The title of the article is wrong. This virus just sends SMS messages, it doesn't call anywhere. (For US readers: in GSM you have something like a pager every mobile, it's called Short Message Service.) It's very easy to spam mobiles, using GSM operators' e-mail or WWW gateways, especially when the e-mails/WWW requests come from different computers of the virus victims and the mobile e-mail addresses are easy to guess, like 123456@sms.yourgsmoperator.com..
Well, to get reasonable coverage, you need lots of investments to the base stations, so the only real possibility now is a combination of a GSM mobile plus some handheld device, or a GSM+handheld in one, like Nokia 9110i. You can make a "data call", i.e. use the mobile as a modem; the Czech providers let you dial-in to their "modems" to get internet for free. Well, for free -- you pay their data charges, which are now some 5c/minute. You get typically 9600 bps, some mobiles and operators do 14.4 kbps, and the Nokia PCMCIA card does even 28.8 kbps full duplex, or 42 kbps in and 14 kbps out. These speeds are very theoretical, depend on the load in the GSM network, which varies a lot.
Well, you dialectize or translate the page for your own use. Like if you had a book in German, and translated it for yourself on a piece of paper. These services don't do any "dissemination", they do it just for the person reading it. On the other hand, what if somebody registered "Sweedish Chef CNN" at AltaVista?
Raging: Text-only Page We're sorry. Raging.com does not support text-only browsers. For a superior text-only search, please visit AltaVista. Why do they need an extra text version for a lightweight search engine, which is ideal for text browsers as it stands, anyway? And besides, this appears only in lynx, it works fine in links and w3m (in case you don't know, cool lynx-like text browsers that can do tables, look them up at freshmeat).
A friend of mine wanted to use it for some typical web-database-publishing system thing, but found out, that the maximum size of data fields (except for BLOBs) is 8kB. Yeah, no longer-than-8kB-articles. It's so hardcoded in Postgres, that you cannot raise this limit without a complete rewrite. At least this was true at the time when he was trying it, perhaps now this limitation has been lifted. (Anybody correct me?)
That is true. I recall seeing a czech construction or building company called UNIX. The people there for sure didn't know anything about operating systems.. Well, perhaps these bricklayers will buy it.:-) BTW, does anybody know any company called Windows? Like, er, glaziers who started their shop in 1995?:-)
Yes. It gets congested even now, even with ETSI power limits, which are much tougher than FCC's, especially in cities in countries where public-band microwave equipment is used a lot. Here (Czech Republic) ISPs use point-to-mutipoint 2.4 GHz microwave to connect leased-line customers without goint through expensive telco, and the air is slowly getting congested. The next step is to obtain private frequencies and not use public band except for "clear" areas..
Interesting that use of this frequency requires frequency hopping, anyone who knows more than me about what this means for interferrence between units?
Well, it does not *require* frequency hopping. In 2.4 GHz, there are both DSSS and FHSS radios.
DSSS (Direct Sequence Spread Spectrum): You basically you statically chose one channel (out of a dozen or so) and then all the communication goes through the chanell, "modulated" by the data, so on chanell #1, 0xff always gets sent on 2.43285938 GHz, and if there's some interference on that frequency (e.g. somebody else is using it), you just cannot transmit 0xff.
FHSS (Frequency Hopping Spread Spectrum): You typically broadcast in whole band (it is allowed by FCC/ETSI regulations), but only few ms at a given frequency. Then you move to another frequency, given by a special sequence. There are lots of these sequences, made up so they interfere very little with each other (e.g. 1 3 4 6 and 1 2 5 3, see that they meet only at 1, in real life there's IIRC 80 frequencies not four as in my example). Kind of like the joke about 1 second late watch never being right. Besides, even if you chose the same sequence as somebody else, if you don't turn your equipment on in exactly the same time, you never cross and both can use full bandwidth the equipment can do.
It's clear that FHSS is much resistant to interference, especially from DSSS pieces of equipment. If you have a DSSS and FHSS equipment sharing same space, DSSS usually dies but FHSS just gets a bit slower.
Again, I am trying to simplify, please don't take this literally. (Yes, 2.43285938 GHz for 0xff in channel #1 is probably wrong.)
Well, I work for an ISP which does wireless Internet access over BreezeNet. I wonder what's so special about enlarging gain and thus reach by adding a high-gain antenna to the equipment. It's possible to do do 40-50 km links with 24-30 dB antennas. Even a bit more with an input amplifier. I don't know the prices of WaveLan cards, or this Apple stuff that apparently has WaveLan in them, because we don't use them as they are very prone to interference (especially somewhere with FHSS traffic), but prices of a pair of BreezeNets and two 24 dB antennas might be something like $1700. And there are even cheaper devices than BreezeNet.
So, "Cheap Long Distance Wireless Networking" already exists. Damned, I am using it as I write this.:-)
There's one caveat though. The FCC/ETSI regulations. From what I know, in the US there's some power limit on the transmitter, and then you basically can use whatever high-gain antenna you like. In Europe, ETSI regulations regulate the total output of the system (the device + antenna gain), so you cannot do very long distance links with this technology, because the larger (=higher-gain) you have, the less power you are allowed to transmit through it..
If you live in ETSI countries, there's a possibility to have fairly cheap high-bandwidth (8 Mbps full-duplex) point to point links over big distances using devices in public 10.3 GHz band (as opposed to 2.4 GHz typically used by BreezeNet, WaveLan et al). This is a bit more expensive (say, $10000 for both points). Here the main manufacturers are CoproSys, Miracle or Alcoma.
I have tried to simplify things, experts please forgive me.
PS It's plain silly to use 2.4 GHz equipment through trees, as the author of the article mentions - leafs block the signal in this band a lot.
Actually, there have been some quarrels between SCO and Microsoft recently. Microsoft wanted them to still maintain compatibility with old Xenix in their OpenServer product, because that's what was in the deal when MS sold Xenix to SCO, but SCO complainted this is useless (who wants Xenix compatibility nowadays)..
Though, you have to consider, that "only telephone costs" might be fairly high. So high, that the price of the dialup access is pretty irreleveant. You pay per minute, I don't know exactly how much in Germany, but here it's something like $2 an hour in peak, and this is the "cheaper" Internet tariff.
nothing much new here, but yet another perspective
on
Perens on Patents
·
· Score: 1
How come there's nothing much new, when it's yet another perspective ?? I thought another perspective also counts as something new..
Well, I haven't tried this 6.2 beta yet, but after 6.1 I am starting to get more and more of the impression that it's going way too much Windows direction. It's no longer about understanding things and how they work, but rather about knowing "when it says this it in fact means this-and-that" and "if you want to do this (e.g. fdisk) you must pretend to do something else to get the first thing done". See also "Insert Driver disk", missing descriptions of packages in 6.1 text install, overall feeling of losing control over the installation of the system, unhelpful help (or none at all) etc.. So in this light it's a piece of good news that the installer got better. To your "root in xterm" problem - just do xhost +root@localhost or something to that effect before you do your su, DON'T do xhost + as it could be pretty dangerous (anybody could send you an X client which would grab your keyboard, snoop your keys and whatnot)..
Is there "the newest QT" codec library for xanim, i.e. in binary form? I think, that given that codecs are usually patented etc. that's probably the best thing we can get -- now. Yes, source or specs would be better, but I doubt they will ever release them..
Yes, you are right, some chess programs do use a kind of pre-selection.. From what I remember from the times I have played chess profesionally (~10 years ago), the chess software works (or used to work those ~ten years ago) this way: it computes all possible positions generated by its move(s) up to some level. Then it evaluates all the positions with some "static function", i.e. counts it's material, centralized King in an ending etc., so each position gets it's ranking. Then the best one is chosen. Now, there are two schools: "pre-selection" and "brutte force". Pre-selection school generates the rankings only few moves ahead, and just trashes positions which look completely silly (i.e. have too low ranking), and only pursues further the promising candidates. The brute force school evaluates the full tree, sometimes getting a really great move. Now, entry-level chess programs use brute force, better programs use the pre-selection method, but the best ones brutte force. Or this is what I learned those 10 years ago, and this is oversimplified, you have opening libraries, endings specialities etc. I wonder what does this add to our discussion of brute force in optimization.:-)
I don't know if you were referring to this (as there's way too many interesting things in KDE 2.0, like the teapot, which pops up a window and beeps when you should take the teabag out of your tea:-), but I saw there beginning of someting what looked too much like the Windows' device manager thing to my taste. You click on Properties, see "Device is operating normally" etc. On the other hand, Windows users would then have NO excuse not to use KDE.:-)
I absolutely agree. I work for an ISP, too, and almost 5% of our customers has switched the routers on their side off for this weekend, most of them probably due to Y2K worries..:-(
Perhaps more troublesome is that GPL (and other free licenses) could be fairly problematic in certain legal systems, namely european continental (read: non-british) law, which normatively orders to have certain things even if you disclaim them.
For example, the disclaimer of warranty could be a problem, GPL isn't good enough as a contract in some legal systems, the author cannot agree to have no compensation for his work, gift taxes problems etc.
I am no lawyer and don't know much about other countries' law, but at least here in Czech Republic the GPL is fairly problematic. (See here, sorry, in Czech only.)
Slashdot Mirror
The title of the article is wrong. This virus just sends SMS messages, it doesn't call anywhere. (For US readers: in GSM you have something like a pager every mobile, it's called Short Message Service.)
It's very easy to spam mobiles, using GSM operators' e-mail or WWW gateways, especially when the e-mails/WWW requests come from different computers of the virus victims and the mobile e-mail addresses are easy to guess, like 123456@sms.yourgsmoperator.com..
Well, to get reasonable coverage, you need lots of investments to the base stations, so the only real possibility now is a combination of a GSM mobile plus some handheld device, or a GSM+handheld in one, like Nokia 9110i. You can make a "data call", i.e. use the mobile as a modem; the Czech providers let you dial-in to their "modems" to get internet for free. Well, for free -- you pay their data charges, which are now some 5c/minute. You get typically 9600 bps, some mobiles and operators do 14.4 kbps, and the Nokia PCMCIA card does even 28.8 kbps full duplex, or 42 kbps in and 14 kbps out. These speeds are very theoretical, depend on the load in the GSM network, which varies a lot.
Well, you dialectize or translate the page for your own use. Like if you had a book in German, and translated it for yourself on a piece of paper. These services don't do any "dissemination", they do it just for the person reading it. On the other hand, what if somebody registered "Sweedish Chef CNN" at AltaVista?
Raging: Text-only Page
We're sorry. Raging.com does not support text-only browsers. For a superior text-only search, please visit AltaVista.
Why do they need an extra text version for a lightweight search engine, which is ideal for text browsers as it stands, anyway? And besides, this appears only in lynx, it works fine in links and w3m (in case you don't know, cool lynx-like text browsers that can do tables, look them up at freshmeat).
Imagine:
A friend of mine wanted to use it for some typical web-database-publishing system thing, but found out, that the maximum size of data fields (except for BLOBs) is 8kB. Yeah, no longer-than-8kB-articles. It's so hardcoded in Postgres, that you cannot raise this limit without a complete rewrite. At least this was true at the time when he was trying it, perhaps now this limitation has been lifted. (Anybody correct me?)
That is true. I recall seeing a czech construction or building company called UNIX. The people there for sure didn't know anything about operating systems.. Well, perhaps these bricklayers will buy it. :-) :-)
BTW, does anybody know any company called Windows? Like, er, glaziers who started their shop in 1995?
OOPS, a typo. :-)
I'd add the wireless Linux resources page at http://www.hpl.hp.com/persona l/Jean_Torrilhes/Linux/.
Yes. It gets congested even now, even with ETSI power limits, which are much tougher than FCC's, especially in cities in countries where public-band microwave equipment is used a lot. Here (Czech Republic) ISPs use point-to-mutipoint 2.4 GHz microwave to connect leased-line customers without goint through expensive telco, and the air is slowly getting congested. The next step is to obtain private frequencies and not use public band except for "clear" areas..
Well, it does not *require* frequency hopping. In 2.4 GHz, there are both DSSS and FHSS radios.
DSSS (Direct Sequence Spread Spectrum): You basically you statically chose one channel (out of a dozen or so) and then all the communication goes through the chanell, "modulated" by the data, so on chanell #1, 0xff always gets sent on 2.43285938 GHz, and if there's some interference on that frequency (e.g. somebody else is using it), you just cannot transmit 0xff.
FHSS (Frequency Hopping Spread Spectrum): You typically broadcast in whole band (it is allowed by FCC/ETSI regulations), but only few ms at a given frequency. Then you move to another frequency, given by a special sequence. There are lots of these sequences, made up so they interfere very little with each other (e.g. 1 3 4 6 and 1 2 5 3, see that they meet only at 1, in real life there's IIRC 80 frequencies not four as in my example). Kind of like the joke about 1 second late watch never being right. Besides, even if you chose the same sequence as somebody else, if you don't turn your equipment on in exactly the same time, you never cross and both can use full bandwidth the equipment can do.
It's clear that FHSS is much resistant to interference, especially from DSSS pieces of equipment. If you have a DSSS and FHSS equipment sharing same space, DSSS usually dies but FHSS just gets a bit slower.
Again, I am trying to simplify, please don't take this literally. (Yes, 2.43285938 GHz for 0xff in channel #1 is probably wrong.)
So, "Cheap Long Distance Wireless Networking" already exists. Damned, I am using it as I write this. :-)
There's one caveat though. The FCC/ETSI regulations. From what I know, in the US there's some power limit on the transmitter, and then you basically can use whatever high-gain antenna you like. In Europe, ETSI regulations regulate the total output of the system (the device + antenna gain), so you cannot do very long distance links with this technology, because the larger (=higher-gain) you have, the less power you are allowed to transmit through it..
If you live in ETSI countries, there's a possibility to have fairly cheap high-bandwidth (8 Mbps full-duplex) point to point links over big distances using devices in public 10.3 GHz band (as opposed to 2.4 GHz typically used by BreezeNet, WaveLan et al). This is a bit more expensive (say, $10000 for both points). Here the main manufacturers are CoproSys, Miracle or Alcoma.
I have tried to simplify things, experts please forgive me.
PS It's plain silly to use 2.4 GHz equipment through trees, as the author of the article mentions - leafs block the signal in this band a lot.
Not now I am afraid: :-)
kocour:~$ host -t soul microsoft.com
Invalid query type soul
I guess somebody must update bind-utils..
Actually, there have been some quarrels between SCO and Microsoft recently. Microsoft wanted them to still maintain compatibility with old Xenix in their OpenServer product, because that's what was in the deal when MS sold Xenix to SCO, but SCO complainted this is useless (who wants Xenix compatibility nowadays)..
Is it GPL-like or BSD-like or YAOSL-type..? Hynek
Though, you have to consider, that "only telephone costs" might be fairly high. So high, that the price of the dialup access is pretty irreleveant. You pay per minute, I don't know exactly how much in Germany, but here it's something like $2 an hour in peak, and this is the "cheaper" Internet tariff.
How come there's nothing much new, when it's yet another perspective ?? I thought another perspective also counts as something new..
Well, I haven't tried this 6.2 beta yet, but after 6.1 I am starting to get more and more of the impression that it's going way too much Windows direction. It's no longer about understanding things and how they work, but rather about knowing "when it says this it in fact means this-and-that" and "if you want to do this (e.g. fdisk) you must pretend to do something else to get the first thing done". See also "Insert Driver disk", missing descriptions of packages in 6.1 text install, overall feeling of losing control over the installation of the system, unhelpful help (or none at all) etc..
So in this light it's a piece of good news that the installer got better.
To your "root in xterm" problem - just do xhost +root@localhost or something to that effect before you do your su, DON'T do xhost + as it could be pretty dangerous (anybody could send you an X client which would grab your keyboard, snoop your keys and whatnot)..
Is there "the newest QT" codec library for xanim, i.e. in binary form? I think, that given that codecs are usually patented etc. that's probably the best thing we can get -- now. Yes, source or specs would be better, but I doubt they will ever release them..
Yes, you are right, some chess programs do use a kind of pre-selection.. From what I remember from the times I have played chess profesionally (~10 years ago), the chess software works (or used to work those ~ten years ago) this way: it computes all possible positions generated by its move(s) up to some level. Then it evaluates all the positions with some "static function", i.e. counts it's material, centralized King in an ending etc., so each position gets it's ranking. Then the best one is chosen. Now, there are two schools: "pre-selection" and "brutte force". Pre-selection school generates the rankings only few moves ahead, and just trashes positions which look completely silly (i.e. have too low ranking), and only pursues further the promising candidates. The brute force school evaluates the full tree, sometimes getting a really great move. Now, entry-level chess programs use brute force, better programs use the pre-selection method, but the best ones brutte force. Or this is what I learned those 10 years ago, and this is oversimplified, you have opening libraries, endings specialities etc. I wonder what does this add to our discussion of brute force in optimization. :-)
I don't know if you were referring to this (as there's way too many interesting things in KDE 2.0, like the teapot, which pops up a window and beeps when you should take the teabag out of your tea :-), but I saw there beginning of someting what looked too much like the Windows' device manager thing to my taste. You click on Properties, see "Device is operating normally" etc. On the other hand, Windows users would then have NO excuse not to use KDE. :-)
I absolutely agree. I work for an ISP, too, and almost 5% of our customers has switched the routers on their side off for this weekend, most of them probably due to Y2K worries.. :-(
For example, the disclaimer of warranty could be a problem, GPL isn't good enough as a contract in some legal systems, the author cannot agree to have no compensation for his work, gift taxes problems etc.
I am no lawyer and don't know much about other countries' law, but at least here in Czech Republic the GPL is fairly problematic. (See here, sorry, in Czech only.)
Right. One can even see Spyglass/NCSA Copyright it in MSIE's About.. thing.