I found opera to be slow, unstable, rendered poorly, and to have a clunky, ugly interface compared to IE. I gave it a couple of months at someone's insistance during the IE5 days, and didn't like it. Tried it a few times since, most of the issues improved, and it may be competative now, but I find Firefox suits my needs well enough and don't feel the urge to try opera yet again, after finding it unpleasant several times before.
IE 5+ was faster and more stable than the alternatives (Netscape, Mozilla, and even [just barely] firefox) on windows, up to IE 6. I'm not sure about 7, speed isn't the only thing in the world, and I switched during 6's lifespan. Actually I find in any OS I've used Firefox 1.5 is less stable than IE under Windows. I can't use Firefox2 because it doesn't work with some webapps I have to work with.
Just because you don't like it and it doesn't have flaws does not mean that ActiveX was not an innovation. Please refer to a dictionary as for/why/.
I've had a very good track record with Word printing consistantly between printers, better than with Open Office (barely) or Acrobat Reader (I had three HP printers and a Xerox at my old office, it was random which would properly print a PDF with images, regardless of the OS that sent the Job, under no circumstances would more than three of them work, often only one would - and which it was, was never consistant).
You can buy a product, and add new things to it to make it better than it was before, that is still innovated.
Look at word is now from where it started, or Windows for that matter.
MS/Bought/ DOS, and they innovated (or in some cases as another slasdotter mentioned, pilfered) quite a bit onto it to get another OS - the 9x branch of Windows.
Structure and design for that still exists in modern Windows. Of course, the holdover stuff is stuff that most of us would rather not have around...
Another thought: IE, look at where it was when MS bought it. Look at it now. Some of the stuff on their is their innovation, some of it is cloned from things others have done. Some of it is good, some of it is bad.
I could create a new mail program with some inventive new feature, or buy the code/license for an existing one and add it. In both cases I'm being just as innovative, but in the latter case, I'm being more effective and efficient about it.
I'm familiar with them, which I use depends on what I'm doing. I usually use portupgrades when I need to fix a screwup..
ex: is had a flag in my make.conf that dind't break anything but made some parts of KDE unstable - rather than finding the package and rebuilding, I just did a portupgrade -Rf x11/kde3 (or was it -rf? I can never remember, I always check that on the man page - I wouldn't use one of the two, too unpredictable what would be installed)
ahh, that's one of the reason I made note of my experience time with the three OSes. I figured FreeBSD was the right choice pretty quickly.
It's a good OS if you don't mind one that requires a bit of typing to do maintenance, and don't mind reading documents for a couple of hours on system maintenance tasks initially so that you know how things "work" and "are done". If you are familiar with the standard unix command line tools, then it's pretty easy, you just need to understand how the application build process works if you want to build your own apps (or thei package system works if you want to use that, which I usually don't).
I like the fact that these three commands, are all in need to install most useful applications, and the fact that there's a ncie directory to search through to find what apps are available for what you want (gentoo has some of these features, and it's the closest thing I find to a Linux distro that is sufficiently useable for me): make install clean #build a program and the programs it needs to run make batch install clean #likewise, but use the default options, don't promt for stuff make config-recursive install clean #prompt for options en-mass first, then install, nice for large ports (ex: KDE) that you want to install and fiddle with some options
very few programs don't "just work" after that. Outside of that, the trick is to make sure you don't try to install multiple conflicting packages. When I was uncertain about stuff, I used to say "just put it all in" - that bit me in the but when I tried to install every Kerberos implementation for KDE - PAM, Heimdal, Kerb5, and one other that I can't remember the name of.
However very few is not all. And sometimes I need to get help - ex: my TV Tuner driver. But the people on the newsgroup and the port maintainer were nice enough, gave me advice, and even that works now.
It's not for everyone, I know plenty of people who/could/ handle it and would still prefer Windows, I know plenty of people who have used them all extensively and prefer Linux (or MacOS to add another to the pot). *Shrug* it's a matter of how you work as to which suits you.
Well, here's my rational. And as a background, I've had maybe a total of 10 months use/admin experience with FreeBSD, and several YEARS worth with Linux and Windows, so my oppinions of Linux/Windows are not from lack of experience with Linux/Windows, and my experience with FreeBSD may be somewhat short, at their best, all three seem equaly pleasant to admin, but when it comes down to the average or worst case, I find that's where FreeBSD shines.
1) In both OSes, I've found installing new programs to be easier than in any distro of Linux that I've used (RH/FC, Ubuntu, Gentoo), namely less failures. Much moreso in Windows than in FreeBSD.
2.f) In FreeBSD, when something breaks, I've found the documentation to be much better than the documentation I find in Linux, and the error messages tend to give a bit better information on finding the source of the issue. I also find things tend to break a bit less often in Linux
2.w) In Windows, when something breaks, which is actually pretty rare in my experience (at least for 2K and XP), there's plenty of documentation online, and in the Windows help files - between the latter and Google with the right error messages pasted in, most errors I've run into aren't hard to solve/bypass
3) The FreeBSD community, on a whole, has been more friendly, and less RTFM than the Linux community. (to me at least)
4.f) FreeBSD is very much oriented to the server/enterprise mindset, with everything geared towards/just working/. Linux tends to be geared more towards what the devs want, which is the latest and greatest at a breakneck pace (though not necessarily with enterprise Linuxes and their derivatives - I ought try out CentOS some time, but FC has always seemed slow and bloated to me, compared to any other OS I've used, which makes me leery of anything based off of a RH distro). Each has their pros and cons (example: better hardware support, a larger selections of applications for any given task, and much nicer looking system administration utilities are major advantages for Linux), I just tend to find the BSD set of advantages more useful for me.
4.w) Windows tries to make everything oriented towards ease of use, so that the end user can get quite a bit done without thinking too much about it. It does abstract a lot of lower level things and make them difficult to get to. Probably the reason why I don't use my Windows box any more, now that everything I need done is done on my BSD box.
5) I find Linux is the only OS where I've spent more time trying to get things working, than with either of the other two.
And please don't call me a moron or stupid because my oppinion and experiences with the various operating systems don't match yours. People work differently with different thigns, I am not telling anyone/not/ to use Linux, I'm just trying to say there are valid reasons people don't use Linux. People see problems and approach problems differently, and thus different methods of execution of a specific task are more or less effective for various people. FreeBSD and Windows are better for me than Linux. I can't tell you which is better for you, you have to decide on your own. I will say if you/havent/ tried it, then you certainly have no right to comment on it, and even if you have tried it, you've no right to insult me because my decisions and oppinions don't match yours. Sorry for this rant, but I've gotten that kind response from similar posts before, it's rather annoying, and it wastes both my time and the time of the writer, while providing nothing productive.
In my case, I'd rather pay $200 or $300 every 2-3 years for windows on my machine than have Linux for free. I've just had too many bad experiences with Linux (and no, I'm not just talking about back in 2000, I'm talking about as recently as attempts from june to september or october of last year, and periodic attempts every six months to a year prior to that as early as 2001 or 2002).
If I didn't have the money, and didn't know about FreeBSD, I probably/would/ pirate Windows so that I could have something useful and functional, or more likely than that, just use an older version that I legitimately have a copy of.
Linux is definetly better than it used to be, and for people who don't administrate their systems (i.e. call on family and friends to do it), it's a perfectly fine OS, but for some people who administrate their own systems, it can still be a challange in a lot of ways.
On the chance you were referring to my comment, I didn't say anything about the Linux Desktop thing, only about the administration tools.
Pretty much any desktop you can get working on Distro A you should be able to get working on Distros B-Z as well. Mileage may vary, but a good admin should be able to get that done either at install time or as a post install time.
So just to reiterate, I wasn't complaining about that:-)
Heck, I run my favorite "Linux" desktop envronment on my FreeBSD box... I'd run it on windows too if it weren't too much of a hassle to set it up.
TCO of Linux being higher than Windows wouldn't completely surprise me given my own personal experience with the OS, though hearing other people's experiences, I would not bet on either outcome. It, in several of it's incarnations, has given me more grief than almost any other OS I've used/administrated (there's only one worse I can think of, sorry/.ers, it's not Windows).
That being said, I'd still like to know - is this weighted per machine on comparison, or per desktop in one set, per server in another, or is it just overall - - If it's the latter, than TCO will be best on whatever system is used least. - If it's the per server/per desktop, then it's a good measure - If it's per machine, whichever has the highest desktop:server would probably win, so it's again unfair/biased.
Also, as it's stated, there are multiple distros; with how differently things are done, I wouldn't except a low TCO for multiple distros. My experience stems from 4 major distributions, totalling maybe 10-12 versions, the administration of different distros seems to be quite high, making multi-distro administration also a challange. That right there tells me this is biased against Linux.
Finally, learning cost: Learning is a sunk cost, and not an over-time cost. Was this TCO over the first year, or was it over a longer time? Did it involve a time-related cost projection? This is relevant because most of the users would have come in knowing how things were done in Windows, but not Linux, some of the admins may have even come in that way. The initial training cost would have been comparatively high compared to the new employee training cost - another VERY important factor that most likely biased this report against Linux. Anyone know if they actually put up facts about this?
A lot of words said and conclusions made in TFA, but at the end of the day, I don't feel any more educated than before - they just gave no useful or novel (/new/ not book or corporation) data.
Patents cannot prevent, limit or contrain the production and distribution of products made using patented technologies and methods provided that said product is distributed without financial or material recompensation either directly or for supporting good and services.
1) The market is improved by keeping patents from harming free distribution of goods and services that in general better the lives of others. [i.e. patents cannot become tools of anti-generocity and anti-benevolacne] 2) The patent holders cannot be abused by other people who want to make money off of their work without providing compensation [if such is desired].
The only way to avoid the dangerous parts of such bugs is to use a VM based language, or something similar.
And those are written in C/C++/Assembly. Which means that you have to trust/they/ didn't make any mistakes in/their/ code, or if it JIT like Java, their conversion of their code into system binary code.
"Not using C" is a fools answer, another reply to this post mentioned proper testable programming paradigms (sp?) and that, regardless of language, is the correct answer.
The Java code may not have memory access, but the JVM does, and it uses that memory access to handle things for the Java code that requries underlying memory access to run.
Basically, you are saying "trust someone elses C Code, but not your own", and "Run parts of your core OS on a virtual machine"
Ex: In java if you create two strings and concatenate them, the VM has to allocate the memory for the strings and the concatenated strings - that's still memory management, and you don't know that there won't be a bug in that any more than there can be a bun in a C coded stack implementation. Or are you saying a potential bug in a VM instruction stack implementation is harmless? Maybe your saying since it's a VM the coders of it can't make mistakes?
No, I have been burned too much by Linux to consider using it seriously, ocassionally I'm convinced to give it another chance, and I am invariably burned it seems.
Slashdot Mirror
Yes, it's the highlander school of consoles!
What planet are you from? Possibly Earth?
I just used lisp as it's the oldest example I can think of...
From 1958, it's pretty old.
Don't forget the paten for an online forum to release news and provide a method for users discussing it!
Oh, and a way to rank the news and the comments on how "good" they are!
I want those two!
On a serious note: isn't the existance of the programming language LISP a good example of prior art in this case.
Very good example. I don't like either product in the least, but I nonetheless will agree that they were innovative.
I found opera to be slow, unstable, rendered poorly, and to have a clunky, ugly interface compared to IE. I gave it a couple of months at someone's insistance during the IE5 days, and didn't like it. Tried it a few times since, most of the issues improved, and it may be competative now, but I find Firefox suits my needs well enough and don't feel the urge to try opera yet again, after finding it unpleasant several times before.
IE 5+ was faster and more stable than the alternatives (Netscape, Mozilla, and even [just barely] firefox) on windows, up to IE 6. I'm not sure about 7, speed isn't the only thing in the world, and I switched during 6's lifespan. Actually I find in any OS I've used Firefox 1.5 is less stable than IE under Windows. I can't use Firefox2 because it doesn't work with some webapps I have to work with.
/why/.
Just because you don't like it and it doesn't have flaws does not mean that ActiveX was not an innovation. Please refer to a dictionary as for
I've had a very good track record with Word printing consistantly between printers, better than with Open Office (barely) or Acrobat Reader (I had three HP printers and a Xerox at my old office, it was random which would properly print a PDF with images, regardless of the OS that sent the Job, under no circumstances would more than three of them work, often only one would - and which it was, was never consistant).
Funny, I don't quite think that's accurate.
If he starts bitching about a game, I know people who will rush out to buy it.
If ever said a game is good, they would rush to put it on their "not even on a very cold day in hell" list.
buying does not indicate a lack of innovation.
/Bought/ DOS, and they innovated (or in some cases as another slasdotter mentioned, pilfered) quite a bit onto it to get another OS - the 9x branch of Windows.
You can buy a product, and add new things to it to make it better than it was before, that is still innovated.
Look at word is now from where it started, or Windows for that matter.
MS
Structure and design for that still exists in modern Windows. Of course, the holdover stuff is stuff that most of us would rather not have around...
Another thought: IE, look at where it was when MS bought it. Look at it now. Some of the stuff on their is their innovation, some of it is cloned from things others have done. Some of it is good, some of it is bad.
I could create a new mail program with some inventive new feature, or buy the code/license for an existing one and add it. In both cases I'm being just as innovative, but in the latter case, I'm being more effective and efficient about it.
I'm familiar with them, which I use depends on what I'm doing. I usually use portupgrades when I need to fix a screwup..
ex: is had a flag in my make.conf that dind't break anything but made some parts of KDE unstable - rather than finding the package and rebuilding, I just did a portupgrade -Rf x11/kde3 (or was it -rf? I can never remember, I always check that on the man page - I wouldn't use one of the two, too unpredictable what would be installed)
sadly, it'll probably be bought by "Sony Computers Online"
ahh, that's one of the reason I made note of my experience time with the three OSes. I figured FreeBSD was the right choice pretty quickly.
/could/ handle it and would still prefer Windows, I know plenty of people who have used them all extensively and prefer Linux (or MacOS to add another to the pot). *Shrug* it's a matter of how you work as to which suits you.
It's a good OS if you don't mind one that requires a bit of typing to do maintenance, and don't mind reading documents for a couple of hours on system maintenance tasks initially so that you know how things "work" and "are done". If you are familiar with the standard unix command line tools, then it's pretty easy, you just need to understand how the application build process works if you want to build your own apps (or thei package system works if you want to use that, which I usually don't).
I like the fact that these three commands, are all in need to install most useful applications, and the fact that there's a ncie directory to search through to find what apps are available for what you want (gentoo has some of these features, and it's the closest thing I find to a Linux distro that is sufficiently useable for me):
make install clean #build a program and the programs it needs to run
make batch install clean #likewise, but use the default options, don't promt for stuff
make config-recursive install clean #prompt for options en-mass first, then install, nice for large ports (ex: KDE) that you want to install and fiddle with some options
very few programs don't "just work" after that. Outside of that, the trick is to make sure you don't try to install multiple conflicting packages. When I was uncertain about stuff, I used to say "just put it all in" - that bit me in the but when I tried to install every Kerberos implementation for KDE - PAM, Heimdal, Kerb5, and one other that I can't remember the name of.
However very few is not all. And sometimes I need to get help - ex: my TV Tuner driver. But the people on the newsgroup and the port maintainer were nice enough, gave me advice, and even that works now.
It's not for everyone, I know plenty of people who
Well, here's my rational. And as a background, I've had maybe a total of 10 months use/admin experience with FreeBSD, and several YEARS worth with Linux and Windows, so my oppinions of Linux/Windows are not from lack of experience with Linux/Windows, and my experience with FreeBSD may be somewhat short, at their best, all three seem equaly pleasant to admin, but when it comes down to the average or worst case, I find that's where FreeBSD shines.
/just working/. Linux tends to be geared more towards what the devs want, which is the latest and greatest at a breakneck pace (though not necessarily with enterprise Linuxes and their derivatives - I ought try out CentOS some time, but FC has always seemed slow and bloated to me, compared to any other OS I've used, which makes me leery of anything based off of a RH distro). Each has their pros and cons (example: better hardware support, a larger selections of applications for any given task, and much nicer looking system administration utilities are major advantages for Linux), I just tend to find the BSD set of advantages more useful for me.
/not/ to use Linux, I'm just trying to say there are valid reasons people don't use Linux. People see problems and approach problems differently, and thus different methods of execution of a specific task are more or less effective for various people. FreeBSD and Windows are better for me than Linux. I can't tell you which is better for you, you have to decide on your own. I will say if you /havent/ tried it, then you certainly have no right to comment on it, and even if you have tried it, you've no right to insult me because my decisions and oppinions don't match yours. Sorry for this rant, but I've gotten that kind response from similar posts before, it's rather annoying, and it wastes both my time and the time of the writer, while providing nothing productive.
1) In both OSes, I've found installing new programs to be easier than in any distro of Linux that I've used (RH/FC, Ubuntu, Gentoo), namely less failures. Much moreso in Windows than in FreeBSD.
2.f) In FreeBSD, when something breaks, I've found the documentation to be much better than the documentation I find in Linux, and the error messages tend to give a bit better information on finding the source of the issue. I also find things tend to break a bit less often in Linux
2.w) In Windows, when something breaks, which is actually pretty rare in my experience (at least for 2K and XP), there's plenty of documentation online, and in the Windows help files - between the latter and Google with the right error messages pasted in, most errors I've run into aren't hard to solve/bypass
3) The FreeBSD community, on a whole, has been more friendly, and less RTFM than the Linux community. (to me at least)
4.f) FreeBSD is very much oriented to the server/enterprise mindset, with everything geared towards
4.w) Windows tries to make everything oriented towards ease of use, so that the end user can get quite a bit done without thinking too much about it. It does abstract a lot of lower level things and make them difficult to get to. Probably the reason why I don't use my Windows box any more, now that everything I need done is done on my BSD box.
5) I find Linux is the only OS where I've spent more time trying to get things working, than with either of the other two.
And please don't call me a moron or stupid because my oppinion and experiences with the various operating systems don't match yours. People work differently with different thigns, I am not telling anyone
In my case, I'd rather pay $200 or $300 every 2-3 years for windows on my machine than have Linux for free. I've just had too many bad experiences with Linux (and no, I'm not just talking about back in 2000, I'm talking about as recently as attempts from june to september or october of last year, and periodic attempts every six months to a year prior to that as early as 2001 or 2002).
/would/ pirate Windows so that I could have something useful and functional, or more likely than that, just use an older version that I legitimately have a copy of.
If I didn't have the money, and didn't know about FreeBSD, I probably
Linux is definetly better than it used to be, and for people who don't administrate their systems (i.e. call on family and friends to do it), it's a perfectly fine OS, but for some people who administrate their own systems, it can still be a challange in a lot of ways.
On the chance you were referring to my comment, I didn't say anything about the Linux Desktop thing, only about the administration tools.
:-)
Pretty much any desktop you can get working on Distro A you should be able to get working on Distros B-Z as well. Mileage may vary, but a good admin should be able to get that done either at install time or as a post install time.
So just to reiterate, I wasn't complaining about that
Heck, I run my favorite "Linux" desktop envronment on my FreeBSD box... I'd run it on windows too if it weren't too much of a hassle to set it up.
the administration of different distros seems to be quite high
TO:
the administration of different distros seems to be quite different
Those are two more valid possibilities.
a statement.
/.ers, it's not Windows).
TCO of Linux being higher than Windows wouldn't completely surprise me given my own personal experience with the OS, though hearing other people's experiences, I would not bet on either outcome. It, in several of it's incarnations, has given me more grief than almost any other OS I've used/administrated (there's only one worse I can think of, sorry
That being said, I'd still like to know -
is this weighted per machine on comparison, or per desktop in one set, per server in another, or is it just overall -
- If it's the latter, than TCO will be best on whatever system is used least.
- If it's the per server/per desktop, then it's a good measure
- If it's per machine, whichever has the highest desktop:server would probably win, so it's again unfair/biased.
Also, as it's stated, there are multiple distros; with how differently things are done, I wouldn't except a low TCO for multiple distros. My experience stems from 4 major distributions, totalling maybe 10-12 versions, the administration of different distros seems to be quite high, making multi-distro administration also a challange. That right there tells me this is biased against Linux.
Finally, learning cost: Learning is a sunk cost, and not an over-time cost. Was this TCO over the first year, or was it over a longer time? Did it involve a time-related cost projection? This is relevant because most of the users would have come in knowing how things were done in Windows, but not Linux, some of the admins may have even come in that way. The initial training cost would have been comparatively high compared to the new employee training cost - another VERY important factor that most likely biased this report against Linux. Anyone know if they actually put up facts about this?
A lot of words said and conclusions made in TFA, but at the end of the day, I don't feel any more educated than before - they just gave no useful or novel (/new/ not book or corporation) data.
Probably not, they are unlikely to look at patents that haven't expired.
Sadly most of MS's patents are post 95 by the article (they had around 100 in 95, 1000 in 99 and 6000 now If I remember correctly?)
Patents cannot prevent, limit or contrain the production and distribution of products made using patented technologies and methods provided that said product is distributed without financial or material recompensation either directly or for supporting good and services.
1) The market is improved by keeping patents from harming free distribution of goods and services that in general better the lives of others. [i.e. patents cannot become tools of anti-generocity and anti-benevolacne]
2) The patent holders cannot be abused by other people who want to make money off of their work without providing compensation [if such is desired].
No, Only SCO would think that the state government of Utah controls the world.
Anybody else would laugh - how the hell do they think that they can make this work, when most of the people in that industry AREN'T IN UTAH!
The only way to avoid the dangerous parts of such bugs is to use a VM based language, or something similar.
/they/ didn't make any mistakes in /their/ code, or if it JIT like Java, their conversion of their code into system binary code.
And those are written in C/C++/Assembly. Which means that you have to trust
"Not using C" is a fools answer, another reply to this post mentioned proper testable programming paradigms (sp?) and that, regardless of language, is the correct answer.
The Java code may not have memory access, but the JVM does, and it uses that memory access to handle things for the Java code that requries underlying memory access to run.
Basically, you are saying "trust someone elses C Code, but not your own", and "Run parts of your core OS on a virtual machine"
Ex: In java if you create two strings and concatenate them, the VM has to allocate the memory for the strings and the concatenated strings - that's still memory management, and you don't know that there won't be a bug in that any more than there can be a bun in a C coded stack implementation. Or are you saying a potential bug in a VM instruction stack implementation is harmless? Maybe your saying since it's a VM the coders of it can't make mistakes?
They payment fee goes in part for tech support that is provided, and potentially software that is not freely distributed.
I can't remember if RHEL has a free download or not, but last I saw, several of the software packages were not free.
No, I have been burned too much by Linux to consider using it seriously, ocassionally I'm convinced to give it another chance, and I am invariably burned it seems.
Nope, I use FreeBSD.
I thought that used the pvr250 port, the pvrxxx port was for the 150 and 500 series (I have a 150 on FBSD)