December 2003 - Legacy - lists.gtalug.org

When will the persecution of MS *end*?!
by plp-ysDPMY98cNQDDBjDh4tngg＠public.gmane.org 20 Dec '03

20 Dec '03

On Fri, 19 Dec 2003, William Park wrote: > On Fri, Dec 19, 2003 at 05:29:24AM -0500, JoeHill wrote: > > > - inability to generate revenue from actual sales. > > > > ...because some company convicted of being an illegal monopoly is > > pulling the rug out from under you. > > This is the case of loser blaming the winner for winning. Real was > trying to win over Microsoft, no? The idea that there need not be winners and looser but that the competition itself is the goal seems to elude you. If the final goal is for all major market share holding companies to win out over all their competitors then there will be 5 kolchozes producing all consumer goods you will ever need, decide what is best for you, and not care a s**t what you really want or need. This has been done before on a grand scale in countries formerly known as 'communist'. Of course you don't want that ? Also it would completely obliterate what is known today as the stock market (which mostly makes its living out of handling bets on who is going to be 1% stronger (or weaker) that someone else now or within a certain given timeframe). Peter -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

C programming question
by plp-ysDPMY98cNQDDBjDh4tngg＠public.gmane.org 20 Dec '03

20 Dec '03

On Fri, 19 Dec 2003, Jim Skehill wrote: > If you are programming in C you can go down to the assembler level and issue > a BIOS interrupt (sorry I don't recall the number) to write to a particular > disk location. > DOS interrupt 25H allows you to write to a particular disk sector, but > seeing that this is the TLUG Mailing list, I don't imagine that's of much > help to you. Except you can't do that in linux since the bios is not functioning while in linux. There are equivalent functions in the kernel, not available from user space. In theory you can acheieve the same effect by using the raw device (e.g. /dev/hda) and use (f)seek & co. to acceess a physical sector (which you get to compute using the disk geometry). You have to be root to do this and the potential for problems is enormous. Peter -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

When will the persecution of MS *end*?!
by plp-ysDPMY98cNQDDBjDh4tngg＠public.gmane.org 20 Dec '03

20 Dec '03

> RealNetwork is like after-market car radio company that is suing GM > because GM has the arrogance to put radio in all GM cars, thereby > shutting out after-market radio companies. You mean, if someone invents something and someone else grabs the idea and uses it with slight changes, just enough to avoid a patent suit, and makes millions out of it by leveraging its monopoly position, it's ok, right ? If M$ had invented the idea they could have monopolized it until forever and nobody would have minded. But they did not. Notice that no-one is suing them over the Word .doc format or .xls or anything like that. Wonder why ? Because it's theirs. Peter -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

When will the persecution of MS *end*?!
by plp-ysDPMY98cNQDDBjDh4tngg＠public.gmane.org 20 Dec '03

20 Dec '03

> Cry babies... RealNetworks lost out in competition. What competition ? Oh, you mean when M$ bundled mediaplayer for free with the os just like it did with the explorer and integrated it as much as possible ? I guess someone should correct the dictionary definition of 'competition' now that it's common usage sense has changed so much ... Peter -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

Waaaay offtopic
by plp-ysDPMY98cNQDDBjDh4tngg＠public.gmane.org 20 Dec '03

20 Dec '03

On Thu, 18 Dec 2003, GDHough wrote: > Now where can I get that flu shot I heard about for my Windows PC? At a gunshop near you. Peter -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

VPN question
by henry-lqW1N6Cllo0sV2N9l4h3zg＠public.gmane.org 19 Dec '03

19 Dec '03

On Sat, 20 Dec 2003, Justin Zygmont wrote: > Just a thought about VPN's and a hope to get back on topic to fun linux > things, I was wondering if anyone knows why an encrypted VPN is so > important using a phone line when it's usually just a point to point link > with no chance of interception? "No chance of interception" is, unfortunately, putting it much too strongly. For one thing, it's a point-to-point link only if the phone company chooses to do it that way. They can easily cut a third party in on the link. They are known to do this for voice conversations, for quality monitoring and for emergency break-in (if it's a major emergency and you've just got to contact Fred and his phone line is constantly busy, an operator can break in on his conversation and tell him he's got an emergency call, if you can convince him/her that it's important enough). They probably don't for modem conversations, but it is possible. They certainly can do it for law-enforcement wiretaps of either. Moreover, with modern electronics it is trivial to tap a phone line undetectably (except for discovery of the tap by physical inspection). It doesn't have to be the police or the Feds doing it. It doesn't necessarily require access to your premises. Henry Spencer henry-lqW1N6Cllo0sV2N9l4h3zg at public.gmane.org -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

VPN question
by jzygmont-tEQKYFGiemxAYG7eUwYNkWD2FQJk+8+b＠public.gmane.org 19 Dec '03

19 Dec '03

Just a thought about VPN's and a hope to get back on topic to fun linux things, I was wondering if anyone knows why an encrypted VPN is so important using a phone line when it's usually just a point to point link with no chance of interception? -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

C programming question
by jingsu-26n5VD7DAF2Tm46uYYfjYg＠public.gmane.org 19 Dec '03

19 Dec '03

For the purposes of just seeing what the kernel does, even in protected mode, with regards to both virtual and physical memory, an easier method may be to run it in an emulator instead of trying to examine the running operating system. Try something like bochs. http://bochs.sourceforge.net/ A colleague of mine uses Bochs for simulating Linux on an x86 for his runtime trace optimization research. Apparently Bochs is a really well constructed and designed piece of work; expecially nice if it becomes necessary to dig under Bochs itself to see what it's doing in terms of emulation. -Jing -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

C programming question
by taavi-LbuTpDkqzNzXI80/IeQp7B2eb7JE58TQ＠public.gmane.org 19 Dec '03

19 Dec '03

On Fri, Dec 19, 2003 at 06:08:04PM -0500, Sergey Kuznetsov wrote: > In this case you can't reach the physical memory. Intel processors ( and > other CPU as well ) protects the access to the real physical memory. > Only kernel have access to it. > You can do it only thru the kernel space as a driver module, or thru the > /proc/kcore, but this is file in format of core file, you should know how to > handle and read it. "Physical memory" is an interesting topic when talking about CPUs with MMUs (memory management units). x86 CPUs since the 286 iirc have MMUs. To read the actual first few hundred bytes of physical, you will probably want to go through DOS, or write your own bootable program and eschew with an OS entirely. The Linux kernel does have the ability to map bits of real physical memory into the standard protected mode address space, though. I just don't know how to do it. Much simpler from DOS. ;) Note that doing something like: char *memory = 0x0202; will let you access memory at address 0x0202, relative to the program as it runs. In DOS, this is actually the 0x0202 byte in real, physical memory(*). In Linux, it'll be whatever's in 0x00000202 as seen by that process. Even in the kernel, it'll be 0x00000202 as seen by the kernel. The MMU can dynamically map physical and logical memory addresses by making use of a TLB (translation lookaside buffer). Because the kernel has to deal with physical devices which are memory mapped, however, it is possible using existing interfaces. You'd want to explore some of the driver and "real mode" memory interfaces, I think. No, it's never as simple as you thought. ;) (*) Of course, you'd want to be sure that you were accessing offset 0x0202 from segment 0x0000. DOS and all other intel real-mode applications have the concept of a "near" and a "far" pointer, the former being 16-bit, and the latter being 24-bit (the middle 8 bits of segment and offset overlap). See, it's NEVER as simple as you thought. ;) -- taa Hackers (the movie) n. A notable bomb from 1995. Should have been titled "Crackers", because cracking is what the movie was about. It's understandable that they didn't however; titles redolent of snack food are probably a tough sell in Hollywood. --From the Jargon File 4.1.2 --http://www.sunsite.ualberta.ca/jargon/ /*eof*/ -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0

Waaaay offtopic
by legrady-bJEeYj9oJeDQT0dZR+AlfA＠public.gmane.org 19 Dec '03

19 Dec '03

Actually, the reverse is the basis of science, ironically. You can't prove Newton's Theory of Gravity, but given Einstein's Theory of Relativity, you can disprove Newton's theory .... or, more precisely, you can prove that Newton's theory is a reasonably approximation, but only until you consider certain situations. And Einstein's theories are disproven by the the Uncertainty Principle, and by String Theory .... which are refinements of Einstein. But it's impossible ( so far ) to prove any of these. Maybe String Theory will be proven, and then we'll know everything. What does this have to do with Perl, kernels, or memory mapping? JoeHill wrote: >On Fri, 19 Dec 2003 22:40:29 -0500 >James Knott <james.knott-bJEeYj9oJeDQT0dZR+AlfA at public.gmane.org> wrote: > > > >> If someone has a claim, the onus is on them to prove their claims, not on >> someone else to prove them wrong. >> >> > >...actually, IIRC it isn't even logically possible to prove a negative. Been >awhile since Formal Logic 101 for me, and many many, er, "memory-loss inducers". > > > -- The Toronto Linux Users Group. Meetings: http://tlug.ss.org TLUG requests: Linux topics, No HTML, wrap text below 80 columns How to UNSUBSCRIBE: http://tlug.ss.org/subscribe.shtml

1 0