Software costs money to make, it basically after a certain amount of time (depending on it's complexity) free software that mimics your software will show up
you mean Linux is going to mimic Vista? That really will mean Linux is not ready for the desktop!
actually, in Soviet Russia, the military pwns the NSA.
(from an article in yesterdays/. about RedHat CEO: "Earlier in the year I was in Russia and RHEL is the most secure operating system certified by the Russian military, therefore there are applications for the Russian military and government that can only run on RHEL. The ironic thing about that is the reason it is so secure is because SE Linux, the core security technology for Linux, was written by the NSA in the US." )
First the military, then the education system, next government... then business?
Windows has switched to CIFS (which I believe is SMBv2) and is a completely different beast, well, a bit different. Its faster for one - I've even had better performance from it than NFS.
So, yes, Microsoft has moved away from SMB already, but it'll be around for quite a while (one of my customers still runs NT4), and Netbios is still used in some places.
(I find it interesting though, in the Linux world, as new protocols become available, the old crufty ones die off quite quickly as the components implementing them stop being maintained and no-one bothers to install them. In the Windows world, such things linger forever.)
I suppose its all about realisation that Linux is making it into corporate environments, and Microsoft now has to do something to keep themselves there.
MS are saying that if you really, really must have a Linux server in your IT shop, they'd better make it so it can connect to the one true corporate user account register, before the people who put the Linux server in decide to try a different LDAP server, maybe even one supplied by Novell.
It makes sense for MS to start doing this, in this way they can keep their dominance in the corporate IT structure, by letting the lowly Linux boxes play in the same playground. The important thing to understand here is that even MS has realised linux is making it big in businesses, that kinda give Linux the seal of approval from MS, not even the most pro-MS, anti-OSS PHB can say its not a valid OS anymore.
Next: an Outlook client... MS won't mind that as it allows them to keep their Exchange systems... until someone builds an Exchange replacement to go with it, and then watch MS share price tumble.
I guess second that. I may be able to figure out Git command-line commands, but the sales guy who checks his documents into our SCM sure won't.
Any serious SCM needs to have a super simple, intuitive, easy to use GUI. And IDE integration. And shell integration. And web interface. And webdav access.
I'm not so sure its a bad thing. Most people just follow the herd, so this guy is shouting "Linux, dude! Its going to be bigger than you thought", hopefully many people who are going to be laid off anyway will listen, retrain to be FOSS-capable and then get re-hired in the new world order.
Ok, the new world order bit might be a little over-optimistic, but the message to re-train is serious.
We have a server room full of old servers that we arranged for a customer, after 3 years, said customer wanted new ones as the lease had run out, so we offered to buy them from the leasing company. They happily quoted us not that much, and we ended up with a room full of stuff.
We also received some networking kit that no-one knew how to use (turned out to be fancy cisco routers) that we sold to a networking company... for the price we paid for the entire lot of kit. Our only problem is that we will have to pay to get rid of the servers in 3 years time when we do it again.
I'm not sure where the best place to find leasing companies is, but a good place to start might be those ebay sellers that only sell lots of BIN kit. Yellow pages after that.
yep, RedHat licences cost more than Windows, but (apart from the other facts already mentioned) you only need to buy 1 to be fully licenced for as many desktops and servers as you like;););)
You can happily run CentOS as much as you like whilst paying RedHat for the 'warm fuzzy' feeling of support that, chance are, you'll never ever use.
Unfortunately not, though I'll see if the link is in my work browser history.. but I did find this which does cover some of the topics mentioned. (the link I referred to was a webcast/ video powerpoint so you can imagine how easy it is to find the damn thing using google).
This link seems to cover some of the same things that I recall- especially the issues of re-hashing strong named assemblies.
There are people who will use 1 browser because it is "the internet". there are people who will use 1 browser because it is God's only browser and there is only one. There are people who will use 1 browser because they cant be bothered to change.
And then there are people who will want to latest, fastest, feature-rich, talked-about browser. And if FF gets "superfast" stickers all over it in the popular press and blogs, people will want to use it. Nobody really wants to be stuck with yesterday's slow old slowness, not in the Internet Age. We've all been conditioned to always go for the upgrade, give us any reason and very many of us will.
I'm not sure about this as its a relatively obscure part of OS/app interaction, but I thought only the parts needed to be paged in were loaded when the shared library was loaded - the right parts are known at compile time as the linker does its thing.
For dynamically loaded libraries, the OS doesn't 'load' the dll, it maps the dll into the app's address space, (you might like to check out rebasing to see what happens when the dll's conflicts with address range already used). Once mapped, only the parts that are needed are loaded in the usual manner (when there is a page fault) as the app uses part of the dll.
However, take the above with a pinch of 'look it up for yourself'.
I do know that.NET apps load all their library code in, especially those that are secured with strong names - the loader has to load it all in order to verify that no-one has tampered with it, and it has to load all dependant libraries and they have... so sometimes your.NET load performance really sucks. Apparently they're going to do something about it (not sure what) but the WPF guys have said that if you want your WPF apps to perform, do not use strongly named assemblies.
Why chain together tools yourself, and manually think about things you really shouldn't be thinking about, when you can have a good filesystem take care of it for you.
Because that's the Unix way - build small components (applications) and chain them together to create something out of the parts. I mean, why have ls and grep when you can have lsgrepsortfind? Really, the point is to have small, easily maintained apps that do 1 thing well than 1 app that does everything possibly well, but more usually poorly as its difficult to maintain and ensure it works properly. Not to mention the bloat when it replicates functionality already provided.
This may not be the best model for a critical component like a filesystem, but on the other hand, reliability of a filesystem is paramount, so keeping it as small as possible is probably a good idea.
You can make an bike out of ordinary paper. I have a pen made of tightly-wrapped paper, its indestructible (well, I'm sure it isn't, but its very strong stuff). A paper bike is certainly feasible, if the paper core of the bike was suitably 'carbon wrapped'.
you can see it... but once you have you'll wish you hadn't:-)
google for bootlog.txt, its a log file of everything Windows does when booting if you enable it. Googling to remind myself of the instructions to enable it tells me that Process Monitor now has an 'enable boot logging' option that's all nice and gui and easy to use.
4) The app store, sdk, etc are built for java. If you have a phone that isn't locked down, you could probably put native code on it, with some work.
5) Also, they use a custom jdk (and libraries), so the java apps have to be built specifically for android
I see a fair bit of complaints about this, a lot of developers who have existing symbian and other c++ based code want to re-use that for Android. Given that the underlying libraries are C++ anyway, it makes sense for Google to release more native-friendly development kits. Encourage them to go native, they will support it in the future if everyone asks. (see this blog)
no, instead of being allocated an IP address, they will give you a block (a/64 IIRC) which will give you enough IPs to address every electronic item you'll ever own.
Its more likely they'll set it up so your devices auto-configure using your ISPs block.
because, to continue your oil analogy, no-one will change until the cost of this increasingly-scare resource rises to a point where its cheaper to migrate to IPv6.
So - when your ISP says, you can have 1 shared IP, but if you want a static one it'll cost you $10 a month.. you'll start thinking about how you can get IPv6 and moving to a competitor who provides it, assuming you can get a home router that supports it... and then other ISPs will see the migration and start offering IPv6 as features, and then we'll be there.
It'll just take those market forces, so if you want to help things along, grab as many static IPs as you can.
ok, I chose poor a poor term... but you're exactly right.
At the moment, I see a lot of devs writing.NET code, and wanting to slap a WPF GUI on them, Mono doesn't support this, and apparently won't for ages. So immediately you see Mono being cross platform only for a subset of applications (eg server). I think MS will be happy with that.
Slashdot Mirror
Software costs money to make, it basically after a certain amount of time (depending on it's complexity) free software that mimics your software will show up
you mean Linux is going to mimic Vista? That really will mean Linux is not ready for the desktop!
actually, in Soviet Russia, the military pwns the NSA.
(from an article in yesterdays /. about RedHat CEO: "Earlier in the year I was in Russia and RHEL is the most secure operating system certified by the Russian military, therefore there are applications for the Russian military and government that can only run on RHEL. The ironic thing about that is the reason it is so secure is because SE Linux, the core security technology for Linux, was written by the NSA in the US." )
First the military, then the education system, next government ... then business?
Windows has switched to CIFS (which I believe is SMBv2) and is a completely different beast, well, a bit different. Its faster for one - I've even had better performance from it than NFS.
So, yes, Microsoft has moved away from SMB already, but it'll be around for quite a while (one of my customers still runs NT4), and Netbios is still used in some places.
(I find it interesting though, in the Linux world, as new protocols become available, the old crufty ones die off quite quickly as the components implementing them stop being maintained and no-one bothers to install them. In the Windows world, such things linger forever.)
I suppose its all about realisation that Linux is making it into corporate environments, and Microsoft now has to do something to keep themselves there.
MS are saying that if you really, really must have a Linux server in your IT shop, they'd better make it so it can connect to the one true corporate user account register, before the people who put the Linux server in decide to try a different LDAP server, maybe even one supplied by Novell.
It makes sense for MS to start doing this, in this way they can keep their dominance in the corporate IT structure, by letting the lowly Linux boxes play in the same playground. The important thing to understand here is that even MS has realised linux is making it big in businesses, that kinda give Linux the seal of approval from MS, not even the most pro-MS, anti-OSS PHB can say its not a valid OS anymore.
Next: an Outlook client... MS won't mind that as it allows them to keep their Exchange systems ... until someone builds an Exchange replacement to go with it, and then watch MS share price tumble.
must be C#, seems everything else is being rewritten in that nowadays :(
I guess second that. I may be able to figure out Git command-line commands, but the sales guy who checks his documents into our SCM sure won't.
Any serious SCM needs to have a super simple, intuitive, easy to use GUI. And IDE integration. And shell integration. And web interface. And webdav access.
I'm not so sure its a bad thing. Most people just follow the herd, so this guy is shouting "Linux, dude! Its going to be bigger than you thought", hopefully many people who are going to be laid off anyway will listen, retrain to be FOSS-capable and then get re-hired in the new world order.
Ok, the new world order bit might be a little over-optimistic, but the message to re-train is serious.
leasing companies of course.
We have a server room full of old servers that we arranged for a customer, after 3 years, said customer wanted new ones as the lease had run out, so we offered to buy them from the leasing company. They happily quoted us not that much, and we ended up with a room full of stuff.
We also received some networking kit that no-one knew how to use (turned out to be fancy cisco routers) that we sold to a networking company ... for the price we paid for the entire lot of kit. Our only problem is that we will have to pay to get rid of the servers in 3 years time when we do it again.
I'm not sure where the best place to find leasing companies is, but a good place to start might be those ebay sellers that only sell lots of BIN kit. Yellow pages after that.
yep, RedHat licences cost more than Windows, but (apart from the other facts already mentioned) you only need to buy 1 to be fully licenced for as many desktops and servers as you like ;););)
You can happily run CentOS as much as you like whilst paying RedHat for the 'warm fuzzy' feeling of support that, chance are, you'll never ever use.
Unfortunately not, though I'll see if the link is in my work browser history.. but I did find this which does cover some of the topics mentioned. (the link I referred to was a webcast/ video powerpoint so you can imagine how easy it is to find the damn thing using google).
This link seems to cover some of the same things that I recall- especially the issues of re-hashing strong named assemblies.
There are people who will use 1 browser because it is "the internet".
there are people who will use 1 browser because it is God's only browser and there is only one.
There are people who will use 1 browser because they cant be bothered to change.
And then there are people who will want to latest, fastest, feature-rich, talked-about browser. And if FF gets "superfast" stickers all over it in the popular press and blogs, people will want to use it. Nobody really wants to be stuck with yesterday's slow old slowness, not in the Internet Age. We've all been conditioned to always go for the upgrade, give us any reason and very many of us will.
I'm not sure about this as its a relatively obscure part of OS/app interaction, but I thought only the parts needed to be paged in were loaded when the shared library was loaded - the right parts are known at compile time as the linker does its thing.
For dynamically loaded libraries, the OS doesn't 'load' the dll, it maps the dll into the app's address space, (you might like to check out rebasing to see what happens when the dll's conflicts with address range already used). Once mapped, only the parts that are needed are loaded in the usual manner (when there is a page fault) as the app uses part of the dll.
However, take the above with a pinch of 'look it up for yourself'.
I do know that .NET apps load all their library code in, especially those that are secured with strong names - the loader has to load it all in order to verify that no-one has tampered with it, and it has to load all dependant libraries and they have ... so sometimes your .NET load performance really sucks. Apparently they're going to do something about it (not sure what) but the WPF guys have said that if you want your WPF apps to perform, do not use strongly named assemblies.
Why chain together tools yourself, and manually think about things you really shouldn't be thinking about, when you can have a good filesystem take care of it for you.
Because that's the Unix way - build small components (applications) and chain them together to create something out of the parts. I mean, why have ls and grep when you can have lsgrepsortfind? Really, the point is to have small, easily maintained apps that do 1 thing well than 1 app that does everything possibly well, but more usually poorly as its difficult to maintain and ensure it works properly. Not to mention the bloat when it replicates functionality already provided.
This may not be the best model for a critical component like a filesystem, but on the other hand, reliability of a filesystem is paramount, so keeping it as small as possible is probably a good idea.
then COM is interoperable, and anything that can create a dlls with a C API are interoperable.
And why do you think these native bits use ridiculous amounts of memory?
because his Java apps use ridiculous amounts of memory and he has to find something else to blame :-)
.NET is interoperable? Oh, you mean with Windows XP and Windows Mobile, of course.
(you can't mean Mono, as that's not Microsoft's .NET, that's be like saying Win32 is a cross-platform library because of Wine).
You can make an bike out of ordinary paper. I have a pen made of tightly-wrapped paper, its indestructible (well, I'm sure it isn't, but its very strong stuff). A paper bike is certainly feasible, if the paper core of the bike was suitably 'carbon wrapped'.
Going from 60 to 30 might qualify as "a very short amount of time"...
In geological time, yes. When you're staring at the screen, no.
but it is a new concept... for Windows, and to Microsoft's view of the world, that's the only thing that matters.
New != already done by someone else, not when they can market it as Microsoft 'innovating'.
you can see it... but once you have you'll wish you hadn't :-)
google for bootlog.txt, its a log file of everything Windows does when booting if you enable it. Googling to remind myself of the instructions to enable it tells me that Process Monitor now has an 'enable boot logging' option that's all nice and gui and easy to use.
4) The app store, sdk, etc are built for java. If you have a phone that isn't locked down, you could probably put native code on it, with some work.
5) Also, they use a custom jdk (and libraries), so the java apps have to be built specifically for android
I see a fair bit of complaints about this, a lot of developers who have existing symbian and other c++ based code want to re-use that for Android. Given that the underlying libraries are C++ anyway, it makes sense for Google to release more native-friendly development kits. Encourage them to go native, they will support it in the future if everyone asks. (see this blog)
what does the version number have to do with whether the product works or not?
no, instead of being allocated an IP address, they will give you a block (a /64 IIRC) which will give you enough IPs to address every electronic item you'll ever own.
Its more likely they'll set it up so your devices auto-configure using your ISPs block.
because, to continue your oil analogy, no-one will change until the cost of this increasingly-scare resource rises to a point where its cheaper to migrate to IPv6.
So - when your ISP says, you can have 1 shared IP, but if you want a static one it'll cost you $10 a month.. you'll start thinking about how you can get IPv6 and moving to a competitor who provides it, assuming you can get a home router that supports it... and then other ISPs will see the migration and start offering IPv6 as features, and then we'll be there.
It'll just take those market forces, so if you want to help things along, grab as many static IPs as you can.
ok, I chose poor a poor term ... but you're exactly right.
At the moment, I see a lot of devs writing .NET code, and wanting to slap a WPF GUI on them, Mono doesn't support this, and apparently won't for ages. So immediately you see Mono being cross platform only for a subset of applications (eg server). I think MS will be happy with that.