Dear movement fellows,
Impact is crucial for our movement, and although metrics will always be imperfect, we must strive to reinvent ourselves and always come up with new innovative ways of measuring what we bring to the Wikimedia projects, to free knowledge, and to human society.
Measuring impact regarding collections of media holds its own challenges and although we have been focusing on this for a while now, much work still lies ahead.
We were inspired by the “bytes added” metric, one of the pinnacles of written content expansion measurement, which goes beyond mere edit count. The same reasoning holds true for media:a puny upload count cannot come close to the real awesomeness.
This is why, as we appreciate that size matters, Wikimedia France quality commitee is proud to introduce its brand new set of metrics: the pixel count and the quality pixel count − since quality is of firstmost importance.
You may query the Pixel count metric for your FDC reports as part of our wm-metrics webapp [1]
Furthermore, an implementation of these new metrics will also ship with our new new (teasing!) product [2]
As of April 1st 2015 Wikimedia France has supported the upload on Wikimedia Commons of:
- 1 229 694 933 639 pixels [3]
- among those pixels, 22 407 932 851 are quality pixels (18,223512%) [4]
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
Confident in the relevance of these new indicators, we would be delighted and honored to see the Pixel count integrated in the Global Metrics.
As always we welcome feedback, hugs and pull requests.
Sincerely, For the quality committee of Wikimedia France Caroline, Jean-Fred, Pierre-Selim and Petit Tigre
[1] https://tools.wmflabs.org/wm-metrics/fdc [2] https://github.com/Commonists/MediaCollectionDB/commit/4c2ab42f83e894c9dd317... [3] http://quarry.wmflabs.org/query/2882 [4] http://quarry.wmflabs.org/query/2886
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
Also, some historic pixels may be worth several newer ones. Pixel valuation is an art as much as it is a science.
-- Marc
Ouais bon, poisson d'avril? :-)
2015-04-01 22:09 GMT+02:00 Marc A. Pelletier marc@uberbox.org:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
Also, some historic pixels may be worth several newer ones. Pixel valuation is an art as much as it is a science.
Thank you for your valuable input, we will think about it for next iterations.
-- Marc
Ouais bon, poisson d'avril? :-)
:-)
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
As always, every tool is developed thinking about Wikipedia and Commons, never all the other sister projects! What about those poor pixels? Are they different from Commons pixels??!?!!1!
Luckily, today the WMF said otherwise: see the "Sister projects" news here https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2015-04-01/News_a...
Aubrey
On Wed, Apr 1, 2015 at 10:12 PM, Pierre-Selim pierre-selim@huard.info wrote:
2015-04-01 22:09 GMT+02:00 Marc A. Pelletier marc@uberbox.org:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute
pixels,
encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
Also, some historic pixels may be worth several newer ones. Pixel valuation is an art as much as it is a science.
Thank you for your valuable input, we will think about it for next iterations.
-- Marc
Ouais bon, poisson d'avril? :-)
:-)
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Pierre-Selim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
They are not free pixels.
Only real free pixels deserve to be counted. Le 1 avr. 2015 23:00, "Andrea Zanni" zanni.andrea84@gmail.com a écrit :
As always, every tool is developed thinking about Wikipedia and Commons, never all the other sister projects! What about those poor pixels? Are they different from Commons pixels??!?!!1!
Luckily, today the WMF said otherwise: see the "Sister projects" news here
https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2015-04-01/News_a...
Aubrey
On Wed, Apr 1, 2015 at 10:12 PM, Pierre-Selim pierre-selim@huard.info wrote:
2015-04-01 22:09 GMT+02:00 Marc A. Pelletier marc@uberbox.org:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute
pixels,
encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
Also, some historic pixels may be worth several newer ones. Pixel valuation is an art as much as it is a science.
Thank you for your valuable input, we will think about it for next iterations.
-- Marc
Ouais bon, poisson d'avril? :-)
:-)
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Pierre-Selim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
I give this project FFFFFF out of a possible FFFFFF.
On Wed, Apr 1, 2015 at 10:06 PM, Christophe Henner < christophe.henner@gmail.com> wrote:
They are not free pixels.
Only real free pixels deserve to be counted. Le 1 avr. 2015 23:00, "Andrea Zanni" zanni.andrea84@gmail.com a écrit :
As always, every tool is developed thinking about Wikipedia and Commons, never all the other sister projects! What about those poor pixels? Are they different from Commons pixels??!?!!1!
Luckily, today the WMF said otherwise: see the "Sister projects" news here
https://en.wikipedia.org/wiki/Wikipedia:Wikipedia_Signpost/2015-04-01/News_a...
Aubrey
On Wed, Apr 1, 2015 at 10:12 PM, Pierre-Selim pierre-selim@huard.info wrote:
2015-04-01 22:09 GMT+02:00 Marc A. Pelletier marc@uberbox.org:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute
pixels,
encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background
pixels
out of the FOV aren't worth as much!
Also, some historic pixels may be worth several newer ones. Pixel valuation is an art as much as it is a science.
Thank you for your valuable input, we will think about it for next iterations.
-- Marc
Ouais bon, poisson d'avril? :-)
:-)
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe:
https://lists.wikimedia.org/mailman/listinfo/wikimedia-l,
mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
-- Pierre-Selim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
On Wed, Apr 1, 2015 at 1:09 PM, Marc A. Pelletier marc@uberbox.org wrote:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
I assume you mean "assigning *mass*" to the pixels. Weight is so Earth-centric!
I would like to join the kawaii-pixel WikiProject. Please let me know when we start debating the relative merits of various color models, and naming conventions, and kawaii-challenged accessibility tools. Thanks!
-- quiddity
I love today. On Apr 1, 2015 5:28 PM, "quiddity" pandiculation@gmail.com wrote:
On Wed, Apr 1, 2015 at 1:09 PM, Marc A. Pelletier marc@uberbox.org wrote:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
I assume you mean "assigning *mass*" to the pixels. Weight is so Earth-centric!
I would like to join the kawaii-pixel WikiProject. Please let me know when we start debating the relative merits of various color models, and naming conventions, and kawaii-challenged accessibility tools. Thanks!
-- quiddity
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Everybody always tries to get rid of the content pixels because they beat up the other pixels, but I tell you what, if you don't give those content creator pixels what they want they're going to take their RGB and go home and THEN where will your silly little projects be without any content pixels, hmmmmmm?
On Wed, Apr 1, 2015 at 5:38 PM, Aleksey Bilogur aleksey.bilogur@gmail.com wrote:
I love today. On Apr 1, 2015 5:28 PM, "quiddity" pandiculation@gmail.com wrote:
On Wed, Apr 1, 2015 at 1:09 PM, Marc A. Pelletier marc@uberbox.org wrote:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute
pixels,
encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background pixels out of the FOV aren't worth as much!
I assume you mean "assigning *mass*" to the pixels. Weight is so Earth-centric!
I would like to join the kawaii-pixel WikiProject. Please let me know when we start debating the relative merits of various color models, and naming conventions, and kawaii-challenged accessibility tools. Thanks!
-- quiddity
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Lol j aime pierre selim!! On Apr 1, 2015 11:52 PM, "Katherine Casey" fluffernutter.wiki@gmail.com wrote:
Everybody always tries to get rid of the content pixels because they beat up the other pixels, but I tell you what, if you don't give those content creator pixels what they want they're going to take their RGB and go home and THEN where will your silly little projects be without any content pixels, hmmmmmm?
On Wed, Apr 1, 2015 at 5:38 PM, Aleksey Bilogur <aleksey.bilogur@gmail.com
wrote:
I love today. On Apr 1, 2015 5:28 PM, "quiddity" pandiculation@gmail.com wrote:
On Wed, Apr 1, 2015 at 1:09 PM, Marc A. Pelletier marc@uberbox.org wrote:
On 15-04-01 03:58 PM, Pierre-Selim wrote:
This is only the beginning: next step is the measurement of cute
pixels,
encyclopedic pixels and amazing pixels.
That metric is all wrong, because it presumes that all pixels are equally valuable. Surely, you should be also assigning weights to pixels depending on how much information they carry - background
pixels
out of the FOV aren't worth as much!
I assume you mean "assigning *mass*" to the pixels. Weight is so Earth-centric!
I would like to join the kawaii-pixel WikiProject. Please let me know when we start debating the relative merits of various color models, and naming conventions, and kawaii-challenged accessibility tools. Thanks!
-- quiddity
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Dear Pierre-Selim,
I look forward to discussing this new metric at the Wikimedia Conference.
I might even take photographs of the deliberations and upload them to Commons in order to improve my personal pixel metric.
Have you figured out a way to translate pixels into multiple languages?
I hope you will document the new pixel metric, and the methods for measuring it, in the Learning Patterns Library.
Regards, Pine On Apr 1, 2015 12:59 PM, "Pierre-Selim" pierre-selim@huard.info wrote:
Dear movement fellows,
Impact is crucial for our movement, and although metrics will always be imperfect, we must strive to reinvent ourselves and always come up with new innovative ways of measuring what we bring to the Wikimedia projects, to free knowledge, and to human society.
Measuring impact regarding collections of media holds its own challenges and although we have been focusing on this for a while now, much work still lies ahead.
We were inspired by the “bytes added” metric, one of the pinnacles of written content expansion measurement, which goes beyond mere edit count. The same reasoning holds true for media:a puny upload count cannot come close to the real awesomeness.
This is why, as we appreciate that size matters, Wikimedia France quality commitee is proud to introduce its brand new set of metrics: the pixel count and the quality pixel count − since quality is of firstmost importance.
You may query the Pixel count metric for your FDC reports as part of our wm-metrics webapp [1]
Furthermore, an implementation of these new metrics will also ship with our new new (teasing!) product [2]
As of April 1st 2015 Wikimedia France has supported the upload on Wikimedia Commons of:
1 229 694 933 639 pixels [3]
among those pixels, 22 407 932 851 are quality pixels (18,223512%) [4]
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
Confident in the relevance of these new indicators, we would be delighted and honored to see the Pixel count integrated in the Global Metrics.
As always we welcome feedback, hugs and pull requests.
Sincerely, For the quality committee of Wikimedia France Caroline, Jean-Fred, Pierre-Selim and Petit Tigre
[1] https://tools.wmflabs.org/wm-metrics/fdc [2]
https://github.com/Commonists/MediaCollectionDB/commit/4c2ab42f83e894c9dd317... [3] http://quarry.wmflabs.org/query/2882 [4] http://quarry.wmflabs.org/query/2886
-- Pierre-Selim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Le Thu, 02 Apr 2015 01:26:07 +0200, Pine W wiki.pine@gmail.com a écrit:
Dear Pierre-Selim,
I look forward to discussing this new metric at the Wikimedia Conference.
I might even take photographs of the deliberations and upload them to Commons in order to improve my personal pixel metric.
Have you figured out a way to translate pixels into multiple languages?
Just walk accross streets or countries and show the pixels to different native speakers: you have translated the pixels. Be aware not to lost some pixels during the translation.
~ Seb35
I hope you will document the new pixel metric, and the methods for measuring it, in the Learning Patterns Library.
Regards, Pine On Apr 1, 2015 12:59 PM, "Pierre-Selim" pierre-selim@huard.info wrote:
Dear movement fellows,
Impact is crucial for our movement, and although metrics will always be imperfect, we must strive to reinvent ourselves and always come up with new innovative ways of measuring what we bring to the Wikimedia projects, to free knowledge, and to human society.
Measuring impact regarding collections of media holds its own challenges and although we have been focusing on this for a while now, much work still lies ahead.
We were inspired by the “bytes added” metric, one of the pinnacles of written content expansion measurement, which goes beyond mere edit count. The same reasoning holds true for media:a puny upload count cannot come close to the real awesomeness.
This is why, as we appreciate that size matters, Wikimedia France quality commitee is proud to introduce its brand new set of metrics: the pixel count and the quality pixel count − since quality is of firstmost importance.
You may query the Pixel count metric for your FDC reports as part of our wm-metrics webapp [1]
Furthermore, an implementation of these new metrics will also ship with our new new (teasing!) product [2]
As of April 1st 2015 Wikimedia France has supported the upload on Wikimedia Commons of:
1 229 694 933 639 pixels [3]
among those pixels, 22 407 932 851 are quality pixels (18,223512%)
[4]
This is only the beginning: next step is the measurement of cute pixels, encyclopedic pixels and amazing pixels.
Confident in the relevance of these new indicators, we would be delighted and honored to see the Pixel count integrated in the Global Metrics.
As always we welcome feedback, hugs and pull requests.
Sincerely, For the quality committee of Wikimedia France Caroline, Jean-Fred, Pierre-Selim and Petit Tigre
[1] https://tools.wmflabs.org/wm-metrics/fdc [2]
https://github.com/Commonists/MediaCollectionDB/commit/4c2ab42f83e894c9dd317... [3] http://quarry.wmflabs.org/query/2882 [4] http://quarry.wmflabs.org/query/2886
-- Pierre-Selim _______________________________________________ Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
Wikimedia-l mailing list, guidelines at: https://meta.wikimedia.org/wiki/Mailing_lists/Guidelines Wikimedia-l@lists.wikimedia.org Unsubscribe: https://lists.wikimedia.org/mailman/listinfo/wikimedia-l, mailto:wikimedia-l-request@lists.wikimedia.org?subject=unsubscribe
wikimedia-l@lists.wikimedia.org