Online Safety Act 2023 and Usenet News

[Clive Page]

I am surprised not to have seen any discussion here, or indeed anywhere, of the possible effects of the UK's Online Safety Act 2023 on Usenet News. I see that some small online forums are already closing down as they don't think they can comply with the Act, see for example:

https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-one-speed-doesnt-fit-all-when-it-comes-to-online-safety

which describes the closure of the London Fixed Gear and Single-Speed forum. I expect others will follow, which is a great shame.

I have not read the Act itself, just some official explainers such as this one:
https://www.gov.uk/government/publications/online-safety-act-explainer/online-safety-act-explainer

From this it looks to me as if Usenet News would be subject to the provisions of the Act but it is not at all clear who would termed the "provider" in this case. I get my news feed from the servers of the Technical University of Berlin (although other fine Usenet servers are available, for the moment). I very much hope that they don't feel that they are a provider of news in the sense covered by the act. In the case of a moderated group like this one, I also hope that the moderators don't feel that it affects them either.

The legislation was obviously aimed at the big boys like Facebook and X/Twitter, but seems to have been drafted so widely that it will have adverse side-effects perhaps quite widely. It may be that eventually case law will show the limits of the applicability of the Act but there may be damage in the mean time.

[Spike]

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere,
of the possible effects of the UK's Online Safety Act 2023 on Usenet
News. I see that some small online forums are already closing down as
https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-one-speed-doesnt-fit-all-when-it-comes-to-online-safety
which describes the closure of the London Fixed Gear and Single-Speed
forum. I expect others will follow, which is a great shame.
https://www.gov.uk/government/publications/online-safety-act-explainer/online-safety-act-explainer
From this it looks to me as if Usenet News would be subject to the
provisions of the Act but it is not at all clear who would termed the
"provider" in this case. I get my news feed from the servers of the
Technical University of Berlin (although other fine Usenet servers are
available, for the moment). I very much hope that they don't feel that
they are a provider of news in the sense covered by the act. In the case
of a moderated group like this one, I also hope that the moderators don't
feel that it affects them either.
The legislation was obviously aimed at the big boys like Facebook and
X/Twitter, but seems to have been drafted so widely that it will have
adverse side-effects perhaps quite widely. It may be that eventually
case law will show the limits of the applicability of the Act but there
may be damage in the mean time.

I can’t comment on the OSA 23 but as far as wide-drafting is concerned I do
recall the Football (Disorder) Act banning face coverings, and which was
used against people who were not football fans, not at a football match,
and not travelling to or from a football facility. ‘The government’ in some
way, shape, or form didn’t like the people concerned, and used the handy
provisions of the FDA against them.

So there will be no surprises if Usenet is caught in some way by the OSA
23. Enjoy it while it lasts, perhaps.

[Jethro_uk]

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere,
of the possible effects of the UK's Online Safety Act 2023 on Usenet
News. I see that some small online forums are already closing down as
https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-one-

speed-doesnt-fit-all-when-it-comes-to-online-safety

Post by Clive Page
which describes the closure of the London Fixed Gear and Single-Speed
forum. I expect others will follow, which is a great shame.
https://www.gov.uk/government/publications/online-safety-act-explainer/

online-safety-act-explainer

Post by Clive Page
From this it looks to me as if Usenet News would be subject to the
provisions of the Act but it is not at all clear who would termed the
"provider" in this case. I get my news feed from the servers of the
Technical University of Berlin (although other fine Usenet servers are
available, for the moment). I very much hope that they don't feel that
they are a provider of news in the sense covered by the act. In the
case of a moderated group like this one, I also hope that the
moderators don't feel that it affects them either.
The legislation was obviously aimed at the big boys like Facebook and
X/Twitter, but seems to have been drafted so widely that it will have
adverse side-effects perhaps quite widely. It may be that eventually
case law will show the limits of the applicability of the Act but there
may be damage in the mean time.

What will the UK be doing about servers hosted outside it's jurisdiction
and accessed by VPN ?

[Nick Odell]

On Tue, 14 Jan 2025 14:17:03 -0000 (UTC), Jethro_uk

Post by Clive Page

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere,
of the possible effects of the UK's Online Safety Act 2023 on Usenet
News. I see that some small online forums are already closing down as
https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-one-

speed-doesnt-fit-all-when-it-comes-to-online-safety

Post by Clive Page
which describes the closure of the London Fixed Gear and Single-Speed
forum. I expect others will follow, which is a great shame.
https://www.gov.uk/government/publications/online-safety-act-explainer/

online-safety-act-explainer

Post by Clive Page
From this it looks to me as if Usenet News would be subject to the
provisions of the Act but it is not at all clear who would termed the
"provider" in this case. I get my news feed from the servers of the
Technical University of Berlin (although other fine Usenet servers are
available, for the moment). I very much hope that they don't feel that
they are a provider of news in the sense covered by the act. In the
case of a moderated group like this one, I also hope that the
moderators don't feel that it affects them either.
The legislation was obviously aimed at the big boys like Facebook and
X/Twitter, but seems to have been drafted so widely that it will have
adverse side-effects perhaps quite widely. It may be that eventually
case law will show the limits of the applicability of the Act but there
may be damage in the mean time.

What will the UK be doing about servers hosted outside it's jurisdiction
and accessed by VPN ?

Some countries have banned the use of VPNs and others only permit the
use of government-approved ones.

https://nordvpn.com/blog/are-vpns-legal/

Nick

[Jethro_uk]

Post by Nick Odell
On Tue, 14 Jan 2025 14:17:03 -0000 (UTC), Jethro_uk

Post by Clive Page

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere,
of the possible effects of the UK's Online Safety Act 2023 on Usenet
News. I see that some small online forums are already closing down as
https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-

one-

Post by Nick Odell

Post by Clive Page
speed-doesnt-fit-all-when-it-comes-to-online-safety

Post by Clive Page
which describes the closure of the London Fixed Gear and Single-Speed
forum. I expect others will follow, which is a great shame.
https://www.gov.uk/government/publications/online-safety-act-

explainer/

Post by Nick Odell

Post by Clive Page
online-safety-act-explainer

Post by Clive Page
From this it looks to me as if Usenet News would be subject to the
provisions of the Act but it is not at all clear who would termed the
"provider" in this case. I get my news feed from the servers of the
Technical University of Berlin (although other fine Usenet servers
are available, for the moment). I very much hope that they don't
feel that they are a provider of news in the sense covered by the
act. In the case of a moderated group like this one, I also hope
that the moderators don't feel that it affects them either.
The legislation was obviously aimed at the big boys like Facebook and
X/Twitter, but seems to have been drafted so widely that it will have
adverse side-effects perhaps quite widely. It may be that eventually
case law will show the limits of the applicability of the Act but
there may be damage in the mean time.

What will the UK be doing about servers hosted outside it's jurisdiction
and accessed by VPN ?

Some countries have banned the use of VPNs and others only permit the
use of government-approved ones.
https://nordvpn.com/blog/are-vpns-legal/

That won't fly in the UK. Not if it's serious about wanting businesses to
operate here.

[Max Demian]

Post by Nick Odell
On Tue, 14 Jan 2025 14:17:03 -0000 (UTC), Jethro_uk

Post by Jethro_uk

Post by Clive Page
The legislation was obviously aimed at the big boys like Facebook and
X/Twitter, but seems to have been drafted so widely that it will have
adverse side-effects perhaps quite widely. It may be that eventually
case law will show the limits of the applicability of the Act but there
may be damage in the mean time.

What will the UK be doing about servers hosted outside it's jurisdiction
and accessed by VPN ?

Some countries have banned the use of VPNs and others only permit the
use of government-approved ones.

Facebook banned. X banned. Twitter banned.

Everyone will have to migrate to the "dark web" to do anything.

[Fredxx]

Post by Max Demian

Post by Nick Odell
On Tue, 14 Jan 2025 14:17:03 -0000 (UTC), Jethro_uk

Post by Jethro_uk

Post by Clive Page
The legislation was obviously aimed at the big boys like Facebook and
X/Twitter, but seems to have been drafted so widely that it will have
adverse side-effects perhaps quite widely.  It may be that eventually
case law will show the limits of the applicability of the Act but there
may be damage in the mean time.

What will the UK be doing about servers hosted outside it's jurisdiction
and accessed by VPN ?

Some countries have banned the use of VPNs and others only permit the
use of government-approved ones.

Facebook banned. X banned. Twitter banned.

That would save huge police resources policing media posts. Where if you
make a complaint for example about a post an ex has made it's an instant
admission of a criminal offence and subsequent prosecution. What's not
to like.

Post by Max Demian
Everyone will have to migrate to the "dark web" to do anything.

[Roland Perry]

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere,
of the possible effects of the UK's Online Safety Act 2023 on Usenet News.

There was some discussion, possibly here, regarding a precursor to this
legislation. I decided that I'd retried, and absent anyone funding me to
campaign on this specific issue, there are plenty of civil society
pressure groups who should have stepped up. Did they not?

[billy bookcase]

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere, of the possible
effects of the UK's Online Safety Act 2023 on Usenet News. I see that some small online
forums are already closing down as they don't think they can comply with the Act, see
https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-one-speed-doesnt-fit-all-when-it-comes-to-online-safety
which describes the closure of the London Fixed Gear and Single-Speed forum. I expect
others will follow, which is a great shame.
https://www.gov.uk/government/publications/online-safety-act-explainer/online-safety-act-explainer
From this it looks to me as if Usenet News would be subject to the provisions of the
Act but it is not at all clear who would termed the "provider" in this case. I get my
news feed from the servers of the Technical University of Berlin (although other fine
Usenet servers are available, for the moment). I very much hope that they don't feel
that they are a provider of news in the sense covered by the act. In the case of a
moderated group like this one, I also hope that the moderators don't feel that it
affects them either.
The legislation was obviously aimed at the big boys like Facebook and X/Twitter, but
seems to have been drafted so widely that it will have adverse side-effects perhaps
quite widely. It may be that eventually case law will show the limits of the
applicability of the Act but there may be damage in the mean time.

All such legislation is only affective as are *the resources* and the
determination of those tasked with actually enforcing it. Which in this case
would appear to be Offcom. Mainly responding to complaints (?)

Assuming that is that is that it isn't part of the legislation that the industry
itself is being required to fund the extra expenditure involved.

Otherwise its simply another case of First World Aspirations meeting the
harsh realities of a Third World Economy.

quote:

We are also suggesting particular documents different stakeholder might be
interested in:

1.. Small/medium sized service providers: may want to start with our digital tool when
available
2.. and review our summary documents where appropriate, before turning to our risk
assessment
3.. guidance, followed by our Codes (supplemented by our other guidance documents).
unquote

https://www.ofcom.org.uk/online-safety/illegal-and-harmful-content/statement-protecting-people-from-illegal-harms-online/

"our digital tool when available" ; "risk assessment" ; "codes"; blah blah blah.

I would humbly suggest that by the time Musk, Zuckerberg, et al have succeeded
in tying up their legal team (all 10 of them) in knots, there simply won't be the
manpower to police or adjudicate on cycling or any other websites hosting discussions.
Which isn't to say such groups might not be spooked into oblivion by opportunistic
legal firms offering superfluous advice.


bb

[Roland Perry]

Post by billy bookcase
All such legislation is only affective as are *the resources* and the
determination of those tasked with actually enforcing it. Which in this case
would appear to be Offcom. Mainly responding to complaints (?)
Assuming that is that is that it isn't part of the legislation that the industry
itself is being required to fund the extra expenditure involved.

OFCOM is partly funded by fees paid by organisations they regulate. Back
in the day I was involved in the discussion to decide which ISPs should
be caught in that net, and I think the answer was "turnover greater than
5 million a year". There's a very long tail of smaller organisations.

A corollary of this is that they'd be very unlikely to take any
enforcement action against anyone in that long tail. There just
aren't that many hours in the day. Especially when such an organisation
is one they've never heard of, isn't registered with them, and so on.

[billy bookcase]

Post by Clive Page
I am surprised not to have seen any discussion here, or indeed anywhere, of the possible
effects of the UK's Online Safety Act 2023 on Usenet News. I see that some small online
forums are already closing down as they don't think they can comply with the Act, see
https://www.theguardian.com/commentisfree/2025/jan/12/note-to-no-10-one-speed-doesnt-fit-all-when-it-comes-to-online-safety
which describes the closure of the London Fixed Gear and Single-Speed forum. I expect
others will follow, which is a great shame.
https://www.gov.uk/government/publications/online-safety-act-explainer/online-safety-act-explainer
From this it looks to me as if Usenet News would be subject to the provisions of the
Act but it is not at all clear who would termed the "provider" in this case. I get my
news feed from the servers of the Technical University of Berlin (although other fine
Usenet servers are available, for the moment). I very much hope that they don't feel
that they are a provider of news in the sense covered by the act. In the case of a
moderated group like this one, I also hope that the moderators don't feel that it
affects them either.
The legislation was obviously aimed at the big boys like Facebook and X/Twitter, but
seems to have been drafted so widely that it will have adverse side-effects perhaps
quite widely. It may be that eventually case law will show the limits of the
applicability of the Act but there may be damage in the mean time.

But is there anything in this legislation which is any more onerous that
are say the existing legal provisions on say, defamation ? I've often read
local forums where statements are made which are clearly defamatory, both of
public figures, local politicians, and other contributors.

This has never seemed to bother the people who host around ten such forums,
around three of which are very active. In one instance in response to a
specific complaint and threat of legal action, an apology was made to a
fellow poster. But other than that, these forums appear to be totally
unmoderated; except possibly for key word filters which block posts
containing instances of profanity swearing etc.

So that would the consequences of breaking any of these Offcom codes be any more
damaging than would the possible consequences publishing defamatory material
as at present? Should it come to anyone's notice in the first place, that is.

bb

[Mark Goodge]

Post by billy bookcase
But is there anything in this legislation which is any more onerous that
are say the existing legal provisions on say, defamation ? I've often read
local forums where statements are made which are clearly defamatory, both of
public figures, local politicians, and other contributors.
This has never seemed to bother the people who host around ten such forums,
around three of which are very active. In one instance in response to a
specific complaint and threat of legal action, an apology was made to a
fellow poster. But other than that, these forums appear to be totally
unmoderated; except possibly for key word filters which block posts
containing instances of profanity swearing etc.
So that would the consequences of breaking any of these Offcom codes be any more
damaging than would the possible consequences publishing defamatory material
as at present? Should it come to anyone's notice in the first place, that is.

The main difference is that defamation is a civil matter, and relies on the
victim to take action against the defamer if they feel it's justified. The
Online Safety Act, on the other hand, creates a set of criminal offences
which can be enforced without the need for a rich victim to fund the action.

In practice, of course, a lot of activity which is theoretically unlawful
under the Act will be below the threshhold for the regulators to care about.
But it may take a while before we have any clarity about where those lines
in the sand will be drawn.

Mark

[Jethro_uk]

Post by Mark Goodge
But it may take a while before we have any clarity about where those
lines in the sand will be drawn.

There's nothing like a vague and arbitrarily enforced law to put the fear
of god into innocent people.

[Spike]

Post by Jethro_uk

Post by Mark Goodge
But it may take a while before we have any clarity about where those
lines in the sand will be drawn.

There's nothing like a vague and arbitrarily enforced law to put the fear
of god into innocent people.

It’s a standard method used by repressive governments to control people.
The Covid regulations and guidance were examples, enthusiastically taken to
the outer fringes of reality by the police, for example where two friends
were having a coffee in a park, which was wrongly interpreted by England’s
Finest as a picnic. People’s caution when parts of the spectrum of artistic
hues are used as descriptors is another.

[Colin Macleod]

Post by Jethro_uk

Post by Mark Goodge
But it may take a while before we have any clarity about where those
lines in the sand will be drawn.

There's nothing like a vague and arbitrarily enforced law to put the fear
of god into innocent people.

This is my concern.

I run a web gateway to Usenet at https://newsgrouper.org.uk , which connects
to the eternal-september.org server in Germany. Newsgrouper is just a one-man
hobby project (I'm a retired programmer). I get around 20 users per day,
about 1/4 of those are within the UK. I allow anyone to connect as a guest
and read. To be able to post I require users to register by sending me an
email. This means I could remove their access if they abuse it by spamming
etc. But the only information I record is that email address.

Eternal-september doesn't carry binary newsgroups, so images and video are not
a concern. I could set up a system that would enable me to block specific
posts or entire groups if I was required to do so. But I certainly don't have
the capacity or willingness to undertake any kind of proactive
monitoring/moderation of posts.

My feeling is that my site is very low risk. But the information I've read
so far about the OSA tells me I need to provide *evidence* under lots of
different categories to *prove* that absence of risk. I don't see how I can
do that. E.g. I would be amazed if any children took an interest in my site,
but I can't prove that this is impossible.

Now it may be that Ofcom will be far too busy with the "big fish" to ever
take the slightest notice of my little site. But I'm anxious about the
possibility that they could, and have started thinking that the safest
course might be to just block access to users within the UK, though I would
certainly not like to do that.

[Mark Goodge]

On Tue, 21 Jan 2025 11:10:04 GMT, Colin Macleod

Post by Colin Macleod
My feeling is that my site is very low risk. But the information I've read
so far about the OSA tells me I need to provide *evidence* under lots of
different categories to *prove* that absence of risk. I don't see how I can
do that. E.g. I would be amazed if any children took an interest in my site,
but I can't prove that this is impossible.

For some of the categories, the evidence you need to provide is the simple
fact that your site cannot carry that kind of material - the risk will
therefore be negligable for every category of image content.

For textual content, the risk isn't really related to children anyway
(that's mostly to do with porn, which for you isn't an issue). So the key
evidence here will be an assessment of how often material of that nature has
ever been on your server in the past. Again, that's likely to be close to
zero, so for these categories, the risk will be low (it can't be negligible,
because, unlike images, it's not impossible).

So provided you can tick off a risk assessment (and Ofcom have promised to
provide sample templates you can just fill in) and show that your risks are
low or negligible for every category (which they will be), then the only
mitigation you will need to take is to ensure that people can contact you if
they see something on your server that shouldn't be there. Given the small
number of users, an email address or contact form will be sufficient.

Post by Colin Macleod
Now it may be that Ofcom will be far too busy with the "big fish" to ever
take the slightest notice of my little site.

It's almost certain that they will never even know about your site unless
someone complains to them about it. And even then, their response will
almost certainly to tell you to get your act together rather than take any
enforcement action.

Post by Colin Macleod
But I'm anxious about the
possibility that they could, and have started thinking that the safest
course might be to just block access to users within the UK, though I would
certainly not like to do that.

That won't help anyway, because you are in the UK so that's a relevant
connection even if one of your users are.

Mark

[Colin Macleod]

Thanks for the feedback.

[Mark Goodge]

On Tue, 14 Jan 2025 12:46:23 +0000, Clive Page <***@page2.eu> wrote:

[On a technical note, there's something seriously wrong with your Usenet
client - it's not wrapping text correctly. I see from the headers that
you're using Thunderbird, which normally gets it right, so you might want to
investigate some of your settings]

Post by Clive Page
https://www.gov.uk/government/publications/online-safety-act-explainer/online-safety-act-explainer
From this it looks to me as if Usenet News would be subject to the provisions
of the Act but it is not at all clear who would termed the "provider" in this
case. I get my news feed from the servers of the Technical University of Berlin
(although other fine Usenet servers are available, for the moment). I very much
hope that they don't feel that they are a provider of news in the sense covered
by the act. In the case of a moderated group like this one, I also hope that
the moderators don't feel that it affects them either.

I'm pretty sure that Usenet would be in scope. It is, after all, the
original U2U service that all the others have copied ideas from. But, for a
text-only service, the risk profile is going to be very low - you're not
going to get porn here, for example.

So, for most providers, compliance is likely to be little more than a tick
box exercise - you merely need to have a risk assessment which shows that
the risk is very low for all the priority categories of content. Usenet
server operators are already legally responsible for material on their
servers - Godfrey v Demon established that some time ago - and the new law
doesn't change that. Usenet also doesn't have any privacy issues which
affect the likes of Microcosm, for example. Everything posted here is
already public, there are no private messages and no end-to-end encryption.
The risks here are, genuinely, low or negligible for almost all the priority
categories, and can easily be shown to be low or negligable.

Ofcom's guidance does, on the face of it, create a lot of bureaucracy for
service providers. You have to read, and understand, the guidance, most of
which will not apply to you but nonetheless needs to be fully comprehended
in order to know what does or does not apply. You have to create, and
maintain, a risk assessment covering at least 17 different categories of
content. You have to have a documented process for allowing users to report
unlawful content. You have to have a documented process for dealing with
reports of unlawful content. And you have to keep all of this documentation
up to date, which means reviewing it regularly.

At least initially, putting this documentation together will be burdonsome
(although once done, annual reviews should be relatively trivial unless the
service itself has changed significantly). And this is the kind of work that
most people whose job is primarily technical (eg, system administration)
will be unfamiliar with. So it will be a challenge. But it's not an
insuperable one.

More generally, the prospect of Ofcom pursuing enforcement against Usenet
and other small, low-risk operators is, I think, close to negligible in
itself. They simply don't have the resources to chase down every small
provider of UGC unless there is evidence that the service is actually being
used to distribute unlawful material. Which, for the vast majority of them,
won't be the case.

Mark

[Jethro_uk]

Post by Mark Goodge
I'm pretty sure that Usenet would be in scope. It is, after all, the
original U2U service that all the others have copied ideas from. But,
for a text-only service, the risk profile is going to be very low -
you're not going to get porn here, for example.

binary newsgroups ?

[Mark Goodge]

On Fri, 17 Jan 2025 18:04:18 -0000 (UTC), Jethro_uk

Post by Jethro_uk

Post by Mark Goodge
I'm pretty sure that Usenet would be in scope. It is, after all, the
original U2U service that all the others have copied ideas from. But,
for a text-only service, the risk profile is going to be very low -
you're not going to get porn here, for example.

binary newsgroups ?

The only places to get those these days are the commercial providers such as
Giganews. They will be able to afford the compliance costs. And, by now,
they will already have plenty of experience dealing with the legal of
ramifications of binaries, given that copyright infringement is one of their
key USPs.

Mark

[Andy Burns]

Post by Mark Goodge
[On a technical note, there's something seriously wrong with your Usenet
client - it's not wrapping text correctly. I see from the headers that
you're using Thunderbird, which normally gets it right, so you might want to
investigate some of your settings]

Thunderbird does get a few things wrong ...

I think it uses lone LFs for line endings instead of CRLF.

It can get fooled into changing multiple spaces into NBSP, maybe lhis
line will trigger it .

It very easily flips into base64 encoding whenever it
sees non-ASCII

But as far as header fields go, Clive's seem the same as mine, but his
TB isn't breaking long lines at all?

[Mark Goodge]

Post by Andy Burns

Post by Mark Goodge
[On a technical note, there's something seriously wrong with your Usenet
client - it's not wrapping text correctly. I see from the headers that
you're using Thunderbird, which normally gets it right, so you might want to
investigate some of your settings]

Thunderbird does get a few things wrong ...
I think it uses lone LFs for line endings instead of CRLF.
It can get fooled into changing multiple spaces into NBSP, maybe lhis
line will trigger it .
It very easily flips into base64 encoding whenever it
sees non-ASCII
But as far as header fields go, Clive's seem the same as mine, but his
TB isn't breaking long lines at all?

The headers say that it's Content-Type: text/plain; charset=UTF-8;
format=flowed. But there aren't any soft line breaks in there, only hard
breaks. And the hard breaks only appear at the end of paragraphs. So no, it
isn't breaking lines at all, even where it should be.

Your post, though, does have soft breaks in it, which my client is correctly
interpreting. I have no idea why they are different.

Mark

[Andy Burns]

Post by Mark Goodge
The headers say that it's Content-Type: text/plain; charset=UTF-8;
format=flowed. But there aren't any soft line breaks in there, only hard
breaks. And the hard breaks only appear at the end of paragraphs. So no, it
isn't breaking lines at all, even where it should be.

@Clive

one setting I've changed from default
mail.strictly_mime_headers=false

defaults I've left alone
plain_text.wrap_long_lines=true
mailnews.wraplength=72

[Clive Page]

Post by Andy Burns

Post by Mark Goodge
The headers say that it's Content-Type: text/plain; charset=UTF-8;
format=flowed. But there aren't any soft line breaks in there, only hard
breaks. And the hard breaks only appear at the end of paragraphs. So no, it
isn't breaking lines at all, even where it should be.

@Clive
one setting I've changed from default
mail.strictly_mime_headers=false
defaults I've left alone
plain_text.wrap_long_lines=true
mailnews.wraplength=72

Thanks to both of you. My setting had mailnews.wraplength=0 and I've
now changed it to 72. I have no idea how I acquired that setting. I
had looked at a few likely culprits but there are so many config
settings it's hard to work out what does what. Hope it is better now.
It does at least seem to be wrapping lines as I type them.

[Mark Goodge]

Post by Clive Page

Post by Andy Burns

Post by Mark Goodge
The headers say that it's Content-Type: text/plain; charset=UTF-8;
format=flowed. But there aren't any soft line breaks in there, only hard
breaks. And the hard breaks only appear at the end of paragraphs. So no, it
isn't breaking lines at all, even where it should be.

@Clive
one setting I've changed from default
mail.strictly_mime_headers=false
defaults I've left alone
plain_text.wrap_long_lines=true
mailnews.wraplength=72

Thanks to both of you. My setting had mailnews.wraplength=0 and I've
now changed it to 72.

And that's fixed it.

Thunderbird's settings aren't always intuitive. That's partly because, like
a lot of open source software, different functionality is added at different
times by different people with different views on how it should be
configured. But once you can penetrate the options controls (and disregard
the leopards), you can customise almost every aspect of its behaviour.

Mark

[Jethro_uk]

And secondly while I am using a text only server, less than 1% of Usenet
traffic is text, most of it being various media, much of which is said
to be in breach of copyright if nothing else.

You say that in a country that bans cannabis and permits alcohol.

[Mark Goodge]

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

And secondly while I am using a text only server, less than 1% of Usenet
traffic is text, most of it being various media, much of which is said to be
in breach of copyright if nothing else.

None of the free, or cheap, Usenet providers carry binaries these days. Even
prior to the Online Safety Act, the propensity of rightsholders to take
action against those who facilitate infringement has made it impractical for
all but the most well-funded commercial providers. Who will be equally well
positioned to ensure compliance with the Act.

Mark

[Roger Hayter]

On 17 Jan 2025 at 20:24:08 GMT, "Mark Goodge"

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

Indeed, but there was a lot of speculation, perhaps based on wishful thinking
or American law, that had Demon taken the offending post down at Mr Godfrey's
request they might have had a defence. Indeed, a test case on those lines
would have been interesting.

Post by Mark Goodge

And secondly while I am using a text only server, less than 1% of Usenet
traffic is text, most of it being various media, much of which is said to be
in breach of copyright if nothing else.

None of the free, or cheap, Usenet providers carry binaries these days. Even
prior to the Online Safety Act, the propensity of rightsholders to take
action against those who facilitate infringement has made it impractical for
all but the most well-funded commercial providers. Who will be equally well
positioned to ensure compliance with the Act.
Mark

[Mark Goodge]

Post by Roger Hayter
On 17 Jan 2025 at 20:24:08 GMT, "Mark Goodge"

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

Indeed, but there was a lot of speculation, perhaps based on wishful thinking
or American law, that had Demon taken the offending post down at Mr Godfrey's
request they might have had a defence. Indeed, a test case on those lines
would have been interesting.

It would have been a defence had they taken down the post on request, yes.
But to do so would require them to accept that the post was defamatory and
that they were liable for it.

More specifically, there is an established principle in defamation law that
if a publisher of the defamatory material is not the author (or editor) of
the material, and has published it in good faith having no reason to believe
that it was defamatory, then if they make an "offer of amends" as soon as
practical on being notified that the material is defamatory, then that will
either forestall action entirely (if the offer is accepted) or provides a
defence (if the offer is not accepted, but the court later decides that it
should have been accepted).

In most cases of "innocent dissemination", an offer of amends that would be
deemed acceptable by a court would not involve any monetary compensation.
Typically, an acceptable offer of amends includes an apology and the
retraction of the defamatory material if possible (eg, deleting it from a
web page or a Usenet server) or, if that isn't possible (eg, for a printed
publication that has already been distributed), the publication of a
subsequent notice of correction that is at least as prominent as the
original defamatory material.

Defamation law isn't intended to punish innocent publishers merely for
publishing something that's later determined to be defamatory, but simply to
mitigate the effects of the defamation on its subject. Monetary compensation
will normally only be awarded by a court against the author of the material
or a non-innocent publisher (that is, someone who publishes, or continues to
publish, the material knowing it is, or is likely to be, defamatory).

In the case of Godfrey v Demon, therefore, had Demon responded by saying
something along the lines of "Yes, sorry, that's clearly defamatory, we've
deleted it as requested" then Godfrey would have had no grounds for action.
Demon's refusal to delete the post moved them into the non-innocent
category; at that point, they were knowingly publishing material that they
knew, or had reason to believe, was defamatory. That was their mistake.

Mark

[Tim Jackson]

On Fri, 17 Jan 2025 20:24:08 +0000, Mark Goodge wrote...

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

I suspect Godfrey v Demon was overtaken shortly afterwards by Regulation
19 (Hosting) of the Electronic Commerce Regulations 2002 (implementing
the InfoSoc Directive).
https://www.legislation.gov.uk/uksi/2002/2013/regulation/19

There might perhaps be an issue as to who is the "recipient" of a
distributed Usenet hosting service: the poster, the viewer, or both? (I
suspect both.)

Post by Mark Goodge

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

Under Regulation 19 it would be failing to take it down when notified.

[Mark Goodge]

Post by Tim Jackson
On Fri, 17 Jan 2025 20:24:08 +0000, Mark Goodge wrote...

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

I suspect Godfrey v Demon was overtaken shortly afterwards by Regulation
19 (Hosting) of the Electronic Commerce Regulations 2002 (implementing
the InfoSoc Directive).
https://www.legislation.gov.uk/uksi/2002/2013/regulation/19

As far as defamation is concerned, that really just reiterates the existing
innocent dissemination principle.

Post by Tim Jackson
There might perhaps be an issue as to who is the "recipient" of a
distributed Usenet hosting service: the poster, the viewer, or both? (I
suspect both.)

That regulation starts off by referring to "the storage of information
provided by a recipient of the service", which would imply that the
recipient of the service is the poster, not the viewer. Of course, there can
be multiple services involved in hosting UGC, but the one that's relevant
here is the service provided to the supplier of information which may later
be determined to be unlawful.

This isn't quite a "mere conduit" defence, because it does make the operator
liable if they don't act on notifications. But it does insulate them from
any action taken related to material of which they could not reasonably have
been aware.

(Mere conduit is covered in regulation 17, but that wouldn't apply to Usenet
anyway because a Usenet server's retention time is typically longer than
would be necessary for the mere end to end transmission of a message).

What's more interesting in the current context is the extent to which the
Online Safety Act affects regulations 17 (mere conduit), 18 (caching) and 19
(hosting). My reading of Ofcom's guidance is that it won't directly override
those regulations by removing those defences, but the OSA requirement for a
robust and accessible reporting mechanism will make it harder for service
providers to argue that they genuinely weren't aware of the presence of
unlawful material on their servers.

Ofcom's guidance does appear to go beyond that for services at high risk of
inadvertantly hosting CSAM or terrorism content by requiring proactive
keyword scanning of text and hash matching[1] of images, which is more
onerous than simply waiting for user reports of unlawful material. But
that's still a means of obtaining awareness, it doesn't create liability for
material that the scanning misses.

On the whole, therefore, I think that provided a service provider is
complying with the OSA requirements for risk assessments, risk mitigation,
robust reporting and, where appropriate, proactive scanning, the Electronic
Commerce Regulations will still provide a defence for any material which
does slip through the net. It's just that the OSA will require providers to
use a net with a smaller mesh.

[1] Which I still can't resist thinking of as "the necessary hashtags".

Mark

[Tim Jackson]

On Sat, 18 Jan 2025 18:22:00 +0000, Mark Goodge wrote...

[Electronic Commerce Regs / InfoSoc Directive]

Tim Jackson wrote...

Post by Mark Goodge

Post by Tim Jackson
There might perhaps be an issue as to who is the "recipient" of a
distributed Usenet hosting service: the poster, the viewer, or both? (I
suspect both.)

That regulation starts off by referring to "the storage of information
provided by a recipient of the service", which would imply that the
recipient of the service is the poster, not the viewer. Of course, there can
be multiple services involved in hosting UGC, but the one that's relevant
here is the service provided to the supplier of information which may later
be determined to be unlawful.

I could have worded my post more clearly.

Yes, I agree that for the purposes of Regulation 19 it has to be the
poster who receives the service. But as Usenet is a distributed
service, the viewer will usually be viewing a copy stored on a
downstream server, not the one it was originally posted on.

Clearly the viewer is receiving a service from the downstream service
provider.

However, in my opinion, the *poster* is also indirectly receiving a
service from that downstream service provider. Which then benefits from
the safety provided by Reg 19 as long as it takes the post down when
notified.

[Mark Goodge]

Post by Tim Jackson
On Sat, 18 Jan 2025 18:22:00 +0000, Mark Goodge wrote...
[Electronic Commerce Regs / InfoSoc Directive]
Tim Jackson wrote...

Post by Mark Goodge

Post by Tim Jackson
There might perhaps be an issue as to who is the "recipient" of a
distributed Usenet hosting service: the poster, the viewer, or both? (I
suspect both.)

That regulation starts off by referring to "the storage of information
provided by a recipient of the service", which would imply that the
recipient of the service is the poster, not the viewer. Of course, there can
be multiple services involved in hosting UGC, but the one that's relevant
here is the service provided to the supplier of information which may later
be determined to be unlawful.

I could have worded my post more clearly.
Yes, I agree that for the purposes of Regulation 19 it has to be the
poster who receives the service. But as Usenet is a distributed
service, the viewer will usually be viewing a copy stored on a
downstream server, not the one it was originally posted on.
Clearly the viewer is receiving a service from the downstream service
provider.
However, in my opinion, the *poster* is also indirectly receiving a
service from that downstream service provider. Which then benefits from
the safety provided by Reg 19 as long as it takes the post down when
notified.

Yes, I agree. It doesn't have to be a contractually provided service. Every
server in the network is providing a service to every author whose material
it carries.

I also don't think the courts would have any trouble reaching that
conclusion, either. Web hosting is increasingly distributed these days; the
use of "cloud" services, CDNs, load-balancing etc means that the physical
server which communicates directly with the user who uploads something may
well be entirely separate (and have a different operator in a different
country) to the server which communicates with the viewer of the content.
But that final server is still providing a service to the originator of the
material, by making it available on their behalf to the viewer. It would be
absurd if technical pedantry meant that the viewer-facing server was not
covered by the regulations. And the courts don't go in for technical
pedantry; if the route from uploader to downloader goes through multiple
servers it's just the same, in law, as if it only goes through one server.

Mark

[Roland Perry]

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

[Roger Hayter]

Post by Roland Perry

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

The complained of post would still be defamatory under any imaginable version
of defamation law.

[Roland Perry]

Post by Roger Hayter

Post by Roland Perry

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

The complained of post would still be defamatory under any imaginable version
of defamation law.

Please show your working, because the law has changed.

[Roger Hayter]

Post by Roland Perry

Post by Roger Hayter

Post by Roland Perry

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember appealed, so
it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an appeal was
felt to be unlikely to succeed. Demon were better resourced than Mr Godfrey.
Had they felt they had a reasonable chance of winning at appeal, they would
have appealed.

And what the case was based on was not Demon
carrying defamatory material per se, but Demon failing to take it down when
asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

The complained of post would still be defamatory under any imaginable version
of defamation law.

Please show your working, because the law has changed.

The post accused him of indulging in an activity which all normal people would
find morally reprehensible. It was indisputably defamatory. Defences are
possible to a defamation claim, but there is no doubt at all that the post
would be found defamatory as a first stage, before any defence was considered.

[Mark Goodge]

Post by Roger Hayter

Post by Roland Perry

Post by Roger Hayter

Post by Roland Perry
What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

The complained of post would still be defamatory under any imaginable version
of defamation law.

Please show your working, because the law has changed.

The post accused him of indulging in an activity which all normal people would
find morally reprehensible. It was indisputably defamatory. Defences are
possible to a defamation claim, but there is no doubt at all that the post
would be found defamatory as a first stage, before any defence was considered.

It's arguable that a post on a fairly obscure Usenet group wouldn't now meet
the "serious harm" requirement of section 1 of the Defamation Act 2013. I
suppose a lot would depend on the probability of the claimant's personal or
professional acquaintances reading the post.

Even had the Act been in force at the time, I think it would have been
relatively easy to make that argument. In 1997, Usenet was by far the
largest online discussion system and it's extremely likely that Dr Godfrey's
colleagues and professional acquaintances would have also been using it. So
a likelihood of serious harm would have been demonstrable, on the balance of
probabilities. And that would be enough to make out the case for defamation;
it would then be down to a defence if one (or more) was applicable.

Roland is right, therefore, that the 2013 Act has raised the bar for
defamation, and it's probable, now, that most things on Usenet wouldn't meet
it unless the claimant could show that there was still a likelihood of
serious harm. And there might be, of course, in a group which still has
substantial use by a particular demographic of which the claimant is a part.
But the claimant would have to make that argument. It couldn't simply be
assumed.

Mark

[Roger Hayter]

On 21 Jan 2025 at 12:10:39 GMT, "Mark Goodge"

Post by Mark Goodge

Post by Roger Hayter

Post by Roland Perry

Post by Roger Hayter

Post by Roland Perry
What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

The complained of post would still be defamatory under any imaginable version
of defamation law.

Please show your working, because the law has changed.

The post accused him of indulging in an activity which all normal people would
find morally reprehensible. It was indisputably defamatory. Defences are
possible to a defamation claim, but there is no doubt at all that the post
would be found defamatory as a first stage, before any defence was considered.

It's arguable that a post on a fairly obscure Usenet group wouldn't now meet
the "serious harm" requirement of section 1 of the Defamation Act 2013. I
suppose a lot would depend on the probability of the claimant's personal or
professional acquaintances reading the post.
Even had the Act been in force at the time, I think it would have been
relatively easy to make that argument. In 1997, Usenet was by far the
largest online discussion system and it's extremely likely that Dr Godfrey's
colleagues and professional acquaintances would have also been using it. So
a likelihood of serious harm would have been demonstrable, on the balance of
probabilities. And that would be enough to make out the case for defamation;
it would then be down to a defence if one (or more) was applicable.
Roland is right, therefore, that the 2013 Act has raised the bar for
defamation, and it's probable, now, that most things on Usenet wouldn't meet
it unless the claimant could show that there was still a likelihood of
serious harm. And there might be, of course, in a group which still has
substantial use by a particular demographic of which the claimant is a part.
But the claimant would have to make that argument. It couldn't simply be
assumed.
Mark

At the time the claimant was to be found daily pontificating on an active
group discussing the country in which his 'offence' allegedly took place. So,
as I say, defence would be very difficult. I believe that he may also have
been living there at least some of the time.

[Roland Perry]

Post by Roger Hayter

Post by Roland Perry

Post by Roger Hayter

Post by Roland Perry

Post by Mark Goodge

A couple of points: Godfrey v. Demon was not so far as I remember
appealed, so it does not set a precedent.

It wasn't appealed, no. But one of the main reasons was that an
appeal was felt to be unlikely to succeed. Demon were better
resourced than Mr Godfrey. Had they felt they had a reasonable
chance of winning at appeal, they would have appealed.

And what the case was based on was not Demon carrying defamatory
material per se, but Demon failing to take it down when asked to.

For that even to be relevant, the question of liability had to be settled
first. If Demon could have defended the case by arguing that they weren't
liable, they would have. But they didn't.

What people need to understand is that Demon ran a Usenet server, and it
was because of that they were pursued for alleged defamation. Since then
the defamation law has changed quite a lot.

The complained of post would still be defamatory under any
imaginable version of defamation law.

Please show your working, because the law has changed.

The post accused him of indulging in an activity which all normal people would
find morally reprehensible. It was indisputably defamatory. Defences are
possible to a defamation claim, but there is no doubt at all that the post
would be found defamatory as a first stage, before any defence was considered.

You've failed to "show your working" under the revised law.