Hohoho. I just spent four hours getting a network card working in windows... hint, it's *not* autodetect, add new hardware or *network configuration*. Nooooo. After three hours getting updated drivers, newer versions, and just ending up with a non functional network device I started tearing through the PCI PnP configuration enough to get it totally not working and eventually somewhere under there finding a PCI Network Device, which, if you loaded the drivers *there* eventually moved over to pretend to be a network card. Knowledge base and vendor support turned up absolutely _nothing_ (yes, I did put the card in the machine). Neither sound or graphics were detected either on initial load; only after vendor supplied installation. A linux newbie installed Redhat 6.2 on the very same machine in less than an hour; all hardware detected and configured without a hitch (altho, I did help her check the hardware for compatibility before purchase). Total installation reboots Windows: 49 Total installation reboots Linux: 1 (plus one for lilo reconfig after windows wiped it) Windows being easy or getting the job done is just an fairy tale. Compared to Linux it's a virtual nightmare to install and 'get the job done'. I prefer the HOWTO's or config files to windows blurting out 'code 2'.
Even more fun is the NYTimes story about the same thing. It also goes into detail about how Microsofts 'polls' about the lack of public support for the prosecution of Microsoft.
Of course, the polls asked people to rank 'prosecute Microsoft' against other actions like 'prosecute airlines who dont service their airplanes', 'overcharging for prescription medicine', etc. When faced with plague, cholera, cancer or a really bad case of flu, most seem to have chosen the flu...
Sure, you trust the government. Does that mean you trust incompetent sleazebag Joe Schmoe who reviewed your 'echelon' calls while you said 'Im going to bomb on this exam today' and you promptly get labeled potential terrorist? Think you could get that struck from your record? After all, you're just another terrorist, you know. Most people trust the government in abstraction, in a democracy, but any government is full of Joe Shmoes who do their 9 to 5 day and couldnt care less if they mislabel one out of a thousand calls. Should the power to label people terrorists really be in such hands? Does catching a few less than smart blabbering criminals justify screwing some innocent people over royally and ruining their lives? Not to mention the purely personal abuse it opens up for?
Oh, so true. I can hardly count the number of applications or systems where, after the programming job is done by some guys fresh out of some computer training program working for some it consultant firm, $ multiple millions have been paid for the hardware, and then the sysadmins sit tearing their hair out trying to keep a system running that is taking ten times the resources it should and basically screaming at the meetings that they have to fix their bloody crap code.
Usually it takes a year or two to get through to management that they've paid for junk that shouldnt see the light of day, and that's usually around the times that the vendors delivering the hardware start looking embarrased, saying they cant sell us hardware that can handle the pressure, we are already overdimensioned by far, and maybe something should be done about the application...
Ok, I can sympathize with not wanting the overhead of X, but if you can afford the overhead of Qt, single chip capacities arent that far off supporting X too...
And, hey, while we are at it, sure, a webbrowser on a toaster must make sense in someones mind, but what _I_ want is my toaster, fridge and oven consoles on my DESKTOP! Which means X!
This question is actually very simple. Each piece of separately copyrighted code is subject to its own license (altho the question of derivatives makes it a bit more complicated). However, these licenses _must_ be compatible or you cannot combine the two separate pieces. For GPL code you can probably place a patch under BSD (without ad clause) license, but the work in whole will still be subject to the GPL (unless the patch code again is extracted and distributed under its own license).
If the license is not compatible with the GPL combining the works will result in an undistributable source, an undistributable binary, etc. Even placing the patch and original source in the same directory or related to eachother on an ftp site may constitute an infringement depending on the interpretation of packaging.
In this case it's probably because they have the comments included on the page (bad design, keep static content and dynamic content separate if possible), and regenerate it on the fly (also bad design, pregenerate static code at useful intervals for best speed).
For a small website it is acceptable, for a major site it would be a disaster, since the server load would be huge for unessesary dynamically generated content (and degrade horrendously as each session lengthens and sucks up resources, multiplying the concurrent sessions many times), and any caching would be impossible (both on-site as you realize you misjudged interest and could have thrown in a fast webcache in a short time and off-site in browsers and webcaches), thus increasing your load with more unnecessary hits.
Well, from my experience of the certification touters, I usually respect someone who has 1 year experience and no certificate more than someone with the same time and a certificate.
The one without a certificate is probably good at his job or he wouldnt have it, but with the one with the certificate you never know if it's just another of the certcrams'r'us guys.
Computer problems add up to a lot more than just a multiple choice quiz where you get to study the questions in advance...
And, of course, using.asp pages with static content is a good way to screw up a lot of caching, which slows down at least my reloading time from a fraction of a second to several thousand times that... so maybe there are other reasons for not using.asp when it isnt needed, eh?
As far as my Unix job searching experience goes, I've never even once been asked for any certification. My current employee didnt even ask any kind of technical questions during the interview. I outlined the range of systems experience I'd had and my programming skills, and they called my references and that was that. Personally, I'd be iffy about a job if they asked for certification, and seriously doubt their employee evaluation skills. Of course, I feel for you. For the Unix related 'certifications' Ive been through there are never any exams. People usually go to those classes to learn and they will use the time until the know the subject, so there is no reason to have an exam. The average class would probably, on being presented with an exam, look at the teacher and go have a beer with the comment that they've been out of gradeschool for a few years now... Then again, I have long hair, so people usually assume Im a unix geek.
Slashdot Mirror
Hohoho. I just spent four hours getting a network card working in windows... hint, it's *not* autodetect, add new hardware or *network configuration*. Nooooo. After three hours getting updated drivers, newer versions, and just ending up with a non functional network device I started tearing through the PCI PnP configuration enough to get it totally not working and eventually somewhere under there finding a PCI Network Device, which, if you loaded the drivers *there* eventually moved over to pretend to be a network card. Knowledge base and vendor support turned up absolutely _nothing_ (yes, I did put the card in the machine). Neither sound or graphics were detected either on initial load; only after vendor supplied installation. A linux newbie installed Redhat 6.2 on the very same machine in less than an hour; all hardware detected and configured without a hitch (altho, I did help her check the hardware for compatibility before purchase). Total installation reboots Windows: 49 Total installation reboots Linux: 1 (plus one for lilo reconfig after windows wiped it) Windows being easy or getting the job done is just an fairy tale. Compared to Linux it's a virtual nightmare to install and 'get the job done'. I prefer the HOWTO's or config files to windows blurting out 'code 2'.
Even more fun is the NYTimes story about the same thing. It also goes into detail about how Microsofts 'polls' about the lack of public support for the prosecution of Microsoft.
Of course, the polls asked people to rank 'prosecute Microsoft' against other actions like 'prosecute airlines who dont service their airplanes', 'overcharging for prescription medicine', etc. When faced with plague, cholera, cancer or a really bad case of flu, most seem to have chosen the flu...
Sure, you trust the government. Does that mean you trust incompetent sleazebag Joe Schmoe who reviewed your 'echelon' calls while you said 'Im going to bomb on this exam today' and you promptly get labeled potential terrorist? Think you could get that struck from your record? After all, you're just another terrorist, you know. Most people trust the government in abstraction, in a democracy, but any government is full of Joe Shmoes who do their 9 to 5 day and couldnt care less if they mislabel one out of a thousand calls. Should the power to label people terrorists really be in such hands? Does catching a few less than smart blabbering criminals justify screwing some innocent people over royally and ruining their lives? Not to mention the purely personal abuse it opens up for?
Oh, so true. I can hardly count the number of applications or systems where, after the programming job is done by some guys fresh out of some computer training program working for some it consultant firm, $ multiple millions have been paid for the hardware, and then the sysadmins sit tearing their hair out trying to keep a system running that is taking ten times the resources it should and basically screaming at the meetings that they have to fix their bloody crap code.
Usually it takes a year or two to get through to management that they've paid for junk that shouldnt see the light of day, and that's usually around the times that the vendors delivering the hardware start looking embarrased, saying they cant sell us hardware that can handle the pressure, we are already overdimensioned by far, and maybe something should be done about the application...
Ok, I can sympathize with not wanting the overhead of X, but if you can afford the overhead of Qt, single chip capacities arent that far off supporting X too...
And, hey, while we are at it, sure, a webbrowser on a toaster must make sense in someones mind, but what _I_ want is my toaster, fridge and oven consoles on my DESKTOP! Which means X!
This question is actually very simple. Each piece of separately copyrighted code is subject to its own license (altho the question of derivatives makes it a bit more complicated). However, these licenses _must_ be compatible or you cannot combine the two separate pieces. For GPL code you can probably place a patch under BSD (without ad clause) license, but the work in whole will still be subject to the GPL (unless the patch code again is extracted and distributed under its own license).
If the license is not compatible with the GPL combining the works will result in an undistributable source, an undistributable binary, etc. Even placing the patch and original source in the same directory or related to eachother on an ftp site may constitute an infringement depending on the interpretation of packaging.
Of course, ianal, etc.
In this case it's probably because they have the comments included on the page (bad design, keep static content and dynamic content separate if possible), and regenerate it on the fly (also bad design, pregenerate static code at useful intervals for best speed).
For a small website it is acceptable, for a major site it would be a disaster, since the server load would be huge for unessesary dynamically generated content (and degrade horrendously as each session lengthens and sucks up resources, multiplying the concurrent sessions many times), and any caching would be impossible (both on-site as you realize you misjudged interest and could have thrown in a fast webcache in a short time and off-site in browsers and webcaches), thus increasing your load with more unnecessary hits.
Well, from my experience of the certification touters, I usually respect someone who has 1 year experience and no certificate more than someone with the same time and a certificate.
The one without a certificate is probably good at his job or he wouldnt have it, but with the one with the certificate you never know if it's just another of the certcrams'r'us guys.
Computer problems add up to a lot more than just a multiple choice quiz where you get to study the questions in advance...
And, of course, using .asp pages with static content is a good way to screw up a lot of caching, which slows down at least my reloading time from a fraction of a second to several thousand times that... so maybe there are other reasons for not using .asp when it isnt needed, eh?
As far as my Unix job searching experience goes, I've never even once been asked for any certification. My current employee didnt even ask any kind of technical questions during the interview. I outlined the range of systems experience I'd had and my programming skills, and they called my references and that was that. Personally, I'd be iffy about a job if they asked for certification, and seriously doubt their employee evaluation skills. Of course, I feel for you. For the Unix related 'certifications' Ive been through there are never any exams. People usually go to those classes to learn and they will use the time until the know the subject, so there is no reason to have an exam. The average class would probably, on being presented with an exam, look at the teacher and go have a beer with the comment that they've been out of gradeschool for a few years now... Then again, I have long hair, so people usually assume Im a unix geek.