cassandra-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From Jeffrey Jirsa <>
Subject Re: DataStax role in Cassandra and the ASF
Date Sun, 06 Nov 2016 18:39:01 GMT
Everything you said is accurate, and I don¹t think anyone¹s debating that.
What I¹m hoping to convey is the method of communication is such that a
SIGNIFICANT number of people interpret the nature of the communication as
unnecessarily antagonistic. You seem to think it¹s necessary, but the
reaction of the community clearly says otherwise.

A person can be 100% right and still come across as a jerk, and the CoC
instructs people to avoid doing so, because it¹s damaging to the

If you ask 100 random people who are neither Cassandra users/developers
nor ASF members about whether or not the communication from the ASF board
members is in this thread is professional, empathetic, friendly, and
likely to build a community, I suspect you¹d find a significant number
that would tell you the communication is none of those things. And THAT is
a problem, too (and it¹s NOT on the same level as mark issues, but if the
question is ³why did Datastax step back from the Apache Cassandra
project², it certainly helps explain why a company might want to do that).

Let¹s build a community, Jim.

On 11/6/16, 12:00 PM, "Jim Jagielski" <> wrote:

>Some clarification.
>Basically, there had been issues w/ DataStax and the PMC for a long,
>long time. It came somewhat to a head in Aug when there was
>a PR/Email about the "Cassandra Summit" with nary a mention
>of Apache at all. None.
>This was after months and months in trying to get DataStax to
>honor our marks. It was this final culmination which which
>resulted in a board member saying "makes me want to jettison".
>At which Jonathan Ellis expressed confusion on what the problem
>was and asking about the context, oblivious to the concern. Someone
>else noted that both the PMC and Cassandra had been "lectured" on
>trademark violations before and said that "one would assume that
>someone learned along the way." Someone then wondered whether
>these recurring issues where due to some fault in the PMC or
>just the normal, expect churn of their being a brand. He
>further stated: "I don't see how we can make it the responsibility
>of the PMC to catch these things". It was then noted that the
>CTO of DataStax is the PMC Chair, as well as co-founder. There
>was then further discussions and "education" on mark guidelines,
>again, with Jake and Aleksey. Aleksey, at least, admitted that
>"If your only success criteria is how well trademark policing is
>performed, then sure, we all failed..."
>More discussion.
>Around this time, one board member referred to below most certainly
>did characterize the "hammer-time" phrase as "premature and
>inflammatory". Others did not. To support that position I will add
>some cut/paste quotes from another director:
>  o Overall, there are a handful of issues here but they look to be easily
>    fixable and - with a little education - preventable in the future.
>  o Given the numbers and seniority of DataStax employees involved with
>    Apache Cassandra it is disappointing that these errors are being made
>    but people make mistakes
>  o The lack of proactive policing of trademarks by the Cassandra
>    PMC is what concerns me
>  o Given the history, I do think the board needs to take some form of
>    action. It has been suggested that the board remove all DataStax
>    employees from the PMC. I agree things are heading in that direction
>    I don't think we are there yet.
>It was after that that someone mentioned that they were on 3 PMC
>and never saw any mark issues with any PMCs other than
>Cassandra (this was a not a director speaking). That is when I
>replied w/ the "I've seen such issues..." response.
>Some take-aways:
> o Mark compliance issues have been ongoing for a long, long
>   time.
> o The PMC and its chair had been involved in these concerns
>   for a long, long time.
>Once all this was done, and this particular issue resolved. The final
>few Emails on the thread close it off with:
>  o Nobody has said commit privs should be removed. Some have discussed
>the potential of removing PMC responsibilities
>  o I would like to see some positive action from the Apache Cassandra
>PMC that they are working on managing this problem.
>  o We all seem to agree that the responsibility for enforcement falls
>first to the PMC, then on VP Branding, and then on the President.
>That is the saga of hammers.
>> On Nov 6, 2016, at 12:57 PM, Jeff Jirsa <> wrote:
>> Now that I have clarity on what can and can't be relayed to the
>>community / dev@, I'm going to reply to this email, and then I suspect
>>I'm done for today, because I'd rather watch football than reply to this
>> On Sat, Nov 5, 2016 at 6:30 AM, Mark Struberg
>><> wrote:
>> Having a bit insight how the board operates (being PMC-chair for 2
>>other TLPs) I can ensure you that the board did handle this very cleanly!
>> I'm going to disagree with this, in a way I hope lets everyone see
>>where things went wrong, and more importantly, the path forward to fix
>> The board correctly identified that Datastax had a majority of the PMC
>>and could exert control.
>> The board correctly identified that Datastax violated trademark
>>policies (multiple times).
>> The board correctly identified that the PMC was not adequately policing
>>Datastax (or really anyone, there were plenty of trademark issues to go
>> The board appears to have incorrectly attributed the lack of policing
>>to the fact that Datastax controlled the PMC. This is an honest mistake.
>>The real blame lies somewhere closer to a lack of understanding of
>>responsibilities, and a lack of visibility into what other parts of
>>Datastax were doing.
>> It's clear I'm not alone in this conclusion - you seem to say the same
>> PS: I strongly believe that the technical people at DataStax really
>>tried to do their best but got out-maneuvered by their marketing and
>>sales people. The current step was just part of a clean separation btw a
>>company and their OSS contributions. It was legally necessary and also
>>important for the overall Cassandra community!
>> Unfortunately, when faced with an example of a trademark issue, there
>>were two very senior members who replied with very hostile,
>>unprofessional responses. One forwarded the example to board@ and
>>private@ with a blanket statement about wanting to "jettison every
>>single Datastax employee from the Apache Cassandra PMC". Another replied
>>with "hammer time?", and  youtube links to Game of Thrones clips were
>>sent. One member of the board (properly, in my opinion) noted that their
>>reactions were premature and inflammatory. Other members of the ASF
>>noted (correctly) that in any sufficiently large organization, it takes
>>process and time to make sure marketing is aware of policies, and the
>>fact that no such process exists isn't cause to jettison the PMC, but it
>>should be something that is corrected.
>> What didn't happen, though, was any admission or acknowledgement that
>>the premature and inflammatory behavior was wrong on the part of the
>>very senior, very vocal folks that said it. Instead, they've continued
>>making inflammatory comments - often because problems continue to happen
>>where they need to be involved, but the tone is such that it's very easy
>>to interpret it as hostile, which makes it very difficult to keep peace
>>in the community.
>> It's often said that when the board acts, they act as a sledgehammer
>>because they have no scalpel. That's true, but the board never actually
>>swung the sledgehammer - they threatened it, but they never needed to
>>jettison every Datastax employee from the PMC, because the Datastax
>>employees actively worked in good faith to correct problems. Sometimes
>>that work was insufficient, and sometimes the PMC as a whole is less
>>responsive than we should be (because many of us are still learning). We
>>(the PMC) have been fairly open about acknowleding our shortcomings, and
>>working to correct them.
>> Unfortunately, while there was acknowledgement from the board that the
>>PMC acted to correct problems (visible in the minutes, we're TRYING to
>>do better), there's never been an acknowledgement that members of the
>>board acted inappropriately - there was, at most, a single statement
>>that it was out of frustration (which appears to be a
>>half-acknowledgement that it may be out of line, but nowhere near an
>>apology for being out of line).
>> I can't speak for Datastax, but if I were in their shoes, and someone
>>threatened to jettison me from the PMC for something I had no prior
>>knowledge of, and then continued to act in an aggressive manner without
>>ever acknowledging that they, too, were wrong, I would also distance
>>myself from that group - not a "take my ball and go home" mentality, but
>>a "these people act in ways that I don't understand, they seem overly
>>hostile, and I should protect myself from them". What's frustrating is
>>that it appears, in many ways, that basic empathy and professionalism on
>>the part of the ASF board members could have potentially prevented this
>>situation entirely. I suspect that members of the ASF who believe the
>>board handled this cleanly re-evaluate that assertion, and ask
>>themselves whether board members acted with empathy, friendliness, and
>>professionalism in their communication with Datastax.
>> If the members of the board take that recommendation to heart, and
>>re-read threads on private@ in an objective manner, and agree with my
>>assertion that they have room for improvement as well, I encourage both
>>the board and Datastax management to reconsider their decisions made in
>>the past few months, for the sake of the community.
>> Because that's why we're all here - the community.

View raw message