The Win32 API basically is the operating system. So of course you need to access the Operating System for an application to work.
It is not. What it is is Microsoft's preferred and best-documented interface to NT. Any Win32 API that deals with a kernel object uses native APIs (e.g. ones beginning with Nt, Ke, Ki, etc.) to do the actual dirty work.
Whether you need to go through the overhead of a windowing system and message dispatch queue just to ping another machine(as the original poster implied) is another story.
The original poster implied no such thing.
CSRSS.EXE is currently required to be loaded and running to even use an app that does no GUI stuff at all. This is unnecessary overhead. The poster did not imply that one needs to use the windows system's messaging facilities to do IPC or networking.
--
"Where, where is the town? Now, it's nothing but flowers!"
Actually, that's required to keep backward compatability with Linux Samba servers.
I think you'll find that it is also required for backward-compatibility with all non-Windows 2000 Microsoft operating systems, too. I doubt that retaining compatibility with Samba (which runs on a hell of a lot more operating systems than Linux, remember) was a major consideration when Active Directory was designed. Please don't blame Samba for AD's design.
The namespace IS flat, because everything is stored in a flat directory for more effecient searching.
This was done purely for backward compatibility reasons. It is not an efficiency measure; they could easily have keys like "rmalda.sales" and "rmalda.geeks" to avoid collisions for users in the same domain but different containers. In fact, containers in AD are purely eye-candy. They're not security principals; only users and groups are.
I believe the actual namespace (the GUID) is about 23 trillion entries max.
An AD domain could only hold significantly fewer than 23 trillion users, even given unlimited memory, disk and network bandwidth. You can't take the largest number AD can use as an index and call that the maximum capacity.
--
"Where, where is the town? Now, it's nothing but flowers!"
In particular, W2K finally has real directory services (which Linux lacks)
Windows 2000's Active Directory is a candy-coated shell over a reworking of the old NT4-style domain system.
An example:
Suppose you create an AD domain called "slashdot.org". You then create two OUs, so you have sales.slashdot.org and geeks.slashdot.org. Being a hardcore geek, Rob Malda will of course be rmalda in the OU geeks.slashdot.org. His username in the underlying domain datanase will also be rmalda.
When Richard Malda join sales, of course you will want to add a user rmalda to sales.slashnet.org. But you can't. Why? Because even if you have objects in separate OUs, their names must still be uniqiue in the domain, because the underlying domain database has a flat namespace.
--
"Where, where is the town? Now, it's nothing but flowers!"
RealPlayer however has repeatedly caused my Win2k to totally lock up, never to recover. I geuss what MS always said about 3rd party drivers being at fault actaully has an element of truth.
RealPlayer is an application, not a kernel-level driver. If an application locks your operating system hard, your operating system has a problem.
--
"Where, where is the town? Now, it's nothing but flowers!"
> Calls to the windowing system for IPC? What
> about named pipes? Mailslots? or TCP/IP?
> Or shared memory?
All of those features are accessed using calls to the Win32 API (bar TCP/IP, which is accessed via the WinSock libraries). Therefore, the Win32 subsystem has to be running for those facilities to be available to the application programmer.
--
"Where, where is the town? Now, it's nothing but flowers!"
And am I correct in my assumption that PGP remains OK as long as you don't create an ADK? Or am I misreading the message?
The problem is that anyone can add an ADK to a public key without affecting the key's fingerprint. In other words, it is perfectly possible for someone to set up a keyserver that adds an ADK for themselves to each key uploaded, and no-one will by any the wiser, unless they examine the key closely.
How they get their hands on the mail encrypted using those keys is of course outside the scope of this post.
Idea: A company could set up an internal auto-ADK-adding keyserver for its employees to use, and of course they have access to the outgoing mail spool.
--
"Where, where is the town? Now, it's nothing but flowers!"
From what I can make of the section regarding GnuPG, it doesn't warn about the presence of the ADK. However, it places but one session key in the cryptogram, a key only recoverable using the user's private key.
But if you get a contaminated version-4 public key, GnuPG will not warn you about it. You should check any and all public keys that you use as decribed in the article. I'm sure the GnuPG team will not be long in adding functionality to do this automatically.
--
"Where, where is the town? Now, it's nothing but flowers!"
Possibly the editors are aiming for maximum bottom-feeder no-think-just-post reactions (more comments, more banner views, yes?). The title you gave your submission just didn't have that. "IBM Kills Monterey" seemingly does.
--
"Where, where is the town? Now, it's nothing but flowers!"
from the looks of inti, signals are still c style callbacks with no type checking?
No.
Looking at the headers available at
http://sources.redhat.com/inti/inti-manual/ I see that each class member representing a signal is a SignalProxy, which seems to be quite like the signalling facilities provided by libsigc++. Looking at the definitition of SignalProxy, its connect method is parameterized for both the type of the signalled object and the member being connected to.
--
"Where, where is the town? Now, it's nothing but flowers!"
I had a quick look at the screen shot for the import/export tool. Helix has struck a balance between providing a usable interface and not concealing too many details.
All the information a power users needs to map from the GUI to what happens 'under the hood' is present, and yet a newbie could use that same tool to connect to whatever network resources he/she desires without difficulty.
--
"Where, where is the town? Now, it's nothing but flowers!"
Application programmers are just lazy now, is all.
A facetious statement at best. Application programmers have to rely on large bodies of code over which they have little or no control, e.g. MFC and other class libraries/frameworks.
And to say that a poor OS encourages good application engineering is silly, to say the least. An OS with memory protection, such as Windows NT, Unix or Mac OS X, makes it much more practical and efficient to debug applictions suffering from stray pointers, etc.
-- "Where, where is the town? Now, it's nothing but flowers!"
Apple succeeded in making Mac OS a best-of-breed operating system for personal computers: Mac OS has set the standards against which modern graphical user interfaces are now modeled.
Not a good start to the article. The operating system that Apple developed is poor at best; a single application crash almost always brings down the machine.
The interface, on the other hand, feels solid and reliable; serious thought went into every part of it.
Nice GUI, shame about the OS.
-- "Where, where is the town? Now, it's nothing but flowers!"
My comment regarding AI was aimed more at the program being able to parse previously unknown data formats (even if they do use XML syntax) to extract the required information. Doing the actual comparisons, etc., would not be beyond us even today, I think, assuming that the program was fed the data is a well-defined format.
An XML schema is not the equivalent of an SGML DTD. XML is simply SGML with restrictions on tag minimization (<em/foo/ versus <em>foo</em>), case sensitivity and some other stuff I forget. XML and SGML DTDs are pretty much alike. XML Schemas are indeed marked-up using an XML DTD.
Currency conversion is rife with special cases. A case in point is converting between currencies in the Euro zone. You don't. What you must do is convert from currency A to Euros, and then from Euro's to currency B.
-- "Where, where is the town? Now, it's nothing but flowers!"
I think the promise of XML is that there will some day be more intelligent programs that can read the "field descriptions", meta-data, or whatever you want to call it, from the DTD. It's a formal way to structure identifiers for data.
The DTD does not specify "meaning" for tags or their contents. The DTD simply specifies which tags can be nested and in what combinations.
To give meaning to the data, you need a schema; I believe that the W3C has proposals for describing XML schemas.
What XML would allow, if people used it properly, is for future programs to make sense of data without having an agreed-upon format
No, it would not. You would need sophisticated AI such as seen in popular movies to do that kind of parsing. Unless two programs agree on a DTD, a schema and what the schema means in relation to the real world, they cannot interoperate.
-- "Where, where is the town? Now, it's nothing but flowers!"
The tablet is nothing to do with.NET per se; it was a bit of flash to cover up the fact very little of what has been announced is actually available right now.
Demos are designed to be cool; it is essential to their nature. Remember the demo of NT 3.5 that was actually pre-recorded and mimed to by the presenter? Demos are not a reflection of reality in many cases.
-- "Where, where is the town? Now, it's nothing but flowers!"
Slashdot Mirror
It is not. What it is is Microsoft's preferred and best-documented interface to NT. Any Win32 API that deals with a kernel object uses native APIs (e.g. ones beginning with Nt, Ke, Ki, etc.) to do the actual dirty work.
The original poster implied no such thing.
CSRSS.EXE is currently required to be loaded and running to even use an app that does no GUI stuff at all. This is unnecessary overhead. The poster did not imply that one needs to use the windows system's messaging facilities to do IPC or networking.
--
"Where, where is the town? Now, it's nothing but flowers!"
I think you'll find that it is also required for backward-compatibility with all non-Windows 2000 Microsoft operating systems, too. I doubt that retaining compatibility with Samba (which runs on a hell of a lot more operating systems than Linux, remember) was a major consideration when Active Directory was designed. Please don't blame Samba for AD's design.
This was done purely for backward compatibility reasons. It is not an efficiency measure; they could easily have keys like "rmalda.sales" and "rmalda.geeks" to avoid collisions for users in the same domain but different containers. In fact, containers in AD are purely eye-candy. They're not security principals; only users and groups are.
An AD domain could only hold significantly fewer than 23 trillion users, even given unlimited memory, disk and network bandwidth. You can't take the largest number AD can use as an index and call that the maximum capacity.
--
"Where, where is the town? Now, it's nothing but flowers!"
Windows 2000's Active Directory is a candy-coated shell over a reworking of the old NT4-style domain system.
An example:
Suppose you create an AD domain called "slashdot.org". You then create two OUs, so you have sales.slashdot.org and geeks.slashdot.org. Being a hardcore geek, Rob Malda will of course be rmalda in the OU geeks.slashdot.org. His username in the underlying domain datanase will also be rmalda.
When Richard Malda join sales, of course you will want to add a user rmalda to sales.slashnet.org. But you can't. Why? Because even if you have objects in separate OUs, their names must still be uniqiue in the domain, because the underlying domain database has a flat namespace.
--
"Where, where is the town? Now, it's nothing but flowers!"
RealPlayer is an application, not a kernel-level driver. If an application locks your operating system hard, your operating system has a problem.
--
"Where, where is the town? Now, it's nothing but flowers!"
> Calls to the windowing system for IPC? What
> about named pipes? Mailslots? or TCP/IP?
> Or shared memory?
All of those features are accessed using calls to the Win32 API (bar TCP/IP, which is accessed via the WinSock libraries). Therefore, the Win32 subsystem has to be running for those facilities to be available to the application programmer.
--
"Where, where is the town? Now, it's nothing but flowers!"
The problem is that anyone can add an ADK to a public key without affecting the key's fingerprint. In other words, it is perfectly possible for someone to set up a keyserver that adds an ADK for themselves to each key uploaded, and no-one will by any the wiser, unless they examine the key closely.
How they get their hands on the mail encrypted using those keys is of course outside the scope of this post.
Idea: A company could set up an internal auto-ADK-adding keyserver for its employees to use, and of course they have access to the outgoing mail spool.
--
"Where, where is the town? Now, it's nothing but flowers!"
From what I can make of the section regarding GnuPG, it doesn't warn about the presence of the ADK. However, it places but one session key in the cryptogram, a key only recoverable using the user's private key.
But if you get a contaminated version-4 public key, GnuPG will not warn you about it. You should check any and all public keys that you use as decribed in the article. I'm sure the GnuPG team will not be long in adding functionality to do this automatically.
--
"Where, where is the town? Now, it's nothing but flowers!"
Wow! More wilful stupidity!
Just because you want it now does not mean Mozilla is late. I am not the only person who is prepared wait for something good.
--
"Where, where is the town? Now, it's nothing but flowers!"
This isn't insightful, it's stupid. The phrase "standards-compliant" refers to W3C standards, such as XML and CSS.
By the way, Mozilla *does* support proxies.
--
"Where, where is the town? Now, it's nothing but flowers!"
Possibly the editors are aiming for maximum bottom-feeder no-think-just-post reactions (more comments, more banner views, yes?). The title you gave your submission just didn't have that. "IBM Kills Monterey" seemingly does.
--
"Where, where is the town? Now, it's nothing but flowers!"
No.
Looking at the headers available at http://sources.redhat.com/inti/inti-manual/ I see that each class member representing a signal is a SignalProxy, which seems to be quite like the signalling facilities provided by libsigc++. Looking at the definitition of SignalProxy, its connect method is parameterized for both the type of the signalled object and the member being connected to.
--
"Where, where is the town? Now, it's nothing but flowers!"
I had a quick look at the screen shot for the import/export tool. Helix has struck a balance between providing a usable interface and not concealing too many details.
All the information a power users needs to map from the GUI to what happens 'under the hood' is present, and yet a newbie could use that same tool to connect to whatever network resources he/she desires without difficulty.
--
"Where, where is the town? Now, it's nothing but flowers!"
"More subtle" is perfecly valid English. The construction works better for some words than other; it's fine in this case.
--
"Where, where is the town? Now, it's nothing but flowers!"
A facetious statement at best. Application programmers have to rely on large bodies of code over which they have little or no control, e.g. MFC and other class libraries/frameworks.
And to say that a poor OS encourages good application engineering is silly, to say the least. An OS with memory protection, such as Windows NT, Unix or Mac OS X, makes it much more practical and efficient to debug applictions suffering from stray pointers, etc.
--
"Where, where is the town? Now, it's nothing but flowers!"
The first sentence I quoted, which you omitted in order to support your flawed retort, states:
Now, operating system means operating system. If he meant GUI, he should have said GUI.
--
"Where, where is the town? Now, it's nothing but flowers!"
Not a good start to the article. The operating system that Apple developed is poor at best; a single application crash almost always brings down the machine.
The interface, on the other hand, feels solid and reliable; serious thought went into every part of it.
Nice GUI, shame about the OS.
--
"Where, where is the town? Now, it's nothing but flowers!"
All good points. My point is that XML is not special or unique. It's like Java; it has its uses.
--
"Where, where is the town? Now, it's nothing but flowers!"
In Europe, reverse-engineering is legal where it is required for interoperability and compatibility. We don't all live in the US, you know.
--
"Where, where is the town? Now, it's nothing but flowers!"
Some random observations:
My comment regarding AI was aimed more at the program being able to parse previously unknown data formats (even if they do use XML syntax) to extract the required information. Doing the actual comparisons, etc., would not be beyond us even today, I think, assuming that the program was fed the data is a well-defined format.
An XML schema is not the equivalent of an SGML DTD. XML is simply SGML with restrictions on tag minimization (<em/foo/ versus <em>foo</em>), case sensitivity and some other stuff I forget. XML and SGML DTDs are pretty much alike. XML Schemas are indeed marked-up using an XML DTD.
Currency conversion is rife with special cases. A case in point is converting between currencies in the Euro zone. You don't. What you must do is convert from currency A to Euros, and then from Euro's to currency B.
--
"Where, where is the town? Now, it's nothing but flowers!"
Risc OS does it better because it anti-aliases the small fonts, and not the big ones. Windows has it backwards.
--
"Where, where is the town? Now, it's nothing but flowers!"
I think you'll find that Windows has been anti-aliasing fonts for at most 5 years. The feature was introduced in the "Plus!" pack for Windows 95.
Acorn's Risc OS, on the other hand, has had anti-aliased outline fonts since approximately 1990.
X apps can anti-alias fonts on the client side; the fact that server-side support is missing is not the unmitigated disaster it appears to be.
--
"Where, where is the town? Now, it's nothing but flowers!"
For those who don't understand, CVS is the Concurrent Versioning System, a program for code change and change conflict management.
--
"Where, where is the town? Now, it's nothing but flowers!"
The DTD does not specify "meaning" for tags or their contents. The DTD simply specifies which tags can be nested and in what combinations.
To give meaning to the data, you need a schema; I believe that the W3C has proposals for describing XML schemas.
No, it would not. You would need sophisticated AI such as seen in popular movies to do that kind of parsing. Unless two programs agree on a DTD, a schema and what the schema means in relation to the real world, they cannot interoperate.
--
"Where, where is the town? Now, it's nothing but flowers!"
Judging by the lack of replies, I would imagine so.
--
"Where, where is the town? Now, it's nothing but flowers!"
--
"Where, where is the town? Now, it's nothing but flowers!"