[Wikimedia-l] Very good news!

classic Classic list List threaded Threaded
4 messages Options
Reply | Threaded
Open this post in threaded view
|

[Wikimedia-l] Very good news!

Milos Rancic-2
This is an extraordinary news for us! For almost 10 years I was hoping
to see that and, finally, I've seen it!

In short, it seems that we reached the bottom in participation in 2014
and that we are now slowly going upwards.

My claim is based on the analysis [1] of the Eric Zachte's
participation statistics on English Wikipedia [2], but I am almost
sure that the rest of the projects more or less mirror it. But,
anyway, I encourage others to check other projects and other relevant
factors and see if their results correlate with what I have found. The
reasons for the change in trends should be also detected.

If we are looking Eric's statistics from 2010 onwards, it is not
immediately obvious if we are going up or down. We reached the peak in
2007 (German Wikipedia somewhat earlier, other projects later, but
English Wikipedia is approximately 50% of our activity and its weight
is too strong for other projects to balance our overall activity).
After that peak, we went down as quickly as we reached the peak. Then,
in 2010, the trends flattened.

However, it was not a stagnation, but barely visible recession.
However, that "barely visible recession" removed approximately 20% of
the very active editors in the period from 2010 to 2014, while the
"visible one" -- from 2007 to 2010 -- was also approximately 20%. At
that point of time, in 2014, the next 10 years would for sure drive
Wikipedia and Wikimedia movement into insignificance.

Comparing such data is also tricky. It's not just necessary to compare
the same months (January 2010 with January 2011, 2012 etc.), but there
could be "freak" months, which are not following general trends.

That's why I used two methods: One is coloring the months by place in
comparison to the months of the previous years. The other is average
number per year.

There are at least a couple of important conclusions:

1) Negative trends have been reversed.

2) Both 2015 and 2016 were not just better than 2013 and 2014, but
even better than 2012, while 2016 is just a little bit worse than
2011!

3) December 2016 was even better than December 2010!

4) I could guess that the period June-November 2016 was worse than the
same period in 2015 because of the political turbulence. Without them
-- as May and December 2016 likely show -- 2016 would be not just
better, but much better than 2015 and maybe even better than 2010.

I would say that the reversal is still fragile and that we should do
whatever we've been doing the last two years. Yes, detecting what
we've been doing good (or bad) is not that easy to detect. But, yes,
better analysis of all of all of the processes should be definitely
done.

I hope that this shows that we are at the beginning of our
Renaissance, Wikimedia Renaissance and that the Dark Wikimedia Age is
behind us! So, please join me in enjoying that fact, even I could be
wrong. It definitely sounds definitely amazing, even it could be just
my imagination! :)

[1] https://docs.google.com/spreadsheets/d/1IXYoTI_nCBhhuJAknH5KL450_D3V67KWTHuoEAh6540/edit?usp=sharing
[2] https://stats.wikimedia.org/EN/ChartsWikipediaEN.htm

--
Milos

_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>
Reply | Threaded
Open this post in threaded view
|

Re: [Wikimedia-l] Very good news!

Alessandro Marchetti
These are data for English wikipedia, right?
You should compare the whole platforms. That's because bilingualism is increasing in many countries but not in the same direction. For example Italian students foreign language skills drastically increased over the last 10 years, so they edit also on English wikipedia. It's not sure that the opposite occurs. It's the same with emigration. Many 25-30 years old PhD students and Postdoc leave Italy and move to other countries. If that area is for example UK (a very common destination) they start to edit enwikipedia. This could happen also in France, whilst many of them have to learn German or Dutch or Swedish and it takes more time.
I "study" flow of users from platform, you can see when I leave welcome message or propose autopatrolled flag here and there and they are most probably asymmetrical at the moment.
If someone is interested, there are some way to try to sample this flows in an objective way. Happy to share with you. 

    Il Domenica 19 Febbraio 2017 3:33, Milos Rancic <[hidden email]> ha scritto:
 

 This is an extraordinary news for us! For almost 10 years I was hoping
to see that and, finally, I've seen it!

In short, it seems that we reached the bottom in participation in 2014
and that we are now slowly going upwards.

My claim is based on the analysis [1] of the Eric Zachte's
participation statistics on English Wikipedia [2], but I am almost
sure that the rest of the projects more or less mirror it. But,
anyway, I encourage others to check other projects and other relevant
factors and see if their results correlate with what I have found. The
reasons for the change in trends should be also detected.

If we are looking Eric's statistics from 2010 onwards, it is not
immediately obvious if we are going up or down. We reached the peak in
2007 (German Wikipedia somewhat earlier, other projects later, but
English Wikipedia is approximately 50% of our activity and its weight
is too strong for other projects to balance our overall activity).
After that peak, we went down as quickly as we reached the peak. Then,
in 2010, the trends flattened.

However, it was not a stagnation, but barely visible recession.
However, that "barely visible recession" removed approximately 20% of
the very active editors in the period from 2010 to 2014, while the
"visible one" -- from 2007 to 2010 -- was also approximately 20%. At
that point of time, in 2014, the next 10 years would for sure drive
Wikipedia and Wikimedia movement into insignificance.

Comparing such data is also tricky. It's not just necessary to compare
the same months (January 2010 with January 2011, 2012 etc.), but there
could be "freak" months, which are not following general trends.

That's why I used two methods: One is coloring the months by place in
comparison to the months of the previous years. The other is average
number per year.

There are at least a couple of important conclusions:

1) Negative trends have been reversed.

2) Both 2015 and 2016 were not just better than 2013 and 2014, but
even better than 2012, while 2016 is just a little bit worse than
2011!

3) December 2016 was even better than December 2010!

4) I could guess that the period June-November 2016 was worse than the
same period in 2015 because of the political turbulence. Without them
-- as May and December 2016 likely show -- 2016 would be not just
better, but much better than 2015 and maybe even better than 2010.

I would say that the reversal is still fragile and that we should do
whatever we've been doing the last two years. Yes, detecting what
we've been doing good (or bad) is not that easy to detect. But, yes,
better analysis of all of all of the processes should be definitely
done.

I hope that this shows that we are at the beginning of our
Renaissance, Wikimedia Renaissance and that the Dark Wikimedia Age is
behind us! So, please join me in enjoying that fact, even I could be
wrong. It definitely sounds definitely amazing, even it could be just
my imagination! :)

[1] https://docs.google.com/spreadsheets/d/1IXYoTI_nCBhhuJAknH5KL450_D3V67KWTHuoEAh6540/edit?usp=sharing
[2] https://stats.wikimedia.org/EN/ChartsWikipediaEN.htm

--
Milos

_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>

   
_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>
Reply | Threaded
Open this post in threaded view
|

Re: [Wikimedia-l] Very good news!

Erik Zachte-3
Some additional links:

Compare total editors per project (deduplicated)
Caveat: only exists for active editors (5+ per month), not for very active editors (100+ edits per month)
https://stats.wikimedia.org/EN/ProjectTrendsEditors.html

Here is a table with total monthly editors for all Wikimedia wikis combined (deduplicated)
https://stats.wikimedia.org/EN/TablesWikimediaAllProjects_AllMonths.htm
and charts showing columns active editors 5+ and very active editors 100+ from that table.
https://commons.wikimedia.org/wiki/File:Active_editors_over_all_Wikimedia_wikis_-_deduplicated.png

As for causes, WereSpielCheckers wrote an essay in 2013:
https://en.wikipedia.org/wiki/User:WereSpielChequers/Going_off_the_boil%3F
and also did an 'In Focus' for Signpost in 2015:
https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2015-08-26/In_focus

As for edits, instead of editors:
On English Wikipedia monthly edits by registered users is visibly higher in 2015 and 2016 than in 2013 and 2014
https://stats.wikimedia.org/EN/PlotEditsEN.png  (blue line)  
BTW way anonymous and bot edits are also higher than their low point in preceding years, but these metrics don't count for active editors trends.
For similar charts for other Wikipedias see https://stats.wikimedia.org/EN/PlotsPngEditHistoryTop.htm

Erik Zachte

P.S. unrelated but good to know if you dive into Wikistats: February 2017 reports are incomplete (under investigation)

-----Original Message-----
From: Wikimedia-l [mailto:[hidden email]] On Behalf Of Alessandro Marchetti
Sent: Sunday, February 19, 2017 6:52
To: Wikimedia Mailing List
Subject: Re: [Wikimedia-l] Very good news!

These are data for English wikipedia, right?
You should compare the whole platforms. That's because bilingualism is increasing in many countries but not in the same direction. For example Italian students foreign language skills drastically increased over the last 10 years, so they edit also on English wikipedia. It's not sure that the opposite occurs. It's the same with emigration. Many 25-30 years old PhD students and Postdoc leave Italy and move to other countries. If that area is for example UK (a very common destination) they start to edit enwikipedia. This could happen also in France, whilst many of them have to learn German or Dutch or Swedish and it takes more time.
I "study" flow of users from platform, you can see when I leave welcome message or propose autopatrolled flag here and there and they are most probably asymmetrical at the moment.
If someone is interested, there are some way to try to sample this flows in an objective way. Happy to share with you.  

    Il Domenica 19 Febbraio 2017 3:33, Milos Rancic <[hidden email]> ha scritto:
 

 This is an extraordinary news for us! For almost 10 years I was hoping to see that and, finally, I've seen it!

In short, it seems that we reached the bottom in participation in 2014 and that we are now slowly going upwards.

My claim is based on the analysis [1] of the Eric Zachte's participation statistics on English Wikipedia [2], but I am almost sure that the rest of the projects more or less mirror it. But, anyway, I encourage others to check other projects and other relevant factors and see if their results correlate with what I have found. The reasons for the change in trends should be also detected.

If we are looking Eric's statistics from 2010 onwards, it is not immediately obvious if we are going up or down. We reached the peak in
2007 (German Wikipedia somewhat earlier, other projects later, but English Wikipedia is approximately 50% of our activity and its weight is too strong for other projects to balance our overall activity).
After that peak, we went down as quickly as we reached the peak. Then, in 2010, the trends flattened.

However, it was not a stagnation, but barely visible recession.
However, that "barely visible recession" removed approximately 20% of the very active editors in the period from 2010 to 2014, while the "visible one" -- from 2007 to 2010 -- was also approximately 20%. At that point of time, in 2014, the next 10 years would for sure drive Wikipedia and Wikimedia movement into insignificance.

Comparing such data is also tricky. It's not just necessary to compare the same months (January 2010 with January 2011, 2012 etc.), but there could be "freak" months, which are not following general trends.

That's why I used two methods: One is coloring the months by place in comparison to the months of the previous years. The other is average number per year.

There are at least a couple of important conclusions:

1) Negative trends have been reversed.

2) Both 2015 and 2016 were not just better than 2013 and 2014, but even better than 2012, while 2016 is just a little bit worse than 2011!

3) December 2016 was even better than December 2010!

4) I could guess that the period June-November 2016 was worse than the same period in 2015 because of the political turbulence. Without them
-- as May and December 2016 likely show -- 2016 would be not just better, but much better than 2015 and maybe even better than 2010.

I would say that the reversal is still fragile and that we should do whatever we've been doing the last two years. Yes, detecting what we've been doing good (or bad) is not that easy to detect. But, yes, better analysis of all of all of the processes should be definitely done.

I hope that this shows that we are at the beginning of our Renaissance, Wikimedia Renaissance and that the Dark Wikimedia Age is behind us! So, please join me in enjoying that fact, even I could be wrong. It definitely sounds definitely amazing, even it could be just my imagination! :)

[1] https://docs.google.com/spreadsheets/d/1IXYoTI_nCBhhuJAknH5KL450_D3V67KWTHuoEAh6540/edit?usp=sharing
[2] https://stats.wikimedia.org/EN/ChartsWikipediaEN.htm

--
Milos

_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>

   
_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>


_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>
Reply | Threaded
Open this post in threaded view
|

Re: [Wikimedia-l] Very good news!

WereSpielChequers-2
In reply to this post by Milos Rancic-2
Hi Milos, You might want to read this signpost article from 2015
<https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2015-08-26/In_focus>
that talked about the rally in very active editors from EN wiki's 2014
nadir. Another interesting measure of the rally, *Time between ten million
edits * <https://en.wikipedia.org/wiki/Wikipedia:Time_Between_Edits>was
hitting 73 days in late 2014 and is now back down to about 60.

Of course what we don't know is how much of the perceived decline was down
to the edit filters and therefore how much of the increase in the last
couple of years is simply because the easy wins for edit filters have been
achieved; Or how much of the decline was due to the rise of smartphones and
tablets where Wikipedia is much more of a broadcast medium with
comparatively few editors.


Jonathan/WereSpielChequers


Message: 2

> Date: Sun, 19 Feb 2017 03:32:24 +0100
> From: Milos Rancic <[hidden email]>
> To: Wikimedia Mailing List <[hidden email]>
> Subject: [Wikimedia-l] Very good news!
> Message-ID:
>         <CAHPiQ2GZdSg7vGYGwKdRqrPcn9+_2BcYcWHGfoXHenUujVeYGA@mail.
> gmail.com>
> Content-Type: text/plain; charset=UTF-8
>
> This is an extraordinary news for us! For almost 10 years I was hoping
> to see that and, finally, I've seen it!
>
> In short, it seems that we reached the bottom in participation in 2014
> and that we are now slowly going upwards.
>
> My claim is based on the analysis [1] of the Eric Zachte's
> participation statistics on English Wikipedia [2], but I am almost
> sure that the rest of the projects more or less mirror it. But,
> anyway, I encourage others to check other projects and other relevant
> factors and see if their results correlate with what I have found. The
> reasons for the change in trends should be also detected.
>
> If we are looking Eric's statistics from 2010 onwards, it is not
> immediately obvious if we are going up or down. We reached the peak in
> 2007 (German Wikipedia somewhat earlier, other projects later, but
> English Wikipedia is approximately 50% of our activity and its weight
> is too strong for other projects to balance our overall activity).
> After that peak, we went down as quickly as we reached the peak. Then,
> in 2010, the trends flattened.
>
> However, it was not a stagnation, but barely visible recession.
> However, that "barely visible recession" removed approximately 20% of
> the very active editors in the period from 2010 to 2014, while the
> "visible one" -- from 2007 to 2010 -- was also approximately 20%. At
> that point of time, in 2014, the next 10 years would for sure drive
> Wikipedia and Wikimedia movement into insignificance.
>
> Comparing such data is also tricky. It's not just necessary to compare
> the same months (January 2010 with January 2011, 2012 etc.), but there
> could be "freak" months, which are not following general trends.
>
> That's why I used two methods: One is coloring the months by place in
> comparison to the months of the previous years. The other is average
> number per year.
>
> There are at least a couple of important conclusions:
>
> 1) Negative trends have been reversed.
>
> 2) Both 2015 and 2016 were not just better than 2013 and 2014, but
> even better than 2012, while 2016 is just a little bit worse than
> 2011!
>
> 3) December 2016 was even better than December 2010!
>
> 4) I could guess that the period June-November 2016 was worse than the
> same period in 2015 because of the political turbulence. Without them
> -- as May and December 2016 likely show -- 2016 would be not just
> better, but much better than 2015 and maybe even better than 2010.
>
> I would say that the reversal is still fragile and that we should do
> whatever we've been doing the last two years. Yes, detecting what
> we've been doing good (or bad) is not that easy to detect. But, yes,
> better analysis of all of all of the processes should be definitely
> done.
>
> I hope that this shows that we are at the beginning of our
> Renaissance, Wikimedia Renaissance and that the Dark Wikimedia Age is
> behind us! So, please join me in enjoying that fact, even I could be
> wrong. It definitely sounds definitely amazing, even it could be just
> my imagination! :)
>
> [1] https://docs.google.com/spreadsheets/d/1IXYoTI_nCBhhuJAknH5KL450_
> D3V67KWTHuoEAh6540/edit?usp=sharing
> [2] https://stats.wikimedia.org/EN/ChartsWikipediaEN.htm
>
> --
> Milos
>
>
>
_______________________________________________
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines and https://meta.wikimedia.org/wiki/Wikimedia-l
New messages to: [hidden email]
Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, <mailto:[hidden email]?subject=unsubscribe>