[ntp:hackers] NTP 4.2.3p7 Released

David L. Mills mills at udel.edu
Thu Jun 22 01:37:59 UTC 2006


Todd,

NIST uses an older version of NTPv4 and has modified it in very minor 
ways. It probably does not have the most recent code. USNO does have 
more recent code, as it has Autokey, but maybe not the most recent code 
with the library in question.

In cases of NIST and USNO and all the ships at sea, is this a purple 
herring? If folks just build and install it, there is no legal issue. My 
concern are the rather large number of embedded systems that have more 
or less ripped off the code with my (our) blessing.

Dave

todd glassey wrote:

> ----- Original Message -----
> From: "David L. Mills" <mills at udel.edu>
> Cc: <hackers at ntp.org>
> Sent: Wednesday, June 21, 2006 2:24 PM
> Subject: Re: [ntp:hackers] NTP 4.2.3p7 Released
>
>
>> Harlan,
>>
>> It's really hard to answer inline; I have to convert everything to
>> fixed-width, then fight with the font when I insert a comment.
>>
>> Todd Glassey raised the issue. I didn't include his message because he
>> marked it private. He said the "new" I/O library was ISC and that it
>> could come with strings.
>
>
> It does - its copyrighted and the ISC's release MUST travel with it, 
> and in
> anything that is packaged with it. That means that this RELEASE ITSELF 
> MUST
> be provided in hardcopy form to the end user to be compliant to the 
> license.
>
>> I have no way to verify that except to invite
>> comment.
>>
>
> So David - we got that the ISC's Event Lib and its I/O processing lib 
> *(are
> they the same thing Paul?) are copyrighted IP's. And since its a
> copyrighted document and IP, its use comes with terms and conditions. The
> idea is whether it makes sense from a legal standpoint to build a 
> second set
> of legal requirements into those already existing with NTP and release
> management issues ithose create for NTP just so that the ISC has more
> justification for existing.
>
> Lets ask NIST if they plan on using the ISC's libraries on their Stratum-1
> Systems - I would be interested to hear Judah's response on this.
>
> ---
>
> By the way - as I have noted several times, Source Forge or one of those
> sites could easily source the code so the issues of whether the ISC is
> providing a critically unreplaceable service are void and not relevant.
>
> The KEY issue is whether it makes sense to encumber the already nebulous
> copyrights of NTP based on the tweaks done to it here and there, with
> another set of IP licenses, which could constrain other parts of the IP.
>
>> My comment about the bind consortium may be missplaced. I understood
>> that if you want the latest bind, you have to join the consortium and
>> cross palms with silver. I'd like to keep the cutting edge of NTP free
>> and open. There is of course always the danger that somebody will take
>> some snapshot, "enhance" it and sell it. That would make two bugstreams
>> and make it really hard to avoid glitches, but it would be legal.
>>
>> As for the event library, I recall it was you. We talked on the phone
>> and you said there might be strings. As for the one-word, have you
>> looked at the (real) copyright page recently?
>>
>> By the way, I would like to avoid any gloss of the copyright page except
>> the original HTML. An acceptable compromise might be to include in such
>> gloss a prominant statement that the gloss is not authoritative and only
>> the original can be considered accurate for legal purpose.
>>
>> I conclude from your message that my fears are unfounded and the
>> embedded NTP code in my EndRun CDMA receiver is legal.
>>
>> Dave
>>
>> Harlan Stenn wrote:
>>
>>> Dave,
>>>
>>> Should you and I take this discussion off-list?
>>>
>>>> I am told the I/O library included in the latest distribution may not
>>>
> be
>
>>>> freely available for derived works, as for example in several products
>>>> now on the market.
>>>
>>>
>>> Who told you, and exactly what I/O library are y'all talking about?
>>>
>>> I am not aware of anything in the base NTP code that has any
>>> restrictions in this regard.
>>>
>>>> My intent has always been that the entire
>>>> distribution is available free of charge and free of restrictions on
>>>> derived works. I violently object if NTP turns into something like the
>>>> gated or bind consortium.
>>>
>>>
>>> Is the last sentence above subsequent or consequent to the previous
>>> sentences I qouted from you?
>>>
>>> If consequent, would you please explain better?
>>>
>>> I understand completely what you mean about gated, but what do you mean
>>> about a 'bind consortium' and what, exactly, are your
>>
> concerns/objections?
>
>>>> I assume the terms and conditions of any
>>>> library, package or addon that appears in the distribution conform in
>>>> spirit to the blanket copyright notics. If this is not the case, hold
>>>> the train.
>>>
>>>
>>> I believe you and I are in agreement there.
>>>
>>> ...
>>>
>>>> This issue came up with the ISC event library and, when it was
>>>
> explained
>
>>>> to me, I elected not to use that. I hope this is not the case with the
>>>> I/O library.
>>>
>>>
>>> From what I can see, the ISC event library is covered by a copyright
>>> agreement that is *better* than the NTP copyright (they already made the
>>> one-word clarification change you and I have discussed), so what is (or
>>> was) the problem there?
>>>
>>> Who explained "it" to you?
>>>
>>> As we are about to convert to the ISC event library, it would be good
>>> for you and me to have a clear understanding and agreement of this
>>> point.
>>>
>>> H
>>>
>> _______________________________________________
>> hackers mailing list
>> hackers at support.ntp.org
>> https://support.ntp.org/mailman/listinfo/hackers
>
>



More information about the hackers mailing list