Team Reports 2016

Team Leaders are encouraged to present a report for their team. (alphabetic order)


AffiliateProgramme


Arbitration

This is a report about the arbitration cases per year (with the focus on 2015 and 2016). The focus is based on case statistics and their changes over the time. How many cases from when were handled at what time.

There are other interesting questions for report about arbitration (nature of cases, summary of rulings, team updates and activity within the team). Those are mostly covered in the quarterly reports. If possible there should also be some kind of report over them, but this will not be part of this overview.

General remark

All data in this report is from the point of view of 2016-06-30, the end date of the financial year.

Exceptions are marked, especially.

Some while ago, I started to take the dates from the arbitration cases, which you can find in the wiki, into a calculation tool and played around with them. As a result you can find some statistics, here.

This data is of interest if one wants to compare the workload (and activity) of the Arbitration team within years (and months). They give a context for the numbers for 2015 / 2016.

Regarding 2016 this data is only noted down until end of financial year of 2015/2016 = 2016-06-30. So it's only half a year.

Duration of cases

As a first topic I would like to address the duration of cases in the arbitration area. As this is something that I did not see covered, before.

To be able to do this, a lot of dates have to be extracted from the case files. Regrettably I was only able to do this for the closed cases, so far. It is much harder to do for the open cases as they may change and then the data needs to be updated, somehow.

Because of this the current presentation will only be about how long the cases that were closed did take.

The first column is the difference to the last duration (so for example in line "quarter year" it's cases that took longer than a month but not longer than 3 months).

The second table shows how many cases in total were closed until then, including the earlier ones.

The last two columns show the same as percentage of all closed cases.

duration of cases after filing

closed after

diff

total

diff %

total %

1 day

21

21

5%

5%

1 week

33

54

8%

14%

1 month

41

95

10%

24%

quarter year

65

160

17%

41%

half year

68

228

17%

58%

1 year

66

294

17%

75%

2 years

67

361

17%

92%

3 years

20

381

5%

97%

> 3 years

11

392

3%

100%

About a 1/4 of the cases could be closed within the first month after they were filed. Over 1/2 of the cases could be closed in the first half of a year after filing and 3/4 of the cases could be closed after a year. After two years there was not much change any more.

So cases were likely to be closed within the first two years after filing [or not, as those cases are not in the presentation]. But most of them were closed after a year or even half a year.

An interesting other perspective is how fast cases could be closed after they were picked up by a CM/Arbitrator team.

The columns are of the same kind.

duration of cases while running

closed after

diff

total

diff %

total %

1 day

90

90

23%

23%

1 week

47

137

12%

35%

1 month

97

234

25%

60%

quarter year

58

292

15%

74%

half year

35

327

9%

83%

1 year

33

360

8%

92%

2 years

21

381

5%

97%

3 years

6

387

2%

99%

> 3 years

5

392

1%

100%

Now about 1/4 of the cases could be handled within one day. Way over 1/2 were closed within one month. Within three months, about 3/4 were could be finished. If cases took over a year not much was changed afterwards.

This shows, three things (Keep them in mind for the end of this report):

  1. The waiting time of cases is a none-trivial part in the duration of cases. Or re-phrased: There were too many cases filed for arbitration to manage them quickly.
  2. Cases that were picked up were mostly dealt with quickly.
  3. If cases could not be closed within three months, they were mostly dragged in length.

A lot of those longer cases are of some nature where other teams did not respond for some while or needed some more time or the like. Or which were possibly uncomfortable for another team. Or where quite complex questions were asked (sometimes at the edge of what arbitration should be used for).

Also a lot of cases could have been closed some time earlier, but took a lot of time for documentation until they could be closed.

Any attempt to reduce the number of waiting unnecessary cases seems to be helpful for closing cases in general. As this would reduce the waiting time.

It also would help if there were more people active in the arbitration area.

From experience and when looking at the cases a huge amount of time is used for documentation/transparency. Maybe we are going too far with this. But on the other hand we get positive feedback that this is quite helpful. We also get complaints that we do not document enough. (That is a usual element of recent "appeals".)

The documentation should be the task of the Case Manager. When we look at the boost of handled cases when new arbitrators are trained first as Case Manager and before they pick up arbitration themselves, it also shows that much more cases could be closed quickly. (You can see this below. By comparing "new names" in the list of closed cases.)

I repeat my request to board / DRO to consider to allow persons of the support team to be Case Managers again, as this is also what DRP states how it should be. It would at least help to ask if there would be volunteers for such tasks.

It also would help to cover parts of the documentation with a software solution, somehow.

Distribution of cases over time

The following table shows in which years the cases of which years were closed. (Rows are the closing years, columns are the years where the dispute was filed.)

closed Cases from years

closed in year

opened in year

cases

2007

2008

2009

2010

2011

2012

2013

2014

2015

2016

all

2007

2

2

2008

2

3

5

2009

1

57

58

2010

38

56

94

2011

8

35

55

98

2012

18

15

32

2013

3

17

11

7

39

2014

4

6

6

19

35

2015

2

2

2

6

3

15

2016

1

1

8

2

2

14

open

4

14

13

22

12

33

18

3

119

all

4

4

108

111

109

56

25

66

23

5

511

The comment to this table in 2014 was:

This diagram is interesting as it shows, that mostly cases from the last few years get closed. Older cases currently do not seem to have a good chance to get closed.

There are probably multiple reasons for this. One is that most of the older cases already have a team treating it, which probably got to a state where it was not possible to proceed, further.

New arbitrator who are looking for a case will not be able to work on those old cases. They tend to pick up cases which seem to be a little bit urgent. This mostly are new (or newer) cases, as the old ones which were urgent are mostly solved (or at least handled by another arbitrator).

Even as this did not completely change in 2015/2016, as some new cases just are urgent and have to be done quickly, this pattern was at least somewhat broken. There were older cases that got some focus, even as the majority was from 2014.

As this was the year (and remains to be the year) with the most open cases, this is not as astonishing.

Currently (starting at the end of the time covered by this report and afterwards) there is some initiative to get rid of a lot of the old cases, some more.

Status of cases

The following table is about the status of the filed cases over the years and not only about which were closed.

It gives an idea which cases are open to be picked up and treated by new arbitrators and which may be "stalled" with a sitting arbitrator in that case.

As I've presented a comparable table in 2014 I kept the data for 2014 about cases in work and in "waiting for an Arbitrator" and added the status of 2016-06-30 as "now". By this it is possible to compare the progress some more. It gives a hint about how many cases in work were finished or what new cases were picked up.

The last column is split as well for cases in 2015 and those in 2016. I decided against a split for financial years, here, as that would have to reach into 2014, again and in total would be much harder to create and to analyse.

Status of cases over the years

year

filed

closed

open

in work

waiting

closed

2014

now

2014

now

2015

2016

2007

4

4

0

0

0

0

0

0

0

2008

4

4

0

0

0

0

0

0

0

2009

108

104

4

4

4

1

0

0

1

2010

111

97

14

15

13

2

1

2

1

2011

109

96

14

14

12

2

2

2

0

2012

56

34

21

10

8

13

13

2

0

2013

25

13

12

4

4

8

8

0

0

2014

66

33

33

17

10

30

23

6

8

2015

23

5

18

0

4

0

14

3

2

2016

5

2

3

0

1

0

2

0

2

Total

511

392

119

64

56

56

63

15

14

At the end of 2014 9 cases were in the OTRS.

At the moment of the creation of this report the following number of possible disputes are in the OTRS:

If we do not count the disputes in the OTRS (as for some it seems to be unclear if they will become cases), one case more could be closed than were opened in the period covered in this report. If we assume that all disputes in the OTRS will become cases, there were 3 cases more filed than could be closed.

When we compare the "in work" with "waiting" columns, the number of "waiting" cases, compared to those "in work" has increased. They more or less have switched over. Some running and old waiting cases were closed while the cases filed in the report period

Filed cases

The report of 2014 contained the following comment:

Until the end of 2011 also cases where non-assurer wanted to close their account were treated by Arbitration and afterwards they could be handled by Support.

If one takes this into consideration one has to say that 66 new cases is a lot to manage by volunteers. That is one new case all 5.5 days!

For 2015 and 2016 the picture is completely different. A lot less cases are filed, especially in 2016.

But the nature of the cases has also changed.

In 2014 a huge majority of cases were filed by only a hand full of claimants. Also three types of cases were relatively often filed in 2014:

  1. delete account cases, due to a change of CCA
  2. reviews over bugs / emergency activities
  3. Inter-team disputes, where a member of one team filed against another team or team member

In 2015 the later type dominated completely. But this time it was not just between teams, but it was mostly against arbitrators (including but not limited to appeals). Also two cases were filed about "defamatory statements" And two times someone filed a dispute to protect arbitrators against possible attacks out of other teams.

Only 12 out of the 27 filed cases (incl. OTRS) were NOT of this nature. Among the remaining cases there are also some review requests over execution of some activities and the like, where some may also go in a comparable direction.

In total the number of filed cases dropped drastically, with big gaps starting with May 2015.

Filed cases per year and month

Month

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Total

2007

2

1

1

4

2008

1

2

1

4

2009

1

7

6

9

23

13

9

7

1

22

10

108

2010

11

23

21

6

6

4

8

9

6

7

3

7

111

2011

11

11

15

7

6

11

10

7

10

16

5

109

2012

9

4

7

7

6

7

6

2

1

6

1

56

2013

1

1

1

3

2

1

2

1

3

1

4

5

25

2014

9

3

3

8

3

6

3

3

13

10

4

1

66

2015

6

2

1

2

1

3

2

2

3

1

23

2016

1

1

3

-

-

-

-

-

-

5

Total

47

45

56

40

34

57

36

36

40

31

58

31

511

In earlier years there were also now and again months without cases. But that was quite seldom and the number of cases was mostly much higher, anyway.

This gets especially drastic in 2016, where we are speaking about the first half of 2016. Only 5 cases were filed in that time. After arbitration had started to "get running" in 2009 never so few disputes were filed in half a year at all.

One element may have been, that in May a precedents ruling was given that made it nearly unnecessary that delete account cases were filed with arbitration but could be closed by support, even if they were assurer accounts. There was already another case in that direction around the change of year 2014 / 2015. This alone does explain the effect.

A much bigger effect had the escalation between a group of people around the first board of 2015 / 2016 and arbitration.

While that board was installed only three cases were filed in total:

  1. a review over parts of the AGM where that board was installed
  2. a review over the "suspension" of arbitrators by that board
  3. a review over the call for an SGM to remove that board

Closed Cases

This is mirrored by the closed cases statistic:

Closed cases per year and month

Month

Jan

Feb

Mar

Apr

May

Jun

Jul

Aug

Sep

Oct

Nov

Dec

Total

2007

2

2

2008

1

1

2

1

5

2009

3

1

5

8

6

1

4

1

14

15

58

2010

8

9

3

2

11

7

9

5

6

25

9

94

2011

11

18

18

16

2

1

2

4

8

6

7

5

98

2012

10

4

3

2

4

1

1

1

1

2

1

2

32

2013

2

2

6

1

1

1

22

4

39

2014

4

3

3

1

4

1

2

3

6

2

6

35

2015

4

6

1

1

2

1

15

2016

1

13

-

-

-

-

-

-

14

Total

36

41

42

23

13

41

18

17

25

21

72

43

392

The closed cases dropped again drastically starting with May 2015. Again this is the time where the review over one of the arbitration cases (or more precisely over an arbitrator) was filed.

Afterwards only 4 cases were closed in 2014. Including that review-case.

Worse: between November 2015 and May 2016 no case at all was closed. This is the period of the first board of financial year 2015/2016.

It is hard to describe such a long period of 5 month other than with "arbitration was stopped".

But if one is honest, the time where arbitration was nearly stopped is even longer. Within one year (mid May 2015 to April 2016) only 4 cases were closed. There never were so few closed cases within 12 months, ever.

All in all we have lost more or less one year of arbitration time.

To finish with something positive: Interestingly, this is not directly visible, if one looks only at the number of closed case. Because after the new board was installed the activity increased drastically, again. The reason for this is mostly that a new member joined the arbitration team and training for him started.

This lead to the fact that in the first half of 2016 nearly as many cases as in complete 2015 were closed. (Or more precise in the second quarter of 2016, after the new board was installed.) This is a comparable number to earlier years.

Team changes

None of the resigned arbitrators so far was active to hand over their cases. In total

Additional events to report, which were affecting arbitration team (including appeals)

Remark: The two arbitrators (Eva Stöwe and Philipp Dunkel) that were mostly affected by above board activities were handling

This report was written by


Assurance


ATE


Audit Team


Critical System Administrator Team


Events team support


Education

Issuing of CATS certificates has come to halt during early 2016.

Initially the support crisis of CAcert prevented the verification of Assurer status of applicants, but now I cannot motivate myself to invest the necessary work into the whole verification and creation process.

At the moment I don't assume that I will invest significant work into the education team during the next year. If there would be any people interested in the position of Education Officer I'd be happy to resign this position. As I currently do not know of such a person I'll refrain from resigning to act at least as a contact person.

I'll continue to do critical support for CATS, though it would really make much sense if there'd be someone else able to do the job in case I am not able to react quickly enough. I have tried to document CATS as good as I can at Software/CATS/Developer, but my experience tells me that this will probably not be enough for someone new to take over the job quickly.


EventsTeam

As in previous years CAcert was active in 2015/2016 at several Linux- and OpenSource events.

Sometime the attendance at these events was organized by members of the events-team, sometimes it was organized by volunteers not being part of the events team.

On Froscon 2015 a development-room was booked, several meetings were organized and assurances done at our booth.

On OpenRheinRuhr, Fosdem, Chaos Communication Congress and at "Chemnitzer Linux-Tage" we had a stand to inform about current progress at CAcert, answer questions and assure applicants.

But ...

Up to today nearly the complete team changed. The teamlead resigned and most team-members got silent in the first half of 2016. Currently board helds the events-teamlead.

Currently there is a quite small team organizing attendance at Froscon and OpenRheinRuhr (and probably other events, too).


Infrastructure

The Infrastructure team lost most members in the aftermath of the SGM in April. Efforts have been started by Jan to get properly versioned infrastructure documentation that is automatically built from sources in Git. The result is browsable at https://infradocs.cacert.org/. The state of the infrastructure has been described in a series of mails in June:

The migration to more auditable system administration via Puppet has been started in November by setting up a puppet master container on [SystemAdministration/Systems/Infra02] and but has not progressed as much as desired due to time constraints. Missing administrators for most containers are still a pressing issue.


New Root & Escrow Project (NRE)


Organisation Assurance Team


Policy Group

Since the update of the CCA and setting the other policies to POLICY status no further policy decisions were done.

However in January of 2015 there was an issue with the correct version of the CCA. Due to a mixup in the code during the transport to production server for the other policies the old version of the CCA was enabled, again. This was fixed later and affected members were informed by a mail.

A topic that was shortly touched twice, was a possible update of the CPS regarding especially additional, explicite signing options for the root certificates. But both times the discussion stopped quickly. But it should be marked as a topic to come back to, in dure time.

Following a ruling in a20150420.1 multiple discussions about the relation of the different heads of power came up from different directions. In this context a proposal for a new policy Policy On Heads Of Power was provided and a discussion was started. One element of this proposal is to add a new executive head called "Cabinet" that could be placed in the community.

So far there were only proposals but no deep discussion. But the number of started discussions in this direction as well as some need to disentangle CAcert Inc out of other policies, hopefully will bring us back to those topics, again. As a need for such a policy seems to be there.

Additional to those content wise remarks it has to be noted that after the AGM 2015-11-22 there were some motions from board to interfere with policies or otherwise with topics in the authority of Policy Group. Those motions later were reverted by the board that was installed by SGM 2016-04-10.

There hopefully will be a more elaborate report after the change of the year to Policy Group and community.


PublicRelations


Software Development Team

The software development team including software testers, developers and assessors continued their work to improve and fix the existing CAcert software.

Two milestones were reached within 2015/2016, but unfortunately it was not possible to finish this:

The first one was as test of doing the root-key-resigning-procedure during Froscon 2015 (August). After detecting small flaws there (mainly typos) a second run was done in the evening.

In March the resigning of the production root key was done at BIT datacenter in Ede.

But:

Up to today the software-assessment team is nearly not existent: The teamlead resigned some days ago (just before the end of the financial year 2015/2016). The remaining members are either silent or quite busy with other tasks.

Since August 2015 no software-patches were sent to the critical team.


Support Team

To give the current status of support some details of the previous financial year should be known. During the FY 2014/2015 there was a suspension of a support member to force a re-training for him. Unfortunately this (support-internal) training was not done within FY 2014/2015.

In Summer 2015 (and especially on Froscon) this escalated badly. The result was a resignation of the only being active supporter at that time. Board then tried to bring up support to work again, unfortunately without success until the AGM.

After the AGM things changed:

The supporter came back and started doing support-work again, a "retraining" was done for the other supporter, but he did not get back access until the SGM in April. As far I know this "retraining" was not sufficient enough to give a "GO" from abritration and/or board.

Immediately after the SGM both members resigned, so support was in a non-working state again.

At the end of May it was possible get a new support-member into the team by an arbitation ruling. This member is currently working trough over 800 support-tickets.

Triage:

Triage is the "preselection"-team within the suport-team. Until the SGM there was one member sorting the incoming mails and (sometimes) forwarding or informing the committee about urgent issues. After he resigned around the SGM, mails sent to support got stuck there.

After it was possible to restart support in May/June there was a call for new members for triage and support. Currently there are two members doing Triage and (maybe) trained for doing support work later.


AGM/TeamReports/2016 (last edited 2016-12-18 19:21:29 by EvaStöwe)