You hit something with something they have the skills to do, haven't yet done, and watch how their mind works. Once the recruiters know what kind of interview you give, you get the right kinds of candidates.
That doesn't sound impossible to find I've worked with those people. But security clearance cuts your ability to pick anything else rather dramatically. Generally you are better off grooming security clearances.
The original article listed the 3 reasons the slots were hard to fill, "including lack of available applicants, applicants looking for more pay and lack of experience"
So in other words employers who don't recruit, don't pay much and aren't willing to train are having trouble. Well good.
I think Microsoft has already lost mindshare in the consumer space. Apple has set the bar to the point that Microsoft products are seen as worse but cheaper. In the consumer space MS is sitting around doing little. Microsoft OTOH is focusing more and more on the highly profitable enterprise market which Apple doesn't want.
It wouldn't shock me if 10 years from now, MS is a enterprise OS and doesn't have anything in consumer.
How many people who support the GPL enforcement support pirating commercial music? Most GPL advocates would like to see weaker copyright laws knowing this would equally to GPL software and music.
Linux is big and strong enough, that a modified version in a specific product is unlikely to have anything important Linux really needs.
Of course it could. Lots if not most things inside the Linux kernel itself come from a small group or even an individual. With other software products that are part of the Linux eco system this is even more common. For example twenty years ago I was using Scientific Word for Windows which allowed me to do calculations (like Maple) and TeX. That's far more advanced than something like LyX which is still very popular.
The creator of Busybox just got sick of helping the SFLC line its pockets by discouraging the use of free software. I think in the end he disowned Busybox and started a new project to do the same thing, under the BSD licence.
Just in case anyone is paying attention to the AC here... The creator of Busybox was Bruce Perens who went on to be the 2nd head of Debian where he authored the Free Software Guidelines. He worked with HP on their move into the Linux community and founded several more Linux projects. He currently works with the government of Norway in a Linux related role.
We don't know that. We haven't had problems like a serious war yet. Between the Napoleanic wars and WWI world trade was massive about 1/3rd of the economy. WWI shut those systems down and they never recovered. What happens if we were to lose connection with India for years? We've had a very stable world for a while now, periods of stability come and go.
Now besides the meta issues... I'd just comment that losing indigenous IT seems to have paired quite often with drops in productivity gains for US operations. So it might very well have been the case that there was a 1x surge in savings followed by a long term administrative problem. We need better paired case studies.
How office will survive, when there are cloud document systems that are much more reliable than traditional office on a desktop ever can be might be fair question
Office + Sharepoint is way ahead of any cloud solution. Office + Sharepoint + Dynamics + Universal Communicator +.... is very rich. That's how they survive. The problem is that maybe 10% of businesses use the whole Microsoft suite.
Its more than that. This used to be called Display Processors
Term used to refer to a well-known effect whereby function in a computing system family is migrated out to special-purpose peripheral hardware for speed, then the peripheral evolves toward more computing power as it does its job, then somebody notices that it is inefficient to support two asymmetrical processors in the architecture and folds the function back into the main CPU, at which point the cycle begins again
There had been several cycles of this already by the late 1960s.
Christ. Does this mean we will have to go though all of this again? Maybe by that time I'll be senile enough to not notice.
Yes about every fifteen years for the rest of your life. What you will really enjoy is in about 5 years people will be pitching thick clients with their tremendously devices specific features as an innovation. You are already seeing this on mobile where the move is away from generic and towards a specific app wrapping the functions of specific sites and once you have that offline modes and...
nor do I see the client OS as a factor in cloud computing (isn't that the whole point?).
Depends on the cloud model. There were some excellent cloud solutions (though they weren't called that) I used in the 1990s with Active X. Often the purpose of cloud models is to make distribution of software easier, you don't necessarily need client diversity. Though with
a) Many more macs b) Mobile c) Linux servers being in the majority
A windows only solution is unlikely to be as acceptable. On the other hand one where only certain clients worked right... I could easily picture that.
where ISP/carrier bandwidth caps are becoming prevalent?
Bandwidth is more or less increasing steadily. There have always been problems on the edges of various pricing models which moves home / small business models from "all you can eat" to "all you can eat with some limitations". If you legitimately need more bandwidth, you can buy as much bandwidth as you want.
Names are useful for the creative team who doesn't know which backend their creative materials (look and feel) are going with. When the creative starts, it might be Fedora 16.2, 17, 17.5, 18... Apple has proven that these names from creative can be used for branding and Debian... have followed suit.
It makes sense. You don't like cute names you aren't the target.
So what software do you think companies will use on a linux platform to read and modify Office documents? I bet that will be OpenOffice. There is your common application.
Maybe. I can think of some alternatives though. 1) Linux maintains its tradition of fragmentation and no office suite gets decent market share. Some are on Open Office, some on Gnome Live (AbiWord, Evolution, Gnumeric...), some on Calligra (was KOffice) others using Google docs. And then from there lots of diversity for example Kexi vs. Qercus for your Calligra database. That's not even counting the minor suite or the ports from other OSes that would happen as Linux became more mainstream for example there would be a Microsoft Office for Linux most likely and Apple's Pages/Numbers/Keynote suite. Office apps are rare in that they have a mono culture, created by Microsoft during the transition to windows. I suspect that an office mono culture never develops, we go back to the situation in the 1980s when there was genuine diversity of productivity applications.
2) Its worth noting a minor point that with diversity, some of the basic choices about productivity applications get questioned. Open Office is extremely conservative in its direction, moreso even than Microsoft Office. For example WYSIWYG (Windows) vs. WYMIWYG (Unix). With the rise of presentation platforms like HTML and not tying documents closely to particular printers it is doubtful WYSIWYG would have won. Associative databases might work better for casual database constructions. If those choices get reopened I don't see Open Office as the winner. Diversity becomes mandatory as applications want opposite feature sets.
3) Compound this even further by noting that Linux has nothing remotely like Sharepoint. What makes Office so good for corporate America is the sharing and collaboration features. Those haven't developed on the Linux eco system at all.
___
Here you just blow you own argument. If it is possible for a browser like Firefox (a common application) to download an execute a virus written in Perl or Python that virus will work on all flavours of Lunux. Commonality breeds venerability.
Diversity kills this already. Take the very first line of this virus:
#!/usr/bin/Perl or #!/usr/bin/perl or #!/usr/bin/perl6 or #!/bin/perl or #!/opt/local/bin/perl
etc... That's how diversity works. More importantly since Linux systems aren't designed to get software from a wide range of locations "download and run" can be hard. For example the system can require an explicit chmod before it will run a downloaded executable. It can run the executable with low permissions (i.e. as nobody) for safety. It can do all sorts of things from a security standpoint that are impossible on Microsoft because Microsoft has to allow end users to easily download and run stuff.
All of that does not matter if the desktop users do not update their OS or software.
Reread this I addressed it. This doesn't have to do with users, this has to do with developer coordination. You are still trying to think like the Linux security would be the same as Microsoft's with updates going to end users to address specific harms. Linux updates could be once a year. As for would they do it, both Microsoft and OSX have found nagging and easy one click to work well to get people to update. No reason not to follow their example.
My point is that one can not guarantee that the Linux found on desktops in the future will be as secure as the Linux today.
I agree with that. We have no idea what a Linux that was on 90% of desktops would look like. We also agree that Linux isn't secure because of great design.
but phrase the question slightly differently... If today's Linuxes were being used by 90% of all desktop users would they be far more secure than Windows, here I would answer yes.
Actually it does. cross platform viruses require a common application that is they come from a situation of lack of diversity, like open office viruses. As far as web scripts that is Linux as a conduit, which I wasn't addressing. A diverse eco system helps against catching the viruses not so much about carrying them.
LAMP is a server package and not a desktop package.
I understand that, though it had huge influence on Linux's development. In the same way WordStar had huge influence on the development of the PC platform. So for example all desktop users use applications that assume Perl/Python scripting is available. Lots of desktop Unix applications use MySQL either as an option or mandatory, the way PC apps use Microsoft's application databases. Finally quite a few desktop Linux users run some apps that use web configuration (apache pointed to 127.0.0.1). LAMP is still very much alive even on desktop.
Microsoft's pricing strategy creates a much more definitive line than would naturally exist between desktop and server. Were Linux dominant I suspect it would be like Unix culture where the line is much fuzzier.
This is only effective if the desktops are updated when a new patch comes out. Servers are almost always updated because there are people to do that. Getting a desktop user to update may not happen as often.
That's not what makes it effective. What is effective that a security layer or configuration can be added across the board. So for example when Microsoft wanted to move to divide user and system dll's they couldn't do anything drastic without breaking applications. Linux distributors can just change the configure step and recompile the applications.
All I am trying to point out is that the article poster is making a huge assumption when looking at today's numbers and stating that malware on Linux is not and never will be an issue.
I understand what you are pointing out. And I'm saying the diversity and the role of distributions makes it impossible assuming a popular desktop Linux that looks like today's Linuxes. Now that might be impossible and it might be that any popular Linux would look like Android and thus lose these advantages.
And the, " the production of mallware for Linux will approach the production for Windows " theory is what you are lacking evidence for. You provided some evidence that there was an increase for OSX. But there are many possibilities where the production increases but never gets very high. In the case of Linux, but not for OSX, the diversity of Linux might raises the cost of malware. Also the Linux distributions acting as distributors for 3rd party applications has already shown itself more abel to respond to security threats by coordinating directly between hundreds of applications for security increases. So it is entirely possible that malware never finds Linux suitable for the same reasons that commercial software has found Linux annoying to create for. Further Linux having moved to fully public servers very early (i.e. LAMP) has been focused on security in environments more hostile than the desktop for 2 decades. So no I don't think Linux is ever likely to have the same kinds of problems as Windows even if it had 90% of the desktop market share. Design choices have real consequences and the advantages Linux has over Windows with respect to viruses and worms are among those real consequences. So that case IMHO isn't even interesting.
Let me just point out that Android proves that the advantage is diversity and server orientation not something inherent to Linux's construction. Take away those two factors and you end up with Android which is looking like it will be as bad as the situation faced by Windows.
The case of OSX IMHO is more interesting. There you have slightly less diversity than Windows but you have a leader that has and wields the kind of centralized power that a Linux distribution does, and possibly even more. Could Apple's willingness to break legacy applications and their ability to move the developer community allow them to stay permanently ahead fo the virus and malware groups. I suspect the answer is yes.
It doesn't make any sense to believe that there will be less publishers because e-publishing lowers the bar to entry. Therefore, there will be more publishers, even if sales are lower, as more and more small players attempt to inhabit niches.
The POD revolution and now the ebook revolution don't lower the bar of entry for publishers they lower the bar of entry for authors.
publisher1 = major 6 players publisher2 = independent publishers looking for books that sell 20k+ publisher3 = self publishing companies looking mainly to sell services to authors
There is no question the publisher3 market is exploding. The publisher1 market is shrinking, though top books still exist. The publisher2 market is in danger of being wiped out.
No what I'm saying is that this project took Comcast 5 years... it cost them a lot and caused a major shift in focus. I didn't say it went badly it was expensive and complex which Comcast admits. Moreover, Comcast of all the major carriers has it the easiest because they don't have complex enterprise services.
The self publishing houses have editorial staffs. The author can purchase editorial services from many of them. Good editorial service, similar to what a publisher would do, start at about $5k for a book. The POD self service people were mainly interested in selling about $500-1500 in services, while offering higher end packages; and the new lower end ebook only are interested in selling even less. The opportunity is trying to get self published authors to buy more services. I see opportunity for a credit agency to partner with a POD house and extend credit.
Ultimately though.... the awful quality is part of what makes the POD market so disruptive. They can offer books that target sales volumes in the hundreds (and at the new price points dozens) and thus appeal to an incredibly narrow niche.
Publishers would be fine with making ebooks. That's not a dramatic shift.
The issue publisher's can't cope with is given low POD costs and now ebooks incredibly niche titles are eating up the market, including the ebook market, for mid range sales titles.
Slashdot Mirror
Problem solving is easy to test for.
You hit something with something they have the skills to do, haven't yet done, and watch how their mind works. Once the recruiters know what kind of interview you give, you get the right kinds of candidates.
That doesn't sound impossible to find I've worked with those people. But security clearance cuts your ability to pick anything else rather dramatically. Generally you are better off grooming security clearances.
The original article listed the 3 reasons the slots were hard to fill, "including lack of available applicants, applicants looking for more pay and lack of experience"
So in other words employers who don't recruit, don't pay much and aren't willing to train are having trouble. Well good.
I imagine they could do the same thing. But it is easier to have a name. Particularly if they are working on multiple projects. Then that name sticks.
I think Microsoft has already lost mindshare in the consumer space. Apple has set the bar to the point that Microsoft products are seen as worse but cheaper. In the consumer space MS is sitting around doing little. Microsoft OTOH is focusing more and more on the highly profitable enterprise market which Apple doesn't want.
It wouldn't shock me if 10 years from now, MS is a enterprise OS and doesn't have anything in consumer.
He was talking about manufacturers. To get what you want requires major scientific innovation.
How many people who support the GPL enforcement support pirating commercial music? Most GPL advocates would like to see weaker copyright laws knowing this would equally to GPL software and music.
Linux is big and strong enough, that a modified version in a specific product is unlikely to have anything important Linux really needs.
Of course it could. Lots if not most things inside the Linux kernel itself come from a small group or even an individual. With other software products that are part of the Linux eco system this is even more common. For example twenty years ago I was using Scientific Word for Windows which allowed me to do calculations (like Maple) and TeX. That's far more advanced than something like LyX which is still very popular.
The creator of Busybox just got sick of helping the SFLC line its pockets by discouraging the use of free software. I think in the end he disowned Busybox and started a new project to do the same thing, under the BSD licence.
Just in case anyone is paying attention to the AC here... The creator of Busybox was Bruce Perens who went on to be the 2nd head of Debian where he authored the Free Software Guidelines. He worked with HP on their move into the Linux community and founded several more Linux projects. He currently works with the government of Norway in a Linux related role.
There are phones with excellent battery life:
1 Motorola Droid Razr Maxx 19.78
2 Apple iPhone 4 (with 3G off) 14.55
3 Apple iPhone 3GS (with 3G off) 13.4
4 HTC Legend 12.75
5 RIM BlackBerry Curve 9360 12
That doom and gloom didn't come true either.
We don't know that. We haven't had problems like a serious war yet. Between the Napoleanic wars and WWI world trade was massive about 1/3rd of the economy. WWI shut those systems down and they never recovered. What happens if we were to lose connection with India for years? We've had a very stable world for a while now, periods of stability come and go.
Now besides the meta issues... I'd just comment that losing indigenous IT seems to have paired quite often with drops in productivity gains for US operations. So it might very well have been the case that there was a 1x surge in savings followed by a long term administrative problem. We need better paired case studies.
How office will survive, when there are cloud document systems that are much more reliable than traditional office on a desktop ever can be might be fair question
Office + Sharepoint is way ahead of any cloud solution. Office + Sharepoint + Dynamics + Universal Communicator + .... is very rich. That's how they survive. The problem is that maybe 10% of businesses use the whole Microsoft suite.
Its more than that. This used to be called Display Processors
Term used to refer to a well-known effect whereby function in a computing system family is migrated out to special-purpose peripheral hardware for speed, then the peripheral evolves toward more computing power as it does its job, then somebody notices that it is inefficient to support two asymmetrical processors in the architecture and folds the function back into the main CPU, at which point the cycle begins again
There had been several cycles of this already by the late 1960s.
Christ. Does this mean we will have to go though all of this again? Maybe by that time I'll be senile enough to not notice.
Yes about every fifteen years for the rest of your life. What you will really enjoy is in about 5 years people will be pitching thick clients with their tremendously devices specific features as an innovation. You are already seeing this on mobile where the move is away from generic and towards a specific app wrapping the functions of specific sites and once you have that offline modes and...
nor do I see the client OS as a factor in cloud computing (isn't that the whole point?).
Depends on the cloud model. There were some excellent cloud solutions (though they weren't called that) I used in the 1990s with Active X. Often the purpose of cloud models is to make distribution of software easier, you don't necessarily need client diversity. Though with
a) Many more macs
b) Mobile
c) Linux servers being in the majority
A windows only solution is unlikely to be as acceptable. On the other hand one where only certain clients worked right... I could easily picture that.
where ISP/carrier bandwidth caps are becoming prevalent?
Bandwidth is more or less increasing steadily. There have always been problems on the edges of various pricing models which moves home / small business models from "all you can eat" to "all you can eat with some limitations". If you legitimately need more bandwidth, you can buy as much bandwidth as you want.
Names are useful for the creative team who doesn't know which backend their creative materials (look and feel) are going with. When the creative starts, it might be Fedora 16.2, 17, 17.5, 18... Apple has proven that these names from creative can be used for branding and Debian... have followed suit.
It makes sense. You don't like cute names you aren't the target.
So what software do you think companies will use on a linux platform to read and modify Office documents? I bet that will be OpenOffice. There is your common application.
Maybe. I can think of some alternatives though.
1) Linux maintains its tradition of fragmentation and no office suite gets decent market share. Some are on Open Office, some on Gnome Live (AbiWord, Evolution, Gnumeric...), some on Calligra (was KOffice) others using Google docs. And then from there lots of diversity for example Kexi vs. Qercus for your Calligra database. That's not even counting the minor suite or the ports from other OSes that would happen as Linux became more mainstream for example there would be a Microsoft Office for Linux most likely and Apple's Pages/Numbers/Keynote suite. Office apps are rare in that they have a mono culture, created by Microsoft during the transition to windows. I suspect that an office mono culture never develops, we go back to the situation in the 1980s when there was genuine diversity of productivity applications.
2) Its worth noting a minor point that with diversity, some of the basic choices about productivity applications get questioned. Open Office is extremely conservative in its direction, moreso even than Microsoft Office. For example WYSIWYG (Windows) vs. WYMIWYG (Unix). With the rise of presentation platforms like HTML and not tying documents closely to particular printers it is doubtful WYSIWYG would have won. Associative databases might work better for casual database constructions. If those choices get reopened I don't see Open Office as the winner. Diversity becomes mandatory as applications want opposite feature sets.
3) Compound this even further by noting that Linux has nothing remotely like Sharepoint. What makes Office so good for corporate America is the sharing and collaboration features. Those haven't developed on the Linux eco system at all.
___
Here you just blow you own argument. If it is possible for a browser like Firefox (a common application) to download an execute a virus written in Perl or Python that virus will work on all flavours of Lunux. Commonality breeds venerability.
Diversity kills this already. Take the very first line of this virus:
#!/usr/bin/Perl or
#!/usr/bin/perl or
#!/usr/bin/perl6 or
#!/bin/perl or
#!/opt/local/bin/perl
etc... That's how diversity works. More importantly since Linux systems aren't designed to get software from a wide range of locations "download and run" can be hard. For example the system can require an explicit chmod before it will run a downloaded executable. It can run the executable with low permissions (i.e. as nobody) for safety. It can do all sorts of things from a security standpoint that are impossible on Microsoft because Microsoft has to allow end users to easily download and run stuff.
All of that does not matter if the desktop users do not update their OS or software.
Reread this I addressed it. This doesn't have to do with users, this has to do with developer coordination. You are still trying to think like the Linux security would be the same as Microsoft's with updates going to end users to address specific harms. Linux updates could be once a year. As for would they do it, both Microsoft and OSX have found nagging and easy one click to work well to get people to update. No reason not to follow their example.
My point is that one can not guarantee that the Linux found on desktops in the future will be as secure as the Linux today.
I agree with that. We have no idea what a Linux that was on 90% of desktops would look like. We also agree that Linux isn't secure because of great design.
but phrase the question slightly differently... If today's Linuxes were being used by 90% of all desktop users would they be far more secure than Windows, here I would answer yes.
This has no effect on cross platform viruses or web scripts. http://en.wikipedia.org/wiki/Linux_malware [wikipedia.org]
Actually it does. cross platform viruses require a common application that is they come from a situation of lack of diversity, like open office viruses. As far as web scripts that is Linux as a conduit, which I wasn't addressing. A diverse eco system helps against catching the viruses not so much about carrying them.
LAMP is a server package and not a desktop package.
I understand that, though it had huge influence on Linux's development. In the same way WordStar had huge influence on the development of the PC platform. So for example all desktop users use applications that assume Perl/Python scripting is available. Lots of desktop Unix applications use MySQL either as an option or mandatory, the way PC apps use Microsoft's application databases. Finally quite a few desktop Linux users run some apps that use web configuration (apache pointed to 127.0.0.1). LAMP is still very much alive even on desktop.
Microsoft's pricing strategy creates a much more definitive line than would naturally exist between desktop and server. Were Linux dominant I suspect it would be like Unix culture where the line is much fuzzier.
This is only effective if the desktops are updated when a new patch comes out. Servers are almost always updated because there are people to do that. Getting a desktop user to update may not happen as often.
That's not what makes it effective. What is effective that a security layer or configuration can be added across the board. So for example when Microsoft wanted to move to divide user and system dll's they couldn't do anything drastic without breaking applications. Linux distributors can just change the configure step and recompile the applications.
All I am trying to point out is that the article poster is making a huge assumption when looking at today's numbers and stating that malware on Linux is not and never will be an issue.
I understand what you are pointing out. And I'm saying the diversity and the role of distributions makes it impossible assuming a popular desktop Linux that looks like today's Linuxes. Now that might be impossible and it might be that any popular Linux would look like Android and thus lose these advantages.
And the, " the production of mallware for Linux will approach the production for Windows " theory is what you are lacking evidence for. You provided some evidence that there was an increase for OSX. But there are many possibilities where the production increases but never gets very high. In the case of Linux, but not for OSX, the diversity of Linux might raises the cost of malware. Also the Linux distributions acting as distributors for 3rd party applications has already shown itself more abel to respond to security threats by coordinating directly between hundreds of applications for security increases. So it is entirely possible that malware never finds Linux suitable for the same reasons that commercial software has found Linux annoying to create for. Further Linux having moved to fully public servers very early (i.e. LAMP) has been focused on security in environments more hostile than the desktop for 2 decades. So no I don't think Linux is ever likely to have the same kinds of problems as Windows even if it had 90% of the desktop market share. Design choices have real consequences and the advantages Linux has over Windows with respect to viruses and worms are among those real consequences. So that case IMHO isn't even interesting.
Let me just point out that Android proves that the advantage is diversity and server orientation not something inherent to Linux's construction. Take away those two factors and you end up with Android which is looking like it will be as bad as the situation faced by Windows.
The case of OSX IMHO is more interesting. There you have slightly less diversity than Windows but you have a leader that has and wields the kind of centralized power that a Linux distribution does, and possibly even more. Could Apple's willingness to break legacy applications and their ability to move the developer community allow them to stay permanently ahead fo the virus and malware groups. I suspect the answer is yes.
It doesn't make any sense to believe that there will be less publishers because e-publishing lowers the bar to entry. Therefore, there will be more publishers, even if sales are lower, as more and more small players attempt to inhabit niches.
The POD revolution and now the ebook revolution don't lower the bar of entry for publishers they lower the bar of entry for authors.
publisher1 = major 6 players
publisher2 = independent publishers looking for books that sell 20k+
publisher3 = self publishing companies looking mainly to sell services to authors
There is no question the publisher3 market is exploding. The publisher1 market is shrinking, though top books still exist. The publisher2 market is in danger of being wiped out.
No what I'm saying is that this project took Comcast 5 years... it cost them a lot and caused a major shift in focus. I didn't say it went badly it was expensive and complex which Comcast admits. Moreover, Comcast of all the major carriers has it the easiest because they don't have complex enterprise services.
The self publishing houses have editorial staffs. The author can purchase editorial services from many of them. Good editorial service, similar to what a publisher would do, start at about $5k for a book. The POD self service people were mainly interested in selling about $500-1500 in services, while offering higher end packages; and the new lower end ebook only are interested in selling even less. The opportunity is trying to get self published authors to buy more services. I see opportunity for a credit agency to partner with a POD house and extend credit.
Ultimately though.... the awful quality is part of what makes the POD market so disruptive. They can offer books that target sales volumes in the hundreds (and at the new price points dozens) and thus appeal to an incredibly narrow niche.
That was in the context of introducing new APIs. That was not in the context of dropping support.
Publishers would be fine with making ebooks. That's not a dramatic shift.
The issue publisher's can't cope with is given low POD costs and now ebooks incredibly niche titles are eating up the market, including the ebook market, for mid range sales titles.