A question of morality. Exactly my point: if it is doubtful whether or not obtaining certain "resources" is morally acceptable, any research to profit on those would be morally doubtful too.
I think your analogy is a little far-fetched. Pharmaceutical companies aren't actually starving, and eating and getting drunk are not necessary stages in a process. But once you asked, if you know the beggar to have some money, yes, I think you shouldn't give him food, except for something you will dispose of anyway.
I agree with you that his decision was a wise one, just that I think it was too ambiguous.
I can't see how supporting research on already-obtained cells differs in practice from funding resarch for stem cell farming. If there is more money available for a given stage in a process, wouldn't some of the money that would be originally employed in that stage be diverted to research in other stages now much more in need?
Am I missing something fundamental or is this really just GWB hedging against criticism?
I can't think of a good reasons to avoid Conectiva Linux, but OTOH I see not a single one to use it. Among the arguments one generally hears in Brazil about why to use CL:
Translation - Their translations to Portuguese are light-years away from professional grade (as unfortunately seems to be the norm with Linux material in Portuguese over the net.) As long as one can functionally read English (which I believe to be the case of people who are into UNIX administration), what one gets from an original English text makes more sense than most translations. If you are a vanilla user who can't read English, well, I do not know how well-translated other distros are, but beating CL in that field isn't hard at all.
90-day support - Now *that* is a joke! I phoned them 3 times with the following questions when I first installed Linux:
How do I display more than 25 lines in a terminal screen? The answer was to add a line like "vga=788" (?) in lilo.conf, but the answer I got was: "This is not covered by basic support, sir. You need to upgrade."
How do I get things compiled after I install gcc and GNU tools?? Answer: install the "development" libraries too, you insignificant beginer! The answer I got: "This is not covered by basic support, sir. You need to upgrade."
How do I access my DOS partition from Linux? Answer: You either edit fstab or use linuxconf to create a mounting point for the filesystem and configure it to mount automatically upon startup. But I got the same standard answer.
Well, to be 100% honest, this was back in 1999, but from my point of view these were all legitimate installation questions to which I had to dig for the answers myself, without Internet access at home at the time.
Hardware support - Conectiva claims to support lots of cheap hardware commonly found in old PC's. I don't think this is not true for some other distros too. And this is becoming less important by the day.
Security - OK, they make patches available over their website really quick, but the default installation of CL 4.0, as I later concluded, left a lot of unnecessary services open. I don't know about CL 6.0 wich I'm using now in my DT machine, for I chose the packages myself.
Easy installation - CL 6.0 defaults to a graphic install, which failed to recognise the video card in my notebook, making it impossible to use. I installed SuSE 6.0, a breeze, but I was too much used to CL's (RH's) way of doing things. Then I installed Mandrake, even easier and recognised everything except, obviously, the WinModem.
There are other things about Conectiva that I don't like, but in this post I'll stick to mention why the arguments they use are not important for many people. If one can't read English at all, I'd say "go CL", despite the rough translation. Otherwise, yes, CL is pretty much just another distro, maybe among the best as a server distro, probably not as a desktop.
Unless Windows now allows similar remote operation...
And from the article:
Needless to say that there's plenty of previews, reviews and columns discussing how XP is built on a new engine, a protected kernel mode architecture,....., built in remote accessibility,.... as well as a plethora of other changes and upgrades when compared to previous Win 9x or even NT based Operating Systems.
Now that you mention, the fact that one can't execute code remotely in a standard W9x/NT box (don't know about W2K, I think it's the same) is a feature that really boosts the security of a well configured NT box. If WXP allows for remote log-in MS should be really careful about what a default install will let open!
I did. Both the grey and the green text looked just as green up front; I thought CT was trying to be cutesy. I read on and forgot about it, but when I read your post it caught my attention: grey looked grey again, which is a good thing...
I think that what really happened was that because that page background is green, the white in this page tends to look rosy. making the grey look green in contrast.
The 8088 was a crippled 8086, kind of the "celeron" of it's day. It was limited to 8-bits externally, though it is 16-bits internally. Because of the 8088, early PC's were limited to being able to address 1MB of RAM. This is probably one reason the 640K DOS limit hung around for so long (still a factor in the early-mid 90's).
From where I see, the 8088 was more a hack than a slipup, for it existence allowed a lot of people to own a PC.
I don't know how old you are but in those days peripherals using 16-bit data buses where much more expensive.
The 640K RAM address limit is, as you say, a DOS limit.
I live in Brazi too, and when I read the front story I scanned the posts looking for someone saying *anything* positive about UPS, so as to bash the sucker to death. It seems I'm not finding anyone to bash today... I'm not telling my experience with them, for it was not nearly as dramatic as yours, but I'm not ordering anything that ships via UPS again.
Mainsoft has licensed the source to windows NT not just for IE. The source to NT is important cause IE uses windows APIs.
Whats the point of having an API? To provide services in a way that allows one to write/port an application without having to worry about the inner workings of the OS/DLL, as long as the interface specification is followed, right? So, why should they need the source code?
The only reason I can think of is that the interface specificaton does not correspond 100% to NTs functionality. This may be due to malice (providing an advantage to those who have acces to the source) or a poor job (either the implementation doesnt do what it si supposed to or the interface is not well documented). I really believe it is a little of each in this case.
That said, I have serious doubts about the quality of Mainsoft-ported software. I have never used any of their products, Im talking from a design point of view. Ill quote Kernighan/Pike on this:
Among the issues to be worked out in a design are (...) Information hiding: what information is visible and what is private? An interface must provide straightforward access to the components while hiding the details of the implementation so they can be changed without affecting users. [K/P, The practice of programming, p.85]
So, having the source to NT may be the fastest way to port MS-Windows to unix, but that is a symptom that something is very wrong somewhere along the line.
This is the TIFF specification (very useful) but as TIFF supports JPEG compression the scheme is well explained. GIF compression is explained too. It is 385k. If bandwidth is really an issue, e-mail me and I can send you a.zip (which is still 269kb)
Over the years I found there is only one thing that allows you to really clean your keyboard: a screwdriver! unless you take your keyboard apart I don't see how you can get those crannies&nooks clean.
On a side note, everytime someone asks me to take a look at his/her computer for some reason or if I'm shopping for used stuff, one of the first things I do is to turn the keyboard upside-down and shake it. Some keyboards have a history...
You know what really pisses me off? These Telcos keep advertising how they are changing the world and connecting every single corner on the planet when in fact they provided little more than a lousy service, as a rule, and only where it is profitable.
Apart from this rant, in the meantime I think you should take a look at how to improve your modem connection. I have a 56k PCI card that never gave me more than 14.4k. Theres no ISDN where I live and DSL is US$150,00/mo (not guaranteed to work above 64k!). No cable either (half a mile away). I can't afford a direct connection for home use.
So I bought an external USR and I could get 28,8k. I then replaced some 100ft of old wiring by a new twisted pair (which I overtwisted mysef) and went after what was going on. I had to replace my voltage stabilizer (you know, those switched bastards) and I could get 33,6k during the night and anything between 40 and 49k during the day. Odd, isn't it? It turned out that the dimer I have in my room (which probably works by generating a square PWM) was injecting a hell of a noise in the electrical circuit of the entire house. Turn the damn thing off and the connection gets clean. Unplugging the phone from the line when you connect is also reported to work in some cases. I saw a page about modem fine tuning a couple of months ago, I think it was 4 or 5 links away from the modem How-to, but I can't find the link now.
Although the original poster probably knows about it, for those reading this discussion out of the same interest, try the Linux Advocacy mini-HOWTO. The information here is not so organized as the poster proposes but some very good hints can be found.
My personal opinion about the question is that sharing your own experience is a key convincing tool.
No, it isn't excessive. Actually, posting this story was one of the best services/. did for OSS in the last weeks.
My point is: as said in the article, OSS is here to stay and it is to everyone's benefit that the distance that separates OSS from proprietary software be shortened. Companies can benefit from this, but most don't do it out of ignorance. One just can't like something he/she doens't know, and RMS is a very good choice of a lecturer on the subject. Having such a testimonial made public will certainly encourage other companies to follow suit, thus the great service.
Now, the baby-sitting part: I assume RMS was there at no charge, so taking care of his basic needs is just the most reasonable thing to do. When you are away from home it is very comforting to have some compensation of this sort. Producing some tea is no big deal. And the original writer stressed that the most important thing is to provide RMS with some conectivity. He is a busy man and it is not nice to monopolize his attention for so long.
At the end of the day, arranging such a visit seems to be quite a good deal for any software vendor.
Not that I am more enlightened but what I think happened was that this power supply was somehow short-circuited and current was being drained through grounding. I've seen this happen before.
Actually part of the reason why I never get much done at home is my wife literally dragging me away from the computer
Well, we can't have the best of both words at the same time, right? Sometimes I also get pulled from the computer by my girlfriend, mainly if shhhe notic s I'm pstin g at slashddort. I try ot resitst fdgifgi gegb hhhhhhhhhh hhhhhhhh hhhhhhhhh hoclMXvnhm "XDV A"X
Congrats for your marriage! I wish you both my best.
Now, that's what you did to solve the problem, but can you tell us what you were doing wrong in the first place, in your opinion? From what I read it was a case of burnout. Did you engage in too tight a schedule before being blocked? Do you consume much caffeine? Were you too anxious about the marriage? There must be a lot of us in a borderline state (I certainly am) and I'd be glad if you could share your experience.
I've recently started using KDE and the one thing that bugs me is the lack of monitor enlargement for many things. I don't currently own a 21" monitor and one of the nice things under Windows is that you will soon be able to emulate a 21" monitor using a 15" one - tricky, but better than nothing. Considering X is the home of the GUI it seems strange that KDE is so 15" unfriendly...
Does anyone know whether this is going to improve or not, or if there are any utilities which can help in this matter?
Very interesting post this one of yours, I could agree with you 100% in the first line, and gradually go down to 0% on the last line.
If I read you correctly, the burden to avoid buffer overflows should be put on the programmer, not on the hardware or on the machine. Then you start advocating that the compiler/library should take care of this! Isn't that like just laying the safety net a couple feet higher?
As a general rule, I think dumb tasks should be left for the machine, the noble ones for the programmer. Checking whether a given input validation is a potential door for an exploit is the programmer's responsibility.
Another thing that you seem to suggest is that quick and dirty test code should be the basis for production code, thus holes could migrate to the final product. I think the ideal solution is to discard test code altogether and start from scratch, but who does that, right? One other approach, which I actually use and believe many do, is to check for errors (at least return values) from the very beginning, so as those parts which will inevitably be cut and pasted to the production source are structured in a way that adding extra checks would be much easier (favouring using heap memory, encasing calls in try/catch blocks in C++, for example)
In ever more Western democracies, governments seem to be intent on limiting people from having access to encryption and electronic privacy. The United States has Carnivore watching its people with Echelon watching the world; the United Kingdom has Regulation of Investigatory Powers (RIP) Bill; and Australia now has The Telecommunications (Interception) Legislation Amendment Bill.
If you adjust one thing in this paragraph the whole scheme makes a lot of sense: Echelon is not an U.S.-only program, rather a multinational initiative conceived by these very countries you mention, and others. And Echelon predates theses national laws by a few years.
What I read from this is that Echelon has actually produced interesting material, but this material cannot be used as evidence, given its illegal nature. So, these bills are enacted so as to allow for legal evidence collection; law enforcement agencies probably know who to look after already.
Slashdot Mirror
A question of morality. Exactly my point: if it is doubtful whether or not obtaining certain "resources" is morally acceptable, any research to profit on those would be morally doubtful too.
I think your analogy is a little far-fetched. Pharmaceutical companies aren't actually starving, and eating and getting drunk are not necessary stages in a process. But once you asked, if you know the beggar to have some money, yes, I think you shouldn't give him food, except for something you will dispose of anyway.
I agree with you that his decision was a wise one, just that I think it was too ambiguous.
I can't see how supporting research on already-obtained cells differs in practice from funding resarch for stem cell farming. If there is more money available for a given stage in a process, wouldn't some of the money that would be originally employed in that stage be diverted to research in other stages now much more in need?
Am I missing something fundamental or is this really just GWB hedging against criticism?
I can't think of a good reasons to avoid Conectiva Linux, but OTOH I see not a single one to use it. Among the arguments one generally hears in Brazil about why to use CL:
Translation - Their translations to Portuguese are light-years away from professional grade (as unfortunately seems to be the norm with Linux material in Portuguese over the net.) As long as one can functionally read English (which I believe to be the case of people who are into UNIX administration), what one gets from an original English text makes more sense than most translations. If you are a vanilla user who can't read English, well, I do not know how well-translated other distros are, but beating CL in that field isn't hard at all.
90-day support - Now *that* is a joke! I phoned them 3 times with the following questions when I first installed Linux:
Well, to be 100% honest, this was back in 1999, but from my point of view these were all legitimate installation questions to which I had to dig for the answers myself, without Internet access at home at the time.How do I display more than 25 lines in a terminal screen? The answer was to add a line like "vga=788" (?) in lilo.conf, but the answer I got was: "This is not covered by basic support, sir. You need to upgrade."
How do I get things compiled after I install gcc and GNU tools?? Answer: install the "development" libraries too, you insignificant beginer! The answer I got: "This is not covered by basic support, sir. You need to upgrade."
How do I access my DOS partition from Linux? Answer: You either edit fstab or use linuxconf to create a mounting point for the filesystem and configure it to mount automatically upon startup. But I got the same standard answer.
Hardware support - Conectiva claims to support lots of cheap hardware commonly found in old PC's. I don't think this is not true for some other distros too. And this is becoming less important by the day.
Security - OK, they make patches available over their website really quick, but the default installation of CL 4.0, as I later concluded, left a lot of unnecessary services open. I don't know about CL 6.0 wich I'm using now in my DT machine, for I chose the packages myself.
Easy installation - CL 6.0 defaults to a graphic install, which failed to recognise the video card in my notebook, making it impossible to use. I installed SuSE 6.0, a breeze, but I was too much used to CL's (RH's) way of doing things. Then I installed Mandrake, even easier and recognised everything except, obviously, the WinModem.
There are other things about Conectiva that I don't like, but in this post I'll stick to mention why the arguments they use are not important for many people. If one can't read English at all, I'd say "go CL", despite the rough translation. Otherwise, yes, CL is pretty much just another distro, maybe among the best as a server distro, probably not as a desktop.
And from the article:
Needless to say that there's plenty of previews, reviews and columns discussing how XP is built on a new engine, a protected kernel mode architecture,....., built in remote accessibility,.... as well as a plethora of other changes and upgrades when compared to previous Win 9x or even NT based Operating Systems.
Now that you mention, the fact that one can't execute code remotely in a standard W9x/NT box (don't know about W2K, I think it's the same) is a feature that really boosts the security of a well configured NT box. If WXP allows for remote log-in MS should be really careful about what a default install will let open!
Wouldn't it be GNU/China?
Anyone else notice this?
I did. Both the grey and the green text looked just as green up front; I thought CT was trying to be cutesy. I read on and forgot about it, but when I read your post it caught my attention: grey looked grey again, which is a good thing...
I think that what really happened was that because that page background is green, the white in this page tends to look rosy. making the grey look green in contrast.
Now that you mentioned it, it could be through faulty grounding...
The 8088 was a crippled 8086, kind of the "celeron" of it's day. It was limited to 8-bits externally, though it is 16-bits internally. Because of the 8088, early PC's were limited to being able to address 1MB of RAM. This is probably one reason the 640K DOS limit hung around for so long (still a factor in the early-mid 90's).
From where I see, the 8088 was more a hack than a slipup, for it existence allowed a lot of people to own a PC.
I don't know how old you are but in those days peripherals using 16-bit data buses where much more expensive.
The 640K RAM address limit is, as you say, a DOS limit.
I live in Brazi too, and when I read the front story I scanned the posts looking for someone saying *anything* positive about UPS, so as to bash the sucker to death. It seems I'm not finding anyone to bash today...
I'm not telling my experience with them, for it was not nearly as dramatic as yours, but I'm not ordering anything that ships via UPS again.
Interesting but, where do Trolls fit in this Dwarf-Elf-Human scheme?
Mainsoft has licensed the source to windows NT not just for IE. The source to NT is important cause IE uses windows APIs.
Whats the point of having an API? To provide services in a way that allows one to write/port an application without having to worry about the inner workings of the OS/DLL, as long as the interface specification is followed, right? So, why should they need the source code?
The only reason I can think of is that the interface specificaton does not correspond 100% to NTs functionality. This may be due to malice (providing an advantage to those who have acces to the source) or a poor job (either the implementation doesnt do what it si supposed to or the interface is not well documented). I really believe it is a little of each in this case.
That said, I have serious doubts about the quality of Mainsoft-ported software. I have never used any of their products, Im talking from a design point of view. Ill quote Kernighan/Pike on this:
Among the issues to be worked out in a design are (...) Information hiding: what information is visible and what is private? An interface must provide straightforward access to the components while hiding the details of the implementation so they can be changed without affecting users. [K/P, The practice of programming, p.85]
So, having the source to NT may be the fastest way to port MS-Windows to unix, but that is a symptom that something is very wrong somewhere along the line.
Where I say "GIF compression" please read "LZW compression".
Try this: /asn/developer/PDFS/TN/TIFF6.pdf
http://partners.adobe.com
This is the TIFF specification (very useful) but as TIFF supports JPEG compression the scheme is well explained. GIF compression is explained too. It is 385k. If bandwidth is really an issue, e-mail me and I can send you a .zip (which is still 269kb)
I wonder how long it will take until cable companies start to hire outsiders to scan their networks...
Fatal server error.
Caught signal 11.
Over the years I found there is only one thing that allows you to really clean your keyboard: a screwdriver! unless you take your keyboard apart I don't see how you can get those crannies&nooks clean.
On a side note, everytime someone asks me to take a look at his/her computer for some reason or if I'm shopping for used stuff, one of the first things I do is to turn the keyboard upside-down and shake it. Some keyboards have a history...
You know what really pisses me off? These Telcos keep advertising how they are changing the world and connecting every single corner on the planet when in fact they provided little more than a lousy service, as a rule, and only where it is profitable.
Apart from this rant, in the meantime I think you should take a look at how to improve your modem connection. I have a 56k PCI card that never gave me more than 14.4k. Theres no ISDN where I live and DSL is US$150,00/mo (not guaranteed to work above 64k!). No cable either (half a mile away). I can't afford a direct connection for home use.
So I bought an external USR and I could get 28,8k. I then replaced some 100ft of old wiring by a new twisted pair (which I overtwisted mysef) and went after what was going on. I had to replace my voltage stabilizer (you know, those switched bastards) and I could get 33,6k during the night and anything between 40 and 49k during the day. Odd, isn't it? It turned out that the dimer I have in my room (which probably works by generating a square PWM) was injecting a hell of a noise in the electrical circuit of the entire house. Turn the damn thing off and the connection gets clean. Unplugging the phone from the line when you connect is also reported to work in some cases. I saw a page about modem fine tuning a couple of months ago, I think it was 4 or 5 links away from the modem How-to, but I can't find the link now.
Good luck.
Although the original poster probably knows about it, for those reading this discussion out of the same interest, try the Linux Advocacy mini-HOWTO. The information here is not so organized as the poster proposes but some very good hints can be found.
My personal opinion about the question is that sharing your own experience is a key convincing tool.
No, it isn't excessive. Actually, posting this story was one of the best services /. did for OSS in the last weeks.
My point is: as said in the article, OSS is here to stay and it is to everyone's benefit that the distance that separates OSS from proprietary software be shortened. Companies can benefit from this, but most don't do it out of ignorance. One just can't like something he/she doens't know, and RMS is a very good choice of a lecturer on the subject. Having such a testimonial made public will certainly encourage other companies to follow suit, thus the great service.
Now, the baby-sitting part: I assume RMS was there at no charge, so taking care of his basic needs is just the most reasonable thing to do. When you are away from home it is very comforting to have some compensation of this sort. Producing some tea is no big deal. And the original writer stressed that the most important thing is to provide RMS with some conectivity. He is a busy man and it is not nice to monopolize his attention for so long.
At the end of the day, arranging such a visit seems to be quite a good deal for any software vendor.
Not that I am more enlightened but what I think happened was that this power supply was somehow short-circuited and current was being drained through grounding. I've seen this happen before.
Actually part of the reason why I never get much done at home is my wife literally dragging me away from the computer
Well, we can't have the best of both words at the same time, right? Sometimes I also get pulled from the computer by my girlfriend, mainly if shhhe notic s I'm pstin g at slashddort. I try ot resitst fdgifgi gegb hhhhhhhhhh hhhhhhhh hhhhhhhhh hoclMXvnhm "XDV A"X
Congrats for your marriage! I wish you both my best.
Now, that's what you did to solve the problem, but can you tell us what you were doing wrong in the first place, in your opinion? From what I read it was a case of burnout. Did you engage in too tight a schedule before being blocked? Do you consume much caffeine? Were you too anxious about the marriage? There must be a lot of us in a borderline state (I certainly am) and I'd be glad if you could share your experience.
BTW, nice first post!
I've recently started using KDE and the one thing that bugs me is the lack of monitor enlargement for many things. I don't currently own a 21" monitor and one of the nice things under Windows is that you will soon be able to emulate a 21" monitor using a 15" one - tricky, but better than nothing. Considering X is the home of the GUI it seems strange that KDE is so 15" unfriendly...
:)
Does anyone know whether this is going to improve or not, or if there are any utilities which can help in this matter?
Apart from this gripe though, KDE is nice
Very interesting post this one of yours, I could agree with you 100% in the first line, and gradually go down to 0% on the last line.
If I read you correctly, the burden to avoid buffer overflows should be put on the programmer, not on the hardware or on the machine. Then you start advocating that the compiler/library should take care of this! Isn't that like just laying the safety net a couple feet higher?
As a general rule, I think dumb tasks should be left for the machine, the noble ones for the programmer. Checking whether a given input validation is a potential door for an exploit is the programmer's responsibility.
Another thing that you seem to suggest is that quick and dirty test code should be the basis for production code, thus holes could migrate to the final product. I think the ideal solution is to discard test code altogether and start from scratch, but who does that, right? One other approach, which I actually use and believe many do, is to check for errors (at least return values) from the very beginning, so as those parts which will inevitably be cut and pasted to the production source are structured in a way that adding extra checks would be much easier (favouring using heap memory, encasing calls in try/catch blocks in C++, for example)
Just my $0,02
In ever more Western democracies, governments seem to be intent on limiting people from having access to encryption and electronic privacy. The United States has Carnivore watching its people with Echelon watching the world; the United Kingdom has Regulation of Investigatory Powers (RIP) Bill; and Australia now has The Telecommunications (Interception) Legislation Amendment Bill.
If you adjust one thing in this paragraph the whole scheme makes a lot of sense: Echelon is not an U.S.-only program, rather a multinational initiative conceived by these very countries you mention, and others. And Echelon predates theses national laws by a few years.
What I read from this is that Echelon has actually produced interesting material, but this material cannot be used as evidence, given its illegal nature. So, these bills are enacted so as to allow for legal evidence collection; law enforcement agencies probably know who to look after already.
It would have been cracked probably, but just because DVD encryption is crap. I am not a cryptographer, but this seems to be the consensus.