Actually I like the desktop in Linux and that's my main motivation for using it (That and the tight integration with all the typical GNU tools.)
My experience does not match yours. Netscape works fine for me, and that includes even using the G2 realvideo player to watch the new Weird Al video. Yes, it is certainly true that many plugins don't exist for Linux/Netscape, but I can't stand the sites that use them even in Windows where the plugins are there. They are generally not worth it and I find myself disabling the plugins to stop the annoyances.
And when it comes to looks, small fonts are seen as more desirable by some, myself included. I really don't like the default settings in windows/netscape or windows/ie because they use these garishly huge fonts that end up making one browser window eat up your whole screen. I like being able to have two windows up side-by-side and that means small fonts. But whichever way your preference goes, disliking defaults is a silly reason to bag a system when the defaults are user-configurable options.
And I see about 19 variable-width fonts and 8 fixed-width fonts in my Netscape font dialog.
Mostly what I like about the Unix model in general (from which Linux inherits a lot of its 'feel'), is that there are lots of things that are configurable. Although they might not be easy to configure, the list of things you can change is bigger than in Windows. For example, by altering the 'X resource Database' file that comes with netscape (netscape.ad) I have picked different fonts for the main menus and button labels, and changed the colors so that the toolbar panel is green, the main menu is black, and so on. Yes, this is hardly simple, but it is very powerful. It's like being able to change registry settings in a system where there are actual english comments in there to explain everything.
So yes, I *do* use it because of the look and feel.
People have been comparing Linux advocacy to Mac, Amiga, and OS/2 advocacy and claiming that Linux will suffer the same fate due to the similarities. This comparasin is invalid for the following reasons:
Mac, Amiga, and OS/2 were owned and produced by companies - companies that have to have a large enough market or else they drop the product. Linux is not in the same boat. Its existence is not at the whim of this company or that.
Mac, Amiga, and OS/2 were/are things that were more popular before, and the flames started when their popularity fell. That made for the 'sore loser' image of advocates. Linux started out not popular and the flames started when the popularity increased. This is a totally different pattern.
This was an example of the kind of childish flamebait that we have come to expect from these folks. No I'm not talking about the zealot e-mails. I am talking about Mindcraft only publishing a tiny fraction of the mail they received and trying to give the impression that this is indicative of the whole. There were a large number of reasoned, level headed people who mailed them with detailed reasons for their disagreement with Mindcraft. We see no representation of them.
At first I was willing to give them the benefit of the doubt and assume that Mindcraft were being honest and gave it a real try. I was assuming that there was some honest reason that their test disagreed with every bit of anecdotal evidence out there.
But this latest tactic of theirs proves me wrong. They *are* biased shills and this proves it. You don't take a community as large as the linux community and act like 12 e-mail messages represents the whole. That's bigoted bullshit on their part.
When I was in the market for a job I wanted to filter out any employer that was too dumb or too cheap to be able to use the internet. I was sick and tired of working for a place that transferred programs to the clients using UUCP over a modem (this was 1996, and not in an internet-poor area, so we had no excuse. It was the source of much embarassment when I had to tell clients we couldn't just FTP stuff over because we're lame.)
So anyway, to filter out similarly unaware companies, I put my resume online as a HTML page and then gave myself the firm rule that all I would ever send to people, on paper or e-mail is the URL that points at it. Any place that either (1) didn't have the ability to figure out what to do with it, or (2) didn't have the time to have a human read a resume before rejecting it, would be somewhere I wouldn't enjoy working. It worked. Now I found a job I actually like and have been here for 2-3 years with no intention of leaving soon.
Most pornography is rather lacking in this regard.
As is most porn-filtering software. And that's the big problem with this fucking stupid law. It's the same problem with the recent Australian law too:
It is not possible to filter 'bad' words without intelligent context-sensitive reading (a human). A program trying to block out text about breasts would end up also blocking text about breast cancer and cookbooks with scandalous passages describing chicken breast meat, for example. Whether you desire to have this stuff blocked or not is irrelevant if it isn't possible to do it right.
In light of #1 above, it is completely unfair to hold the library responsible for the content it receives from outside sources. It is exactly like finding a pornographic magazine in a plain brown wrapper in the mail, finding out that it was ordered by your son, and then choosing to blame this on the postman who delivered it to your house, who couldn't have known what it was without breaking it open and reading your mail.
There are two reasons for disliking this bill. On the one hand the censorship it is trying to accomplish is inexcusable, and on the other hand it isn't even practically enforcable anyway. Seeing as how we don't stand a snowball's chance in hell of convincing people that the first reason is sufficient to shoot it down, I reccomend concentrating on the second - the practicality of enforcement.
The Cable TV in the school example is not appropriate for the following reason: Playboy and Cinemax cost extra. This is a strong incentive against the school buying them, just as strong as not letting little kids see the naughty bits. It's more akin to choosing to buy paperbacks instead of hardcover books, or buying a simple board with hooks on it for the football scoreboard instead of an electronic display. That is very different from blocking something that comes at no extra charge and is part of the standard service.
Any government agent who abuses their power should be subject to either life inprisonment or execution. THAT would deter abuses more than a million new weak laws that never get invoked.
It would also prevent anyone from ever wanting to become a law enforcement agent. I agree that gummint agents should be punished for transgressions just like everyone else, but I cannot agree to the notion that they should be punished more than an ordinary citizen who commits the same crime.
I can't speak to the other points you made, since I never used NT Terminal server, but I can say that your comparasin of # of processes means nothing. Unix tends to use a large number of smaller processes and NT tends to use a small number of larger processes. This evolved this way because the overhead for making a new process in Unix is smaller than it is in NT. An NT programer might choose to start a thread as opposed to a process for one of two reasons: 1 - It makes sense for the design of what he's doing, or 2 - To attain acceptable performance, even if it is a confusing design. The Unix programer doesn't have the 2nd incentive since in most Unixen creating processes and creating threads take about the same amount of time.
He seems to believe that the fact that Redhat sells a box in stores prevents people from downloading the release for free.
That the following two sets of people are identical sets:
{ people who told him Redhat will only make money from support and services }
{ Opensource advocates }
(There is certainly much overlap in those sets, in fact the first set might be entirely contained within second. But they are not the same set, and thus proving the first set of people wrong does not prove the second set wrong.)
In the words of the immortal Bugs, "What a Maroon!".
Washington was burned by the British, not Canadians.
The war of 1812 was between the US and the UK, not Canada. The UK just happened to use Canada as their staging ground, because those northern colonies were still loyal to them. That's a bit different than saying the country of Canada that we see today was responsible.
If I want the system to stop discriminating against me by artificially limiting me to 10 open sockets, I needed (once upon a time) that registry change. MS made sure I had to buy a bunch of crap on top in order to get it (legally) changed. This is (was) bullshit. In order to run a *DIFFERENT* piece of server software effectively on NT, like Netscape's server, I'd have had to buy the version of NT that comes with IIS. This was bundling of a much worse degree than the stuff the DOJ is complaining about. They required the purchase of MS's server-type software before decrippling the system so it would work with anyone else's server-type software.
I noticed this in the article, talking about the goals of this 'enhancement':
1) fork() This implementation of fork() will clone the running interpreter and create a new interpreter with its own thread running in the same process. Goal is functional equivalence to fork() on UNIX systems without the performance hit of NT's process creation overhead.
Uhm, aren't they basically admitting publicly that NT's process creation model is dog-slow? It's pretty strange to hear them admit that outright.
And also, isn't that just a horribly bad idea to alter the Perl interpeter so that when a perl script does a 'fork' that it ends up making just a thread instead of a process? It sounds like a good idea for performance, but pre-existing scripts that *expect* their forks to create processes are going to run into trouble on this. (One reason for forking in a server is so that if the sub-server dies from a bug the main server is still going - this change blows that idea out of the water.)
Atheists are not a majority. It must have been an unusual, nonrandom selection of people in your class if the majority were right-wingers and atheists. Either that or you are using the overly broad definition of atheism that some Christains are guilty of and labelling every other religion as atheist.
The easiest way around this stupid idea, assuming it even gets implemented, is to open lots of proxies for our ozzie friends to use that are not on port 80.
The idiots who make this kind of software don't realize that there is nothing sacred in the use of port 80 - you don't *have* to use port 80 to be a web server, and you can't very well censor every single port number without disabling the whole net.
You are operating under the premise that the following two statements are contradictory:
"Linux is better because it's open source."
"MS Open Source is useless."
They are not contradictory. Linux was opensource from day 1. MS was not. Taking a pre-existing large closed bohemoth of code and opening it at a later date is not very useful. Especially when the plan is to only open some small parts of it (read the article), and leave the rest closed. With the way everything in Windows is interdependant, opening up only part of the code is not very useful: "Oh, look, here they call the foobar function - Is that very efficient? I don't know, I can't see the foobar function..."
...against buying a name without intending to actually use it. It's one thing to talk about buying a name indenting to sell it to someone else who needs it (what's already illegal), but it's even worse (IMO) to buy a name just for the intent of preventing someone else from being able to use it at all and just sitting on it.
If Bush buys bushsucks.com, he should be required to actually use it.
The only reason I dual boot into NT at all these days is to use Visual Cafe. So when VMware came along I thought, *great*! I'll be able to use this for VisualCafe and I can stop dual-booting.
But the performance of Cafe under VMware is horrid, and I can't figure out why. Every other program runs very fast in NT inside VMware, even graphical ones like CorelDraw. Word is fast, Excell is fast, Netscape is fast, IE is fast. But Visual Cafe is dog slow and I can't figure out why it is any different. (It isn't signifigantly slower than those other programs under native NT.)
I've tried everything. I've tried giving most of my 256 MB ram to vmware. I've tried running VMware before any other application, to ensure it has a greater likelyhood of grabbing real RAM and not swap space. I've tried running bare with nothing but VMware running. I'm using the accellerated DGA-ified X server from VMware. This is really bugging me because Visual Cafe is the only app I really need under NT - the other apps that can run fine are ones I don't use.
Has anyone out there had a similar experience and can give me a magic bullet to fix this?
The Permida2 video cards (3Dlabs) suck under VMware. The X server provided by VMware, that supposedly has the better DGA stuff in it, does not work. (well, it runs, but the DGA extensions don't work - so its no different than using a stock Xfree86 Xserver.) This is a fairly common card in workstations, so this is a big sticking point.
In the judging criteria, they stated that they measured domain density as the number of commercial.com addresses in an area.
That's not fair to places with loads of networking 'service' via.edu or.gov.
I can't figure out what the logic was behind restructing it to.com, other than to deliberately favor commercial use of the nyet as somehow "more important" than noncomercial use.
Okay, so opensource crypto software becomes exportable, but only in source-code form, from what it seems. What about if you make an rpm of crypto software that includes the binary in addition to the source? Is that still disallowed? This is a problem for things like Linux distributions that want to have Apache working 'out of the box'.
Don't forget that San Jose is part of the Silicon Valley area. Just because you see something there geared toward geeks doesn't mean it will spread to the rest of the country. The San-Jose Mercury News has always been a bit more technically saavy than your average news rag.
A little x-ray every so often won't hurt you, but what about people who fly a *lot* like salesmen, or the pilots and flight attendants who pass through several scanners per day?
Before you say that xrays are harmless, ask yourself why the x-ray machine operator in a hostpital (who has to do several x-rays per day) stands behind that shield?
"And with every 5000 frequent flyer miles, you get a free lead apron!"
The University where I work is one that is in the Internet2 consortium. Crud. Just what we need - more 'mandatory' Wintel crap floating around campus. (I don't like the "donating machinery" part of the deal. MS is only doing this so that Windows doesn't get left behind by Internet2. They want to ensure that there's enough Internet2 work done *on* Windows by giving this donation. Money speaks louder than sound technical decisions.
I seriously doubt that your lawnmower gets 100 mpg. Even a walk-behind push mower, without driven wheels doesn't get that much. (Consider how many days you would have to walk to get 100 miles, consider that you use up a typical 1-2 gallon lawmower tank in an afternoon.) It can only get worse when you talk about a self-propelled mower. is your mower electric by any chance?
Slashdot Mirror
My experience does not match yours. Netscape works fine for me, and that includes even using the G2 realvideo player to watch the new Weird Al video. Yes, it is certainly true that many plugins don't exist for Linux/Netscape, but I can't stand the sites that use them even in Windows where the plugins are there. They are generally not worth it and I find myself disabling the plugins to stop the annoyances.
And when it comes to looks, small fonts are seen as more desirable by some, myself included. I really don't like the default settings in windows/netscape or windows/ie because they use these garishly huge fonts that end up making one browser window eat up your whole screen. I like being able to have two windows up side-by-side and that means small fonts. But whichever way your preference goes, disliking defaults is a silly reason to bag a system when the defaults are user-configurable options.
And I see about 19 variable-width fonts and 8 fixed-width fonts in my Netscape font dialog.
Mostly what I like about the Unix model in general (from which Linux inherits a lot of its 'feel'), is that there are lots of things that are configurable. Although they might not be easy to configure, the list of things you can change is bigger than in Windows. For example, by altering the 'X resource Database' file that comes with netscape (netscape.ad) I have picked different fonts for the main menus and button labels, and changed the colors so that the toolbar panel is green, the main menu is black, and so on. Yes, this is hardly simple, but it is very powerful. It's like being able to change registry settings in a system where there are actual english comments in there to explain everything.
So yes, I *do* use it because of the look and feel.
At first I was willing to give them the benefit of the doubt and assume that Mindcraft were being honest and gave it a real try. I was assuming that there was some honest reason that their test disagreed with every bit of anecdotal evidence out there.
But this latest tactic of theirs proves me wrong. They *are* biased shills and this proves it. You don't take a community as large as the linux community and act like 12 e-mail messages represents the whole. That's bigoted bullshit on their part.
So anyway, to filter out similarly unaware companies, I put my resume online as a HTML page and then gave myself the firm rule that all I would ever send to people, on paper or e-mail is the URL that points at it. Any place that either (1) didn't have the ability to figure out what to do with it, or (2) didn't have the time to have a human read a resume before rejecting it, would be somewhere I wouldn't enjoy working. It worked. Now I found a job I actually like and have been here for 2-3 years with no intention of leaving soon.
As is most porn-filtering software. And that's the big problem with this fucking stupid law. It's the same problem with the recent Australian law too:
- It is not possible to filter 'bad' words without intelligent context-sensitive reading (a human). A program trying to block out text about breasts would end up also blocking text about breast cancer and cookbooks with scandalous passages describing chicken breast meat, for example. Whether you desire to have this stuff blocked or not is irrelevant if it isn't possible to do it right.
- In light of #1 above, it is completely unfair to hold the library responsible for the content it receives from outside sources. It is exactly like finding a pornographic magazine in a plain brown wrapper in the mail, finding out that it was ordered by your son, and then choosing to blame this on the postman who delivered it to your house, who couldn't have known what it was without breaking it open and reading your mail.
There are two reasons for disliking this bill. On the one hand the censorship it is trying to accomplish is inexcusable, and on the other hand it isn't even practically enforcable anyway. Seeing as how we don't stand a snowball's chance in hell of convincing people that the first reason is sufficient to shoot it down, I reccomend concentrating on the second - the practicality of enforcement.The Cable TV in the school example is not appropriate for the following reason: Playboy and Cinemax cost extra. This is a strong incentive against the school buying them, just as strong as not letting little kids see the naughty bits. It's more akin to choosing to buy paperbacks instead of hardcover books, or buying a simple board with hooks on it for the football scoreboard instead of an electronic display. That is very different from blocking something that comes at no extra charge and is part of the standard service.
It would also prevent anyone from ever wanting to become a law enforcement agent. I agree that gummint agents should be punished for transgressions just like everyone else, but I cannot agree to the notion that they should be punished more than an ordinary citizen who commits the same crime.
I can't speak to the other points you made, since I never used NT Terminal server, but I can say that your comparasin of # of processes means nothing. Unix tends to use a large number of smaller processes and NT tends to use a small number of larger processes. This evolved this way because the overhead for making a new process in Unix is smaller than it is in NT. An NT programer might choose to start a thread as opposed to a process for one of two reasons: 1 - It makes sense for the design of what he's doing, or 2 - To attain acceptable performance, even if it is a confusing design. The Unix programer doesn't have the 2nd incentive since in most Unixen creating processes and creating threads take about the same amount of time.
- That making money is anti-opensource.
- He seems to believe that the fact that Redhat sells a box in stores prevents people from downloading the release for free.
- That the following two sets of people are identical sets:
In the words of the immortal Bugs, "What a Maroon!".- { people who told him Redhat will only make money from support and services }
- { Opensource advocates }
(There is certainly much overlap in those sets, in fact the first set might be entirely contained within second. But they are not the same set, and thus proving the first set of people wrong does not prove the second set wrong.)The war of 1812 was between the US and the UK, not Canada. The UK just happened to use Canada as their staging ground, because those northern colonies were still loyal to them. That's a bit different than saying the country of Canada that we see today was responsible.
If I want the system to stop discriminating against me by artificially limiting me to 10 open sockets, I needed (once upon a time) that registry change. MS made sure I had to buy a bunch of crap on top in order to get it (legally) changed. This is (was) bullshit. In order to run a *DIFFERENT* piece of server software effectively on NT, like Netscape's server, I'd have had to buy the version of NT that comes with IIS. This was bundling of a much worse degree than the stuff the DOJ is complaining about. They required the purchase of MS's server-type software before decrippling the system so it would work with anyone else's server-type software.
1) fork() This implementation of fork() will clone the running interpreter and create a new interpreter with its own thread running in the same process. Goal is functional equivalence to fork() on UNIX systems without the performance hit of NT's process creation overhead.
Uhm, aren't they basically admitting publicly that NT's process creation model is dog-slow? It's pretty strange to hear them admit that outright.
And also, isn't that just a horribly bad idea to alter the Perl interpeter so that when a perl script does a 'fork' that it ends up making just a thread instead of a process? It sounds like a good idea for performance, but pre-existing scripts that *expect* their forks to create processes are going to run into trouble on this. (One reason for forking in a server is so that if the sub-server dies from a bug the main server is still going - this change blows that idea out of the water.)
Atheists are not a majority. It must have been an unusual, nonrandom selection of people in your class if the majority were right-wingers and atheists. Either that or you are using the overly broad definition of atheism that some Christains are guilty of and labelling every other religion as atheist.
Rush is a band, not the Big Fat Idiot on TV.
The idiots who make this kind of software don't realize that there is nothing sacred in the use of port 80 - you don't *have* to use port 80 to be a web server, and you can't very well censor every single port number without disabling the whole net.
- "Linux is better because it's open source."
- "MS Open Source is useless."
They are not contradictory. Linux was opensource from day 1. MS was not. Taking a pre-existing large closed bohemoth of code and opening it at a later date is not very useful. Especially when the plan is to only open some small parts of it (read the article), and leave the rest closed. With the way everything in Windows is interdependant, opening up only part of the code is not very useful: "Oh, look, here they call the foobar function - Is that very efficient? I don't know, I can't see the foobar function..."If Bush buys bushsucks.com, he should be required to actually use it.
But the performance of Cafe under VMware is horrid, and I can't figure out why. Every other program runs very fast in NT inside VMware, even graphical ones like CorelDraw. Word is fast, Excell is fast, Netscape is fast, IE is fast. But Visual Cafe is dog slow and I can't figure out why it is any different. (It isn't signifigantly slower than those other programs under native NT.)
I've tried everything. I've tried giving most of my 256 MB ram to vmware. I've tried running VMware before any other application, to ensure it has a greater likelyhood of grabbing real RAM and not swap space. I've tried running bare with nothing but VMware running. I'm using the accellerated DGA-ified X server from VMware. This is really bugging me because Visual Cafe is the only app I really need under NT - the other apps that can run fine are ones I don't use.
Has anyone out there had a similar experience and can give me a magic bullet to fix this?
The Permida2 video cards (3Dlabs) suck under VMware. The X server provided by VMware, that supposedly has the better DGA stuff in it, does not work. (well, it runs, but the DGA extensions don't work - so its no different than using a stock Xfree86 Xserver.) This is a fairly common card in workstations, so this is a big sticking point.
That's not fair to places with loads of networking 'service' via .edu or .gov.
I can't figure out what the logic was behind restructing it to .com, other than to deliberately favor commercial use of the nyet as somehow "more important" than noncomercial use.
Okay, so opensource crypto software becomes exportable, but only in source-code form, from what it seems. What about if you make an rpm of crypto software that includes the binary in addition to the source? Is that still disallowed? This is a problem for things like Linux distributions that want to have Apache working 'out of the box'.
Don't forget that San Jose is part of the Silicon Valley area. Just because you see something there geared toward geeks doesn't mean it will spread to the rest of the country. The San-Jose Mercury News has always been a bit more technically saavy than your average news rag.
Before you say that xrays are harmless, ask yourself why the x-ray machine operator in a hostpital (who has to do several x-rays per day) stands behind that shield?
"And with every 5000 frequent flyer miles, you get a free lead apron!"
The University where I work is one that is in the Internet2 consortium. Crud. Just what we need - more 'mandatory' Wintel crap floating around campus. (I don't like the "donating machinery" part of the deal. MS is only doing this so that Windows doesn't get left behind by Internet2. They want to ensure that there's enough Internet2 work done *on* Windows by giving this donation. Money speaks louder than sound technical decisions.
Or maybe your 100 mpg meant "meters per gallon"?