For backups I use a wonderful (free as in beer) program called XXCLONE (www.xxclone.com). The program does a complete backup of your installation to any other partition or hard drive and then makes it bootable. I use it with a 60 gig laptop drive I got on eBay I put in an external enclosure. This way if my internal drive eats it, all I have to do is swap the drives out. It works like a charm, I've had the pleasure of using it before when migrating from a 20 gig drive to a new 60 gig one - it is literally a two click process.
If they don't do what I need them to do, then no, 126,119 programs is not enough. When it comes to doing exactly what I need done, it's simple - one program is enough.
Take the bibliographic program "EndNote," for example. Nothing in the Linux world comes close to what this program does in conjunction with MS Word. Nothing. OpenOffice? No. Bibus? A good project, but not there yet...(I can't wait until it is).
Qantity can simply not make up for quality when it comes to specialized software. If EndNote was ported to Linux, along with a solid statistics program with a good GUI, I'll bet many more academics would make the switch. (WINE doesn't work with EndNote, nor does SPSS or STATA.)
The lack of these types of specialized programs has kept me and others in my field from being a full-time Linux users. In this day and age, our department (and I'm sure most others) would benefit greatly, monetarily, by using Linux.
The responses you've gotten to your post show why Linux isn't being adopted as quickly as it could.
For every useful reply suggesting how to remedy your problem, you get two condescending remarks saying you somehow should just know what to do. Geez...
Yes, you are right that people who are "coding Flash-based web sites deserve swift kicks to the buttocks." But, that does not change the fact that they ARE still coding them, and people want an OS or browser that works well with them.
Aside from offering a poor product at a high price, their customer service is frustrating as all get-out. It took me hours and hours to cancel my account in 2003. Their cancellation process is so difficult, it ends up being comparable to theft.
For those of you who don't loathe AOL enough, The Consumerist recently posted their customer retention guide. It's an interesting read and shows just how institutionalized their horrible policies are: http://www.consumerist.com/consumer/exclusive/aol- retention-manual-uploaded-in-full-188310.php
Antivirus and antispyware companies will have a customer base for another decade at least because not everyone will make the jump to Vista at once. Many people still run Windows 98 and even *gasp* Windows 95. The many W95,W98, and XP users will provide income for antivirus companies for a long time ahead.
It works very well right out of the box for odd hardware. I'm a fulltime Kubuntu 5.10 user. I have a Sony R505EL laptop that has a relatively strange hardware configuration. The CD-ROM is on a docking bay that the computer communicates with via firewire.
Most linux distros get to a certain point in the install and then "lose" the CD drive because they don't detect and configure this odd setup correctly.
There are ways around this, but it can be a major pain in the arse to screw with modules during install. The problem occurs with Slackware, Debian, Mandrake, and Suse (yes, my ISP loves me for downloading all these iso's).
Anyway, Kubuntu detects my configuration and installs without a hitch. I thought about switching to Debian just last week and figured it would install fine because Kubuntu does. However, I was wrong.
0.3% can absolutely be a statistically significant deviation given a large enough sample size. This is one of the problems in studies with huge samples; lots of things appear to be significantly related.
In the end it is up to the author of this or any other study to prove causation (remember, correlation is not equal to causation) and to discuss the implications of his or her findings. This is the case regardless of sample size. The author should make an argument, based on sound logic, as to whether the general public should be concerned their findings. In this case it is a 0.3% increase in disease. The benefits that come with being highly educated surely outweigh the small increased likelihood of disease.
The responder above has it right, what is interesting is that the finding is statistically significant AND it goes in the opposite direction than current theory predicts.
So, this new finding does not support previous studies. This fact is interesting, but not likely to be remarkable to people who do not study this sort of thing. The fact that the effect is significant in the opposite direction is remarkable and of interest to a wider audience.
And remember, a small percentage of "significant" findings are false positives. This is the nature of statistics. In medical research, the the odds of getting an incorrect result on a statistical test of significance is usually less than one per 1000 (i.e. "p" is equal to or less than 0.001). In other words, if you ran the same test 1000 times on the same data, you would get a significant result in error one time or less.
Last, don't forget that 14% of all statistics are made up on the spot.
Slashdot Mirror
For backups I use a wonderful (free as in beer) program called XXCLONE (www.xxclone.com). The program does a complete backup of your installation to any other partition or hard drive and then makes it bootable. I use it with a 60 gig laptop drive I got on eBay I put in an external enclosure. This way if my internal drive eats it, all I have to do is swap the drives out. It works like a charm, I've had the pleasure of using it before when migrating from a 20 gig drive to a new 60 gig one - it is literally a two click process.
If they don't do what I need them to do, then no, 126,119 programs is not enough. When it comes to doing exactly what I need done, it's simple - one program is enough.
Take the bibliographic program "EndNote," for example. Nothing in the Linux world comes close to what this program does in conjunction with MS Word. Nothing. OpenOffice? No. Bibus? A good project, but not there yet...(I can't wait until it is).
Qantity can simply not make up for quality when it comes to specialized software. If EndNote was ported to Linux, along with a solid statistics program with a good GUI, I'll bet many more academics would make the switch. (WINE doesn't work with EndNote, nor does SPSS or STATA.)
The lack of these types of specialized programs has kept me and others in my field from being a full-time Linux users. In this day and age, our department (and I'm sure most others) would benefit greatly, monetarily, by using Linux.
The responses you've gotten to your post show why Linux isn't being adopted as quickly as it could. For every useful reply suggesting how to remedy your problem, you get two condescending remarks saying you somehow should just know what to do. Geez...
Yes, you are right that people who are "coding Flash-based web sites deserve swift kicks to the buttocks." But, that does not change the fact that they ARE still coding them, and people want an OS or browser that works well with them.
Aside from offering a poor product at a high price, their customer service is frustrating as all get-out. It took me hours and hours to cancel my account in 2003. Their cancellation process is so difficult, it ends up being comparable to theft. For those of you who don't loathe AOL enough, The Consumerist recently posted their customer retention guide. It's an interesting read and shows just how institutionalized their horrible policies are: http://www.consumerist.com/consumer/exclusive/aol- retention-manual-uploaded-in-full-188310.php
Antivirus and antispyware companies will have a customer base for another decade at least because not everyone will make the jump to Vista at once. Many people still run Windows 98 and even *gasp* Windows 95. The many W95,W98, and XP users will provide income for antivirus companies for a long time ahead.
This reminds me so much of the Nintendo Power Glove! Awesome! http://www.gamerevolution.com/oldsite/articles/ran ts/powerglove.jpg
It works very well right out of the box for odd hardware. I'm a fulltime Kubuntu 5.10 user. I have a Sony R505EL laptop that has a relatively strange hardware configuration. The CD-ROM is on a docking bay that the computer communicates with via firewire. Most linux distros get to a certain point in the install and then "lose" the CD drive because they don't detect and configure this odd setup correctly. There are ways around this, but it can be a major pain in the arse to screw with modules during install. The problem occurs with Slackware, Debian, Mandrake, and Suse (yes, my ISP loves me for downloading all these iso's). Anyway, Kubuntu detects my configuration and installs without a hitch. I thought about switching to Debian just last week and figured it would install fine because Kubuntu does. However, I was wrong.
but does it run DOOM?
...this is my favorite. Check it out. http://www.pocketmod.com/
0.3% can absolutely be a statistically significant deviation given a large enough sample size. This is one of the problems in studies with huge samples; lots of things appear to be significantly related.
In the end it is up to the author of this or any other study to prove causation (remember, correlation is not equal to causation) and to discuss the implications of his or her findings. This is the case regardless of sample size. The author should make an argument, based on sound logic, as to whether the general public should be concerned their findings. In this case it is a 0.3% increase in disease. The benefits that come with being highly educated surely outweigh the small increased likelihood of disease.
The responder above has it right, what is interesting is that the finding is statistically significant AND it goes in the opposite direction than current theory predicts.
So, this new finding does not support previous studies. This fact is interesting, but not likely to be remarkable to people who do not study this sort of thing. The fact that the effect is significant in the opposite direction is remarkable and of interest to a wider audience.
And remember, a small percentage of "significant" findings are false positives. This is the nature of statistics. In medical research, the the odds of getting an incorrect result on a statistical test of significance is usually less than one per 1000 (i.e. "p" is equal to or less than 0.001). In other words, if you ran the same test 1000 times on the same data, you would get a significant result in error one time or less.
Last, don't forget that 14% of all statistics are made up on the spot.
Kyle-
Q: Why not "just switch to Mac/OSX"? A: Some of us like having more than one mouse button.