Org mailing list

[Thomas Gramstad <thomas.gramstad AT ub.uio.no>]

Hi,

I posted this talk suggestion a few weeks ago, but never got a 
response. (Did it make it to the list at all?)

It's a long post from Amelia Andersdotter about a talk that was 
planned in Oslo this spring. That talk didn't happen for 
logistical reasons (plane strike), so I think it would be nice 
and interesting to have it at FSCONS in november. I would like 
the meeting tonight to vote on it and hopefully approve it.

Thomas


---------- Forwarded message ----------
Date: Sat, 27 Apr 2019 13:00:47 +0200
From: Amelia Andersdotter <amelia AT andersdotter.cc>
To: Thomas Gramstad <thomas.gramstad AT ub.uio.no>
Cc: Amelia Andersdotter <amelia.andersdotter AT piratpartiet.se>
Subject: Re: Invitasjon til PIR landsmøte i Oslo 27-28 april 2019
Resent-From: <thomas.gramstad AT ub.uio.no>

Jag lade i alla fall lite tid igår på att fundera på vad jag ville säga:

Greetings and salutations.

My name is Amelia Andersdotter and for the past 15 years I've been
working on digital rights in various capacities. I started out caring
about copyright and surveillance issues - heated debates in the Swedish
society, some of which will be remembered by people who were old enough
in 2005-2009. I served as an European Parliament member on behalf of the
Pirate Party between 2011 and 2014. Since 2015 I run a data protection
oriented NGO in Sweden called Dataskydd.net and for the past couple of
years I've worked with British NGO ARTICLE19 on their program to assess
the impact of technical design choices in standards organizations on the
ability of individuals to exercise their human rights.

To disclose some more of my background, I have a bachelor's degree in
mathematics and I once wrote a bachelor thesis in business law oriented
towards competition and anti-trust in the telecommunications sector.
I've been enrolled in a master's program for applied mathematics at
Uppsala university since 2017... It is going somewhere.

Thomas Gramstad kindly asked me to share some thoughts on the state of
information policy today. It's a broad field, and encompasses many
things from innovation policy to human rights and democracy, but I will
try to narrow it to examples that I believe contextualize issues
relevant for northern Europe.

I also want to quickly make the caveat that I am not too familiar with
the Norwegian situation, and while Norway is not in the European Union -
which is where most information is determined from a Swedish perspective
- and that therefore, should you find me Sweden or EU centric, these are
the reasons. I should also clarify that I am speaking on behalf of any
organization for which I'm currently working, but representing my own
observations accumulated over the past fifteen years.

I will cover three separate topics today, that for some of you may feel
old, but I will justify this perspective. Firstly, I want to cover
national security and its relation to the internet in broad terms.
Secondly, I want to cover public sector institutions, their struggle for
bestness through computers and connections and the way in which this
religious approach to computers is not actually very useful. And lastly,
I shall go back to a core topic of information policy that continues to
be important for individuals today: copyright.

In information policy, new topics crop up by the day.

New surveillance laws, new interesting means for law enforcement or
companies to encroach on individual freedoms and opportunities, new
corporate rights aiming to foreclose competition or restrict
individuals, and so forth. We are living in a time where governments and
companies are becoming increasingly more powerful, while individuals are
increasingly less powerful.

For a short while, around the Snowden revelations, there were
discussions around this. I find the reflection of Barton Gellman
particularly fitting: it is as if we are behind a one-way mirror, where
we, as citizens and consumers, are increasingly more transparent to
companies and governments, while they are increasingly more opaque to us
[Cato Institute, 2013].

Good news are few and far between, and to add insult to injury,
governments all over the world are now beating the cyber war-drums. Your
router, which just yesterday was a consumer security disaster by virtue
of the poor procurement practises at telecommunications operators and
the lack of oversight, is now a war-zone, effectively limiting the
government's sense of urgency in applying even the few norms that were
already in place.

It is an issue.

Meanwhile, human rights courts are increasingly bogging themselves down
in state perspectives, rather than individual perspectives. At the end
of the day, as a society, we have not advanced too far from the
Westphalian model. The European Union, which started out a brilliant
post-modernist experience of having states compromise and negotiate,
rather than colonially pushing their views on one another in a contest
of power, is increasingly modernist. We see it not the least in the
adoption of a stale French proposal to promote European prominence by
introducing heavy regulatory requirements on US web services in
copyright law, in direct contradiction with what most successful web
businesses (few of which are, by the way, based in France) would argue
Europe needs.

These are long-term trends. The re-modernization of state-to-state
relation and state-to-citizen relations, the return to a world-view
where loyalty to king and country goes first, and loyalty with the
individual citizen or a private business always comes second. And, on
the internet, the disastrous consequence that using the internet as a
battle field implies that each one of us, however unknowingly, are
caught in the cross-fire.

The European Union faces some interesting challenges in this respect.

In the past four years alone, the European Union has produced not one,
but two, new laws aiming at harmonizing network and information security
rules around the EU. The reason for this is of course that most
information and communication technologies are developed by and
distributed by the private sector to consumers who are either
individuals, civilian public sector institutions or themselves private
sector companies. Think about an wifi access point set up at a
kintergarden, or a printer installed in a local office.

For most of these producers and consumers of ICTs, being in a warzone is
inconvenient. It limits their choices, makes the products available more
expensive, and risks reducing the quality of the service by restrictions
imposed from national requirements. The European Union is trying to fix
this problem, by ensuring that any security-motivated, nationalist,
protectionist measures are at least harmonious in the European trade area.

But EU member states are, in general, not stupid, and they recognise
when someone is trying to integrate competences that they consider core
to their identities in a global world in a market place. The first
European attempt at harmonization, the Network and Information Security
Directive of 2015, was therefore watered down to the point of being
useless. The only definitive requirement contained therein, is an
obligation on network and information providers to do breach notifications.

It is the third European act which requires breach notifications, the
others being the ePrivacy directive of 2009 and the General Data
Protection Regulation of 2016. And only the ePrivacy directive of 2009
recognises the important role that transparency around security issues
towards consumer plays in ensuring market actors get their act together,
and gradually act to improve their security.

I was alerted recently to how the implementation of the Network and
Information Security Directive has gone poorly in Sweden. Almost two
years after the entry into effect of the Swedish national law, many
organizations still don't know if they are covered. I feel vindicated by
this knowledge, because already in 2015 it was obvious that the Swedish
government had speedily acted to produce its own local Network and
Information Security law to compete with the European framework. The
Swedish home-brew legislation takes a national security perspective, in
theory could be filled with more stringent requirements that are more
poorly adapted to a market place, and primarily - does not concern
itself with individuals or consumers at all, since it is not in the
nature of a government security apparatus to do so. I had predicted that
these frameworks would compete, and that market actors, if they were
wise, would recognise that the nationalist aire around national security
laws is so much stronger than the consumer-oriented market policies of
the EU, and therefore choose to follow only the Swedish home-brew law.
The European Commission must have felt immediately that the Network and
Information Security Directive was a failure. They acted even in 2016 to
create a Cyber-Security Regulation, which, by virtue of being a
regulation, becomes immediately applicable law in all the member states.
No sooner had the dust settled after the initial  Network and
Information Security Directive than were the member states and European
Parliament back in negotiations over security issues. Now the European
Union has a system for adopting cyber-security standards.

The speed with which new initiatives are launched, and new laws passed,
at the scale of the European Union, is in itself an indication of
failure. The fact of the matter is, member states, in general, are not
willing to give up national security competencies, and this is harming
the inner market. It is still expected of all of us that we accept that
we are treated worse by our governments, and worse by our companies and
service providers, because we are expected to have this grand idea of
loyalty of with an antequated idea of a modernist state imposing itself
on us and on our neighbours.

I find this field so fascinating, because it is really a breaking-point
in history. It is where we choose whether to be lulled back into the
19th century or whether to advance into the 21st.

For Europeans, now more than at any time, we can actually choose - well,
within limits because we are constrained by the particularly strong
modernist traditions in some EU states - whether we believe the world is
a competition between civilisations, or whether the world itself is in
fact a civilisation.

Let me contextualize this with an example that may seem trivial.

About two years ago, I was working over summer as an op-ed writer for a
local paper in mid-north west of Sweden. I came across the Swedish
village of Grövelsjön to the very west of Älvdalen municipality,
bordering Norway and the equally delightful village of Grøvelsjøn, which
I had the great pleasure of visiting. These areas, around Älvdalen,
Särna and Idre, are exceptionally well-connected.

The burning light of one man, former telecoms equipment developer Agne
Fredriksson, ensured around 2010 that the villages of the Swedish side
of the border would make use of European funds to roll-out fibre optic
cable village networks. Now, in Grövelsjön and Sorkmyren, the lights are
on and the connections are 100 Mbit symmetric. I visited this place, and
there is something special about having your phone out of range of the
3G network while still having excellent data traffic connectivity.

The village networks of Älvdalens municipality in turn created the
opportunity for the municipal broadband coordinator to focus on
backhaul. In many municipalities of Sweden, you see, fibre optic
roll-out has been achieved with city networks - a prominent example of
which is Stockholm STOKAB - that operate both backhaul and last-mile.
But Älvdalen is geographically large, and has a small population -
building both backhaul and last-mile is very expensive. By being able to
focus the municipal resources on backhaul, while having every village
own their own local last-mile fibre rings, Älvdalen's municipality is
now one of few municipalities in all of the mid-west north set to hit
the national broadband targets of 95% high-speed coverage.

An issue for this region is that it is very mountainous. Building
backhaul is not only expensive because of the nature of the thing, but
also because of the landscape in which it has to be constructed. I was
informed that one way of solving it relatively cheaply would have been
to build a small loop of fibre into Grøvelsjøn, Norway, to avoid a
mountain. This would have had many advantages: the cost for deployment
would have been considerably reduced, while the citizens of Grøvelsjøn,
Norway, could have enjoyed 100 Mbit symmetric the same as their Swedish
peers a few kilometers away. Unfortunately, for reasons of lack of
market harmonization and of national security, this had been impossible.

In Norway, and someone can correct me if this has now changed on been
improved, it turned out the telecoms market is divided into regional
monopolies, where companies are allocated territories on which they are
the sole decider of what infrastructure goes down. In Grøvelsjøn,
Norway, this local provider was a cupper-based ADSL company.

Additionally, in Sweden there is a mass-surveillance law which dictates
that all traffic covering the border must be interceptable by national
intelligence. So building the backhaul into Norway actually presents a
problem for both the national security establishment and for the
municipality - how to solve this precarious problem of the small fibre
loop?

The solution was of course to make the less efficient choice of running
the backhaul around a mountain, a much longer, more expensive distance
that also deprived the citizens of Grøvelsjøn, Norway, of high-speed
symmetric connectivity.

It is but one of many sad examples of how old nationalist, protectionist
policies stop active, community-minded Scandinavians co-operating for a
better future. And while I recognise the gist of this story is that
things on the Swedish border worked out fine, while it was only our
Norwegian friends who continue suffering under low-quality internet
connections, I cannot be satisfied when any household that could have
cheaply, easily and logically been given access to the wealth of the
world wide web and internet connectivity is not given this access
because of what must reasonably considered trivialities.

These issues - infrastructure policy and national security - are age-old
issues. They seem boring, but they have a big impact on our societies.
When I was asked to make a statement on what issues I believe will be
important going forward, they are such issues: the avalanche of many
new, bad ideas must not cause us to lose sight of the many old problems
that are still around.

I am not necessarily a believer in trend-setting politics, but I do
believe politics should solve outstanding issues.

I can add to this a different outstanding issue, that is somehow not
only related to the internet and the world wide web, but to the very
nature of computation itself.

For a few years now, artificial intelligence and machine learning have
been trending topics in the global debates on privacy, autonomy and,
indeed, national security. These discussions started around the same
time that I was working a lot on the Swedish implementation of the data
protection regulation, and therefore had fresh in my mind my
investigations into 1960s and 1970s computerization discussions in
Scandinavia and Europe at large. It is surprising how slowly discussions
on computers are advancing - one could argue they are still in the same
place that they were half a century ago.

For one, any new information and communication technology brings with it
the hope of more efficiency and greater precision. It is envisaged that
we, as a society, will become better, faster, and stronger, from the
deployment of many, many machines. In fact, one could argue, society -
not just Scandinavia, but globally - has the naive impression that
machines can save us from ourselves.
After seeing the discussions on machines as decision-makers over half a
century, I am no longer optimistic about this perspective.

The internet is great, that's true, but it is great precisely because we
all get to know one another, hang out, chat, and exchange culture or
texts and banter, but computers in general do not remove from society
any inherent issues that arise from human-to-human interactions.

Let me contextualise this with two examples from the public authorities
of the state which issues my passports.

Firstly, when Sweden computerized its public and private sector in the
1960s, big benefits were predicted for everyone. Sweden would be
efficient and at the forefront and have national companies leading the
global revolution. By the 1970s, many of the big benefits envisaged of
centralized computing had failed to materialize. Not only were things
not very much more efficient, but it also turned out that the new
centralized computer systems introduced a range of maintenance problems
which served to shift demand for labour to other parts of the
organization - but worse, because unlike for railroads and telephony,
the government frequently had to relied on private companies to acquire
the right expertise.

The government, puzzled by this circumstances, observed that people had
failed to adapt themselves to the machines, and that further benefits
would surely arise down the line.

By the end of the 1970s, it was of course becoming obvious that machines
were not infallible. They could have bugs, that is, malfunction, and
when they were connected to each other there could be other, even
deliberate, security issues.

In the beginning of the 1980s, the government realized that dealing with
all these issues cost a lot of money. While there can be no particular
political responsibility attributed to what happened next, it is true
that the Swedish public institutions at this developed business models
to cover the costs of the computers that today make them a slightly more
sinister version of Google. Most large Swedish public databases are, by
law, constituted to provide careful profiles of citizens to advertisers.
This is true for the tax database, for the traffic registry, for the
general population database and for the corporate registry.

Because by law citizens cannot, as they do on web services, spoof
information, that is, hide their identities behind junk information,
these government profiling activities are not, as in the case of web
services, machine-learning based. There is no need for the government to
resort to statistical models to explain the properties of personal
properties of any individual member of the public, because they have
enough data to be deterministic.

Now, I will be the first person to concede that the 1990s and early
2000s actually weren't that bad. Broadband roll-out, the personal
computer reform and peer-to-peer networks all worked out for people. The
tax authority was also reformed in a way that enabled them to make good
use of computers: for the vast majority of my life, declaring taxes in
Sweden has, for most people, been as simple as sending a text message.
These types of reforms are helpful and inspire confidence that computers
can in fact achieve good things.

You will forgive me, I hope, for not delaying too much on these hopeful
topics at this time, but I will come back to them later.

Unfortunately, all these good aspects of internet and connectivity have
not exactly been embraced by our political classes. By 2010, for
instance, the Swedish government had relabeled "computerization" as
"digitalization" and appeared to be back in the 1970s. Again, there were
predictions of Swedish "bestness" - poorly defined, and ever in the
future. Narratives emerged where it was not the case that the internet
connected people, and enabled friendship and cultural exchange, but
where, in fact, the internet was cesspit of low morality, and where
individuals attacked other other individuals.

It was no longer the case that the state, through thoughtful use of
information and communications technologies, could serve its people. It
was the case that individual members of the public were harming other
individuals of the public, and that the state's primary role was in
punishing individuals who were internet users.

We are still broadly stuck in these 1970s charades. Still striving for
bestness, that no one knows it is, and still stuck in forgetting that a
government has duties with respect to the people it serves.

I say, with blissful nostalgia of the 1990s and the early 2000s, that we
should not ask what we can do for our governments, but what our
governments can do for us.

And it is true that we should expect more of our governments than for
them to pit us against one another.

Secondly, let me touch a bit on law enforcement issues, and in
particular those demands for new law enforcement competencies that
require intrusive, unmonitored surveillance such as data retention, data
requests or hacking.

Again, I may come to be a bit Sweden-centric - it goes with the
territory, a-hah, and in spite of all my internationalism, I still feel
the confinement of the Westphalian state.

It has been frequently raised, especially by those that fear the
internet and computers, that these platforms are a bit like the wild,
wild west and that they must be regulated. I have increasingly come to
share this view, but not from the perspective that it is normally raised
- but rather from the perspective of the impunity with which our
government authorities act on these platforms.

Data retention is an age-old debate that rose to prominence again as the
European Court of Justice, the EU:s highest court, declared it
disproportional and unnecessary in 2016, favouring instead measures
targetted at specific individuals who are specifically suspected of a crime.

But it is not new on the agenda. There are Swedish police officers who
have been working since I was in third grade of elementary school on
creating the global conditions for data retention. While I cherish the
1990s and the early 2000s, I also acknowledge that in the background
there were dark forces at play.

It has been litigated many times. As many as nine European
constitutional courts had declared data retention illegal before Sweden
had its data retention law installed in 2012, as a consequence of a
European directive passed in 2006, in direct defiance of most advice and
available evidence.

You would think, that in such a challenged situation, with such little
evidence to back up the efficiency of data retention, that Swedish law
enforcement, having nagged for 20 years to get this tool, would have
ensured that they gathered statistical evidence on the usefulness of
this tool. But you would think wrongly.

In fact, law enforcement efficiency in Sweden has been on the decline
for a couple of decades and is frequently and aggressively raised in the
media. The drop in efficiency increased after 2012, and a national-wide
re-organization in 2015, I am told, caused them to not have any data on
the efficiency of data retention at all.

Mysteriously, this is not unlike every other EU state that has data
retention rules in place. It is allegedly the silver bullet that will
stop society's decay, and yet it is impossible to demonstrate how
exactly this is so.

If one is religious about data collection, and believe that computers
will bring bestness, efficiency and save us from ourselves, then of
course that belief might be sufficient to sustain the optimistic view
that a lack of results today, may not persist into the future. But I
believe that in the data retention discussions, more than in any other
society debate on surveillance, we can see the flaws of this religion.

For one, Sweden is a fairly well-regulated society and assesses the
efficiency of its public institutions a lot. To the point of
inefficiency, some would claim.

In the case of law enforcement, we have direct evidence that police
officers in Sweden are bad at handling evidence, suffer from inferior
internal organizational planning systems, and are additionally unskilled
at investigating crime scenes. As far as I can determine, they are not
trained in interrogation techniques, and are not offered to be trained
in interrogation techniques.

But religion is not rational, and the public institution in charge of
evaluating police activities has increasingly come to be seen as a
conspiracy-mongerer, devalued and insulted for covering up facts, and
generally portrayed as incompetent. Not just by weird people on Twitter
either, but by reasonable members of the public.

In law enforcement activities on the internet and around computers, we
are increasingly living in a religious society, that puts beliefs above
facts, and that addresses facts by throwing mud at the institutions in
charge of it. At best, this is silly. At worst, it causes the Swedish
minister of interior affairs to unrighteously threaten Swedish
telecommunications operators with fines and withdrawn licenses, because
he is unwilling to take responsibility for real problems in the police
force, that cause real problems for police officers and the public when
real crimes are committed.

If one is religious around data collection, believing that eventually,
surely, it must save us from ourselves, this insistence on over-looking
the evidence, and turning a blind eye to the real human rights concerns
caused by these methods, can be justified on faith alone.

A challenge therefore, for individual engaged in information politics,
is how to counter this religion.

The internet is great, that's true, but it is great precisely because we
all get to know one another, hang out, chat, and exchange culture or
texts and banter. Computers in general, or data, do not remove from
society any inherent issues that arise from human-to-human interactions.

It brings me to my promised finale, which addresses a topic that
continues to be of great concern and which influences internet users
around the world, every day, in a restrictive way that also, for the
European Union, brings the particular challenge of creating community
and camaraderie.

I am of course talking about copyright, and the botched reform of the
European communities recently passed.

To bring some perspective to this issue, let me point out that I can
physically move my body between Malmö and Köpenhamn in about 30 minutes
by train at no great cost or effort to myself. I basically just get a
ticket and sit, and suddenly I'm across the border.

Going the other way has been made more difficult by the border controls
enacted by the stressed-out government in Stockholm, that unfortunately
feels the economic development and fortitude of the southern parts of
their territory is not so useful to them. But I would still argue it's
not more difficult than, say, a 40 or 50 minute train ride, with a brief
stop at Copenhagen Airport.

A leg-stretcher. Let's call it a public health initiative.

If I were to attempt moving an mp3-file containing music across the same
border, I could easily be stuck in license negotiations for years before
it succeeded. If I were anything other than a bigger commercial entity,
it would be seriously challenging to me. If it is an mp4-file,
containing not only audio but also video, I will argue that I have next
to no chance of transferring the file legally without institutional backing.

This is of course ridiculous. We, allegedly, have free movement of
people in the European Union, but we cannot move the things that connect
people - culture, experiences, shared happiness and references - across
the same borders. For the longest of time, sharing cultural experiences
with Germans has been a massive pain, for every time I try I get the
response that they find the video is not available in their country.

The European Union, of course, has been trying to address these issues.
But every time they hit a wall: they get rid of geoblocking, except for
audiovisual content - so the very thing that caused geoblocking to be an
issue is still geoblocked, while now, at last, I can order stuff from
amazon.de, which of course I never did because I don't speak German, as
don't most people of the European Union.

They try to make special rules for cultural heritage, but end up
imposing on librarians a duty of due diligence. I ended up being able to
access the train schedule of Luxembourg to Brussels for 1862, which
again was great, but only for about 2 seconds and, if we're honest,
knowing when people could depart from Luxembourg to  Brussels in 1862
may be of academic interest to a small range of individuals in the
European Union, but doesn't really cut it if we're trying to build
relationships between broad groups of Scandinavians and Luxembourgians
or Belgians in the 2010s or 2020s.

A fairly known Swedish company, Spotify, exemplifies the failures of the
European markets as much as it exemplifies how easily accessible
streaming services could have been a good market opportunity for
creative industries going forward. Their launch pattern was Sweden,
Netherlands, UK and the US. Only after establishing themselves on the US
market were they able to come back to the EU to establish themselves in
more jurisdictions here. Meanwhile, many Europeans - especially the more
tech savvy ones - had been using Spotify by proxies, essentially
cheating the service by claiming they were listening from a jurisdiction
in which they were not based (and doing so to Spotify's detriment, since
Spotify's license requirements would have not actually covered such
consumption).

It is surely a failure of the European inner market that new services
cannot approach it without first establishing a strong presence on the
US market. It gives a big advantage to US companies who, of course, by
default establish themselves on the US market before coming to Europe.
An EU company has twice the work, and must be twice as good, to have
half the chance of becoming an EU company at all, as does a US company.
These are real economic problems, that the EU should really want to solve.

Some video streaming services have come to be popular across the
European Economic Area as well - US video streaming services, by the
logic I just presented. The response of the EU was to force these
services to have a certain percentage of locally produced content -
because it does not register with our politicians that the issue is
really that European companies cannot establish themselves in Europe
without establishing themselves in the US and this is economically
inefficient and very harmful. Instead of lowering the bar for market
entry, it is raised.

This is sadly a familiar pattern of all information policies of the
European Union and its member states.

For instance, the European Union was really the originating point of
several of the largest, most popular and cherished popular platforms on
the internet. Services like Oink, The Pirate Bay and Demonoid had
world-wide reach and, like no other services, connected people around
shared cultural experiences, large and small.

It is no surprise that nationalism and fear feeds off of the lack of
understanding of the other - and pirate copying movies and music could
have contributed to mitigating such ignorant resentment between peoples.

Instead, the European Union banned these services, just like they have
now raised market entry barries for video streaming services while
continuously failing to make stronger efforts at facilitating the
establishment of EU cultural services.

A lot people may not like it, but that's how it. The European Union was
really good at the one thing, and then it banned that thing, because
that's how Europe rolls. And then it aggressively complains about being
a digital colony, because complaining is, apparently, how to sort things
out, while taking care of yourself isn't.

To finish off, I would like to stress that these issues that I have
raised only allude to a larger framework of lack of European strategy,
and the increasing hostility from governments to ordinary people.

Ultimately, the very broadest question is whether Europeans are enemies
of Europe, or friends of Europe, and I believe they are friends of
Europe and that such a mind-set should dictate how politics in any area
is developed.

I believe it is in all of our interest that our interactions and
communications are at the forefront of any political discussion, because
only through our interactions, friendships and joint efforts can we
solve other outstanding issues, whether they be the unequal distribution
of income, the relative powerlessness of our governments in the face of
investment capital, the climate crisis or any other things.

It is precisely because it our cross-border friendship and mutual
interactions that we must continously raise and address the problem of
our governments religiously holding on to mass-surveillance as a
catch-all solutions that all evidence points to being a different
problem entirely, that we must continously attempt to resist the image
of citizens as the natural enemy of their governments and each other,
and the fear of any European (or for that matter American) economic
web-based activities so prevalent in our cultural establishments.

This is what I hope you will work for, and that I will continue working
for with the means at my disposal.

Thank you for your kind attention.

    ----