[Standards-JIG] Cspace - much hype, but a DHT isn't enough for most people

[Carlo v. Loesch]

ah! cspace! so much hype, so few innovation.. using a p2p technology
for a completely wrong purpose.. but maybe i'm wrong, so let's talk
about it..

thomasasta at gmx.net typeth:
| The RSA-Key is the long key for emailfooters and the optional CSpace-ID is the short easy to remember number for exchange with the ladies in a tramp ;-)

the lady is a tramp in a tram, i see, but what's sexy
about an ICQ-like number? when all of humanity has a CSpace-ID,
will they be hard to remember because they are ten digits long?

and what if your address servers matching up the cspace-id to
your public key are bought up by aol?

| No, this is insecure, so a server as a transport hosting millions of private key would underdetermien the security. So the only option is to integrate Cspace in the clients, as with jabber-encryption the servers are useless, as they route only white nose traffic, so the server hop is a WASTE.

the server hop is not so much a waste, as it offers services
for the time you are offline.. storing messages, handing out your
mobile phone number to your friends in need.. if cspace gets
popular, people will start running "bouncers" just as with IRC.

| Encrytion in jabber makes jabberservers obsolete, this is the claim!

i'm not sure about jabber servers but PSYC servers surely aren't
obsoleted by this yet. try to convince me. go read what a psyced.org
server can do and tell me kademlia does anything like that.  :)

| The binding between Cspac and Jabber is the enxryption: Every jabber client with encryption does not need servers. Its a WASTE routing !

it may be unnecessary routing to go through a home server, yes. but you can
exchange end-user ip numbers and start talking directly to each other.
jabber doesn't provide that, but PSYC does. how you look up the ip number
of the recipient is a different issue - a public DHT network has its
vulnerabilities, i prefer having my own home server handle the job for me.

and i prefer a uniform like psyc://psyced.org/~lynX to accept incoming
requests for me, then i can still use PK to make sure the person i am
talking to is the same person i was talking to last week. i am also
willing to accept xmpp:lynX at ve.symlynX.com if necessary.

| As there are python jabbers and enxrypted jabber there is basis for a discussion, and in general: why do we need jabber with servers, if we can call cspace jabber without servers? As CSpace-ID is optional, we could make jabber servers working as well with kademlia DHT and RSA-Keys additionall, so if any jabber server is offline, the network would still work, as it is decentral !

throwing away jabber S2S isn't really such a bad idea, but i'd at least
replace it with something better than with a public DHT..

the network is decentralized.. well, if the user management is
decentralized then you cannot do any specific things for this user
at all. like if i tell my psyced server to forward all private
messages from my latest hot flirt directly by sms to my mobile
phone, how do you do that with cspace.. i suppose with a bouncer, huh?

it seems cspace is much more like IRC than like psyc/jabber really.
on IRC people do not expect their servers to do anything personal
for them. they just do what they do for everybody. and the distributed
data base operates similarely like DHT, only a lot less efficiently.

DHT isn't all uninteresting, but it doesn't just have strengths.
one thing i wanted to know, if you ask the DHT for a person's ip number,
will it tell you the ip number, or will it forward your request to the
person? the latter is okay, the first would be a major security issue.

also in DHT, you don't have a hierarchy like in DNS - so if a client
is released with a certain set of root servers with it, aren't those
"first contact" servers going to be a bottleneck?

| > cspace in Jabber to jabber client, than they should leave jabber for secure
| > Cspace protocol... of course this is very acid to jabber as here a leave of

i know better reasons to leave jabber.. ;)

| > architecture of encrypted messages is nonsense and they could to it IP to
| > IP. They only need a lookup service to get the IP adress or RSA-key of the
| > client.

yeah yeah it's all so simple in a simple world but the world isn't simple
and often you can't even connect each other ip to ip. so what are you going
to do, install proxies? or.. haha.. bouncers?

| > > Once this understood, with the first coded p2p Messenger, which is now
| > out with CSpace, we can get rid of the jabber servers.

some people love simple solutions so much, they even switch off their
brains for it. then again, some people love not simple solutions like XML.

| > > understood the revolution/turnaround?

no revolution, just yet another chat system.
this time without a proper application-specific resolution system.
as if a DHT wasn't just a complex to run as anything else.
and a DHT is oligarchic. who gets to be in charge of the DHT?

-- 
--» Carlo v. Loesch --» http://symlynX.com » psyc://psyced.org/~lynX
	        xmpp:lynX at ve.symlynX.com » irc://psyced.org/#symlynX
	      the CryptoChat » https://psyced.org/LynX/?room=symlynX