>send a really powerful signal [...] and you can permanently fry that transponder,
No, you can't permanently damage a satellite this way. The path loss to a satellite in geostationary orbit is around 200dB. Estimate 50dB with a massive dish on the ground and 30dB gain on the satellite. Assume you need to get a watt of power into the satellite to physicall damage the front end. You need aproximatly 1 TeraWatt into your dish. The voltage part of the electromagnetic wave will exceed the breakdown voltage of air and you will just produce a lot of plasma in the beampath above your dish. (You can exceed the voltage breakdown of air with a really powerfull laser and get sparks in mid air.)
> and my desktop support won't upgrade the service pack - so no USB ports are useable on my machine... Eh? There is no official USB support for NT4 in any service pack. A few peripherals come with a ground up USB implementation for that specific bit of hardware. I seem to recall there is a third party add on that costs money and supports a very limited amount of hardware.
>I think they're at SP6 for that one.... SP6a actually, SP6 had some problems.
Microsoft officially stops selling NT4 licenses and providing support on the 30th of July, which is next monday.
I will still be running it for some time to come. NT4 includes version 2 of IE. IE2 is so old it dosn't support http1.1 and can't access virtual hosted sites cutting it off from a lot of the web.
With mozilla it is practical to run NT4 without installing a later version of IE. Installing IE4+ and ending up with bits of IE jammed into the system DLL's significantly slows down NT. Without IE it is pleasent to use on a 200MHZ machine given plenty of RAM.
I want a new version of NT4 with updated drivers and USB support which I would happily roll out instead of 2k. Microsoft plans to maximise revenue direct otherwise.
I'm experimenting with using a speech synthersizer to read out ebooks to listen to in the bath or while traveling. They can be compressed to really low bitrate mp3's. You can listen to an audio book while walking down the street unlike a handheld device. Less chance of missing your stop on the bus if you can stare out of the window instead of at a handheld.
Any reccomendations for good freeware text to speech programs? I'm trying out freetts at the moment.
Re:This is obvious
on
42-Volt Autos
·
· Score: 2, Insightful
I remember an article in an IEE magazine about this a couple of years ago.
A nominal 12V car battery produces about 13.8V when fully charged. When it is being charged the voltage across the terminals can be as high as 15.5volts. The voltage also varies with temperature and load.
Electrical systems must meet more stringent safety requirements if they use 50V or more. 42V was chosen to keep within the 50V limit during charging.
Some people are confused about the reasons for a higer voltage. The amount of power used is volts*amps. With three times the voltage things like head lights that will be made to use the same power will require a third of the current so the wires can be thinner.
Simplifying somewhat, with a 12v supply and an 4ohm speaker a transformerless amplifier can supply about 24watts rms (0.707*v^2/R), which is more than enough for everyone except boy racers. Move to 42V and an amp can supply 72 watts per channel (marketing will call it 150Watts music power) without an expensive and bulky lump of iron and copper.
IP6 allocations are not permanent, you don't own ip6 addresses and you can't get PI(provider independent) blocks. To get a range of ip6 addresses you have to get them from your ip6 gateway provider or be a big or important enough network operator or institution.
Slashdot has inserted some spaces into the key I posted in the parent message, to import my key please remove the spaces, including any
whitespace in front of WASTE_PUBLIC_KEY
I spoke too soon, It's just a display bug. I didn't realise how the key management(or lack of) works.
Instead of just disconnecting and reconnecting with no obvious clue about what wrong waste should pop up a box saying "You have the other clients public key be he dosn't have yours!" or vice versa.
There is lots fo room to improve the key management in Waste
It looks like Nullsoft won't be accepting bug reports for waste so I'l mention my bug list here.
It looks like it is using a signed 16bit variable for the port number somwhere, if a client is set to use a port above 32767 it gets treated as a negative number and other clients can't connect.
A minor one is the the port number setting is not saved if you change it then close the settings dialog without pressing the update button.
I'l put my temporary waste node on a different prt, see my journal.
I'd like to play with Waste and see how well it works with a group of people spread across the internet. Would anyone like to form a waste network of random slashdotters? I'm busy now so this will probably not start rolling until the coming weekend.
Leave a message in my slashdot journal or use my current throwaway email address zaphodbond@yahoo.com
I suspect the thing we need most is someone with a static ip or domain name to announce thier public key and leave waste running continuously for a few days.
>P2P wastes bandwidth. As in efficiency. How much bytes traverse >the wire to download 100 MB via FTP? via HTTP? via scp? >Compare that to P2P.
It vaires considerably been filesharing systems. My rough guess based on watching various p2p clients, including primary/supernode traffic, not including TCP overhead, assuming long term use is
gnutella/0.4 1-2Gbyte traffic for 100Mb download winmx/kazaa/G2MP 150-250MB edonkey 120MB
If you compare to the www you need to consider how many web searchs and pages people have to go to to find the file or information they want.
>H2k also had an "orgy" that flopped All the H2K participants who were vaguly in posession of some mental faculitys worked out that there was zero chance of any women showing up. About three people bought a tshirt and turned up for the orgy, plus the two organisers and a couple of dozen feds and cops who wanted to ensure no minors were involved.
And to stay on topic Mr Draper did not attend.
Oh, and I heard it was a quarter stick of dynamite. I think, at the time I was in the Blarney Rock watching Cyberjunkie beat all comers in the who can drink(neck) a four pint jug of beer fastest knockout competition against the CDC and the Chaos Computer Club.
Those of you who are interested in the development of peer to peer
systems such as bittorrent may be interested in the Codecon conference
which took place last month. There were some very interesting panels.
Bram Cohen the author of bittorrent is also the main codecon organisner.
The audio recording of the talks and panels at codecon can be downloaded with
bittorrent. It maxed my downstream at 50KB/sec, someone else reported 200KB/s down.
I find Microsoft's explaination for not fixing this RPC problem unconvincing. I suspect that if they wanted to they could add a check for
malformed packets in wahtever bit of code listens on port 135. It might not be pretty or high performance but I think it would work. Any experts on windows architecture reading?
NT4 is my favorite version of windows. I keep a sacrficial install around to test new software. By being carefull about what gets installed I'v had uptimes of 100+ days from NT machines and reboots are usually hardware related. It is possible to run NT4 without IE4/5/6 so you don't have IE intergrated into all the system dll's bogging it down.
NT4 workstation is available cheaply. At large computer shows there is usually a trader with a few cd+license packs for about E25 each.
I hope to use NT4 for another five years or so, until I can't buy hardware with NT support.
Hang on a sec, we could really do with more work for geeks at the moment. If a load of corporations are pushed into upgrading their fleets of NT4 machines, with all the attendant problems that go with buggering about with computers, that means more work for geeks. Yah microsoft! Where's that alpha copy of windows longhorn...
I love moz as a browser. I rarely mention the mail client bugs because I should go and look at the source and help instead of complaining. There are lots of good clients around and my personal experiance is that a text only email client is better, firstly because of security holes (mainly thinking of outlook) and secondly because loading images in spam email shows up in the logs on the spammers web server, telling them the email has been read and resulting in lots more spam.
It will be a very good thing if minotaur fix the mozilla mail client and get the fixes back into the mozilla tree.
bugzilla.mozilla.org lists bugs where the mail client fails to retreive mail which have been open for years. I stopped using mozilla as a mail client due to bug 58301 which has been open for nearly three years.
1) Shareaza works reasonably well though the lack of a usefull anti leech mechanism is a drawback. A leaf and hub architecture is a reasonable compromise, it usually dosn't allow searching of the whole network. A distributed hash table architecture would allow global search but DHT is vulnerable to an attacker with modest resources flooding the search space with junk.
2)I'm interested in the reliable communiaction over UDP part of the protocol.
Two peers, neither of whom accept incoming connections due to connection sharing or firewalling can't exchange files with any current p2p software.
I'm a bit surprised that no p2p project has tried to do UDP connection splicing to allow two peers, both behind internet connection sharing (NAT) to talk to each other. It not be possible but I havn't come across anyone who has tried at all. It dosn't have to work for everyone to be usfull. Just allowing 10% of NATed hosts to communicate would be worth the effort.
NAT routers usually allow outgoing UDP connections, normally the remote machine will be listening on a UDP prt (most common use is port 53 for domain name lookups. The local computer sends out a packet with a unique local port number (sequentially or randomly assigned) and a remote port of 53. When a UDP reply arrives from the internet, it also has a remote port number of 53 and the chosen local port number. The connection sharing machine looks at the local port number and compares it to a table of known connections. If it matches an entry it knows which of the computers behind the NAT it is intended for and forwards it.
In theory, if two NATed hosts send out UDP packets at the same time, using the same port numbers the connection will penetrate the NATs. It requires the help of another machine but in a p2p system you have plenty of machines that can pass on some small messages.
The only problem I can see is that a NAT may change the local port number. I hope that cheap cable/dsl routers only change the local port if necessary, or if they do will use predicatable numbers.
This technique will also be usfull for getting through statefull firewalls.
I need to research this more. Am I missing somthing? I'm aware that shifting files over UDP requires TCP compatible flow control so it dosn't hog all the bandwidth.
3)Due to the number of people behind transparent proxys that hijack all outgoing connections to port 80 and the number of people who set their filesharing clients to listen on 80, I think is is a pity Mike didn't take the opportunity to add a faculity to encapsulate traffic with a fake HTTP GET header so it will go through http-only proxys. Kazaa does somthing like this.
>as if there can only be one message per frequency
There are already systems allowing radio users such as
taxi's and security guards to use the same frequencys.
The same frequency is often allocated to firms in
geographically seperate locations. A system called CTCSS
is used so that even if a signal from the base
transmitter of a building reaches the walkie talkie
of a security guard miles away it dosn't come out
of the speaker. CTCSS sends a low frequency tone along
with the voice, the receivers only turn on the audio
output when the correct tone is detected.
Security guards don't talk on their radio all the time
and the wanted signal are usually closer and stronger
so it works well.
Digital trunked radio systems, similar to cellular phone systems are also gaining ground.
Microsoft purchased 23.6 percent of the British cable TV company Telewest for $2.6B a few years ago. The current price of telewest shares values the whole company at about $100M. Considering Telewest is billions in dept I think it is still overvalued, zero would be closer to the mark.
From the point of view of building wireless communitys, you have to have every non-ubergeek with a windows machine, a wireless card and a white stick antenna acting as a forwarding node. That means a packaged solution that just means running setup.exe and having "meshnet" appear in control panel->network with a few easy options eg community name(default), public/private, enable forwarding.
The end user should just see it as a window that shows what machines are in range, allowing browsing of drive shares, personal web servers etc.
It absolutly has to support current ip4 applications, eg filesharing, IRC etc.
Allocating ip address is slightly tricky. You can't rely on any one node to allocate addresses. The radio hams have a whole class A allocated to them, 44.x.x.x
Due to regulations and little use of tcp/ip by hams there is nothing reachable from the internet in this block. Packets to 44.x.x.x are routed to the university of California and dropped, the half of the block suggested for use outside the USA will certainly never be reachable.
Wireless networking is radio done by amateurs, just not specifically licenced ones. My idea is to for everyone to simply chose a random address in 44.x.x.x with a 255.255.255.252 netmask and let software sort out the routing. Forwarding ip-broadcast traffic isn't practical in a city-wide network, so we just have to accept that machines won't appear in network neighbourhood. Of course clusters of friends close to each other can subnet themselves together, assuming they have someone who understands the details of ip networking.
We could use 10.x.x.x but thats bound to end up conflicting with somones home/company/school LAN.
As for giving machines names, going back to hosts files would be a big step back, we need a distributed peer to peer name lookup solution so that when someone using the name "Fred.consume.net" is a couple of hops away you can find them.
Nodes providing internet connection would do NAT. You would either set your default gateway to the 44 address of your prefered connection or just let the routing software choose the best route out.
All the individual problems needed to do this have been solved well already. All the work that has gone into creating the like of the BGP routing protocol can be reused for this.
I havn't bneen able to find CHALKBOARD using a few search engines, it's too common a term. Anyone have a link to this software please? zaphodbond@yahoo.com
Slashdot Mirror
>send a really powerful signal [...] and you can permanently fry that transponder,
No, you can't permanently damage a satellite this way.
The path loss to a satellite in geostationary orbit is around 200dB. Estimate 50dB with a massive dish on the ground and 30dB gain on the satellite. Assume you need to get a watt of power into the satellite to physicall damage the front end.
You need aproximatly 1 TeraWatt into your dish. The voltage part of the electromagnetic wave will exceed the breakdown voltage of air and you will just produce a lot of plasma in the beampath above your dish. (You can exceed the voltage breakdown of air with a really powerfull laser and get sparks in mid air.)
> and my desktop support won't upgrade the service pack - so no USB ports are useable on my machine...
Eh? There is no official USB support for NT4 in any service pack. A few peripherals come with a ground up USB implementation for that specific bit of hardware. I seem to recall there is a third party add on that costs money and supports a very limited amount of hardware.
>I think they're at SP6 for that one....
SP6a actually, SP6 had some problems.
Microsoft officially stops selling NT4 licenses
and providing support on the 30th of July, which is next monday.
I will still be running it for some time to come.
NT4 includes version 2 of IE. IE2 is so old it dosn't support http1.1 and can't access virtual hosted sites cutting it off from a lot of the web.
With mozilla it is practical to run NT4 without installing a later version of IE. Installing IE4+ and ending up with bits of IE jammed into the system DLL's significantly slows down NT. Without IE it is pleasent to use on a 200MHZ machine given plenty of RAM.
I want a new version of NT4 with updated drivers and USB support which I would happily roll out instead of 2k. Microsoft plans to maximise revenue direct otherwise.
Any reccomendations for good freeware text to speech programs? I'm trying out freetts at the moment.
I remember an article in an IEE magazine about this a couple of years ago.
A nominal 12V car battery produces about 13.8V when fully charged. When it is being charged the voltage across the terminals can be as high as
15.5volts. The voltage also varies with temperature and load.
Electrical systems must meet more stringent safety requirements if they use 50V or more. 42V was chosen to keep within the 50V limit during charging.
Some people are confused about the reasons for a higer voltage. The amount of power used is volts*amps. With three times the voltage things like head lights that will be made to use the same power will require a third of the current so the wires can be thinner.
Simplifying somewhat, with a 12v supply and an 4ohm speaker a transformerless amplifier can supply about 24watts rms (0.707*v^2/R),
which is more than enough for everyone except boy racers.
Move to 42V and an amp can supply 72 watts per channel (marketing will call it 150Watts music power) without an expensive and bulky lump of iron and copper.
IP6 allocations are not permanent, you don't own ip6 addresses
and you can't get PI(provider independent) blocks.
To get a range of ip6 addresses you have to get them from your
ip6 gateway provider or be a big or important enough network operator
or institution.
Maharito is talking nonsense, ethernet is very well standardised at the link layer. There is no problem at all mixing different brands of NIC.
I posted it in my journal as well here
I spoke too soon, It's just a display bug.
I didn't realise how the key management(or lack of) works.
Instead of just disconnecting and reconnecting with no obvious clue about what wrong waste should pop up a box saying "You have the other clients public key be he dosn't have yours!" or vice versa.
There is lots fo room to improve the key management in Waste
WASTE_PUBLIC_KEY 20 1024 Throwaway182 45ABE25118CBA 3B5B071656D 3D8F878C43B8 606B77406F2 6E31C15F2F15 62F7CE1AD3B 3798EB8FBCCB A4621EE811L IC_KEY_END
E29A094E7E3D5C85FF1A145C38B134E6BEDD
0357CA29FDE2CBE2B9B143CB638578615295D
F984DB513A72CE0F7B7A80DC3DF6E63390771
455208F5CEC43DEFF4782624794A12A36FBE4
10684BE68E1ECCAB0003010001
WASTE_PUB
It looks like Nullsoft won't be accepting bug reports for waste so I'l
mention my bug list here.
It looks like it is using a signed 16bit variable for the port number somwhere, if a client is set to use a port above 32767 it gets treated
as a negative number and other clients can't connect.
A minor one is the the port number setting is not saved if you change it then close the settings dialog without pressing the update button.
I'l put my temporary waste node on a different prt, see my journal.
I'd like to play with Waste and see how well it works with a group of people spread across the internet. Would anyone like to form
a waste network of random slashdotters?
I'm busy now so this will probably not start rolling until the
coming weekend.
Leave a message in my slashdot journal or use my current
throwaway email address zaphodbond@yahoo.com
I suspect the thing we need most is someone with a static
ip or domain name to announce thier public key and leave
waste running continuously for a few days.
I'm not using a network name in waste.
>P2P wastes bandwidth. As in efficiency. How much bytes traverse
>the wire to download 100 MB via FTP? via HTTP? via scp?
>Compare that to P2P.
It vaires considerably been filesharing systems. My rough guess based on watching various p2p clients, including primary/supernode traffic, not including TCP overhead, assuming long term use is
gnutella/0.4 1-2Gbyte traffic for 100Mb download
winmx/kazaa/G2MP 150-250MB
edonkey 120MB
If you compare to the www you need to consider how many
web searchs and pages people have to go to to find the file or information they want.
There is a BBC news article about Welsh Police impounding cars because the owners used cooking oil as fuel without paying fuel tax here.
>H2k also had an "orgy" that flopped
All the H2K participants who were vaguly in posession of some mental
faculitys worked out that there was zero chance of any women showing up.
About three people bought a tshirt and turned up for the orgy,
plus the two organisers and a couple of dozen feds and cops who
wanted to ensure no minors were involved.
And to stay on topic Mr Draper did not attend.
Oh, and I heard it was a quarter stick of dynamite. I think, at the time I was
in the Blarney Rock watching Cyberjunkie beat all comers in the who can drink(neck) a four pint
jug of beer fastest knockout competition against the CDC and the Chaos Computer Club.
Perhaps they can steal a few good ideas from shouldexist.org
Bram Cohen the author of bittorrent is also the main codecon organisner. The audio recording of the talks and panels at codecon can be downloaded with bittorrent. It maxed my downstream at 50KB/sec, someone else reported 200KB/s down.
The current version of maxblast, version 3 only works if a maxtor drive is installed in the machine.
I find Microsoft's explaination for not fixing this RPC problem unconvincing. I suspect that if they wanted to they could add a check for malformed packets in wahtever bit of code listens on port 135. It might not be pretty or high performance but I think it would work. Any experts on windows architecture reading?
NT4 is my favorite version of windows. I keep a sacrficial install around to test new software. By being carefull about what gets installed I'v had uptimes of 100+ days from NT machines and reboots are usually hardware related. It is possible to run NT4 without IE4/5/6 so you don't have IE intergrated into all the system dll's bogging it down.
NT4 workstation is available cheaply. At large computer shows there is usually a trader with a few cd+license packs for about E25 each.
I hope to use NT4 for another five years or so, until I can't buy hardware with NT support.
Hang on a sec, we could really do with more work for geeks at the moment. If a load of corporations are pushed into upgrading their fleets of NT4 machines, with all the attendant problems that go with buggering about with computers, that means more work for geeks. Yah microsoft! Where's that alpha copy of windows longhorn...
It will be a very good thing if minotaur fix the mozilla mail client and get the fixes back into the mozilla tree.
bugzilla.mozilla.org lists bugs where the mail client fails to retreive mail which have been open for years. I stopped using mozilla as a mail client due to bug 58301 which has been open for nearly three years.
1) Shareaza works reasonably well though the lack of a usefull anti leech
mechanism is a drawback. A leaf and hub architecture is a reasonable compromise,
it usually dosn't allow searching of the whole network.
A distributed hash table architecture would allow global search but DHT is
vulnerable to an attacker with modest resources flooding the search space with
junk.
2)I'm interested in the reliable communiaction over UDP part of the protocol.
Two peers, neither of whom accept incoming connections due to connection sharing
or firewalling can't exchange files with any current p2p software.
I'm a bit surprised that no p2p project has tried to do UDP connection splicing
to allow two peers, both behind internet connection sharing (NAT) to talk to
each other. It not be possible but I havn't come across anyone who has tried at all.
It dosn't have to work for everyone to be usfull. Just allowing 10% of
NATed hosts to communicate would be worth the effort.
NAT routers usually allow outgoing UDP connections, normally the remote
machine will be listening on a UDP prt (most common use is port 53 for domain
name lookups. The local computer sends out a packet with a unique local port
number (sequentially or randomly assigned) and a remote port of 53.
When a UDP reply arrives from the internet, it also has a remote port number
of 53 and the chosen local port number. The connection sharing machine
looks at the local port number and compares it to a table of known connections.
If it matches an entry it knows which of the computers behind the NAT
it is intended for and forwards it.
In theory, if two NATed hosts send out UDP packets at the same time, using
the same port numbers the connection will penetrate the NATs. It requires the
help of another machine but in a p2p system you have plenty of machines that
can pass on some small messages.
The only problem I can see is that a NAT may change the local port number.
I hope that cheap cable/dsl routers only change the local port if necessary,
or if they do will use predicatable numbers.
This technique will also be usfull for getting through statefull firewalls.
I need to research this more. Am I missing somthing?
I'm aware that shifting files over UDP requires TCP compatible flow control
so it dosn't hog all the bandwidth.
3)Due to the number of people behind transparent proxys that hijack all outgoing
connections to port 80 and the number of people who set their filesharing
clients to listen on 80, I think is is a pity Mike didn't take the opportunity
to add a faculity to encapsulate traffic with a fake HTTP GET header so it will
go through http-only proxys. Kazaa does somthing like this.
There are already systems allowing radio users such as taxi's and security guards to use the same frequencys.
The same frequency is often allocated to firms in geographically seperate locations. A system called CTCSS is used so that even if a signal from the base transmitter of a building reaches the walkie talkie of a security guard miles away it dosn't come out of the speaker. CTCSS sends a low frequency tone along with the voice, the receivers only turn on the audio output when the correct tone is detected.
Security guards don't talk on their radio all the time and the wanted signal are usually closer and stronger so it works well.
Digital trunked radio systems, similar to cellular phone systems are also gaining ground.
Microsoft purchased 23.6 percent of the British cable TV company Telewest for $2.6B a few years ago. The current price of telewest shares values the whole company at about $100M. Considering Telewest is billions in dept I think it is still overvalued, zero would be closer to the mark.
From the point of view of building wireless communitys, you
have to have every non-ubergeek with a windows machine,
a wireless card and a white stick antenna acting as a
forwarding node. That means a packaged solution that
just means running setup.exe and having "meshnet" appear
in control panel->network with a few easy options eg
community name(default), public/private, enable forwarding.
The end user should just see it as a window that shows
what machines are in range, allowing browsing of drive
shares, personal web servers etc.
It absolutly has to support current ip4 applications,
eg filesharing, IRC etc.
Allocating ip address is slightly tricky. You can't rely
on any one node to allocate addresses.
The radio hams have a whole class A allocated to them,
44.x.x.x
Due to regulations and little use of tcp/ip by hams
there is nothing reachable from the internet in this
block. Packets to 44.x.x.x are routed to the university
of California and dropped, the half of the block
suggested for use outside the USA will
certainly never be reachable.
Wireless networking is radio done by amateurs, just
not specifically licenced ones.
My idea is to for everyone to simply chose a random
address in 44.x.x.x with a 255.255.255.252 netmask
and let software sort out the routing. Forwarding
ip-broadcast traffic isn't practical in a city-wide
network, so we just have to accept that machines won't
appear in network neighbourhood. Of course clusters of
friends close to each other can subnet themselves
together, assuming they have someone who understands
the details of ip networking.
We could use 10.x.x.x but thats bound to end up conflicting
with somones home/company/school LAN.
As for giving machines names, going back to hosts files
would be a big step back, we need a distributed peer to
peer name lookup solution so that when someone using
the name "Fred.consume.net" is a couple of hops away you
can find them.
Nodes providing internet connection would do NAT.
You would either set your default gateway to the
44 address of your prefered connection or just let
the routing software choose the best route out.
All the individual problems needed to do this have been
solved well already. All the work that has gone into
creating the like of the BGP routing protocol can
be reused for this.
I havn't bneen able to find CHALKBOARD using a few search engines, it's too common a term. Anyone have a link to this software please?
zaphodbond@yahoo.com