more when it's private. I'm not sure what kind of answer you actually expect.

Dec 26, 16 / Cap 25, 00 19:37 UTC

A very good way to test how strong is a cipher algorithm is to make it available in public , hiding it sometimes it's not a wise thing because you don't realy know if it is secure or not someone else could find a backdoor or a flaw which can make it useless

Dec 26, 16 / Cap 25, 00 20:19 UTC

The thing about security through obscurity is it's the poorest form of security available, only one slight step better than having none - typically being only a given amount of time before it is discovered, and less if people actually look for - as understood by anyone that has ever studied or even considered security and from what I can make out thus far, this is the only layer of security applicably employed. If you're doing security properly you can detail this intimately with condifence with regards to policies and procedures employed and it will only serve to prevent most people from even trying, as they will understand their failure has already been assured.

If you're not going to take this subject seriously - and instead just claim you are - then this will endanger not only the various projects success but the citizens themselves.

Precise details are not entirely required, it's not as if I'm asking for copies of keys, but what is required is trust that my - and others - data is being handled in a secure and responsible fashion. To take the previous examples offered by this organisation as a measure, then I am of absolutely zero confidence that any serious thought has been placed into this subject - and serious thought is warrented, considering the reach and impact of data security in operations. Especially as time progresses.

Failure to publish as open source isn't entirely a concern of mine at this moment in time, however it's not something I'm proud to hear for the least reason of failure to adhere to Asgardian ethics of freedom of knowledge. Minimally, it would allow for community-based contributions which will rapidly decrease development times and offer the widest range of skills input.

The answer I would actually expect is as indicated by the original questioning. Oddly enough. Who would of thought.

What measures have been taken to ensure the data you've already collected using "the cloud" have been sanitised? How can you be sure you've nuked out the data mailchimp hold? Even worse, companies with proven track record for moral and ethical violations with regards to use of the data they hold, like facebook. What measures have been taken to ensure data security, authenticity, and integrity of this site and or relevant databases and othersuch backend services?

Another glaring fail I've noticed thus far is the lack of authenticaiton when entering data. Assuming I am me simply because this browser has been used on this machine to connect to this service is incredibly unwise and a feature not adopted by any security conscious organisation, anywhere, as it would anticipate I can keep this machine secure. It should be, having taken measures to assure this as I don't appear to be as simple as most on this subject but relying on that isn't clever. It's common for some users to allow third parties use of their hardware, or the use of third party hardware as a minimal issue. I find that an interesting feature because sensibly not trusting the browser itself for such means, I don't allow it to save any such details. This would imply there's some sort of cookie/authetication token stored in the browsers temporary stoarge pool, which in most OS deployments isn't treated with any particular protections from other users and/or software - or via HTML5's "local storage" option, which in it's default incarnations are notably insecure. It's certianly not something that should be recycled across multiple sessions. I understand this is possibly deployed in the name of ease of use for the "less advanced" citizens, but failure to authenticate the end user is indeed as claimed instantly raises many questions with regards to the data found input. It's certianly a problem waiting to happen.

  Updated  on Dec 27, 16 / Cap 26, 00 02:26 UTC, Total number of edits: 1 time
Reason: typo