[OT] Arrogant filters

How can we punish filter driver developers who fail unknown IOCTLS and
FSCTLS rather than passing them down to let others have a stab at them? I am
thinking in terms of some sort of excommunication. Or maybe some sort of
award of Hubris; presented to the company that is shipping such a driver.

Some folks have no business writing code for the real world. You would think
that the word “FILTER” would trigger the correct synapse… A little
abstraction… Oh well, maybe I am just dreaming.

Jamey Kirby, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

Easy - Microsoft can add code to Windows to force all drivers to pass
some sort of mini-WHQL test on the fly before allowing them to be installed.

Jamey Kirby wrote:

How can we punish filter driver developers who fail unknown IOCTLS and
FSCTLS rather than passing them down to let others have a stab at them? I am
thinking in terms of some sort of excommunication. Or maybe some sort of
award of Hubris; presented to the company that is shipping such a driver.

Some folks have no business writing code for the real world. You would think
that the word “FILTER” would trigger the correct synapse… A little
abstraction… Oh well, maybe I am just dreaming.

Jamey Kirby, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

Of course, if they’re signed, they’ve already gone through WHQL, but too
often I’ve downloaded the most prominently displayed updates from driver
vendors on the web to find out I have to click through the ‘unsigned
driver’ dialog to get them on my machine. Clicking through that dialog
is probably a reflex now for most users.

Nick Ryan wrote:

Easy - Microsoft can add code to Windows to force all drivers to pass
some sort of mini-WHQL test on the fly before allowing them to be
installed.

Jamey Kirby wrote:

> How can we punish filter driver developers who fail unknown IOCTLS and
> FSCTLS rather than passing them down to let others have a stab at
> them? I am
> thinking in terms of some sort of excommunication. Or maybe some sort of
> award of Hubris; presented to the company that is shipping such a driver.
>
> Some folks have no business writing code for the real world. You would
> think
> that the word “FILTER” would trigger the correct synapse… A little
> abstraction… Oh well, maybe I am just dreaming.
> Jamey Kirby, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
>
>
>
>
>
>
>
>

Jamey, do you believe in the tooth fairy too? To most of them the only
“trigger” they know is the horse. Maybe I am giving them too much credit.
There are a few really talented file systems filter driver writers and the
masses whose talent lies in areas other than computer software design and
development. How many times have you seen a post from some FSF writer who
said they put in the hooks to handle IRP_MJ_CREATE and wonder why their
development system won’t boot anymore. No FastIo routines or read or write
or even passthru. They also think they can save the fileobject and
FsContext in the dispatch and have it usable. Add that to the “advanced”
writers who say they want to open the file themselves in the dispatch
function and then decide if they should allow real create to proceed. Then
they complain that they see another create for the same file and can’t
figure out what to do with it. Maybe they should just keep trying to create
the create until they get the pretty blue screen.

How about those who think that the old sources for filemon are a good
starting place for an active FSF? ASAIK, there are no good examples of
active FSFs available for free. Those of us who have done them, don’t give
that much work and pain away for free. BTW, the examples in the IFS Kit are
passive filters. Another good one is those who want to change the name and
have it automagically work on the correct volume. Maybe we should ask
Microsoft to just handle it if it is same file system doing the processing.
Should make the Microsoft code very interesting in trying to share device
extensions. Should do wonders for security too.

Would someone stop the world, I wanna get off. If you are into pain run
several of the antivirus products on the same system. Somebody better hurry
up and invent Isaac Asimov’s positronic brain robot.

“Jamey Kirby” wrote in message news:xxxxx@ntdev…

How can we punish filter driver developers who fail unknown IOCTLS and
FSCTLS rather than passing them down to let others have a stab at them? I am
thinking in terms of some sort of excommunication. Or maybe some sort of
award of Hubris; presented to the company that is shipping such a driver.

Some folks have no business writing code for the real world. You would think
that the word “FILTER” would trigger the correct synapse… A little
abstraction… Oh well, maybe I am just dreaming.

Jamey Kirby, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com

Then it better send a fairly random ioctl down, or filter writers will write
code to pass that one specific ioctl. Come to think of it, the sequence has
to vary too, so that they don’t write heruistics to “pass the first 3
requests after a start is received” or some such.

Loren

----- Original Message -----
From: “Nick Ryan”
Newsgroups: ntdev
To: “Windows System Software Developers Interest List”
Sent: Thursday, August 14, 2003 3:54 PM
Subject: [ntdev] Re: [OT] Arrogant filters

> Easy - Microsoft can add code to Windows to force all drivers to pass
> some sort of mini-WHQL test on the fly before allowing them to be
installed.
>

Well, the focus has to be on blocking the clueless developer from
getting his code on the system. The malicious developer is always going
to find a way to game the system. Of course, the line is blurred when
you have a developer sufficiently non-clueless to do whatever is
required to pass the test but still clueless enough to realize what the
test is trying to TEST (i.e. the presence of a general solution to a
problem).

Loren Wilton wrote:

Then it better send a fairly random ioctl down, or filter writers will write
code to pass that one specific ioctl. Come to think of it, the sequence has
to vary too, so that they don’t write heruistics to “pass the first 3
requests after a start is received” or some such.

Loren

----- Original Message -----
From: “Nick Ryan”
> Newsgroups: ntdev
> To: “Windows System Software Developers Interest List”
> Sent: Thursday, August 14, 2003 3:54 PM
> Subject: [ntdev] Re: [OT] Arrogant filters
>
>
>
>>Easy - Microsoft can add code to Windows to force all drivers to pass
>>some sort of mini-WHQL test on the fly before allowing them to be
>
> installed.
>
>
>
>
>

I was concerned about Igor Clueless that says “why does the system refuse to
load my driver?” and Emil Moreclueless replying “you are supposed to pass
the first 4 ioctls down to the test driver, you have to do this by counting,
because for unknown reason the ioctls aren’t always the same and you can’t
check for specific code to pass”. Keep in mind that Igor and Emil may not
have their conversation on this mailing list where someone could object…

Loren

----- Original Message -----
From: “Nick Ryan”
Newsgroups: ntdev
To: “Windows System Software Developers Interest List”
Sent: Thursday, August 14, 2003 5:09 PM
Subject: [ntdev] Re: [OT] Arrogant filters

> Well, the focus has to be on blocking the clueless developer from
> getting his code on the system. The malicious developer is always going
> to find a way to game the system. Of course, the line is blurred when
> you have a developer sufficiently non-clueless to do whatever is
> required to pass the test but still clueless enough to realize what the
> test is trying to TEST (i.e. the presence of a general solution to a
> problem).
>
> Loren Wilton wrote:
>
> > Then it better send a fairly random ioctl down, or filter writers will
write
> > code to pass that one specific ioctl. Come to think of it, the sequence
has
> > to vary too, so that they don’t write heruistics to “pass the first 3
> > requests after a start is received” or some such.
> >
> > Loren
> >
> > ----- Original Message -----
> > From: “Nick Ryan”
> > Newsgroups: ntdev
> > To: “Windows System Software Developers Interest List”

> > Sent: Thursday, August 14, 2003 3:54 PM
> > Subject: [ntdev] Re: [OT] Arrogant filters
> >
> >
> >
> >>Easy - Microsoft can add code to Windows to force all drivers to pass
> >>some sort of mini-WHQL test on the fly before allowing them to be
> >
> > installed.

Jamey Kirby wrote:

How can we punish filter driver developers who fail unknown IOCTLS and
FSCTLS rather than passing them down to let others have a stab at them? I am
thinking in terms of some sort of excommunication. Or maybe some sort of
award of Hubris; presented to the company that is shipping such a driver.

I think this award is a great idea. I also want to nominate:

(1) Companies that ship debug binaries full of spew that gets in my way
when I’m trying to debug *my* driver. Second prize goes to companies
that ship debug user-mode apps and DLLs.

(2) USB function drivers that don’t pass internal IOCTLs down the stack,
which cuts off upper filters from the device.

(3) Any driver that forwards IRPs to the PDO instead of the immediately
lower device object. Microsoft’s own RNDIS driver did this, at least
during beta, which made it pretty hard to put a lower filter in place to
work around a firmware bug.

(4) Filters that queue IRPs and then present them in a different thread
context, possibly breaking reasonable assumptions by the function driver
writer about METHOD_NEITHER requests.


Walter Oney, Consulting and Training
Basic and Advanced Driver Programming Seminars
Check out our schedule at http://www.oneysoft.com

Walter,

You forgot shipping a driver that doesn’t pass the driver verifier.
Of course that means we would need a lot of space to post the winners.

Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting

Jamey Kirby wrote:
> How can we punish filter driver developers who fail unknown IOCTLS and
> FSCTLS rather than passing them down to let others have a stab at them? I am
> thinking in terms of some sort of excommunication. Or maybe some sort of
> award of Hubris; presented to the company that is shipping such a driver.

I think this award is a great idea. I also want to nominate:

(1) Companies that ship debug binaries full of spew that gets in my way
when I’m trying to debug *my* driver. Second prize goes to companies
that ship debug user-mode apps and DLLs.

(2) USB function drivers that don’t pass internal IOCTLs down the stack,
which cuts off upper filters from the device.

(3) Any driver that forwards IRPs to the PDO instead of the immediately
lower device object. Microsoft’s own RNDIS driver did this, at least
during beta, which made it pretty hard to put a lower filter in place to
work around a firmware bug.

(4) Filters that queue IRPs and then present them in a different thread
context, possibly breaking reasonable assumptions by the function driver
writer about METHOD_NEITHER requests.


Walter Oney, Consulting and Training
Basic and Advanced Driver Programming Seminars
Check out our schedule at http://www.oneysoft.com

Sometimes that’s the exact effect you want to achieve, no ? :slight_smile:

Alberto.

-----Original Message-----
From: Jamey Kirby [mailto:xxxxx@storagecraft.com]
Sent: Thursday, August 14, 2003 6:40 PM
To: Windows System Software Developers Interest List
Subject: [ntdev] [OT] Arrogant filters

How can we punish filter driver developers who fail unknown IOCTLS and
FSCTLS rather than passing them down to let others have a stab at them? I am
thinking in terms of some sort of excommunication. Or maybe some sort of
award of Hubris; presented to the company that is shipping such a driver.

Some folks have no business writing code for the real world. You would think
that the word “FILTER” would trigger the correct synapse… A little
abstraction… Oh well, maybe I am just dreaming.

Jamey Kirby, Windows DDK MVP
StorageCraft Corporation
xxxxx@storagecraft.com
http://www.storagecraft.com


Questions? First check the Kernel Driver FAQ at
http://www.osronline.com/article.cfm?id=256

You are currently subscribed to ntdev as: xxxxx@compuware.com
To unsubscribe send a blank email to xxxxx@lists.osr.com

The contents of this e-mail are intended for the named addressee only. It
contains information that may be confidential. Unless you are the named
addressee or an authorized designee, you may not copy or use it, or disclose
it to anyone else. If you received it in error please notify us immediately
and then destroy it.

How are you doing to debug your own drivers that do not pass any tests YET
on this system if they will fail to load? ))

I guess there is no solution that will prevent anybody with the root
rights from loading kernel module.

Anton Kolomyeytsev

Easy - Microsoft can add code to Windows to force all drivers to pass
some sort of mini-WHQL test on the fly before allowing them to be installed.

Jamey Kirby wrote:

> How can we punish filter driver developers who fail unknown IOCTLS and
> FSCTLS rather than passing them down to let others have a stab at them? I am
> thinking in terms of some sort of excommunication. Or maybe some sort of
> award of Hubris; presented to the company that is shipping such a driver.
>
> Some folks have no business writing code for the real world. You would think
> that the word “FILTER” would trigger the correct synapse… A little
> abstraction… Oh well, maybe I am just dreaming.
>
> Jamey Kirby, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
>
>
>
>
>
>
>
>

>ASAIK, there are no good examples of

active FSFs available for free. Those of us who have done them, don’t give
that much work and pain away for free. BTW, the examples in the IFS Kit
are
passive filters.

There are no good examples because Microsoft probably doesn’t have the time
and/or inclination to produce such. No one else can because the licensing
restrictions of the IFS kit STRICTLY forbid such. You couldn’t possibly
post a free sample driver that uses ntifs.h to any public location, even if
you wanted to, without breaking several provisions in the IFS kit licensing
agreement.

AND, it sounds like you are saying not all is well in file system land.
*gasp*

Hmmm, still gotta wonder then about those licensing restrictions on the IFS
kit. Doesn’t seem the sort of setup a maker of a stable OS, who is worried
about 3rd party driver quality, would want to keep in place.


Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm

“David J. Craig” wrote in message
news:xxxxx@ntdev…
>
> Jamey, do you believe in the tooth fairy too? To most of them the only
> “trigger” they know is the horse. Maybe I am giving them too much credit.
> There are a few really talented file systems filter driver writers and the
> masses whose talent lies in areas other than computer software design and
> development. How many times have you seen a post from some FSF writer who
> said they put in the hooks to handle IRP_MJ_CREATE and wonder why their
> development system won’t boot anymore. No FastIo routines or read or
write
> or even passthru. They also think they can save the fileobject and
> FsContext in the dispatch and have it usable. Add that to the “advanced”
> writers who say they want to open the file themselves in the dispatch
> function and then decide if they should allow real create to proceed.
Then
> they complain that they see another create for the same file and can’t
> figure out what to do with it. Maybe they should just keep trying to
create
> the create until they get the pretty blue screen.
>
> How about those who think that the old sources for filemon are a good
> starting place for an active FSF? ASAIK, there are no good examples of
> active FSFs available for free. Those of us who have done them, don’t
give
> that much work and pain away for free. BTW, the examples in the IFS Kit
are
> passive filters. Another good one is those who want to change the name
and
> have it automagically work on the correct volume. Maybe we should ask
> Microsoft to just handle it if it is same file system doing the
processing.
> Should make the Microsoft code very interesting in trying to share device
> extensions. Should do wonders for security too.
>
> Would someone stop the world, I wanna get off. If you are into pain run
> several of the antivirus products on the same system. Somebody better
hurry
> up and invent Isaac Asimov’s positronic brain robot.
>
> “Jamey Kirby” wrote in message
news:xxxxx@ntdev…
>
> How can we punish filter driver developers who fail unknown IOCTLS and
> FSCTLS rather than passing them down to let others have a stab at them? I
am
> thinking in terms of some sort of excommunication. Or maybe some sort of
> award of Hubris; presented to the company that is shipping such a driver.
>
> Some folks have no business writing code for the real world. You would
think
> that the word “FILTER” would trigger the correct synapse… A little
> abstraction… Oh well, maybe I am just dreaming.
>
> Jamey Kirby, Windows DDK MVP
> StorageCraft Corporation
> xxxxx@storagecraft.com
> http://www.storagecraft.com
>
>
>
>

> There are no good examples because Microsoft probably doesn’t have the time

and/or inclination to produce such. No one else can because the licensing
restrictions of the IFS kit STRICTLY forbid such. You couldn’t possibly
post a free sample driver that uses ntifs.h to any public location, even if
you wanted to, without breaking several provisions in the IFS kit licensing
agreement.

Hmm, how do you read this? You mean the SysInternals duo were breaking
their license all of the years they had FileMon source code posted on
their web site? The IFS kit license (I’m looking at the XP SP1 final
release license) does include a number of strange provisions, but
nothing I see that would prohibit the public distribution of code that
merely includes ntifs.h, as long as the code is not a derivative of one
of Microsoft’s samples.


Nick Ryan (MVP for DDK)

Well, I am no lawyer but reading the latest IFS EULA available at this link:

http://www.microsoft.com/whdc/ddk/ifskit/ServerIFSEULA.mspx

I notice a couple of interesting items right off:

(i) Microsoft Windows Server 2003 IFS source code, sample development source
code, tools, and other object code (“IFS Code”)

  1. GRANT OF LICENSE. This EULA grants you the following rights:

. a. SOFTWARE PRODUCT.
.
.
.
You may reproduce, license and distribute the IFS Code as part of the
IFS Drivers in object code form only to your end users, provided you first
obtain a valid digital signature from Microsoft for the IFS Drivers pursuant
to any additional instructions in the SOFTWARE PRODUCT and you comply with
the Redistribution Requirements described below.

. b. Redistribution Requirements. You may reproduce and distribute an
unlimited number of copies of the IFS Code as described above in object code
form, provided that: (i) you first obtain a valid digital signature from
Microsoft for the IFS Drivers pursuant to instructions in the IFS Kit; (ii)
you distribute the IFS Code only in conjunction with and as a part of your
IFS Drivers for the OS Product; (iii) the IFS Code only operates in
conjunction with the OS Product identified in Section 1(a) above; (iv) you
do not use Microsoft’s name, logo, or trademarks to advertise, market or
promote your IFS Drivers for the OS Product without the express written
permission of Microsoft; (v) you include a valid copyright notice on your
IFS Drivers; (vi) you agree to indemnify, hold harmless, and defend
Microsoft from and against any claims or lawsuits, including attorney’s
fees, that arise or result from the use or distribution of your IFS Drivers;
and (vii) you notify Microsoft of your date of distribution of the IFS
Drivers by e-mail to xxxxx@microsoft.com (or such other e-mail
address as Microsoft may designate) no later than sixty (60) days prior to
such date.

Okay, so to answer your question, if FileMon derived in anyway from sample
source in the IFS kit, then yes distributing source to same was breaking the
law. Perhaps this is the real reason source is no longer available.

Further, FileMon’s approach to the problem may well have been unique, as
their code did not do what a typical file system filter driver should. In
order to write a proper file system filter driver, I don’t know how you
could avoid deriving your work from IFS source. Proper information just
isn’t available from any other source, and I am positive MS could make this
claim successfully in court if they were so inclined.

In addition, I notice the ONLY tool maker in this space is shipping their
tools in object form.

But, all this is moot as the file system driver space doesn’t have any
problems with 3rd party drivers right?


Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm

“Nick Ryan” wrote in message news:xxxxx@ntdev…
>
> > There are no good examples because Microsoft probably doesn’t have the
time
> > and/or inclination to produce such. No one else can because the
licensing
> > restrictions of the IFS kit STRICTLY forbid such. You couldn’t possibly
> > post a free sample driver that uses ntifs.h to any public location, even
if
> > you wanted to, without breaking several provisions in the IFS kit
licensing
> > agreement.
>
> Hmm, how do you read this? You mean the SysInternals duo were breaking
> their license all of the years they had FileMon source code posted on
> their web site? The IFS kit license (I’m looking at the XP SP1 final
> release license) does include a number of strange provisions, but
> nothing I see that would prohibit the public distribution of code that
> merely includes ntifs.h, as long as the code is not a derivative of one
> of Microsoft’s samples.
>
> –
> Nick Ryan (MVP for DDK)
>
>
>

“Bill McKenzie” wrote in message

> In addition, I notice the ONLY tool maker in this space is shipping their
> tools in object form.
>
Uh, the OSR Filter Driver Development Kit sure has a lot of source code in
it, so I have to question the above. Yes, the File System Development Kit
is mostly binary, but then again some of it is licensed code so what else
would it be.

Don Burn (MVP, Windows DDK)
Windows 2k/XP/2k3 Filesystem and Driver Consulting

Well as long as the source isn’t IFS sample derived source, or as long as no
litigation is ensued, I guess all is well. And I stand corrected on the
format of the source of those tools.

Back to the point, the IFS EULA seems fairly clear about “sample development
source
code,” being shippable in object format only whether or not what it says is
followed by anyone.

Also, the following section would seem to completely disallow any free
samples:

f. Publicly Available Software. Your license rights to the Sample Code are
conditioned upon you (a) not incorporating Identified Software into, or
combining Identified Software with, the Sample Code or a derivative work
thereof; (b) not distributing Identified Software in conjunction with the
Sample Code or a derivative work thereof; and (c) not using Identified
Software in the development of a derivative work of the Sample Code.
“Identified Software” means software which is licensed pursuant to terms
that (i) create, or purport to create, obligations for Microsoft with
respect to the Sample Code or derivative work thereof or (ii) grant, or
purport to grant, to any third party any rights or immunities under
Microsoft’s intellectual property or proprietary rights in the Sample Code
or derivative work thereof. Identified Software includes, without
limitation, “Publicly Available Software”. “Publicly Available Software”
means any software that requires as a condition of use, modification and/or
distribution of such software that such software or other software
incorporated into, derived from or distributed with such software be (a)
disclosed or distributed in source code form; (b) be licensed for the
purpose of making derivative works; or (c) be redistributable at no charge.

Again, I am certainly no lawyer, but this sure doesn’t look approachable to
me.


Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm

“Don Burn” wrote in message news:xxxxx@ntdev…
>
>
> “Bill McKenzie” wrote in message
>
> > In addition, I notice the ONLY tool maker in this space is shipping
their
> > tools in object form.
> >
> Uh, the OSR Filter Driver Development Kit sure has a lot of source code in
> it, so I have to question the above. Yes, the File System Development Kit
> is mostly binary, but then again some of it is licensed code so what else
> would it be.
>
> Don Burn (MVP, Windows DDK)
> Windows 2k/XP/2k3 Filesystem and Driver Consulting
>
>
>
>
>

“Bill McKenzie” wrote in message
news:xxxxx@ntdev…
>
>
> Again, I am certainly no lawyer, but this sure doesn’t look approachable
to
> me.
>

The FileMon source on the web significantly predated the IFS kit. In fact,
the VERY first file system filter driver source code was written by OSR’s
Tony Mason and was released on Compuserve, back when those forums where the
only game in town for Windows NT driver support. It created QUITE a stir at
the time.

Those of you who are new to world of NT drivers, like Bill, won’t remember
that we had to wait something like 5 years before there WAS a publically
avialable IFS kit. I won’t rehash all the history, it’s painful, and old,
and gone, and doesn’t matter that much anymore.

The IFS Kit license isn’t what most of us in the community would wish for,
obviously. But you’ll note that it DOES have a pretty reasonable balance…
Like, for example, we can have this forum and discuss the IFS Kit contents.
That is, at least, SOMEthing. Every few years we hear hopefull signs that
the license might become a bit less restrictive. But given how long it took
to GET an IFS Kit, and how hard we had to fight for it, it’s FAR better than
the alternative (which is a source license).

Peter
OSR

> was released on Compuserve

Compuserve, WOW!! So what was it like riding your Tyranasaurus Rex to
work? :slight_smile:

The IFS Kit license isn’t what most of us in the community would wish for,
obviously. But you’ll note that it DOES have a pretty reasonable
balance…
Like, for example, we can have this forum and discuss the IFS Kit
contents.
That is, at least, SOMEthing. Every few years we hear hopefull signs that
the license might become a bit less restrictive. But given how long it
took
to GET an IFS Kit, and how hard we had to fight for it, it’s FAR better
than
the alternative (which is a source license).

Agreed. Further, I hope enough discussions like this will cause some at the
MS to once again reconsider their lawyer’s prose.


Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm

“Peter Viscarola” wrote in message news:xxxxx@ntdev…
>
>
> “Bill McKenzie” wrote in message
> news:xxxxx@ntdev…
> >
> >
> > Again, I am certainly no lawyer, but this sure doesn’t look approachable
> to
> > me.
> >
>
> The FileMon source on the web significantly predated the IFS kit. In
fact,
> the VERY first file system filter driver source code was written by OSR’s
> Tony Mason and was released on Compuserve, back when those forums where
the
> only game in town for Windows NT driver support. It created QUITE a stir
at
> the time.
>
> Those of you who are new to world of NT drivers, like Bill, won’t remember
> that we had to wait something like 5 years before there WAS a publically
> avialable IFS kit. I won’t rehash all the history, it’s painful, and old,
> and gone, and doesn’t matter that much anymore.
>
> The IFS Kit license isn’t what most of us in the community would wish for,
> obviously. But you’ll note that it DOES have a pretty reasonable
balance…
> Like, for example, we can have this forum and discuss the IFS Kit
contents.
> That is, at least, SOMEthing. Every few years we hear hopefull signs that
> the license might become a bit less restrictive. But given how long it
took
> to GET an IFS Kit, and how hard we had to fight for it, it’s FAR better
than
> the alternative (which is a source license).
>
> Peter
> OSR
>
>
>
>

I believe section f is again referring to use of Microsoft’s Sample
Code, not anything you may come up with yourself (it’s basically saying
that you can’t place Microsoft’s code under something like an
open-source license). I think the capitalization of ‘Sample Code’ is the
key here.

Bill McKenzie wrote:

Well as long as the source isn’t IFS sample derived source, or as long as no
litigation is ensued, I guess all is well. And I stand corrected on the
format of the source of those tools.

Back to the point, the IFS EULA seems fairly clear about “sample development
source
code,” being shippable in object format only whether or not what it says is
followed by anyone.

Also, the following section would seem to completely disallow any free
samples:

f. Publicly Available Software. Your license rights to the Sample Code are
conditioned upon you (a) not incorporating Identified Software into, or
combining Identified Software with, the Sample Code or a derivative work
thereof; (b) not distributing Identified Software in conjunction with the
Sample Code or a derivative work thereof; and (c) not using Identified
Software in the development of a derivative work of the Sample Code.
“Identified Software” means software which is licensed pursuant to terms
that (i) create, or purport to create, obligations for Microsoft with
respect to the Sample Code or derivative work thereof or (ii) grant, or
purport to grant, to any third party any rights or immunities under
Microsoft’s intellectual property or proprietary rights in the Sample Code
or derivative work thereof. Identified Software includes, without
limitation, “Publicly Available Software”. “Publicly Available Software”
means any software that requires as a condition of use, modification and/or
distribution of such software that such software or other software
incorporated into, derived from or distributed with such software be (a)
disclosed or distributed in source code form; (b) be licensed for the
purpose of making derivative works; or (c) be redistributable at no charge.

Again, I am certainly no lawyer, but this sure doesn’t look approachable to
me.


Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm

“Don Burn” wrote in message news:xxxxx@ntdev…
>
>>
>>“Bill McKenzie” wrote in message
>>
>>
>>>In addition, I notice the ONLY tool maker in this space is shipping
>
> their
>
>>>tools in object form.
>>>
>>
>>Uh, the OSR Filter Driver Development Kit sure has a lot of source code in
>>it, so I have to question the above. Yes, the File System Development Kit
>>is mostly binary, but then again some of it is licensed code so what else
>>would it be.
>>
>>Don Burn (MVP, Windows DDK)
>>Windows 2k/XP/2k3 Filesystem and Driver Consulting
>>
>>
>>
>>
>>
>
>
>
>
>


Nick Ryan (MVP for DDK)

Well, here is the phrase that would encompass anything I can imagine
creating in this space:

“Sample Code or a derivative work”

I don’t know how you could create a non-derivative work. How else could you
possibly know the correct way to create a file system related driver without
following the IFS samples, or some samples derived from the IFS samples?
The IFS IS the only source of this information.


Bill McKenzie
Compuware Corporation
Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
http://frontline.compuware.com/nashua/patches/utility.htm

“Nick Ryan” wrote in message news:xxxxx@ntdev…
>
> I believe section f is again referring to use of Microsoft’s Sample
> Code, not anything you may come up with yourself (it’s basically saying
> that you can’t place Microsoft’s code under something like an
> open-source license). I think the capitalization of ‘Sample Code’ is the
> key here.
>
> Bill McKenzie wrote:
>
> > Well as long as the source isn’t IFS sample derived source, or as long
as no
> > litigation is ensued, I guess all is well. And I stand corrected on the
> > format of the source of those tools.
> >
> > Back to the point, the IFS EULA seems fairly clear about “sample
development
> > source
> > code,” being shippable in object format only whether or not what it says
is
> > followed by anyone.
> >
> > Also, the following section would seem to completely disallow any free
> > samples:
> >
> > f. Publicly Available Software. Your license rights to the Sample Code
are
> > conditioned upon you (a) not incorporating Identified Software into, or
> > combining Identified Software with, the Sample Code or a derivative work
> > thereof; (b) not distributing Identified Software in conjunction with
the
> > Sample Code or a derivative work thereof; and (c) not using Identified
> > Software in the development of a derivative work of the Sample Code.
> > “Identified Software” means software which is licensed pursuant to terms
> > that (i) create, or purport to create, obligations for Microsoft with
> > respect to the Sample Code or derivative work thereof or (ii) grant, or
> > purport to grant, to any third party any rights or immunities under
> > Microsoft’s intellectual property or proprietary rights in the Sample
Code
> > or derivative work thereof. Identified Software includes, without
> > limitation, “Publicly Available Software”. “Publicly Available Software”
> > means any software that requires as a condition of use, modification
and/or
> > distribution of such software that such software or other software
> > incorporated into, derived from or distributed with such software be (a)
> > disclosed or distributed in source code form; (b) be licensed for the
> > purpose of making derivative works; or (c) be redistributable at no
charge.
> >
> > Again, I am certainly no lawyer, but this sure doesn’t look approachable
to
> > me.
> >
> > –
> > Bill McKenzie
> > Compuware Corporation
> > Watch your IRPs/IRBs/URBs/SRBs/NDIS pkts with our free WDMSniffer tool:
> > http://frontline.compuware.com/nashua/patches/utility.htm
> >
> >
> > “Don Burn” wrote in message news:xxxxx@ntdev…
> >
> >>
> >>“Bill McKenzie” wrote in message
> >>
> >>
> >>>In addition, I notice the ONLY tool maker in this space is shipping
> >
> > their
> >
> >>>tools in object form.
> >>>
> >>
> >>Uh, the OSR Filter Driver Development Kit sure has a lot of source code
in
> >>it, so I have to question the above. Yes, the File System Development
Kit
> >>is mostly binary, but then again some of it is licensed code so what
else
> >>would it be.
> >>
> >>Don Burn (MVP, Windows DDK)
> >>Windows 2k/XP/2k3 Filesystem and Driver Consulting
> >>
> >>
> >>
> >>
> >>
> >
> >
> >
> >
> >
>
> –
> Nick Ryan (MVP for DDK)
>
>
>