Jump to content

Talk:Requests for comment/X!'s Edit Counter

Add topic
From Meta, a Wikimedia project coordination wiki
Latest comment: 10 years ago by SlimVirgin in topic New discussion

Problem

[edit]

I can't talk about the tool which doesn't work for me and my bot. Infovarius (talk) 20:44, 23 June 2013 (UTC)Reply

What?—cyberpower ChatHello! 20:55, 23 June 2013 (UTC)Reply
That's because you have many edits (over 155k). Snowolf gives some usable/useful examples at the top of the RfC page. Mathonius (talk) 20:56, 23 June 2013 (UTC)Reply
The labs version isn't restricted to 45,000 edits. The labs version, due to more usable resources, allows stats for users that have up to 500,000 edits now.—cyberpower ChatHello! 20:58, 23 June 2013 (UTC)Reply
The toolserver version isn’t restricted to 45,000 edits either, otherwise I wouldn’t get any results there for de:, but I’m getting such results for > 50,000 edits. So, there’s no such restriction there either. --Geitost diskusjon 08:18, 24 June 2013 (UTC)Reply

When was the opt-in scheme imposed on Edit Counter?

[edit]

That was 2 or 3 years ago, wasn't it? Varlaam (talk) 07:02, 28 June 2013 (UTC)Reply

Poor format for !vote

[edit]

I think that all available options should be listed towards the top of the page; I've gotten the impression from some of the comments that some people are only thinking that there are the first two options available. Also, one can select multiple options (e.g. the last option, but opt-out rather than opt-in individually). I really think that this needs to be reformatted. It's also mentioning, per my comment in the "Remove opt-in completely" section, that these !voters likely view the "switch to opt-out option" as a second-best option. --Jackson Peebles (talk) 21:47, 28 June 2013 (UTC)Reply

All available options are listed at the top of the page in the Contents box. --AFBorchert (talk) 04:46, 29 June 2013 (UTC)Reply

IP votes

[edit]

At least consider their messages, please. :) PiRSquared17 (talk) 00:44, 1 July 2013 (UTC)Reply

Also discussed at Talk:Privacy policy/Call for input (2013)

[edit]

Here. — Racconish Tk 09:34, 3 July 2013 (UTC)Reply

Alleged difference between Wikimedia Labs and Toolserver TOS

[edit]

What exactly is the alleged difference between Toolserver TOS and the draft of Wikimedia LAbs TOS that is applicable to this tool ? I see the Wikimedia Labs TOS states "No collection or storage of private data or personally identifiable information without the express authorization of the individuals". Thanks, — Racconish Tk 10:07, 3 July 2013 (UTC)Reply

The privacy policy of the toolserver goes beyond that as it is not just concerned about collection or storage of personal data but also its processing (i.e. data mining). Let me allow to quote the relevant part:
  • Tools that allow profiling of individual user's activity (beyond what can easily be achieved directly on the public wiki sites) must only be applied with the respective user's consent (opt-in).
    Note: analysis of publically available data (data mining) may well lead to information that compromizes the privacy of individuals (profiling). The fact that anyone could in theory perform this analysis does not justify the publication of such information. Only if the data is available just as easily from the Wiki iteself, or if the data in no way interferes with the privacy of individuals, can it be made available on the toolserver without the user's consent. See also w:Data Mining#Privacy concerns and ethics and w:Profiling practices#Risks and issues.
--AFBorchert (talk) 11:49, 3 July 2013 (UTC)Reply
Thanks for the clarification. Yes, Toolserver TOS are very explicit. Though I don't think Wikimedia Labs draft of TOS authorizes such "profiling". Cheers, — Racconish Tk 08:42, 4 July 2013 (UTC)Reply

Opt-in votes

[edit]

Just some random ones:

Uhm... --Ricordisamoa 07:44, 1 July 2013 (UTC)Reply

Ever heard of SUL? NNW (talk) 08:07, 1 July 2013 (UTC)Reply
Ever seen this?? -jkb- 09:38, 1 July 2013 (UTC)Reply
Naturally, voters come from all Wikimedia projects, not only from Meta. --Martina Nolte (talk) 12:17, 1 July 2013 (UTC)Reply
Indeed. In fact, most users will rarely post/edit on Meta, only when there something happening which is of relevance to the Wikimedia wikis they're mostly editing, so it's perfectly natural that some user's first edits on Meta were made now in this context. For example, I think that users Sakra and Schreiber are quite active in German Wikipedia. Gestumblindi (talk) 12:31, 31 July 2013 (UTC)Reply
Also Grueslayer now has 1,188 edits since Mar 28, 2010 on de-WP and Jossi2 has 8,263 edits since Oct 17, 2004 there. Sakra has 5,106 edits since Oct 09, 2011 14:05 there, Schreiber has 9,547 edits since Nov 08, 2004 15:08 there and also is sysop since February 2008. Danapit is active on en-Wikivoyage since March, Jethwarp has 17,522 edits since Sep 14, 2009 on en-WP, Zananiri has 980 edits since Feb 05, 2009 there. --Geitost diskusjon 08:09, 1 August 2013 (UTC)Reply

Wikichecker

[edit]

Many of the users might be unaware that at Wikichecker similar information and graphs are already available and all you have to do is to type in the name of the user or article in the appropriate box. The only difference is that it is ad supported. I think that this information needs to be incorporated into the Rfc page. Participants need to know that a similar service already exists without any requirement for any opt-in. DiptanshuTalk 18:04, 6 July 2013 (UTC)Reply

And that it's not a global tool whereas X!'s Edit Counter is.—cyberpower ChatHello! 18:35, 6 July 2013 (UTC)Reply
Avaiable for en.wikipedia only, as far as I can see. -jkb- 19:29, 6 July 2013 (UTC)Reply
Also seems to work with the French, Japanese and Russian Wikipedias. --Stefan2 (talk) 21:15, 6 July 2013 (UTC)Reply
Wikichecker is a service offered by somebody not directly affiliated with the WMF. Earlier in this discussion page you find this link indicating the vastness and diversities of the Wikipedias present in a global sense. But, this tool serves good utility of administrative purpose of understanding an user's usage patterns. Nobody is here to fight and we are not fighting each other. The purpose should be to enrich the resource that we already have and building upon the ones required. As such, I find the entire discussion meaningless as whatever people are considering as private data does not remain private under the copyleft after entering the public domain. Sorry if I am wrong. DiptanshuTalk 03:52, 7 July 2013 (UTC)Reply
But however, I do not understand why it should be that important for this voting. -jkb- 08:52, 7 July 2013 (UTC)Reply
It seems to me that the main purpose of keeping an opt-in is to restrict the access to personal editing data for ones who prefer it that way, or to restrict it with an objective of screening 'explicit' data from minors. The irony of the fact is that it would be pointless doing so in case a similar service is already available even if somebody does not opt-in for the service. The opt-in is therefore like placing a gate-keeper on one entrance of a hall when at least one other unguarded entrance is already present. I think this information is definitely relevant to the 'intelligent' people who are voting on the effectivity of an opt-in because the presence of another entrance makes it practically a pointless discussion. DiptanshuTalk 09:51, 8 July 2013 (UTC)Reply
No, it isn't. Just because someone does it, it doesn't mean that we have to do it, too. This discussion might influence the privacy policy for Wikimedia projects and then it might be forbidden to have such a "service" for everybody on Wikimedia servers. NNW (talk) 10:09, 8 July 2013 (UTC)Reply
Thanks NNW. I find your explanation more acceptable. I would like to withdraw from my stand. DiptanshuTalk 11:31, 8 July 2013 (UTC)Reply

Consensus

[edit]

It looks like this proposal will not succeed. Nonetheless, the opt-in requirement should be removed for the English Wikipedia (only) per consensus there. -- King of 04:12, 10 July 2013 (UTC)Reply

That would seem to be the result, but it's really up to the tool's maintainer to decide what exactly to do, as long as it doesn't break any of WMF's rules or actual laws (e.g., with enwiki, which decided to remove opt-ins). Edit: He shouldn't override the global consensus, but what to do on a wiki-by-wiki basis is up to him. PiRSquared17 (talk) 04:20, 10 July 2013 (UTC)Reply
Oh wait! I can't remove it just to piss everyone off? Darn.  ;-)—cyberpower ChatHello! 16:15, 11 July 2013 (UTC)Reply
I'm not sure that there is actually "a proposal" here. It looks to me like cyberpower just wanted to know what people thought was a good idea. People told him what they thought. He can now take that information and do whatever he wants with it (consistent with laws and TOU at Labs). WhatamIdoing (talk) 19:46, 12 July 2013 (UTC)Reply

Can I ask how the consensus turned out this way? I understand it's not a democratic decision but more people said, "We don't want this," than people who said, "We want this". I'll admit I'm annoyed it turned out this way, and that's why I'm here - I want to understand how this "consensus" was achieved, because as far as I saw, that wasn't it. LazyBastardGuy 22:47, 22 July 2013 (UTC)Reply

I'm looking for the village pump discussion but I can't find it. All I find is a link pointing me back here. LazyBastardGuy 22:51, 22 July 2013 (UTC)Reply
Here is the discussion. Actually, some of the votes were influenced by a mention in Der Kurier, which was mostly pro-opt-in. Of course, their opinions should count. PiRSquared17 (talk) 22:54, 22 July 2013 (UTC)Reply
I never said anyone's opinions shouldn't count. But that's the German WP and I'm from the English one, and this decision hardly seemed 'local'. I'll leave you guys alone after I say one last thing: It's a little strange that a shorter discussion on the village pump seemed more influential than the broad one happening here. I'm a little T'd at that, but who cares... Anyways, I'm done. Cheers. LazyBastardGuy 23:00, 22 July 2013 (UTC)Reply
The purely numerical result of this here is something like 260:200 for opt-in. At the same time, the decision only for enwiki is the other way round. So where is the problem? --MF-W 23:51, 22 July 2013 (UTC)Reply

┌─────────────────────────────────┘
I'm also concerned that people are saying there is consensus to remove the opt-in for editors on the English Wikipedia, even though this RfC is clearly against removing it. Cyberpower started an RfC on the village pump on 4 June, but cut it short on 22 June to start this one, and directed people from there to here. So it doesn't seem right that the English RfC is now regarded as definitive, when editors there were told to post to this RfC instead.

Should there not be a full, 30-day, widely advertised RfC on the English WP for something that affects every Wikipedian? SlimVirgin (talk) 03:29, 23 July 2013 (UTC)Reply

Ok, I didn't understand the concerns were about interpreting the enwiki rfc outcome, not this one. --MF-W 13:15, 23 July 2013 (UTC)Reply
SlimVirgin makes a great point about the concerns mentioned. I have held off the plans to remove opt-in on the English Wikipedia for now pending a fully advertised RfC.—cyberpower ChatHello! 04:49, 23 July 2013 (UTC)Reply

Few questions

[edit]
  1. Can the tool have the option to ignore opt in for researchers (researcher flag) [1]?
  2. Is the tool open source? What would prevent someone from copying it and removing the opt in part which keeps making it mostly useless? --Piotrus (talk) 04:02, 18 August 2013 (UTC)Reply
@Piotrus and Cyberpower678:
  1. The tool currently has no way to identify what user you are. I guess it could use TUSC, to let users see their own things and let researchers as well, but this is unlikely to happen, as it could violate privacy.
  2. See tools:~tparis/pcount/source.php/toollabs:xtools/pcount/source.php. Of course, a determined individual can get this info through their own version of the tool, another tool, or writing their own script to get info via the API or normal interface. Nobody is denying that. Other than X!'s edit counter, numerous other tools with no opt-in exist. The point is to make it so that this information is not directly available on Wikimedia or Toolserver servers, IIRC.
PiRSquared17 (talk) 17:56, 31 August 2013 (UTC)Reply
@PiRSquared17: why wouldn't it be legitimate for anyone to host such tool (without opt-in feature) on Wikimedia Labs? There's no private data collection, and only WMF could prevent such tools from being hosted on their servers: the community should not have any say in the matter. --Ricordisamoa 03:25, 1 September 2013 (UTC)Reply
This discussion was only relevant to X!'s tool. The maintainer of the script wanted some input. If someone wants to put another version with no opt-in, that would technically be okay, but it would violate this consensus, and could lead to disputes. PiRSquared17 (talk) 03:39, 1 September 2013 (UTC)Reply
Exactly. X!'s Edit Counter is widely used everywhere. So naturally, the consensus of privacy is going to be respected. Anybody can make their own and the question also is, will it be accepted as the new edit counter. Anybody is free to make their own edit counter, however, I have been warned that someone is trying to clone X!'s edit counter and hack the opt requirement part. I have disabled open access to the code to this specific tool as a result, for the time being.—cyberpower ChatHello! 14:39, 1 September 2013 (UTC)Reply
It has been made irrevocably available under the terms of the GNU GPL. --Ricordisamoa 14:51, 1 September 2013 (UTC)Reply
I have no objections to the code being used, but will not allow a cloned hacked copy of it to be used to defy consensus.—cyberpower ChatHello! 14:54, 1 September 2013 (UTC)Reply
Also, hosting closed-source code violates the Labs Terms of use. --Ricordisamoa 14:57, 1 September 2013 (UTC)Reply
Why are you set on cloning the tool without opt-in?—cyberpower ChatHello! 15:07, 1 September 2013 (UTC)Reply
Maybe just because he wants to make a point that the RfC really didn't prevent it from happening. The WMF would need to create a Labs user-profiling policy for any real change to happen. PiRSquared17 (talk) 15:09, 1 September 2013 (UTC)Reply
I have re-enabled the source program.—cyberpower ChatHello! 15:12, 1 September 2013 (UTC)Reply
@Ricordisamoa: do you wish to reply to the last question from Cyberpower678? PiRSquared17 (talk) 16:23, 1 September 2013 (UTC)Reply
I'm not going to fork the tool, but to write a better one from scratch. --Ricordisamoa 05:18, 2 September 2013 (UTC)Reply
I've now published toollabs:ricordisamoa/dui --Ricordisamoa 21:16, 7 September 2013 (UTC)Reply
"Deep User Inspector"? If the point is beeing creepy, point made... it doesn't even has to be run. - Nabla (talk) 21:42, 7 September 2013 (UTC)Reply

New discussion

[edit]

For anyone still interested in whether to retain the opt-in, I've opened a discussion at en:Wikipedia talk:Requests for comment/User analysis tool. SlimVirgin (talk) 23:16, 9 May 2014 (UTC)Reply