Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Andy B
On February 26, 2019 at 11:22:27 AM, Nate Graham
(n...@kde.org(mailto:n...@kde.org)) wrote:

>  On Tue, 26 Feb 2019 10:37:56 -0700 Christian Loosli wrote 
> > That all in mind is why I think the approach was maybe not the best one and
> > should be reconsidered for future events and articles.
> >
> > Thanks and kind regards,
> >
> > Christian
>
>
> I'm in agreement with Christian about this. Conducting this process behind 
> closed doors seems to have backfired, as evidenced by the technical problems 
> that happened anyway and the controversy and drama that we're all now 
> embroiled in. Regardless of intentions, these results should be a clear sign 
> that the process didn't work and needs to be re-examined. And it's very 
> disappointing to hear that a part of the reason for this process was 
> specifically to exclude a particular person. It's no great secret that Paul 
> and Jonathan don't have a great working relationship, but this problem needs 
> to be addressed openly. The two of you trying to work around one another, 
> occasionally writing passive aggressive emails on the mailing list is not how 
> the problem will get solved; instead it will just get bigger over time.
>
> We don't need secrecy, politicking, and personality-driven conflicts like 
> this in KDE. When there's a problem, let's handle it like adults so we can 
> all get back to the business of making the finest open-source software the 
> world has ever seen.
>
> Nate
>

Thank you all for your comments. I am sure now it is a lesson learned
and we will be more vigilant of these situations in the future. It is
part of the challenge of working with a large community across many
countries.

We all make assumptions that we feel are right but when placed in a
larger context they don’t always work the way it was intended.

I feel the community has spoken and we are in agreement. Let’s learn
from it and now move forward with the amazing work we deliver to many
people in the world.

Hugs.

Andy
KDE Board Member


Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Nate Graham
  On Tue, 26 Feb 2019 10:37:56 -0700 Christian Loosli  
wrote  
 > That all in mind is why I think the approach was maybe not the best one and  
 > should be reconsidered for future events and articles.  
 >  
 > Thanks and kind regards,  
 >  
 > Christian 


I'm in agreement with Christian about this. Conducting this process behind 
closed doors seems to have backfired, as evidenced by the technical problems 
that happened anyway and the controversy and drama that we're all now embroiled 
in. Regardless of intentions, these results should be a clear sign that the 
process didn't work and needs to be re-examined. And it's very disappointing to 
hear that a part of the reason for this process was specifically to exclude a 
particular person. It's no great secret that Paul and Jonathan don't have a 
great working relationship, but this problem needs to be addressed openly. The 
two of you trying to work around one another, occasionally writing passive 
aggressive emails on the mailing list is not how the problem will get solved; 
instead it will just get bigger over time.

We don't need secrecy, politicking, and personality-driven conflicts like this 
in KDE. When there's a problem, let's handle it like adults so we can all get 
back to the business of making the finest open-source software the world has 
ever seen.

Nate



Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Christian Loosli
Am Dienstag, 26. Februar 2019, 17:54:38 CET schrieb Paul Brown:

Hello Paul, 

> > The workboard item is https://phabricator.kde.org/T10477,  it wasn't
> > tagged KDE promo, it wasn't sent to the dot-editors list
> 
> This is true. However, there were good reasons for keeping things under
> wraps:
> 
> Firstly nobody wanted it to pop up on some place like Reddit, have a bunch
> of people cascade into the servers before they were ready, then moan on
> line how KDE can't get anything right and "bring back KDE 3!". Safeguarding
> KDE's reputation is one of Promo's prime directives.

well, in my opinion we managed quite the opposite, to be honest. 
Not only did we publish an article that was wrong and looked a bit 
unprofessional, personally I also think having at least some more testing 
before going public by a group would have been helpful. 

First of all, we did have performance issues when it got live. First due to a 
bug that is, as far as I am aware, now fixed. Now due to the bridge being the 
shared matrix bridge, which is under quite a load, hence having a couple of 
seconds of delay between sending messages and seeing them / message order 
mixup, which is to be solved (likely by switching over to a dedicated bridge)

That, and the few bugs reported (and some of them fixed) in the matrix kde 
support channel right after release are all things I think could have been 
ironed out before release if tested. 

In addition to that, the internal-only approach seems to have lead to a rather 
biased / sided article which, according to Lazlo, still comes across as biased 
/ sided. This is of course debateable, but I can see where he is coming from. 

So I think some "not getting it right" moaning is warranted, and the "bring 
back KDE 3" people will always be there, and everything looks a nail if all 
you have is a hammer, so no matter what we do and how we do it: it can be used 
as an argument.

That all in mind is why I think the approach was maybe not the best one and 
should be reconsidered for future events and articles. 

> Cheers

Thanks and kind regards, 

> Paul

Christian




Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Laszlo Papp
Hi Paul,

Thank you for your feedback.

I personally still have concerns with the dot article. I am not a native
English speaker, so I am happy to stand corrected if I am wrong.

But to me, it still reads with some bias towards Matrix. I would like the
article to read as Matrix is yet another choice, not a better or worse
choice than IRC. This is to respect everyone's choice for communication
equally and fairly. The everyone to their own principle.

The principle in KDE is the manifesto, I assume. That is matrix as well as
IRC agnostic. This is a good thing.

I think it would be better not to claim at large that Matrix or IRC is
better worse than the other. It is up to the individual non-paid volunteers
to decide on pair-to-pair or group basis to decide what works for them
best. In this way, I would like the article in the end of the day to read
fairly and without favoritism.

Best regards,
Laszlo Papp

On Tue, Feb 26, 2019 at 4:54 PM Paul Brown  wrote:

> Hello Jonathan,
>
> > The workboard item is https://phabricator.kde.org/T10477,  it wasn't
> > tagged KDE promo, it wasn't sent to the dot-editors list
>
> This is true. However, there were good reasons for keeping things under
> wraps:
>
> Firstly nobody wanted it to pop up on some place like Reddit, have a bunch
> of
> people cascade into the servers before they were ready, then moan on line
> how
> KDE can't get anything right and "bring back KDE 3!". Safeguarding KDE's
> reputation is one of Promo's prime directives.
>
> So far, by the way, so good: no outlet and no social media platform has
> had
> anything bad to say about the decision.
>
> Getting back to being discrete until ready, the people of Matrix were
> particularly and understandably nervous about this deployment. Apart from
> the
> hit to their reputation a botched deployment would have had, they also
> have
> patrons and a disastrous, unplanned and premature release could give their
> sponsors second thoughts about supporting them again.
>
> > and I wasn't pinged
>
> As for this... well, as we are being frank here: You are not the most
> discrete
> of people and not anybody's first choice for keeping things quiet.
>
> Don't get me wrong, being forthcoming is a much appreciated quality in
> most
> situations involving Free Software. Your no-nonsense and upfront
> personality
> has made you a beacon of transparency, not only for KDE, but in FLOSS in
> general.
>
> But those particular personality traits would've not helped with this
> task,
> unfortunately.
>
> > (I'm the only active volunteer Dot editor).
>
> I'm not sure this is true, but if it is, it is probably a great time to
> ask
> for more volunteers.
>
> Hello Community Mailing List readers! Posts for Dot editors are now open!
> Requirements are good language skills. Knowledge of editing and
> copywriting a
> plus. Drop buy the Promo mailing list, IRC or Matrix room and we'll get
> you
> sorted.
>
> > I've tried to discuss problems in promo with the e.V. board and CWG in
> > the past when long term contributors have left, when the team was
> > changed from a community team to a closed access team,
>
> This is, to put it mildly, a misstatement. Anyone is free to join the
> Promo
> working group and everybody there helps decide what Promo does and shapes
> the
> strategy we follow.
>
> > when our
> > mailing lists were micro managed or when I was insulted for organising
> > a conference stall
>
> Are you sure? What did they say? What were the circumstances? I find it
> hard to
> believe it was gratuitous. Something like "Don't help us, you moron!"
> sounds
> way out of character of everybody I know in Promo. I feel more context is
> needed.
>
> But, before you answer, is it relevant to this discussion? Is it relevant
> to
> the discussion about Promo? If not, maybe start another thread or take it
> to
> CWG and see if they can solve it.
>
> > but I've only been dismissed or ignored
>
> Anything but, I'd say. When you wanted a policy for accessing social
> media,
> Promo dropped everything to make it happen. You have been active and
> contributing to Promo ever since (something which is much appreciated, by
> the
> way) and everything you have suggested has been taken into consideration
>
> ... Unless I have missed something. If this is the case, please tell what
> it
> is and we'll give it a second go.
>
> > and the
> > community at large seems happy for that to happen so I can't offer any
> > assurances of changes.
>
> Maybe that is because the community at large sees things differently from
> you.
> Maybe you are confabulating several non-related things which don't all
> reflect
> reality. Maybe if you took a step back and tried to see things from other
> people's points of view, you may see why sometimes not everything can be
> carried out exactly as you would wish. Making concessions is part of
> working
> within a community. Not everything important for you is important for
> everybody else.
>
> Cheers
>
> Paul
> --
> 

Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Paul Brown
Hello Jonathan,

> The workboard item is https://phabricator.kde.org/T10477,  it wasn't
> tagged KDE promo, it wasn't sent to the dot-editors list 

This is true. However, there were good reasons for keeping things under wraps:

Firstly nobody wanted it to pop up on some place like Reddit, have a bunch of 
people cascade into the servers before they were ready, then moan on line how 
KDE can't get anything right and "bring back KDE 3!". Safeguarding KDE's 
reputation is one of Promo's prime directives.

So far, by the way, so good: no outlet and no social media platform has had 
anything bad to say about the decision.

Getting back to being discrete until ready, the people of Matrix were 
particularly and understandably nervous about this deployment. Apart from the 
hit to their reputation a botched deployment would have had, they also have 
patrons and a disastrous, unplanned and premature release could give their 
sponsors second thoughts about supporting them again.

> and I wasn't pinged

As for this... well, as we are being frank here: You are not the most discrete 
of people and not anybody's first choice for keeping things quiet.

Don't get me wrong, being forthcoming is a much appreciated quality in most 
situations involving Free Software. Your no-nonsense and upfront personality 
has made you a beacon of transparency, not only for KDE, but in FLOSS in 
general.

But those particular personality traits would've not helped with this task, 
unfortunately.

> (I'm the only active volunteer Dot editor).

I'm not sure this is true, but if it is, it is probably a great time to ask 
for more volunteers.

Hello Community Mailing List readers! Posts for Dot editors are now open! 
Requirements are good language skills. Knowledge of editing and copywriting a 
plus. Drop buy the Promo mailing list, IRC or Matrix room and we'll get you 
sorted.

> I've tried to discuss problems in promo with the e.V. board and CWG in
> the past when long term contributors have left, when the team was
> changed from a community team to a closed access team,

This is, to put it mildly, a misstatement. Anyone is free to join the Promo 
working group and everybody there helps decide what Promo does and shapes the 
strategy we follow.

> when our
> mailing lists were micro managed or when I was insulted for organising
> a conference stall

Are you sure? What did they say? What were the circumstances? I find it hard to 
believe it was gratuitous. Something like "Don't help us, you moron!" sounds 
way out of character of everybody I know in Promo. I feel more context is 
needed.

But, before you answer, is it relevant to this discussion? Is it relevant to 
the discussion about Promo? If not, maybe start another thread or take it to 
CWG and see if they can solve it.

> but I've only been dismissed or ignored

Anything but, I'd say. When you wanted a policy for accessing social media, 
Promo dropped everything to make it happen. You have been active and 
contributing to Promo ever since (something which is much appreciated, by the 
way) and everything you have suggested has been taken into consideration

... Unless I have missed something. If this is the case, please tell what it 
is and we'll give it a second go.

> and the
> community at large seems happy for that to happen so I can't offer any
> assurances of changes.

Maybe that is because the community at large sees things differently from you. 
Maybe you are confabulating several non-related things which don't all reflect 
reality. Maybe if you took a step back and tried to see things from other 
people's points of view, you may see why sometimes not everything can be 
carried out exactly as you would wish. Making concessions is part of working 
within a community. Not everything important for you is important for 
everybody else.

Cheers

Paul
-- 
Promotion & Communication

www: http://kde.org
Mastodon: https://mastodon.technology/@kde
Facebook: https://www.facebook.com/kde/
Twitter: https://twitter.com/kdecommunity




Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Scott Harvey
Okay, I can see the point made by Laszlo and Helio. I retract my request
after further consideration.

I'm a part of the community, as is this discussion, and so I suppose it
should stay on the community mailing list.

Apologies for intervening... please continue.

-Scott

On Tue, Feb 26, 2019 at 8:49 AM Laszlo Papp  wrote:

> Hi Scott,
>
> I can see your point, so not trying to challenge it.
>
> I just feel that open means open, not open like in facebook context. In a
> completely open environment, it is not just the success and easiness that
> becomes available, but some respectful and fair arguments, difficulties,
> etc, that need addressing.
>
> I feel that KDE as a community will eventually benefit from feedback and
> discussions like this if the community takes proper actions going forward.
>
> Best regards,
> Laszlo Papp
>
> On Tue, Feb 26, 2019 at 2:31 PM Scott Harvey  wrote:
>
>> Jonathan, et al -
>>
>> Can I respectfully ask that this debate/dispute be moved elsewhere?
>>
>> I've been on hiatus from my role as a minor KDE contibutor for a few
>> months. It's not encouraging to resume paying attention only to find
>> another argument in progress.
>>
>> I suppose it could be argued that this maillist is intended for community
>> discussion and that this is indeed a community issue... I just don't feel
>> it's good for morale (mine, at least).
>>
>>
>> -Scott (sharvey)
>>
>> On Tue, Feb 26, 2019 at 7:34 AM, Jonathan Riddell 
>> wrote:
>>
>> The workboard item is https://phabricator.kde.org/T10477 , it wasn't
>> tagged KDE promo, it wasn't sent to the dot-editors list and I wasn't
>> pinged (I'm the only active volunteer Dot editor). I've tried to discuss
>> problems in promo with the e.V. board and CWG in the past when long term
>> contributors have left, when the team was changed from a community team to
>> a closed access team, when our mailing lists were micro managed or when I
>> was insulted for organising a conference stall but I've only been dismissed
>> or ignored and the community at large seems happy for that to happen so I
>> can't offer any assurances of changes. Jonathan On Tue, 26 Feb 2019 at
>> 11:46, Christian Loosli  wrote:
>>
>> Hi Jonathan, thanks for the wrap-up. I am less interested in pointing
>> blame, and more interested in - how this could have happened - what our
>> learnings are so this doesn't happen again in the future? It still is
>> unclear to me how non-true accusations without further explanation made it
>> into the article. Even for people who are not familiar with the subject,
>> this imho should never happen. If you are not sure, you don't throw around
>> accusations of things being insecure. It bothers me even more that there is
>> a lengthy discussion on the subject (and a follow up survey and result)
>> available to the people who participated in this, the article looked to me
>> like this discussion, survey and result (that we did put a lot of time and
>> effort in) were ignored. From what I gathered it even was given to the
>> right people to proof-read, but the article was released without waiting
>> for a reply. How can that happen, and why was it so urgent to push that
>> article out? So to avoid this in the future, I'd like to see us following a
>> process that does involved proof-reading by people familiar with the
>> subject, so we look as professional as we as KDE should be by now, and
>> usually are. As a last but not least, I'm also not terribly happy when
>> people involved were also the ones still, in public, making statements
>> against one of the technologies we decided to use and support, stating we
>> should abandon them. Together with the flawed article this doesn't look
>> good. I'd love to see people at least try to not let their personal views
>> bias them too much, especially not when a group decision was made. I have
>> my personal views and preferences on this too, but I try my best to accept
>> the decision taken and support it. Thanks and kind regards, Christian
>>
>>


Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Laszlo Papp
Hi Scott,

I can see your point, so not trying to challenge it.

I just feel that open means open, not open like in facebook context. In a
completely open environment, it is not just the success and easiness that
becomes available, but some respectful and fair arguments, difficulties,
etc, that need addressing.

I feel that KDE as a community will eventually benefit from feedback and
discussions like this if the community takes proper actions going forward.

Best regards,
Laszlo Papp

On Tue, Feb 26, 2019 at 2:31 PM Scott Harvey  wrote:

> Jonathan, et al -
>
> Can I respectfully ask that this debate/dispute be moved elsewhere?
>
> I've been on hiatus from my role as a minor KDE contibutor for a few
> months. It's not encouraging to resume paying attention only to find
> another argument in progress.
>
> I suppose it could be argued that this maillist is intended for community
> discussion and that this is indeed a community issue... I just don't feel
> it's good for morale (mine, at least).
>
>
> -Scott (sharvey)
>
> On Tue, Feb 26, 2019 at 7:34 AM, Jonathan Riddell  wrote:
>
> The workboard item is https://phabricator.kde.org/T10477 , it wasn't
> tagged KDE promo, it wasn't sent to the dot-editors list and I wasn't
> pinged (I'm the only active volunteer Dot editor). I've tried to discuss
> problems in promo with the e.V. board and CWG in the past when long term
> contributors have left, when the team was changed from a community team to
> a closed access team, when our mailing lists were micro managed or when I
> was insulted for organising a conference stall but I've only been dismissed
> or ignored and the community at large seems happy for that to happen so I
> can't offer any assurances of changes. Jonathan On Tue, 26 Feb 2019 at
> 11:46, Christian Loosli  wrote:
>
> Hi Jonathan, thanks for the wrap-up. I am less interested in pointing
> blame, and more interested in - how this could have happened - what our
> learnings are so this doesn't happen again in the future? It still is
> unclear to me how non-true accusations without further explanation made it
> into the article. Even for people who are not familiar with the subject,
> this imho should never happen. If you are not sure, you don't throw around
> accusations of things being insecure. It bothers me even more that there is
> a lengthy discussion on the subject (and a follow up survey and result)
> available to the people who participated in this, the article looked to me
> like this discussion, survey and result (that we did put a lot of time and
> effort in) were ignored. From what I gathered it even was given to the
> right people to proof-read, but the article was released without waiting
> for a reply. How can that happen, and why was it so urgent to push that
> article out? So to avoid this in the future, I'd like to see us following a
> process that does involved proof-reading by people familiar with the
> subject, so we look as professional as we as KDE should be by now, and
> usually are. As a last but not least, I'm also not terribly happy when
> people involved were also the ones still, in public, making statements
> against one of the technologies we decided to use and support, stating we
> should abandon them. Together with the flawed article this doesn't look
> good. I'd love to see people at least try to not let their personal views
> bias them too much, especially not when a group decision was made. I have
> my personal views and preferences on this too, but I try my best to accept
> the decision taken and support it. Thanks and kind regards, Christian
>
>


Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Helio Chissini de Castro
Yes, i agreed to move elsewhere..
...as long the intention is just hide from the community and bury in
different place

This discussion should be public and visible to KDE community, as this is
the KDE spirit ever. If something was done wrong, everyone should be able
to understand and help to fix

We are not a closed club, or are we ?

On Tue, 26 Feb 2019 at 15:31, Scott Harvey  wrote:

> Jonathan, et al -
>
> Can I respectfully ask that this debate/dispute be moved elsewhere?
>
> I've been on hiatus from my role as a minor KDE contibutor for a few
> months. It's not encouraging to resume paying attention only to find
> another argument in progress.
>
> I suppose it could be argued that this maillist is intended for community
> discussion and that this is indeed a community issue... I just don't feel
> it's good for morale (mine, at least).
>
>
> -Scott (sharvey)
>
> On Tue, Feb 26, 2019 at 7:34 AM, Jonathan Riddell  wrote:
>
> The workboard item is https://phabricator.kde.org/T10477 , it wasn't
> tagged KDE promo, it wasn't sent to the dot-editors list and I wasn't
> pinged (I'm the only active volunteer Dot editor). I've tried to discuss
> problems in promo with the e.V. board and CWG in the past when long term
> contributors have left, when the team was changed from a community team to
> a closed access team, when our mailing lists were micro managed or when I
> was insulted for organising a conference stall but I've only been dismissed
> or ignored and the community at large seems happy for that to happen so I
> can't offer any assurances of changes. Jonathan On Tue, 26 Feb 2019 at
> 11:46, Christian Loosli  wrote:
>
> Hi Jonathan, thanks for the wrap-up. I am less interested in pointing
> blame, and more interested in - how this could have happened - what our
> learnings are so this doesn't happen again in the future? It still is
> unclear to me how non-true accusations without further explanation made it
> into the article. Even for people who are not familiar with the subject,
> this imho should never happen. If you are not sure, you don't throw around
> accusations of things being insecure. It bothers me even more that there is
> a lengthy discussion on the subject (and a follow up survey and result)
> available to the people who participated in this, the article looked to me
> like this discussion, survey and result (that we did put a lot of time and
> effort in) were ignored. From what I gathered it even was given to the
> right people to proof-read, but the article was released without waiting
> for a reply. How can that happen, and why was it so urgent to push that
> article out? So to avoid this in the future, I'd like to see us following a
> process that does involved proof-reading by people familiar with the
> subject, so we look as professional as we as KDE should be by now, and
> usually are. As a last but not least, I'm also not terribly happy when
> people involved were also the ones still, in public, making statements
> against one of the technologies we decided to use and support, stating we
> should abandon them. Together with the flawed article this doesn't look
> good. I'd love to see people at least try to not let their personal views
> bias them too much, especially not when a group decision was made. I have
> my personal views and preferences on this too, but I try my best to accept
> the decision taken and support it. Thanks and kind regards, Christian
>
>


Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Scott Harvey

Jonathan, et al -

Can I respectfully ask that this debate/dispute be moved elsewhere?

I've been on hiatus from my role as a minor KDE contibutor for a few 
months. It's not encouraging to resume paying attention only to find 
another argument in progress.


I suppose it could be argued that this maillist is intended for 
community discussion and that this is indeed a community issue... I 
just don't feel it's good for morale (mine, at least).



-Scott (sharvey)

On Tue, Feb 26, 2019 at 7:34 AM, Jonathan Riddell  
wrote:
The workboard item is  ,  it 
wasn't

tagged KDE promo, it wasn't sent to the dot-editors list and I wasn't
pinged (I'm the only active volunteer Dot editor).

I've tried to discuss problems in promo with the e.V. board and CWG in
the past when long term contributors have left, when the team was
changed from a community team to a closed access team, when our
mailing lists were micro managed or when I was insulted for organising
a conference stall but I've only been dismissed or ignored and the
community at large seems happy for that to happen so I can't offer any
assurances of changes.

Jonathan

On Tue, 26 Feb 2019 at 11:46, Christian Loosli > wrote:


 Hi Jonathan,

 thanks for the wrap-up.
 I am less interested in pointing blame, and more interested in

 - how this could have happened
 - what our learnings are so this doesn't happen again in the future?

 It still is unclear to me how non-true accusations without further 
explanation
 made it into the article. Even for people who are not familiar with 
the
 subject, this imho should never happen. If you are not sure, you 
don't throw

 around accusations of things being insecure.
 It bothers me even more that there is a lengthy discussion on the 
subject (and
 a follow up survey and result) available to the people who 
participated in
 this, the article looked to me like this discussion, survey and 
result (that

 we did put a lot of time and effort in) were ignored.

 From what I gathered it even was given to the right people to 
proof-read, but
 the article was released without waiting for a reply. How can that 
happen, and

 why was it so urgent to push that article out?

 So to avoid this in the future, I'd like to see us following a 
process that
 does involved proof-reading by people familiar with the subject, so 
we look as

 professional as we as KDE should be by now, and usually are.

 As a last but not least, I'm also not terribly happy when people 
involved were

 also the ones still, in public, making statements against one of the
 technologies we decided to use and support, stating we should 
abandon them.

 Together with the flawed article this doesn't look good.
 I'd love to see people at least try to not let their personal views 
bias them
 too much, especially not when a group decision was made. I have my 
personal
 views and preferences on this too, but I try my best to accept the 
decision

 taken and support it.

 Thanks and kind regards,

 Christian






Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Jonathan Riddell
The workboard item is https://phabricator.kde.org/T10477 ,  it wasn't
tagged KDE promo, it wasn't sent to the dot-editors list and I wasn't
pinged (I'm the only active volunteer Dot editor).

I've tried to discuss problems in promo with the e.V. board and CWG in
the past when long term contributors have left, when the team was
changed from a community team to a closed access team, when our
mailing lists were micro managed or when I was insulted for organising
a conference stall but I've only been dismissed or ignored and the
community at large seems happy for that to happen so I can't offer any
assurances of changes.

Jonathan

On Tue, 26 Feb 2019 at 11:46, Christian Loosli  wrote:
>
> Hi Jonathan,
>
> thanks for the wrap-up.
> I am less interested in pointing blame, and more interested in
>
> - how this could have happened
> - what our learnings are so this doesn't happen again in the future?
>
> It still is unclear to me how non-true accusations without further explanation
> made it into the article. Even for people who are not familiar with the
> subject, this imho should never happen. If you are not sure, you don't throw
> around accusations of things being insecure.
> It bothers me even more that there is a lengthy discussion on the subject (and
> a follow up survey and result) available to the people who participated in
> this, the article looked to me like this discussion, survey and result (that
> we did put a lot of time and effort in) were ignored.
>
> From what I gathered it even was given to the right people to proof-read, but
> the article was released without waiting for a reply. How can that happen, and
> why was it so urgent to push that article out?
>
> So to avoid this in the future, I'd like to see us following a process that
> does involved proof-reading by people familiar with the subject, so we look as
> professional as we as KDE should be by now, and usually are.
>
> As a last but not least, I'm also not terribly happy when people involved were
> also the ones still, in public, making statements against one of the
> technologies we decided to use and support, stating we should abandon them.
> Together with the flawed article this doesn't look good.
> I'd love to see people at least try to not let their personal views bias them
> too much, especially not when a group decision was made. I have my personal
> views and preferences on this too, but I try my best to accept the decision
> taken and support it.
>
> Thanks and kind regards,
>
> Christian
>
>


Re: KDE now has its own Matrix infrastructure

2019-02-26 Thread Christian Loosli
Hi Jonathan, 

thanks for the wrap-up. 
I am less interested in pointing blame, and more interested in 

- how this could have happened
- what our learnings are so this doesn't happen again in the future?

It still is unclear to me how non-true accusations without further explanation 
made it into the article. Even for people who are not familiar with the 
subject, this imho should never happen. If you are not sure, you don't throw 
around accusations of things being insecure. 
It bothers me even more that there is a lengthy discussion on the subject (and 
a follow up survey and result) available to the people who participated in 
this, the article looked to me like this discussion, survey and result (that 
we did put a lot of time and effort in) were ignored.

>From what I gathered it even was given to the right people to proof-read, but 
the article was released without waiting for a reply. How can that happen, and 
why was it so urgent to push that article out? 

So to avoid this in the future, I'd like to see us following a process that 
does involved proof-reading by people familiar with the subject, so we look as 
professional as we as KDE should be by now, and usually are. 

As a last but not least, I'm also not terribly happy when people involved were 
also the ones still, in public, making statements against one of the 
technologies we decided to use and support, stating we should abandon them.
Together with the flawed article this doesn't look good. 
I'd love to see people at least try to not let their personal views bias them 
too much, especially not when a group decision was made. I have my personal 
views and preferences on this too, but I try my best to accept the decision 
taken and support it. 

Thanks and kind regards, 

Christian