classic Accountability Theatre
172 posts • joined 2 May 2007
ah, that's interesting.
Is it not possible (I naively assumed this was routine) to have a "provisional" authorisation code which would deal with that situation? (Ideally confirmed by a "signature" from the customer, but let's not run before we can walk...)
Someone help me understand...
I presume they don't store payment card details. (if that assumption is wrong, then all bets are off and I withdraw my question)
So, assuming they don't, yes they need to process the data, but presumably that's done in a couple of secure sessions (one with the customer, one with the Card Issuer) but once they've received a payment authorisation, they have no further legit use for the data. So how has an attacker breached their defences? Are the secure communication protocols broken? or what...
Completely Agree but don't have the space or time to answer the questions in your final para
The short version is
1 Incentivise the use of private notarised personal data "wallets" securely stored in various devices and capable of providing the answer to some questions without revealing actual data (eg whether someone is above or below an age constraint can be revealed without revealing date of birth). Also capable - with the co-operation of couriers who buy into the idea in order to feed off the "privacy preferred" market - of supplying one time "address keys" which even the courier can expose only in sufficient detail for their current sorting requirements. (but the merchant or supplier never gets to see or store)
2 in the few instances where data really does need to be warehoused, compartmentalise it so that one warehouse may hold, for example, address data but not names or other private data; while another might hold dates of birth etc. (Only linkable with more one time keys etc)
3 impose strict video-logged access controls on such data warehouses so that if any human access the protected data, (publicly) trusted auditors will always be a) notified and b) able to discover exactly who, when, why and where they accessed the data (and, of course, have full legal rights to blow the whistle if they spot anything underhand).
Can someone who understands the technology please explain why we'd still be using rotating platters in 10 or 20 years time? Surely the evolutionary path has switched to solid state...
(that is, if the Law Commission qualifies as a Government source - not sure of their constitutional position)
the full quote from the Law Commission reveals that someone actually knows what they're talking about:
"We have concerns, however, as to whether the use of Verify would be sufficient to protect testators from undue influence and impersonation. Verify does not currently ensure that the person entering the information is in fact the person he or she is purporting to be; rather it focuses on verifying that the person exists. While the involvement of witnesses generally provides some protection against fraud and undue influence, Verify does not currently have any facility for the participation of witnesses. Furthermore, Verify relies on passwords to control access to the service. There is a risk, therefore, that testators will give their passwords to family members or carers, and might be pressured to do so by persons wanting to abuse them."
If only these people could also advise the authoritarians on end to end encryption...
I too bully my customers and colleagues into trying to take Privacy issues more seriously though, to date I haven't been paid (or even thanked) for my efforts.
I'd be very interested if you could point to a more formal version of your argument which I could thrust under my resistors noses...
Probably too late for you to notice this reply but if you do see it, I would dearly love to see evidence of that result (multiple collisions when entries compared to each other) . Not challenging your veracity. In fact I really hope it's true and there is some published evidence to support it. Would just love to be able to use that argument and wave it in certain faces...
am I missing some subtlety here?
Saying "No" - like Clapper did - would only be risky if, like Clapper's example, it was a lie
Ergo, the answer is obviously Yes. The only security threat might lie in explaining in exactly which circumstances the illegal spying on citizens takes place. (such as when the citizen is talking to someone not on American soil - which I believe is already exempted, but you get the gist)
in some of the software I develop I use deliberate random errors in certain dialogues, to spot humans trying to answer questions without inspection or thought.
It occurs to me that something similar is required for the "Level 3" driverless cars (which are supposed to be able to handle almost all situations but still need close human monitoring). i.e. the software should regularly (but randomly) send false alarms to the control panel and measure the time and accuracy with which the human deals with them. If their response time exceeds a safe threshold, take the earliest opportunity to park the car and cede full control to the human (with an auto reset of, say, the next day?)
The actual monitoring (data capture) is not (or need not be) the issue. The issue is human access to - for whatever purpose - the captured data. Surveillance is merely one of the purposes.
The problem, as ever, is that we the people have no means of determining when the technology and relevant data is switched from one mode (capture) to the other (access). This is the gap I've previously and repeatedly referred to as "Accountability Theatre".
The short version is that there are always going be justifiable cases for surveillance. For a traffic based example, I rather like the idea that we could develop an AI system to watch motorway traffic in order to identify genuinely dangerous situations emerging in real time. This would include things like some imbecile driving the wrong way, or a car randomly weaving in a manner likely to indicate someone falling asleep at the wheel. Or someone driving at twice the speed limit in busy traffic. Where such threats are identified, the AI prods a Human and they can raise an alarm, send out a traffic cop, turn on the warning signals etc etc.
I doubt that anyone is going to argue with that kind of use of surveillance. Where it crosses the line into authoritarianism is, for example, with John Robson's suggestion that it could also be used to "enforce appropriate speed limits". This is a grey area. Certainly, as hinted above, some speeds would qualify for the alarm surveillance mentioned above . Driving at 80 on a reasonably clear motorway does not. That said, we should have no issue with the data captured being used, after a serious incident (eg a fatal accident) to see to what extent either speeding or careless driving contributed to the accident.
And in all cases, where data is accessed for any reason whatsoever, by a human actor, it should not be technically possible for such access to take place, without it being subject to the most rigorous surveillance of all, with the data being provably captured to an immutable audit trail.
simple solution, at least for cases like this where deep pocket is trying to with-hold an award. Allow them to appeal as long and often as they like; with a 1% interest charge PER DAY from the date of the original award. Payable only if they lose, of course...
Sumatra easily lightest weight stable pdf reader.
Was using Foxit as well for my occasional pdf editing needs. Then spotted reference (above) to Libre Office Draw being able to do the job. Tested it and it works. Bit clunky (go into edit mode/ saves to an odg file / leaves it it read only mode/ enable editing / edit / save / export as pdf) but more than happy to put up with that in order to liberate myself from foxit...
Would have attached this as thankyou reply to who-ever it was who posted that Libre Office tip but damned if I can find the comment now!
Why should we trust the Authority's own response to an FOI inquiry?
No one we trust has unconstrained access to the data for audit purposes. It is much more likely that we're seeing an (unknown) fraction of the real offences - enough to make us believe a proper job has been done - than reality.
Until ALL authorities are under a mandatory requirement to implement audited access control with the data protected by an immutable audit trail, all attempts at "oversight" should be treated as, at best, suspect, at worst, criminal deception.
Honest question to which I suspect I can guess the answer but do not know.
Where are the highest broadband speeds? The answer to that might be provided by posts like this
The crucial question is whether any of those have retained or discarded net neutrality. My belief is that they've all retained it but I can't find sources to sustain that belief. If my conjecture is valid, why would anyone anywhere be arguing to discard neutrality? Why would even the greediest American capitalists not seek to emulate the success of their Asian exemplars?
Or am I missing something?
It very definitely IS a serious question. My own attempt at a serious answer is here but in short, mind reading technology can only be prevented from being a totalitarian wet dream if we force governments to accept some of the protections it also makes possible (principally the ability to block authentication when "coercion" is detected) so that no one can be forced to disclose anything without their informed consent. Please see also my comments here
Without any effort on my part, I spotted no less than 3 stories in today's Reg postings which are related to this issue and all require the same general answer*: a solution to Accountability Theatre
In short, what data they store and access is the wrong (or, at least, secondary) target. There are major issues with those criteria (how securely they are stored, by whom, where, why etc) but the issues that are not being discussed are:
1) How do we know data is accurate, complete and unmodified?
2) How do we know who accessed it, when and why?
The only rational and provable answer is that all relevant data must be protected on an immutable audit trail and that access must be made technically impossible without going through a logging procedure which captures (at least) the credentials, including biometrics (preferably photograph) of the person requesting access, together with the reason for access, every keystroke or mouse movement made during the session and their login/logout times; all of which is itself protected by an immutable audit trail.
The principle role of the Law in this context, is to mandate those technical measures and to render illegal the storage of or access to any sensitive data by any other means. It also needs to empower a body, genuinely trusted by the public and independent of government and/or those holding the data, to access, for the purpose of oversight and audit, (without constraint, other than the need to maintain confidentiality where it is in the public interest) whenever they or we have legitimate concerns, any and all relevant records as well as the immutable audit trail.
*The other relevant stories are
That's as plausible a strategy as "Win Game"
I did give fleeting thought to starting up a petition along the lines of criminalising uninformed authoritarian comment on matters they know nothing about but that eliminates virtually all political discussion, which, while desirable, is even more unrealistic.
Public education is, in my view, the only realistic way to defeat the bastards in the long term. It does not require that every voter understands the fundamental ethics, let alone the fundamental mechanics of secure communications. All it requires is moral comprehension by a significant minority, say 20% or so, of the implications of criminalising secure communications. That's enough to ensure, when the relevant test cases come before a jury, that the case is dismissed with the same finality as we've seen (in the UK) with certain infamous attempts to use the Official Secrets Act. (eg Peter Wright)
This could work in the UK and Commonwealth countries which use the UK legal model. Not much use in those European countries which don't use juries and not much use in the USA, where jury-rigging is standard, but we can't let the perfect be the enemy of the good.
As to how we educate the masses, I think we need to start with the lowest common denominator - the Daily Mail - and persuade an appropriate hack to write the story from the angle that those nasty civil servants are trying to curtail their liberty. Writing the more balanced and rational version for the broadsheets would be relatively trivial as half of them are already on side.
(if you don't already know how, download WinAeroTweaker and look under the Network Options)
From then on, Windows may inform you that updates exist but that it won't download them until you're NOT on a metered connection. When you're ready (after running wushowhide.diagcab and "hiding" any updates you don't want, like I mentioned last week) you do NOT have to unset the metered connection. Just press the download button and it will proceed as normal.
And I've just checked on a W10 Home machine which stupidly allowed itself to update to the anniversary edition, and that hack still works
unfortunately I don't have that option. My principle programming language is still Visual Foxpro and that only co-operates with windoze
apologies. Memory failure on my part.
the real name is at the end of this link!
The privacy problem is closely related to the loss of control over Updates. The fixes for one are useful for the other. All those mentioned below are free of charge.
You can take reasonable control with a combination of Spybot Antibeacon (as well as "Immunise" on the first tab, remember to select all the optional telemetry blocks on the second tab) and Winaerotweaker, which will let you do such useful things as setting your ethernet connection to "metred" which stops Windoze updates in its tracks (because they fear class actions caused by forcing users to download GB on $/gb connections). You can also use it to disable many of the auto updates and rebooting after update.
In the Pro or Enterprise versions you can also use gpedit to force W-update into "Notify Only" mode, but that won't prevent "Security" updates.
However, be aware that MS is writing its own countermeasures to these countermeasures. For example, many of the IP addresses blocked by Spybot AntiBeacon have now been hard coded around by subsequent updates.
Finally download Wushowdiag.cab, which MS were forced to release, I believe, as a consequence of another court case resulting from an update borking one or more users systems. It is presented as a "troubleshooter" but what it really does is allow you to preview all outstanding updates and select those you don't want. Those "hidden" updates will then be ignored when you choose to permit an update.
For reasons adequately spelt out in the article, proving the truth or falsity of a given claim is almost impossible unless its a reference to empirical data already in the public domain.
The fall back position is to assess the consensus around the issue. Even that has major issues (eg, the consensus regarding the Theory of Evolution among Creationists is somewhat adrift of the Consensus among Scientists) but it wouldn't require rocket science to narrow the searches for consensus to "widely trusted sources".
The first problem for the software to solve would be the categorisation of the claims being reviewed. Once categorised, they could limit their searches for consensus to those sources "agreed" to be relevant to the categories.
Version 1 might be a simple summary of the arguments and conclusions found in those sources which seem to be relevant to the claim under review. Version 27.1 will inform the user not just of the summary arguments and conclusions, but make them aware of "trusted" disputation (again from "reputable" sources) and also cross reference anonymous tags (pre-shared among friends and colleagues) from those who indicate they approve the review and those who disapprove it. The client would then present a Review Summary along the following lines:
The Claim that The Theory of Evolution is an adequate explanation for biological diversity and speciation (etc) achieves a wide consensus among 99.3% of sources trusted within the relevant field. It achieves 63% consensus among Socially Trusted sources.
Of your own contacts who have registered an opinion, 98% accept the consensus. 2% reject it.
The Theory is disputed by a significant minority of the population who favour a religiously based explanation widely referred to as "Creationism" or "Creation Science"
Sources: Link 1
LInk 2 etc
This approach is pretty objective and doesn't confuse the issue by trying to define truth, merely summarising global opinion and leaving it up to the reader to decide where there own loyalty lies.
The multiple use of quotation marks highlights the desperate need for what I call "Trust Anchors". I'm working on that. More later...
Surely the key difference is the obligation Wiki places - and polices - on authors to provide sources. Anyone doing serious research might find a good or bad summary of the current state of knowledge in a Wikipedia article but that's only ever a starting point. You then go to the sources and make your own mind up about their credibility and the overall credibility of the article. If you are not impressed, and sufficiently motivated, you can then edit the article to try to correct any errors you think you've identified. And you'll be required to post your own sources (or criticised for not)
Those features alone place them in a different universe to the Daily Wail. The only time they ever publish sources is when required to by the source and if you've ever tried to correct one of their egregious errors you'll be aware of how futile that effort can be.
Be interesting to see if I get a similar bunch of down-votes for supporting your criticism. I posted a detailed critique of Power of Nightmare back in 2005, before things really kicked off with ISIS etc. Almost everything I've seen since has reinforced my then opinions.
Curtis is damn good at presenting novel points of view. His problem, I think, is that he doesn't challenge his own views strongly enough before presenting them as though they were solid conclusions. Frankly, too often, he comes across as someone who formulated an opinion, then went looking for ways to justify it.
That said, he's always educational and often entertaining. I downloaded Hypernormalisation last night (6 GB ferchrissake) and so far I''ve only watched the first 10 mins, to see if I'd want to watch the rest.
and, to any US based readers: If you're prepared to organise a Class Action against the bastards, I'm hereby pledging the first $100 towards the fund...
1) for those who have already applied the tweaks, do they remain tweaked?
2) can we use wushowdiag.cab to block the
(and if so, does anyone know what KB number to look out for. Sergey Tkachenko's article does tell us it's build 14393 and Martin Brinkmann refers to Build 1607 so that's a clue)
3) if all else fails, will host file ip blocking still work?
which is not quite as bad as no encryption at all but means that any skype conversation is accessible to the American TLAs. (A bit like Blackberry BIS [encrypted with their keys] v Blackberry BES [encrypted with your keys])
I went looking for Skype End to End encryption the day after Microsoft bought Skype and was told, explicitly (though I cannot now retrace the source) that it wasn't possible because of the way Skype works (routed through a central server). So unless something has dramatically changed (which would be a major step in exactly the opposite direction of where these things are going) you should not be trusting Skype encryption for anything more serious than keeping script kiddies out of your hair...
for what happens when you insist on a crypto "back door"...
Nobody would contest the desirability of knowing exactly what was in the killers' head and history. Preferably before they get to commit their crime.
And it is not just conceivable but likely that within 10-20 years, we will have technology capable of ferreting that information out of any head.
Once that is possible, it will be plausible to argue that, for example, airlines should be allowed to put every passenger through such a mind scanner, in order to ensure that no-one with evil intent against the aircraft is permitted to board.
Society is divided into two groups. The authoritarians and their followers form one group and they will argue in favour of allowing the mind-scanners and insisting that we all step through them,
Once they've conceded that for something as serious as air travel, it will be only a matter of time before they concede it for (in roughly descending order) weeding out Pedophiles, Rapists, Tax dodgers, Trolls, and Dissidents
Those who understand Liberty and the nature of threats like the above will probably have to fight the authoritarians literally to the death in what may come to be known as Humanity's Final War.
The current Apple battle is an early skirmish in that war.
Pick your sides now and be sure of a good seat...
...to make it illegal for politicians (or anyone else in a potential policy making position) to make proposals regarding Security unless they can first demonstrate a clear understanding of the deep interdependence between Privacy and Security.
I'll leave the finer details as an exercise for the class...
in the end, the paint dries...
and that ain't the Judean People's front, or any religion, or any non-religion.
The common enemy is Authoritarianism; which is NOT the same as "Authority".
Authoritarianism is the dangerous psychological condition which afflicts many, but not all, of those who acquire Authority.
Authoritarianism is the belief that you are "right" or "have rights" BECAUSE you have authority. It is easily the most dangerous and destructive belief common amongst human beings.
Be all of which as it may, and, at the risk of what may look like self-promotion, can I implore those of you who have shown genuine passion about this issue to take a gander at my fictional "History of Digital Telepathy" which is a "review" written about half a century in our current future, looking back on what I obviously would love to see happen.
It touches on most the major issues raised in this discussion and I would welcome feedback from what constitutes a more informed audience than I usually get.
Bernard, I'm obviously on your side, against the schill/troll/authoritarian follower or whatever MB may be. But your example is illustrative of the problem we have in spelling out the harm. In short, that story is an example of abuse of process but, in the end, the actual harm will probably be limited to a bit of intimidation which, in this instance, will fail. One or two cops will be disciplined (but almost certainly keep their jobs). No great threat to the man in the street (as they see it).
As a privacy fundamentalist, I have tried, for years, to construct an argument that makes the "if you've nothing to hide" brigade wake up and smell the coffee. The reason it's almost impossible is that all the potential harms remain exactly that - "potential" - for the vast majority of the herd. How many Wildebeest refuse to cross the river, on the annual migration, despite the fact that, last year, about 2 dozen of their number were eaten by crocodiles?
Very few citizens get to see their neighbours and friends victimised by the bully state. And in the tiny communities where the victimisation reaches more visible levels, a relatively small amount of targeting is enough to intimidate the rest of them back into toeing the line.
The real and lasting harm is to the Social Psyche, which learns to internalise the new restrictions and repressions and when to kowtow to authority. This is particularly obvious in the USA, where the level of real harm caused to citizens by the State is vastly greater than any other "first world" nation but still hasn't created the kind of backlash which is required to make the bastards back down.
That "successful" model of naked authoritarianism is encouraging all the other "civilised" nations to tread, more or less warily, down that same path. But until we start seeing more of our own innocent citizens gunned down in the street by steroid pumped uniformed bullies, (which I genuinely do not think will happen in most of Europe) the average Joe is going to remain content with the "soft authoritarianism" we have adopted and consider it a fair price to pay for our perceived security in the street.
And if we had a couple of Paris style attacks in the UK, my guess is that the populace would bend over and allow themselves to be further butt-fucked for the sake of "keeping us safe"
they have forced me into that position by becoming my enemy. Naturally, the arrangement is reciprocal. I suspect this conclusion will be shared by most other UK based Reg readers. Others have their own reasons for opposing their own states.
In short, I'd much rather face the miniscule risks of a terrorist attack against me and mine, than the colossal risks they're about to impose on us, without any realistic opportunity for us to opt out.
why do all those hostile to or ignorant of the terms of the Transhumanist debate always ask the wrong question: "If you could live forever, would you want to?"
Transhumanism isn't about immortality, it's about Omortality (optional mortality) i.e. living for as long as you still think it's a good idea.
With David Anderson's report, we finally look like we may be moving in the right direction.
However, prior authorisation by his proposed new judicial body, while useful and often necessary, is certainly not sufficient and occasionally impractical.
What is absolutely vital is complete and routine data-capture (to an immutable audit trail) of the entire surveillance decision-making process and subsequent implementation. This will allow us to rewind whatever they did, after the event, to see whether what they did was necessary and proportionate. Personally I would prefer that data to be available on demand to what we might call a Public Auditing Jury. This would render the process democratic, but I would accept his new judicial body as an interim compromise.
The most important point is that it's usually not what they do or who authorises an operation which matters. There will always be occasions in the field where actions are necessary prior to the possibility of consultation and authorisation. What matters is that everything they do is recorded, so we can review what was done in our name. Any material activity not found to be recorded on the Audit Trail would be an automatic criminal offence, as would any attempt to prevent access to the audit trail by the oversight body (be that Jury or Judges)
Hilarious. If we didn't have 10/10 cloud cover I bet we'd see the red glow of humiliated embarrassment from this side of the atlantic.
Remind me. Why would CENTCOM have a Twitter feed? Would there be any connection with PR motives?
Beautiful. Makes 2015 already a good year...
Well, good, if we can pretend, for a while, that we haven't already lost over 2,500 souls to the sociopathic kilers who sponsored this attack...
assuming the facts are as reported, any half decent prosecutor could make a strong case in the European Courts (probably also the UK Courts) that what the government has done amounts to "abuse of process" and would thus nullify their attempt at bypassing the EU legislation. If there are any concerned copyright holders with sufficiently deep pockets, you might want to consider funding such a prosecution...
Thanks David, I'll give it a try...
Fully agree. And your justifiable rant reminds me of the question I've been meaning to put to anyone more experienced with the windoze environment than I am.
Is it not possible to create either a registry setting or registry checker or firewall setting (etc etc) which simply and automatically blocks any attempt at installing not just Chrome but ANYTHING which we don't explicitly (and consciously - for example by being required to enter a randomly generated PIN rather than just clicking a button or pressing Enter) permit? AND, having said NO, will never permit any future attempt to ask us about the same app again unless we go in and re-enable the question for that particular application.
I'm very familiar with many of the registry change blocking shields (on my systems, ZoneAlarm blocks them, Avast blocks them and I've got Regwatcher alerting me to any changes) and even with those aids and few decades of windoze experience, some still get through. So I've acquired considerable expertise in removing the bastards, but prevention would be so much better than cure and protecting my clients, friends and family from such intrusions is virtually impossible.
Your post is extremely apposite and well re-engineered.
I would guess your downvotes result from ignorance and failure to recognise your parody of the famous translation of Martin Niemoller's 1946 poem which, for the benefit of those who've obviously never encountered it, I include here:
First they came for the Jews
and I did not speak out
because I was not a Jew.
Then they came for the Communists
and I did not speak out
because I was not a Communist.
Then they came for the trade unionists
and I did not speak out
because I was not a trade unionist.
Then they came for me
and there was no one left
to speak out for me.
Biting the hand that feeds IT © 1998–2017