Our somewhat lengthy comments on Part 1 of the Home Office paper to the EU Parliament - "Liberty and Security - Striking the Right Balance" -
Retention of telecommunications traffic data:
Right from the start, this section of the paper attempts to blur and confuse exactly what "Data Retention" is, presumably in the hope that the Members of the European Parliament and the public will fail to understand the difference between legal Communications Data Traffic Analysis, of say, mobile phone itemised bills, which are currently available to the authorities, and the proposals to retain such data well beyond the time that they would normally have been deleted or anonymised, in accordance with the necessary principles of Data Protection, which are the law in most European Union countries.
As the EDRI-gram analysis of this section of the Paper points out:
"While Watkin promised the indignant MEPs that the UK Presidency would present a major new study into the effectiveness and viability of the measures the next day, the disappointment was enormous when the UK presidency paper was finally released. In stead of scientific research into the need and benefits of data retention in Europe, as urgently requested by the LIBE members, the paper is a thin narrative with some emotional examples of the use of traffic data in police investigations."
In other words, a typical NuLabour Home Office policy document, an embarrassment to the UK public in front of our European neighbours.
"The study mentions 4 examples of the use of traffic data, all of which seem to fall within a 3 months framework. All these data should thus have been available by the telephony companies for their own purposes irrespective of any law on data retention."
The statement on the retention of Internet Data are surprising, and it contradicts ome of the examples given.
Where is the quantitative research on how long serious criminals or terrorists keep using their old mobile phones / old mobile phone numbers ?
There is no estimate of the huge costs that would be required to securely access and search the huge amount of retained data, which seems to be planned. This is not free, and it easily costs more than the raw data storage.
Retention of telecommunications traffic dataSummary
- Retained communication data has been crucial in unravelling terrorist networks and solving serious crimes.
- That will remain the case because terrorists and serious criminals have little choice but to communicate by phone or internet.
- In the UK, over half of all data requirements in terrorist investigations are for data over six months old.
- Only data generated or processed for business purposes is retained, so the extra cost for business is associated solely with keeping data for longer than it otherwise would be.
- The UK experience is that these costs are modest: £875 000 for one major mobile network.
- In the case of the internet, the requirement is to retain information about log-ins and log-outs alone.
Mobile phones and e-mails have a central role in the daily lives of people in the EU. For many of us they are indispensable tools in life. Sadly, the same is true for criminals. However, the technology that makes it easier for them to carry out their criminality can also be their downfall. The information that is automatically generated when they use their phones, mobiles and the internet can be as useful to those investigating crime as the physical DNA or fingerprints that can be left at the crime scene. Of course, it can also be used to corroborate alibis. In the UK it is an essential tool for investigating serious crime and is being used to find those behind the bomb attacks in London.
What is it?
Communications traffic data is the information that is generated automatically when their services are used; including information about where on a network a communication originates or terminates, the devices through which the communication is made and received and about the time the communication is made.
The service providers need this information themselves for legitimate business reasons enabling communications and managing their networks, for fraud detection and revenue collection. They currently keep differing sorts of data for different lengths of time. For example, information required simply for enabling the communication is not required by the provider immediately; data about network management is needed a little longer; and for fraud detection and revenue collection longer still. The periods of time that the information is kept for these purposes varies from provider to provider as they are all currently deleted as soon as they have outgrown that provider’s business need. An example of the data collected is attached to this paper.
How is it used?
Communications data can assist an enquiry by providing a link between people, times and places which may lead to the identification of witnesses, forensic opportunities or the criminal’s financial assets. Drug and people traffickers conduct much of their business through the use of communication services and there are instances when mobile telephones have been used to detonate bombs. The ability to trace to whom a phone is registered, what calls were made, when and where they were made, whether answered or unanswered, means investigators can establish or refute certain events leading to a crime and the individuals associated with it. Terrorists and many serious criminals have already been brought to justice in the UK and other countries with the help of retained data.
Over half of the 30 million or so mobile phones in the UK use pre-paid SIM cards, and most of these are either not registered, or are registered in the wrong person's name , perfectly legitimately e.g. when they are given to other family members as presents.
In Sweden, a bomb threat was made to police by e-mail that a bomb had been placed at Stockholm Central Station. Using logs of the allocation of IP addresses retained by an Internet Service Provider, the source of the e-mail was traced to a public library in Stockholm. The library staff was able to provide information to the police that enabled the identification of the offender.
Did the Swedish Police really wait 12 months before investigating a bomb threat against their own Police HQ the largest Railway Station in the country ? It is likely that their investigation used current business data, not Retained Data.
According to the paragraph on Internet Data in this paper, the proposal would actually prevent this form of analysis of emails and of IP address data for Retained Data !
Why do we need a European model?Most people acknowledge that in our globalised world cross-border crime is more prevalent than before. All of us agree that greater international co-operation is needed to tackle the problem.
However, as explained above, even within countries different service providers retain data for different lengths of time and so the ability of law enforcement to investigate serious crime is determined by the business practices of the particular service provider that a suspect, victim or witness of crime happens to use or have used.
And with different practices across the EU this element of chance is magnified even further. We are still in the relatively early stages of the investigation into the London bombings but it is clear already that there are international links that may lead to those who encouraged and supported the attacks. Fortunately Italy retains this information for as long as the UK does but should we really have to rely on chance?
The tracking of the terrorist suspect accused of being one of the July 21st failed bomb attackers through Italy to Rome, is an example of real time or near real time, targetted Communications Data analysis.
This is not evidence for Data Retention of every innocent persons data for 12 months over and above the period of time which the Mobile Phone Companies keep their logs files for normal business purposes.
In at least one case in the UK, the ability of the police and intelligence agencies to identify a terrorist network on the basis of an initial lead has depended on access to retained telecommunications data which revealed links between individuals otherwise invisible to investigators.
How many of these links were revealed from the current business data, and how many were revealed by analysis of 12 month or older retained data ?
This appears to be another deliberate confusion of "Data Retention" with "Communications Data Reaffic Analysis of data held for current business purposes".
The draft Framework Decision would provide a legal basis for retention of specified data for the purpose of investigation, detection and prosecution of crime and terrorism. It would provide clarity for the telecommunications providers, law enforcement and safeguards for the public.
It absolutely does not provide any clarity !
This Home Office paper only adds to the confusion.
What safeguards for the public ?
- 10 years in prison for the misuse of retained communications data ?
- The forced resignation of the politiicans nominally in charge of the organisations who allow such abuse to occur ?
- Financial compensation the for unecessary invasions of privacy of individuals ?
If there were safeguards along those lines, then we might be convinced, but there are none, so we are still sceptical.
In the UK a suspect was eliminated from a murder investigation with unsuccessful call data. He gave evidence that he did not know the victim was dead and had, in fact been, calling her all day. His mobile phone call records showed no calls to the victim (because no calls to her had incurred a charge). The mobile phone company did not keep a record of connected-unanswered calls.However because calls from a mobile to a landline will pass via the cheapest route, they can be present on another provider’s interconnectivity records, which incur a charge (from one provider to another) and thus a record is generated. Communications data evidence showed 27 connected-unanswered calls between the suspects’s mobile and the victim’s landline, corroborating the man's explanation and eliminating him from enquiries.
What a peculiar example ! If the only record of the mobile phone calls was through the interconnect billing record, rhen this does not have any Mobile Phone Cell ID Location Data in it either.
How does this establish an alibi ? The suspect could have been using his mobile phone from right beside the murder victim !
None of the previous, more detailed "shopping lists" for telecommunications Communications Data Retention have included interconnectivity records between telecomms providers.
Presumably this "alibi" was investigated using the existing records kept as part of the normal business procedures.
How many times has "Communications Data" been used to corroborate an alibi, rather than to put someone under suspicion ?
Again, this is not evidence for Data Retention of 12 months over and above the period of time which the Phone Companies keep their logs files for normal business purposes.
ProportionalityIn the UK telecommunications data is used to investigate serious crimes. IT is also only used proportionally. Law-enforcement targets their requests for access to data as required by the investigation. For example initially they may only ask for data collected on the day or in the vicinity of a murder. In the case of a terrorist attack they will seek a larger amount of data for a bigger area. In each case the request must be proportionate to the investigation. This is however only possible if the data is retained by the service provider.
Where are the safeguards to prevent "fishing expeditions" or "trawling of databases" via "data warehousing" technology ?
Why 12 months?
Presumably because the Home Office has failed over and over again to make out any believable case for a longer period of retention ?
Concerns have been expressed that retaining data for 12 months or longer is excessive. But data is often needed that is over 6 months old. Examples from Ireland make clear that data that is significantly older is sometimes used. A recent study of the requirements for disclosure of data made by the police in the UK established that the majority of data required (85%) was less than six months old. However, where data between 7 and 12 months old was required, it was used to investigate the most serious crimes, mostly murder.
These statistics are of no value, unless the Home Office also publishes how many of the "requirements for disclosure of data"
actually lead to major lines of inquiry, and were not simply "fishing expeditions" or "clutching at straws" guesses.
That is for two reasons: firstly, the nature of the crimes may mean that the culprits do more to conceal their tracks so a crime (and therefore a suspect) may not come to light for weeks or months – for example where a dead body is only found months after a murder has taken place. In such circumstances investigators need to be able to look back in time to establish with who that victim may have been in contact at the time of death.
The attempted or sometimes successful, concealment of the very fact that a crime has been committed, is something which has goimg on since the dawn of history.
Mandatory Retention of Communications Traffic Data will do nothing to prevent that.
Secondly, in such crimes there will always be an expectation that such investigations will exhaust every line of enquiry and will run for longer, and that the acquisition of older data will be proportionate to the aim of detecting such offences.
Really ? So no stone remains unturned when it comes to invesigating old or even ancient crimes ? These investigations are allocated more resources than current crimes ?
Perhaps in some distopian police state, but not where we the taxpaying public have to fund such investigations, they are not.
In 2002 an aggravated burglary took place in a UK city. Over the following months similar offences occurred across the same city. In each instance the victims were elderly single people or couples, who were tied up, assaulted and robbed of cash, valuables and credit cards. As the series of offences progressed the investigators wanted to determine if any mobile phone could be identified as being in the vicinity of more than one of the offences. This line of enquiry was rejected three times by the senior investigating officer as disproportionate to the offences then under investigation.The victim of the ninth burglary died three months after being assaulted. By the time of her death at least fourteen offences were linked together and an offence of manslaughter was being investigated. This made the investigation of mobile phone location data and the acquisition of private communications data more proportionate to the aim of detecting the offenders.
Mobile phone data was obtained which proved links between the offences. Because data from the time of the earliest burglaries had been deleted evidence of the full extent of the criminality was lost. Several arrests have been made.
This is not a valid argument for Data Retention !
Most people will be puzzled why the first violent "aggravated burglary" was not treated as a Serious Crime, i.e. one defined under the Regulation of Investigatory Powers Act 2000,
The Regulation of Investigatory Powers Act 2001 section 81 (3) General Interpretation appears to already allow both Communications Data and actual Phone Intereptions to be applied for where crimes of violence are concerned.
- "(3) Those tests are-
(a) that the offence or one of the offences that is or would be constituted by the conduct is an offence for which a person who has attained the age of twenty-one and has no previous convictions could reasonably be expected to be sentenced to imprisonment for a term of three years or more;
(b) that the conduct involves the use of violence, results in substantial financial gain or is conduct by a large number of persons in pursuit of a common purpose."
Why isn’t intercepting their telecommunications sufficient?Interception of telecommunications plays a very important part in tackling crime in the United Kingdom. However, it relies on suspicion in advance of a criminal act. It is entirely possible, with all the resources the UK has put in to tackling terrorism, for people who are completely unknown to the authorities to commit the most heinous acts. In other crimes there may be very little pre-meditation before the crime.
Yes, and how exactly would Data Retention be of any use against terrorists who are "completely unknown" ?
Note how the Home Office paper glosses over the peculiar situation in the UK where "interception of communications" is forbidden by law, from being used as evidence in court, it can only be used for "intelligence" - a situtation which astonishes most of the other European Union countries.
Won’t they use other means of communicating?In the future some criminals and terrorists will adapt their use of technology to make the retention of this data a less important tool for investigators. However, with the current level of communications technology, this data is a necessary part of many investigations as it is difficult for terrorists and criminals to communicate without using the telephone or internet.
Rubbish ! Criminals and terrorists have always communicated face to face, and will continue to do so in the future.
CostsExperience in the UK and Ireland is that the costs associated with the retention of communications data are not excessive. For example, in its biggest project to date the UK Government is working with a national mobile phone network which represents a substantial share of the UK market and which presently retains all of its traffic data for two days. This data has a high degree of detail which can be valuable to investigators and includes outgoing and incoming answered and unanswered calls together with details of the geographical location of mobile equipment. After two days some of the data is retained for billing purposes and some is normally destroyed.
Following discussion with the UK Government the mobile phone network is proposing to retain all the data for 12 months. The total cost of doing so and providing a tool to retrieve specific data is £875,000 (€1.2m) and the UK Government is funding this project.
We must be told which UK Mobile phone Network is proposing to contravene the Data Protection Act 1998 in this way, by not obtaining the informed prior consent of each of its individual customers, to retain and to process personal data which is no longer required for business purposes.
We shall be asking the Home Office, and if necessary the Information Commissioner, OfComm the telecomms industry Regulator and the individual Mobile Phone Networks, if necessary.
Whichever of the Mobile Phone networks it is, they will by virtue of this scheme with the Home Office, be acting as Public Authorities and will therefore come under the auspices of the Freedom of Information Act 2000.
Note that the "cost" figure appears to be simply for the cost of the data storage.
The true cost of accessng such data securely and proportionately, and with a proper audit trail and checking of authorisations, will be many times this.
A Ghanaian national whose family owned what was described as a 'gold-field' came to London to sell a sample of gold. His contacts decided not to buy. The Ghanaian then went to the Netherlands for the same purpose. He was kidnapped, apparently at Schipol Airport, and a ransom demand was made to his family in Ghana, in the sum of £500,000.The family reported the ransom to police in Ghana who contacted the UK police. The UK and Dutch authorities co-operated with the investigation. After four days there had been no contact from the hostage for more than 24 hours and the UK police were contacted to help identify the hostage’s London contact. With the hostage’s life at risk it was proportionate to seek relevant traffic data and the UK police identified a high frequency of calls from a hotel payphone to a Belgian mobile and to the family of the hostage in Ghana.
The Belgian mobile belonged to those guarding the hostage. Enquiries led the Belgian authorities to the place where the hostage was being held. He was rescued, having suffered severe torture. Arrests were made in Liege and in London.
How would retaining data for a year have helped in this case ?
Surely all the Communications Traffic Data Analysis took place on current business records, at the very most a week old ?
This is another example of, presumably deliberately, confusing "Data Retention" with "Communications Data Traffic Analysis of Current business data records".
For such a significant source of evidence these costs are not considered excessive for government or law-enforcement when set against other costs involved in pursuing criminals or terrorists.In the UK our police can and do pay for communications data. In a typical murder case they may spend approx £50,000 (€72 400) on communications data, this rises to approx £500,000 (€724 000) for a terrorist investigation. This is a relatively small amount of their law enforcement budget and is not expensive when compared to other costs such as forensics: in a murder case forensics can exceed £500,000 (€724 000) and to forensically examine one cigarette end will cost the police approx £800 (€1158).
Internet data
We recognise the difficulty of retaining large amounts of internet traffic data and it is not the aim of the Framework Decision to require industry to retain this data. The Framework Decision only requires industry to retain information on when and where an individual logs-on and logs off the internet which will require substantially fewer resources.
On the face of it, this is an astonishing climbdown by the UK Home Office.
The misguided European Union plans for Data retention talked of 3 years retention with no distinctions being made between different types of internet data protocols and log files.
The UK's own Voluntary Code of Practice Appendix A (,pdf) under Part 11 of the Anti-terrorism, Crime and Security Act 2001:
- "EMAIL DATA 6 months
e.g. Log-on (authentication user name, date and time of log-in/log-off, IP address
logged-in from)
Sent email (authentication user name, from/to/cc email addresses, date and time
sent)
Received email (authentication user name, from/to email addresses, date and
time received)
ISP DATA 6 months
e.g. Log-on (authentication user name, date and time of log-in/log-off, IP address
assigned)
Dial-up: CLI and number dialled
Always-on: ADSL end point/MAC address (If available)
WEB ACTIVITY LOGS 4 days
e.g. Proxy server logs (date/time, IP address used, URL’s visited, services)
The data types here will be restricted solely to Communications Data and exclude content of communication. This will mean that storage under this code can only take place to the level of www.homeoffice.gov.uk/……."
Can it really be true that Internet usage, using many and various data protocols, will be virtually exempt from this Data Retention scheme ?
It appears to contradict the "Swedish bomb threat by email and IP address" example given above, in the unlikely event that the Swedish Police waited for several months to try to track down a bomb threat against their own Police HQ.
What use is "when and where an individual logs-on and logs off the internet" when so much of the Internet activity in the UK now involves "always on" broadband services, and via GPRS and 3GPP mobile phones, WiFi wireless local/wide area networks etc. ?
Does the sneaky wording including the word "individual" hide an attempt to force some sort of repressive Communist Chinese crackdown on cyber cafés and registration or licensing and identification of individual internet users ?
Such a scheme would obviously also be totally impracticable, but that is no barrier to the panopticon fantasies which seem to surround the topic of Data Retention in the corridors of the Home Office and the European Union.
Zero data callsUnconnected calls are not included within the scope of the Framework Decision. Connected but unanswered calls are included within the scope because these can be signals to accomplices or used as a way of detonating explosives. The data call record attached shows some unconnected calls.
If a mobile phone is used as a bomb detonator, rather than just as a source of a cheap electonic timing device, as in the Madrid bombings, then the detonation phone call could come from anywhere in the world, i.e. outside the European Union
Surely any bomb explosion will be investigated on the day of the explosion and not 12 months or more later ?
How will mandatory Data Retention of the personal data of the European Union's 450 million people's help at all ?
Completely innocent "zero data calls" happen all the time.
How are investigators meant to magically distinguish between a "subliminal channel" "wait for 3 rings and then start the next phase of the crime" type message, and a normal mobile phone "uncompleted call" caused by network traffic congestion etc ?
They would need access to, and the analytical skills to interpret even temporary local problems at every single Mobile Phone Cell base Station transmitter, and fixed line telephone exchange and to retain a history of all of these for the data retention period.
Various telecomms industry bodies have expressed fears over the costs of implementing this "zero data call" requirement.
Data ProtectionThe UK understands the data protection concerns around this issue and there is absolutely no doubt that when data is retained by a service provider it must be stored securely.
Under the Framework Decision neither the police nor any public authority will have unrestricted access to the retained data. This will be governed by national law. Of course, they will only be permitted access for the reasons set out Framework Decision in the Directive, namely the investigation, detection and prosecution of crime.
Rules on access are regulated at a national level and police and other authorities will therefore have to meet the national standards to access private information.
The Presidency agrees that further clarity on the data protection rules relating to the third pillar would be helpful and we look forward to the Commission producing a proposal on data protection in the third pillar later this year.
What safeguards are there to prevent a Police or intelligence agency or even a private sector consultancy or subscontractoe from regularly submitting repeated piecemeal requests, which result in a cloned copy of the Retained Data, which is not destroyed or anonymised after even the 12 month Data Retention period ?
What is there to prevent such data being transferred to another European Union country or even outside of the EU, e.g. the USA or Russia, where our Data Protection laws do not apply ? Nothing, and it is increasingly likely thatthis will happen under the various "anti-terrorism" data sharing plans that Home Secretary Charles Clarke was alluding to in his Introduction to this paper and his speech during which he stressed some of the points in this paper.
[Simplified fake example of a Call Data Record omitted for brevity]
We note, as do other people who have actually worked with mobile phone data, such as the fourth place blog, this alleged example of a Call Data Record seems to be rather sparse, compared with real ones.
Data retention inconsistancies"Slugging through this data retention proposal I see an example of a call data record with these fields:
Date & Time Of Call
Calling MSISDN
Called MSISDN
Video Call flag
SMS event flag
Record Type (e.g. mobile originated, unanswered etc)
Duration In SecondsSuffice to say that this portrayal of a CDR is very much simpilified. The seven fields given here are actually more like 30 to 50 fields. I notice there is a video call flag, so this is could be the "3" network. I also notice that there is no cell ID field in the example"
Compare and contrast this with the earlier United Kingdom Home Office proposal for voluntary data retention here in the UK:
Voluntary Code of Practice Appendix A (,pdf) under Part 11 of the Anti-terrorism, Crime and Security Act 2001
"APPENDIX A Data retention: expansion of data categoriesSUBSCRIBER INFORMATION 12 months
(From end of subscription/last change)
Subscriber details relating to the person
e.g. Name, date of birth, installation and billing address, payment methods,
account/credit card detailsContact information (information held about the subscriber but not verified
by the CSP) e.g. Telephone number, email
address Identity of services subscribed to (information determined by the
communication service provider)
e.g. Customer reference/account number,list of services subscribed to
Telephony: telephone number(s), IMEI, IMSI(s)
Email: email address(es), IP at registration
Instant messaging: Internet Message Handle, IP at registration
ISP - dial-in: Log-in, CLI at registration (if kept)
ISP - always-on: Unique identifiers, MAC address (if kept), ADSL end points,
IP tunnel addressTELEPHONY DATA 12 months
e.g. All numbers (or other identifiers e.g. name@bt) associated with call (e.g.
physical/presentational/network assigned CLI, DNI, IMSI, IMEI, exchange/divert
numbers)Date and time of start of call
Duration of call/date and time of end of call
Type of call (if available)
Location data at start and/or end of call, in form of lat/long reference.
Cell site data from time cell ceases to be used.
IMSI/MSISDN/IMEI mappings.
For GPRS & 3G, date and time of connection, IMSI, IP address assigned.
Mobile data exchanged with foreign operators; IMSI & MSISDN, sets of GSM
triples, sets of 3G quintuples, global titles of equipment communicating with or about the subscriber.SMS, EMS and MMS DATA 6 months
e.g. Calling number, IMEICalled number, IMEI
Date and time of sending
Delivery receipt - if available
Location data when messages sent and received, in form of lat/long reference."
Some of this, especially the sets of GSM
triples, sets of 3G quintuples i.e. the cryptographic keys which also protect the confidentiality of the conversations or data messages over the air, goes well beyond the equivalent landline telephone itemised billing.
Are you really convinced by this paper, through which the UK Home Office seeks to influence the European Parliament into accepting its controversial plans for Data Retention ?
If not, the you may wish to sign the online Data Retention is No Solution online petition, urging the European Parliament to reject these illegal, invasive, disproportionate and ineffective European Union wide Mandatory Data Retention plans.
Stockholm Central Station is accually the central RAILWAY station, not a police station.
@ Marcus - thanks for the correction - the point about Current Business Data, rather than Retained Data still stands though, as any bomb threat to a Railway Station reported to the Police should have been dealt with as much speed and vigour as one against a a Police Station.