<

 F.A.O.  Ian Taylor MP

 total 11 pages inc. cover

DaveBird , http://www.xemu.demon.co.uk/index.html
xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx 
ON BEHALF OF W.MidlandsBranch of the National Council for Civil Liberies


to:  Ian Taylor MP,  minister of telecommunications,       25/NOV/1996
     Department of Trade & Industry, 1 Victoria St,
     London, SW1H 0ET, ENGLAND.

re:  Communiqué of the Telecomms Council re unlawful & harmful content


Dear Ian Taylor,

I enclose our response to the Communiqué, and hope you will be willing to 
make some of the points in it on behalf of the UK at the Telecomms 
Council on Thursday (28th November).

Liability should occur in the place where sending, viewing, exracting to keep.
or deliberate republishing by a reader, occurs.  The best model is 
conditional Common Carrier. Senders can occur loss financially 
and to their rights from 
removal of material on unproven and possibly false complaints.  Mandatory 
removal should be by criminal warrant or civil injunction----if there is 
not a successful case before a jury, the owner should be compensated. 
Measures of prior restraint are generally ineffective.

We believe there should be a right of anonymous free speech. At most, 
disclosure of a name should be ordered only when there is a strong case 
for a criminal conviction leading to long imprisonment.

Filtering systems have distinct limitations. It would be disproportionate  
to enforce child ratings on all adult material which is quite content to 
be blocked from child access---rating of webpages should be on a true 
voluntary basis.  We concur that the American RSACi system is totally 
unsuitable, and would like to see a firm recommendation to that effect 
for ISPs which talk about prematurely committing to a failed system.

I look forward to hearing your views on our response to the communiqué 
(full text enclosed), and hope you have a fruitful conference.


Yours Sincerely,

Dave Bird, for W.Mids. NCCL group.

IN REPLY TO:  A COMMUNICATION  TO THE E.P., COUNCIL & COMMITTEES ON
            ''Illegal and harmful content on the Internet''

P04. INTRODUCTION. The nature of unlawful and distasteful material.
LIABILITY OF ACCESS PROVIDERS should be conditional common carrier.
LIABILITY SHOULD BE LOCATED where the specific act of sending, viewing, copying or redistributing information takes place.

P05. UNLAWFUL MATERIAL is innocent until proved guilty and must be balanced with rights against allegations proved false. Examples are given of false or oppressive removal of information. CIVIL LAW should operate by court order, with possible redress.

P06. CRIMINAL LAW should operate similarly by warrant. BLOCKING ACCESS is unwise and often unfeasible.
P07. The two examples from the communiqué are analysed. Prior restraint is an ineffective strategy. Law enforcement works best by the (‘Net) community’s consent.

P08. ANONYMOUS FREE SPEECH IS DEFENDED. Three option are given: true free speech, release of names only for very serious crimes, or last resort checks & balances.
P09. LAWFUL MATERIAL...is best described as ''distasteful''. It should be filtered for minors, not for (unwilling) adults. FILTERING SYSTEMS---characteristics analysed.

p10. THE VOLUNTARY SYSTEM. Ratings have trouble with changing materials. It would be disproportionate to put child ratings on adult discourse There are some problems with current UK developments in this area.
WHICH RATINGS SYSTEM
p11. We need proper local systems as per the last Communiqué.

THE INTRODUCTION AND POINT 3, while emphasising the positive potential of the Internet, identify certain areas of unlawful material to which current laws should be applied & distasteful material where the chief remedy is to provide the means by which others may keep them from vulnerable groups such as minors. Unlawful materials were variously civil law [libel, copyright, commercial & personal confidentiality], and criminal law [fraud, national security e.g. terrorist activity], some of it widely varying [obscenity, incitement to racial hatred]. It is unlikely on a world scale countries could agree whether any hatespeech laws were constitutional, let alone to all outlaw holocaust denial; any collaborative framework must allow for this wide variation.

{A}.LIABILITY---ACCESS PROVIDERS.
The best model is that anyone acting as access provider:
(i) must deal with any complaint that personal mail in harassment designed to put the recipient in genuine fear, and may make regulations on volume or technical construction irrespective of content as needed to prevent deliberate blocking or stop massive repetition of the same item,
(ii) must otherwise carry all lawful material from their users irrespective of any lawful content and especially of political or religious opinions, and all material addressed to their users unless the users ask to refuse it; and must offer safe passage to mail in transit from and to a third country without reading or interfering with it, as with ordinary post.
The sender should have control over any message sent, and any specific recipients also have an interest in it: the access provider should indemnify them that it will be safely delivered, if the access provider wishes to be free of liability for content conveyed.

LIABILITY---WHERE?
The liability for sending a message into the Internet should occur at the place where it is put onto the Internet. The liability for viewing, keeping, or deliberately circulating onward by a reader, of any message from the Internet should occur in each case at the place where this is done. Viewing should be on the same basis as viewing a TV program; taking a copy outside the Internet software or viewing after the connection is dropped as possessing a permanent copy; deliberately republishining onwards to others by the Internet or other media as distribution. While there is some scope for uniformity between similar countries the basic principle should be as for extradition, that police cooperate in two states if the act is a serious criminal offence [not politically motivated] in both.

The Internet occurs in a context where countries like Iran or Singapore may be involved, or there are the very different legal and constitutional provisions of the United States to contend with. There should be full subsidiarity, with no country trying to export uniquely local laws to other jurisdictions. This may lead to some anomalies....but no worse, and no more avoidable, than when taking print magazines or viewing television across international boundaries and outside the E.U.

{B}UNLAWFUL MATERIAL.
It seems a strange omission that Freedom of Expression is not mentioned in the case of material which is alleged to be unlawful but has not yet been proved so in the courts---that the author is innocent until proved guilty, and that if his material is suppressed on a false charge then he has suffered a wrong which should be compensated and, if foreseeable, prevented. There could be problems, either with state bodies abusing power or at least subject to no balancing oversight, and especially with individual in civil law able to impose arbitrary extra-legal demands and doing so falsely.

Examples are:

• In Holland the ‘Church’ of Scientology sued author Karin Spaink to remove material said to reveal their secrets but only substantiated a small part of their claim, the clear intent being to silence her by threats. In America, they abused an arrangement with Netcom online to remove material, by having them do so based on false statements.
• In the UK the Clubs & Vice unit has repeatedly proceeded by threats and seizures which never reach jury trial and which a jury might well not sustain. They attempted to have newsgroups removed by a similar process of background threats which is unlikely to stand up in court. We therefore believe that any removal of material must lead to a guaranteed jury trial if resisted, to ensure justice for both parties.
• In the UK, newspaper owner Robert Maxwell repeatedly used writs to gag publication which never proceeded to trial to suppress information on his financial misdeeds. Thousands of innocent people lost their pensions as a result. These regulations would be ensuring that thousands could be cheated again in a similar situation and the Internet silenced from comment, especially if he could also issue writs to get the names of anyone giving the information anonymously.

UNLAWFUL MATERIAL---CIVIL LAW.
Civil law matters such as copyright and breach of confidence should be dealt with as follows.
(i) The monitoring organisation should have a ''cease and desist'' letter passed to the user, which would of course say that prolonging the publication would be likely to increase the damages. The ISP would ask the user for a response [if none, then they delete the material anyway].
(ii) If they have prima facie evidence and can convince a court they have a good chance of victory, and they are suffering a real and increasing loss as publication continues, they can get a court order for its removal. If they the strength of the case and the urgency of the situation is sufficient, the order can be in the first place ex parte.

However, this should be balanced as follows. They should indemnify the defendant that they will within three months apply for a jury trial. If the defendant resists and the trial either fails to proceed or rules for the defence, then damages should be due. In order that nobody can ''buy'' gagging orders for money, the court should also be obliged to consider whether orders have been overturned or based on false information--from related plaintiffs and/or on related material--before and refuse orders in such cases; and the court should automatically punish for contempt anybody obtaining an order based on deliberate false information.

UNLAWFUL MATERIAL---CRIMINAL LAW.
The simplest system is that the police and prosecutors be able on prima facie evidenceeither to issue warning letters saying that noncompliance may increase the penalty on the user, or to obtain a court warrant for the removal of material if there is a sevre and increasing criminal mischeif by it staying. The balancing provision would be that the court should be obliged to order damages paid, or the defendant might sue separately, if they resisted and a date for jury trial was not applied for within three months, or if such a trial ruled against the prosecution. The judge or magistrate issuing the warrant would also have a duty in law not to issue it if there was a record of similar failed warrants. None of this would prejudice immediate measures being taken and upheld to remove obviously unlawful materials such as trading in hard drugs. If the system is based on (criminal) warrants or (civil) court orders then there is no ambiguity in the treatment of access providers. Of course they must obey warrants and court orders, or answer to the issuing court for their refusal. At other times they should be absolved of responsibility for the users’ messages which they carry.

The complainant may wish to pursue criminal warrants or court orders in a number of countries. There is bound to be, though some parts of the law can be standardised, a diversity of law in the different nations [especially on such divergent standards as obscenity]. It should be the responsibility of the complaining nation to only issue requests which they believe to the best of their knowledge are applicable in the receiving country’s law; but of the receiving country to check for certain. Damages for wrongful complaints should simply be multiplied by the number of countries involved, all collectable through the one court in the complaining country.

{C} BLOCKING ACCESS.
The response to material which is against the civil or criminal law in the receiving country is that local laws nevertheless apply to any citizen who either takes or forwards a copy---that is an act within local jurisdiction. There have been attempts, historically, to prevent the spread of particular news stories before into (e.g.) my own country. At the time of the abdication, stories were cut out of continental newspapers on arrival which said anything about Edward and Mrs Simpson. In the 1980s, injunctions were used to prevent discussion of a book called Spycatcher, which discussed an earlier attempted rightwing coup within the security services. Neither was very successful.

Attempts to prevent the dissemination of particular news or small quotes are likely to prevent a reaction where enforcement of the law against a few provokes mass defiance, humiliating the law as ineffective overall -- a situation no wise government will provoke. It is like King Knut putting his throne on the beach and ordering the tide to turn the tide back. Knut did this to show flattering couriers the limits of his power. It is to be hoped the telecommunications council will not also need to literally demonstrate the limitations of such powers.

Lessons should be learned from the two incidents mentioned in Germany. The attack on xs4all’s connectivity for carrying a leftwing magazine was received with hostility and resistance: countries attacking connectivity itself are viewed with similar dislike as individual vandals doing the same. And, although CompuServe reacted passively to the first batch of extra-legal threats, a recent Reuters report says they are now threatening to remove their operation from Germany. We must remember that the Internet allows competition in a worldwide market. There will always be somebody else outside unreasonable restrictions. It could be that little would remain inside except the access and the consumers, while the employment and profit generation has moved to a more congenial climate. If in that case attempts were made to isolate the EU, then either they would fail anyway, or they would succeed in making the EU a backwater on the world scene cut off from major developments elsewhere.

It is unlikely that any one piece of news, information or imagery can be made to no longer exist. In the case of child pornography, offers to sell illegal products, etc, the most productive activity for law enforcement is to watch anyone foolish enough to access it and take copies or order the illegal goods, and prosecute them under local law for actual offences. The public should be aware that laws (and the extent of enforcement) vary from place to place, and just because someone can talk about things which are unlawful for you locally does not mean you can get involved in unlawful acts locally. It was seen in UK that confidential material, pages messages, cannot be left unlocked and dealt with by prior threat. First, people must take proactive steps to protect confidences with strong encryption. Second, without abandoning prior restraint entirely, it has to be recognised that you can’t put information back in the bottle once it’s out, and the most productive action is to concentrate on penalties after publication---particularly in attacking profits made from any unlawful use of others’ copyright.

Law enforcement works best based on consent of the community involved. Attempts by large state or private organisation to suppress news of discreditable activity are met with resistance. On the other hand, the good sense of the Internet community can be trusted in that pretty much everyone intensely dislikes child porn, heroin importing, etc, and can be relied on themselves to cooperate against such things. This trust can, however, be entirely lost----

recent attempts to approach people with blackmail threats that they will be under new penalties, raids, etc, if they do not cooperate in (e.g.) compulsory rating of all lawful material will simply damage cooperation and good will.

{D}ANONYMOUS FREE SPEECH
If there is any free speech at all then it means free speech where one is not in fear of unreasonable penalty for speaking out. This includes holding a political or religious viewpoint of any stripe which may put you at risk of discrimination or penalty if openly known---at worst, the threat is of murder for dissenting against certain powerful religions. It also includes risk of persecution by employers on releasing information which may mean, at worst, people are likely to be killed, maimed or suffer serious loss as a result of its continued secrecy. Anonymous speech is therefore vital to any true free speech right.

Nobody will trust encryption where any stranger has the key, and nobody will trust anonymity where there are records of each message (or pseudonymous forwarding where the forwarding information can be decrypted locally). The only trustworthy option is that there is no record of the source of messages, or of the identity of pseudonyms, to be attacked in the first place. The most likely response if this is made impossible locally is that outside countries will offer it--for a small dollar fee per year--and be quite resourceful at evading any attempt to block them.

We have three suggestions on anonymity:

• OPTION I. Make no mistake, we believe that true anonymity is essential to free speech, and nothing else can substitute.
• OPTION II. If you will not accept this, then we think orders to reveal anonymous messages should be made only by the highest courts of each nation, and only where there is a strong likelihood that there will be a criminal conviction before a jury resulting in a sentence of at least five years imprisonment. The data protection laws should not only force the remailers to log addresses, but impose a large fine and/or jail terms for anyone breaking the confidentiality without such an order. There should be no civil access--however large the amount involved--and all such orders should be reported annually to the national parliament.
• OPTION III. If you really went worse than this, under severe protest, then we suggest the following. (1) Such orders would still be made only by the highest courts, and only when there was a strong likelihood of success for long imprisonment or for large sums of money [or other exceptionally serious reasons] in civil matters. The anonymous person would have the right to be heard first, which could be in writing and/or by representative if necessary. (2) It would be an over-riding reason to refuse the order if [i] the disputed matter is of an essentially religious or political nature, or [ii] the material involved concerns public safety or matters in which the public have an interest or is itself a revelation of serious criminal or unlawful acts, or [iii] the anonymous person is in genuine fear of any unlawful harm or injury. he state would also indemnify them in full against any unlawful or unreasonable harm which resulted to them (in civil matters then suing the original plaintiff to recover the cost), which could be a large sum if they were killed or injured as a result of exposure.

None of this would be any barrier to a publisher of commercial works finding out who caused him severe loss of revenue by circulating his copyright material.

{B}DISTASTEFUL MATERIAL.
If the description that e.g. obscenity is ''harmful'' to either adults or minors, then diligent research should be able to discover consistently that and how a particular sort of harm is occurring by some particular mechanism--and do so in increasing detail with increasing examination. This is not the case. A better description would therefore be ''distasteful''. The same goes for offensive racist material, unless it is the case that particular material has provably incited a particular act when criminal law applies.

We do not believe adults should have to reate their material so children can have access to it or read only material fof given ratings, unless they freely and individually chose to. In the case of children, there is a public will that they are not yet able to make fully informed choices so parents or teachers have every right to limit what distasteful material they can access. We obviously support freedom of expression in this area; and would like to see a single market given that individual states are never likely to agree on whether a particular level of ''obscenity'', or whether holocaust denial, should actually be criminal.

FILTERING SYSTEMS.
There are a number of questions about such systems which your categorisation does not fully cover.

• (i) Is it a blacklist which allows everything except specifically known ''bad'' material to be viewed, or a whitelist which allows only specifically known ''good'' material to be views and blocks everything else---the former provides more sites, the latter more certainty.
• (ii) Does it have categories other than just a single yes/no, and does if work on a graduated scale: these give the recipients more flexibility about what they allow onto their systems.
• (iii) Is it humanly filtered? Software filters can only handle text not images, and mistake the UK town of ''Scunthorpe'' for a rude word---but human filtering is labour intensive.
• (iv) Does the censor approve individual sites, or only set the criteria by which others generate their own ratings subject to verification. This would be humanly checked, allow several categories with graduated scales, and be effectively a whitelist [sites which chose not to use it would be automatically blocked from any system set up to accept only certain levels so it was suitable for children].

THE VOLUNTARY PRINCIPLE.
There are some obvious difficulties with ratings systems originally designed for children’s computer games. One is that they fail to cope with a webpage as a changing medium---some users might be changing a couple of files most days of the week, as they log on to collect their newsgroups. A fixed and registered rating is not flexible enough to cope with this. Another is the inherent entertainment bias. Sex or violence can only be added to a computer game as a gratuitous choice to add spice: rating systems do not cope with the idea that it is inherent to the theme of a work of art, or coolly and factually reported as what in fact happened in a news item (and it is difficult to see how they could).

At least in the UK we have newspapers which--within the law and certain broad limits of taste--will report an interview containing swearwords or show partial nudity or, of course, report such incidents as sexual murders. On the other hand we have children’s comics which are carefully designed for different agegroups. The newspapers do not say that today they are suitable for 9 year olds re swearing and 12 year olds re nudity: it would be an absurd and disproportionate measure to make them rate their material on its suitability for young children. Some discourse (commercial material aimed at the whole family) obviously plays very safe, other discourse (academic analysis or news reports) would consider it an unwarranted intrusion to be rated or messed around in this way.

We are therefore disturbed at statement by a UK ISP, possibly under outside pressures, that they will compel a ''voluntary'' rating system on their users. All users, even most of those who argue that voluntary ratings are a good idea, are universally disgusted with the idea of compulsion; the resulting argument and resistance would probably make the whole thing a complete mess anyway. We would appreciate a statement that there will be no universal compulsion to put children’s ratings on all pages even those who wish to be unrated & inaccessible to minors. There is also the precedent of misleadingly named British Board of Film Classification---which does not just make advisory classifications on which user choice can be based, but is now a censorship system which can block cinema or video distribution by refusing a rating. This understandably makes people suspicious of any new such scheme.

WHICH RATINGS SYSTEM?
If a ratings system is to be used, then it should be roughly of the above type based on PICS. We agree with the communiqué that, while everyone might share a system with particular categories and grades, the rating should be done locally by each state or its large ISPs reflecting local standards as to what is acceptable in e.g. swearing.

The system should not be proprietary, but by having a ''club'' which accept each others encrypted signatures (or even share a small pool of signatures) as ''reliable'' and recommended to their users. The system should be designed specifically for webpages, and the questions should be very basic and quick to answer e.g. no more than one page of A4 to explain four grades for each of four categories. The system should by locally administered and accountable, and all altering of ratings or exclusion from the ratings scheme publicly reported on usenet to ensure fair play.

The American RSACi scheme has implicitly and explicitly been criticised in the communiqué and elsewhere for several reasons -- it is proprietary, not local, not accountable, the questionnaire is very complex, etcetera. We would appreciate a specific recommendation that RSACi not be used, again in view of suggestions that it be prematurely introduced in the UK and then leave us incompatible with other EU states.Because adults can in certain circumstances harm minors by excluding material, special low ratings should apply for child abuse help (so it cannot be blocked e.g. by an abuser) and AIDS information (so it cannot be blocked by adults with an odd attitude to sex).

.

.

.

.