1 |
On 01/15/17 14:23, Michał Górny wrote: |
2 |
> |
3 |
> What do you think? |
4 |
> |
5 |
I think this proposal is utterly unworkable in practice. While the |
6 |
intention is rather obvious, and heavily geared toward actual |
7 |
contributing members of the community at large, the proposed |
8 |
definitional scope and structure are incompatible with actual workloads |
9 |
already in place. |
10 |
|
11 |
To provide some perspective to those unfamiliar with the actual volumes |
12 |
in consideration here, just on the forums there are typically several |
13 |
"users" manually banned per day for posting spam, and perhaps a dozen or |
14 |
two profiles manually banned because the profiles themselves were spam, |
15 |
in addition to that there are typically hundreds (in some cases |
16 |
thousands) of accounts which are effectively automatically banned due to |
17 |
their spam content or at the very least matching reported user profiles |
18 |
on Stop Forum Spam[1]. Opening a Council bug for each of these would be |
19 |
an insurmountable workload if done manually, and at the very least a |
20 |
ludicrous volume of completely pointless mail to all Council members; |
21 |
but it is *exactly* what would be required by this proposal. |
22 |
|
23 |
As for the potential counterargument that bots could be easily dropped |
24 |
from the definition of "user" in this context, there is no general way |
25 |
to distinguish a bot from a non-bot user in full generality, and several |
26 |
ways in which non-bot users and bots could effectively share accounts so |
27 |
it would all need reported regardless. |
28 |
|
29 |
Note that the above is not considering any actions taken with regard to |
30 |
contributing users, which are by comparison quite rare, though one could |
31 |
consider locking a topic to be a "disciplinary action" which would |
32 |
require still more Council bugs, warnings regarding borderline behavior |
33 |
would require still more Council bugs. |
34 |
|
35 |
As it stands, disciplinary actions are handled per medium and channel, |
36 |
with appeals going first to those with direct authority over that medium |
37 |
or channel, then to ComRel, then the Council. This is simple, |
38 |
consistent, and most of all it is on the whole effective; all while |
39 |
minimizing the amount of make work. If there is meant to be an implicit |
40 |
argument that this is somehow insufficiently documented, by all means |
41 |
make that point, ask people to document things more pervasively, do not |
42 |
discard a working system because someone could not be bothered to read |
43 |
the documentation. |
44 |
|
45 |
[1] http://www.stopforumspam.com/forum/index.php |