bantering with punk was Re: What advantage does Signal protocol have over basic public key encryption?

Punk-BatSoup-Stasi 2.0 punks at tfwno.gf
Sun Jan 31 22:43:25 PST 2021


On Mon, 1 Feb 2021 00:05:13 -0500
Karl <gmkarl at gmail.com> wrote:

> >> It's really a lot of work to look up information to back up every
> >> single point I make.
> >
> > 	.... A lot of work? Finding one reddit link?
> <snip>
> > 	I would never make such assumption. It's a perfect example of a 'non
> > sequitur'. It doesn't follow that X being easy means X is being done.
> 
> Punk, my schizophrenia is really bad.  

	As far as I'm concerned 'schizophrenia' is fascist psychobabble, just like 'ADD', 'autism', 'paranoia' and the like. Notice that 'autism' has been invented by a literal austrian nazi. 

	I would still use the word/concept "crazy" to refer to people who are detached from reality (in various ways), but it's a common sense/philosophical concept, not pseudo-scientific bullshit used for political manipulation. 

	Anyway, my point is, I don't think you should say you have 'schizophrenia' and I don't think you're crazy either. 


> 
> More to the point, you argue absolutely everything I say in many
> areas, so it is not very effective to look everything up for evidence,
> if it is relatively obviously.


	fair enough - we disagree about some things being relatively obvious tho =P 

 
> Here is a proof for you:
> There are X billion people on the planet.  Each person is exposed to Y
> different things and has Z different skills.  Each person has A%
> downtime, during which they combine their Z skills with the Y things
> they are exposed to.  Over the course of somebody's life, they will
> sample the combinations of Z with Y, A% of the time.
> 
> Not only that, but they probably work a job B% of their remaining
> time, where they also combine Z with Y in ways that other people
> choose.
> 
> Now, their Z skills arrived from experience and practice, in exposure
> to the Y things around them.  So, since they have these skills at all,
> we can infer that there is a smaller set of skills Z_2, which they
> used in thorough learning combination with Y_2, the things they were
> exposed to earlier in life.
> 
> Ahhhh I don't know enough off-hand probability to make it look like a
> real proof ;P


	at best you could say that if something is easy, then it's more likely to be done. But wait, even such an assertion isn't necessarily or strictly true. 

	Naive counter example : it's very easy to smash your fingers with a hammer but very few people would do that on purpose. 
	

	Anyway you can easily prove your point by listing the 'alternative', federated networks that use signal software. 



> 
> But the upshot is that if something is everywhere, and obviously
> doable, people _do_ it.  People _try_ things.  If you give a set of
> paints to a group of people without paints, and wait, you will produce
> new, creative paintings.  It is what happens.  Every single time.
> Unless somebody else is exposing them to artistry otherwise or
> preventing them from engaging in it, which does not happen
> universally.
> 


> >> >         On the other hand you cannot use the 'signal service' at
> >> > signal.org without registering. As a side note of sorts :  "Signal is a
> >> > registered trademark in the United States and other countries.". Plus :
> >> >
> >> >         "You agree to use our Services only for legal, authorized, and
> >> > acceptable purposes. "
> >> >         ('acceptable'? 'authorized'? 'legal'? LMAO)
> >> >
> >> >         "Signal’s Rights. We own all copyrights, trademarks, domains,
> >> > logos, trade dress, trade secrets, patents, and other intellectual
> >> > property rights associated with our Services."
> >>
> >> Trademark law doesn't seem to be a good point against anything here,
> >> to me.  It is normal in the USA to register trademarks for words.
> >
> > 	
> > 	Yes, 'normal' means it follows 'norms' or 'laws'. Following insane US laws
> > is 'normal' only in that circular sense. Apart from that, it is insane.
> 
> This doesn't seem to be a logical argument point, as registering a
> trademark is not following a law, 

	Registering a 'trademark' is 'invoking' the 'protection' of 'trademark law'. Which is nothing but having the state criminals on your side using violence against people who would 'infringe' 'trademark' 'law'.

	
> and norms are emergent whereas laws are prescriptive.

	https://en.wiktionary.org/wiki/norm

	"A rule that is imposed by regulations and/or socially enforced by members of a community."

	That's hardly different from a 'prescriptive law'. 'Normal' also means 'typical' but that doesn't help your case either. Trademark is indeed typical of the western cesspool. Typical garbage. 

	Anyway, I think their use of 'trademark law' is in poor taste, but arguably isn't that big of a deal. 



> 
> Do you do anything in your free time, punk-stasi?
> 
> I had a really unpleasant experience today where I lost a lot of
> memory parts.  I cried for these lost memories.
> 
> Do you code in python?  Machine learning algorithms are big in python.
> They're not hard to use.


	I code in python but I code dumb software that (hopefully) only does what I tell it to do. 


> 
> Part of your name is "batsoup".  When I was caving it was so
> wonderful.  Caves are incredibly beautiful inside.  It is like a
> climbing gym, everywhere, often for miles and miles, sometimes
> thousands of feet under the soil.  Some people camp out in caves
> during long mapping trips.

	Caving sounds like a really nice activity. I live in the pampas though and this place is probably one of the flatest parts of the earth. 


	(I added 'batsoup' to my nick in reference to the "raw bat soup that the chinese eat and 
that caused the current 'covid' genocide" - In other words I added 'batsoup' to my nick to mock grarpamp and co.) 



> 
> I went into a new cave that had just been discovered once.  We had to
> crawl through these tiny passageways, both sideways and up-and-down!
> We also had to rapell down cliffs using ropes.  Eventually we got to a
> room 300 ft wide and tall, with a giant waterfall coursing straight
> down the middle of it.  That waterfall had cut the whole room.  People
> climbed up the sides of the room, and their headlamps looked like
> stars sparkling in the dark underground distance.  Around the edges of
> the cave the water pooled, and flowed further into more unexplored
> earth.
> 
> Caves often have crystals in them.  Bats hang upside down.  There are
> also other creatures, insects and other things.
> 
> It would be easy to live deep in a cave using water for electricity.
> But you'd need to manage your toilet well.


	I've been seriously thinking about the benefits of living underground when it comes to avoiding surveillance...

 
> I have always loved being alone, silent, and in the darkness.  Caves
> have these things to such an extreme.

		
	Silence is nice. 


 
> >> This makes it harder for other people to co-opt them and pretend to be
> >> you.  They are only legally protecting their name and logo.
> >>
> >> Yes, their mainstream client centralises registration, probably for
> >> reliable identity management, a choice that has been discussed
> >> extensively.
> >
> >
> > 	.... has been discussed by whom?
> 
> They used to discuss it in bug reports and feature requests.  It's
> also been discussed on this list extensively.
> 
> https://en.wikipedia.org/wiki/Zooko%27s_triangle
> 
> > 	So anyway, my conclusion again : the signal PROTOCOL is more robust under
> > CERTAIN attacks. The SIGNAL COMPANY is an untrustworthy organization that
> > was 'founded' with 3 millions from the pentagon and 50 millions from some
> > silicon valley oligarch and WHATSAP asshole.
> 
> Punk is participating in the incredibly harmful targeting of community
> work.  

	I don't view this in terms of anti (or pro) community. If anything I'm critical of the 'community organizer'...


> He makes it obvious he is doing this, so he is an ally who
> should not be trusted.  He has also acted to support this list, I
> believe.
> 
> The idea of sharing mistrust of a nonprofit is interesting, as
> nonprofits are also made to support the ends of profit-motivated
> organisations.


	Well, as I imagine you're aware, there are a lot of 'non govt organizations' out there that are just a facade for govcorp. 

	just a sample : https://en.wikipedia.org/wiki/List_of_projects_supported_by_George_Soros



> 
> The idea of pointing out behaviors that sound untrustworthy is also interesting.
> 
> The continued question is how do we know whom to trust.


	Nobody should be trusted? I mean, without looking at the dictionary definition, my understanding of 'trust' is something like : trust is the assumption, more or less unwarranted, that something isn't harmful. So in cases where security is critical, trust looks like a bad idea.


> 
> Maybe a much better question, is whom can we trust better than whom
> else, and in what ways, and for what?

	yeah.
	



More information about the cypherpunks mailing list