RE: [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Steven M Christey]

If a license has a clause that requires insecure behavior, then the behavior 
itself can still be classified using an existing CWE.

Informally, from a CWE perspective, it doesn’t matter  forced the 
insecure behavior to enter the product, or what phase of the SDLC 
(requirements, design, implementation error, configuration error, etc.) – 
ideally, CWE characterizes the insecure behavior itself. This notion does get 
extended to certain properties of code that can contribute to security issues 
(CWE-1121: Excessive McCabe Cyclomatic Complexity), so to me, a broader 
question is whether “having a risky license” fits as an “undesired property of 
behavior” or if it’s something else.

One other note – we are tracking several proposals about expanding CWE scope 
that need to be discussed by the entire CWE community sometime in the future, 
and this active debate about licenses is another scope discussion, which might 
fit under “human processes that exist outside of the product itself.”

- Steve


From: Kurt Seifried 
Sent: Thursday, November 9, 2023 3:30 PM
To: Roguski, Przemyslaw 
Cc: jonathan.w.hood6@army.mil; Hatfield, Arthur 
; CWE Research Discussion 

Subject: Re: [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing 
(UNCLASSIFIED)

What if a license has a clause that requires an insecure or problematic 
setting/configuration/behavior? Someone did a parody licence called the " 
Insecure License" but I wouldn't put it past sometime to have done this for 
real. 

What if a license has a clause that requires an insecure or problematic 
setting/configuration/behavior? Someone did a parody licence called the " 
Insecure License" but I wouldn't put it past sometime to have done this for 
real. What about audit clauses (e.g. can I audit the NSA usage of my product) 
and other forms of information disclosure?

On Thu, Nov 9, 2023 at 11:46 AM Przemyslaw Roguski 
mailto:progu...@redhat.com>> wrote:
Hi Jon, Thank you for accepting different opinions and I'm really happy that we 
have this discussion here. To be honest I never consider licensing issues as a 
potential problem that could be considered as a software weakness. But it seems

Hi Jon,

Thank you for accepting different opinions and I'm really happy that we have 
this discussion here.
To be honest I never consider licensing issues as a potential problem that 
could be considered as a software weakness.
But it seems that such a clarification is required.
Let me repeat what I said before, regardless of the declared licence (valid or 
not) still you can be impacted if you run a malicious software/code.
Like in Arthur's example.
Thank you Arthur for your great examples!

Jon, let me try to provide one more clarification. If your license management 
software will accept invalid licence and recognize it as a valid one, then yes, 
it's a licence related weakness, but it's a weakness in your software, not a 
problem with the wrong license itself that was provided. Depending on the root 
cause in such a case it could be "CWE-184: Incomplete List of Disallowed 
Inputs" weakness related to the license management software. You can blame the 
invalid license, but in fact the root cause and the weakness is in the license 
management software.

Like it was said a few times, licence by itself, regardless if valid or not, 
cannot lead to vulnerability, hence cannot be classified as a weakness as well.
Invalid licence can lead to legal problems, but this is a different story.

I hope it helps.

Best regards,
Przemek


Przemyslaw Roguski

Security Architect, Product Security

Red Hat Poland<https://www.redhat.com/>

progu...@redhat.com<mailto:progu...@redhat.com>IM: rogue
<https://red.ht/sig>
 <https://red.ht/sig>
On Thu, Nov 9, 2023 at 6:49 PM Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 
 wrote:<https://red.ht/sig>
I understand the position better with this analogy; thank 
you.<https://red.ht/sig>
 <https://red.ht/sig>
I do believe that it is not a comparable analogy. Raising energy prices are not 
a property of the software. A software license is a property of the software, 
so the argument you make here is based off of an initial assertion that seems 
incorrect. Just because the fix for the weakness is voluntary doesn’t mean it’s 
not a weakness (IE: voluntarily stop using untrusted components, CWE-1357), 
though license enforcement may not be voluntary in all 
cases.<https://red.ht/sig>
 <https://red.ht/sig>
“It doesn’t do anything to stop the execution of that software on any system 
not under your direct control where it’s already running” – I’m arguing that it 
does. When you incorporate software in violation of the license, you expose 
your product to injunction or restraint which absolutely can apply to the 
executing software directly under your control. For example, if the Home Depot 
website used software without licensing it, the software 

Re: [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Kurt Seifried]

se adherence with license management software or request that your
>> users adhere to it voluntarily, it’s still an availability vulnerability
>> that should also have a CWE to categorize it.
>>
>>
>>
>> Thank you for helping me understand the position better. I do not think I
>> agree with it, but do understand the position better than I did before.
>>
>>
>>
>> Jon
>>
>>
>>
>> *From:* Hatfield, Arthur 
>> *Sent:* Thursday, November 9, 2023 9:54 AM
>> *To:* Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) <
>> jonathan.w.hood6@army.mil>; Przemyslaw Roguski ;
>> Steven M Christey 
>> *Cc:* CWE Research Discussion 
>> *Subject:* Re: [Non-DoD Source] Re: Request for CWE: Improper Licensing
>> (UNCLASSIFIED)
>>
>>
>>
>> You don't often get email from arthur_hatfi...@homedepot.com. Learn why
>> this is important <https://aka.ms/LearnAboutSenderIdentification>
>>
>> Look at it this way:
>>
>>
>>
>> Licensing issues are not a property of software, but of the society and
>> economy around the software.
>>
>>
>>
>> A buffer overflow in a driver will crash your computer and make it
>> unavailable any time data passes through it in a particular way, no matter
>> who is causing that data to go through that buffer or why. A GPL-violation
>> lawsuit will only stop you from distributing software if you voluntarily
>> settle or you lose the lawsuit, and even then that’s basically going to
>> require voluntary action on your part to stop using and/or distributing the
>> software. It doesn’t do anything to stop the execution of that software on
>> any system not under your direct control where it’s already running.
>> Availability of the software in this case is not affected by a “coding
>> weakness,” but by your organizational response to social, legal, and
>> economic pressure.
>>
>>
>>
>> If we put license issues in CWE, then we might as well put rising energy
>> costs in CWE. A surprise jump in your power bill could affect the
>> availability of your application if you respond to the bill by voluntarily
>> turning off your computer.
>>
>>
>>
>>
>>
>> *RT Hatfield* *|* *Staff Cybersecurity Analyst **|* *BS CS, CCITP, CISSP*
>>
>> *The Home Depot | **Cyber Threat Intelligence*
>>
>> * arthur_hatfi...@homedepot.com
>>
>> * c...@homedepot.com
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>>
>> INTERNAL USE
>>
>> *From: *Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) <
>> jonathan.w.hood6@army.mil>
>> *Date: *Thursday, November 9, 2023 at 10:26 AM
>> *To: *Przemyslaw Roguski , Steven M Christey <
>> co...@mitre.org>
>> *Cc: *CWE Research Discussion 
>> *Subject: *[EXTERNAL] [EXT] RE: [Non-DoD Source] Re: Request for CWE:
>> Improper Licensing (UNCLASSIFIED)
>>
>> I respectfully disagree with this. Using a license incorrectly causes an
>> availability issue directly, and availability is one of the cybersecurity
>> principles that represent weaknesses and vulnerabilities by the definitions
>> I am aware of.
>>
>>
>>
>> Can you please help me understand what definition CWE is using for each?
>> The nearest definitions I can find are: “A ‘weakness’ is a condition in a
>> software, firmware, hardware, or service component that, under certain
>> circumstances, could contribute to the introduction of vulnerabilities” (
>> https://cwe.mitre.org/about/new_to_cwe.html). Following the
>> vulnerability theory (
>> https://cwe.mitre.org/documents/vulnerability_theory/intro.html)
>> suggests that we need to have a PRODUCT implementing FEATURE by performing
>> certain BEHAVIORS that operate on RESOURCES. I will assume these
>> definitions in my disagreement below, and acknowledge that my basic
>> definitions of some of these terms may be off.
>>
>>
>>
>> The core question is therefore: is a license violation a vulnerability by
>> the vulnerability theory used by the CWEs? I argue in the affirmative. You
>> state, “No doubt that invalid licenses are a serious problem from the
>> security perspective, but it's more a supply chain issue and legal
>> problem.” Then a PRODUCT implementing software with a “supply chain issue”
>> or “legal problem” to achieve its behavior on its resources produces an
>> availability security impact against PRODUCT. If you want to identify it
>> more generally as “supply chain issue” or “legal insuffici

Re: [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Hatfield, Arthur]

I think the main distinction that I want to be observed is that the CWEs are 
about vulnerabilities that are intrinsic to a piece of software (firmware, 
hardware) due to defects in its design or implementation, leading to unwanted 
behavior of the software (firmware, hardware) itself. Licensing issues involve 
a business process vulnerability due to defects in organizational practices 
around the software (e.g. not paying bills, or not respecting copyright law) 
leading to sticky business situations that can make continuing to rely on a 
given software component odious to the organization’s decision-makers.

RT Hatfield | Staff Cybersecurity Analyst | BS CS, CCITP, CISSP
The Home Depot | Cyber Threat Intelligence
• arthur_hatfi...@homedepot.com<mailto:arthur_hatfi...@homedepot.com>
• c...@homedepot.com<mailto:c...@homedepot.com>





INTERNAL USE
From: Przemyslaw Roguski 
Date: Thursday, November 9, 2023 at 1:45 PM
To: Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 

Cc: Hatfield, Arthur , CWE Research Discussion 

Subject: [EXTERNAL] Re: [EXT] RE: [Non-DoD Source] Re: Request for CWE: 
Improper Licensing (UNCLASSIFIED)
Hi Jon, Thank you for accepting different opinions and I'm really happy that we 
have this discussion here. To be honest I never consider licensing issues as a 
potential problem that could be considered as a software weakness. But it seems

Hi Jon,

Thank you for accepting different opinions and I'm really happy that we have 
this discussion here.
To be honest I never consider licensing issues as a potential problem that 
could be considered as a software weakness.
But it seems that such a clarification is required.
Let me repeat what I said before, regardless of the declared licence (valid or 
not) still you can be impacted if you run a malicious software/code.
Like in Arthur's example.
Thank you Arthur for your great examples!

Jon, let me try to provide one more clarification. If your license management 
software will accept invalid licence and recognize it as a valid one, then yes, 
it's a licence related weakness, but it's a weakness in your software, not a 
problem with the wrong license itself that was provided. Depending on the root 
cause in such a case it could be "CWE-184: Incomplete List of Disallowed 
Inputs" weakness related to the license management software. You can blame the 
invalid license, but in fact the root cause and the weakness is in the license 
management software.

Like it was said a few times, licence by itself, regardless if valid or not, 
cannot lead to vulnerability, hence cannot be classified as a weakness as well.
Invalid licence can lead to legal problems, but this is a different story.

I hope it helps.

Best regards,
Przemek


Przemyslaw Roguski

Security Architect, Product Security

Red Hat Poland 
[redhat.com]<https://urldefense.com/v3/__https:/www.redhat.com/__;!!M-nmYVHPHQ!IT_WQ6o4kOATVHOVzPwf7E_vHmtTOGWTno9gvKZW9p-B4E3liOnptTlKyzWvlRvkXiqu5T8sRizeClxJ9g6DOl5e6w$>

progu...@redhat.com<mailto:progu...@redhat.com>IM: rogue
[Image removed by 
sender.][red.ht]<https://urldefense.com/v3/__https:/red.ht/sig__;!!M-nmYVHPHQ!IT_WQ6o4kOATVHOVzPwf7E_vHmtTOGWTno9gvKZW9p-B4E3liOnptTlKyzWvlRvkXiqu5T8sRizeClxJ9g5eCEVQgQ$>

On Thu, Nov 9, 2023 at 6:49 PM Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 
mailto:jonathan.w.hood6@army.mil>> wrote:
I understand the position better with this analogy; thank you.

I do believe that it is not a comparable analogy. Raising energy prices are not 
a property of the software. A software license is a property of the software, 
so the argument you make here is based off of an initial assertion that seems 
incorrect. Just because the fix for the weakness is voluntary doesn’t mean it’s 
not a weakness (IE: voluntarily stop using untrusted components, CWE-1357), 
though license enforcement may not be voluntary in all cases.

“It doesn’t do anything to stop the execution of that software on any system 
not under your direct control where it’s already running” – I’m arguing that it 
does. When you incorporate software in violation of the license, you expose 
your product to injunction or restraint which absolutely can apply to the 
executing software directly under your control. For example, if the Home Depot 
website used software without licensing it, the software may have a license 
enforcement mechanism (and would have the legal authority to) shut off the 
software once the license becomes unverified, or Home Depot may receive an 
injunction to stop using the unlicensed software. Either of these scenarios 
would directly affect the availability of the Home Depot website and are 
reflected by an underlying coding weakness of relying on or incorporating 
improperly licensed components. Home Depot would not have a choice or voluntary 
decision in such a case, and even if it did, it's still a quantifiable threat 
to the software.

“If we put license issues in CWE, then we might as 

Re: [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Przemyslaw Roguski]

buffer or why. A GPL-violation
> lawsuit will only stop you from distributing software if you voluntarily
> settle or you lose the lawsuit, and even then that’s basically going to
> require voluntary action on your part to stop using and/or distributing the
> software. It doesn’t do anything to stop the execution of that software on
> any system not under your direct control where it’s already running.
> Availability of the software in this case is not affected by a “coding
> weakness,” but by your organizational response to social, legal, and
> economic pressure.
>
>
>
> If we put license issues in CWE, then we might as well put rising energy
> costs in CWE. A surprise jump in your power bill could affect the
> availability of your application if you respond to the bill by voluntarily
> turning off your computer.
>
>
>
>
>
> *RT Hatfield* *|* *Staff Cybersecurity Analyst **|* *BS CS, CCITP, CISSP*
>
> *The Home Depot | **Cyber Threat Intelligence*
>
> * arthur_hatfi...@homedepot.com
>
> * c...@homedepot.com
>
>
>
>
>
>
>
>
>
>
>
> INTERNAL USE
>
> *From: *Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) <
> jonathan.w.hood6@army.mil>
> *Date: *Thursday, November 9, 2023 at 10:26 AM
> *To: *Przemyslaw Roguski , Steven M Christey <
> co...@mitre.org>
> *Cc: *CWE Research Discussion 
> *Subject: *[EXTERNAL] [EXT] RE: [Non-DoD Source] Re: Request for CWE:
> Improper Licensing (UNCLASSIFIED)
>
> I respectfully disagree with this. Using a license incorrectly causes an
> availability issue directly, and availability is one of the cybersecurity
> principles that represent weaknesses and vulnerabilities by the definitions
> I am aware of.
>
>
>
> Can you please help me understand what definition CWE is using for each?
> The nearest definitions I can find are: “A ‘weakness’ is a condition in a
> software, firmware, hardware, or service component that, under certain
> circumstances, could contribute to the introduction of vulnerabilities” (
> https://cwe.mitre.org/about/new_to_cwe.html
> <https://cwe.mitre.org/about/new_to_cwe.html>).
> Following the vulnerability theory (
> https://cwe.mitre.org/documents/vulnerability_theory/intro.html
> <https://cwe.mitre.org/documents/vulnerability_theory/intro.html>)
> suggests that we need to have a PRODUCT implementing FEATURE by performing
> certain BEHAVIORS that operate on RESOURCES. I will assume these
> definitions in my disagreement below, and acknowledge that my basic
> definitions of some of these terms may be off.
>
>
>
> The core question is therefore: is a license violation a vulnerability by
> the vulnerability theory used by the CWEs? I argue in the affirmative. You
> state, “No doubt that invalid licenses are a serious problem from the
> security perspective, but it's more a supply chain issue and legal
> problem.” Then a PRODUCT implementing software with a “supply chain issue”
> or “legal problem” to achieve its behavior on its resources produces an
> availability security impact against PRODUCT. If you want to identify it
> more generally as “supply chain issue” or “legal insufficiency,” it’s still
> a vulnerability that directly affects availability, incurs technical debt,
> and inflicts reputation/brand damage (
> https://cwe.mitre.org/cwraf/creatingyourownvignettes.html
> <https://cwe.mitre.org/cwraf/creatingyourownvignettes.html>).
> I believe that more specific supply chain or legal issues are appropriate
> as well (and license violations would be a specific example), but these two
> general classes of vulnerabilities you’ve identified also meet the criteria
> for becoming a CWE. CWE-1357 almost meets some of this definition as well.
> Instead of a license-violating component being “not sufficiently trusted to
> meet expectations for security” (with availability being part of the
> security definition), it would be nice to have a CWE that can refer to a
> component (trusted or not) that in fact does not meet security expectations
> because of “supply chain” or “legal” vulnerabilities.
>
>
>
> Can you please further explain why a “supply chain issue and legal
> problem” is an abuse of the weakness definition? I feel like you
> acknowledge it’s a weakness while also saying it’s an abuse of the
> definition of a weakness, which indicates that I’m not understanding some
> of your argument. You lose me at “Invalid license itself cannot lead to a
> vulnerability just like that.” How is a coding weakness that affects
> availability not a distinct, individual vulnerability, regardless of what
> other vulnerabilities may also be in the software?
>
>
>
> Sincere thanks for your response and interaction,
>
> Jon
>

[EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA)]

I understand the position better with this analogy; thank you.

 

I do believe that it is not a comparable analogy. Raising energy prices are not 
a property of the software. A software license is a property of the software, 
so the argument you make here is based off of an initial assertion that seems 
incorrect. Just because the fix for the weakness is voluntary doesn’t mean it’s 
not a weakness (IE: voluntarily stop using untrusted components, CWE-1357), 
though license enforcement may not be voluntary in all cases.

 

“It doesn’t do anything to stop the execution of that software on any system 
not under your direct control where it’s already running” – I’m arguing that it 
does. When you incorporate software in violation of the license, you expose 
your product to injunction or restraint which absolutely can apply to the 
executing software directly under your control. For example, if the Home Depot 
website used software without licensing it, the software may have a license 
enforcement mechanism (and would have the legal authority to) shut off the 
software once the license becomes unverified, or Home Depot may receive an 
injunction to stop using the unlicensed software. Either of these scenarios 
would directly affect the availability of the Home Depot website and are 
reflected by an underlying coding weakness of relying on or incorporating 
improperly licensed components. Home Depot would not have a choice or voluntary 
decision in such a case, and even if it did, it's still a quantifiable threat 
to the software.

 

“If we put license issues in CWE, then we might as well put rising energy costs 
in CWE.” Whether the energy is cut off through physical means (battery 
overload) or you can find a repeatable way to intentionally get a reliable 
power source shut off because the user can’t pay the power bill, it’s still a 
valid CWE-920 in my opinion. Likewise, whether you force license adherence with 
license management software or request that your users adhere to it 
voluntarily, it’s still an availability vulnerability that should also have a 
CWE to categorize it.

 

Thank you for helping me understand the position better. I do not think I agree 
with it, but do understand the position better than I did before.

 

Jon

 

From: Hatfield, Arthur  
Sent: Thursday, November 9, 2023 9:54 AM
To: Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 
; Przemyslaw Roguski ; 
Steven M Christey 
Cc: CWE Research Discussion 
Subject: Re: [Non-DoD Source] Re: Request for CWE: Improper Licensing 
(UNCLASSIFIED)

 


You don't often get email from arthur_hatfi...@homedepot.com 
<mailto:arthur_hatfi...@homedepot.com> . Learn why this is important 
<https://aka.ms/LearnAboutSenderIdentification> 



Look at it this way:

 

Licensing issues are not a property of software, but of the society and economy 
around the software. 

 

A buffer overflow in a driver will crash your computer and make it unavailable 
any time data passes through it in a particular way, no matter who is causing 
that data to go through that buffer or why. A GPL-violation lawsuit will only 
stop you from distributing software if you voluntarily settle or you lose the 
lawsuit, and even then that’s basically going to require voluntary action on 
your part to stop using and/or distributing the software. It doesn’t do 
anything to stop the execution of that software on any system not under your 
direct control where it’s already running. Availability of the software in this 
case is not affected by a “coding weakness,” but by your organizational 
response to social, legal, and economic pressure.

 

If we put license issues in CWE, then we might as well put rising energy costs 
in CWE. A surprise jump in your power bill could affect the availability of 
your application if you respond to the bill by voluntarily turning off your 
computer.

 

 

RT Hatfield | Staff Cybersecurity Analyst | BS CS, CCITP, CISSP

The Home Depot | Cyber Threat Intelligence

*  <mailto:arthur_hatfi...@homedepot.com> arthur_hatfi...@homedepot.com

*  <mailto:c...@homedepot.com> c...@homedepot.com

 

 

 

 

 

INTERNAL USE

From: Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 
mailto:jonathan.w.hood6@army.mil> >
Date: Thursday, November 9, 2023 at 10:26 AM
To: Przemyslaw Roguski mailto:progu...@redhat.com> >, 
Steven M Christey mailto:co...@mitre.org> >
Cc: CWE Research Discussion mailto:cwe-research-list@mitre.org> >
Subject: [EXTERNAL] [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper 
Licensing (UNCLASSIFIED)

I respectfully disagree with this. Using a license incorrectly causes an 
availability issue directly, and availability is one of the cybersecurity 
principles that represent weaknesses and vulnerabilities by the definitions I 
am aware of.

 

Can you please help me understand what definition CWE is using for each? The 
nearest definitions I can find are: “A ‘weakness’ i

[EXT] Re: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Hatfield, Arthur]

Look at it this way:

Licensing issues are not a property of software, but of the society and economy 
around the software.

A buffer overflow in a driver will crash your computer and make it unavailable 
any time data passes through it in a particular way, no matter who is causing 
that data to go through that buffer or why. A GPL-violation lawsuit will only 
stop you from distributing software if you voluntarily settle or you lose the 
lawsuit, and even then that’s basically going to require voluntary action on 
your part to stop using and/or distributing the software. It doesn’t do 
anything to stop the execution of that software on any system not under your 
direct control where it’s already running. Availability of the software in this 
case is not affected by a “coding weakness,” but by your organizational 
response to social, legal, and economic pressure.

If we put license issues in CWE, then we might as well put rising energy costs 
in CWE. A surprise jump in your power bill could affect the availability of 
your application if you respond to the bill by voluntarily turning off your 
computer.


RT Hatfield | Staff Cybersecurity Analyst | BS CS, CCITP, CISSP
The Home Depot | Cyber Threat Intelligence
• arthur_hatfi...@homedepot.com<mailto:arthur_hatfi...@homedepot.com>
• c...@homedepot.com<mailto:c...@homedepot.com>






INTERNAL USE
From: Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 

Date: Thursday, November 9, 2023 at 10:26 AM
To: Przemyslaw Roguski , Steven M Christey 

Cc: CWE Research Discussion 
Subject: [EXTERNAL] [EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper 
Licensing (UNCLASSIFIED)
I respectfully disagree with this. Using a license incorrectly causes an 
availability issue directly, and availability is one of the cybersecurity 
principles that represent weaknesses and vulnerabilities by the definitions I 
am aware of.

Can you please help me understand what definition CWE is using for each? The 
nearest definitions I can find are: “A ‘weakness’ is a condition in a software, 
firmware, hardware, or service component that, under certain circumstances, 
could contribute to the introduction of vulnerabilities” 
(https://cwe.mitre.org/about/new_to_cwe.html). Following the vulnerability 
theory (https://cwe.mitre.org/documents/vulnerability_theory/intro.html) 
suggests that we need to have a PRODUCT implementing FEATURE by performing 
certain BEHAVIORS that operate on RESOURCES. I will assume these definitions in 
my disagreement below, and acknowledge that my basic definitions of some of 
these terms may be off.

The core question is therefore: is a license violation a vulnerability by the 
vulnerability theory used by the CWEs? I argue in the affirmative. You state, 
“No doubt that invalid licenses are a serious problem from the security 
perspective, but it's more a supply chain issue and legal problem.” Then a 
PRODUCT implementing software with a “supply chain issue” or “legal problem” to 
achieve its behavior on its resources produces an availability security impact 
against PRODUCT. If you want to identify it more generally as “supply chain 
issue” or “legal insufficiency,” it’s still a vulnerability that directly 
affects availability, incurs technical debt, and inflicts reputation/brand 
damage (https://cwe.mitre.org/cwraf/creatingyourownvignettes.html). I believe 
that more specific supply chain or legal issues are appropriate as well (and 
license violations would be a specific example), but these two general classes 
of vulnerabilities you’ve identified also meet the criteria for becoming a CWE. 
CWE-1357 almost meets some of this definition as well. Instead of a 
license-violating component being “not sufficiently trusted to meet 
expectations for security” (with availability being part of the security 
definition), it would be nice to have a CWE that can refer to a component 
(trusted or not) that in fact does not meet security expectations because of 
“supply chain” or “legal” vulnerabilities.

Can you please further explain why a “supply chain issue and legal problem” is 
an abuse of the weakness definition? I feel like you acknowledge it’s a 
weakness while also saying it’s an abuse of the definition of a weakness, which 
indicates that I’m not understanding some of your argument. You lose me at 
“Invalid license itself cannot lead to a vulnerability just like that.” How is 
a coding weakness that affects availability not a distinct, individual 
vulnerability, regardless of what other vulnerabilities may also be in the 
software?

Sincere thanks for your response and interaction,
Jon

From: Przemyslaw Roguski 
Sent: Sunday, November 5, 2023 1:21 PM
To: Steven M Christey 
Cc: Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 
; CWE Research Discussion 

Subject: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

You don't often get email from progu...@redhat.com. Learn why this is 
important<https://aka.ms/LearnAboutSenderIdentif

[EXT] RE: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

[Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA)]

I respectfully disagree with this. Using a license incorrectly causes an 
availability issue directly, and availability is one of the cybersecurity 
principles that represent weaknesses and vulnerabilities by the definitions I 
am aware of.

 

Can you please help me understand what definition CWE is using for each? The 
nearest definitions I can find are: “A ‘weakness’ is a condition in a software, 
firmware, hardware, or service component that, under certain circumstances, 
could contribute to the introduction of vulnerabilities” 
(https://cwe.mitre.org/about/new_to_cwe.html). Following the vulnerability 
theory (https://cwe.mitre.org/documents/vulnerability_theory/intro.html) 
suggests that we need to have a PRODUCT implementing FEATURE by performing 
certain BEHAVIORS that operate on RESOURCES. I will assume these definitions in 
my disagreement below, and acknowledge that my basic definitions of some of 
these terms may be off.

 

The core question is therefore: is a license violation a vulnerability by the 
vulnerability theory used by the CWEs? I argue in the affirmative. You state, 
“No doubt that invalid licenses are a serious problem from the security 
perspective, but it's more a supply chain issue and legal problem.” Then a 
PRODUCT implementing software with a “supply chain issue” or “legal problem” to 
achieve its behavior on its resources produces an availability security impact 
against PRODUCT. If you want to identify it more generally as “supply chain 
issue” or “legal insufficiency,” it’s still a vulnerability that directly 
affects availability, incurs technical debt, and inflicts reputation/brand 
damage (https://cwe.mitre.org/cwraf/creatingyourownvignettes.html). I believe 
that more specific supply chain or legal issues are appropriate as well (and 
license violations would be a specific example), but these two general classes 
of vulnerabilities you’ve identified also meet the criteria for becoming a CWE. 
CWE-1357 almost meets some of this definition as well. Instead of a 
license-violating component being “not sufficiently trusted to meet 
expectations for security” (with availability being part of the security 
definition), it would be nice to have a CWE that can refer to a component 
(trusted or not) that in fact does not meet security expectations because of 
“supply chain” or “legal” vulnerabilities.

 

Can you please further explain why a “supply chain issue and legal problem” is 
an abuse of the weakness definition? I feel like you acknowledge it’s a 
weakness while also saying it’s an abuse of the definition of a weakness, which 
indicates that I’m not understanding some of your argument. You lose me at 
“Invalid license itself cannot lead to a vulnerability just like that.” How is 
a coding weakness that affects availability not a distinct, individual 
vulnerability, regardless of what other vulnerabilities may also be in the 
software?

 

Sincere thanks for your response and interaction,

Jon

 

From: Przemyslaw Roguski  
Sent: Sunday, November 5, 2023 1:21 PM
To: Steven M Christey 
Cc: Hood, Jonathan W CTR USARMY DEVCOM AVMC (USA) 
; CWE Research Discussion 

Subject: [Non-DoD Source] Re: Request for CWE: Improper Licensing (UNCLASSIFIED)

 


You don't often get email from progu...@redhat.com.  
 Learn why this is important



Hi All, 

 

In my personal opinion, adding new weakness or renaming existing one to 
something more "licensing" related is abuse of the weakness definition.

We defined the weakness and vulnerability definitions a long time ago and any 
licensing problems do not fit the weakness use case.

 

The real-world examples provided in this thread indicate that there were 
license problems, but it's only a side effect of the problem.

Let me explain it in a different way. If you use a component or 3rd party 
software where it is a good, correct licence, but that component is not 
maintained correctly or has some vulnerabilities that some day lead to an 
exploit and successful attack, then it doesn't matter that there was a correct 
licence. 
No doubt that invalid licenses are a serious problem from the security 
perspective, but it's more a supply chain issue and legal problem.

Invalid licence itself cannot lead to a vulnerability just like that. There 
must be another weakness that can lead to a vulnerability.

Licences should be registered and monitored similarly to the components 
(artifacts) provenance, which is in scope of SBOMs.

Hence adding a new weakness licensing related is not a good idea in my opinion.

 

Best regards,

Przemek

 

Przemyslaw Roguski

Security Architect, Product Security