you sure thats all you have open? willing to share your IP address with us all to check?
I find on linux you tend to have more need for a firewall. Linux will often be running RPC, and like you say X (and I know at least KDE) use ports too that should be firewalled.
I have to agree with the person you're commenting to.
Firewalls are not useful for an individual system if you don't have things running on ports that can be abused.
Windows does make this very hard, while Linux it is trivial and by default has less running. Turning off the extras is a chore under Windows and can be futile if you fire up an app that opens up a port on demand -- and you don't know about it.
Right now, I have plenty of Linux games to play, so I'm not playing Windows games under WineX.
The only thing that annoys me about WineX is that as the libs get updated for the distribution of Linux I use, WineX no longer works. If I'm not subscribed to WineX, I can no longer play the games. (I don't dual boot...too much of a bother.)
While my stack of native Linux games is a few times higher, I have sucessfully played a quite a few games using WineX (from memory); Diablo II (before Wine could), Civ. III, Black & White, Alice, Heavy Metal FAKK (bargin bin: hey, $5 is $5). There are a couple more, but I don't remember them.
Before buying any of these, I did check the list not only for the ratings but for comments. Civ. III was a bit of a pain to deal with, though worth it for the number of hours of play it provided.
Now that I'm playing Savage so much, I don't see a need for a subscription. If Civ. IV/V/... comes out, I'll consider it again for sure.
Are MAC gamers paying these assholes 5$/mnth for MAC games?
No, they are paying about $30-60 for a one-time payment for a single game. In WineX terms, that's 6-12 months -- OK, on sale $20 or 4 months.
For one, you're paying for a single game.
For the other, you're paying for the ability to play multiple games.
The two aren't similar.
While I'm not currently a WineX subscriber -- Savage is too fun all by itself -- I can easily see joining up again for a 1 year subscription. $60/year is trivial if you play alot of games and want to use Linux exclusively.
Calling Transgaming names even when they spell out what works and how well seems a bit strained.
I'd highly recommend you at least head over to Groklaw and read the text transcript of Professor Moglen's Harvard speech. It will open your eyes.
In my other comment, I mentioned that I skimmed the Profs speach. Well, it turns out I lied -- or was simply wrong because it was unintentional. I did read the whole thing, and I do agree with what he said if not in exactly the same spirit.
Even after re-reading the speach, I did not see a section where he said that SCO was actually in violation of the GPL.
SCO said they did not respect the GPL, called it nasty names, and then went on to abide by it. This seems to be an improvement over how Linksys acted a few weeks/months ago.
If you can find a section of the speach or any other part where he says that SCO is actually in violation of the GPL -- not just talking as if they are ignoring the GPL -- I'd like to see it.
When they stated that bunch of gobbledygook, that is *precisely* when they violated the GPL. By not holding it up as the perfectly good license for software that is free/libre, and being unwilling to abide by its terms, they have violated the license under paragraph four of the license. They are no longer able to distribute, copy, modify, etc, any of the Linux/GNU software, or any other software licensed under the GPL.
Would that hold up in a courtroom? Understand, I'll be in the crowds that cheer when SCO gets what they really deserve and all this sillyness goes away. If a pack of rabbid dogs stalks them, who am I to complain?...yet, saying and doing are two different things.
SCO has made public statements that don't agree with the GPL and don't think it is legitimate. A reasonable person could (if I'm reasonable!) say that they are acting as if the GPL were legitimate even if they say otherwise.
A similar example (though I admit this is a big stretch): A wife and husband argue, and one says "I'm going to kill you!".
If the guilty spouse then acts on that statement by poisoning the target spouse, they have commited murder.
If they don't act, they haven't done anything illegal (barring restraining orders, etc...).
If the guilty spouse never said they were going to kill the other one, yet they do indeed kill by poisioning the target spouse, they have commited murder.
#3 isn't an issue. #2 and #1 could apply; mostly a #2 though SCO's actions could be a serious attempt at #1.
I'd highly recommend you at least head over to Groklaw and read the text transcript of Professor Moglen's Harvard speech. It will open your eyes.
I'll read it all this time, though I didn't see this the first time round when I skimmed.
Q. Combination lost and Unit Locked. How Do I remove the DEFCON 1 if the device can not be unlocked?
A. If the battery is installed, then you will first want to disable its electronic alarm.
Break open the battery compartment to remove the battery
Cut the cable with a wire cutter (note: an ordinary paper cutter will not do); if the battery is still in the unit when the cable is cut, the alarm will sound continuously.
Nope. SCO believe that all GPL'd software is really public domain, since the GPL is (allegedly) unconstitutional. And they think that, by getting the GPL ruled invalid, suddenly all this code will become public domain and fair game for anyone who wants it.
Is SCO corporation actually out of compliance with the GPL, or are they just talking as if they don't actually comply?
I would't shed a tear if SCO were metaphorically torn to bits and beaten into the ground with the bloody bits that remain...though I'm straining to see the wish backed by facts.
By attempting to get users of Linux to cough up fees for SCO's extra license for Linux, GNU, and Fyodor's NMAP program as one of many other GPL'd program distributed with Linux and SCO's own product offerings, SCO is in violation of the GPL. Fyodor is well within his rights to terminate SCO's license to use the software. Once upon a time, Caldera very happily distributed GNU/Linux and Nmap for free, with all the conditions intact to remain in compliance with the GPL. By attempting to charge an additional licensing fee and dictating conditions that are not in line with the GPL's own, they are in violation.
Clear as mud, right??
If SCO were suing for supposed improper tech in NMAP, I'd agree.
SCO is suing IBM over a contract issue that they claim involves code that ended up in Linux (the kernel).
In Fydor's case, he can't use your reasons to terminate SCO's licence as long as SCO complies with the word of the text in regaurds to NMAP.
Note that part of the GPL specifically prohibits denying the licence for some users based on preference. Other licences allow this type of discrimination, though the GPL does not.
Examples:
As a member of the NRA, you might want to have only NRA members use the code and not to hand it over to non-NRA members -- the GPL does not allow this.
As an anti-war protest, you attempt to disallow the use of the code by the military or specifically in weappons systems. The GPL does not allow this either.
Maybe I'm missing something...though a clarification/correction would be welcome.
We'll never get to this point if every time two people disagree they split the project. Project forks are good to an extent - but I think that we lose a lot more than we gain because of it. At least MS has a meeting and decides how to continue - we, the OSS community, just get pissed off and branch.
How exactly?
(Meaning: Are your comments related to "The Luxury of Ignorance", or are you raising a largely unrelated yet existing complaint?)
The older systems are growing more secure, because the virus writers are going after the newere ones.
Win9x;
Good: Less complex, so fewer places to exploit.
Bad: All programs run as 'root'/'administrator' and no architectural protections at all beyond the system crashing (intentionally) after a priv. operation has occured (usually a program bug).
WinNT/Win2000/WinXP/...;
Good: System enforces 'root'/'administrator' access.
Bad: Most configurations and users do not respect this separation.
While these are not complete reasons, they do cover the major areas.
Security has little to do with popularity or attention. Win9x can't be hardened, and many of the older attacks still work against it as it is actually used...so why bother inventing more?
The older versions of the Linux kernel seem to be alive, well, and still being patched for security flaws. In fact, the most recent kernel release is 2.2.26.
The reason for this is simple: Motivation.
Microsoft isn't motivated to patch software they are not making money on. They are motivated *not* to make changes since that can push users to upgrade. Since the code is closed, they are the only group that can act on this motivation effectively.
Linux 2.0.x and 2.2.x are maintained by people and corporations who use those kernels and are motivated to keep them secure. Since the code is open, anyone with this motivation can make corrections even if the changes are not widely distributed or placed in the main branch.
Why is this surprising? Most users use a computer as a toaster, and mostly by rote. Imagine if tommorrow morning your toaster has a new lever on it labelled "PBT 1.0". What is PBT? It is Perfect Brown Toast. But did you KNOW that?
Well, I'd hover my finger over the button for a second or two.
Doesn't anyone else find it extremely cumbersome and security error prone to allow processes to open listen ports as they wish? Isn't there an equivalent to ipfilter in the Windows kernel?
Agreed. That doesn't make any sense. While I know folks can add-on tools like Zone Alarm, not having a built-in configuration for this seems strange.
Along those lines though, the per-process/app/server block of ZA and other Windows firewalls could have some uses on Linux. I guess with SE Linux, that will come along for-free, though I don't know that for a fact. Anyone?
Over all, the 'Unix way' of doing things should be natural to you now. It is harder to do tasks under Windows using Unix methods, though you should be able to see where gaps are in the default Windows configuration and guess that there are Unix-style tools available to fill those holes. For the moment, try and avoid using those methods. You won't forget them, though...
2. Learn to live with Windows.
The best way to learn something is to force yourself to do it. So, even if you have a Linux box at home consider setting up a Windows system and use that exclusively for a couple months. (In all honesty, I don't do this, though I am moderately annoyed or comfortable at times using any OS...Windows, Linux, Mac....)
3. Do you really need admin access to your workstation? If so, make your case and present it to them. Do you need to test/install new applications? Do you need to run an application that only runs in admin mode?
Root/Administrator access should be a rare situation not the norm.
Also, remember that even on your typical Linux system having root access is not going to mean what it does now in a few years (if not sooner). SE Linux and other changes will change what root should be able to do for many distributions.
Some people at Microsoft do listen, you just have to make a bit of an effort to find them. Curiously, a comment from the developer in question was that the dev teams love direct contact with customers prepared to give them helpful information about bugs or feature requests, they just wish the PR people would stop getting in the way.:-)
...a perfect example between the difference in OSS and closed source.
I've worked both sides of the fence, and realize the differences. There are base motivations that drive each to do things differently. Still, I was stunned when I asked Theodore Tso a question in email a few years back, and he not only responded quickly but even sent a patch for me to try out!
I like your analogy, though most spreadsheet users can't program in VBA let alone more serious languages. Moving to a dedicated development environment is too scary for most of the VBA users.
If you do heavy VBA, though, switching to a better tool is a wise choice.
NTFS does support hardlinks easily via the command line but only to files. http://www.microsoft.com/technet/treeview/default. asp?url=/technet/prodtechnol/winxppro/proddocs/fsu til.asp [microsoft.com]
I still haven't found how to do softlinks but it is sort of pointless if you can only do it to files.
That's something that is being ignored by folks who say "Windows supports links!"; yes, but not very well. Another "technically correct, practically useless" feature. So close...just not quite right.
AFAIK, they use a custom versioning/simultaneous build/test tool. its not really reasonable to expect them to use source safe for the scope of their main projects...
Sounds pathetic! Good thing I don't have to use SS...I have to use, oh, nevermind, ClearCase. [walks away feeling depressed]
NT on NTFS supports both, and the documentation is not ambiguous.
OK, now use them without writing a program. What practical good are they the way MS has devised them?
Uhh... what? Explorer and related shell services are the only things that deal with.lnk files, and they have absolutely nothing to do with the underlying filesystem.
That's the PROBLEM! If you used the real thing, you'd see that.lnk files are a bad idea crudely implemented.
You don't use Windows much, do you.
I've used Windows including Beta and Alpha releases since 1.x. The registry -- for example -- is a good idea, though also poorly implemented. MS has some catchup work to do.
Others have already addressed your bad definitions of hard and soft links.
They did? In practical use, the support isn't there. Everything is an add-on, and not built in. (And yes, I've used the tools already mentioned...I'm the one who is attempting to get our admins to use them.)
In Nautilus, right click a file or folder and click "Make Link". "ls -l" confirms it: regular symbolic link.
Excellent!
Launchers are meant for, yes, "launching" a command. They need to be a file since the file needs to store infomation on the icon to use, some comment information, etc. A link has none of this information, so it's not a suitable replacement.
For programs, that makes some sense (but not much to me since program data can be handled as meta data seperately from the binary and a softlink could be used for the actual object).
For partitions and other parts of file systems, I can't understand why.desktop/launchers are used at all. Since all file systems are a mount point (Ex: anything mentioned in/etc/fstab; Floppy drive, CD, network mount points...), and directories are linkable (Ex: ~/home), a double link would make more sense for these.
Maybe I'm spoiled, though OS/2 (RIP) did do this one thing right and consistantly.
Actually, Windows does support softlinks, which are termed as junctions.
Yes, and they are a PITA to use. Can't execute a link over the network, for example.
To add to confusion,
Windows only supports softlinks -- not hardlinks -- yet refers to the junctions as hardlinks in many documents.
Windows does not support softlinks off of the current partition.
Definitions...
Softlink: A pointer to a file resource that acts like the target except when deleted. If deleted, the target is not also deleted.
Hardlink: A pointer to a file resource on the local partition. If removed, the target is also removed.
Windows has pseudo links, and even attempts to fake the.lnk files for some -- but not all -- utilities. This limited support is only available on NTFS since Windows doesn't by default support anything except NTFS for this type of work.
Gripe: KDE and Gnome do not allow creation of soft/hard links and use the dummy file method (like Windows) when setting up a 'link'. I've put in a request that this be changed in KDE, though what is needed for complete consistancy is OS-level support for various device types so it is transparent to the GUI.
Bottom line: I don't know if Windows 2000 and Windows XP can support what the original person commented on...so that's the reason why SourceSafe does the goofy work arounds not needed elsewhere.
I didn't want to make a big deal about this earler, though...I know about the event log, and use it since there's little else. By default, the logs are almost useless in comparison to the same type of logs on many Unix/Unix-like systems.
Having anything appear in the Windows logs seems to be a big deal since very few non-MS apps use the standard logs, let alone most MS apps and servers themselves.
For example, under Linux, each line in/var/log/messages* starts with something like this;
Feb 20 23:28:10 systemname.domain.url
The details that complete the line run from the basic facts to the obscure (but handy);
cups: cupsd shutdown succeeded
kernel: cdrom: This disc doesn't have any tracks I recognize!
last message repeated 4 times
cups: cupsd startup succeeded
su(pam_unix)[10412]: session opened for user root by ausername(uid=534)
gconfd (ausername-19685): GConf server is not in use, shutting down.
gconfd (ausername-19685): Exiting
kernel: ksysguardd: numerical sysctl 7 2 1 is obsolete.
If I search the net for these, I'll find something that describes what each does. For Windows many fewer messages are saved and they are seldom mentioned in online resources. When they are, the notes are not very informative. Note that this is a general observation and I have seen counter examples, though the quantity and quality favors Unix overwhelmingly.
Most every subprocess has this type of coverage under a Unix/Unix-like OS, not just Linux. In Windows, most don't. Even when the apps do, it's difficult to check what user had what process if they don't currently own it...and even then, it's not nearly as easy as Unix.
I've spent days attempting to figure out something -- usually a resource related problem -- that "just happened" under Windows but not while I was staring at the process meters. Windows does make it easy to figure out what's happening at the current moment, but not any better than top.
Anoyances with both: Top shows shared memory (so you have to be aware of duplicates). Windows shows a confusing list of what general system memory is in use; it's not obvious, in fact it seems to fudge the truth on a regular basis.
My background includes memory diagnostics tools designed for Windows-only and system ananalysis tools, though I've seen others complain about the same things I see.
Slashdot Mirror
I find on linux you tend to have more need for a firewall. Linux will often be running RPC, and like you say X (and I know at least KDE) use ports too that should be firewalled.
I have to agree with the person you're commenting to.
Firewalls are not useful for an individual system if you don't have things running on ports that can be abused.
Windows does make this very hard, while Linux it is trivial and by default has less running. Turning off the extras is a chore under Windows and can be futile if you fire up an app that opens up a port on demand -- and you don't know about it.
The only thing that annoys me about WineX is that as the libs get updated for the distribution of Linux I use, WineX no longer works. If I'm not subscribed to WineX, I can no longer play the games. (I don't dual boot...too much of a bother.)
While my stack of native Linux games is a few times higher, I have sucessfully played a quite a few games using WineX (from memory); Diablo II (before Wine could), Civ. III, Black & White, Alice, Heavy Metal FAKK (bargin bin: hey, $5 is $5). There are a couple more, but I don't remember them.
Before buying any of these, I did check the list not only for the ratings but for comments. Civ. III was a bit of a pain to deal with, though worth it for the number of hours of play it provided.
Now that I'm playing Savage so much, I don't see a need for a subscription. If Civ. IV/V/... comes out, I'll consider it again for sure.
No, they are paying about $30-60 for a one-time payment for a single game. In WineX terms, that's 6-12 months -- OK, on sale $20 or 4 months.
For one, you're paying for a single game.
For the other, you're paying for the ability to play multiple games.
The two aren't similar.
While I'm not currently a WineX subscriber -- Savage is too fun all by itself -- I can easily see joining up again for a 1 year subscription. $60/year is trivial if you play alot of games and want to use Linux exclusively.
Calling Transgaming names even when they spell out what works and how well seems a bit strained.
In my other comment, I mentioned that I skimmed the Profs speach. Well, it turns out I lied -- or was simply wrong because it was unintentional. I did read the whole thing, and I do agree with what he said if not in exactly the same spirit.
Even after re-reading the speach, I did not see a section where he said that SCO was actually in violation of the GPL.
SCO said they did not respect the GPL, called it nasty names, and then went on to abide by it. This seems to be an improvement over how Linksys acted a few weeks/months ago.
If you can find a section of the speach or any other part where he says that SCO is actually in violation of the GPL -- not just talking as if they are ignoring the GPL -- I'd like to see it.
(really!)
Would that hold up in a courtroom? Understand, I'll be in the crowds that cheer when SCO gets what they really deserve and all this sillyness goes away. If a pack of rabbid dogs stalks them, who am I to complain? ...yet, saying and doing are two different things.
SCO has made public statements that don't agree with the GPL and don't think it is legitimate. A reasonable person could (if I'm reasonable!) say that they are acting as if the GPL were legitimate even if they say otherwise.
A similar example (though I admit this is a big stretch): A wife and husband argue, and one says "I'm going to kill you!".
#3 isn't an issue. #2 and #1 could apply; mostly a #2 though SCO's actions could be a serious attempt at #1.
I'll read it all this time, though I didn't see this the first time round when I skimmed.
If the cover is broken, though, the whole gadget is useless, right? For $50, I'm not complaining. Anyone should be able to remember a 4 digit code.
A. If the battery is installed, then you will first want to disable its electronic alarm.
Break open the battery compartment to remove the battery
Cut the cable with a wire cutter (note: an ordinary paper cutter will not do); if the battery is still in the unit when the cable is cut, the alarm will sound continuously.
Yes, but is saying they don't accept it enough if they actually comply with the licence in practice?
Is SCO corporation actually out of compliance with the GPL, or are they just talking as if they don't actually comply?
I would't shed a tear if SCO were metaphorically torn to bits and beaten into the ground with the bloody bits that remain...though I'm straining to see the wish backed by facts.
Details and corrections appreciated!
Clear as mud, right??
If SCO were suing for supposed improper tech in NMAP, I'd agree.
SCO is suing IBM over a contract issue that they claim involves code that ended up in Linux (the kernel).
In Fydor's case, he can't use your reasons to terminate SCO's licence as long as SCO complies with the word of the text in regaurds to NMAP.
Note that part of the GPL specifically prohibits denying the licence for some users based on preference. Other licences allow this type of discrimination, though the GPL does not.
Examples:
Maybe I'm missing something...though a clarification/correction would be welcome.
How exactly?
(Meaning: Are your comments related to "The Luxury of Ignorance", or are you raising a largely unrelated yet existing complaint?)
Win9x;
Good: Less complex, so fewer places to exploit.
Bad: All programs run as 'root'/'administrator' and no architectural protections at all beyond the system crashing (intentionally) after a priv. operation has occured (usually a program bug).
WinNT/Win2000/WinXP/...;
Good: System enforces 'root'/'administrator' access.
Bad: Most configurations and users do not respect this separation.
While these are not complete reasons, they do cover the major areas.
Security has little to do with popularity or attention. Win9x can't be hardened, and many of the older attacks still work against it as it is actually used...so why bother inventing more?
The reason for this is simple: Motivation.
Microsoft isn't motivated to patch software they are not making money on. They are motivated *not* to make changes since that can push users to upgrade. Since the code is closed, they are the only group that can act on this motivation effectively.
Linux 2.0.x and 2.2.x are maintained by people and corporations who use those kernels and are motivated to keep them secure. Since the code is open, anyone with this motivation can make corrections even if the changes are not widely distributed or placed in the main branch.
Well, I'd hover my finger over the button for a second or two.
Agreed. That doesn't make any sense. While I know folks can add-on tools like Zone Alarm, not having a built-in configuration for this seems strange.
Along those lines though, the per-process/app/server block of ZA and other Windows firewalls could have some uses on Linux. I guess with SE Linux, that will come along for-free, though I don't know that for a fact. Anyone?
Over all, the 'Unix way' of doing things should be natural to you now. It is harder to do tasks under Windows using Unix methods, though you should be able to see where gaps are in the default Windows configuration and guess that there are Unix-style tools available to fill those holes. For the moment, try and avoid using those methods. You won't forget them, though...
The best way to learn something is to force yourself to do it. So, even if you have a Linux box at home consider setting up a Windows system and use that exclusively for a couple months. (In all honesty, I don't do this, though I am moderately annoyed or comfortable at times using any OS...Windows, Linux, Mac....)
Root/Administrator access should be a rare situation not the norm.
Also, remember that even on your typical Linux system having root access is not going to mean what it does now in a few years (if not sooner). SE Linux and other changes will change what root should be able to do for many distributions.
...a perfect example between the difference in OSS and closed source.
I've worked both sides of the fence, and realize the differences. There are base motivations that drive each to do things differently. Still, I was stunned when I asked Theodore Tso a question in email a few years back, and he not only responded quickly but even sent a patch for me to try out!
If you do heavy VBA, though, switching to a better tool is a wise choice.
I still haven't found how to do softlinks but it is sort of pointless if you can only do it to files.
That's something that is being ignored by folks who say "Windows supports links!"; yes, but not very well. Another "technically correct, practically useless" feature. So close...just not quite right.
Sounds pathetic! Good thing I don't have to use SS...I have to use, oh, nevermind, ClearCase. [walks away feeling depressed]
It doesn't work over the network. You have to remote connect to the target machine using VNC/PC Anywhere/TS and bring up a 'local' file browser.
Handy for 1 machine, impractical as a method for managing a network (when dealing with admins who don't quite get it and just see buttons to click).
OK, now use them without writing a program. What practical good are they the way MS has devised them?
That's the PROBLEM! If you used the real thing, you'd see that .lnk files are a bad idea crudely implemented.
I've used Windows including Beta and Alpha releases since 1.x. The registry -- for example -- is a good idea, though also poorly implemented. MS has some catchup work to do.
They did? In practical use, the support isn't there. Everything is an add-on, and not built in. (And yes, I've used the tools already mentioned...I'm the one who is attempting to get our admins to use them.)
Excellent!
For programs, that makes some sense (but not much to me since program data can be handled as meta data seperately from the binary and a softlink could be used for the actual object).
For partitions and other parts of file systems, I can't understand why .desktop/launchers are used at all. Since all file systems are a mount point (Ex: anything mentioned in /etc/fstab; Floppy drive, CD, network mount points...), and directories are linkable (Ex: ~/home), a double link would make more sense for these.
Maybe I'm spoiled, though OS/2 (RIP) did do this one thing right and consistantly.
Yes, and they are a PITA to use. Can't execute a link over the network, for example.
To add to confusion,
Windows only supports softlinks -- not hardlinks -- yet refers to the junctions as hardlinks in many documents.
Windows does not support softlinks off of the current partition.
Definitions...
Softlink: A pointer to a file resource that acts like the target except when deleted. If deleted, the target is not also deleted.
Hardlink: A pointer to a file resource on the local partition. If removed, the target is also removed.
Windows has pseudo links, and even attempts to fake the .lnk files for some -- but not all -- utilities. This limited support is only available on NTFS since Windows doesn't by default support anything except NTFS for this type of work.
Gripe: KDE and Gnome do not allow creation of soft/hard links and use the dummy file method (like Windows) when setting up a 'link'. I've put in a request that this be changed in KDE, though what is needed for complete consistancy is OS-level support for various device types so it is transparent to the GUI.
Bottom line: I don't know if Windows 2000 and Windows XP can support what the original person commented on...so that's the reason why SourceSafe does the goofy work arounds not needed elsewhere.
Having anything appear in the Windows logs seems to be a big deal since very few non-MS apps use the standard logs, let alone most MS apps and servers themselves.
For example, under Linux, each line in /var/log/messages* starts with something like this;
The details that complete the line run from the basic facts to the obscure (but handy);
kernel: cdrom: This disc doesn't have any tracks I recognize!
last message repeated 4 times
cups: cupsd startup succeeded
su(pam_unix)[10412]: session opened for user root by ausername(uid=534)
gconfd (ausername-19685): GConf server is not in use, shutting down.
gconfd (ausername-19685): Exiting
kernel: ksysguardd: numerical sysctl 7 2 1 is obsolete.
If I search the net for these, I'll find something that describes what each does. For Windows many fewer messages are saved and they are seldom mentioned in online resources. When they are, the notes are not very informative. Note that this is a general observation and I have seen counter examples, though the quantity and quality favors Unix overwhelmingly.
Most every subprocess has this type of coverage under a Unix/Unix-like OS, not just Linux. In Windows, most don't. Even when the apps do, it's difficult to check what user had what process if they don't currently own it...and even then, it's not nearly as easy as Unix.
I've spent days attempting to figure out something -- usually a resource related problem -- that "just happened" under Windows but not while I was staring at the process meters. Windows does make it easy to figure out what's happening at the current moment, but not any better than top.
Anoyances with both: Top shows shared memory (so you have to be aware of duplicates). Windows shows a confusing list of what general system memory is in use; it's not obvious, in fact it seems to fudge the truth on a regular basis.
My background includes memory diagnostics tools designed for Windows-only and system ananalysis tools, though I've seen others complain about the same things I see.