On the one hand I fully agree with you. Don't complain if your computer is not working with the latest software/websites/hardware if you don't bother to keep it up-to-date.
On the other hand sometimes you don't have any other choice, as you might not be in the position to change the OS.
I am a working for an insurance company (one of the BIG ones), and all PCs in our agencies in Germany are still running Windows 2000.
The company's failure to have a proper upgrade strategy in place, it's NIH syndrome combined with the lack of (well trained) software developers and it's officialism caused massive delays in rolling out a sucessor platform/product.
Luckily they finally started this year with the roll-out on all agencies PCs - ie. manual install of a new OS and apps, and we're talking about 8000 machines... that process should be finished by Q4.
CIGS of course doesn't address other problems with solar adoption, such as durability over time, public acceptance and investment, etc.
According to the first article linked in TFA:
"CIGS also doesn't degrade in sunlight like other thin-film technologies."
Actually I own a rollable CIGS solar panel (not longer produced, and not to be confused with this one that uses amorphous silicon). Whenever those panels pop up on eBay they create a ot of interest and are sold at a high price.
I doubt there will be a problem with public acceptance, if the price gets to the same level or lower level as current solar panels. The primary reason why solar technology is not so successful in being adopted is the high initial investment. Of course we all would also like to see a increase in efficiency, but in all fairness... even with 100% efficient energy conversion we'll get at most 350W/sqm
The sources I found claim that CIGS cells last 40+ years. Another advantage over mono-/polycrystalline silicon is that CIGS can also operate quite well under cloudy conditions. AFAIK such solar panels have been in use by NASA for many years already, one major reason is their stability and longetivity, but NASA now prefers even more efficient, specialised solar cell technology.
IIRC compared to other thin-film solar panel technologies CIGS is the one with the highest efficiency (at least it used to be), but it is still far short what a ordinary (non thin-film/flexible) mono- and polycrystalline silicon cells can archive.
If that is the case then I will welcome anyone breaking IPv6 with a new NAT solution, for this reason. I don't pay for installing an extra phone in my house, so I don't expect to have to do the same with IPv6.
That is not the idea. Instead you will get a whole subnet assigned to you. For example I have a IPv6 subnet with a 48-bit prefix, i.e. the first 48-bits are fixed by my ISP, but I can use the rest (128 - 48 = 80bit) to assign to any devices I have. Thats 2^80 IPv6 addresses just for me, and that didn't cost me anything (extra)! When IPv6 will become more popular most people will probably 'only' get a 64bit subnet though, which is still more than enough for home use I guess...
BTW, forgot to mention that I actually know one guy that was involved in the MP3 development (Jürgen Koller). Haven't met him for several years however, but he seems to still work at that place... here you can find a picture of him and an address where to send the beer;)
Drinking age is starting at 16 in Germany, except for 'hard drinks' like spirits (ie. distilled stuff in general). BTW, smoking is also allowed at 16 years, and you can go to pubs (NOT nightclubs), but you have to leave at midnight.
No limitations once you've become legally grown up (18 years), then you can also get your car driving license. Funny thing though is that married teenagers are legally treated as 'grown ups' in many cases.
The other thing that bothered me, has to do with CD Writing. It seems that in kernel developer land, more people have SCSI CD Writers, and IDE CD Writers are unusual. But in the real world, IDE CD writers are far more common, and SCSI CD Writers are prohibitively expensive.
But the whole CD Writing support treats IDE CD Writers like some kind of odd hardware from the top to the bottom,
because they are;)
Modern IDE ATAPI CD/DVD writers use the 'SCSI-3 mmc' command set. So basically they are SCSI Writers in IDE disguise. But internally they are not full SCSI devices, as they only support a small subset of all SCSI commands.
As a kernel hacker, what would you do?
1): write a SCSI Emulation layer that makes those IDE ATAPI devices act as full, real SCSI devices (from the software point of view). No redundant code, easier to maintain if done correctly.
2): write a specific IDE ATAPI driver and one specific driver for SCSI devices. Some redundant code, but more 'optimised' for each type of device
Linux has both.
1) was not maintained for quite a while, and obviously was badly written at first, thus nobody wanted to clean it up and it was maked 'deprecated'. But it still offers more features than 2).
OTOH 2) has DMA support which makes it attractive for many people, but is quite new and some app developers are afraid to support it, since the API might still change.
and the developer of the cdrecord client has been totally unprofessional about the whole thing.
Nobody has to love Joerg Schilling, and I often don't like his attitude either.
However everybody is entitled to have his/her opinion, and in his case he thinks that everything except the SCSI way sucks (he loves Solaris and FreeBSD because they have good SCSI support). So naturally he despises 2), whereas the rest of the linux developer try to make it the default.
He produces the best CD/DVD writing application for linux, and he knows more that most other people when it comes to CD/DVD writing. And he has a point when he says that you should still use the ide-scsi driver whenever possible, because ide-cd driver is NOT complete and still lacks important features. The problem is that since he is quite a SCSI fanatic many people are not taking him seriously.
Is Godwin's law suspended for this discussion, BTW (given the topic)"?
Yes.
Godwin's law only applies when somebody wants to win an argument by referring to Hitler/Holocaust in a discussion that is in no way related to that topic.
BTW, Godwin's law is also not applicable when somebody tries to enforce it in order to end a discussion.
No, it does refer to the hardware in general. I own several Suns, HPs and administered SGIs and Alphas.
The HP 9000 series are built like tanks. Very heavy, CPUs bigger than your CD drive. Nothing (at least no conventional weapons) will destroy them.
SGI and Alpha often come in surprisingly cheap cases, but their interior is more reliable than the more robust looking Suns (the RAM and the HDDs being their weak points).
Well, IIRC the old NeXTSTEP (3.3 and below) were using Objective-C in the kernel, at least in some drivers. I don't think this is still the case with MacOS X.
I fully agree that Objective-C is not a viable option for archiving the highest possible performance (and that is what many users what in a kernel, besides stability). However NeXTSTEP/OPENSTEP/MacOS X is based on Mach (which already does some sort of message passing), and the Objective-C message passing was/is highly optimized and integrated into the Mach subsystem, so that the performance impact is rather low...
On the subject of partisan sniping, I particularly like Bush's new ads, the one's with all the wolves circling the camera, implying that the terrorists want Kerry to win.
am I the only one who couldnt spot the difference in quality between the different cables?
Yes, it is difficult to tell, but if you really look closely, you'll see a difference in the quality of the pictures. A good way to find out it to open each picture (1, 2, 3) in a browser-tab (not in a new window); then flip between the tabs and you'll notice the difference.
In the coax picture you'll notice the 'color bleed' and distortions: on the face (cheek, mouth) of the referee, and on his left arm (especially when compared to the composite picture). Finally the s-video screenshot is slighly sharper and has more contrast (ie. not as blurred) as the other pictures, as you can see if you look at the audience on the right side
Slashdot Mirror
Your company just sounds like the company I'm working at!
On the one hand I fully agree with you. Don't complain if your computer is not working with the latest software/websites/hardware if you don't bother to keep it up-to-date.
... that process should be finished by Q4.
On the other hand sometimes you don't have any other choice, as you might not be in the position to change the OS.
I am a working for an insurance company (one of the BIG ones), and all PCs in our agencies in Germany are still running Windows 2000.
The company's failure to have a proper upgrade strategy in place, it's NIH syndrome combined with the lack of (well trained) software developers and it's officialism caused massive delays in rolling out a sucessor platform/product.
Luckily they finally started this year with the roll-out on all agencies PCs - ie. manual install of a new OS and apps, and we're talking about 8000 machines
Oh BTW, they're updating to Windows XP.
What are you talking about? Those tags are there already! They are named "Insightful" and "Interesting" respectively ...
According to the first article linked in TFA:
Actually I own a rollable CIGS solar panel (not longer produced, and not to be confused with this one that uses amorphous silicon). Whenever those panels pop up on eBay they create a ot of interest and are sold at a high price.
I doubt there will be a problem with public acceptance, if the price gets to the same level or lower level as current solar panels. The primary reason why solar technology is not so successful in being adopted is the high initial investment. Of course we all would also like to see a increase in efficiency, but in all fairness
The sources I found claim that CIGS cells last 40+ years. Another advantage over mono-/polycrystalline silicon is that CIGS can also operate quite well under cloudy conditions. AFAIK such solar panels have been in use by NASA for many years already, one major reason is their stability and longetivity, but NASA now prefers even more efficient, specialised solar cell technology.
IIRC compared to other thin-film solar panel technologies CIGS is the one with the highest efficiency (at least it used to be), but it is still far short what a ordinary (non thin-film/flexible) mono- and polycrystalline silicon cells can archive.
... are they Norwegian Blue?
Me and china. What a combination!
Put your toaster on fec0::/10 and it won't be routable. There you go: secure.
Site-Local scoped addresses (FEC0::/10) have been deprecated as of September 2004 (see RFC3879).
If that is the case then I will welcome anyone breaking IPv6 with a new NAT solution, for this reason. I don't pay for installing an extra phone in my house, so I don't expect to have to do the same with IPv6.
...
That is not the idea. Instead you will get a whole subnet assigned to you. For example I have a IPv6 subnet with a 48-bit prefix, i.e. the first 48-bits are fixed by my ISP, but I can use the rest (128 - 48 = 80bit) to assign to any devices I have. Thats 2^80 IPv6 addresses just for me, and that didn't cost me anything (extra)! When IPv6 will become more popular most people will probably 'only' get a 64bit subnet though, which is still more than enough for home use I guess
BTW, forgot to mention that I actually know one guy that was involved in the MP3 development (Jürgen Koller). Haven't met him for several years however, but he seems to still work at that place ... here you can find a picture of him and an address where to send the beer ;)
Drinking age is starting at 16 in Germany, except for 'hard drinks' like spirits (ie. distilled stuff in general). BTW, smoking is also allowed at 16 years, and you can go to pubs (NOT nightclubs), but you have to leave at midnight.
No limitations once you've become legally grown up (18 years), then you can also get your car driving license. Funny thing though is that married teenagers are legally treated as 'grown ups' in many cases.
Haven't you also noticed that it's mainly Taco that does the dupes? Sure, there are others but Taco makes most of them.
...
... kinda similar to some politicians I know.
That's because he has a split personality, and every single one of his multiple personalities don't know what the others are doing
Yeah, and when a snail packet is too big you can fragment it with a knife ;)
The deepest drill hole on land goes down some 12 km.
Yes, also see my other comment.
TFA claims it is the deapest hole drilled into the seafloor and that they had drilled into the lower section of Earth's crust for the first time.
But those guys have drilled much deeper holes before on land.
The most gogeous [sic] sales staffer I know also has an IQ that raises the average of any room she's ever walked into ...
Fahrenheit or Celsius?
Why bother? Hurd should be finished anytime soon!
By your definition a Bradley MA2A tank is 'portable'.
The other thing that bothered me, has to do with CD Writing. It seems that in kernel developer land, more people have SCSI CD Writers, and IDE CD Writers are unusual. But in the real world, IDE CD writers are far more common, and SCSI CD Writers are prohibitively expensive.
;)
But the whole CD Writing support treats IDE CD Writers like some kind of odd hardware from the top to the bottom,
because they are
Modern IDE ATAPI CD/DVD writers use the 'SCSI-3 mmc' command set. So basically they are SCSI Writers in IDE disguise. But internally they are not full SCSI devices, as they only support a small subset of all SCSI commands.
As a kernel hacker, what would you do?
1): write a SCSI Emulation layer that makes those IDE ATAPI devices act as full, real SCSI devices (from the software point of view). No redundant code, easier to maintain if done correctly.
2): write a specific IDE ATAPI driver and one specific driver for SCSI devices. Some redundant code, but more 'optimised' for each type of device
Linux has both.
1) was not maintained for quite a while, and obviously was badly written at first, thus nobody wanted to clean it up and it was maked 'deprecated'. But it still offers more features than 2).
OTOH 2) has DMA support which makes it attractive for many people, but is quite new and some app developers are afraid to support it, since the API might still change.
and the developer of the cdrecord client has been totally unprofessional about the whole thing.
Nobody has to love Joerg Schilling, and I often don't like his attitude either. However everybody is entitled to have his/her opinion, and in his case he thinks that everything except the SCSI way sucks (he loves Solaris and FreeBSD because they have good SCSI support). So naturally he despises 2), whereas the rest of the linux developer try to make it the default.
He produces the best CD/DVD writing application for linux, and he knows more that most other people when it comes to CD/DVD writing. And he has a point when he says that you should still use the ide-scsi driver whenever possible, because ide-cd driver is NOT complete and still lacks important features. The problem is that since he is quite a SCSI fanatic many people are not taking him seriously.
Is Godwin's law suspended for this discussion, BTW (given the topic)"?
Yes.
Godwin's law only applies when somebody wants to win an argument by referring to Hitler/Holocaust in a discussion that is in no way related to that topic. BTW, Godwin's law is also not applicable when somebody tries to enforce it in order to end a discussion.
Don't complain.
I live in Ireland and for EUR 47/month (about USD $60) I get
- 512/128 kbps
- 16GB cap (either up or down)
- dynamic IP, daily disconnect
No, it does refer to the hardware in general. I own several Suns, HPs and administered SGIs and Alphas.
The HP 9000 series are built like tanks. Very heavy, CPUs bigger than your CD drive. Nothing (at least no conventional weapons) will destroy them.
SGI and Alpha often come in surprisingly cheap cases, but their interior is more reliable than the more robust looking Suns (the RAM and the HDDs being their weak points).
Well, IIRC the old NeXTSTEP (3.3 and below) were using Objective-C in the kernel, at least in some drivers. I don't think this is still the case with MacOS X.
...
I fully agree that Objective-C is not a viable option for archiving the highest possible performance (and that is what many users what in a kernel, besides stability).
However NeXTSTEP/OPENSTEP/MacOS X is based on Mach (which already does some sort of message passing), and the Objective-C message passing was/is highly optimized and integrated into the Mach subsystem, so that the performance impact is rather low
On the subject of partisan sniping, I particularly like Bush's new ads, the one's with all the wolves circling the camera, implying that the terrorists want Kerry to win.
Funny thing, as (some) terrorists actually support the reelection of Dubya
am I the only one who couldnt spot the difference in quality between the different cables?
Yes, it is difficult to tell, but if you really look closely, you'll see a difference in the quality of the pictures. A good way to find out it to open each picture (1, 2, 3) in a browser-tab (not in a new window); then flip between the tabs and you'll notice the difference.
In the coax picture you'll notice the 'color bleed' and distortions: on the face (cheek, mouth) of the referee, and on his left arm (especially when compared to the composite picture). Finally the s-video screenshot is slighly sharper and has more contrast (ie. not as blurred) as the other pictures, as you can see if you look at the audience on the right side
Perhaps this is what you are looking for