I saw this over vacation and have been meaning to respond.
I think Thehelpfulone raises a very important point. Echo does rely in email notifications. Also, email is where users actually are and until new users get into the habit of coming to our sites directly, email is a way to draw them back in.
Does anyone have the answer to question #1 below (why email is optional in the first place?).
Howie
On Sat, Dec 29, 2012 at 12:59 AM, Thehelpfulone <thehelpfulonewiki@gmail.com
wrote:
Hi guys,
Sending this to the main EE list as I imagine E2 may also have some useful input. I just watched one of the account creation videos from the UserTesting.com dashboard which Steven also published at https://commons.wikimedia.org/wiki/File:ACUX_user_test_2.ogv and an interesting point that the tester made was that she was happy that the email address was optional, because that meant that she didn't have to give it, and that she never gave optional information.
A couple of questions:
- Does anyone know why we made email optional in the first place - was it
because it wasn't a core part of the wiki functionality given that it was previously only used for password resets, or was it something to do with storing the least amount of data about our users as possible?
- With Echo and Flow and all the new notification stuff that is planned,
and given Email notifications https://www.mediawiki.org/wiki/Echo/Feature_requirements#Email_notificationsstill play a big part of it, should we be considering trying to actually get *more *email addresses for users who sign up and so either make it less clear that email is optional (probably not so good from a usability standpoint?) or even go so far as to make email required?
As an alternative to this we could give reasons why providing your email address is a good, but I worry that putting too much detail on the new account creation page would cause a TL;DR like the tester made quite clear at https://commons.wikimedia.org/wiki/File:Account_creation_test_2.ogv.
If this discussion has already been had, apologies - I've just joined this list and couldn't find anything from a brief skim of the archives.
Ryan
Thehelpfulone http://meta.wikimedia.org/wiki/User:Thehelpfulone
ee mailing list ee@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
On 01/10/2013 05:57 PM, Howie Fung wrote:
I saw this over vacation and have been meaning to respond.
I think Thehelpfulone raises a very important point. Echo does rely in email notifications. Also, email is where users actually are and until new users get into the habit of coming to our sites directly, email is a way to draw them back in.
Does anyone have the answer to question #1 below (why email is optional in the first place?).
It's been like that as long as I can remember. It has been used for Special:EmailUser for a long time, though, so it's not true that it was recently only for resets.
I assume the answer is a tautology, "so people don't have to give their email". Probably, the goal was to encourage signups by minimizing the required personal information.
But it's clearly a trade-off. If they don't have an email, they can't use password reset (which alone is important), and things like Echo emails will not work out of the box.
Matt Flaschen
There is a tradeoff between requiring the field and getting the user through the process. This is mostly a practical concern. Is anyone aware of a community/cultural rational?
We should think about experimenting with this at some point since we're expanding the use of email. And depending on the community considerations, we could even consider making email a required field. Or do what Gmail does and make it optional, but not label it as such.
This isn't a high priority for the e2/e3 teams right not (e.g., I'd want to look at the confirm email loop first), but something to keep on the backlog.
Howie
On Thu, Jan 10, 2013 at 3:07 PM, Matthew Flaschen mflaschen@wikimedia.orgwrote:
On 01/10/2013 05:57 PM, Howie Fung wrote:
I saw this over vacation and have been meaning to respond.
I think Thehelpfulone raises a very important point. Echo does rely in email notifications. Also, email is where users actually are and until new users get into the habit of coming to our sites directly, email is a way to draw them back in.
Does anyone have the answer to question #1 below (why email is optional in the first place?).
It's been like that as long as I can remember. It has been used for Special:EmailUser for a long time, though, so it's not true that it was recently only for resets.
I assume the answer is a tautology, "so people don't have to give their email". Probably, the goal was to encourage signups by minimizing the required personal information.
But it's clearly a trade-off. If they don't have an email, they can't use password reset (which alone is important), and things like Echo emails will not work out of the box.
Matt Flaschen
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
On Thu, Jan 10, 2013 at 10:46 PM, Howie Fung hfung@wikimedia.org wrote:
There is a tradeoff between requiring the field and getting the user through the process. This is mostly a practical concern. Is anyone aware of a community/cultural rational?
The community rationale comes from a very old and entrenched part of Wikimedia: the idea that people have a right to complete and total anonymity, even when editing. It is also because the norm was set before the ubiquity of easy to use Web mail accounts (we predate Gmail by a few years, for instance), so users at the time had the idea that semi-anonymous email was not as ridiculously easy to set up and use.
As Matt says, it's pretty much a tautology at this point, at least among the old guard. Example of that: https://meta.wikimedia.org/wiki/Research_talk:Account_creation_UX#Requiring_...
To speak to the registration process: the tradeoff is almost not worth thinking about. Even among the users in testing who actually noticed email was optional, they still would fill it out much of the time. In the age of notifications and password recovery systems, users have a clear mental model of why email is helpful.
On 01/11/2013 02:22 AM, Steven Walling wrote:
The community rationale comes from a very old and entrenched part of Wikimedia: the idea that people have a right to complete and total anonymity, even when editing.
I don't think it was just a philosophical thing.
I think it was probably pragmatically understood that the fewer fields people need to fill the more likely there are to finish. This is particularly true when it's personal info.
It is also because the norm was set before
the ubiquity of easy to use Web mail accounts (we predate Gmail by a few years, for instance), so users at the time had the idea that semi-anonymous email was not as ridiculously easy to set up and use.
I agree that both technology and attitudes have shifted. People are more willing to give out their email now, which is partly (but probably not mostly) because *some* users know about and use disposable emails.
As Matt says, it's pretty much a tautology at this point, at least among the old guard. Example of that: https://meta.wikimedia.org/wiki/Research_talk:Account_creation_UX#Requiring_...
To speak to the registration process: the tradeoff is almost not worth thinking about. Even among the users in testing who actually noticed email was optional, they still would fill it out much of the time.
Of course, the users who were around initially were not the same population on average as the ones we're now testing and reaching out to.
Early adopters are more likely to notice things like a field not being required.
In the age of notifications and password recovery systems, users have a clear mental model of why email is helpful.
I agree.
Matt Flaschen
On 11 January 2013 07:43, Matthew Flaschen mflaschen@wikimedia.org wrote:
On 01/11/2013 02:22 AM, Steven Walling wrote:
The community rationale comes from a very old and entrenched part of Wikimedia: the idea that people have a right to complete and total anonymity, even when editing.
I don't think it was just a philosophical thing.
I think it was probably pragmatically understood that the fewer fields people need to fill the more likely there are to finish. This is particularly true when it's personal info.
It is also because the norm was set before
the ubiquity of easy to use Web mail accounts (we predate Gmail by a few years, for instance), so users at the time had the idea that semi-anonymous email was not as ridiculously easy to set up and use.
I agree that both technology and attitudes have shifted. People are more willing to give out their email now, which is partly (but probably not mostly) because *some* users know about and use disposable emails.
Agreed. I think we're overstating it by saying "users had te idea that semi-anonymous email was not as ridiculously easy to set up and use" - look, I was using Hotmail accounts before Wikipedia existed. Gmail did not come up with the idea of easy-register pseudonymous email accounts.
As Matt says, it's pretty much a tautology at this point, at least among the old guard. Example of that:
https://meta.wikimedia.org/wiki/Research_talk:Account_creation_UX#Requiring_...
To speak to the registration process: the tradeoff is almost not worth thinking about. Even among the users in testing who actually noticed email was optional, they still would fill it out much of the time.
Of course, the users who were around initially were not the same population on average as the ones we're now testing and reaching out to.
Early adopters are more likely to notice things like a field not being required.
In the age of notifications and password recovery systems, users have a clear mental model of why email is helpful.
I agree.
Matt Flaschen
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down. If we mandate email we're also dramatically sizing up the pool of people that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
Oliver said:
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
This is a good point. In my work with subject recruitment policy in Wikipedia, I worried that using SpecialEmail to contact users circumvents the primary means for Wikipedians to notice when other people are being spammed. For example, if I receive an email from a potential spammer, it's currently impossible for me to tell how many other users were contacted, so its hard to know if I should just make a post in the Village Pump to see if there are others or quickly find an admin on IRC to shut this user down. This is my primary reason for recommending that subject recruitment requests happen on wiki -- so that Wikipedians can most easily track its use.
If we mandate email we're also dramatically sizing up the pool of people
that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
I just ran a few queries to check how much we'd be sizing up our pool of emailable editors. In the last year, 380,281 users registered an account and made at least on edit. Of these users, 233,733 (61.5%) provided an email address and 128,572 (55.0%) confirmed it. Assuming that the same number of people register accounts once we require an email address and the same proportion of people that provide an email address will confirm it, requiring an email address could increase the pool of email spammable editors for new cohorts by about 80k users (a 63% increase) from 128k to ~200k/year.
-Aaron
On Fri, Jan 11, 2013 at 4:18 AM, Oliver Keyes okeyes@wikimedia.org wrote:
On 11 January 2013 07:43, Matthew Flaschen mflaschen@wikimedia.orgwrote:
On 01/11/2013 02:22 AM, Steven Walling wrote:
The community rationale comes from a very old and entrenched part of Wikimedia: the idea that people have a right to complete and total anonymity, even when editing.
I don't think it was just a philosophical thing.
I think it was probably pragmatically understood that the fewer fields people need to fill the more likely there are to finish. This is particularly true when it's personal info.
It is also because the norm was set before
the ubiquity of easy to use Web mail accounts (we predate Gmail by a few years, for instance), so users at the time had the idea that semi-anonymous email was not as ridiculously easy to set up and use.
I agree that both technology and attitudes have shifted. People are more willing to give out their email now, which is partly (but probably not mostly) because *some* users know about and use disposable emails.
Agreed. I think we're overstating it by saying "users had te idea that semi-anonymous email was not as ridiculously easy to set up and use" - look, I was using Hotmail accounts before Wikipedia existed. Gmail did not come up with the idea of easy-register pseudonymous email accounts.
As Matt says, it's pretty much a tautology at this point, at least among the old guard. Example of that:
https://meta.wikimedia.org/wiki/Research_talk:Account_creation_UX#Requiring_...
To speak to the registration process: the tradeoff is almost not worth thinking about. Even among the users in testing who actually noticed email was optional, they still would fill it out much of the time.
Of course, the users who were around initially were not the same population on average as the ones we're now testing and reaching out to.
Early adopters are more likely to notice things like a field not being required.
In the age of notifications and password recovery systems, users have a clear mental model of why email is helpful.
I agree.
Matt Flaschen
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down. If we mandate email we're also dramatically sizing up the pool of people that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
-- Oliver Keyes Community Liaison, Product Development Wikimedia Foundation
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
On 11 January 2013 17:45, Aaron Halfaker aaron.halfaker@gmail.com wrote:
Oliver said:
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
This is a good point. In my work with subject recruitment policy in Wikipedia, I worried that using SpecialEmail to contact users circumvents the primary means for Wikipedians to notice when other people are being spammed. For example, if I receive an email from a potential spammer, it's currently impossible for me to tell how many other users were contacted, so its hard to know if I should just make a post in the Village Pump to see if there are others or quickly find an admin on IRC to shut this user down. This is my primary reason for recommending that subject recruitment requests happen on wiki -- so that Wikipedians can most easily track its use.
If we mandate email we're also dramatically sizing up the pool of people
that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
I just ran a few queries to check how much we'd be sizing up our pool of emailable editors. In the last year, 380,281 users registered an account and made at least on edit. Of these users, 233,733 (61.5%) provided an email address and 128,572 (55.0%) confirmed it. Assuming that the same number of people register accounts once we require an email address and the same proportion of people that provide an email address will confirm it, requiring an email address could increase the pool of email spammable editors for new cohorts by about 80k users (a 63% increase) from 128k to ~200k/year.
-Aaron
Some interesting statistics - do we know how many users have actually been spammed through the MediaWiki interface though? If spamming is a serious problem then we could just reduce the number of emails someone can send out per day like we have for the 6 account creation/day limit per user account/IP at the moment. In fact some metrics on how many users (both new and older ones) use the MediaWiki Special:EmailUser interface could be interesting too if you can generate those.
agreed – on a related note, I have data on email registration/authentication/notifications for MoodBar users here: http://toolserver.org/~dartar/fd_notify/
On Jan 11, 2013, at 10:16 AM, Thehelpfulone thehelpfulonewiki@gmail.com wrote:
On 11 January 2013 17:45, Aaron Halfaker aaron.halfaker@gmail.com wrote: Oliver said: I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a lot of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
This is a good point. In my work with subject recruitment policy in Wikipedia, I worried that using SpecialEmail to contact users circumvents the primary means for Wikipedians to notice when other people are being spammed. For example, if I receive an email from a potential spammer, it's currently impossible for me to tell how many other users were contacted, so its hard to know if I should just make a post in the Village Pump to see if there are others or quickly find an admin on IRC to shut this user down. This is my primary reason for recommending that subject recruitment requests happen on wiki -- so that Wikipedians can most easily track its use.
If we mandate email we're also dramatically sizing up the pool of people that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
I just ran a few queries to check how much we'd be sizing up our pool of emailable editors. In the last year, 380,281 users registered an account and made at least on edit. Of these users, 233,733 (61.5%) provided an email address and 128,572 (55.0%) confirmed it. Assuming that the same number of people register accounts once we require an email address and the same proportion of people that provide an email address will confirm it, requiring an email address could increase the pool of email spammable editors for new cohorts by about 80k users (a 63% increase) from 128k to ~200k/year.
-Aaron
Some interesting statistics - do we know how many users have actually been spammed through the MediaWiki interface though? If spamming is a serious problem then we could just reduce the number of emails someone can send out per day like we have for the 6 account creation/day limit per user account/IP at the moment. In fact some metrics on how many users (both new and older ones) use the MediaWiki Special:EmailUser interface could be interesting too if you can generate those.
-- Thehelpfulone http://meta.wikimedia.org/wiki/User:Thehelpfulone _______________________________________________ EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
Hmn. Theoretically things like the CU log will contain emailuser activity, but I'm not sure how easy it is to parse en masse - or if it's possible to distinguish between spam and the lack thereof ;p.
I could try going for "emails sourced from blocked accounts" as a highly back-of-the-envelope calculation.
On 11 January 2013 18:16, Thehelpfulone thehelpfulonewiki@gmail.com wrote:
On 11 January 2013 17:45, Aaron Halfaker aaron.halfaker@gmail.com wrote:
Oliver said:
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
This is a good point. In my work with subject recruitment policy in Wikipedia, I worried that using SpecialEmail to contact users circumvents the primary means for Wikipedians to notice when other people are being spammed. For example, if I receive an email from a potential spammer, it's currently impossible for me to tell how many other users were contacted, so its hard to know if I should just make a post in the Village Pump to see if there are others or quickly find an admin on IRC to shut this user down. This is my primary reason for recommending that subject recruitment requests happen on wiki -- so that Wikipedians can most easily track its use.
If we mandate email we're also dramatically sizing up the pool of people
that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
I just ran a few queries to check how much we'd be sizing up our pool of emailable editors. In the last year, 380,281 users registered an account and made at least on edit. Of these users, 233,733 (61.5%) provided an email address and 128,572 (55.0%) confirmed it. Assuming that the same number of people register accounts once we require an email address and the same proportion of people that provide an email address will confirm it, requiring an email address could increase the pool of email spammable editors for new cohorts by about 80k users (a 63% increase) from 128k to ~200k/year.
-Aaron
Some interesting statistics - do we know how many users have actually been spammed through the MediaWiki interface though? If spamming is a serious problem then we could just reduce the number of emails someone can send out per day like we have for the 6 account creation/day limit per user account/IP at the moment. In fact some metrics on how many users (both new and older ones) use the MediaWiki Special:EmailUser interface could be interesting too if you can generate those.
-- Thehelpfulone http://meta.wikimedia.org/wiki/User:Thehelpfulone
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
These stats are indeed helpful. The conversion rates are lower than I had anticipated. Over the entire funnel, we're losing the ability to communicate via email with 67% of our registered users.
Given these numbers, I'd like for us to think about experimenting with removing the "optional" label, as well as improving the confirmation flow. We'll need to discuss the spam issue a little more closely -- I'm not sure I buy the argument entirely, but we should discuss.
(This, of course, needs to be prioritized against existing work, so I'm not sure where this would fall right now).
Howie
On Fri, Jan 11, 2013 at 10:16 AM, Thehelpfulone <thehelpfulonewiki@gmail.com
wrote:
On 11 January 2013 17:45, Aaron Halfaker aaron.halfaker@gmail.com wrote:
Oliver said:
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
This is a good point. In my work with subject recruitment policy in Wikipedia, I worried that using SpecialEmail to contact users circumvents the primary means for Wikipedians to notice when other people are being spammed. For example, if I receive an email from a potential spammer, it's currently impossible for me to tell how many other users were contacted, so its hard to know if I should just make a post in the Village Pump to see if there are others or quickly find an admin on IRC to shut this user down. This is my primary reason for recommending that subject recruitment requests happen on wiki -- so that Wikipedians can most easily track its use.
If we mandate email we're also dramatically sizing up the pool of people
that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
I just ran a few queries to check how much we'd be sizing up our pool of emailable editors. In the last year, 380,281 users registered an account and made at least on edit. Of these users, 233,733 (61.5%) provided an email address and 128,572 (55.0%) confirmed it. Assuming that the same number of people register accounts once we require an email address and the same proportion of people that provide an email address will confirm it, requiring an email address could increase the pool of email spammable editors for new cohorts by about 80k users (a 63% increase) from 128k to ~200k/year.
-Aaron
Some interesting statistics - do we know how many users have actually been spammed through the MediaWiki interface though? If spamming is a serious problem then we could just reduce the number of emails someone can send out per day like we have for the 6 account creation/day limit per user account/IP at the moment. In fact some metrics on how many users (both new and older ones) use the MediaWiki Special:EmailUser interface could be interesting too if you can generate those.
-- Thehelpfulone http://meta.wikimedia.org/wiki/User:Thehelpfulone
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
On 11 January 2013 18:33, Howie Fung hfung@wikimedia.org wrote:
These stats are indeed helpful. The conversion rates are lower than I had anticipated. Over the entire funnel, we're losing the ability to communicate via email with 67% of our registered users.
Given these numbers, I'd like for us to think about experimenting with removing the "optional" label, as well as improving the confirmation flow. We'll need to discuss the spam issue a little more closely -- I'm not sure I buy the argument entirely, but we should discuss.
Yeah, my argument isn't "we're opening ourselves up to spam, so we
shouldn't do it (anything is justified with a sufficient benefit) but more to highlight the fact that if we do this we could see increased uptake/demand for us to actually fix up Special:EmailUser, which'll take N hours of dev time. I just want to make sure we don't ignore possible disruptions to our work in the long-term.
(This, of course, needs to be prioritized against existing work, so I'm not sure where this would fall right now).
Howie
On Fri, Jan 11, 2013 at 10:16 AM, Thehelpfulone < thehelpfulonewiki@gmail.com> wrote:
On 11 January 2013 17:45, Aaron Halfaker aaron.halfaker@gmail.comwrote:
Oliver said:
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a *lot* of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
This is a good point. In my work with subject recruitment policy in Wikipedia, I worried that using SpecialEmail to contact users circumvents the primary means for Wikipedians to notice when other people are being spammed. For example, if I receive an email from a potential spammer, it's currently impossible for me to tell how many other users were contacted, so its hard to know if I should just make a post in the Village Pump to see if there are others or quickly find an admin on IRC to shut this user down. This is my primary reason for recommending that subject recruitment requests happen on wiki -- so that Wikipedians can most easily track its use.
If we mandate email we're also dramatically sizing up the pool of people
that bad-faith trolls can take a shot at; we may find the community demanding a rebuild of Special:EmailUser when the vandals of the world notice what we've done.
I just ran a few queries to check how much we'd be sizing up our pool of emailable editors. In the last year, 380,281 users registered an account and made at least on edit. Of these users, 233,733 (61.5%) provided an email address and 128,572 (55.0%) confirmed it. Assuming that the same number of people register accounts once we require an email address and the same proportion of people that provide an email address will confirm it, requiring an email address could increase the pool of email spammable editors for new cohorts by about 80k users (a 63% increase) from 128k to ~200k/year.
-Aaron
Some interesting statistics - do we know how many users have actually been spammed through the MediaWiki interface though? If spamming is a serious problem then we could just reduce the number of emails someone can send out per day like we have for the 6 account creation/day limit per user account/IP at the moment. In fact some metrics on how many users (both new and older ones) use the MediaWiki Special:EmailUser interface could be interesting too if you can generate those.
-- Thehelpfulone http://meta.wikimedia.org/wiki/User:Thehelpfulone
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
EE mailing list EE@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/ee
On Fri, Jan 11, 2013 at 10:33 AM, Howie Fung hfung@wikimedia.org wrote:
Given these numbers, I'd like for us to think about experimenting with removing the "optional" label, as well as improving the confirmation flow. We'll need to discuss the spam issue a little more closely -- I'm not sure I buy the argument entirely, but we should discuss.
I think we could easily run this test in E3. If all we're doing is removing the label but not changing the config, we could probably even A/B it.
On 01/11/2013 07:18 AM, Oliver Keyes wrote:
I think one thing we may need to consider is what this binds us into. So, there's a very good reason as to why a /lot/ of the old guard have disabled email or not provided accounts - the emailing system built into MediaWiki sucks. It can be, will be and has been used for wide-scale abuse that it's pretty difficult to shut down.
We could require email, but just disable "Enable e-mail from other users" by default. This is a setting, but it's true by default now.
Also, I think a large part of the old guard know how to use email solely for password resets (there's only four checkboxes, and only two are checked by default, "Enable e-mail from other users, and "E-mail me when my user talk page is changed").
In my experience, EmailUser is used occasionally (e.g. for me, sometimes for https://en.wikipedia.org/wiki/Wikipedia:Resource_exchange and adoption/mentoring), but not all that often. If I had to guess, I'd say new users without mentors would use it less.
We could change the wording of MediaWiki:nowikiemailtext (which shows when you try to email someone that doesn't allow it) to say something along the lines of:
"Users choose whether they want to receive email, and it is disabled by default for new users. This user currently does not allow it.
Consider posting {{tl|please enable email}} on [[User:Joe|Joe]]'s talk page to request they enable it. They are not required to do so and can change their mind."
We could JavaScriptify that too if we wanted.
Matt Flaschen