Short post: Offense on the Defense

Apart from love, here too all is fair. Hence, the offense may be pushed into defense every once in a while. Yes, think that one through.
Or, that is misinterpreting it. Offense and defense do a danse macabre while the content fights out at higher abstraction levels. Think that one through ..!

[Edited to add: this link, and this one. Others apply as well.]

OK, ’nuff for now, and this:
DSC_0705
[Not even unique, as a NY wedgie; only just (…) the prettiest]

Preventing detection

At last, there’s a resurgence of non-preventative infosec (#ditchcyber) efforts. As, e.g., here (in Duts though the orig would be Engrish ..?) and here (a decent one, almost making the right point; co-typical ..? and on second reading, a bit empty of actual actionable advice). Hinting at leaving the Prevention Imperative and refocusing on Resilience.
Because ‘deperimetrisation’ may have clouded the longer-term, more strategic failure of locking oneself in and shooing away the so grossly underestimated enemies by one’s own utterly ridiculous overestimation of … authority, power, capabilities and competences, considered-self-evident importance (quod non…). The dumb not realising how dumb they actually are…

We’ve said this before, over and over again. And we’ll say it again. Because the Laggards (hey remember yesterday’s post?) still haven’t got it, deeply enough into their veins.

But, we have a start of that at last. Why only now? Because even the most conservative (sic) can no longer hold the fort (sic) of box-shipping at all levels? Anyway:
DSC_0804
[Rebound into the heavens!]

Why ‘cyber’s still a dud

[Oh yes @CyberTaters will warp the pings re this post. And #ditchcyber!]

For one, all (sic) of ‘cybersecurity’ (quod non) is incomprehensible to those that consider themselves ‘leaders’ in one way or another in practices where actual infosec should be top of mind. Since the (for quite too large a part) despicable mice (of this story) don’t see their own folly, these kindergarten emperors will be found to wear their new clothes well… but not ‘get’ what it takes to start developing ideas how to actually lead in the infosec field. Starting with debunking Internet myths and hype-FUD but also starting the sea changes needed to achieve something (if maybe not everything).

For another, since all the hype-FUD only leads to Technology focusing, where those that would still not have thus-focused houses on order should be fired; decades of developments would have to have been easily dealt with – though it is rocket science, it’s hence not that hard. Hey, designing and building a probe to Pluto, isn’t there an app for that?
Leaving the other 99.9% (well…) of work in the area of People (and don’t start me on Process..! see my posts over the past couple of weeks). Which, even if it would be understood what needs to be done in that field, would be known to be near impossible to pull off, let alone in the short term.

Hence by simple (?) logic, ‘cyber’whatever is a dud.

Sobering:
DSCN2508
[You know where, or not; every corner needs to be beautiful…]

Scaling ‘security’

Availability: 99.9% (per year).
‘Security’ (the C, the I) … nothing. Or, the infeasible 100.0% XOR nothing.

We may have a major issue here…

Well, we do have OSSTMM on one hand, and the seriously innovative, very important Secrecy stuff on the other.
But can we answer the question “How secure are we“..? Indeed, OSSTMM gives us a number – for the operational and technical elements. How ’bout integrating the tactical, strategic, and non-tech stuff like hooman behaviour ..? And still make it somewhat understandable to the clueless (Csomethings and other involved in the utterly useless nonsensical area designated by the pejorative joke label ‘governance’; all with the exceptions acknowldged of course); other than the above % per year estimates that are interpreted so badly..!
Oh and things like failure rates from e.g., FMAE, as presented like ‘dam can stand a one-in-a-thousand-year flood’ also don’t work – dam can break today, and tomorrow, and the statistic may very well still be valid!

Maybe it’s key to first find how to whack the notion of “1-in-1000yrs means I don’t have to worry for another 999 years” fallacy. Psychology it is but so security should be..! As many of Bruce Schneier-et-al’s posts prove (?), FUD and other angle fail so miserably.

The time (decades) we’ll need to turn around the psychos, allow us some leeway to develop suitable Scale(s?) of Security. But let’s not wait for the end of those decades before embarking on the exploratory first steps of that. You suggestions, please, today.

[Edited ahead of posting, to add: This here piece on the (declining) half-life of secrets; definitely something to include in the above ‘metrics’. ..?]

For the eye candy:
DSCN4499
[Zurenborg again, slightly edited – who’ll do the colour corrections for me?]

Before it disappears: Told you so

Oh, before it returns to oblivion; re the Hacker Team hackback: I’ll just join the endless queue of Told You So’ers with reference to this.
Noting that there is a confusing connection to the illegalise-encryption-cum-mandated-government-backdoors stupidity that keeps coming back like whack-a-mole, to put it very, very friendly.

OK, leaving you with:
DSCN4521
[Antwerp beauty, untiltshifted]

The need for a new security framework

… I feel the need for it. A new security framework.

Because what we have, is based on outdated models. Of security. Of organisations. Of how the world turns.
Bureaucracy doesn’t cut it no more. The very idea of hierarchically stacked framework sets (COSO/CObIT/ISO27k1:2013/…) likewise, is stale.
And the bottom-up frameworks en vogue, e.g., OSSTMM (if you don’t know what that is all (sic) about, go in shame and find out!) and core work like Vicente Aceituno Canal’s, haven’t found traction enough yet, nor are they integrated soundly enough (yet!!) into further bottom-up overarching approaches. Ditching the word ‘framework’ as that is tainted.

But what then? At least, OSSTMM. And physical security. And SMAC. And IoT. And Privacy (European style, full 100.0%, mandatory). And business-organising disruption, exploded labour markets, geopolitics, et al.

OK. Who of you has pointers to such an Utopia ..? [Dystopian angles intended]

Unrelated:
DSCN6146
[Your guess. Not Nancy. But is it Reims ..?]

I am not me. Myself: nope, neither.

Now that infosec has become to lean so much on the People side of things – as in theory all things Tech have been solved, for decades already just not implemented to any degree of seriousness..! and ‘process’ having been exposed as utter nonsense ‘management’ babble – it is strange to see that psychology hasn’t come to the fore much, much more. Even when pundits and others, and the minions like Yours Truly even, have posted over and over again that no tech system however perfect can stand the assault of through, e.g., casual negligence and unattentive error let alone gullibility and other vices.

E.g., in the area of IAM. Where I, the construct, the behind-the-persona ego I recognise as such, is constantly changing. In my case, developing fast, forward, up. In your case… well, let’s be nice to one another so I’ll remain silent.
And all sorts of avatars are developing as substitute for you and me within systems. See, with AI mushrooming lately, avatar ‘development’ may quite easily, soon, surpass ‘you’ in being ..?

Back to the story line: It’s just not userIDs anymore; context-aware and -inclusive, capability- and rights-attached constructs they are, and integrating with the Avatar Movement (Rise of the Machines, yes) to morph into actual beings that might soon pass Turing for comparability to/with humanoid identities. We’ll be on equal footing, then, or soon after, bland dumbed-down versions of personas/egos.

But How Is This Relevant … Ah, the clue of today’s post: Because social engineering, phishing etc. play on the weaknesses of humans to be able to impersonate. So, either stop the weaknesses (as vulnerabilities; eternally impossible) logical-OR stop the impersonation (the assumption of avatars/personas by attackers; taking down their masks). The latter, by at least being aware that the avatar, the persona, isn’t the actual person. How to get that into systems, and at the same time recognising ‘actual’ avatars/personas i.e., the link between those and the right real persons behind the masks even when considering through human weakness the persona has been ‘compromised’ …? That will solve so many infosec troubles…
But heyhey, I don’t have a clue like you do. Or do you ..? Very much would like to hear ..!

[Edited to add before publishing: Hold Press; include this on behavioural stuff]

DSCN2608
[“Riga”..? Aptly French?]

Nice note

Just a long-form quote this time, by Norm Laudermilch:

In addition, we should stop using the term “advanced threat” to describe the threats we see every day. It’s too common to hear a recently breached company point to a “very sophisticated cyber attack perpetrated by a nation-state”, which makes it sound like this was something undetectable and impossible to stop. Gartner analyst Neil MacDonald calls this the “dog ate my homework” excuse. More likely we find that it was just another piece of malware cranked out by one of the latest exploit toolkits, delivered via spear-phishing or targeted malvertising, perpetrated not by highly advanced nation-state adversaries but by comparatively low-tech cyber crime gangs. Even if a nation-state attacker crafts an extraordinarily unique and complex malware payload, they’re probably using the common delivery vectors mentioned above. Why? Because these attacks work every time.

Emphasis mine and I second. Until quantumcrypto is cracked, each, any and all cracks are of sophistication Zero. Or One, at most. Combining the most basic of ‘attacks’ i.e. exploits of negligence. Read the full article, and agree. Oh, and [self-plug] there could be side benefits in sloppiness, like this – IF deployed properly. And have your press release at hand, like this one.

So, …
DSC_1024
[Surpreme court; would you want your ball there?]

As predicted; a next container move

Actually, the speed of development of this, is bigger than it seems. Both on the impact and on the implementation side. It’s just that it’s out of sight for most.

Any suggestions how this impacts Security ..?

For now:
DSCN5522
[Next time we’ll take fresh pics; DC]

Trigger seeding

In defense of sloppy account management …
Sort of. Rather, deliberately sloppy account management.

Reading through this in particular, and that, I wondered: Would there not be a nice part of a solution in seeding your user accounts database(s) with fake accounts, to act as tripwires ..? They could be given no access to anything, or access only to honeypot-like info / environments. And then trigger the alarm when accessed – by intruders, or by own security staff or auditors when doing surveillance of controls functioning.
Somehow also, I have a gut feeling there’s some hidden secondary effects in this. Any of you who has given this some more thought already, and have info on this ..? Much appreciated.

For now, this:
DSCN1106
[This makes me look fat. La Défense again.]

Maverisk / Étoiles du Nord