The GPL governs how you use the code. It dictates how you must behave if you change it.
No it doesn't! This was exactly the point which was being made. I can take GPL code and change it in any way I like and use that modified version and the GPL license imposes no conditions on me at all.
However, if I take my modified version of the code and pass (or sell) it on to someone else then the GPL kicks in. As soon as I distribute it I have to then release my modified code under the GPL.
Try using modern Ubuntu and Fedora 6 test 3. Boot up time is so slow it's not even funny. Most people these days run laptops, so that is VERY IMPORTANT. And starting the Xserver is so slow, it's fscking pathetic. REALLY annoying. So when I logout, it takes 20-30 seconds to log back in. This is on a Yonah system with 512MiB RAM (macbook)
Personally I've not found a problem with boot time on either the Ubuntu or Fedora systems I run, but maybe I expect less (or have less services running).
The bigger point about this is that on my MacBook I can't really tell you how long it takes to boot because I never shut it down. The best solution for laptops is not to speed up booting, it's to make suspend work properly. My Mac is a joy to use on this front - it shuts down in about 5 seconds, and wakes up just as quickly.
To be fair, Linux is getting MUCH better in this regard. Suspend support has really come on in the
last year or so. If this can be made reliable then who cares how long your machine takes to boot.
From what I've seen the problem is very widespread - however it isn't always noticed.
I'm on my third MacBook Pro (for other problems, not the noise!), and all the ones I've had have made this noise, as has every other one I've seen. However probably more than 50% of people who've listened to my machine couldn't hear the noise.
The whine is very high pitched and it seems that lots of people just don't hear that frequency. However I can tell you that if you do hear it it's like fingernails down a blackboard. It nearly drove me mad until I found QuietMBP which immediately shut it up.
I think this is the best suggestion. You probably don't need to buy stuff, just let it be known that you're prepared to take the old bits that loads of people will have sitting in a back room somewhere.
A great example of what can be done this way is the computing museum at Bletchley Park in the UK. It's basically just a big shed where they've set up all the donated equipment they could get hold of. It's laid out roughly in chronological order and everything that can run does.
It's all very hands on. The volunteers are keen to know if you can get anything working which currently doesn't and many of the machines are set up to actually do something which the public can interact with. It all feels very amateur - but all the better for that.
Unfortunately the article is a bit light on details. It says that Sun are going to make the JRE easier to redistribute but that on it's own isn't enough for many distros. It would also have to be at least able to be repackaged (so it goes somewhere more friendly that the Sun supplied RPM) and preferably modified (to make it play nicer with the rest of the system) before it's really useful.
Also, it's a shame it seems they're only going to include the JRE. Nice and easy for linux users to run java programs. Shame they won't be able to write any...
This doesn't seem like as much of a breakthrough as they're claiming. PCR is basically a system where you can amplify DNA by putting it through a series of heating / cooling cycles in the presence of a thermostable enzyme which does the actual amplification. Labs already use expensive peltier heaters/coolers to make this pretty efficient.
All this company have done is make a machine which heats up and cools down faster. The basic biochemistry is still the same. For most PCR reactions the rate limiting step isn't the ramping between temperatures but rather the length of time you have to leave to let the enzyme to let it copy the DNA (normally calculated around 30secs per 1000bases - though it's probaby faster than that).
The only big win for this would be if you're amplifying very small stretches of DNA (a few tens of bases) when the temperature ramping times could be significant. Even so it's still going to be far from instantaneous.
The cutting down of the core distribution was an intentional decision and is arguably a good rather than a bad thing. It arose from the (eventual) appearance of Fedora extras which is the community maintanined set of packages for fedora. This marks a passing of control from Redhat to the community and pushes Fedora towards being the community oriented distribution it was promised to be. You can expect that the core FC5 distribution will be even smaller than FC4 with more packages moved to Extras.
What was bad about the FC4 release is that Extras has yet to be tied in to the installer, which means that the installing or upgrading of these packages must be done post install. One of the goals for FC5 is to have Extras available at install time.
I think that a more realistic view of bioinformatics is slowly emerging. A few years back everyone was saying that purely theory based bioinformatics was going to replace much of the current bench work - and this has largely failed to materialise.
What has happened is that bioinformatics has become more pervasive in the general research community, and this has presented opportunities for pure bioinformaticians. I think that there are two potential roles for bioinformaticians:
Development/Research - These are the people who develop new analysis techniques and write the tools which eventually become available to everyone. They are usually grant funded and work like any other research group.
(What I do - and am very happy with thankyou!) Practical Bioinformatics - Bioinformatians who work alongside wet bench groups to help manage and analyse their data. They use and adapt the techniques developed and make them accesible to bench science.
Unfortunately it seems that a lot of the new batch of bioinformatics graduates (mostly MScs) don't fall easily into either of these camps. They often transfer straight from a biology BSc to a bioinformatics MSc, and emerge at the end with computing skills too basic to take up a development post, and have no experience to be able to undertake a practical or advisory role. Not surprisingly they then have problems finding the job they expected to be waiting for them.
Slashdot Mirror
No it doesn't! This was exactly the point which was being made. I can take GPL code and change it in any way I like and use that modified version and the GPL license imposes no conditions on me at all.
However, if I take my modified version of the code and pass (or sell) it on to someone else then the GPL kicks in. As soon as I distribute it I have to then release my modified code under the GPL.
Personally I've not found a problem with boot time on either the Ubuntu or Fedora systems I run, but maybe I expect less (or have less services running).
The bigger point about this is that on my MacBook I can't really tell you how long it takes to boot because I never shut it down. The best solution for laptops is not to speed up booting, it's to make suspend work properly. My Mac is a joy to use on this front - it shuts down in about 5 seconds, and wakes up just as quickly.
To be fair, Linux is getting MUCH better in this regard. Suspend support has really come on in the last year or so. If this can be made reliable then who cares how long your machine takes to boot.
From what I've seen the problem is very widespread - however it isn't always noticed.
I'm on my third MacBook Pro (for other problems, not the noise!), and all the ones I've had have made this noise, as has every other one I've seen. However probably more than 50% of people who've listened to my machine couldn't hear the noise.
The whine is very high pitched and it seems that lots of people just don't hear that frequency. However I can tell you that if you do hear it it's like fingernails down a blackboard. It nearly drove me mad until I found QuietMBP which immediately shut it up.
I think this is the best suggestion. You probably don't need to buy stuff, just let it be known that you're prepared to take the old bits that loads of people will have sitting in a back room somewhere.
A great example of what can be done this way is the computing museum at Bletchley Park in the UK. It's basically just a big shed where they've set up all the donated equipment they could get hold of. It's laid out roughly in chronological order and everything that can run does.
It's all very hands on. The volunteers are keen to know if you can get anything working which currently doesn't and many of the machines are set up to actually do something which the public can interact with. It all feels very amateur - but all the better for that.
Anyone who gets the chance to visit should do so. Their web site is at http://www.retrobeep.org/
Unfortunately the article is a bit light on details. It says that Sun are going to make the JRE easier to redistribute but that on it's own isn't enough for many distros. It would also have to be at least able to be repackaged (so it goes somewhere more friendly that the Sun supplied RPM) and preferably modified (to make it play nicer with the rest of the system) before it's really useful.
Also, it's a shame it seems they're only going to include the JRE. Nice and easy for linux users to run java programs. Shame they won't be able to write any...
This doesn't seem like as much of a breakthrough as they're claiming. PCR is basically a system where you can amplify DNA by putting it through a series of heating / cooling cycles in the presence of a thermostable enzyme which does the actual amplification. Labs already use expensive peltier heaters/coolers to make this pretty efficient.
All this company have done is make a machine which heats up and cools down faster. The basic biochemistry is still the same. For most PCR reactions the rate limiting step isn't the ramping between temperatures but rather the length of time you have to leave to let the enzyme to let it copy the DNA (normally calculated around 30secs per 1000bases - though it's probaby faster than that).
The only big win for this would be if you're amplifying very small stretches of DNA (a few tens of bases) when the temperature ramping times could be significant. Even so it's still going to be far from instantaneous.
The cutting down of the core distribution was an intentional decision and is arguably a good rather than a bad thing. It arose from the (eventual) appearance of Fedora extras which is the community maintanined set of packages for fedora. This marks a passing of control from Redhat to the community and pushes Fedora towards being the community oriented distribution it was promised to be. You can expect that the core FC5 distribution will be even smaller than FC4 with more packages moved to Extras. What was bad about the FC4 release is that Extras has yet to be tied in to the installer, which means that the installing or upgrading of these packages must be done post install. One of the goals for FC5 is to have Extras available at install time.
What has happened is that bioinformatics has become more pervasive in the general research community, and this has presented opportunities for pure bioinformaticians. I think that there are two potential roles for bioinformaticians:
Unfortunately it seems that a lot of the new batch of bioinformatics graduates (mostly MScs) don't fall easily into either of these camps. They often transfer straight from a biology BSc to a bioinformatics MSc, and emerge at the end with computing skills too basic to take up a development post, and have no experience to be able to undertake a practical or advisory role. Not surprisingly they then have problems finding the job they expected to be waiting for them.