I don't think it's really correct to call anything in Silverlight a "web app" not only is it a client based runtime... but it really didn't get the necessary features to support a full featured webapp until Silverlight 3. What Silverlight is... is a step between a clickonce app, and a web page. I think MS's message is going to start becoming along these lines: "If you can do it in HTML5 the do, if you can't then use Silverlight." What that message says to me is that Silverlight is for apps that you want to make REALLY REALLY easy to deploy and are willing to live with the restrictions Silverlight places on that application.
<rant>Personally I think the tragedy of Silverlight... and indeed.NET in general is that MS has not made it more open, not so open as Java where it degrades into detrimental infighting, but open in the sense that it would be easy for Novell to implement Moonlight and Mono. I am one of the many.NET developers who wishes that Mono had brought WPF to linux. Why? Because at least then I could make apps that don't look like shit on linux. I presume that it never happened due to licensing issues over DirectX with MS, which is really a shame. Mono as a client runtime is a tragedy, what it could have done, and what it did is good... but not really any better than java. While the server runtime is worth talking about, it is as far as I am aware not used extensively, thus making it a failure.</rant>
Right now, I'm using C#, Windows.Forms, and Managed DirectDraw for graphics. DirectDraw is very fast at creating and rendering graphics, it does a good job. It's a little tricky to set up, but performance is much better than System.Drawing.
I do get tons of warnings that my code is using classes which are marked as deprecated, and the built in XML documentation is also drowned out with a big long warning about how DirectDraw is deprecated, blah blah, then finally when you look at the last sentence, there's your actual documentation.
Also, the program is made for a Windows XP machine, and refuses to run on another Windows 7 machine. I have no idea which dependency it doesn't like.
For C#, SDL is a non-starter. There is an SDL library available, but you are restricted to only using the single SDL window that's created, you can't make other windows or controls in that window. There is also a SDL panel control for Windows Forms, but it's a bad joke that draws to the screen by assigning System.Drawing.Bitmap objects.
The one SDL object per form...that is because WinForms is very thinly wrapped MFC/GDI+ and while for very simple UI applications it works great. It is fundamentally limited due to the fact that it is legacy reaching back to the 16bit windows days and is dependent on USER and GDI objects (these are very limited resources) and also has NO hardware acceleration on most machines.
As for DirectDraw... that IS the reason it doesn't work on Windows 7, Microsoft doesn't even ship those libraries to 7 unless they are specifically installed by the user, and why would they! They have Direct2D, a well designed (but currently buggy because it's new) framework. If you're programming.NET and want access to that power just use WPF which is a thinly veiled layer over DirectX10 (with a built in back compatiblity layer for XP) with some basic work already done for you. Why would you waste your time programming on an obsolete (and soon to be removed) component, when there is a fully supported, well documented framework already available!
While I know the OP wasn't interested really in 3D as a C# guy I have to give a shout out to XNA which has some excellent 3d capabilities from a managed language, I wish that MS would allow others to implement XNA as it would be a major boon to linux gaming if they did
This headline should really be fixed as it fails to differentiate between Cook's Illustrated and the offending Cook's source, the former is often simply called "Cook's" so a more specific headline (in my opinion) is in order.
A client side plugin does not satisfy FIPS as it has to be an integrated portion of the system, this is because the requirement also includes the ability for network administrators to enforce S/MIME on all messages on the network, which cannot be enforced using a client side plugin. Sadly this sort of requirement is why most of the US government still uses IE as well.
Sad but usually true.... at the company that I work (which for a large variety of reasons shall remain anonymous) we make extensive use of MS's Hyper-V virtualization. The smart admin would go, "ok I want Hyper-V that's ok.... let's install server core and make more efficient use of our resources" (Hyper-V restricts the number of VM's based on RAM usage, so and Ideal Hyper-V box is one with less processor but stuffed to the gills with RAM). Do they? NO! Why? Because they are scared of using the tools MS has provided them that would allow them to easily remotely administer it (Remote Server Administration Tools, Powershell, Win-RM) all of which are fully capable of scripting (most new ones are built on powershell and ship with powershell extentions).
Powershell is arguably just as powerful as bash if not comparable to some of the extension shells, the ability to directly access both COM and the.NET framework providing massive flexibility to an admin (and the ability to write their own tools in C#). However most Windows admins still RDP to the server... and still install server full... why because it's what they are comfortable with.
In the end I think you're right... linux/unix admins simply do know more about the system they work on, most windows admins couldn't tell you when they might want to run ngen.exe on their.NET app or not... which is sad
Commenting is not enough... you need to make sure you clean up the comments in a file before resubmitting it... dead comments can be more dangerous than dead code... as dead code at least doesn't run... dead comments mislead the person following later into believing a lie a lie that could potentially have major impacts on the software.
Lastly... write meaningful test cases... if your tests only prove the happy path, that's OK... but if they prove that the happy path is the only path.. that is better.
you are making the assumption that that is the sole goal of the spammers... I don't think that is a safe assumption. Nor do I think that it is safe to assume that people won't get suckered by it anyway.
Agreed wholeheartedly, locking down IPtables is also a good Idea. I set up a linux webserver a coupla months back, it has three ports you can hit from either IPv4 or IPv6: 22,80, and 443 the only ports that should ever be open on that box. I would have bothered with SSHDFilter too but I just set up keys.
I'm running Server 2008 with the domain and forest at 2008 level, ALL of my machines are set up to use KerberosV5 and LDAP, the only ones that even occasionally give me trouble about it are some legacy Server 2003 boxes (XP seems to work fine or maybe I'm just deluding myself, which is what I would think too, if I hadn't checked the logs) and the ONLY reason they give me trouble about it is because they were originally connected to an SBS2003 domain. Vista and 7, if I have a problem with them, it's that occasionally they get it in their mind to be too secure for the legacy boxes, actually occasionally is the wrong word as it has only happened once.... So to say that Windows WON'T work with Kerberos and LDAP is just plain wrong, could the utilities for it be better, yes... and I won't argue that windows has a long way to go. However nor will I say that Samba's efforts are wasted, although from my perspective as I do mostly windows admin it is a little bit ironic, that as far as I can tell their interaction with the Microsoft AD folks has mostly been a help to the MS guy's to make AD more reliable and secure.
Why am I the one posting this... (sighs) anyway this is all 20-20 hindsight but let it be a lesson to always follow the rules of security (Yes it's on technet, yes MS should follow their own rules). Failure to do so will result in this in this case you failed on.
Law #2: If a bad guy can alter the operating system on your computer, it's not your computer anymore
Law #3: If a bad guy has unrestricted physical access to your computer, it's not your computer anymore.
Regardless, you should get your own hardware in a co-lo you've background checked and have references for. Once you've done that you should do what every respectable admin has been doing for years, turn off direct root access, Set up Public Key Authentication, and for the love of all that is secure turn off password auth for SSH. If you do that then any unauthorized access to your box is in violation of 18 U.S.C. 1030 and is punishable under the law.
Intel does in fact have some support for BCD (note this is not IEEE 754r) in their processors using the BCD opcode, this could potentially be used to implement IEEE 754r
But the GC does not solve two things:
2) Taking up tons of RAM because of unnecessary allocations (I've seen Java code that allocates MBs in tight loops...)
I can't answer your first one because I would have do more research than I feel like doing at the moment, however your second comment makes me laugh, it is quite apparent that you like most programmers that I know of java,.net, and other managed languages don't understand the single most critical thing any programmer must understand about these languages. They are reference based meaning that every time you create a new variable you allocate more ram, however if you had defined the variable in a c style manner at the top of the block outside the loop then the memory is only allocated once.
WPF: a set managed bindings for the Direct3D API designed to allow.NET applications to scale to large resolutions without looking like utter crap (like winforms or GDI/GDI+ would).
Direct2D: a set of APIs built on top of Direct3D that allows for applications to use vector graphics and hardware accelerated scaling without the pain of directly dealing half the crap in Direct3D. Note: this is NOT a replacement for DirectDraw!!!
Direct3D: a set of APIs originally designed to allow applications (read games) to take advantage of hardware acceleration, as the apparent flaws in GDI/GDI+ became apparent it became the API of choice for application programmers worth their salt to use so their apps wouldn't look horrible when scaled to high resolution, however because Direct3D doesn't use device independent pixels the same flaw that affects GDI/GDI+ still affected these apps (unless they did their own scaling code)
GDI/GDI+: a set of API's that were originally put in to windows 95 to allows application to access the DWM (desktop window manager) and create decent looking applications, in Windows XP GDI was deprecated in-favor of the improved (the author chokes) GDI+. Several things should be noted here, first GDI/GDI+ do have support for scaling, however they required the programmer to implement it, something most either didn't care to do, or didn't want to do. GDI/GDI+ is no longer hardware accelerated in Vista on non WDDM (Windows Display Driver Model) devices (thing 945gm). Finally GDI/GDI+ is probably the largest source of memory bloat in many windows applications (and consequently why IE has soooo much memory bloat)
I believe you are what would be best described as a "Blue Dog" Fiscally, and constitutionally conservative while socially liberal, this used to be known as a republican until the Republican party decided that the religious right should be part of their base, despite the fact that the only commonality they had was the word "Conservative." Now these traditional republicans are given the name RINO by the likes of Sarah Palin. At some point this faction will break away from the Republicats and form a true opposition, however until then we can do our best to vote for candidates we truly support.
LyX is nice, and for my algorithms HW it was essential, it just had symbols that even Word 2007 didn't have, (Word 2003's equation editor is a joke), and once I learned the keyboard shortcuts it was a lot faster too. Word 2007 actually uses many of the same shortcuts (when in equation editor mode) as Lyx (it seemed to me at the time that MS was trying to make a poor copy. and did a half way decent job of it given that Word is not a typesetting program). Windows 7 has a math pad, lacking a wacom or other reasonable input device I was unable to adequately test it. However given the Macbook Pro LyX is probably the best answer because LyX has a Mac distribution.
I seem to be having Deja'vu with a little company called SiCortex that no longer exists (the Wikipedia article is out of date). Why? Because nobody wanted to rewrite their software for a machine less than 0.1% of the market used. The other reason is that most software in the HPC world is written to be GCC/Intel compatible so porting to PGI was interesting. This chip might go somewhere if they can market it as a coprocessor with BLAS libraries, however even if they try to do that their going up against IBM with its CELL blades and it's mixed platform BLAS as used by Roadrunner which is currently the number one on the top500.
That too has been a focus of this whole fiasco, the fact that the plug-in exists isn't a problem, nor is the functionality it provides, which is critical to many enterprise FF users. The core of the whole thing and what has pissed most people off on both sides is that both MS and Mozilla took action without customer consent, effectively choosing for us. First MS for installing it, then Mozilla for disabling it. The resounding consensus has been: "Could you at least ask first?" By acting without consent both Microsoft and Mozilla have shown that they don't think their users can make an informed consenting decision about how they wish to use their browsers. Such an action is disrespectful and disgraceful of both parties. Microsoft should know better having been burned before, and Mozilla because such an action is against the core principles on which Firefox was marketed.
Mike, any user NOT installing the IE updates on Windows is an idiot, because the COM components of IE are used in many applications. Thus not patching IE even if they haven't opened it in ages is the stupidest thing they could ever do (followed by not updating Flash Player) for the security of their system. So saying that people won't install the patch because it has the letters IE in the name is bull. The patch is listed as a CRITICAL update, not recommended but CRITICAL. On the other hand should MS introduce an optional update to install an updated version of the plug-ins? I'm thinking so...
To those who are going to make the inevitable comments about the use of the COM IE components supporting MS browser monopoly: your right, but there is no guaranteed alternative, and I have yet to see a COM interface for FF, or Chrome.
Protocols and other intellectual property made under contract to the military cannot be patented, the implementation can, but the standard cannot, courtesy of the US Constitution
With the same processor as the Zune HD, I wonder if anyone will port casual games from one to the other? I don't think they would port anything large but at least in theory it could happen....
An interesting work around to the implicit right to privacy in the 9th amendment. The States are authorized to gather and store such information under the 10th amendment, however there is nothing preventing the FBI from asking kindly for access.
Slashdot Mirror
I don't think it's really correct to call anything in Silverlight a "web app" not only is it a client based runtime... but it really didn't get the necessary features to support a full featured webapp until Silverlight 3. What Silverlight is... is a step between a clickonce app, and a web page. I think MS's message is going to start becoming along these lines: "If you can do it in HTML5 the do, if you can't then use Silverlight." What that message says to me is that Silverlight is for apps that you want to make REALLY REALLY easy to deploy and are willing to live with the restrictions Silverlight places on that application.
<rant>Personally I think the tragedy of Silverlight... and indeed .NET in general is that MS has not made it more open, not so open as Java where it degrades into detrimental infighting, but open in the sense that it would be easy for Novell to implement Moonlight and Mono. I am one of the many .NET developers who wishes that Mono had brought WPF to linux. Why? Because at least then I could make apps that don't look like shit on linux. I presume that it never happened due to licensing issues over DirectX with MS, which is really a shame. Mono as a client runtime is a tragedy, what it could have done, and what it did is good... but not really any better than java. While the server runtime is worth talking about, it is as far as I am aware not used extensively, thus making it a failure.</rant>
Right now, I'm using C#, Windows.Forms, and Managed DirectDraw for graphics. DirectDraw is very fast at creating and rendering graphics, it does a good job. It's a little tricky to set up, but performance is much better than System.Drawing. I do get tons of warnings that my code is using classes which are marked as deprecated, and the built in XML documentation is also drowned out with a big long warning about how DirectDraw is deprecated, blah blah, then finally when you look at the last sentence, there's your actual documentation.
Also, the program is made for a Windows XP machine, and refuses to run on another Windows 7 machine. I have no idea which dependency it doesn't like.
For C#, SDL is a non-starter. There is an SDL library available, but you are restricted to only using the single SDL window that's created, you can't make other windows or controls in that window. There is also a SDL panel control for Windows Forms, but it's a bad joke that draws to the screen by assigning System.Drawing.Bitmap objects.
The one SDL object per form...that is because WinForms is very thinly wrapped MFC/GDI+ and while for very simple UI applications it works great. It is fundamentally limited due to the fact that it is legacy reaching back to the 16bit windows days and is dependent on USER and GDI objects (these are very limited resources) and also has NO hardware acceleration on most machines.
As for DirectDraw... that IS the reason it doesn't work on Windows 7, Microsoft doesn't even ship those libraries to 7 unless they are specifically installed by the user, and why would they! They have Direct2D, a well designed (but currently buggy because it's new) framework. If you're programming .NET and want access to that power just use WPF which is a thinly veiled layer over DirectX10 (with a built in back compatiblity layer for XP) with some basic work already done for you. Why would you waste your time programming on an obsolete (and soon to be removed) component, when there is a fully supported, well documented framework already available!
While I know the OP wasn't interested really in 3D as a C# guy I have to give a shout out to XNA which has some excellent 3d capabilities from a managed language, I wish that MS would allow others to implement XNA as it would be a major boon to linux gaming if they did
To para quote a former boss of mine who worked on the Windows 95 team:
"Oh we never coded any thing in Perl... if we needed anything we just asked Larry to do it"
-David Theilen
This headline should really be fixed as it fails to differentiate between Cook's Illustrated and the offending Cook's source, the former is often simply called "Cook's" so a more specific headline (in my opinion) is in order.
A client side plugin does not satisfy FIPS as it has to be an integrated portion of the system, this is because the requirement also includes the ability for network administrators to enforce S/MIME on all messages on the network, which cannot be enforced using a client side plugin. Sadly this sort of requirement is why most of the US government still uses IE as well.
Sad but usually true.... at the company that I work (which for a large variety of reasons shall remain anonymous) we make extensive use of MS's Hyper-V virtualization. The smart admin would go, "ok I want Hyper-V that's ok.... let's install server core and make more efficient use of our resources" (Hyper-V restricts the number of VM's based on RAM usage, so and Ideal Hyper-V box is one with less processor but stuffed to the gills with RAM). Do they? NO! Why? Because they are scared of using the tools MS has provided them that would allow them to easily remotely administer it (Remote Server Administration Tools, Powershell, Win-RM) all of which are fully capable of scripting (most new ones are built on powershell and ship with powershell extentions).
Powershell is arguably just as powerful as bash if not comparable to some of the extension shells, the ability to directly access both COM and the .NET framework providing massive flexibility to an admin (and the ability to write their own tools in C#). However most Windows admins still RDP to the server... and still install server full... why because it's what they are comfortable with.
In the end I think you're right... linux/unix admins simply do know more about the system they work on, most windows admins couldn't tell you when they might want to run ngen.exe on their .NET app or not... which is sad
Commenting is not enough... you need to make sure you clean up the comments in a file before resubmitting it... dead comments can be more dangerous than dead code... as dead code at least doesn't run... dead comments mislead the person following later into believing a lie a lie that could potentially have major impacts on the software. Lastly... write meaningful test cases... if your tests only prove the happy path, that's OK... but if they prove that the happy path is the only path.. that is better.
you are making the assumption that that is the sole goal of the spammers... I don't think that is a safe assumption. Nor do I think that it is safe to assume that people won't get suckered by it anyway.
Agreed wholeheartedly, locking down IPtables is also a good Idea. I set up a linux webserver a coupla months back, it has three ports you can hit from either IPv4 or IPv6: 22,80, and 443 the only ports that should ever be open on that box. I would have bothered with SSHDFilter too but I just set up keys.
"Won't Work With" ....
I'm running Server 2008 with the domain and forest at 2008 level, ALL of my machines are set up to use KerberosV5 and LDAP, the only ones that even occasionally give me trouble about it are some legacy Server 2003 boxes (XP seems to work fine or maybe I'm just deluding myself, which is what I would think too, if I hadn't checked the logs) and the ONLY reason they give me trouble about it is because they were originally connected to an SBS2003 domain. Vista and 7, if I have a problem with them, it's that occasionally they get it in their mind to be too secure for the legacy boxes, actually occasionally is the wrong word as it has only happened once.... So to say that Windows WON'T work with Kerberos and LDAP is just plain wrong, could the utilities for it be better, yes... and I won't argue that windows has a long way to go. However nor will I say that Samba's efforts are wasted, although from my perspective as I do mostly windows admin it is a little bit ironic, that as far as I can tell their interaction with the Microsoft AD folks has mostly been a help to the MS guy's to make AD more reliable and secure.
Why am I the one posting this... (sighs) anyway this is all 20-20 hindsight but let it be a lesson to always follow the rules of security (Yes it's on technet, yes MS should follow their own rules). Failure to do so will result in this in this case you failed on.
Regardless, you should get your own hardware in a co-lo you've background checked and have references for. Once you've done that you should do what every respectable admin has been doing for years, turn off direct root access, Set up Public Key Authentication, and for the love of all that is secure turn off password auth for SSH. If you do that then any unauthorized access to your box is in violation of 18 U.S.C. 1030 and is punishable under the law.
Intel does in fact have some support for BCD (note this is not IEEE 754r) in their processors using the BCD opcode, this could potentially be used to implement IEEE 754r
Agreed this is in the best traditions of the Apple 1
But the GC does not solve two things: 2) Taking up tons of RAM because of unnecessary allocations (I've seen Java code that allocates MBs in tight loops...)
I can't answer your first one because I would have do more research than I feel like doing at the moment, however your second comment makes me laugh, it is quite apparent that you like most programmers that I know of java, .net, and other managed languages don't understand the single most critical thing any programmer must understand about these languages. They are reference based meaning that every time you create a new variable you allocate more ram, however if you had defined the variable in a c style manner at the top of the block outside the loop then the memory is only allocated once.
I believe you are what would be best described as a "Blue Dog" Fiscally, and constitutionally conservative while socially liberal, this used to be known as a republican until the Republican party decided that the religious right should be part of their base, despite the fact that the only commonality they had was the word "Conservative." Now these traditional republicans are given the name RINO by the likes of Sarah Palin. At some point this faction will break away from the Republicats and form a true opposition, however until then we can do our best to vote for candidates we truly support.
LyX is nice, and for my algorithms HW it was essential, it just had symbols that even Word 2007 didn't have, (Word 2003's equation editor is a joke), and once I learned the keyboard shortcuts it was a lot faster too. Word 2007 actually uses many of the same shortcuts (when in equation editor mode) as Lyx (it seemed to me at the time that MS was trying to make a poor copy. and did a half way decent job of it given that Word is not a typesetting program). Windows 7 has a math pad, lacking a wacom or other reasonable input device I was unable to adequately test it. However given the Macbook Pro LyX is probably the best answer because LyX has a Mac distribution.
Have you used the Nvidia drivers under linux, CUDA is about the only thing they're good for
Screw Vista!! Windows 7 FTW! although seriously would that box even run Crysis...., I'm not actually sure it would...
I seem to be having Deja'vu with a little company called SiCortex that no longer exists (the Wikipedia article is out of date). Why? Because nobody wanted to rewrite their software for a machine less than 0.1% of the market used. The other reason is that most software in the HPC world is written to be GCC/Intel compatible so porting to PGI was interesting. This chip might go somewhere if they can market it as a coprocessor with BLAS libraries, however even if they try to do that their going up against IBM with its CELL blades and it's mixed platform BLAS as used by Roadrunner which is currently the number one on the top500.
That too has been a focus of this whole fiasco, the fact that the plug-in exists isn't a problem, nor is the functionality it provides, which is critical to many enterprise FF users. The core of the whole thing and what has pissed most people off on both sides is that both MS and Mozilla took action without customer consent, effectively choosing for us. First MS for installing it, then Mozilla for disabling it. The resounding consensus has been: "Could you at least ask first?" By acting without consent both Microsoft and Mozilla have shown that they don't think their users can make an informed consenting decision about how they wish to use their browsers. Such an action is disrespectful and disgraceful of both parties. Microsoft should know better having been burned before, and Mozilla because such an action is against the core principles on which Firefox was marketed.
Mike, any user NOT installing the IE updates on Windows is an idiot, because the COM components of IE are used in many applications. Thus not patching IE even if they haven't opened it in ages is the stupidest thing they could ever do (followed by not updating Flash Player) for the security of their system. So saying that people won't install the patch because it has the letters IE in the name is bull. The patch is listed as a CRITICAL update, not recommended but CRITICAL. On the other hand should MS introduce an optional update to install an updated version of the plug-ins? I'm thinking so...
To those who are going to make the inevitable comments about the use of the COM IE components supporting MS browser monopoly: your right, but there is no guaranteed alternative, and I have yet to see a COM interface for FF, or Chrome.
Protocols and other intellectual property made under contract to the military cannot be patented, the implementation can, but the standard cannot, courtesy of the US Constitution
With the same processor as the Zune HD, I wonder if anyone will port casual games from one to the other? I don't think they would port anything large but at least in theory it could happen....
An interesting work around to the implicit right to privacy in the 9th amendment. The States are authorized to gather and store such information under the 10th amendment, however there is nothing preventing the FBI from asking kindly for access.