Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
Yuvi Panda T http://yuvi.in/blog
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the user
agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitizationor https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection: https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there: https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.org wrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed.
I don't think that "the wishes of the community" have been established and the whole point of checkuser is that it allows for fingerprinting.
On Thu, Mar 27, 2014 at 4:20 AM, Nuria Ruiz nuria@wikimedia.org wrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the user
agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitization or https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.org wrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
+1. I'm totally down for keeping less information around, but if it gets in the way of people doing their job?
Rather than having an ethical debate over it, we could always test the actual usefulness with Science. That way we'd be able to see how much granularity each additional component adds to the data.
On 27 March 2014 07:15, Aaron Halfaker ahalfaker@wikimedia.org wrote:
Including more information on the UA, while being covered by legal under
the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed.
I don't think that "the wishes of the community" have been established and the whole point of checkuser is that it allows for fingerprinting.
On Thu, Mar 27, 2014 at 4:20 AM, Nuria Ruiz nuria@wikimedia.org wrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the
user agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitizationor https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.org wrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]:
https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
-- Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Rather than having an ethical debate over it, we could always test the
actual usefulness with Science. That way we'd be able to see how much granularity each additional component adds to the data. I kind of feel we are going backwards as we throughly discussed this point, technical info and references regarding entropy and user agents and fingerprinting can be found here: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
On Thu, Mar 27, 2014 at 3:49 PM, Oliver Keyes okeyes@wikimedia.org wrote:
+1. I'm totally down for keeping less information around, but if it gets in the way of people doing their job?
Rather than having an ethical debate over it, we could always test the actual usefulness with Science. That way we'd be able to see how much granularity each additional component adds to the data.
On 27 March 2014 07:15, Aaron Halfaker ahalfaker@wikimedia.org wrote:
Including more information on the UA, while being covered by legal under
the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed.
I don't think that "the wishes of the community" have been established and the whole point of checkuser is that it allows for fingerprinting.
On Thu, Mar 27, 2014 at 4:20 AM, Nuria Ruiz nuria@wikimedia.org wrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the
user agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitizationor https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.orgwrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]:
https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
-- Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Oliver Keyes Research Analyst Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Repost, because filtering; there might be a point of confusion here that's causing the problem. As I understand it, the user agent sanitisation is expected to apply to EventLogging data, and data in the Analytics pipeline, but not data streaming into MediaWiki proper - namely, the cu_changes table. Nuria, is that the case?
On 27 March 2014 08:16, Nuria Ruiz nuria@wikimedia.org wrote:
Rather than having an ethical debate over it, we could always test the
actual usefulness with Science. That way we'd be able to see how much granularity each additional component adds to the data. I kind of feel we are going backwards as we throughly discussed this point, technical info and references regarding entropy and user agents and fingerprinting can be found here: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
On Thu, Mar 27, 2014 at 3:49 PM, Oliver Keyes okeyes@wikimedia.orgwrote:
+1. I'm totally down for keeping less information around, but if it gets in the way of people doing their job?
Rather than having an ethical debate over it, we could always test the actual usefulness with Science. That way we'd be able to see how much granularity each additional component adds to the data.
On 27 March 2014 07:15, Aaron Halfaker ahalfaker@wikimedia.org wrote:
Including more information on the UA, while being covered by legal under
the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed.
I don't think that "the wishes of the community" have been established and the whole point of checkuser is that it allows for fingerprinting.
On Thu, Mar 27, 2014 at 4:20 AM, Nuria Ruiz nuria@wikimedia.orgwrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the
user agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitizationor https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single
person using the most up-to-date version of the app on a particular platform would >be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.orgwrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote: > Hello! > > We are getting closer to a general release of the Wikipedia Android > and iOS apps, and I think we should standardize on a User-Agent > format. The old app just appended an identifier in front of the > phone's default UA[1] but I think we can do better, to avoid privacy > concerns[2]. > > How about: > > WikipediaApp/<version> <OS>/<form-factor>/<version> > > This gives us all the info we need (App version, OS, Form Factor > (Tablet / Phone) and OS version) without giving away too much. It is > also fairly simple to construct and parse. > > For the latest alpha, my Nexus 4 would generate > > WikipediaApp/32 Android/Phone/4.4 > > While an iOS device might generate > > WkipediaApp/2.0 iOS/Phone/7.1 > > form-factor would just be Phone|Tablet for now, and can be expanded > later if necessary. > > Thoughts? > > [1]: https://www.mediawiki.org/wiki/Mobile/User_agents#Apps > [2]: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization > -- > Yuvi Panda T > http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Oliver Keyes Research Analyst Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
I don't think that "the wishes of the community" have been established and
the whole point of checkuser is that it allows for fingerprinting. Not permanently, the "stewards" only require IP/UA data for 90 days. That timeperiod falls into the range we are keeping operational data, this discussion pertains to data we are keeping long term.
We have no piece of community feedback encouraging us to store user agents, quite the opposite. All feedback we have regarding user agent data gathering points into the direction of "do not store longterm user agent data not agreggated".
On Thu, Mar 27, 2014 at 3:15 PM, Aaron Halfaker ahalfaker@wikimedia.orgwrote:
Including more information on the UA, while being covered by legal under
the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed.
I don't think that "the wishes of the community" have been established and the whole point of checkuser is that it allows for fingerprinting.
On Thu, Mar 27, 2014 at 4:20 AM, Nuria Ruiz nuria@wikimedia.org wrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the
user agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitizationor https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.org wrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]:
https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
-- Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Garry,
Replying again to clarify matters a bit.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
There are two types of data we are considering when it comes to data gathering and storage: operational and application data.
Operational data is retained for a 90 day period, it is logged per request and not manipulated in any way. The second type of data is application data, data that comes from logging an application event (like event logging) or tracks the usage of a feature. This data will be aggregated to avoid privacy concerns.
The google doc that describes user agent data collection does so for application data, i.e. data we wish to retain long term. https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
I think perhaps the confusion here comes from not defining who are the consumers of the user agent format mobile is proposing.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
By whom? how? via checkuser extension or other system?. It is worth having in mind that a mobile application is not a website (i.e. requests do not come from a browser, they come from an http client) and thus you might not be able to detect false accounts in the same fashion. For example, it is not strange that all users of a telco appear to come from a small set of IP addresses. In that case the IP bit of a request is not very significant when it comes to uniquely identify a user.
On Thu, Mar 27, 2014 at 10:20 AM, Nuria Ruiz nuria@wikimedia.org wrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the user
agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitization or https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.org wrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]: https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
It seems like we've resolved the original misunderstanding, and Mobile can work it out with the Functionaries directly what the impact on CheckUser will be. Unless Analytics engineering wants to volunteer their copious free time[1] to rebuild CheckUser, meta-debates around the tool are probably not the way to go.
[1]hahahahahahhahahaaaaaa
On 28 March 2014 02:56, Nuria Ruiz nuria@wikimedia.org wrote:
Garry,
Replying again to clarify matters a bit.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
There are two types of data we are considering when it comes to data gathering and storage: operational and application data.
Operational data is retained for a 90 day period, it is logged per request and not manipulated in any way. The second type of data is application data, data that comes from logging an application event (like event logging) or tracks the usage of a feature. This data will be aggregated to avoid privacy concerns.
The google doc that describes user agent data collection does so for application data, i.e. data we wish to retain long term.
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
I think perhaps the confusion here comes from not defining who are the consumers of the user agent format mobile is proposing.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
By whom? how? via checkuser extension or other system?. It is worth having in mind that a mobile application is not a website (i.e. requests do not come from a browser, they come from an http client) and thus you might not be able to detect false accounts in the same fashion. For example, it is not strange that all users of a telco appear to come from a small set of IP addresses. In that case the IP bit of a request is not very significant when it comes to uniquely identify a user.
On Thu, Mar 27, 2014 at 10:20 AM, Nuria Ruiz nuria@wikimedia.org wrote:
As a checkuser, user agents are an important part of my workflow for
identifying that multiple accounts are owned by the same person.
So I'm going to have to argue for including more information in the
user agent.
Including more information on the UA, while being covered by legal under the new privacy policy, really goes agains the wishes of the community as they do not wish to be finger printed. See: https://www.mediawiki.org/wiki/Talk:EventLogging/UserAgentSanitizationor https://meta.wikimedia.org/wiki/Talk:Privacy_policy There has been plenty more discussions about this on analytics e-mail list.
Your proposed user agent would basically mean that every single person
using the most up-to-date version of the app on a particular platform would
be indistinguishable from each other. This would, unfortunately, lead to
lots of innocent users getting blocked as sockpuppets.
However, note that the UA " WikipediaApp/<version> <OS>/<form-factor>/<version>" clearly satisfies the use case of the mobile team. It provides as much information as they need from their user without sending any private data.
Can you please list what is your use case? Namely how are you identifying "false" accounts. Perhaps relying on the user agent to do so is not the best strategy going forward. Have in mind that with the old privacy policy UA data needed to be discarded after 90 days. With the new policy there is more legal room but given community feedback analytics team is planning on aggregating all UA information in the future. This means that UA data will not be stored (or reported) per user or request but rather agreggated (as in "4% of users use iPhone").
We gathered recently information from all teams as to use cases pertaining UA data collection:
https://office.wikimedia.org/wiki/Analytics/Internal/EventLogging/PrivateDat... .
Let's talk about your use case and add it to the document that already exists describing usages of user agent data, this document was sent out to all teams couple months ago but there is no description of your use case there:
https://docs.google.com/a/wikimedia.org/document/d/1bp6qrvYi0Mh7l0s1psGnXEEN...
On Wed, Mar 26, 2014 at 11:20 PM, Dan Garry dgarry@wikimedia.org wrote:
Hey Yuvi,
As a checkuser, user agents are an important part of my workflow for identifying that multiple accounts are owned by the same person. So I'm going to have to argue for including more information in the user agent. Your proposed user agent would basically mean that every single person using the most up-to-date version of the app on a particular platform would be indistinguishable from each other. This would, unfortunately, lead to lots of innocent users getting blocked as sockpuppets.
Here's an example of a user agent from an iPhone using Safari: Mozilla/5.0 (iPhone; U; CPU iPhone OS 4_3_1 like Mac OS X; zh-tw) AppleWebKit/533.17.9 (KHTML, like Gecko) Version/5.0.2 Mobile/8G4 Safari/6533.18.5
Look at all of that wonderful information! ;-) In general, the more information you can include without breaching the user's privacy, the better.
I'd be happy to work with you on this.
Thanks, Dan
P.S. You may also want to consult with the legal team, to ensure that an unacceptable levels of private information are not given out. They would also make a complement for me; I would likely be pulling in the direction of "MOAR INFORMATION!", whereas they would likely be pulling in the direction of "LESS INFORMATION!". :-)
On 26 March 2014 15:00, Yuvi Panda yuvipanda@gmail.com wrote:
Add Analytics to cc, as I think they'll be interested as well :)
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
Hello!
We are getting closer to a general release of the Wikipedia Android and iOS apps, and I think we should standardize on a User-Agent format. The old app just appended an identifier in front of the phone's default UA[1] but I think we can do better, to avoid privacy concerns[2].
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
This gives us all the info we need (App version, OS, Form Factor (Tablet / Phone) and OS version) without giving away too much. It is also fairly simple to construct and parse.
For the latest alpha, my Nexus 4 would generate
WikipediaApp/32 Android/Phone/4.4
While an iOS device might generate
WkipediaApp/2.0 iOS/Phone/7.1
form-factor would just be Phone|Tablet for now, and can be expanded later if necessary.
Thoughts?
[2]:
https://www.mediawiki.org/wiki/EventLogging/UserAgentSanitization
-- Yuvi Panda T http://yuvi.in/blog
-- Yuvi Panda T http://yuvi.in/blog
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
-- Dan Garry Associate Product Manager for Platform Wikimedia Foundation
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote:
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote:
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
-- Yuvi Panda T http://yuvi.in/blog
And here's a patchset just posted for review for the iOS app.
https://gerrit.wikimedia.org/r/126188
Here are examples of what the UAs will look like in the logs:
WikipediaApp/2.0-alpha-2014-04-11 (Android/4.4.2) WikipediaApp/iOS-Wikipedia-4.0-4.0 (iPhone OS/7.1)
(I suspect "4.0-4.0" will change in the latter, though, as production build is approached.)
On Fri, Apr 4, 2014 at 4:04 PM, Oliver Keyes okeyes@wikimedia.org wrote:
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote:
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
-- Yuvi Panda T http://yuvi.in/blog
-- Oliver Keyes Research Analyst Wikimedia Foundation
Mobile-l mailing list Mobile-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/mobile-l
Cool. I'm going to run the ua-parser over them tomorrow to see if it recognises the devices or not, if that works?
On 15 April 2014 18:31, Adam Baso abaso@wikimedia.org wrote:
And here's a patchset just posted for review for the iOS app.
https://gerrit.wikimedia.org/r/126188
Here are examples of what the UAs will look like in the logs:
WikipediaApp/2.0-alpha-2014-04-11 (Android/4.4.2) WikipediaApp/iOS-Wikipedia-4.0-4.0 (iPhone OS/7.1)
(I suspect "4.0-4.0" will change in the latter, though, as production build is approached.)
On Fri, Apr 4, 2014 at 4:04 PM, Oliver Keyes okeyes@wikimedia.org wrote:
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote:
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
-- Yuvi Panda T http://yuvi.in/blog
-- Oliver Keyes Research Analyst Wikimedia Foundation
Mobile-l mailing list Mobile-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/mobile-l
So, it identifies the first one as Android, but can't pick out version number, and identifies the second as running Mobile Safari, but can't pick out the OS or device. I would recommend tweaking and testing these strings before deploying them if we want accurate device numbers (and we totally want accurate device numbers). Happy to help with the 'testing' part. Will chip in on gerrit directly.
On 15 April 2014 19:17, Oliver Keyes okeyes@wikimedia.org wrote:
Cool. I'm going to run the ua-parser over them tomorrow to see if it recognises the devices or not, if that works?
On 15 April 2014 18:31, Adam Baso abaso@wikimedia.org wrote:
And here's a patchset just posted for review for the iOS app.
https://gerrit.wikimedia.org/r/126188
Here are examples of what the UAs will look like in the logs:
WikipediaApp/2.0-alpha-2014-04-11 (Android/4.4.2) WikipediaApp/iOS-Wikipedia-4.0-4.0 (iPhone OS/7.1)
(I suspect "4.0-4.0" will change in the latter, though, as production build is approached.)
On Fri, Apr 4, 2014 at 4:04 PM, Oliver Keyes okeyes@wikimedia.orgwrote:
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote:
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
-- Yuvi Panda T http://yuvi.in/blog
-- Oliver Keyes Research Analyst Wikimedia Foundation
Mobile-l mailing list Mobile-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/mobile-l
-- Oliver Keyes Research Analyst Wikimedia Foundation
Sounds good. Available today? I have an open calendar 1105-1355 and 1405-COB.
-Adam
On Wed, Apr 16, 2014 at 9:15 AM, Oliver Keyes okeyes@wikimedia.org wrote:
So, it identifies the first one as Android, but can't pick out version number, and identifies the second as running Mobile Safari, but can't pick out the OS or device. I would recommend tweaking and testing these strings before deploying them if we want accurate device numbers (and we totally want accurate device numbers). Happy to help with the 'testing' part. Will chip in on gerrit directly.
On 15 April 2014 19:17, Oliver Keyes okeyes@wikimedia.org wrote:
Cool. I'm going to run the ua-parser over them tomorrow to see if it recognises the devices or not, if that works?
On 15 April 2014 18:31, Adam Baso abaso@wikimedia.org wrote:
And here's a patchset just posted for review for the iOS app.
https://gerrit.wikimedia.org/r/126188
Here are examples of what the UAs will look like in the logs:
WikipediaApp/2.0-alpha-2014-04-11 (Android/4.4.2) WikipediaApp/iOS-Wikipedia-4.0-4.0 (iPhone OS/7.1)
(I suspect "4.0-4.0" will change in the latter, though, as production build is approached.)
On Fri, Apr 4, 2014 at 4:04 PM, Oliver Keyes okeyes@wikimedia.orgwrote:
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote:
As followup: QChris did the actual work of reading the relevant RfC for me, and for now I've settled on:
WikipediaApp/<version> (<OS>/<version>; <form-factor>)
https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding this to Android, I suppose an iOS patch will come later. Checkuser work will probably take a bit more time to materialize.
Thanks everyone for chiming in!
-- Yuvi Panda T http://yuvi.in/blog
-- Oliver Keyes Research Analyst Wikimedia Foundation
Mobile-l mailing list Mobile-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/mobile-l
-- Oliver Keyes Research Analyst Wikimedia Foundation
-- Oliver Keyes Research Analyst Wikimedia Foundation
Sure; just run by my desk when you have a free moment. I've checked in with Yuvi and we've fixed it for Android (the existing code shooould work, there was just a delta between 'how the code works' and 'how the code was reported to work'; I'm going to work on iOS for a bit now.
On 16 April 2014 09:26, Adam Baso abaso@wikimedia.org wrote:
Sounds good. Available today? I have an open calendar 1105-1355 and 1405-COB.
-Adam
On Wed, Apr 16, 2014 at 9:15 AM, Oliver Keyes okeyes@wikimedia.orgwrote:
So, it identifies the first one as Android, but can't pick out version number, and identifies the second as running Mobile Safari, but can't pick out the OS or device. I would recommend tweaking and testing these strings before deploying them if we want accurate device numbers (and we totally want accurate device numbers). Happy to help with the 'testing' part. Will chip in on gerrit directly.
On 15 April 2014 19:17, Oliver Keyes okeyes@wikimedia.org wrote:
Cool. I'm going to run the ua-parser over them tomorrow to see if it recognises the devices or not, if that works?
On 15 April 2014 18:31, Adam Baso abaso@wikimedia.org wrote:
And here's a patchset just posted for review for the iOS app.
https://gerrit.wikimedia.org/r/126188
Here are examples of what the UAs will look like in the logs:
WikipediaApp/2.0-alpha-2014-04-11 (Android/4.4.2) WikipediaApp/iOS-Wikipedia-4.0-4.0 (iPhone OS/7.1)
(I suspect "4.0-4.0" will change in the latter, though, as production build is approached.)
On Fri, Apr 4, 2014 at 4:04 PM, Oliver Keyes okeyes@wikimedia.orgwrote:
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the next alpha goes out for Android (in a few hours) it should have the new user agent!
On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com wrote: > As followup: QChris did the actual work of reading the relevant RfC > for me, and for now I've settled on: > > WikipediaApp/<version> (<OS>/<version>; <form-factor>) > > https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for adding > this to Android, I suppose an iOS patch will come later. Checkuser > work will probably take a bit more time to materialize. > > Thanks everyone for chiming in!
-- Yuvi Panda T http://yuvi.in/blog
-- Oliver Keyes Research Analyst Wikimedia Foundation
Mobile-l mailing list Mobile-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/mobile-l
-- Oliver Keyes Research Analyst Wikimedia Foundation
-- Oliver Keyes Research Analyst Wikimedia Foundation
Okay, worked it out for iOS. You want something like...
WikipediaApp/v2.0 (iPhone OS 4_4; [tablet/mobile])
The last bit is only necessary if we want to be able to identify (broadly) whether it's a mobile or tablet device. I think we do, but I am prepared to be overruled :). With the above user agent, ua-parser provides:
{'device': {'family': 'Other'}, 'os': {'major': '4', 'patch_minor': None, 'minor': '4', 'family': 'iOS', 'patch': None}, 'user_agent': {'major': None, 'minor': None, 'family': 'Mobile Safari', 'patch': None}, 'string': ' WikipediaApp/v2.0 (iPhone OS 4_4; tablet)'}
On 16 April 2014 09:47, Oliver Keyes okeyes@wikimedia.org wrote:
Sure; just run by my desk when you have a free moment. I've checked in with Yuvi and we've fixed it for Android (the existing code shooould work, there was just a delta between 'how the code works' and 'how the code was reported to work'; I'm going to work on iOS for a bit now.
On 16 April 2014 09:26, Adam Baso abaso@wikimedia.org wrote:
Sounds good. Available today? I have an open calendar 1105-1355 and 1405-COB.
-Adam
On Wed, Apr 16, 2014 at 9:15 AM, Oliver Keyes okeyes@wikimedia.orgwrote:
So, it identifies the first one as Android, but can't pick out version number, and identifies the second as running Mobile Safari, but can't pick out the OS or device. I would recommend tweaking and testing these strings before deploying them if we want accurate device numbers (and we totally want accurate device numbers). Happy to help with the 'testing' part. Will chip in on gerrit directly.
On 15 April 2014 19:17, Oliver Keyes okeyes@wikimedia.org wrote:
Cool. I'm going to run the ua-parser over them tomorrow to see if it recognises the devices or not, if that works?
On 15 April 2014 18:31, Adam Baso abaso@wikimedia.org wrote:
And here's a patchset just posted for review for the iOS app.
https://gerrit.wikimedia.org/r/126188
Here are examples of what the UAs will look like in the logs:
WikipediaApp/2.0-alpha-2014-04-11 (Android/4.4.2) WikipediaApp/iOS-Wikipedia-4.0-4.0 (iPhone OS/7.1)
(I suspect "4.0-4.0" will change in the latter, though, as production build is approached.)
On Fri, Apr 4, 2014 at 4:04 PM, Oliver Keyes okeyes@wikimedia.orgwrote:
Cool. Let me know when some examples go out and I'll take a look at how our standardised UA parser treats them for the mobile metrics.
On 4 April 2014 11:52, Yuvi Panda yuvipanda@gmail.com wrote:
> https://gerrit.wikimedia.org/r/#/c/123117/ got merged, so when the > next alpha goes out for Android (in a few hours) it should have the > new user agent! > > > > On Wed, Apr 2, 2014 at 3:54 AM, Yuvi Panda yuvipanda@gmail.com > wrote: > > As followup: QChris did the actual work of reading the relevant RfC > > for me, and for now I've settled on: > > > > WikipediaApp/<version> (<OS>/<version>; <form-factor>) > > > > https://gerrit.wikimedia.org/r/#/c/123117/ is a WIP patch for > adding > > this to Android, I suppose an iOS patch will come later. Checkuser > > work will probably take a bit more time to materialize. > > > > Thanks everyone for chiming in! > > > > -- > Yuvi Panda T > http://yuvi.in/blog >
-- Oliver Keyes Research Analyst Wikimedia Foundation
Mobile-l mailing list Mobile-l@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/mobile-l
-- Oliver Keyes Research Analyst Wikimedia Foundation
-- Oliver Keyes Research Analyst Wikimedia Foundation
-- Oliver Keyes Research Analyst Wikimedia Foundation
On Wed, Apr 16, 2014 at 10:07:44AM -0700, Oliver Keyes wrote:
Okay, worked it out for iOS. You want something like...
WikipediaApp/v2.0 (iPhone OS 4_4; [tablet/mobile])
The “v” after the slash in “WikipediaApp/v2.0” looks unusual. While the RfC does not forbid them, none of the RfC's examples comes with such a “v”, nor do we see “v”s used much in the wild.
Best regards, Christian
Okay, I updated https://gerrit.wikimedia.org/r/#/c/126188/.
Here's what the UA looks like:
WikipediaApp/4.0 (iPhone OS 7.1; Phone)
Is that right?
Note, the #.# will change from 4.0 to a #.#.# per Apple guidelines. Looks like the YAML covers that.
The form factor can take on "Phone", "Tablet", or "Other". If another form factor idiom is introduced, I suppose we'll want to update the code to reflect another form factor idiom and not just use "Other".
-Adam
On Wed, Apr 16, 2014 at 12:45 PM, Christian Aistleitner < christian@quelltextlich.at> wrote:
On Wed, Apr 16, 2014 at 10:07:44AM -0700, Oliver Keyes wrote:
Okay, worked it out for iOS. You want something like...
WikipediaApp/v2.0 (iPhone OS 4_4; [tablet/mobile])
The "v" after the slash in "WikipediaApp/v2.0" looks unusual. While the RfC does not forbid them, none of the RfC's examples comes with such a "v", nor do we see "v"s used much in the wild.
Best regards, Christian
-- ---- quelltextlich e.U. ---- \ ---- Christian Aistleitner ---- Companies' registry: 360296y in Linz Christian Aistleitner Gruendbergstrasze 65a Email: christian@quelltextlich.at 4040 Linz, Austria Phone: +43 732 / 26 95 63 Fax: +43 732 / 26 95 63 Homepage: http://quelltextlich.at/
Hi Adam,
On Wed, Apr 16, 2014 at 02:10:01PM -0700, Adam Baso wrote:
Okay, I updated https://gerrit.wikimedia.org/r/#/c/126188/.
Here's what the UA looks like:
WikipediaApp/4.0 (iPhone OS 7.1; Phone)
Is that right?
Looks great to me.
Extra hug for getting rid of those stupid iUnderscores and using proper dots *hug*.
I didn't dare to suggest it. But I am really glad to see them gone \o/
Note, the #.# will change from 4.0 to a #.#.# per Apple guidelines. Looks like the YAML covers that.
Yes. I double-checked locally and for
WikipediaApp/4.0 (iPhone OS 7.1.0; Phone)
ua-parser detects OS as family: iOS, major: 7, minor: 1, patch: 0.
Thanks!
Have fun, Christian
Hi Oliver,
On Wed, Apr 16, 2014 at 09:15:53AM -0700, Oliver Keyes wrote:
So, it identifies the first one as Android, but can't pick out version number,
you're lagging behind master. Android version should be correctly picked since
https://github.com/tobie/ua-parser/commit/e9d5238513b3184ef0cbcb6e4c403a20f4...
and identifies the second as running Mobile Safari, but can't pick out the OS or device.
I would recommend tweaking and testing these strings before deploying them [...]
Regardless of how you tweak the User-Agent strings ... how would you get ua_parser the to report the User-Agent family as “WikipediaApp”?
You would have to teach ua_parser about it.
And if we have to teach ua_parser something anyways ... we might as well stick with standards for our User-Agents and teach ua_parser to extract not only the User-Agent, but also to be more robust when extracting OS information.
That would benefit us and ua_parser.
It's just a simple two line patch [1].
if we want accurate device numbers (and we totally want accurate device numbers).
Device information is not at all included in the User-Agent. And that's actually good. No need to leak all over the Internet who uses which device.
But as device information is not included in the User-Agent, we cannot parse it out to get per device numbers.
Have fun, Christian
[1] Something along the lines of (probably do not want to split version number parts at -, but do not know)
git diff HEAD^ diff --git a/regexes.yaml b/regexes.yaml index 3ecd0b4..cfdf595 100644 --- a/regexes.yaml +++ b/regexes.yaml @@ -1,6 +1,8 @@ user_agent_parsers: #### SPECIAL CASES TOP ####
+ - regex: '(WikipediaApp)/([^-]*)-([^-]*)-([^ ]*) ' + # HbbTV standard defines what features the browser should understand. # but it's like targeting "HTML5 browsers", effective browser support depends on the model # See os_parsers if you want to target a specific TV @@ -645,7 +647,7 @@ os_parsers: # iOS # http://en.wikipedia.org/wiki/IOS_version_history ########## - - regex: '(CPU OS|iPhone OS|CPU iPhone) (\d+)[_.](\d+)(?:[_.](\d+))?' + - regex: '(CPU OS|iPhone OS|CPU iPhone)[ /](\d+)[_.](\d+)(?:[_.](\d+))?' os_replacement: 'iOS'
# remaining cases are mostly only opera uas, so catch opera as to not catch iphone spoofs
On 16 April 2014 10:21, Christian Aistleitner christian@quelltextlich.atwrote:
Hi Oliver,
On Wed, Apr 16, 2014 at 09:15:53AM -0700, Oliver Keyes wrote:
So, it identifies the first one as Android, but can't pick out version number,
you're lagging behind master. Android version should be correctly picked since
https://github.com/tobie/ua-parser/commit/e9d5238513b3184ef0cbcb6e4c403a20f4...
Good catch! Updated at my end.
and identifies the second as running Mobile Safari, but can't pick out the OS or device.
I would recommend tweaking and testing these strings before deploying them [...]
Regardless of how you tweak the User-Agent strings ... how would you get ua_parser the to report the User-Agent family as "WikipediaApp"?
You would have to teach ua_parser about it.
And if we have to teach ua_parser something anyways ... we might as well stick with standards for our User-Agents and teach ua_parser to extract not only the User-Agent, but also to be more robust when extracting OS information.
That would benefit us and ua_parser.
It's just a simple two line patch [1].
Sure; for app identification we could just handle it ourselves - we probably want to avoid pushing WM-specific strings upstream.
if we want accurate device numbers (and we totally want accurate device numbers).
Device information is not at all included in the User-Agent. And that's actually good. No need to leak all over the Internet who uses which device.
But as device information is not included in the User-Agent, we cannot parse it out to get per device numbers.
It's not at the moment, but it could be, and I think that just including device */class/* (tablet versus mobile versus other) would probably be fine. I don't see how this would be 'leak[ing] all over the internet'.
Have fun, Christian
[1] Something along the lines of (probably do not want to split version number parts at -, but do not know)
git diff HEAD^ diff --git a/regexes.yaml b/regexes.yaml index 3ecd0b4..cfdf595 100644 --- a/regexes.yaml +++ b/regexes.yaml @@ -1,6 +1,8 @@ user_agent_parsers: #### SPECIAL CASES TOP ####
- regex: '(WikipediaApp)/([^-]*)-([^-]*)-([^ ]*) '
- # HbbTV standard defines what features the browser should understand. # but it's like targeting "HTML5 browsers", effective browser support
depends on the model # See os_parsers if you want to target a specific TV @@ -645,7 +647,7 @@ os_parsers: # iOS # http://en.wikipedia.org/wiki/IOS_version_history ##########
- regex: '(CPU OS|iPhone OS|CPU iPhone) (\d+)[_.](\d+)(?:[_.](\d+))?'
- regex: '(CPU OS|iPhone OS|CPU iPhone)[
/](\d+)[_.](\d+)(?:[_.](\d+))?' os_replacement: 'iOS'
# remaining cases are mostly only opera uas, so catch opera as to not catch iphone spoofs
-- ---- quelltextlich e.U. ---- \ ---- Christian Aistleitner ---- Companies' registry: 360296y in Linz Christian Aistleitner Gruendbergstrasze 65a Email: christian@quelltextlich.at 4040 Linz, Austria Phone: +43 732 / 26 95 63 Fax: +43 732 / 26 95 63 Homepage: http://quelltextlich.at/
Analytics mailing list Analytics@lists.wikimedia.org https://lists.wikimedia.org/mailman/listinfo/analytics
Hi Yuvi,
On Thu, Mar 27, 2014 at 03:30:24AM +0530, Yuvi Panda wrote:
On Thu, Mar 27, 2014 at 3:20 AM, Yuvi Panda yuvipanda@gmail.com wrote:
How about:
WikipediaApp/<version> <OS>/<form-factor>/<version>
TL;DR: Looks good. But what about either of
WikipediaApp/<version> (<OS> <version>; <form-factor>) WikipediaApp/<version> (<OS>; <version>; <form-factor>) WikipediaApp/<version> (<OS>/<version>; <form-factor>)
? They would more closely reflect both the User-Agents in the wild and also RFC 2616 [1].
------------------------------------------------------------
According to RFC 2616's approach to formalize User-Agents [2], your proposed variant would be two “product” [3] tokens.
While the RFC does not forbid doing so, I'd argue that the OS part would better fit a “comment” [4] token.
Firstly, there is RFC's interpretation of using subproducts (2nd, 3rd, ... product) for User-Agent header:
[...] and any subproducts which form a significant part of the user agent.
I do not think that the OS forms a significant /part/ of the user agent. Rather the user agent runs on top of the OS (and might well be specific to the OS). So for me, the RFC imposed metaphor does not work too nicely for the original suggested User-Agent.
Secondly (and that is more relevant), typical existing User-Agents also put OS information into “comment” tokens. Some random examples:
Mozilla/5.0 (Windows NT 6.1; WOW64) Mozilla/5.0 (Linux; Android 4.3; en-gb; SAMSUNG GT-I9505 Build/JSS15J) Opera/9.80 (Windows NT 5.1; U; ru) Dalvik/1.4.0 (Linux; U; Android 2.3.6; GT-S5360 Build/GINGERBREAD) Articles/295 (iPhone; iOS 7.0.4; Scale/2.00)
So I'd put the OS part into a comment token like:
WikipediaApp/<version> (<OS>/<form-factor>/<version>)
But I am not sure about the order and separators in “<OS>/<form-factor>/<version>”.
If “<version>” refers to the version of the “<OS>” I guess it should go with the “<OS>” and not be separated by “<form-factor>”.
Whether it'd be
<OS>/<version> <OS> <version> <OS>; <version>
depends mostly on what you want to convey. The slash variant is closer to the “product” token of the RFC but the RFC does not require to use “product” tokens within a “comment” token. In fact, all of the above three variants can be found in the wild (compare the “comment” token in the above examples)
Regardless how one models the “<OS>” and “<version>”, I would not separate it from “<form-factor>” by a slash. While I could not find it in the RFC, use in the wild pretty much everywhere uses semi-colons to separate different parts of the “comment” token (see examples above).
Hence, from my point of view any of
WikipediaApp/<version> (<OS> <version>; <form-factor>) WikipediaApp/<version> (<OS>; <version>; <form-factor>) WikipediaApp/<version> (<OS>/<version>; <form-factor>)
would more closely reflect both the User-Agents in the wild and also RFC 2616.
Have fun, Christian
[1] http://tools.ietf.org/html/rfc2616 [2] http://tools.ietf.org/html/rfc2616#section-14.43 [3] http://tools.ietf.org/html/rfc2616#section-3.8 [4] Search for "Comments can be included" on http://tools.ietf.org/html/rfc2616#page-17 to find the definition.