Draft message to Community ML - comments, please

[argonel]

Hi Anne,

I've researched this situation as extensively as I can. Having done
so, I do not believe the kind of action you propose is reasonable.

I examined the "Back to basics" page, which was created 17 August 2012
by someone else, 4 months prior to his first posting on it. (2
December 2012)

You state that his efforts on Bugzilla represent a barrage, but
careful examination reveals:

His first bug report was on 2012-04-21. Looks like he accidentally hit
submit before finishing the report, so there's a dupe there that
probably should be closed.

His work on the "Back to basics" page spans 6 months.

The other wiki pages he modified did not and do not bear any warning
or statement of purpose, in fact the Plasma Tasks page is linked with
an open invitation to add new tasks. Your instructions to him were
that unsubstantiated complains on the wiki had no value. One could
conceivably take the previous contents of the Plasma Tasks page to be
a bunch of complaints with no substantiation.

He did edit the main page with a comment that was deemed unacceptable,
it was removed 3 hours later by an admin.

His bug reports were not filed heedlessly, the "Back to basics" page
contains many references to bugs that he didn't report. It appears to
me to be a curated collection, not something created for the purposes
of defamation or harassment.

There are 41 issues on that page, 39 cite bugs. If one were to
consider bugs that are not his but cited on that page, you can give
him credit for de-duping 8 bugs. (Not bad for someone who doesn't seem
to be a developer.)

You requested he back up his claims twice, first on 21 December 2012.
By the end of that day he'd found and added 14 bug numbers, none of
them written by himself. On 19 March you gave him a suggestion for
deciding whether a bug report was of value or not. He still hadn't
filed any bugs himself.

On June 8, he filed most of the bugs. After each one it seems he
pasted the bug number into the wiki. In my opinion its not hard to
create 29 bugs in 6.5 hours from pre-written text. He has responded to
developer questions, adding clarifications and even closing bugs
himself.

With the exception of a brief argument with a developer, which I'd
characterize as one person misinterpreting the word choices made by
another, I didn't see anything alarming in the text of his reports.
His tone is perhaps less objective than preferable, and sometimes his
explanations of the bug is rough in the initial report. I have
co-workers that do a much less acceptable job. I don't even blame him
for a couple of the dupes, because its not necessarily obvious that
his report and the earlier report are related.

In at least two cases the person responding to the bug actually asks
for more reports to be filed, as something that is assumed to be
handled by a library is in fact implemented in the individual
applications.

In several cases he incorrectly added keywords to the bug reports, I'm
not sure the correct use of keywords is documented anywhere. In one
case a triager was much more brusque than I think is acceptable.

After having examined this evidence I do not see much to get excited
about. With some encouragement and guidance this person could turn out
to be a real asset.

Please advise if there is additional information to consider.

Regards, Eli

Footnote:

Status of 31 bugs:

(8) unconfirmed
(2) marked confirmed, one could likely be closed upon examination of
the commits.
(4) marked as fixed.
(4) marked wontfix.
(3) marked worksforme, based on either fixes in a newer version or
inability to reproduce.
(3) marked downstream or upstream.
(1) marked invalid, where I would have marked it upstream.
(6) marked duplicate

Also:
(8) not filed, presumably found as duplicates on his own
(2) no bug # on the wiki

So 29% of the bugs he filed were useful. I should be so fortunate with
bugs reported to *my employer*.

(sorry for the top post, it seems impossible to convince gmail to not
quote the previous text.)

On Mon, Aug 5, 2013 at 5:55 AM, Anne Wilson <annew at kde.org> wrote:
> -----BEGIN PGP SIGNED MESSAGE-----
> Hash: SHA1
>
> Some time ago a user started a page on http://community.kde.org on
> which he detailed things that he considered to be wrong with KDE
> software.  The comments were largely subjective, and in many cases
> demonstrated things that were not the general experience.  At that
> point he was asked to substantiate his complaints by pointing to bug
> reports.
>
> That strategy backfired, as he then began creating a barrage of bug
> reports, creating a good deal of work for triagers who have to decide
> whether the report is a duplicate of some other, or invalid for any
> reason.  Up to now people have been very patient with him, but it has
> come to our notice that he has crossed a line.
>
> He has now begun defacing other pages, adding links to his comments
> page and/or his bug reports.  This basically results in what should be
> "official" information, posted by developers of the application in
> question, being questionable, in that the reader cannot easily
> differentiate his input.
>
> In view of this we have reached the point where we believe that he
> should be banned from posting on the wikis, and his comments
> completely removed.
>
> Anne Wilson
> annew at kde.org
> -----BEGIN PGP SIGNATURE-----
> Version: GnuPG v1.4.14 (GNU/Linux)
> Comment: Using GnuPG with Thunderbird - http://www.enigmail.net/
>
> iEYEARECAAYFAlH/dpkACgkQj93fyh4cnBdY9wCfYHNrjNIp/O3nnwS9zaJBdHae
> zmwAnAyWgBtJEcU78ToayGlRXdvlZE7n
> =6jfB
> -----END PGP SIGNATURE-----
> _______________________________________________
> kde-www mailing list
> kde-www at kde.org
> https://mail.kde.org/mailman/listinfo/kde-www