From maryhawking at tigers.demon.co.uk Mon Dec 6 08:53:47 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Mon, 6 Dec 2010 08:53:47 -0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? Message-ID: <7E5456D53823492AA59C352DE036B5ED@MaryPC> http://www.channelregister.co.uk/2010/12/03/national_addressing_eric_pickles _ordnance_survey/ There was a good deal of discussion when PDS (the Personal Demographics Service http://www.connectingforhealth.nhs.uk/systemsandservices/demographics/pds ) was introduced in the NHS on the risks that it could pose to people with good reason to conceal their addresses and contact details, including groups such as those with abusive ex-partners or families and anyone with a connection to Huntingdon Life Sciences. There is a means of withholding the demographics which might locate you (see Access and Security): the back office still holds the details but they are not displayed. Does anyone know whether there are intended to be facilities for similar withholding of details in this new database, and do these exist in the databases (thought by commentators on the article to be highly inaccurate anyway) to be used to populate the new database, and if so, will they be carried into the new system? PDS declarations of "vulnerability" were not carried forward into the Children's Database, and AFAIAA there were no mechanisms for requesting withholding of contact details. Mary Hawking -------------- next part -------------- An HTML attachment was scrubbed... URL: From pwt at iosis.co.uk Mon Dec 6 09:33:00 2010 From: pwt at iosis.co.uk (Peter Tomlinson) Date: Mon, 06 Dec 2010 09:33:00 +0000 Subject: Is an (inaccurate) national address database a threat to privacy and sometimes personal security? In-Reply-To: <7E5456D53823492AA59C352DE036B5ED@MaryPC> References: <7E5456D53823492AA59C352DE036B5ED@MaryPC> Message-ID: <4CFCADCC.3030005@iosis.co.uk> Just as a little aside, my local NHS Health Centre has the wrong spelling for the name of the street where I live, and told me that they were not able to correct it... (Mary, I think I have correctly adjusted the title of your post.) Peter On 06/12/2010 08:53, Mary Hawking wrote: > > http://www.channelregister.co.uk/2010/12/03/national_addressing_eric_pickles_ordnance_survey/ > > There was a good deal of discussion when PDS (the Personal > Demographics Service > http://www.connectingforhealth.nhs.uk/systemsandservices/demographics/pds > ) was introduced in the NHS on the risks that it could pose to people > with good reason to conceal their addresses and contact details, > including groups such as those with abusive ex-partners or families > and anyone with a connection to Huntingdon Life Sciences. > > There is a means of withholding the demographics which might locate > you (see Access and Security): the back office still holds the details > but they are not displayed. > > Does anyone know whether there are intended to be facilities for > similar withholding of details in this new database, and do these > exist in the databases (thought by commentators on the article to be > highly inaccurate anyway) to be used to populate the new database, and > if so, will they be carried into the new system? > > PDS declarations of ?vulnerability? were not carried forward into the > Children?s Database, and AFAIAA there were no mechanisms for > requesting withholding of contact details. > > //Mary Hawking// > From ukcrypto at sourcetagged.ian.co.uk Mon Dec 6 14:52:40 2010 From: ukcrypto at sourcetagged.ian.co.uk (Ian Mason) Date: Mon, 6 Dec 2010 14:52:40 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: <7E5456D53823492AA59C352DE036B5ED@MaryPC> References: <7E5456D53823492AA59C352DE036B5ED@MaryPC> Message-ID: <748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> On 6 Dec 2010, at 08:53, Mary Hawking wrote: > http://www.channelregister.co.uk/2010/12/03/national_addressing_eric_pickles_ordnance_survey/ > > There was a good deal of discussion when PDS (the Personal > Demographics Servicehttp://www.connectingforhealth.nhs.uk/systemsandservices/demographics/pds > ) was introduced in the NHS on the risks that it could pose to > people with good reason to conceal their addresses and contact > details, including groups such as those with abusive ex-partners or > families and anyone with a connection to Huntingdon Life Sciences. > There is a means of withholding the demographics which might locate > you (see Access and Security): the back office still holds the > details but they are not displayed. > > Does anyone know whether there are intended to be facilities for > similar withholding of details in this new database, and do these > exist in the databases (thought by commentators on the article to be > highly inaccurate anyway) to be used to populate the new database, > and if so, will they be carried into the new system? > PDS declarations of ?vulnerability? were not carried forward into > the Children?s Database, and AFAIAA there were no mechanisms for > requesting withholding of contact details. > > Mary Hawking > > I think there's some misunderstanding here. As I understand it, this new announcement concerns a database of just addresses, not people's addresses. That is, it's like the Postcode Address File (PAF), a way to validate an address and ensure that it's used in one correct form. In itself it does nothing other than show that an address exists, not who lives there or what that address is used for. Omitting entries from such a database might be more of a security risk than listing them - their very omission makes them 'interesting'. For typical addresses with a building number it would be trivial to extract a list of omitted addresses. e.g. If there was a No. 5 and No. 7 Acacia Avenue then the absence of No. 6 from the database is interesting, suspicious or probative depending on your intent. A geographically targeted search could find 'missing' properties in an area where you suspected that your ex-wife, government minister, MI5 safe house or whatever was resident. Ian From igb at batten.eu.org Mon Dec 6 18:51:51 2010 From: igb at batten.eu.org (Ian Batten) Date: Mon, 6 Dec 2010 18:51:51 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: <748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> References: <7E5456D53823492AA59C352DE036B5ED@MaryPC> <748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> Message-ID: <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> > > I think there's some misunderstanding here. As I understand it, this new announcement concerns a database of just addresses Correct. This dataset does not contain names of people: it merely contains a list of street addresses. Unless you've changed the name of your house to include your name, it is of no value to your opponents. I think Pickles' announcement probably also marks the end of the PO monopoly on postcodes, which would be a very good thing. ian From maryhawking at tigers.demon.co.uk Tue Dec 7 08:01:17 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Tue, 7 Dec 2010 08:01:17 -0000 Subject: Is an (inaccurate) national adress database a treat to privacyand sometimes personal security? In-Reply-To: <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> References: <7E5456D53823492AA59C352DE036B5ED@MaryPC><748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> Message-ID: Sorry, my mistake Mary Hawking -----Original Message----- From: Ian Batten [mailto:igb at batten.eu.org] Sent: 06 December 2010 18:52 To: UK Cryptography Policy Discussion Group Subject: Re: Is an (inaccurate) national adress database a treat to privacyand sometimes personal security? > > I think there's some misunderstanding here. As I understand it, this new announcement concerns a database of just addresses Correct. This dataset does not contain names of people: it merely contains a list of street addresses. Unless you've changed the name of your house to include your name, it is of no value to your opponents. I think Pickles' announcement probably also marks the end of the PO monopoly on postcodes, which would be a very good thing. ian From Ross.Anderson at cl.cam.ac.uk Tue Dec 7 13:33:46 2010 From: Ross.Anderson at cl.cam.ac.uk (Ross Anderson) Date: Tue, 07 Dec 2010 13:33:46 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: Message-ID: Mary wrote: >There was a good deal of discussion when PDS (the Personal Demographics Service >http://www.connectingforhealth.nhs.uk/systemsandservices/demographics/pds ) >was introduced in the NHS on the risks that it could pose to people with >good reason to conceal their addresses and contact details, including groups >such as those with abusive ex-partners or families and anyone with a >connection to Huntingdon Life Sciences. I have knowledge of a recent case where a woman turned up at A&E with serious injuries after her ex-husband found her address from a relative who was a ward clerk at a trust. Neither the woman's GP nor A&E knew how to stop-note her on PDS, or even that it might have been a good idea to do so. This security failure is, I suspect, due to a perverse incentive on the Department of Health. The Department does not want to highlight the fact that its PDS system is available to over 800,000 users and is thus insecure. Officials have spent a huge amount of effort trying to centralise IT systems in the hope that this will help them manage the NHS better. They don't want to diss their own baby. But virtually no-one else has the ability to educate doctors on the scale and at the speed that would be advisable. An interesting feature of this case is that the woman is now suing the NHS for failing to advise her about PDS, the risk, and the advisability of being stop-noted. I hope she wins a bundle, and that it's public. That would finally cause trusts, GPs and their insurers to pay attention. Ross From igb at batten.eu.org Tue Dec 7 17:34:07 2010 From: igb at batten.eu.org (Ian Batten) Date: Tue, 7 Dec 2010 17:34:07 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: References: Message-ID: <605FF117-8019-44C2-B65D-DC03ADB58F1A@batten.eu.org> > The Department does not want to highlight > the fact that its PDS system is available to over 800,000 users and > is thus insecure. The nonsense about "oh well, they're all vetted" should hopefully be rendered slightly less convincing to government by mentioning Bradley Manning. Whatever your views on l'affaire Wikileaks, one positive side-effect of it will be that governments will realise that "secrets" accessible to millions of people aren't secret at all. By the by, back in the day, a company I was working with disabled USB and floppy ports on computers, to reduce data leakage (the argument ran that although the disabling was by-passable, it removed the "I didn't know I wasn't allowed to" argument). One has to wonder why the US Military didn't do likewise, and whether the much-vaunted NHS security policy will do so. ian From pwt at iosis.co.uk Wed Dec 8 11:03:16 2010 From: pwt at iosis.co.uk (Peter Tomlinson) Date: Wed, 08 Dec 2010 11:03:16 +0000 Subject: Is an (inaccurate) national adress database a treat to privacyand sometimes personal security? In-Reply-To: <605FF117-8019-44C2-B65D-DC03ADB58F1A@batten.eu.org> References: <605FF117-8019-44C2-B65D-DC03ADB58F1A@batten.eu.org> Message-ID: <4CFF65F4.8080404@iosis.co.uk> I'm really puzzled now. Is PDS just addresses, or are patient names (always or just in some doctors' practices) linked to their addresses on PDS? Peter On 07/12/2010 17:34, Ian Batten wrote: >> The Department does not want to highlight >> the fact that its PDS system is available to over 800,000 users and >> is thus insecure. > The nonsense about "oh well, they're all vetted" should hopefully be rendered slightly less convincing to government by mentioning Bradley Manning. Whatever your views on l'affaire Wikileaks, one positive side-effect of it will be that governments will realise that "secrets" accessible to millions of people aren't secret at all. > > By the by, back in the day, a company I was working with disabled USB and floppy ports on computers, to reduce data leakage (the argument ran that although the disabling was by-passable, it removed the "I didn't know I wasn't allowed to" argument). One has to wonder why the US Military didn't do likewise, and whether the much-vaunted NHS security policy will do so. > > ian > From igb at batten.eu.org Wed Dec 8 11:54:25 2010 From: igb at batten.eu.org (Ian Batten) Date: Wed, 8 Dec 2010 11:54:25 +0000 Subject: Is an (inaccurate) national adress database a treat to privacyand sometimes personal security? In-Reply-To: <4CFF65F4.8080404@iosis.co.uk> References: <605FF117-8019-44C2-B65D-DC03ADB58F1A@batten.eu.org> <4CFF65F4.8080404@iosis.co.uk> Message-ID: <5E50048C-70E2-49D9-B726-57D0F0E94EDB@batten.eu.org> On 08 Dec 10, at 1103, Peter Tomlinson wrote: > I'm really puzzled now. Is PDS just addresses, or are patient names (always or just in some doctors' practices) linked to their addresses on PDS? PDS is names. The Ordnance Survey dataset Mary originally referred to does not have names. Conflating the two is not helpful. ian From maryhawking at tigers.demon.co.uk Wed Dec 8 23:28:15 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Wed, 8 Dec 2010 23:28:15 -0000 Subject: Is an (inaccurate) national adress database a treat to privacy andsometimes personal security? In-Reply-To: References: Message-ID: I got interested in how to render a patient vulnerable on PDS when the initial leaflet came round, advising patients to see their GPs - with no information whatsoever for GPs on how to do it - and loads of misinformation when I tried to find out! Most GPs won't know: check it out before you start.. Mary Hawking Few months ago I did it for my very first patients needing this. It worked. -----Original Message----- From: Ross Anderson [mailto:Ross.Anderson at cl.cam.ac.uk] Sent: 07 December 2010 13:34 To: ukcrypto at chiark.greenend.org.uk Subject: Is an (inaccurate) national adress database a treat to privacy andsometimes personal security? Mary wrote: >There was a good deal of discussion when PDS (the Personal Demographics Service >http://www.connectingforhealth.nhs.uk/systemsandservices/demographics/pds ) >was introduced in the NHS on the risks that it could pose to people with >good reason to conceal their addresses and contact details, including groups >such as those with abusive ex-partners or families and anyone with a >connection to Huntingdon Life Sciences. I have knowledge of a recent case where a woman turned up at A&E with serious injuries after her ex-husband found her address from a relative who was a ward clerk at a trust. Neither the woman's GP nor A&E knew how to stop-note her on PDS, or even that it might have been a good idea to do so. This security failure is, I suspect, due to a perverse incentive on the Department of Health. The Department does not want to highlight the fact that its PDS system is available to over 800,000 users and is thus insecure. Officials have spent a huge amount of effort trying to centralise IT systems in the hope that this will help them manage the NHS better. They don't want to diss their own baby. But virtually no-one else has the ability to educate doctors on the scale and at the speed that would be advisable. An interesting feature of this case is that the woman is now suing the NHS for failing to advise her about PDS, the risk, and the advisability of being stop-noted. I hope she wins a bundle, and that it's public. That would finally cause trusts, GPs and their insurers to pay attention. Ross From maryhawking at tigers.demon.co.uk Fri Dec 10 15:24:34 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Fri, 10 Dec 2010 15:24:34 -0000 Subject: outsourcing GP appointments to India: is this legal under DPA? Message-ID: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> http://www.pulsetoday.co.uk/story.asp?sectioncode=23 &storycode=4128022&c=2 This is a story about NHS Shared Business Services (DH plus Steria) urging GP Consortia to make massive savings by outsourcing back-office functions and above all appointments to India. Regardless of the practicalities - and I cannot see this being popular with either GPs or patients - would this outsourcing of NHS data be legal under DPA? Even if already being done? Mary Hawking -------------- next part -------------- An HTML attachment was scrubbed... URL: From amidgley at gmail.com Sat Dec 11 02:01:36 2010 From: amidgley at gmail.com (Adrian Midgley) Date: Sat, 11 Dec 2010 02:01:36 +0000 Subject: outsourcing GP appointments to India: is this legal under DPA? In-Reply-To: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> References: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> Message-ID: On 10 December 2010 15:24, Mary Hawking wrote: > > Regardless of the practicalities ? and I cannot see this being popular with > either GPs or patients ? would this outsourcing of NHS data be legal under > DPA? Even if already being done? > > IANAL, but probably not. I interviewed one provider of a dictation service with the typing done in india, who would not have it that any data at all left the UK, since the typists were working on the server that was in the UK. I was unable to satisfy myself whether he believed it, and was therefore stupid and ignorant, or knew it to be untrue and was repeating the story anyway in the hope of gaining profit. -- Adrian Midgley http://www.defoam.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From fjmd1a at gmail.com Sat Dec 11 10:10:38 2010 From: fjmd1a at gmail.com (Francis Davey) Date: Sat, 11 Dec 2010 10:10:38 +0000 Subject: outsourcing GP appointments to India: is this legal under DPA? In-Reply-To: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> References: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> Message-ID: On 10 December 2010 15:24, Mary Hawking wrote: > http://www.pulsetoday.co.uk/story.asp?sectioncode=23&storycode=4128022&c=2 > > This is a story about NHS Shared Business Services (DH plus Steria) urging > GP Consortia to make massive savings by outsourcing back-office functions > and above all appointments to India. > > Regardless of the practicalities ? and I cannot see this being popular with > either GPs or patients ? would this outsourcing of NHS data be legal under > DPA? Even if already being done? > With care, namely suitable contractual arrangements or binding corporate rules (where done internally) it can be done. There is an infelicity in that the law does not properly specify processor to processor arrangements, but this would likely be controller to processor. -- Francis Davey From amidgley at gmail.com Sat Dec 11 18:29:45 2010 From: amidgley at gmail.com (Adrian Midgley) Date: Sat, 11 Dec 2010 18:29:45 +0000 Subject: outsourcing GP appointments to India: is this legal under DPA? In-Reply-To: References: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> Message-ID: Binding across jurisdictions is not easy. Bear in mind also that this is probably not a system intended to use appointments in the same way as a general practice currently uses them - the aim woulld be that someone who rings and asks for an appointment is given an appointment, or upsold to some other service, and the appointment is only fairly likely to be where the patient would currently go. Bad system. If it served the purposes of us or our patients, annd was economic, and safe, we would have set it up some time ago. -- Adrian Midgley http://www.defoam.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From maryhawking at tigers.demon.co.uk Sun Dec 12 08:28:02 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Sun, 12 Dec 2010 08:28:02 -0000 Subject: outsourcing GP appointments to India: is this legal under DPA? In-Reply-To: References: <3908C7790EDD48B4B907A3B57EA0623E@MaryPC> Message-ID: Interesting point, Adrian. Do you have any inside knowledge about any intention to regard all appointments as equal by NHS Business Services - or DH? I would have thought that if you call an on-line service for an appointment with a particular company, it is unlikely that you would be offered one with a business rival. But then, any time I have made an appointment over the phone it has not been via a call centre owned by a department with its own agenda! Mary Hawking _____ From: Adrian Midgley [mailto:amidgley at gmail.com] Sent: 11 December 2010 18:30 To: UK Cryptography Policy Discussion Group Subject: Re: outsourcing GP appointments to India: is this legal under DPA? Binding across jurisdictions is not easy. Bear in mind also that this is probably not a system intended to use appointments in the same way as a general practice currently uses them - the aim woulld be that someone who rings and asks for an appointment is given an appointment, or upsold to some other service, and the appointment is only fairly likely to be where the patient would currently go. Bad system. If it served the purposes of us or our patients, annd was economic, and safe, we would have set it up some time ago. -- Adrian Midgley http://www.defoam.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From amidgley at gmail.com Sun Dec 12 17:46:06 2010 From: amidgley at gmail.com (Adrian Midgley) Date: Sun, 12 Dec 2010 17:46:06 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> References: <7E5456D53823492AA59C352DE036B5ED@MaryPC> <748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> Message-ID: The same applies to Google Streetview. THe holes are potentially interesting. The general move to providing the State's data to the citizens is a good one. -- Adrian Midgley http://www.defoam.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From james.davis at ja.net Tue Dec 14 15:27:02 2010 From: james.davis at ja.net (James Davis) Date: Tue, 14 Dec 2010 15:27:02 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: References: <7E5456D53823492AA59C352DE036B5ED@MaryPC> <748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> Message-ID: <4D078CC6.3090201@ja.net> -----BEGIN PGP SIGNED MESSAGE----- Hash: SHA1 On 12/12/10 17:46, Adrian Midgley wrote: > The same applies to Google Streetview. THe holes are potentially > interesting. There are a lot of inconsistently and incorrectly numbered and named streets out there. I'm confident that this would overwhelm any interesting omissions in the data. James - -- James Davis +44 1235 822229 PGP: 0xD1622876 Senior CSIRT Member 0300 999 2340 (+44 1235 822340) Lumen House, Library Avenue, Didcot, Oxfordshire, OX11 0SG -----BEGIN PGP SIGNATURE----- Version: GnuPG v1.4.10 (GNU/Linux) Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ iEYEARECAAYFAk0HjMYACgkQhZi14NFiKHaUvQCfSvZFQv1KdbbUEYsPuml7L2Q8 CSEAn3nWXMpoP5Y5ZVF+OYqBk/s97ref =QKpV -----END PGP SIGNATURE----- JANET(UK) is a trading name of The JNT Association, a company limited by guarantee which is registered in England under No. 2881024 and whose Registered Office is at Lumen House, Library Avenue, Harwell Science and Innovation Campus, Didcot, Oxfordshire. OX11 0SG From Andrew.Cormack at ja.net Wed Dec 15 10:04:08 2010 From: Andrew.Cormack at ja.net (Andrew Cormack) Date: Wed, 15 Dec 2010 10:04:08 +0000 Subject: Is an (inaccurate) national adress database a treat to privacy and sometimes personal security? In-Reply-To: <4D078CC6.3090201@ja.net> References: <7E5456D53823492AA59C352DE036B5ED@MaryPC> <748EF3AF-8263-43C3-826B-14336164342E@sourcetagged.ian.co.uk> <5608D481-CF9F-4FE2-A4D9-F62740AFA759@batten.eu.org> <4D078CC6.3090201@ja.net> Message-ID: <61E52F3A5532BE43B0211254F13883AE015B73@EXC001> Apparently some people considered the holes in Streetview Germany sufficiently interesting that they got "egged"... http://www.bbc.co.uk/news/technology-11827862 Andrew > -----Original Message----- > From: ukcrypto-bounces at chiark.greenend.org.uk [mailto:ukcrypto- > bounces at chiark.greenend.org.uk] On Behalf Of James Davis > Sent: 14 December 2010 15:27 > To: UK Cryptography Policy Discussion Group > Subject: Re: Is an (inaccurate) national adress database a treat to > privacy and sometimes personal security? > > -----BEGIN PGP SIGNED MESSAGE----- > Hash: SHA1 > > On 12/12/10 17:46, Adrian Midgley wrote: > > The same applies to Google Streetview. THe holes are potentially > > interesting. > > There are a lot of inconsistently and incorrectly numbered and named > streets out there. I'm confident that this would overwhelm any > interesting omissions in the data. > > James > > - -- > James Davis +44 1235 822229 PGP: 0xD1622876 > Senior CSIRT Member 0300 999 2340 (+44 1235 822340) > Lumen House, Library Avenue, Didcot, Oxfordshire, OX11 0SG > -----BEGIN PGP SIGNATURE----- > Version: GnuPG v1.4.10 (GNU/Linux) > Comment: Using GnuPG with Mozilla - http://enigmail.mozdev.org/ > > iEYEARECAAYFAk0HjMYACgkQhZi14NFiKHaUvQCfSvZFQv1KdbbUEYsPuml7L2Q8 > CSEAn3nWXMpoP5Y5ZVF+OYqBk/s97ref > =QKpV > -----END PGP SIGNATURE----- > > JANET(UK) is a trading name of The JNT Association, a company limited > by guarantee which is registered in England under No. 2881024 > and whose Registered Office is at Lumen House, Library Avenue, > Harwell Science and Innovation Campus, Didcot, Oxfordshire. OX11 0SG > From marcus at connectotel.com Wed Dec 15 13:14:59 2010 From: marcus at connectotel.com (Marcus Williamson) Date: Wed, 15 Dec 2010 13:14:59 +0000 Subject: Crown Prosecution Service Delays Ruling on BT and Phorm UK Privacy Invasions Message-ID: Crown Prosecution Service Delays Ruling on BT and Phorm UK Privacy Invasions http://www.ispreview.co.uk/story/2010/12/15/crown-prosecution-service-delays-ruling-on-bt-and-phorm-uk-privacy-invasions.html From igb at batten.eu.org Wed Dec 15 13:20:45 2010 From: igb at batten.eu.org (Ian Batten) Date: Wed, 15 Dec 2010 13:20:45 +0000 Subject: Crown Prosecution Service Delays Ruling on BT and Phorm UK Privacy Invasions In-Reply-To: References: Message-ID: On 15 Dec 2010, at 13:14, Marcus Williamson wrote: > > Crown Prosecution Service Delays Ruling on BT and Phorm UK Privacy Invasions > > http://www.ispreview.co.uk/story/2010/12/15/crown-prosecution-service-delays-ruling-on-bt-and-phorm-uk-privacy-invasions.html Why don't they just say "we're not going to prosecute" and have done with it? Who on earth do they think they're kidding? I think an interesting FoI request might be the number of man-hours worked on this case, broken down by months. Because I bet it's zero, and they're hoping to compost the paper until everyone's forgotten about it. ian From marcus at connectotel.com Thu Dec 16 00:21:22 2010 From: marcus at connectotel.com (Marcus Williamson) Date: Thu, 16 Dec 2010 00:21:22 +0000 Subject: Contactless bank cards In-Reply-To: <4CE656A4.9010001@iosis.co.uk> References: <9dVUDsEe8p4MFAsN@perry.co.uk> <2D3A2407-3A22-49A9-B637-30180BC104E5@sourcetagged.ian.co.uk> <0EFE4206-66FB-444F-89E7-6ACC006B0194@batten.eu.org> <52A14F4F-82AB-4087-8B0E-0CA298DD1316@sourcetagged.ian.co.uk> <4CE643AD.4060603@iosis.co.uk> <4CE64FDC.9090007@callnetuk.com> <4CE656A4.9010001@iosis.co.uk> Message-ID: <7qmig6lg2t2nrsi2faqjr4nvmeo713slb2@4ax.com> Wembley Arena becomes first 'contactless' venue in Britain Wembley Arena has leveraged its sponsorship deal with Barclaycard to become the UK?s first music venue to introduce a contactless payment system. (more...) http://www.musicweek.com/story.asp?sectioncode=1&storycode=1043648&c=1 From mikie.simpson at gmail.com Fri Dec 17 11:38:07 2010 From: mikie.simpson at gmail.com (Michael Simpson) Date: Fri, 17 Dec 2010 11:38:07 +0000 Subject: Clouds! Message-ID: It seems that cloud computing is the new shiny up here in sunny scotland. Specifically use of cloud computing infrastructure to reduce costs in the NHS. Microsoft appear to be front and centre in the various presentations being given to the civil service and the buzz is growing about use of their Azure service. Whilst i understand and have used cloud services (mainly EC2 since its inception) for some specific tasks such as rapid prototyping of web applications for demo purposes and also for a highly scalable render farm experiment i am at a bit of a loss to see why it should be used in lieu of the normal "infrastructure as a service" or software as a service provided those companies already contracted (and paid for) by the taxpayer. -maybe for speeding up rendering of MRI images or scaling the intraweb server automagically so that when everyone hits it at 0900 it remains snappy, both of which could be achieved by utilising virtual machines or distributed clients to take advantage of *many* wasted cpu cycles that the scottish NHS already has. The point of this mail though is to bring to your attention the recent revelation by Jon Honeyball in this month's "PC Pro" after he managed to corner Bob Muglia - president of MSFT's server and tools business and asked him about the sanctity of data stored in their EU cloud. Jon was told that if the dept of homeland security (or presumably any other agency that really wants to) asks for any data held by MSFT anywhere then it will be transferred to the US datacentre and handed over "no ifs, no buts." caveat emptor mike From maxsec at gmail.com Fri Dec 17 13:53:44 2010 From: maxsec at gmail.com (Martin Hepworth) Date: Fri, 17 Dec 2010 13:53:44 +0000 Subject: Clouds! In-Reply-To: References: Message-ID: Big if and but, the DPA of which they should be aware. but sometimes the USA does have a clue about information retieval.. http://www.oxfordmail.co.uk/archive/2010/12/16/News+%28Cakenews%29/8740181.Formula_One_style_technology_used_in_court_case/ -- Martin Hepworth Oxford, UK On 17 December 2010 11:38, Michael Simpson wrote: > It seems that cloud computing is the new shiny up here in sunny > scotland. Specifically use of cloud computing infrastructure to reduce > costs in the NHS. Microsoft appear to be front and centre in the > various presentations being given to the civil service and the buzz is > growing about use of their Azure service. Whilst i understand and have > used cloud services (mainly EC2 since its inception) for some specific > tasks such as rapid prototyping of web applications for demo purposes > and also for a highly scalable render farm experiment i am at a bit of > a loss to see why it should be used in lieu of the normal > "infrastructure as a service" or software as a service provided those > companies already contracted (and paid for) by the taxpayer. > -maybe for speeding up rendering of MRI images or scaling the intraweb > server automagically so that when everyone hits it at 0900 it remains > snappy, both of which could be achieved by utilising virtual machines > or distributed clients to take advantage of *many* wasted cpu cycles > that the scottish NHS already has. > > > > The point of this mail though is to bring to your attention the > recent revelation by Jon Honeyball in this month's "PC Pro" after he > managed to corner Bob Muglia - president of MSFT's server and tools > business and asked him about the sanctity of data stored in their EU > cloud. Jon was told that if the dept of homeland security (or > presumably any other agency that really wants to) asks for any data > held by MSFT anywhere then it will be transferred to the US datacentre > and handed over "no ifs, no buts." > > caveat emptor > > mike > > -------------- next part -------------- An HTML attachment was scrubbed... URL: From maryhawking at tigers.demon.co.uk Sat Dec 18 09:08:36 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Sat, 18 Dec 2010 09:08:36 -0000 Subject: Clouds! In-Reply-To: References: Message-ID: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> Does that mean that, assuming that it is correct that if the Department of Homeland Security asked for any data held in MS controlled clouds it would be handed over without further ado, it is illegal, under DPA for anyone in any sector to use Cloud technology - or at least any Cloud application in which MS was involved - for anything containing personal data? Mind you, if NHS Business is exporting backoffice functions such as typing letters and making appointments to India, and now this, it would appear that the NHS doesn't feel DPA applies to them - if someone can save a few pence "in these times of unprecedented financial stringency"! Mary Hawking _____ From: Martin Hepworth [mailto:maxsec at gmail.com] Sent: 17 December 2010 13:54 To: UK Cryptography Policy Discussion Group Subject: Re: Clouds! Big if and but, the DPA of which they should be aware. but sometimes the USA does have a clue about information retieval.. http://www.oxfordmail.co.uk/archive/2010/12/16/News+%28Cakenews%29/8740181.F ormula_One_style_technology_used_in_court_case/ -- Martin Hepworth Oxford, UK On 17 December 2010 11:38, Michael Simpson wrote: It seems that cloud computing is the new shiny up here in sunny scotland. Specifically use of cloud computing infrastructure to reduce costs in the NHS. Microsoft appear to be front and centre in the various presentations being given to the civil service and the buzz is growing about use of their Azure service. Whilst i understand and have used cloud services (mainly EC2 since its inception) for some specific tasks such as rapid prototyping of web applications for demo purposes and also for a highly scalable render farm experiment i am at a bit of a loss to see why it should be used in lieu of the normal "infrastructure as a service" or software as a service provided those companies already contracted (and paid for) by the taxpayer. -maybe for speeding up rendering of MRI images or scaling the intraweb server automagically so that when everyone hits it at 0900 it remains snappy, both of which could be achieved by utilising virtual machines or distributed clients to take advantage of *many* wasted cpu cycles that the scottish NHS already has. The point of this mail though is to bring to your attention the recent revelation by Jon Honeyball in this month's "PC Pro" after he managed to corner Bob Muglia - president of MSFT's server and tools business and asked him about the sanctity of data stored in their EU cloud. Jon was told that if the dept of homeland security (or presumably any other agency that really wants to) asks for any data held by MSFT anywhere then it will be transferred to the US datacentre and handed over "no ifs, no buts." caveat emptor mike -------------- next part -------------- An HTML attachment was scrubbed... URL: From amidgley at gmail.com Sat Dec 18 09:34:06 2010 From: amidgley at gmail.com (Adrian Midgley) Date: Sat, 18 Dec 2010 09:34:06 +0000 Subject: Clouds! In-Reply-To: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> References: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> Message-ID: On 18 December 2010 09:08, Mary Hawking wrote: > Does that mean that, assuming that it is correct that if the Department > of Homeland Security asked for any data held in MS controlled clouds it > would be handed over without further ado, it is illegal, under DPA for > anyone in any sector to use Cloud technology ? or at least any Cloud > application in which MS was involved - for anything containing personal > data? > > Encrypt. -- Adrian Midgley http://www.defoam.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From rich at annexia.org Sat Dec 18 15:35:14 2010 From: rich at annexia.org (Richard W.M. Jones) Date: Sat, 18 Dec 2010 15:35:14 +0000 Subject: Clouds! In-Reply-To: References: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> Message-ID: <20101218153514.GA23315@annexia.org> On Sat, Dec 18, 2010 at 09:34:06AM +0000, Adrian Midgley wrote: > On 18 December 2010 09:08, Mary Hawking wrote: > > Does that mean that, assuming that it is correct that if the Department > > of Homeland Security asked for any data held in MS controlled clouds it > > would be handed over without further ado, it is illegal, under DPA for > > anyone in any sector to use Cloud technology ? or at least any Cloud > > application in which MS was involved - for anything containing personal > > data? > > Encrypt. Not sure what this single word answer implies, but I'll tell you a true story: probably second most popular question I get about libguestfs (after "does it run on Ubuntu?") is "OMG you're revealing the files on virtual machines bypassing all the file permissions! How can I stop this!?" I have to tell them that you can already read whatever is in a cloud virtual machine, encrypted or not. libguestfs just makes it easy. Apart from some very speculative schemes[1], if you want to do computing in someone else's cloud, someone else is going to have complete access to the data on those virtual machines, and whether you've encrypted the disks or not won't make a blind bit of difference. Amazon tell us that they divide up storage and virtualization so one Amazon administrator can't read the keys out of VM memory and use them to decrypt the hard disk stored somewhere else, but you have to assume that two Amazon admins could do this, and in any case take what they say on trust. [Not picking on Amazon here BTW, this applies equally to any public cloud, they are just the most popular one] Private cloud APIs where the physical machines and storage are under your own control excepted from the above. Rich. [1] Here would I add a reference to a paper about using mathematical transforms to perform computing on encrypted data without decrypting it, but I'm afraid I can't find it right this minute ... -- Richard Jones Red Hat From amidgley at gmail.com Sun Dec 19 02:55:47 2010 From: amidgley at gmail.com (Adrian Midgley) Date: Sun, 19 Dec 2010 02:55:47 +0000 Subject: Clouds! In-Reply-To: <20101218153514.GA23315@annexia.org> References: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> <20101218153514.GA23315@annexia.org> Message-ID: If one only uses the cloud (whatever that turns out to be) for data storage, then encryption of that data before passing it to the storage, and decryption locally after retrieving it provides a degree of reassurance against a company with extensive hopes of government sales, or which has considerable earnings within the sphere of influence of a government, dragging copies of one's data from wherever it might be (indeterminate, presumably, if it is in a cloud) and handing it over. But yes, hard to process information without being able to decrypt it. Perhaps not impossible - the contents of each data field (but not each metadata field) might be encrypted against a key held by the user, and then passed to the server, which can assemble to fields into a correct record without decrypting the actual text. It wouldn't work very well for Read Codes if they were stored in a relational database fashion, but if they were stored like EMIS does it would reconstruct the visible notes. ANd it would not support searches of a large population quickly, as a relational database does, however it would work as well as anything does with a document model of record, which I continue to think is potentially much better for the doctors and others making the record, and no worse for those using it for day to day medicine. Putting notes you've made on a computer someone else controls, for whatever level of in-group you care to adopt (me, me and my partners, my consortium, the NHS, all EEC health services, the world, the inhabitants of the observable universe including our entire future lightcone) is dumb. If you want to control access to them. -- Adrian Midgley http://www.defoam.net/ -------------- next part -------------- An HTML attachment was scrubbed... URL: From igb at batten.eu.org Sun Dec 19 09:36:50 2010 From: igb at batten.eu.org (Ian Batten) Date: Sun, 19 Dec 2010 09:36:50 +0000 Subject: Clouds! In-Reply-To: <20101218153514.GA23315@annexia.org> References: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> <20101218153514.GA23315@annexia.org> Message-ID: > > [1] Here would I add a reference to a paper about using mathematical > transforms to perform computing on encrypted data without decrypting > it, but I'm afraid I can't find it right this minute ... Homomorphic encryption. Fascinating, but not practical today, nor likely to become so any time soon. And the definition of "computation" it uses is not hugely useful for the purposes this list is concerned with. ian From igb at batten.eu.org Sun Dec 19 09:41:20 2010 From: igb at batten.eu.org (Ian Batten) Date: Sun, 19 Dec 2010 09:41:20 +0000 Subject: Clouds! In-Reply-To: <20101218153514.GA23315@annexia.org> References: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> <20101218153514.GA23315@annexia.org> Message-ID: > > Apart from some very speculative schemes[1], if you want to do > computing in someone else's cloud, someone else is going to have > complete access to the data on those virtual machines, and whether > you've encrypted the disks or not won't make a blind bit of > difference. It's a pretty active area of research, though, so I don't think we should assume this is axiomatic. For most cases, the desired position is not that access to the data is impossible, rather that it's impractical relative to the value of the data to the attacker. In some cases, schemes which rely on the security properties given by processors and support logic (for example, "Trusted Execution" as leveraged by the CMU "Flicker" project headed by Jonathan McClure) may be sufficient. ian From rich at annexia.org Sun Dec 19 09:57:27 2010 From: rich at annexia.org (Richard W.M. Jones) Date: Sun, 19 Dec 2010 09:57:27 +0000 Subject: Clouds! In-Reply-To: References: <909B86ED5B60400FB5D3B56219EC2B30@MaryPC> <20101218153514.GA23315@annexia.org> Message-ID: <20101219095727.GA32351@annexia.org> On Sun, Dec 19, 2010 at 09:36:50AM +0000, Ian Batten wrote: > > > > [1] Here would I add a reference to a paper about using mathematical > > transforms to perform computing on encrypted data without decrypting > > it, but I'm afraid I can't find it right this minute ... > > Homomorphic encryption. Fascinating, but not practical today, nor > likely to become so any time soon. And the definition of > "computation" it uses is not hugely useful for the purposes this list > is concerned with. That's the one. Wikipedia has a good description: https://secure.wikimedia.org/wikipedia/en/wiki/Homomorphic_encryption#Fully_homomorphic_encryption Any interesting papers on it I should read? Rich. -- Richard Jones Red Hat From lists at internetpolicyagency.com Sun Dec 19 10:30:32 2010 From: lists at internetpolicyagency.com (Roland Perry) Date: Sun, 19 Dec 2010 10:30:32 +0000 Subject: Clouds! In-Reply-To: References: Message-ID: In article , Michael Simpson writes > The point of this mail though is to bring to your attention the >recent revelation by Jon Honeyball in this month's "PC Pro" after he >managed to corner Bob Muglia - president of MSFT's server and tools >business and asked him about the sanctity of data stored in their EU >cloud. Jon was told that if the dept of homeland security (or >presumably any other agency that really wants to) asks for any data >held by MSFT anywhere then it will be transferred to the US datacentre >and handed over "no ifs, no buts." Cloud computing is the new growth area for regulators interested in data privacy and cross-border issues. All the major institutions are looking at it - EU, Council of Europe, OECD and so on. It's a problem because the (ostensibly simpler) jurisdictional issues of non-cloud computing haven't really been sorted out yet. But the good news is that (at a senior level anyway) the major suppliers understand the problem and are happy to advise potential customers that it's not the solution for them, if certain essential (to their application) safeguards, like knowing exactly which country your data is held in, are lacking. It sounds like MSFT are being equally up front about the Law Enforcement aspects. > >caveat emptor It's not compulsory to use cloud computing, if it's not suitable for the application. But I agree that potential users need to know what questions to ask. -- Roland Perry From maryhawking at tigers.demon.co.uk Mon Dec 20 08:24:47 2010 From: maryhawking at tigers.demon.co.uk (Mary Hawking) Date: Mon, 20 Dec 2010 08:24:47 -0000 Subject: Clouds! In-Reply-To: References: Message-ID: Of course, the problem with HMG and NHS is that, even if they know the right questions to ask, they appear to have a rule either not to ask them or to ignore any answers suggesting this might not be quite what is needed! Mary Hawking -----Original Message----- From: Roland Perry [mailto:lists at internetpolicyagency.com] Sent: 19 December 2010 10:31 To: ukcrypto at chiark.greenend.org.uk Subject: Re: Clouds! In article , Michael Simpson writes > The point of this mail though is to bring to your attention the >recent revelation by Jon Honeyball in this month's "PC Pro" after he >managed to corner Bob Muglia - president of MSFT's server and tools >business and asked him about the sanctity of data stored in their EU >cloud. Jon was told that if the dept of homeland security (or >presumably any other agency that really wants to) asks for any data >held by MSFT anywhere then it will be transferred to the US datacentre >and handed over "no ifs, no buts." Cloud computing is the new growth area for regulators interested in data privacy and cross-border issues. All the major institutions are looking at it - EU, Council of Europe, OECD and so on. It's a problem because the (ostensibly simpler) jurisdictional issues of non-cloud computing haven't really been sorted out yet. But the good news is that (at a senior level anyway) the major suppliers understand the problem and are happy to advise potential customers that it's not the solution for them, if certain essential (to their application) safeguards, like knowing exactly which country your data is held in, are lacking. It sounds like MSFT are being equally up front about the Law Enforcement aspects. > >caveat emptor It's not compulsory to use cloud computing, if it's not suitable for the application. But I agree that potential users need to know what questions to ask. -- Roland Perry From mikie.simpson at gmail.com Mon Dec 20 10:39:20 2010 From: mikie.simpson at gmail.com (Mike Simpson) Date: Mon, 20 Dec 2010 10:39:20 +0000 Subject: Clouds! In-Reply-To: References: Message-ID: <09771C89-1530-4562-9E74-E98C0165953F@gmail.com> On 20 Dec 2010, at 08:24, "Mary Hawking" wrote: > Of course, the problem with HMG and NHS is that, even if they know the right > questions to ask, they appear to have a rule either not to ask them or to > ignore any answers suggesting this might not be quite what is needed! > > Mary Hawking >> As it took Jon 2 years of asking various managers within msft before finally getting an answer by cornering bob in a corridor at the PDC I really wouldn't paint that as being "upfront". I also don't think the question will be asked up here without sotto voce prompting. I do think that "private" clouds would be the best way to have patient data on a cloud-based system -physical access and that - and am looking at openstack and also redhat's PaaS (probably once it filters down to CentOS which is our main platform of choice) for this. However I am still at a loss as to why it seems a good idea to add cloud based infrastructure and costs to the already heady mix of local physical servers running NT4 - 2k3, a spattering of old school unix, and the datacentre in central Scotland supplying citrix metaframe access to most of us. A rationalisation of the current setup would seem more appropriate rather than looking for the next magic wand. In summary though, without explicit legal contracts stating that your patient data will not leave the datacentre in the UK or EU under any circumstances then you cannot proceed with that company's cloud solution without breaching the DPA - no matter what they promise is "coming soon". Does that seem fair? mike ps please don't think I have an issue with msft per se. It is just that they tend to be front and centre when shiny gets presented and there us a degree of coziness that irks me somewhat. From lists at internetpolicyagency.com Mon Dec 20 11:30:15 2010 From: lists at internetpolicyagency.com (Roland Perry) Date: Mon, 20 Dec 2010 11:30:15 +0000 Subject: Clouds! In-Reply-To: <09771C89-1530-4562-9E74-E98C0165953F@gmail.com> References: <09771C89-1530-4562-9E74-E98C0165953F@gmail.com> Message-ID: In article <09771C89-1530-4562-9E74-E98C0165953F at gmail.com>, Mike Simpson writes >On 20 Dec 2010, at 08:24, "Mary Hawking" > wrote: > >> Of course, the problem with HMG and NHS is that, even if they know the right >> questions to ask, they appear to have a rule either not to ask them or to >> ignore any answers suggesting this might not be quite what is needed! >> >> Mary Hawking >>> > >As it took Jon 2 years of asking various managers within msft before >finally getting an answer by cornering bob in a corridor at the PDC I >really wouldn't paint that as being "upfront". Having been at a major intergovernmental event with MSFT present, and admitting what I recounted, I think they were. >In summary though, without explicit legal contracts stating that your >patient data will not leave the datacentre in the UK or EU under any >circumstances then you cannot proceed with that company's cloud >solution without breaching the DPA - no matter what they promise is >"coming soon". Agreed, and that's what the suppliers do seem happy to come clean about. Although I don't doubt there's may still be some need to "join up" their intergovernmental position with that of the salesmen on the ground. -- Roland Perry From igb at batten.eu.org Mon Dec 20 13:20:05 2010 From: igb at batten.eu.org (Ian Batten) Date: Mon, 20 Dec 2010 13:20:05 +0000 Subject: Clouds! In-Reply-To: <09771C89-1530-4562-9E74-E98C0165953F@gmail.com> References: <09771C89-1530-4562-9E74-E98C0165953F@gmail.com> Message-ID: <46FC0F41-8971-439E-8121-D808EC283976@batten.eu.org> > > In summary though, without explicit legal contracts stating that your patient data will not leave the datacentre in the UK or EU under any circumstances then you cannot proceed with that company's cloud solution without breaching the DPA - no matter what they promise is "coming soon". I have spoken to people in government on the matter, who said that governance and jurisdiction was the main problem confronting cloud solutions, and was today pretty much an eliminator for use for any project where CESG's remit runs. The implication was that even encrypted data would be a problem (on availability rather than confidentiality or integrity grounds). I've spoken to the UK evangelist for Amazon cloud services and he was able to talk about the certifications that their data centres have (specifically HIPPA for US health data) but confirmed that although they might be able to offer EU-only data from their Dublin datacentre they would not be able to offer UK-only as they don't operate EC3/S3/etc from any UK base. It's hardly any secret that the big players in UK government IT services are building UK-based private cloud infrastructures to address this market, where they can offer location and personnel clearance guarantees. It does strike me, speculating wildly, that one consequence of the economic problems in Eire may be to make it somewhat less of a low-tax honeypot. I'm not convinced that the sole reason for US companies setting up in Eire is low tax --- you can never underestimate the nostalgia and longing of the Irish-American community towards "the old country" --- but it's clearly one of the primary drivers, and if Eire's tax regime swung closer to the EU median then other locations would be as attractive: the UK, for example. Microsoft's BPOS offering again comes out of Eire datacentres and again has governance issues. What's interesting as well is that the UK government market isn't actually attractive enough to cause MSFT and Amazon to stir themselves from their Irish lair. ian From fw at deneb.enyo.de Sun Dec 26 18:05:34 2010 From: fw at deneb.enyo.de (Florian Weimer) Date: Sun, 26 Dec 2010 19:05:34 +0100 Subject: Clouds! In-Reply-To: (Michael Simpson's message of "Fri, 17 Dec 2010 11:38:07 +0000") References: Message-ID: <87wrmwl89d.fsf@mid.deneb.enyo.de> * Michael Simpson: > caveat emptor The elephant in the room is that most large organizations feel that they have as much control over their own IT today as they would have over a cloud service provider (if not less). If you're still running NT in any capacity, there's probably some truth to it. Even with our own IT infrastructure, most of us face an average gross data leakage rate of around 10 kbps per TB stored, simply due to hard drive failures. This doesn't sound much given today's bandwidths, but it's actually quite a lot, particularly if you have to assume that your vendor turns against you and exerts control over the data transmission (which is technically feasible). From marcus at connectotel.com Tue Dec 28 17:08:51 2010 From: marcus at connectotel.com (Marcus Williamson) Date: Tue, 28 Dec 2010 17:08:51 +0000 Subject: =?ISO-8859-1?Q?Banks_attempt_to_su?= =?ISO-8859-1?Q?ppress_maths_studen?= =?ISO-8859-1?Q?t's_expos=E9_of_chip_?= =?ISO-8859-1?Q?and_pin?= Message-ID: Banks attempt to suppress maths student's expos? of chip and pin http://www.independent.co.uk/news/education/education-news/banks-attempt-to-suppress-maths-students-expos233-of-chip-and-pin-2170396.html From rich at annexia.org Tue Dec 28 19:06:01 2010 From: rich at annexia.org (Richard W.M. Jones) Date: Tue, 28 Dec 2010 19:06:01 +0000 Subject: Banks attempt to =?iso-8859-1?Q?suppre?= =?iso-8859-1?Q?ss_maths_student's_expos=E9?= of chip and pin In-Reply-To: References: Message-ID: <20101228190601.GA31257@annexia.org> On Tue, Dec 28, 2010 at 05:08:51PM +0000, Marcus Williamson wrote: > > Banks attempt to suppress maths student's expos? of chip and pin > > http://www.independent.co.uk/news/education/education-news/banks-attempt-to-suppress-maths-students-expos233-of-chip-and-pin-2170396.html Ross Anderson's [who is on this list] response is gold: http://www.cl.cam.ac.uk/~rja14/Papers/ukca.pdf I really hope that letter sent those fools off to look up Erasmus, and they learned a little bit. Rich. -- Richard Jones Red Hat From tugwilson at gmail.com Tue Dec 28 17:43:21 2010 From: tugwilson at gmail.com (John Wilson) Date: Tue, 28 Dec 2010 17:43:21 +0000 Subject: =?ISO-8859-1?Q?Re=3A_Banks_attempt_to_suppress_maths_student=27s_expos?= =?ISO-8859-1?Q?=E9_of_chip_and_pin?= In-Reply-To: References: Message-ID: Here's the link to the original blog post which has links to both letters http://www.lightbluetouchpaper.org/2010/12/25/a-merry-christmas-to-all-bankers/ Ross' reply is a classic. John Wilson From bogus@does.not.exist.com Sun Dec 12 21:12:03 2010 From: bogus@does.not.exist.com () Date: Sun, 12 Dec 2010 21:12:03 -0000 Subject: No subject Message-ID: really!) I suspect that in current law an agent has to be human. But there = are definitely legal problems around the status of "software agents", so th= at may be the way the law is heading. If I find time to investigate I'll tr= y to remember to report back here. =20 > > But it seemed that if that resulted in the secretary acting *as* the > >boss for a particular subset of his work then the secretary/boss *is* > >the intended recipient and the interception problem goes away. That > >seemed a reasonable fit for the paper-based world where, if I get a > >letter signed "pp CEO" then I treat it as coming from the CEO, even > >though it's very obvious that it hasn't. >=20 > But what about inbound items (email or postal) marked "Private and > Confidential"? Not sure how widespread the practice is, but the executives of my employer = have two different e-mail addresses, one of which I suspect is accessible t= o PAs and the other isn't. Whether that is enforced by technology or practi= ce I don't know. > >It also seemed to make it the boss's responsibility to define the > >extent of actions for which the secretary could act as agent, and if > >the boss doesn't make that clear then it's their problem and not the > >secretary's. It seemed a bit unfair to us if the poor secretary > carried > >the can for misinterpreting unclear instructions, which seems to be > >another consequence of trying to justify it as interception-with- > consent :( >=20 > I think we are back in the situation I was describing earlier - there > might be one outcome dictated by common sense (based on a deeper > understanding of these 'private sector interception' issues than > perhaps > was exposed in the ICOA Review in 1999), and another by the way the > current law is drafted. I suspect we are. But what's puzzling is that I would expect MPs to have un= derstood the PA situation from their own experience so to have talked about= it. Since, as far as I know, they didn't, that made me wonder if there was= an obvious (to them) answer to the problem that we techies were missing. B= ut it may be that all the examples in their world are covered by clear defi= nitions of the doormat/BT box etc. and it's only in things like e-mail, voi= cemail, Centrex, with which they weren't familiar in 1999 that the location= of the doormat isn't clear, so the possibility that the PA is on the "wron= g" side of it arises. Andrew > Roland. >=20 > >-- > >Andrew Cormack, Chief Regulatory Adviser, JANET(UK) >=20 > -- > Roland Perry