I had two Li batterys for my cheap CTX laptop. They started off lasting an hour, After six months they last twenty minutes and not long after were dead. Since I allways haul it around in its padded suitcase I got three 6Volts 10AH lead acid batterys for about E30 total which fit in a section of the bag. Added about 1Kg to the weight. The mains PSU puts out 20V DC so the 18V from the batterys runs it nicly for about 10 hours. I made a little adaptor with a diode and a resistor so I can trickle charge the batterys with the laptop psu, though after a trip I usually put the batterys on proper charger.
My laptop has a built in switching regulator and runs happily from 10.5V to 20V but some, like those Sony Vios I can't afford need, a regulated input.
I hook the three batterys together with some short wire with spade connectors. I didn't want to put an inline fuse in each wire so if I ever get the connections seriously wrong there will be some fireworks that won't be appriciated on a packed train.:-O
My years of ham radio, building electronic projects and seeing idiots weld spanners to industrial UPS battery stacks have taught me to be carefull with high-current batterys.
>introduce just one upload from my computer, and I suddenly find that my HTTP requests take forever
If you limit your upload rate to slighlty less than the maximum allowed by your connection then you will have less latency on your other connections. If you try to upload as fast a s possible the buffers in your computer/NAT router/cablemodem fill up and keystrokes in ssh sessions and suchlike have to wait for everything else in the buffer to go through first. With 128Kbit/sec up try limiting kazaa to 11 or 12KB up.
You don't need a verisgn certificate. I suggest adding an email public key field to the DNS record for domains and making new email clients that sign each message. When a message is received, the mail client can check the digital signature. To send a verified messaged with somedomain.com in the from address, you must either control the domain or be given the private key by the domain owner. Once the domain name part of an email addres can't be forged, realtime blacklists can be used to dispose of all email from that domain. I'm not sure if the cost of buy registering thousands of throwaway domains is enough to sink spammers buisness models completly but it will put it up a bit and preventing spammers putting any address on an email would make filtering much easier.
I don't think that a proper anti-spam email system will arrive in the next decade unless it grows out of an instant messaging system. The only way to change SMTP is for Microsoft to put the new protocol into outlook and then put out a version which uses it by default in three or four years.
>what if you put in a big screen and played DVDs while people were doing their laundry? That might prove a hell of lot more profitable.
Profitable for the MPAA when they sue the proprieter for unlicensed public performance of their copyrighed works.
Re:A *real* anti-leech/anti attacker system propos
on
Gnutella2?
·
· Score: 1
>Have each person maintain a list of trusted users. >Clients could automatically add a point of trust per data unit downloaded
This is where your scheme falls down. In a real p2p net you constantly interact with people you have not communicated with before. You don't exchange files with the same people often enough for "you scratch my back, I'l scratch yours" to give a worthwhile benifit.
The traffic required to gather the information needed to keep track of individual users is more of a burden than leeches. When user A uploads to user B only they know that it happened, if user A says he uploaded but user B says he didn't then who do you trust? You end up with a circular problem finding a critical mass of people you trust to rely on.
There is also the privacy/incrimination issue of tracking individual users uploads and downloads.
I thought about this a lot and tried to create algorithms for working out how much to trust someone.
I'm mainly trying to come up with a workable distributed effective anti leech system. The only solution that seems workable to me is for the users to form groups. The group members trust each other. People who want the same sort of files will group together, groups of pr0nsters, electronica fans, ebook people etc.
When people in group X have sent you more bytes than you have sent to them, you give people from group X priority in your upload queue. Group members are mostly people who upload a lot so being friends with a whole group generally works out ok.
You need some crypto so that group members can prove their membership and people can't falsly claim to be a member of a group that uploads a lot.
The implementation details will need tweaking after some actual s/w is written. My current vision is that a group grows from a single member who holds the group private key. The founder grants membership to people who upload to him. Group members send reccomendations to the founder for new members. When the group grows large it may be necessary to spread the private key to several members. I imagine a successful group having a couple of hundred members.
When someone is given upload priority because they are in a group the uploader says "you get to jump the queue because groups members x,y,z have uploaded to me". That information is shared and used within the group. To get rid of people who stop uploading after joining a group, the members form a new group with a new key every month or so, they descided amongst themselves who gets invited into the new group based on who uploaded within the group and who enhanced the groups reputation with the rest of the net. This is also necessary when the keyholder stops using the net/formats his drive etc.
People can be a member of multiple groups and new one are allways forming. You can only credit your uploads to a couple of groups so obviously you credit the groups that you have benifited from being a member of and drop membership of useless groups.
With many people only being online part time and the need to find colleagues on dynamic ips to exchange info, things like forming a new group out of an old one may take several weeks.
I bought SuSe 8.0 and found the version of KDE it installs is annoyingly slow on a K6-300.
I gather it was a new major release of KDE, if I get 8.1 is it likly to be faster?
And , no, I'm not willing to put up with the horrible UI's of all the other window managers. Nor am I going to manually install a new version of KDE, my time is worth more than a boxed set of CD's I'd go back to SuSe 6.4 but it dosn't have automatic update so I'd have to spend three days reading security bullitens and manually patching to get in into a safe state to connect to the internet, a default install gets hacked by a script kid within a day.
Oh and a hint that might save a few people some time. I loaned SuSe 8.0 to a couple of people who wanted to try out linux on old machines, they found that SuSe arn't joking about the 64MB RAM minimum spec, the installer just falls over.
Hong Kong is now part of China. I saw a rumour, can't remeber where, that there are secret undersea fibre optic cables running from Hong Kong to Macow to supply corporations with an uncensored internet connection.
>the glasgow store reboots each machine when you log off, wiping the disk
I know, that was the case in all the stores when I used them two years ago. I was refering to the two day cycle. Around 3am every day they get everyone in half the store to move to a PC in the other half then reboot all the machines including the 25% which are frozen, displaying BIOS error messages or "Slackware linux Login:"
Since nobody ever had a pen we used the handle of a Mcdonalds plastic spoon to poke through the hole in the underside of the bench to prod the power button and restart crashed machines ourselves. Unless there was a woman in a skirt on the other side of the wooden partition, in which case we got the staff to crawl on the floor for us.:-)
The CD burners in easy everything were not self service. You saved your files to a network drive then took your ticket to the front desk. The orange sweat-top clad wage slave on the till would have a look at how much data you had and offer to delete some to make it fit on one CD. It would be equivalent to going to kinko's and asking the staff to photocopy a book for you.
When they charged UKP1 (E1.5) to burn a CD the 4 writers per store were often all in use. I stopped going when they put the price up to E7.5 including a supplied blank.
Their system was fairly crap, on several ocassions I descided to leave when the trains started around 6am sunday morning having been there for a couple of days (no toilets!) and would still be there at noon after ftping my files to a different computer because the till computer could not see the network share of the original one.
I'm told the security is improved now but an aquantance of mine figured out how to install linux on the machines, the staff had no computer knowlege and never noticed us using bash on the console and the lack of the annoying banner ad bar. The firewall meant giving shells to people on IRC meant using cron and netcat to make an outgoing connection with a shell attached. They reboot all the machines every few days which copys a fresh image over the entire harddrive.
Re:This is how it works in layman's terms
on
Voices in Your Head
·
· Score: 1
I have an electronic book from the mid-1980's, Electronics Explained by Peter Laurie,
which has a circuit for a "selective shouter". That design is intended for
use with more widly spaced speakers than described in the article but its the same
basic thing. I suspect most of the innovation in Norris's design is the tight
beam ultrasonic speakers.
This relies on human ears not having an entirly linear response. Human ears
being very good at receiving sound I imagine a significant volume of ultrasound
is needed. I work in the audio industry, we use inaudible tones for
monitoring speaker lines where sound systems are used for emergency announcments.
I can say from experiance that a few watts of 20KHz left on in the workshop
gives me a headache.
Human ears are very insensative at 50KHz. 20-30Khz would be more appropriate. I think the laurie design used a 20Khz carrier and
translated 0.3-3Khz speech signals to 20.3-23Khz
I can hear loud sounds at 14-15Khz. At 20Khz I can't hear anything but standing
next to a horn speaker putting out a fw watts I can feel somthing, an I'm
getting a headache feeling, possibly my eyeballs vibrating.
A lot of open source projects could benifit from automated
testing. The wine project has spent nine years working
on translating the windows API to X and reproducing
the windows GUI.
Many
people think wine is a vital tool
for greater use of linux.
They rely on people downloading new
versions and testing windows programs using whatever
versions of X, system libraries and windows DLL's they
happen to have and then posting bug reports in a newsgroup.
There are still lots of quirks with relativly simple windows programs
not redrawing or displaying dialogs correctly.
I believe that wine would progress much faster if the wine developers
could make a change then click test and later, with no more effort
get a report saying "3 tests which failed previously now pass, 1 test
which passed on the last version now fails".
I will be very impressed if somone comes up with a free automated
testing tool that allows tests to be set up easily and is powerfull
enough to provide human readable results "the bitmap is drawn 1 pixel
to the left", "text as expected but wrong font" "dialog appeared behind
other window" etc.
If there is a bug in a commonly used part of beep such as the framing code it could cause all protocols created with it to be vulnerable.
TCP/IP and protocols running over it have had plenty of problems in the past. From machine crashing exploits like winnuke, boink, and teardrop to the reccent apache vulnerability. Problems have take a long time to be found even in widly used protocols.
I expect that people who use a medium level language instead of creating their own protocol handling routines are unlikly to spend time making serious effort to audit the compiler for security problems. The BEEP project page talks about learning previous protocol work but don't specificaly mention security that I noticed.
Ok, I can karma whore too,
Ryan Lackey, the sealand CTO has a couple of pictures on his
home page
He's a cool geezer. He has shown up at london 2600 and DNScon in the past.
Based on my observations of people I know, I won't consume anything containg aspartame. The sample size isn't big enough for the statistions but I'm convinced.
I know two people who used to drink a litre or more of diet soft drink per day, they have poor memorys, no initative and don't notice whats going on around them.
I noticed another friend was not as with it as usuall. He's normally articulate but he had started forgetting words in the middle of sentences. When quizzed he mentioned he was drinking a lot of diet drinks. This friend is also a heavy mobile user, he has the smallest model of nokia with an ineffient internal antenna, it made my ear feel hot when I used it in a place with low signal strength,though after the call the plastic case of the phone was cool. Maybe there is an alternative explanation there.
I find it annoying that aspartame is included in things I don't expect like vitamin pills and cider. It is in most brands of yogurt except natural yougurt. I sometimes have to buy natural yogurt and add my own fruit.
Anyone isn't bored of the aspartame debate yet might like to google for 'aspartame toxicity'.
> What bothers me about p2p is the fact that 1 search = 1 packet. This is not true on the gnutella network. A search query is a packet, a gnutella packet containing a search for "britney" is 15 bytes or so. Every gnutella node passes on search querys to all their peers by putting the search querys into a TCP socket going to the peer. The computers TCP/IP stack breaks up the stream of data into packets, addresse the packets to the other computer and sends then out onto the internet. With TCP overhead, about 97 searchs for "britney" can fit in one 1500 byte TCP packets.
Ultrapeers were reccently added to gnutella which has helped to make it slightly less horribly ineffient. A search query will usually be 1 per packet when talking to an ultrapeer but only over the one hop to the ultrapeer.
You mention compression, I agree that compressing the search querys and result going over gnutella connections would make a big difference.Its such a good idea it's been included in version 0.6 of the protcol spec see http://groups.yahoo.com/group/the_gdf/message/7053
Indeed, until it broke reccently I could happily read slashdot on the bus to work on the half-vga screen on my my nokia 9110. I used the built in browser for that as the charges go from E0.03/min to E0.40/min at 8am. Off peak I telnet (with s/key one time passwords) to my home PC and use lynx, IRC from the pub etc. I not sure wether to get it repaired to to pay less for a second hand one off ebay with no guarantee. I can't manage without an electronic organiser/nagging device to tell me I should have bee somewhere five minutes ago but I'm not going to carry one and a phone.
When I was in the states last year I was amazed to find I could not buy a pay-as-you-talk mobile for less than USD200. I wanted one to use for ten days then bin when I left. Here they are E45 from newsagents. Amazingly in the USA you have to pay for incoming calls to mobiles!! The mobiles have normal numbers mixed in with landlines so you don't know if you a phoning one or not.
I just give it a quick try out of interest to see if it works from behind a NAT box or when it's firewalled. The answer seems to be no, though I may just be unlucky.
On p2p filesharing networks I'v observed about two thirds of the users are firewalled/NATed and a lot of users don't have enough networking knowledge to set up port forwarding on their NAT/ICS/wingate PC or adsl router.
If node is looking for another node to get an incoming feed from, it can directly connect to firewalled nodes so it has to get a message to the firewalled node via another node that the firewalled node has open a connection to.
The webpage say it dosn't currently optimise the network topology, hopfully it will cope with firewalled nodes when that is done.
Being behind a NAT means its effectivly firewalled and even if portforwarding is set up to allow it to receive incoming connections it won't know its own internet reachable ip address, it just sees a 192.168.x.x style address (rfc1918). The filesharing client edonkey copes by having edonkey servers tell client what their IP addresses are when they connect. The author could implement that in streamer, clientclient.
ISDN at around 27ukp/month, plus 20+ for a flat-rate ISP
Tiscali's anytime package supports 64K ISDN for 15UKP/mth. Disconnects every 2 hours but they don't hassle anyone for heavy usage. Quality varies regionally, I can stay connected all weekend but some people report frequent busy signals.
You could try forging a bounce message from: Mailer-Daemon@yourdomain.
You can usually just change your email address in your mail client though there are a few ISP's don't allow outgoing email with a from line that isn't the users assigned mailbox.
Alternativly just deliver it straight to port 25, as per example;
RFC2821
me@vax:~ > host -t MX Newsletters.Microsoft.com Newsletters.Microsoft.com mail is handled (pri=10) by Newsletters.Microsoft.com
me@vax:~ > telnet Newsletters.Microsoft.com 25 Connected to Newsletters.Microsoft.com.
220 newsletters.microsoft.com Microsoft ESMTP MAIL Service, Version: 5.0.2195.4905
HELO mydomain.com 250 newsletters.microsoft.com Hello [62.64.219.7]
MAIL FROM: Mailer-Daemon@mydomain.com 250 2.1.0 Mailer-Daemon@mydomain.com....Sender OK
RCPT TO: 0_28145_1E184A2F-7C3F-D111-9D3F-0000F84121EB_CA@Ne wsletters.Microsoft.com 250 2.1.5 0_28145_1E184A2F-7C3F-D111-9D3F-0000F84121EB_CA@Ne wsletters.Microsoft.com
data 354 Start mail input; end with .
Subject: Mail delivery failed: returning message to sender
This message was created automatically by mail delivery software (Exim).
A message that you sent could not be delivered to one or more of its
recipients. This is a permanent error. The following address(es) failed:
username@mydomain.com
SMTP error from remote mailer after RCPT TO::
host mydomain [192.100.1.81]: 550 Unknown local user 'username'
.
note the empty line to seperate the subject from the body.
I'v been an ocassional user of PGP for year, first the DOS client then GPG on linux.
A friend of mine tried to use the freeware NA windows version. Hes a typical windows user and won't read instructions. After giving him a five minute talk saying "Other people use you public key to write messages to you, only you can read the message with your private key etc". Days later I call in at his house and he had not managed to use it. The user interface was horrible. Despite having used command line PGP for user and having a quick look at the help I couldn't find his keyring or work out how to use it from a quick look at the menus.
I can't imagine what the staff working on PGP were doing, certainly not useability
There were three background processes running on his already unstable win98 machine poping up box's demanding he type in his details and register. I think he reinstalled windows in the end. People who use PGP are gneerally a bit paranoid, annoying them by trying to make tem register seems pointless.
>I, for one, could save serious buckage on my electrical bill if I put a pedal-type charger under my desk while I worked.
When I last looked at a power bill some time ago electricity cost me the equivalent of about 30 cents per kilowatt hour.
Have a look at this page.
http://www.princeton.edu/pr/pwb/99/1108/powerd.s ht ml
It says students could only produce 200watts for ten minutes in tests. Allowing for mechanical and converter effiency, probably only 100watts of electricity.
Do you think 3 cents per hour of pedaling is worthwhile? Do you want to get sweaty pedaling hard while using your computer?
My 2 year old laptop draws 20 watts idling, 26 when the mouse is moving and 33 accessing the HDD.
When a batterys is charged only 2/3 of the energy is stored, the rest is lost as heat.
The numbers suggested in the article seem optomistic to me
>You type somthing wrong and it actually goes to an MSN search page I'v been cold called several times by telemarketers offering to see me links to my website when someone types somthing into the IE6 address bar that isn't a valid domain name. They want the equivalent of aprox USD150 setup fee and USD150 per phrase. I consider this to be another example of abuse of thier monopoly. They are attempting to use their control of most peoples browser to gain an unfair advantage over existing search engines and to some extent bypass the domain name system itself.
I'v used PGP about twice a year for the last 4 years when I'v come across someone else willing to make the effort to use it. I'v allways used command line clients. When I want to use it it takes a while to find which harddrive my keys are on and two minutes to read the command line options, import the keys since I will have changed distro since I last used it, encrypt or decrypt a message and I'm done. A friend who is barely computer literate ask me to help him with the PGP for win freeware he'd just installed. After a couple of minutes explanation of public keys and priivate keys I had a go with the windows program. I couldn't work out what was going on. There is a key utility that shows various keys but dosn't tell you if they are the public keys or the private keys its showing. I was lost despite knowing what I wanted to do and my friend who is a typical windows user and cannot be made to read more than once sentance of the help even when beaten with a stick had no chance. PGP freeware edition also has three or four background processes running all the time, very unwelcome on an already less than stable win98 machine. Very dissapointing, I really want more use of crypto. I'm daydreaming about a p2p app that does filesharing instant messaging, slashdot style discussion with the hard crypto and trust netwroks happing without clueless users even noticing.
Slashdot Mirror
I had two Li batterys for my cheap CTX laptop. They started off lasting an hour, After six months they last twenty minutes and not long after were dead. Since I allways haul it around in its padded suitcase I got three
:-O
6Volts 10AH lead acid batterys for about E30 total which fit in a section of the bag. Added about 1Kg to the weight. The mains PSU puts out 20V DC so the 18V from the batterys runs it nicly for about 10 hours. I made a little adaptor with a diode and a resistor so I can trickle charge the batterys with the laptop psu, though after a trip I usually put the batterys on proper charger.
My laptop has a built in switching regulator and runs happily from 10.5V to 20V but some, like those Sony Vios I can't afford need, a regulated input.
I hook the three batterys together with some short wire with spade connectors. I didn't want to put an inline fuse in each wire so if I ever get the connections seriously wrong there will be some fireworks that won't be appriciated on a packed train.
My years of ham radio, building electronic projects and seeing idiots
weld spanners to industrial UPS battery stacks have taught me to be
carefull with high-current batterys.
>introduce just one upload from my computer, and I suddenly find that my HTTP requests take forever
If you limit your upload rate to slighlty less than the maximum allowed by your connection then you will have less latency on your other connections. If you try to upload as fast a s possible the buffers in your computer/NAT router/cablemodem fill up and keystrokes in ssh sessions and suchlike have to wait for everything else in the buffer to go through first.
With 128Kbit/sec up try limiting kazaa to 11 or 12KB up.
You don't need a verisgn certificate. I suggest adding an email public key field to the DNS record for domains and making new email clients that sign each message.
When a message is received, the mail client can check the digital signature. To send a verified messaged with somedomain.com in the from address, you must either control the domain or be given the private key by the domain owner.
Once the domain name part of an email addres can't be forged, realtime blacklists can be used to dispose of all email from that domain.
I'm not sure if the cost of buy registering thousands of throwaway domains is enough to sink spammers buisness models completly but it will put it up a bit and preventing spammers putting any address on an email would make filtering much easier.
I don't think that a proper anti-spam email system will arrive in the next decade unless it grows out of an instant messaging system.
The only way to change SMTP is for Microsoft to put the new protocol into outlook and then put out a version which uses it by default in three or four years.
>what if you put in a big screen and played DVDs while people were doing their laundry? That might prove a hell of lot more profitable.
Profitable for the MPAA when they sue the proprieter for unlicensed public performance of their copyrighed works.
>Have each person maintain a list of trusted users.
>Clients could automatically add a point of trust per data unit downloaded
This is where your scheme falls down. In a real p2p net you constantly
interact with people you have not communicated with before. You
don't exchange files with the same people often enough for "you
scratch my back, I'l scratch yours" to give a worthwhile benifit.
The traffic required to gather the information needed to keep track
of individual users is more of a burden than leeches.
When user A uploads to user B only they know that it happened,
if user A says he uploaded but user B says he didn't then who do you trust?
You end up with a circular problem finding a critical mass of people
you trust to rely on.
There is also the privacy/incrimination issue of tracking individual
users uploads and downloads.
I thought about this a lot and tried to create algorithms for working
out how much to trust someone.
I'm mainly trying to come up with a workable distributed effective anti
leech system. The only solution that seems workable to me is for the
users to form groups. The group members trust each other. People who
want the same sort of files will group together, groups of pr0nsters,
electronica fans, ebook people etc.
When people in group X have sent you more bytes than you have sent to them,
you give people from group X priority in your upload queue.
Group members are mostly people who upload a lot so being friends with a
whole group generally works out ok.
You need some crypto so that group members can prove their membership
and people can't falsly claim to be a member of a group that uploads
a lot.
The implementation details will need tweaking after some actual s/w
is written. My current vision is that a group grows from a single
member who holds the group private key. The founder grants membership
to people who upload to him. Group members send reccomendations to
the founder for new members. When the group grows large it may be
necessary to spread the private key to several members.
I imagine a successful group having a couple of hundred members.
When someone is given upload priority because they are in a group
the uploader says "you get to jump the queue because groups members
x,y,z have uploaded to me". That information is shared and used
within the group. To get rid of people who stop uploading after
joining a group, the members form a new group with a new key
every month or so, they descided amongst themselves who gets invited
into the new group based on who uploaded within the group and
who enhanced the groups reputation with the rest of the net.
This is also necessary when the keyholder stops using the net/formats his drive etc.
People can be a member of multiple groups and new one are allways forming. You can only credit your uploads to a couple of groups
so obviously you credit the groups that you have benifited from
being a member of and drop membership of useless groups.
With many people only being online part time and the need
to find colleagues on dynamic ips to exchange info, things
like forming a new group out of an old one may take several
weeks.
I bought SuSe 8.0 and found the version of KDE it installs is annoyingly slow on a K6-300.
I gather it was a new major release of KDE, if I get 8.1 is it likly to be faster?
And , no, I'm not willing to put up with the horrible UI's of all the other window managers.
Nor am I going to manually install a new version of KDE, my time is worth more than a boxed set of CD's
I'd go back to SuSe 6.4 but it dosn't have automatic update so I'd have to spend three days reading security bullitens and manually patching to get in into a safe state to connect to the internet, a default install gets hacked by a script kid within a day.
Oh and a hint that might save a few people some time. I loaned SuSe 8.0 to a couple of people who wanted to try out linux on old machines, they found that SuSe arn't joking about the 64MB RAM minimum spec, the installer just falls over.
Hong Kong is now part of China. I saw a rumour, can't remeber where, that there are secret undersea fibre optic cables running from Hong Kong to Macow to supply corporations with an uncensored internet connection.
NTK mentioned this in their usual style in this edition, third paragraph in the news section.
>the glasgow store reboots each machine when you log off, wiping the disk
:-)
I know, that was the case in all the stores when I used them two years ago.
I was refering to the two day cycle. Around 3am every day they get everyone in half the store to move to a PC in the other half then reboot all the machines including the 25% which are frozen, displaying BIOS error messages or "Slackware linux Login:"
Since nobody ever had a pen we used the handle of a Mcdonalds plastic spoon to poke through the hole in the underside of the bench to prod the power button and restart crashed machines ourselves. Unless there was a woman in a skirt on the other side of the wooden partition, in which case we got the staff to crawl on the floor for us.
The CD burners in easy everything were not self service.
You saved your files to a network drive then took your ticket to the front desk. The orange sweat-top clad wage slave on the till would have a look at how much data you had and offer to delete some to make it fit on one CD. It would be equivalent to going to kinko's and asking the staff to photocopy a book for you.
When they charged UKP1 (E1.5) to burn a CD the 4 writers per store were often all in use. I stopped going when they put the price up to E7.5 including a supplied blank.
Their system was fairly crap, on several ocassions I descided to leave when the trains started around 6am sunday morning having been there
for a couple of days (no toilets!) and would still be there at noon after ftping my files to a different computer because the till computer could not see the network share of the original one.
I'm told the security is improved now but an aquantance of mine figured out how to install linux on the machines, the staff had no computer knowlege and never noticed us using bash on the console and the lack of the annoying banner ad bar. The firewall meant giving shells to people on IRC meant using cron and netcat to make an outgoing connection with a shell attached. They reboot all the machines every few days which copys a fresh image over the entire harddrive.
This relies on human ears not having an entirly linear response. Human ears being very good at receiving sound I imagine a significant volume of ultrasound is needed. I work in the audio industry, we use inaudible tones for monitoring speaker lines where sound systems are used for emergency announcments. I can say from experiance that a few watts of 20KHz left on in the workshop gives me a headache.
Human ears are very insensative at 50KHz. 20-30Khz would be more appropriate. I think the laurie design used a 20Khz carrier and translated 0.3-3Khz speech signals to 20.3-23Khz I can hear loud sounds at 14-15Khz. At 20Khz I can't hear anything but standing next to a horn speaker putting out a fw watts I can feel somthing, an I'm getting a headache feeling, possibly my eyeballs vibrating.
A lot of open source projects could benifit from automated testing. The wine project has spent nine years working on translating the windows API to X and reproducing the windows GUI.
Many people think wine is a vital tool for greater use of linux. They rely on people downloading new versions and testing windows programs using whatever versions of X, system libraries and windows DLL's they happen to have and then posting bug reports in a newsgroup. There are still lots of quirks with relativly simple windows programs not redrawing or displaying dialogs correctly.
I believe that wine would progress much faster if the wine developers could make a change then click test and later, with no more effort get a report saying "3 tests which failed previously now pass, 1 test which passed on the last version now fails".
I will be very impressed if somone comes up with a free automated testing tool that allows tests to be set up easily and is powerfull enough to provide human readable results "the bitmap is drawn 1 pixel to the left", "text as expected but wrong font" "dialog appeared behind other window" etc.
If there is a bug in a commonly used part of beep such as the
framing code it could cause all protocols created with it to
be vulnerable.
TCP/IP and protocols running over it have had plenty of
problems in the past. From machine crashing exploits like
winnuke, boink, and teardrop to the reccent apache
vulnerability. Problems have take a long time to be found
even in widly used protocols.
I expect that people who use a medium level language instead of
creating their own protocol handling routines are unlikly to
spend time making serious effort to audit the compiler for
security problems. The BEEP project page talks about learning
previous protocol work but don't specificaly mention security
that I noticed.
Ok, I can karma whore too, Ryan Lackey, the sealand CTO has a couple of pictures on his home page He's a cool geezer. He has shown up at london 2600 and DNScon in the past.
Based on my observations of people I know, I won't consume anything containg aspartame. The sample size isn't big enough for the statistions but I'm convinced.
I know two people who used to drink a litre or more of diet soft drink per day, they have poor memorys, no initative and don't notice whats going on around them.
I noticed another friend was not as with it as usuall. He's normally articulate but he had started forgetting words in the middle of sentences. When quizzed he mentioned he was drinking a lot of diet drinks.
This friend is also a heavy mobile user, he has the smallest model of nokia with an ineffient internal antenna, it made my ear feel hot when I used it in a place with low signal strength,though after the call the plastic case of the phone was cool. Maybe there is an alternative explanation there.
I find it annoying that aspartame is included in things I don't expect like vitamin pills and cider. It is in most brands of yogurt except natural yougurt. I sometimes have to buy natural yogurt and add my own fruit.
Anyone isn't bored of the aspartame debate yet might like to google for 'aspartame toxicity'.
> What bothers me about p2p is the fact that 1 search = 1 packet.
3
This is not true on the gnutella network. A search query is a packet, a gnutella packet containing a search for "britney" is 15 bytes or so. Every gnutella node passes on search querys to all their peers by putting the search querys into a TCP socket going to the peer. The computers TCP/IP stack breaks up the stream of data into packets, addresse the packets to the other computer and sends then out onto the internet. With TCP overhead, about 97 searchs for "britney" can fit in one 1500 byte TCP packets.
Ultrapeers were reccently added to gnutella which has helped to make it slightly less horribly ineffient. A search query will usually be 1 per packet when talking to an ultrapeer but only over the one hop to the ultrapeer.
You mention compression, I agree that compressing the search querys and result going over gnutella connections would make a big difference.Its such a good idea it's been included in version 0.6 of the protcol spec see http://groups.yahoo.com/group/the_gdf/message/705
Indeed, until it broke reccently I could happily read slashdot on the bus to work on the half-vga screen on my my nokia 9110. I used the built in browser for that as the charges go from E0.03/min to E0.40/min at 8am. Off peak I telnet (with s/key one time passwords) to my home PC and use lynx, IRC from the pub etc. I not sure wether to get it repaired to to pay less for a second hand one off ebay with no guarantee. I can't manage without an electronic organiser/nagging device to tell me I should have bee somewhere five minutes ago but I'm not going to carry one and a phone.
When I was in the states last year I was amazed to find I could not buy a pay-as-you-talk mobile for less than USD200. I wanted one to use for ten days then bin when I left. Here they are E45 from newsagents. Amazingly in the USA you have to pay
for incoming calls to mobiles!! The mobiles have normal numbers mixed in with landlines so you don't know if you a phoning one or not.
I just give it a quick try out of interest to see if it works from behind a NAT box or when it's firewalled. The answer seems to be no, though I may just be unlucky.
On p2p filesharing networks I'v observed about two thirds of the users are firewalled/NATed and
a lot of users don't have enough networking knowledge to set up port forwarding on their NAT/ICS/wingate PC or adsl router.
If node is looking for another node to get an
incoming feed from, it can directly connect to firewalled nodes so it has to get a message to the firewalled node via another node that the firewalled node has open a connection to.
The webpage say it dosn't currently optimise the network topology, hopfully it will cope with firewalled nodes when that is done.
Being behind a NAT means its effectivly firewalled and even if portforwarding is set up to allow it to receive incoming connections it won't know its own internet reachable ip address, it just sees a 192.168.x.x style address (rfc1918).
The filesharing client edonkey copes by having
edonkey servers tell client what their IP addresses are when they connect. The author could implement that in streamer, clientclient.
ISDN at around 27ukp/month, plus 20+ for a flat-rate ISP Tiscali's anytime package supports 64K ISDN for 15UKP/mth. Disconnects every 2 hours but they don't hassle anyone for heavy usage. Quality varies regionally, I can stay connected all weekend but some people report frequent busy signals.
I wonder if it's time to change all the links from NC4.79 to Mozilla
Please do, mozilla is a big improvement on NC4.7
You could try forging a bounce message from: Mailer-Daemon@yourdomain.
You can usually just change your email address in your mail client though there are a few ISP's don't allow outgoing email with a from line that isn't the users assigned mailbox.
Alternativly just deliver it straight to port 25, as per example;
RFC2821me@vax:~ > host -t MX Newsletters.Microsoft.com
Newsletters.Microsoft.com mail is handled (pri=10) by Newsletters.Microsoft.com
me@vax:~ > telnet Newsletters.Microsoft.com 25
Connected to Newsletters.Microsoft.com.
220 newsletters.microsoft.com Microsoft ESMTP MAIL Service, Version: 5.0.2195.4905
HELO mydomain.com
250 newsletters.microsoft.com Hello [62.64.219.7]
MAIL FROM: Mailer-Daemon@mydomain.com
250 2.1.0 Mailer-Daemon@mydomain.com....Sender OK
RCPT TO: 0_28145_1E184A2F-7C3F-D111-9D3F-0000F84121EB_CA@N
250 2.1.5 0_28145_1E184A2F-7C3F-D111-9D3F-0000F84121EB_CA@N
data
354 Start mail input; end with .
Subject: Mail delivery failed: returning message to sender
This message was created automatically by mail delivery software (Exim).
A message that you sent could not be delivered to one or more of its recipients. This is a permanent error. The following address(es) failed:
username@mydomain.com
SMTP error from remote mailer after RCPT TO::
host mydomain [192.100.1.81]: 550 Unknown local user 'username'
.
note the empty line to seperate the subject from the body.
I'v been an ocassional user of PGP for year, first the DOS client then GPG on linux.
A friend of mine tried to use the freeware NA windows version. Hes a typical windows user and won't read instructions. After giving him a five minute talk saying "Other people use you public key to write messages to you, only you can read the message with your private key etc". Days later I call in at his house and he had not managed to use it. The user interface was horrible. Despite having used command line PGP for user and having a quick look at the help I couldn't find his keyring or work out how to use it from a quick look at the menus.
I can't imagine what the staff working on PGP were doing, certainly not useability
There were three background processes running on his already unstable win98 machine poping up box's demanding he type in his details and register. I think he reinstalled windows in the end. People who use PGP are gneerally a bit paranoid, annoying them by trying to make tem register seems pointless.
>I, for one, could save serious buckage on my electrical bill if I put a pedal-type charger under my desk while I worked.
s ht ml
When I last looked at a power bill some time ago electricity cost me the equivalent of about 30 cents per kilowatt hour.
Have a look at this page.
http://www.princeton.edu/pr/pwb/99/1108/powerd.
It says students could only produce 200watts for ten minutes in tests. Allowing for mechanical and converter effiency, probably only 100watts of electricity.
Do you think 3 cents per hour of pedaling is worthwhile? Do you want to get sweaty pedaling hard while using your computer?
My 2 year old laptop draws 20 watts idling, 26 when the mouse is moving and 33 accessing the HDD.
When a batterys is charged only 2/3 of the energy is stored, the rest is lost as heat.
The numbers suggested in the article seem optomistic to me
>You type somthing wrong and it actually goes to an MSN search page
I'v been cold called several times by telemarketers offering to see me links to my website when someone types somthing into the IE6 address bar that isn't a valid domain name. They want the equivalent of aprox USD150 setup fee and USD150 per phrase. I consider this to be another example of abuse of thier monopoly. They are attempting to use their control of most peoples browser to gain an unfair advantage over existing search engines and to some extent bypass the domain name system itself.
I'v used PGP about twice a year for the last 4 years when I'v come across someone else willing to make the effort to use it. I'v allways used command line clients. When I want to use it it takes a while to find which harddrive my keys are on and two minutes to read the command line options, import the keys since I will have changed distro since I last used it, encrypt or decrypt a message and I'm done.
A friend who is barely computer literate ask me to help him with the PGP for win freeware he'd just installed. After a couple of minutes explanation of public keys and priivate keys I had a go with the windows program. I couldn't work out what was going on. There is a key utility that shows various keys but dosn't tell you if they are the public keys or the private keys its showing. I was lost despite knowing what I wanted to do and my friend who is a typical windows user and cannot be made to read more than once sentance of the help even when beaten with a stick had no chance.
PGP freeware edition also has three or four background processes running all the time, very unwelcome on an already less than stable win98 machine.
Very dissapointing, I really want more use of crypto. I'm daydreaming about a p2p app that does filesharing instant messaging, slashdot style discussion with the hard crypto and trust netwroks happing without clueless users even noticing.