systemd works great for millions of people. I am sure most of them did not even notice the change (unless their distro was previously using sysvinit and their boot got much faster). So either you only tried systemd on a really buggy distro, maybe a development release, or you did had some complicated set of custom boot scripts that broke, or you have a really odd hardware setup. Most distros still write the old ascii log files, so i don't see how debugging could be harder.
High energy physics often involves setting up systems for data acquisition, experiment control, data management, etc, that need to run for 10 or 15 years with minimal maintenance or change. That requires pretty long support cycles.
He said "Newton's laws work just fine for explaining pretty much anything a human can put their hands on".
So it would seem that you are the idiot who can't read. Someone needs to go back to pedant school.
Are they not using the trick of selling chips with a defective core as a lower core count (like the old Phenom X3). I assumed that was why you get strange numbers like 61 cores.
Nobody disagrees that ionising radiation can cause cancer. But at the small doses that members of the public were exposed to there is valid debate over whether the increase in cancer risk is zero, almost zero or just tiny. Sensational headlines spread fear and stress, which likely is having a worse effect on people than the actual radiation.
If you give everyone an ultrasound then you find lots of nodules and cysts. Repeat the study in other parts of japan unaffected by radiation and you get the same. http://www.nature.com/articles...
The article has a pretty good bibliography. Reading though the actual journal articles they all seem to agree, when you do mass screening you find lots of potentially cancerous lumps, more than you'd expect from national cancer stats, but you find the same even in uncontaminated areas.
OpenOffice has known security issues (its not safe to open untrusted documents with it). Given how few developers are still left at OO, it could be months or years before a fixed version is released (they recommend that you manually delete one of its library files).
"Volkswagen’s rigging of emissions tests for 11m cars means they may be responsible for nearly 1m tonnes of air pollution every year, roughly the same as the UK’s combined emissions for all power stations, vehicles, industry and agriculture, a Guardian analysis suggests." http://www.theguardian.com/bus...
Separating out everything into modules is going to be hard. Even PCs which have no space/weight constraints are not fully modular. Much harder for a phone where things like the CPU and RAM are typically soldered on top of each other.
But you can take a more modest goal of making the major components replaceable, repairable and upgradable, for example, https://www.fairphone.com/2015...
Many of the developers were already fed up with Sun's poor stewardship of the OpenOffice.org project, hence the go-oo project which hosted many improvements that Sun did not integrate and was the basis of OpenOffice packages that most Linux distributions shipped. Oracle was just the final straw that catalysed a true fork (go-oo was more of a patch set that needed to follow OOo). LO used the GPL because that was pretty much the only option (and maybe the developers think it is better for LO than a permissive licence). LO made big strides, for example in code clean up and build systems even before Oracle decided dump the code on the Apache foundation. The permissive was probably to keep IBM happy, as they have previously released closed derivatives of OO.
What code exactly have LO stolen from AOO? There is nothing new in AOO to be worth taking, LO has always been a step ahead of AOO (and is about 3 steps ahead now). The only possible example is the sidebar, but that was developed by IBM, not Oracle or Apache.
Personally I think Apache should de-list AOO on the grounds that they can't even produce security updates in a timely manor.
i've had thinkpads with both the top led and the backlit keyboard. I prefer the latter, but it is quite nice to have the LED illuminate your work space a little. Maybe there could be room for both.
Because (assuming it is like fairphone 1) it does not ship with the google apps and services. You can use fdroid to install opensource apps and whichever search engine you want.
From a scientific simulation point of view the worst possible outcome of a simulation is a result that looks plausible but is in fact wrong. I'd much prefer my program to halt or crash, so that I can start to look where the bug is.
Yes sometimes its a pain. Say I want to measure the velocity by looking at the time taken to cross some region, and sometimes I have zero length regions. Yes i'll get an error, headdesk and add a suitable check. But it will take knowledge of the problem to work out if I should just not report a velocity for that region, or maybe assume it is the same as the previous one, or average the previous and next. If the expression just returned 0 (or 1 (seen as x/x=1)), and that happened to be within the range of plausible values, then it could take weeks or months to spot the issue.
Slashdot Mirror
systemd works great for millions of people. I am sure most of them did not even notice the change (unless their distro was previously using sysvinit and their boot got much faster). So either you only tried systemd on a really buggy distro, maybe a development release, or you did had some complicated set of custom boot scripts that broke, or you have a really odd hardware setup. Most distros still write the old ascii log files, so i don't see how debugging could be harder.
High energy physics often involves setting up systems for data acquisition, experiment control, data management, etc, that need to run for 10 or 15 years with minimal maintenance or change. That requires pretty long support cycles.
Isn't that what the Berkeley Earth study tried to show http://berkeleyearth.org/about...
He said "Newton's laws work just fine for explaining pretty much anything a human can put their hands on".
So it would seem that you are the idiot who can't read. Someone needs to go back to pedant school.
Are they not using the trick of selling chips with a defective core as a lower core count (like the old Phenom X3). I assumed that was why you get strange numbers like 61 cores.
Nobody disagrees that ionising radiation can cause cancer. But at the small doses that members of the public were exposed to there is valid debate over whether the increase in cancer risk is zero, almost zero or just tiny. Sensational headlines spread fear and stress, which likely is having a worse effect on people than the actual radiation.
I would not want to move my children to somewhere that is increasing its fossil fuel use.
If you give everyone an ultrasound then you find lots of nodules and cysts. Repeat the study in other parts of japan unaffected by radiation and you get the same.
http://www.nature.com/articles...
The article has a pretty good bibliography. Reading though the actual journal articles they all seem to agree, when you do mass screening you find lots of potentially cancerous lumps, more than you'd expect from national cancer stats, but you find the same even in uncontaminated areas.
No, it finds many tiny (often harmless) tumours, nodules and cysts, that would go unnoticed normally.
Nope—There’s No Thyroid Cancer Epidemic in Fukushima
http://thebreakthrough.org/ind...
Elevated by increased screening, not by radiation. http://thebreakthrough.org/ind...
OpenOffice has known security issues (its not safe to open untrusted documents with it). Given how few developers are still left at OO, it could be months or years before a fixed version is released (they recommend that you manually delete one of its library files).
LibreOffice fixed the issue months ago.
"It is estimated that the effects of NO 2 on mortality are equivalent to 23,500 deaths annually in the UK" -- UK dept. for environment and rural affairs https://consult.defra.gov.uk/a... http://www.bbc.co.uk/news/busi...
"Volkswagen’s rigging of emissions tests for 11m cars means they may be responsible for nearly 1m tonnes of air pollution every year, roughly the same as the UK’s combined emissions for all power stations, vehicles, industry and agriculture, a Guardian analysis suggests." http://www.theguardian.com/bus...
Cyanogen Inc and CyanogenMod are separate things.
Separating out everything into modules is going to be hard. Even PCs which have no space/weight constraints are not fully modular. Much harder for a phone where things like the CPU and RAM are typically soldered on top of each other.
But you can take a more modest goal of making the major components replaceable, repairable and upgradable, for example, https://www.fairphone.com/2015...
Companies won't use any software that they can't make closed source derivatives of. That why no company uses MS Office.
Many of the developers were already fed up with Sun's poor stewardship of the OpenOffice.org project, hence the go-oo project which hosted many improvements that Sun did not integrate and was the basis of OpenOffice packages that most Linux distributions shipped. Oracle was just the final straw that catalysed a true fork (go-oo was more of a patch set that needed to follow OOo). LO used the GPL because that was pretty much the only option (and maybe the developers think it is better for LO than a permissive licence). LO made big strides, for example in code clean up and build systems even before Oracle decided dump the code on the Apache foundation. The permissive was probably to keep IBM happy, as they have previously released closed derivatives of OO.
What code exactly have LO stolen from AOO? There is nothing new in AOO to be worth taking, LO has always been a step ahead of AOO (and is about 3 steps ahead now). The only possible example is the sidebar, but that was developed by IBM, not Oracle or Apache.
Personally I think Apache should de-list AOO on the grounds that they can't even produce security updates in a timely manor.
Not putting all my eggs in one basket. Imagine loosing 2TB at once.
Personally I use 100s of 8GB drivers and a nest of usb hubs and adapters. When they die (a few per week) I only loose a few GB.
Also some GPL derived drivers that they have been distributing to their customers. https://twitter.com/mjg59/stat...
i've had thinkpads with both the top led and the backlit keyboard. I prefer the latter, but it is quite nice to have the LED illuminate your work space a little. Maybe there could be room for both.
Main issues are mining, manufacturing, life cycle and social entrepreneurship. https://www.fairphone.com/road...
A chunk of the extra cost comes from small volume production. But anyway they are transparent about all the costs https://www.fairphone.com/proj...
Because (assuming it is like fairphone 1) it does not ship with the google apps and services. You can use fdroid to install opensource apps and whichever search engine you want.
From a scientific simulation point of view the worst possible outcome of a simulation is a result that looks plausible but is in fact wrong. I'd much prefer my program to halt or crash, so that I can start to look where the bug is.
Yes sometimes its a pain. Say I want to measure the velocity by looking at the time taken to cross some region, and sometimes I have zero length regions. Yes i'll get an error, headdesk and add a suitable check. But it will take knowledge of the problem to work out if I should just not report a velocity for that region, or maybe assume it is the same as the previous one, or average the previous and next. If the expression just returned 0 (or 1 (seen as x/x=1)), and that happened to be within the range of plausible values, then it could take weeks or months to spot the issue.