CCDP : What we know

After another day of confusion around the Government’s plans for increased surveillance powers, it now appears there is some back-tracking and the bill will only be a ‘draft’.

Here’s an update of some key points, while Privacy International have highlighted some serious issues with a briefing given to Liberal Democrat MPs.

Key issues:

  • The Coalition Agreement pledged: “We will end the storage of internet and email records without good reason.”
  • The plans are expected to be announced in the Queen’s speech, meaning it is almost impossible they become law before the Olympics
  • Plans to install GCHQ’s own physical equipment to give them real-time access to data appear incompatable with proper judicial oversight of surveillance
  • The new proposals may remove the ability of service providers to challenge requests for data – Google last year only complied with 63% of requests
  • It is far from clear if this is technically possible, particularly where secure communications are used and it is impossible to expose ‘header’ information without also exposing content.
  • This will cost taxpayers billions, but also mean businesses incur new costs at a time when the Government is asking them to invest in high-speed internet infrastructure
  • The fact that this is not a ‘central’ database is being used as a smokescreen to detract from Lib Dem and Conservative promises in opposition
  • The 7/7 Inquest Coroner’s report discusses the issues involved with large amounts of data and surveillance: One point states “It is unlikely these could have been detected by surveillance given the large number of untraceable “operational” phones used by the bombers and only attributed to them once their identities and details were known.”

Compare and contrast:

Jacqui Smith: “Communications data is used as important evidence in 95% of serious crime cases and in almost all security service operations since 2004.”

Theresa May: “Such data has been used in every security service terrorism investigation and 95 per cent of serious organised crime investigations over the last ten years.”

A final word from the Prime Minister, David Cameron:

“Faced with any problem, any crisis – given any excuse – Labour grasp for more information, pulling more and more people into the clutches of state data capture… And the Government doesn’t want to stop with the basic information. They want the most complex, important, personal information there is… Scare tactics to herd more disempowered citizens into the clutches of officialdom, as people surrender more and more information about their lives, giving the state more and more power over their lives. If we want to stop the state controlling us, we must confront this surveillance state.”


  1. Gary Walsh
    4th April 2012

     OK I know government is always seeing how far it can go in terms of monitoring our private communications based on ever-increasingly spurious claims and unsubstantiated concernw over security, but I firmly believe everyone has it wrong on this occasion. To me this smells of classic spin. Take an emotive subject like privacy, increase the pressure, sit back and watch it simmer. The real thorn in the government’s side at present is the NHS.

    A sparse little press release about snooping that affects absolutely everyone is perfect fodder designed to draw attention away from real on-going events that draw much controversy.

    Seems to be working too. They say ‘Jump’, the media says ‘Is this high enough?’

    All this Spin Is Making Me Dizzy or Meanwhile, Back At The NHS’ 

  2. gregorylent
    4th April 2012

    the tragedy of the kinds of minds attracted to government is too much to bear.

  3. Stephi
    4th April 2012

    I am with Gary Walsh the spin is making me dizzy. NHS far more important.

  4. Stephi
    4th April 2012

    NHS – Saturday March 17th there was a complet media blackout ready to push the bill through the following Monday. Gary Walsh you may have missed this unless you were on twitter that evening.


    This is what happened !

  5. CCDP What we know | Libertarian Home
    4th April 2012

    […] What we know April 4, 2012By Simon GibbsCCDP What we know – Big Brother […]

  6. Britse regering wil elektronisch verkeer van elke inwoner in het oog houden »
    4th April 2012

    […] al vanuit verschillende hoeken een heleboel (en terechte) kritieken moeten slikken. Onder meer Big Brother Watch en Privacy International toonden zich felle tegenstanders van de draak van een wet. In zoverre dat […]

  7. Stephi
    4th April 2012

    Spain’s SINDE Law same as SOPA

  8. Stephi
    4th April 2012


    9/11 – London 7/7 and Madrid train bombing I am convinced all three were inside jobs.

    Spain’s Azna removed ALL data of the bombings before fleeing to the States straight into the waiting arms of Rupert Murdoch where he sits to this day on the board of News International.

    Zapatero stood up to a full house in Parliament and gave them this devastating news the week he took office!

  9. Roger Thornhill
    4th April 2012

    This “draft” is just a mealy-mouthed attempt to dissolve rebellion from the LibDems.

    Look at the wording: one can drive a double decker bus through it.

    Hypocrisy on three sides. A total lack of principle and understanding of the Rule of Law.

    This new infringement is, quite precisely, UNWARRANTED.

  10. Mj Rogers
    5th April 2012

    Will the CIA have direct or indirect access to the surveillance data ? Is the data to be shared with foreign governments or their agencies ?

  11. Esta Semana (2012/14) – PT: Carros de luxo e altas velocidades; Mais duas siglas a seguir #CCDP e #CISPA; Sucesso: GNU/Linux em Munique, por cá mais do mesmo! « O Vigia
    5th April 2012

    […] history. And not just for those suspected of criminal activities; for everyone." ; CCDP : What we know – esta é a nova sigla a […]

  12. Joseph Hargrove
    10th April 2012

    The premise that digital data can be used as “reliable evidence . . .of serious crime,” poses an unprecedented threat to individual rights and freedoms.  The framers of English common law and their colonial offspring, the framers of the US Constitution, had a profound distrust for government in any and all its forms.  This distrust was based, not on some esoteric political theorising, but on a clear and objective understanding of the fallibility of human nature.

    Information technology has evolved by many orders of magnitude.  Human nature has not. 

    When those who govern *or the special interests on whose behalf they govern) urge us to “Trust the government” they are really saying “Trust us.”  And increasingly, they are saying, “Trust us to operate without regard for centuries of common law. 

    Mr. Thornhill’s comment on the Rule of Law is dead on the mark. 

    The idea that a social contract exists (or should exist) between a society and those it empowers to govern on its behalf is fundamental.  A nation’s laws are (or should be), in effect, the terms and conditions of that social contract.   Historically, at least in the UK and US, the Rule of Law has been seen as protecting a society from the predations of those who govern.  

    In the decades since WWII, that notion has been turned on its head.  Laws have now become the chosen mechanism whereby those who govern seek to impose their will on society at large.  When laws cease to protect individual rights and freedoms, in favor of protecting some abstract notion of collective security, those who govern cannot be trusted to function properly. 

    This is where the premise that digital data is reliable evidence is most perilous. 

    We investigate to eliminate uncertainty. If we had certainty about the answers we seek, we would not need to investigate.  Seeking answers implies asking questions.  And the reality is that, for most practical investigations, it mathematical probabilities that some of the answers retrieved will be wrong. 

    Technically, the problem is one of “classification.”   My database contains a plethora of “facts.”  My question defines some set of criteria whereby the system will make a positive or negative decision regarding each of the facts.

    The process will inherently produce a mixed outcome, comprised of “true positives” (objectively true answers, identified as such by the system); “false positives” (objectively false answers, erroneously identified as true by the system); “true negatives” (objectively false answers, not retrieved by the system as true); and “false negatives” (objectively true answers that the system misses.) 

    Recall that we are, by definition, uncertain about the answer (otherwise we wouldn’t need to be asking the question.  Thus, classification is inherently a statistical process.  And therein lies the rub.  The digital universe is not infinite, but for all practical purposes it is virtually unbounded.  For every “true” positive in the data set, there will be orders of magnitude more false data points that have some probability of being incorrectly identified as true. 

    The inherent behavior of classifiers is most often described by something called the “relative operating characteristic” or ROC curve. (For historical interest, this is mathematically the same as the “receiver operating characteristic” curve used since WWII to describe the performance of radars.) 

    The essence of the ROC for any classifier is best summed up as follows:  If you want to be sure that system delivers all of the true answers, the classifier can do that, if you can live with an arbitrarily high number of false positives.  Conversely I can guarantee that you will get only true answers–if you can live with an arbitrarily high probability of getting no answers at all. 

    This behavior is colloquially known as “no free lunch.”  Or, alternatively, as producing a “matrix of confusion.”

    The point of all this?   When applied to the kinds of data and behaviors involved in potential criminal activity, statistically, mathematically, information systems cannot be relied on to get it right.  

    A case in point:The Google Search that got me to this Web Site returned
    two tailored advertisements: one for a amazing array of watches, bibs,
    and magazines on e-bay, the other a link that will allow me to watch the
    TV show “Big Brother” on-line, for free.  In the case of advertising,
    the statistics work fine.  The hit rate per unit cost to deliver the add
    is astronomically favorable to the vendor; and if I’m not
    interested–no harm, no foul. 

    Translate this into a world where law enforcement relies on
    electronically-retrieved data to identify criminal activity, and where
    courts and juries have been brain-washed into believing that the data
    are “reliable evidence,” and we have a recipe for catastrophe.

    The best discussion of how classifiers behave I’ve encountered is Tom Fawcett’s paper, ROC Graphs: Notes and Practical Considerations for Researchers. I consider it a tour de force, one that finally appears to be getting the play and recognition it has long deserved.  And in this case, Googling the Author and ROC will get you there. 

    Thanks to any and all who waded through all this.

    (C) 2010 J. Hargrove–from a manuscript in progress.

    • Zeke-zero
      12th April 2012

      Fantastic explanation.  Thank you.

  13. Tonight: Privacy International pub meet on Dave’s Surveillence State | Libertarian Home
    11th April 2012

    […] in the pub. We’ll be at the Old Nick in Holborn, London from 5:30pm onwards chatting about CCDP – all are […]

  14. Peter F
    20th May 2012

    Big Brother Watch, I commend your campaign, but you really need to crank up a gear and start bringing these issues to a wider audience. I, for one, was fooled by the pledges made by Cameron and Clegg in opposition to reverse the successive Orwellian statutes brought in under New Labour. However, it appears that no party is immune to the lobbying of the police and security services on this issue – if they had it their way, they’d happily put CCTV in every home in the name of ‘fighting terrorism’.

    It is VERY WORRYING that even the Lib Dems, whom you might expect to ideologically opposed to such measures, have instead appeared to have bought into the brash assurances that the powers will be exercised under ‘strict controls’. This shows that there really is little hope on these matters.

    Please, please try to bring these concerns to a wider audience – right now, the public see opponents of these kind of plans as cranky nerds or conspiracy theorists.

  15. CCDP: Privacy Law and Monitoring the Public | tashalaw
    20th May 2012

    […] as Privacy International, the Open Rights Group, the Foundation for Information Policy Research and Big Brother Watch have organised a conference on the 19th of April at the London School of Economics in order to […]