Slashdot Mirror

Instead of complaining you could at least make a half-hearted effort to find out what the problem is.

We have talked about W7 performance on netbooks which will only allow to run 3 apps. Perfect for an antivirus, a firewall, an antispyware, the WGA... oh crap!

The 3 app limit will only be for the starter edition, which is being aimed at "developing markets." Expect African, Asian, and South American users to be dissatisfied and perhaps unwilling to use Windows 7 when they're targeted.

All of those are kept around for backward compatibility purposes.

Microsoft inherited drive letters and backslashes from CP/M. My only concern with drive letters would be running out of them ("only" 26 available), but how many times have people run into this problem? I certainly never have. I will admit backslashes are somewhat of a pain (i.e. escape characters in languages like C), but it's not a major deal.

What's wrong with the minimize/restore/maximize buttons?

Under the hood Windows actually can move open files, but they chose not to allow this for the sake of simplicity. Aside from having to reboot after pretty much every update, I don't see the big deal with this one, either.

Considering one of Microsoft's top priorities is binary backward compatibility and none of the "junk" you mention is inherently broken in some way, I don't really see what your point is.

We have talked about W7 performance on netbooks which will only allow to run 3 apps. Perfect for an antivirus, a firewall, an antispyware, the WGA... oh crap!

Unless they decide to scrap everything they've written with Gazelle, I seriously doubt Trident is going anywhere:

Excerpt:
Instead of undertaking a significant effort of writing our own HTML parser, ren- derer, and JavaScript engine, we borrow these components from Internet Explorer 7 in a way that does not compromise security. Relying on IEâ(TM)s Trident renderer has a big benefit of inheriting IEâ(TM)s page rendering compatibility and performance.

Gazelle is from Microsoft Research, and their paper discusses the details of the security model - it's not just a marketing claim.

The idea is that every 'origin' (basically a domain name, which is used as the basis for access control in all modern browsers) is separated into its own sandboxed process. If a page on your domain embeds an iframe from an advertiser's domain, the iframe is rendered in a separate process, and all communication is handled through a Browser Kernel which enforces the security constraints (e.g. preventing the advert from touching or rendering anything outside its iframe box, even if an attacker can find a way to execute arbitrary code in it). Plugins are handled in the same way.

Chrome's security model doesn't handle that kind of separation of multiple sites within a single page. But Gazelle sacrifices some backward compatibility (e.g. it removes the document.domain attribute, and it requires all plugins to be rewritten to use the Browser Kernel instead of directly accessing the network or filesystem), which is unlikely to be acceptable in practice.

And Gazelle is certainly not a replacement for the IE engine - it's built on the existing IE7 components for parsing, rendering, scripting, etc. It's research, and the value is its ideas, some of which could perhaps be integrated into current browser engines to improve security. It's not meant to be a real browser engine, but it seems successful as a research experiment.

frosty ones!

goatse

Maybe I am a little too harsh, thats perfectly possible since I'm a full time *nix admin and as such some topics are simply common knowledge for me. But having said that... When I read the article I'm surprised at how much he raves on about Synaptic and how this is the way to install/remove software. I therefor conclude that this was one of the major drawbacks during his last attempt.

Still... I can't help wonder if its safe to conclude that this basically means that this guy never bothered to even glimpse at the Ubuntu manual (or online help screens)? I mean, how hard can it be? You goto the Ubuntu website (where you most likely went anyway to download it), then you click "Support" (or if you're daring click the Documentation link directly). And then you can take your pick, like 8.10.

The reason I'm pointing out the obvious is because if you followed this road (what other is there?) then you'll end up on this page. And when I see that the 2nd link is none other than Adding and removing software then I maybe an arrogant BOFH but I can't help myself saying: "Next time: RTFM!".

Granted.. Normally none of us do that at first and I'll have to admit that a total novice (c|w)ould expect Linux to be as userfriendly as Windows. iow; you can operate it without reading a manual. But if you have tried over and over and you still didn't succeed, wouldn't it be kinda logical to start poking around the Support section before resorting to help and displayed surprises like these?

I mean... Even Microsoft has an extensive support section amongst which How to use XP. Why would Linux be any different?

Maybe I am a little too harsh, thats perfectly possible since I'm a full time *nix admin and as such some topics are simply common knowledge for me. But having said that... When I read the article I'm surprised at how much he raves on about Synaptic and how this is the way to install/remove software. I therefor conclude that this was one of the major drawbacks during his last attempt.

Still... I can't help wonder if its safe to conclude that this basically means that this guy never bothered to even glimpse at the Ubuntu manual (or online help screens)? I mean, how hard can it be? You goto the Ubuntu website (where you most likely went anyway to download it), then you click "Support" (or if you're daring click the Documentation link directly). And then you can take your pick, like 8.10.

The reason I'm pointing out the obvious is because if you followed this road (what other is there?) then you'll end up on this page. And when I see that the 2nd link is none other than Adding and removing software then I maybe an arrogant BOFH but I can't help myself saying: "Next time: RTFM!".

Granted.. Normally none of us do that at first and I'll have to admit that a total novice (c|w)ould expect Linux to be as userfriendly as Windows. iow; you can operate it without reading a manual. But if you have tried over and over and you still didn't succeed, wouldn't it be kinda logical to start poking around the Support section before resorting to help and displayed surprises like these?

I mean... Even Microsoft has an extensive support section amongst which How to use XP. Why would Linux be any different?

You could ebay them, if your time is worth nothing. To prep them, you'd have to mount them on a machine and securely wipe them (on a windows box download sdelete for free from sysinternals.) Use the -z option to wipe free space (critical for cleaning flash drives.)

Old drives are not as energy efficient as modern drives, so they cost more to spin -- a RAID would just be an expensive storage container. So unless you have a need for old, small drives (say an old, small machine) the safest advice would be to destroy them.

I like playing with neodymium magnets, so I take my drives apart and harvest them. Bending and flexing the platters will render them unreadable by almost anyone but the NSA, so unless you're protecting treasonable secrets, it's probably not worth the effort to do much more damage than that. (Be careful, glass platters don't flex - they shatter.) If you are that paranoid, heating them beyond their Curie point will absolutely destroy any stored information.

The odds are quite good that you won't have to bootleg anything. Home Use Program

GPO is also extensible, which is nice: http://technet.microsoft.com/en-us/library/cc736356.aspx

If they haven't "cracked" the protocol by now, they aren't going to.

it does a 'dual path' of both sleep and hibernation most of the time.

Windows so intelligently will run the battery dead in sleep and then lose everything.

So does Windows Vista. It's called hybrid sleep.

It really was, and Microsoft still has text describing it in TechNet.

Ok, let's get this one out of the way...

For one, I don't believe Windows was originally designed to be a multi-user OS, was it? Everything it does that pretends to be has been an afterthought kludge. I honestly don't know if this is the case with NT-based systems so feel free to correct me.

Wrong. NT was very much designed around a multi-user model, they just did not enable any multi-user interfaces beyond telnet. The same multi-user NT level separation and code running today was in the first NT release.

3rd parties were providing multi-user on NT back in 1992-1993 when it first shipped.

NT 4.0 added in RDP, but the multi-user model and concurrent multi-user access was already there, it was just the GUI protocol added.

Wrong -- they didn't just have to introduce the RDP protocol, they had to change the kernel to support multiple Windows sessions on the server side (the terminal services). The Win32 subsystem only supports one interactive user at a time, so the kernel had to be modified to run several such subsystems at a time. in contrast to that, "Since UNIX was developed from the beginning as a multi-user system, it has been optimized to run multiple sessions on a single server.". Straight from the horse's mouth.

Ok, let's get this one out of the way...

For one, I don't believe Windows was originally designed to be a multi-user OS, was it? Everything it does that pretends to be has been an afterthought kludge. I honestly don't know if this is the case with NT-based systems so feel free to correct me.

Wrong. NT was very much designed around a multi-user model, they just did not enable any multi-user interfaces beyond telnet. The same multi-user NT level separation and code running today was in the first NT release.

3rd parties were providing multi-user on NT back in 1992-1993 when it first shipped.

NT 4.0 added in RDP, but the multi-user model and concurrent multi-user access was already there, it was just the GUI protocol added.

Wrong -- they didn't just have to introduce the RDP protocol, they had to change the kernel to support multiple Windows sessions on the server side (the terminal services). The Win32 subsystem only supports one interactive user at a time, so the kernel had to be modified to run several such subsystems at a time. in contrast to that, "Since UNIX was developed from the beginning as a multi-user system, it has been optimized to run multiple sessions on a single server.". Straight from the horse's mouth.

You don't go to http://update.microsoft.com/microsoftupdate/v6/default.aspx?ln=en ?

Any thoughts on how such a thing could be gotten to work?

I have a few, but I'm not sure they are really what you're after.

At the binary level, runas.exe is really the application that exposes "sudo" functionality to other applications. Bizarrely, despite being a console application, it doesn't seem to expose any functionality to select how an application is executed. In this case, is it executed in the parent console or in a new console? This seems to me to be a pretty odd oversight for a console application that could clearly see use for launching other console applications from a command line.

So, to achieve what you want, it seems to me you need to delve into Windows API's. So, you could do what you want in either Cygwin or Cmd through a "helper" application that uses the right API with the right parameters, but this to me seems quite clumsy for what ought to be a simple problem with a simple solution.

There is a potential alternative, but I'm not sure if it's viable for you. PowerShell (a new shell by Microsoft for Windows designed from the ground-up) is both .NET based and can directly interact with the .NET API from the shell. This might present a unique opportunity to solve your problem without resorting to actually building a simple (or not so simple depending on how customizable you need it to be) helper application.

You could, for instance, leverage the ProcessStartInfo Class. An example piece of script that you could adapt for your purposes (you'll need to customize it and probably use different/additional properties):
$runas = new-object System.Diagnostics.ProcessStartInfo
$runas.Arguments = "/noprofile /user:Box\Admin `"C:\example.exe`""
$runas.FileName = "runas.exe"
$runas.UseShellExecute = $false
$runas_proc = [System.Diagnostics.Process]::Start($runas)

The UseShellExecute property should ensure that your program runs in the parent console window, and it's the property I use to do exactly that in a script of mine (though, it doesn't use runas or any UAC functionality). The above class has a CreateNoWindow property as well, but I vaguely recall that not working for me, and someone gave me a technical explanation as to why it isn't applicable in this case but UseShellExecute is. My memory has faded somewhat though, so experimenting with both may well be worthwhile.

PowerShell V2 which is still in development adds many new cmdlets, one being a Start-Process cmdlet which should eliminate the need to script the .NET API directly for this case. Of course, you'd need to be using alpha software (CTP3 is the latest) to do this currently. The prior method would work in V1 and V2.

A quick google search suggests numerous other ways to use UAC in PowerShell through various approaches. PowerShell may well be worth a look, and there's no reason you can't call Cygwin applications from a PowerShell instance.

A quick bit of advice, if you do choose to investigate PowerShell, don't approach it like just another shell, as it's design is fundamentally different from anything on Unix or the prior Cmd shell. I'd recommend reading the wikipedia article on it for a good overview. However, if you do choose to use it, you'll have an extremely powerful shell with a featureset easily on par with the bash or zsh shell. The limitation will be in the frequent lack of 3rd party console applications that Unix has had for a long time, but, as I mentioned before, there's no reason you can't use these from PoSH.

I hope some of this is useful for you :)

Given that the user might still expect all that other stuff to work after "removing" IE, what are you really removing? A windowed presentation with some bookmark functionality?

Well, yes, that, and an address bar. That's also what most people call a "browser". The thing that renders the pages is "rendering engine". IE is the browser, MSHTML/Trident is the rendering engine.

Perhaps someone else can comment on how close Windows is to allowing some other browser vendor to be a plug-in replacement for all that other functionality.

MSHTML is embeddable into applications as an ActiveX control. ActiveX is COM-based, and COM is all about programming against interfaces. In case of MSHTML, that's the IWebBrowser2 interface, and everything that it references. Due to the nature of COM, it is, of course, entirely possible to provide your own implementation of all of these. At some point, the application also has to instantiate a specific COM component implementing that interface - that's identified by a GUID in the registry. Again, it is quite possible to rewrite that registry key so that it points to your implementations instead of MSHTML.

Of course, this is still a very non-trivial task, because there are many subtleties. Applications that host MSHTML can rely on many of its features (custom HTML elements and CSS properties, VML, JavaScript extensions or VBScript, ability to host .NET and ActiveX controls, etc). Those are all documented, but the sheer amount of things that must be handled is staggering.

Firefox 3.1's support for the CSS @font-face rule. With this ability, web developers have the option of specifying web fonts that must be downloaded for their website to appear as they intended.

If you are not logged in to your machine as Administrator/root, then how will this work. Well - it can not ever work, which is good!! I don't want some application to change my system.

Same with Firefox's AutoUpdate "feature" - it assumes that you're logged in as Administrator/root. How else could Firefox update its program directory? Please correct me if I am wrong.

If there's anybody out there still surfing the web as Administrator/root, then ... well - I give up. Then it's your own fault if your system gets messed up.

Microsoft even recommends the Principle of Least Privilege for User Accounts in Windows XP. I wonder how many people know about this and actually use their computer according to these guidelines.

On Linux/Unix this is the first thing you learn - you only login as root when you need to do administrative work (and only then). Web surfing definitely does not fall into that category.

Every version since Win98 has included the Windows Script Host by default. This allows one to automate quite a variety of tasks out of the box using vbscript or javascript. It's a little clunky for some things (e.g. recursive file searches), but is generally flexible enough for most needs.

The software check you did in windows 3.1 is available in windows ( as well as with various 3rd party addins) Details on the native version can be found at http://technet.microsoft.com/en-us/library/bb457006.aspx

article pointed me to [HKEY_CLASSES_ROOT\Folder\shellex\ColumnHandlers\{F9DB5320-233E-11D1-9F84-707F02C10627}] @="PDF Column Info"
Hoping to mitigate the vulnerability, I deleted the key after exporting it. However it does not cure the idiocy of Adobe allowing executables in something supposed to simply describe a document.

sure thing, just down load the source from the repository the type ./configure
make
make install
oh wait......

Well, first you need to sign up before you can get access to the repository. But other than that, I doubt Windows compiles with GNUmake or the GNU configuration tools.

I wonder if they will spend the money to make windows updates work with 'other than IE'.

seeing as the submitted didn't link it (or the 'editors' removed it?)

http://www.microsoft.com/technet/security/bulletin/MS08-066.mspx

Just to note from that security bulletin:
Published: October 14, 2008
Updated: January 13, 2009

This has already been patched for some time. Yes, I know, some are wary of installing patches in case they bring on some other issues, so one word of warning: if you use ZoneAlarm (by jove, why? WHY WHY WHY??), be sure to read the 'list of known issues after applying this patch':
http://support.microsoft.com/kb/956803

seeing as the submitted didn't link it (or the 'editors' removed it?)

http://www.microsoft.com/technet/security/bulletin/MS08-066.mspx

Just to note from that security bulletin:
Published: October 14, 2008
Updated: January 13, 2009

This has already been patched for some time. Yes, I know, some are wary of installing patches in case they bring on some other issues, so one word of warning: if you use ZoneAlarm (by jove, why? WHY WHY WHY??), be sure to read the 'list of known issues after applying this patch':
http://support.microsoft.com/kb/956803

I agree Silverlight is probably better than Flash, but that's setting a rather low bar.

Agreed - this is setting the bar low. A much better bar would be SVG. The 1.2 draft already supports every feature in Flash or Silverlight and is completely open (comes from the people who made HTML if this is new to you). It's completely xml based with the same DOM programming interface as HTML allowing AJAX as we're now seeing for HTML.

And here's where Microsoft is at fault:
SVG is probably already supported to some degree in your browser natively without any plugins (unless your primative enough to still be using IE) which is because Microsoft has been passively refusing to include it in its browser - presumable because it IS a direct competitor to Silverlight. Check out the MSDN thread about it's support that's been running since 2006 with no proper response.

Windows mostly uses counted (unicode) strings in kernel mode

http://msdn.microsoft.com/en-us/library/aa380518(VS.85).aspx

The idea is that you only interact with these strings by calling run time library runctions RtlXxx rather than fiddling around with pointers. It's actually quite OK once you learn how it works.

Symbian is obsessed with string descriptors too, and it has far too many types of them. You also need to push them onto a clean up stack so they can be freed if your code "leaves", a sort of pseudo exception.

Even simple string handling ends up incomprehensible gibberish.

http://descriptors.blogspot.com/2005/05/20-how-do-i-use-heap-based-buffer.html

E.g.

_LIT(KFred, "Fred"); // Allocate a heap descriptor of max length 4
HBufC* heapBuf = KFred().AllocLC();

_LIT(KCyril, "Cyril");
TPtr ptr(heapBuf->Des()); // Modifiable TPtr over heapBuf data area
ptr = KCyril(); // This would panic because max length (4) is exceeded // Instead, we need to do a reallocation // Leave on cleanup stack in case the realloc fails and a leave occurs
heapBuf = heapBuf->ReAllocL(5); // Realloc succeeded, but heapBuf pointer may have changed // We must update the pointer stored on the cleanup stack
CleanupStack::Pop(); // Push it off // Push it back on again
CleanupStack::PushL(heapBuf);

Note this guy knows what he's doing, code from someone who doesn't is much worse than this. Good job he knows that he has to update the pointer on the cleanup stakc beacuse ReAllocL might have changed it.

I was thinking along the same lines.

But to the person asking the question, if you want a full answer then you need to get your site built and make use of stress testing tools such as JMeter for Apache or Microsoft's WAS tool for IIS.

It's not something anyone here can give you a definite answer for without knowing how well your site is implemented and what it actually does.

Look into Transaction Cost Analysis, that's ultimately what you need here, a good start is this article:

http://technet.microsoft.com/en-us/commerceserver/bb608757.aspx

or this one:

http://msdn.microsoft.com/en-us/library/cc261632.aspx

Don't worry that these are MS articles on MS technologies they both still cover the ideas that are applicable elsewhere.

Even though no one here can give you a full answer for the above mentioned reasons, we can at least give you our best guesses and this is where I think the parent poster is spot on, 6 servers is absolute overkill for this kind of load requirements and indeed, unless your application does some pretty intensive processing I see little reason why a single server couldn't do the trick or at least a web/application server and a database server at most.

For ensuring high availability you may indeed need more servers of course and as you mention a requirement for FTP is bandwidth likely to be an issue?

The fact you're only expecting 1000 a day suggest you're not running the biggest of operations and although it's nice to do these things in house it may just be worth you using a hosting provider with an acceptable SLA, at the end of the day they have more experience, more hardware, more bandwidth and can probably even do things a fair bit cheaper than you can. Do you have a generator to allow continued provision of the service should your power fail for an extended period for example? If you receive an unexpected spike in traffic or a DDOS do you have the facility to cope with and resolve that like a big hosting company could?

There are many things I wouldn't ever use an external hosting provider for, but this doesn't sound like one of them.

I was thinking along the same lines.

But to the person asking the question, if you want a full answer then you need to get your site built and make use of stress testing tools such as JMeter for Apache or Microsoft's WAS tool for IIS.

It's not something anyone here can give you a definite answer for without knowing how well your site is implemented and what it actually does.

Look into Transaction Cost Analysis, that's ultimately what you need here, a good start is this article:

http://technet.microsoft.com/en-us/commerceserver/bb608757.aspx

or this one:

http://msdn.microsoft.com/en-us/library/cc261632.aspx

Don't worry that these are MS articles on MS technologies they both still cover the ideas that are applicable elsewhere.

Even though no one here can give you a full answer for the above mentioned reasons, we can at least give you our best guesses and this is where I think the parent poster is spot on, 6 servers is absolute overkill for this kind of load requirements and indeed, unless your application does some pretty intensive processing I see little reason why a single server couldn't do the trick or at least a web/application server and a database server at most.

For ensuring high availability you may indeed need more servers of course and as you mention a requirement for FTP is bandwidth likely to be an issue?

The fact you're only expecting 1000 a day suggest you're not running the biggest of operations and although it's nice to do these things in house it may just be worth you using a hosting provider with an acceptable SLA, at the end of the day they have more experience, more hardware, more bandwidth and can probably even do things a fair bit cheaper than you can. Do you have a generator to allow continued provision of the service should your power fail for an extended period for example? If you receive an unexpected spike in traffic or a DDOS do you have the facility to cope with and resolve that like a big hosting company could?

There are many things I wouldn't ever use an external hosting provider for, but this doesn't sound like one of them.

if your easily offended by anti-Microsoft sentiments you might feel more at home here: http://msdn.microsoft.com/en-au/default.aspx#

It's right on Microsoft's website. Looks to me like msft promised not sue because of vfat?

1. LIMITED LICENSE AND COVENANT NOT TO SUE.

(a) Provided that you comply with all terms and conditions of this Agreement and subject to the limitations in Sections 1(c) - (f) below, Microsoft grants to you the following non-exclusive, worldwide, royalty-free, non-transferable, non-sublicenseable license under any copyrights owned or licensable by Microsoft without payment of consideration to unaffiliated third parties, to reproduce the Specification solely for the purposes of creating portions of products which comply with the Specification in unmodified form.

http://www.microsoft.com/whdc/system/platform/firmware/fatgen.mspx

I think the problem is you are looking on CL. Try contacting the agencies/vendor companies directly, or even http://www.microsoft.com/careers/. Plenty of 6 figure jobs out there, IF you have the right experience.

"Driving Microsoft's Growth by Innovating From India". My grandfather told me it's good to know where you came from, it helps you to understand where you're going.

I have been using ws2k8-e for about 6 months now for my workstation/desktop. It's faster than regular vista and has the features I like about vista while not being full of bloat. I also have ws2k8-e core on my media server.

Nice! How much did the licenses cost?

Assuming that "ws2k8-e" means "Windows Server 2008 Enterprise, and further considering that according to Microsoft WS2K8-E has a list price of $3,999 US, that's a hefty chunk of change... even with OEM pricing it's expensive: 2 licenses at the lowest quoted price would be $4,808 US.

Or, did you go the MSDN Operating Systems route? That's still $699 US, and it's not licensed for production use.

Or did you just obtain it through copyright infringement?

Come on, you can tell us :)

Considering this comment, however, my money's on the last.

I have been using ws2k8-e for about 6 months now for my workstation/desktop. It's faster than regular vista and has the features I like about vista while not being full of bloat. I also have ws2k8-e core on my media server.

Nice! How much did the licenses cost?

Assuming that "ws2k8-e" means "Windows Server 2008 Enterprise, and further considering that according to Microsoft WS2K8-E has a list price of $3,999 US, that's a hefty chunk of change... even with OEM pricing it's expensive: 2 licenses at the lowest quoted price would be $4,808 US.

Or, did you go the MSDN Operating Systems route? That's still $699 US, and it's not licensed for production use.

Or did you just obtain it through copyright infringement?

Come on, you can tell us :)

Considering this comment, however, my money's on the last.

2008 R2, will give much more reasons to upgrade from 2003. Really interesting looks the move to 64bit only architectures, the vastly improved Hyper-V and the emphasis on performance optimizations. Related to the the Hyper-V enhancements and in particular the actual live migration support, I think that would be the moment to position itself as a real alternative to VMWare and Xen based hypervisors. http://www.microsoft.com/windowsserver2008/en/us/R2-top-reasons.aspx/.

I would wager they make that impossible to do since they'd rather you buy their own fancy new back up solution for that. Granted, I've messed with that solution and it's pretty cool, but it ain't cheap.

There are multiple issues which can cause what you describe, the most commonly one i've encountered in the wild is the combination of a WS08 bug (for which there is a hotfix) together with McAfee.

Most likely:
http://support.microsoft.com/kb/959816

Maybe (SMB2 only):
http://support.microsoft.com/kb/948572

Basically: If you have issues like that, don't reboot the servers. Open a PSS case.

There are multiple issues which can cause what you describe, the most commonly one i've encountered in the wild is the combination of a WS08 bug (for which there is a hotfix) together with McAfee.

Most likely:
http://support.microsoft.com/kb/959816

Maybe (SMB2 only):
http://support.microsoft.com/kb/948572

Basically: If you have issues like that, don't reboot the servers. Open a PSS case.

"64-Bit goodness" was available with win2K3 as well so even that's not a reason to go with win2K8.

Quite often "multilanguage" creeps in quite unintentionally; for example, in the last 3 years, I've only worked on one project which didn't contain a significant amount of XSLT - which, like it or not, is a separate language, and quite complicated one at that.

On the other hand, with .NET, it has become a pretty popular approach in Windows land to mix C++ and C# - the former for performance, and also when dealing with messy C/C++ APIs, and the later for high-level business logic and UI; though lines do get blurred sometimes. I think the reason why it is fairly common is because it is so easy to do - you can have a solution with C++ DLLs and C# projects together in it, with a few tricks you can set up dependencies, and calling C API from a DLL is very easy from C# thanks to P/Invoke (and even easier when there's a C++/CLI wrapper) - most of marshalling chores get taken care of automatically. I think this isn't quite as common with Java because JNI is, frankly, a very inconvenient API, and pretty slow as well. Python and others are better, but you still get a lot of boilerplate code in C. Then again, SWIG helps a lot (in all of the above scenarios), and for Python specifically, boost.python is great.

Getting back to .NET, with the late additions, it seems to be going beyond C++ & C# combo now. For one thing, VB actually has some neat features over C# now, most notably language integrated XML literals and queries, so it can be used much in the same way as XQuery. This is actually convenient enough when dealing with a lot of XML that I had used separate VB projects for XML processing in an otherwise fully C# application just for that reason. Then, of course, there's F#, the newcomer that is very handy at certain tasks: text parsing - FParsec is great!; tree and graph processing; and, in general, all sorts of computations, especially parallelizable ones.

Let's also not forget that ALL version of Windows, from NT on up, have been written to comply with C2 security standards:

*snip*

Which proves your point, it's the USER, not the OS, that lends itself toward a lack of security.

So, C2 security standards do not insist it is a bad thing when a simple 40 byte tcp packet can execute code on the remote windows machine, from the time when Windows came with TCP/IP, up until the beginning of 2008?

Then in no way should the word 'security' show up in the title 'The C2 security standard' :P

Oh, and since you took the time to include links, I should too.
http://www.microsoft.com/technet/security/bulletin/ms08-001.mspx

Not true.

READ. LEARN.

http://technet.microsoft.com/en-us/library/cc767094.aspx

I get that you like OSX / Linux better, and that's fine. I like them too, for some things. But to trash ANY OS with false information is living on a belief system, rather than an informed and educated system, my friend :)

You could check their official website though. It's not as if Wikipedia contains everything.

http://www.microsoft.com/esp/

http://msdn.microsoft.com/en-us/subscriptions/aa948864.aspx
Look for "Microsoft Virtual CD Control Tool". They have had that on their site for years...

its been a long time since I did Win32, but I remember when they changed it so applications couldn't "steal" focus from another application if the focused application hadn't seen mouse or keyword activity in X seconds (X configurable through the registry). The number of times the taskbar window flashed was also a configurable registry setting... somehow, though, applications like Outlook could ALWAYS steal focus. I always wondered what API call they used to do that, because I could never find it, and I scoured MSDN.

They used this one (recently documented as part of MS's antitrust settlement with the EU, AIUI).

"Blue Screen of Death" now "Azure Notice of Discomfort" in preparation for new cloud computing initiatives.