My watchflickr tool [1] includes an option to upload an image with a suitable CC license to commons using a "bot" account [2]. So far, I have received no complaints about bad uploads, and from its gallery it seems OK as well (except some duplicate uploads).
Now, anyone can upload an image to flickr, and release it under CC-BY(-SA). Same as wikipedia, right? Except that wikipedia uploads are probably screened much more thoroughly for cases that are clearly not under the given license.
My CommonsHelper tool [3] eases the transfer of images from wikipedia to the commons, and has been used a whooping 93435 times this year. Assuming that every use results in an upload on commons, over 330 images /per day/ enter commons this way, a not unimportant proportion of the 5000 uploads per day, especially considering that it will only take images that have a commons-compatible license.
However, users still have to save the image on their own computer, then upload them under their own user account, which is annoying and time-consuming.
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
I would like to propose the reactivation of that feature. Concernes about unsuitable uploads through the bot account are superflous, IMHO, since images are screened thrice this way: 1. On the wikipedia where the image was originally uploaded 2. By the CommonsHelper (e.g. it will reject "fair use" images from en) 3. On commons, by the usual suspects :-)
Which is two levels of screening more than direct uploads to commons, which were, last time I checked, enabled ;-)
With SUL in sight, new uploads will shift from wikipedias to commons, but there's still a lot of images around: en : >766,000 (but that includes "fair use") de : ~119,000 (no "fair use") fr : ~37,000
All in all, I'd estimate that there's between 0.5 and 1 million images on the wikipedias that would be suitable for commons. You can see how "save locally, then upload manually" annoyance can scale up :-)
Cheers, Magnus
[1] http://tools.wikimedia.de/~magnus/watchflickr.php [2] http://commons.wikimedia.org/wiki/User:File_Upload_Bot_%28Magnus_Manske%29 [3] http://tools.wikimedia.de/~magnus/commonshelper.php
On 16/10/2007, Magnus Manske magnusmanske@googlemail.com wrote:
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
I would like to propose the reactivation of that feature. Concernes about unsuitable uploads through the bot account are superflous, IMHO, since images are screened thrice this way:
- On the wikipedia where the image was originally uploaded
- By the CommonsHelper (e.g. it will reject "fair use" images from en)
- On commons, by the usual suspects :-)
Which is two levels of screening more than direct uploads to commons, which were, last time I checked, enabled ;-)
That's some dubious counting :P
Could CH use a bot similar to the Flickr upload bot? I am really impressed by its method (although it is a little counterintuitive the first time). That way there is a solid record of who the transferrer is... which is the sticking point to allowing "anonymous" transfers, for me.
cheers, Brianna
On 10/16/07, Brianna Laugher brianna.laugher@gmail.com wrote:
On 16/10/2007, Magnus Manske magnusmanske@googlemail.com wrote:
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
I would like to propose the reactivation of that feature. Concernes about unsuitable uploads through the bot account are superflous, IMHO, since images are screened thrice this way:
- On the wikipedia where the image was originally uploaded
- By the CommonsHelper (e.g. it will reject "fair use" images from en)
- On commons, by the usual suspects :-)
Which is two levels of screening more than direct uploads to commons, which were, last time I checked, enabled ;-)
That's some dubious counting :P
My point is that transfering images from wikipedias through CommonsHelper is likely to result in little copyright violations, compared to our working "grab a user name and upload whatever you want" method :-)
Could CH use a bot similar to the Flickr upload bot? I am really impressed by its method (although it is a little counterintuitive the first time). That way there is a solid record of who the transferrer is... which is the sticking point to allowing "anonymous" transfers, for me.
My whole point is "one-click transfer". IMHO it's less (the same at best) work to save locally and upload than go through the counter-intuitive edit-and-click-here loop.
Also, uploading under my account (instead of a bot) might convey some false sense of ownership. I got some "your-image-is-being-deleted" messages over the years for things I only transfered from a wikipedia. I don't know any more about the images than is given in the description; noone bothers to contact the original author on wikipedia. Uploading through the bot would * show the "tagging" admin that it's no use to write on the talk page ('cause its a bot), but look up the original uploader (one click) * ease nuking of bad images ("bot upload, original uploader long gone => bye-bye")
Magnus
"Magnus Manske" magnusmanske@googlemail.com wrote on Tue, 16 Oct 2007 13:56:36 +0100:
[...]
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
I would like to propose the reactivation of that feature.
I used to use a bot made by HardDisk from de.wp to do exactly that, but that one doesn't work any longer.
* Is CommonsHelper able to rename files during the upload? * Does it fix links to new image? * Does it tag the wikipedia image with {{NowCommons}}? * License template switching would be nice as well, see http://tools.wikimedia.de/~luxo/litafi.php for an implementation.
Either way, I'd say: YES!
de : ~119,000 (no "fair use")
... but Panoramafreiheit, Logos and pictures from 2007-100=1907
Maybe you would have to change some minor details on the way it handles {{information}} but I would like to see such a tool working for me again.
Best regards and strong support,
Flo
On 10/16/07, Florian Straub Flominator@gmx.net wrote:
"Magnus Manske" magnusmanske@googlemail.com wrote on Tue, 16 Oct 2007 13:56:36 +0100:
[...]
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
I would like to propose the reactivation of that feature.
I used to use a bot made by HardDisk from de.wp to do exactly that, but that one doesn't work any longer.
- Is CommonsHelper able to rename files during the upload?
Yes.
- Does it fix links to new image?
If you mean, "does it edit the original wikipedia", then no.
- Does it tag the wikipedia image with {{NowCommons}}?
I remember coding a button for this; I probably just turned it off.
- License template switching would be nice as well, see http://tools.wikimedia.de/~luxo/litafi.php for an implementation.
CommonsHelper already does a lot of template transitions, including licenses.
Either way, I'd say: YES!
A binary decision! On a wiki(p|m)edia mailing list! I have to frame this one :-)
de : ~119,000 (no "fair use")
... but Panoramafreiheit, Logos and pictures from 2007-100=1907
I can filter for Panoramafreiheit and Logo categories. Images from 1907 should be OK on Commons, right? We could name it {{PD-old-50}}, if all fails...
Maybe you would have to change some minor details on the way it handles {{information}} but I would like to see such a tool working for me again.
I'm open for suggestion (maybe off-list?).
Best regards and strong support,
Flo
Thanks, Magnus
Maybe you would have to change some minor details on the way it handles {{information}} but I would like to see such a tool working for me again.
I'm open for suggestion (maybe off-list?).
Dann wollen wir mal: Wenn das Lizenztag irgendwas mit "self" sagt, sollte in Source stehen: * selfmade * originally uploaded as ... on [[w:en:Main Paige|en.wp]] (oder so)
Die Description könnte außerdem {{Projektsprache|einzufügender Text}}
Sind nur Vorschläge ...
Kann man eigentlich deinen Uploadbot auch ohne FIST für Flickr-Bilder nutzen? Eventuell könnte man ihn ja sogar bei Flinfo andocken ...
Gruß,
Flo
np: Hans Zimmer - Homeland
On 16/10/2007, Magnus Manske magnusmanske@googlemail.com wrote:
My watchflickr tool [1] includes an option to upload an image with a suitable CC license to commons using a "bot" account [2]. So far, I have received no complaints about bad uploads, and from its gallery it seems OK as well (except some duplicate uploads).
It's a winner. I've been using it on placeholder images on en:wp. WatchFlickr has about one hit in fifty, but that one is worth the effort. Often the image found is crappy and needs cropping and tweaking to be any good, but often enough it's bot-uploadable.
Now, anyone can upload an image to flickr, and release it under CC-BY(-SA). Same as wikipedia, right? Except that wikipedia uploads are probably screened much more thoroughly for cases that are clearly not under the given license.
The one thing I find problematic with the auto-uploading bot is that it accepts the tag on Flickr as valid. I've found far too many cases where the tag on Flickr is clearly invalid and the Flickr uploader's just been sloppy.
That is: have the robot leave the image as requiring human validation, and that should be cleared up.
My CommonsHelper tool [3] eases the transfer of images from wikipedia to the commons, and has been used a whooping 93435 times this year. Assuming that every use results in an upload on commons, over 330 images /per day/ enter commons this way, a not unimportant proportion of the 5000 uploads per day, especially considering that it will only take images that have a commons-compatible license. However, users still have to save the image on their own computer, then upload them under their own user account, which is annoying and time-consuming. CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
I think a checking step like Flickr checking would be suitable.
I would like to propose the reactivation of that feature. Concernes about unsuitable uploads through the bot account are superflous, IMHO, since images are screened thrice this way:
- On the wikipedia where the image was originally uploaded
- By the CommonsHelper (e.g. it will reject "fair use" images from en)
- On commons, by the usual suspects :-)
Tag it as needing step 3 and it'll be fine until then.
All in all, I'd estimate that there's between 0.5 and 1 million images on the wikipedias that would be suitable for commons. You can see how "save locally, then upload manually" annoyance can scale up :-)
Indeed.
Providing you add a tag for "could a human check this please" then anything that makes it easier would be marvellous, thank you!
- d.
On 10/16/07, David Gerard dgerard@gmail.com wrote:
That is: have the robot leave the image as requiring human validation, and that should be cleared up.
I used to add {{flickrreview}}, but since a human has to click on the link in watchflickr, and the license tag is CC on flickr, I skip this now. I can re-add {{flickrreview}} if that is wanted.
Providing you add a tag for "could a human check this please" then anything that makes it easier would be marvellous, thank you!
Is there some generic template for this? Something like {{humanreview}}, preferrably something that has a JavaScript one-click-removal? Should there be one?
Cheers, Magnus
On 16/10/2007, Magnus Manske magnusmanske@googlemail.com wrote:
On 10/16/07, David Gerard dgerard@gmail.com wrote:
That is: have the robot leave the image as requiring human validation, and that should be cleared up.
I used to add {{flickrreview}}, but since a human has to click on the link in watchflickr, and the license tag is CC on flickr, I skip this now. I can re-add {{flickrreview}} if that is wanted.
I'd recommend it. I saw a few too many images that were just a bit too high-quality and I had to decide couldn't reliably be assumed to be by the Flickr uploader. If someone else decides they're OK then fine, but the bot still just trusts what Flickr says.
Providing you add a tag for "could a human check this please" then anything that makes it easier would be marvellous, thank you!
Is there some generic template for this? Something like {{humanreview}}, preferrably something that has a JavaScript one-click-removal? Should there be one?
Similar to Flickr-review I suppose. I dunno, I'm not a Commons admin and don't actually do this sort of janitorial work, I've avoided it so far ;-)
-d.
"Magnus Manske" magnusmanske@googlemail.com wrote on Tuesday, October 16, 2007 7:37 PM:
On 10/16/07, David Gerard dgerard@gmail.com wrote:
That is: have the robot leave the image as requiring human validation, and that should be cleared up.
I used to add {{flickrreview}}, but since a human has to click on the link in watchflickr, and the license tag is CC on flickr, I skip this now. I can re-add {{flickrreview}} if that is wanted.
Why not just watching the contributions of the bot account?
Regards,
Flo
Magnus Manske wrote:
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
Reactivation sounds good to me, with the addition of a review tag. Random commoners may or may not be familiar with the kinds of dubious claims made on many en: images, so it's useful to gather them into a distinct group that can be worked over when they are long gone from "latest files".
Stan
On 10/16/07, Stan Shebs stanshebs@earthlink.net wrote:
Magnus Manske wrote:
CommonsHelper does have the functionality to do direct uploads via the aforementioned bot account, however, that has been deactivated since forever, due to concerns.
Reactivation sounds good to me, with the addition of a review tag. Random commoners may or may not be familiar with the kinds of dubious claims made on many en: images, so it's useful to gather them into a distinct group that can be worked over when they are long gone from "latest files".
OK, state of the union: * WatchFlickr now adds {{flickrreview}} again * CommonsHelper has a "direct upload" switch (off by default) * Direct upload by CommonsHelper will add {{BotMoveToCommons}} with the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
If too many weird things turn up through this channel, tell me and I'll turn it off (or filter with new criteria). Block bot in case of emergency :-)
I'll be tweaking it over the next few days (existing image detection, NowCommons link, etc.). I'm also working on a "unified image detector" tool that will supercede WatchFlickr and MissingImages, eventually.
Cheers, Magnus
On 16/10/2007, Magnus Manske magnusmanske@googlemail.com wrote:
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
All it lacks now is a "This has been moved to commons" template on the original so people know it can be deleted harmlessly.
Nice one! *thunderous applause*
- d.
On Tue, 16 Oct 2007 22:15:26 +0300, Magnus Manske magnusmanske@googlemail.com wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
I'm a bit lost in this conversation. Did any human review the license information and such before or right after the image was transferred to Commons?
On 10/16/07, Samuli Lintula samuli@samulilintula.net wrote:
On Tue, 16 Oct 2007 22:15:26 +0300, Magnus Manske magnusmanske@googlemail.com wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
I'm a bit lost in this conversation. Did any human review the license information and such before or right after the image was transferred to Commons?
Well, * /presumably/ it was checked when it was uploaded in the original wikipedia * /presumably/ the guy who transfers it checked (how else does he know the name?) * /presumably/ the guy who transfers it checks the end result * /presumably/ the good people at commons check it when it is transfered
IMHO that's an improvement over the "normal" upload to commons ("oh, I found a pretty image on the web, I'll upload it to commons!";-)...
If the original wikipedia image has a valid license tag, how should one check anyway? The obvious criteria are: * Image is too large to be a thumbnail stolen from some webpage * Image does not contain borders, logos, or "(c)" texts * Image either has a source given (eg., nasa.gov), or * looks amateurish enough to be made by the original uploader ;-)
Yay for reasonable doubt!
Cheers, Magnus
On 10/16/07, Samuli Lintula samuli@samulilintula.net wrote:
On Tue, 16 Oct 2007 22:15:26 +0300, Magnus Manske magnusmanske@googlemail.com wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
I'm a bit lost in this conversation. Did any human review the license information and such before or right after the image was transferred to Commons?
Well,
- /presumably/ it was checked when it was uploaded in the original
wikipedia
- /presumably/ the guy who transfers it checked (how else does he know
the name?)
- /presumably/ the guy who transfers it checks the end result
- /presumably/ the good people at commons check it when it is transfered
IMHO that's an improvement over the "normal" upload to commons ("oh, I found a pretty image on the web, I'll upload it to commons!";-)...
If the original wikipedia image has a valid license tag, how should one check anyway? The obvious criteria are:
- Image is too large to be a thumbnail stolen from some webpage
- Image does not contain borders, logos, or "(c)" texts
- Image either has a source given (eg., nasa.gov), or
- looks amateurish enough to be made by the original uploader ;-)
Yay for reasonable doubt!
Do I detect a bit of a mocking attitude here? I don't see any reason for that. Why would we want to lose a chance to check images that are being uploaded on Commons? It is very good practice that we check if source is provided, if the source seems plausible, if permission is provided, if there is anything dubious about the image etc. We should try to check these things with fresh uploads and transfers from other projects.
If we are to presume something, I presume that local projects have at least as many copyvios or sourceless/permissionless images as Commons. That is, within the range of 10-30 %.
On 10/17/07, samuli@samulilintula.net samuli@samulilintula.net wrote:
On 10/16/07, Samuli Lintula samuli@samulilintula.net wrote:
On Tue, 16 Oct 2007 22:15:26 +0300, Magnus Manske magnusmanske@googlemail.com wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
I'm a bit lost in this conversation. Did any human review the license information and such before or right after the image was transferred to Commons?
Well,
- /presumably/ it was checked when it was uploaded in the original
wikipedia
- /presumably/ the guy who transfers it checked (how else does he know
the name?)
- /presumably/ the guy who transfers it checks the end result
- /presumably/ the good people at commons check it when it is transfered
IMHO that's an improvement over the "normal" upload to commons ("oh, I found a pretty image on the web, I'll upload it to commons!";-)...
If the original wikipedia image has a valid license tag, how should one check anyway? The obvious criteria are:
- Image is too large to be a thumbnail stolen from some webpage
- Image does not contain borders, logos, or "(c)" texts
- Image either has a source given (eg., nasa.gov), or
- looks amateurish enough to be made by the original uploader ;-)
Yay for reasonable doubt!
Do I detect a bit of a mocking attitude here?
Just a little, maybe :-)
I don't see any reason for that. Why would we want to lose a chance to check images that are being uploaded on Commons? It is very good practice that we check if source is provided, if the source seems plausible, if permission is provided, if there is anything dubious about the image etc. We should try to check these things with fresh uploads and transfers from other projects.
Absolutely. And there's nothing to stop anyone from checking the uploads through the bot account. They are no different from "normal" uploads, except many (most?) of them will already have been checked on a wikipedia, as well as by the "transferer", and they all have an {{Information}} template.
If we are to presume something, I presume that local projects have at least as many copyvios or sourceless/permissionless images as Commons. That is, within the range of 10-30 %.
Probably. But by that reasoning, even if we were to blindly transfer every image from, say, en.wikipedia that has {{GFDL}} in it, 70-90% of the uploads through the bot would be acceptable for commons. Is that the case for "normal" uploads? I'd think that the upload rate of copyvios on commons through "normal" means will increase with the popularity of the project.
I'm not saying that all images on en.wikipedia that are tagged {{GFDL}} are good for commons. All I'm saying is that most of it is, and that the average copyvio rate will likely be lower that what we get through "normal" uploads. /Especially/ for en.wikipedia, where it is (relatively) easy to call something "fair use" instead of GFDL, and be done with it. Those won't even make it through CommonsHelper in the first place.
Cheers, Magnus
Maybe I can give a bit of input from my experience at transferring images from pt.wikipedia to Commons. Local uploads were disabled last year on pt.wiki and from then on a slow, painstaking process of moving images to Commons has started. I don't know what I would do without CommonsHelper ;) The question of which images would be good to transfer or not came about; naturally, people transferring images to Commons didn't want to end up in a "Commons blacklist" for uploading images without all the proper information. So we turned into forming a task force to tackle the problem. All images were tagged with {{a verificar}} (to be verified), which put them in a category where people could go through and check such images. "People" in this context means a small group of users with good knowledge about licensing and such, including two Commons admins. Images deemed to be good to go to Commons were then tagged with {{move-to-commons}} which added the CommonsHelper button. It has been mostly the same group of people transferring the images to Commons anyway. So in this particular case, there was a double screening. It takes a LOT of time (it's a small task force), but we've had very little to no problems with such uploads. The point is: maybe local wikis can develop processes of screening good images to upload to Commons and continue to inform/improve knowledge in local communities on the advantages of using Commons as a global repository, leaving local uploads only for fair use images and such (am I wrong or Icelandic Wikipedia did just this recently?)
Adding a category to a bot-uploaded image is good to have a further level of screening in Commons, in the end it's as Magnus has said: it's another upload to check anyway, regardless of who made it; the category will ease the screening, so thanks for that :) Bots do wacky stuff sometimes too, and I certainly don't trust the Flickr uploads because of the licensing problems already addressed before.
We still have a couple of hundreds of images to go on pt.wiki, so *yay to direct upload!* ;) Sorry for the long message and for the (slight?) off-topic Cheers, Patrícia
Magnus Manske magnusmanske@googlemail.com wrote: On 10/17/07, samuli@samulilintula.net wrote:
On 10/16/07, Samuli Lintula wrote:
On Tue, 16 Oct 2007 22:15:26 +0300, Magnus Manske wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
An example of how it looks: http://commons.wikimedia.org/w/index.php?title=Image%3ARosemary_Clooneys_hom...
I'm a bit lost in this conversation. Did any human review the license information and such before or right after the image was transferred to Commons?
Well,
- /presumably/ it was checked when it was uploaded in the original
wikipedia
- /presumably/ the guy who transfers it checked (how else does he know
the name?)
- /presumably/ the guy who transfers it checks the end result
- /presumably/ the good people at commons check it when it is transfered
IMHO that's an improvement over the "normal" upload to commons ("oh, I found a pretty image on the web, I'll upload it to commons!";-)...
If the original wikipedia image has a valid license tag, how should one check anyway? The obvious criteria are:
- Image is too large to be a thumbnail stolen from some webpage
- Image does not contain borders, logos, or "(c)" texts
- Image either has a source given (eg., nasa.gov), or
- looks amateurish enough to be made by the original uploader ;-)
Yay for reasonable doubt!
Do I detect a bit of a mocking attitude here?
Just a little, maybe :-)
I don't see any reason for that. Why would we want to lose a chance to check images that are being uploaded on Commons? It is very good practice that we check if source is provided, if the source seems plausible, if permission is provided, if there is anything dubious about the image etc. We should try to check these things with fresh uploads and transfers from other projects.
Absolutely. And there's nothing to stop anyone from checking the uploads through the bot account. They are no different from "normal" uploads, except many (most?) of them will already have been checked on a wikipedia, as well as by the "transferer", and they all have an {{Information}} template.
If we are to presume something, I presume that local projects have at least as many copyvios or sourceless/permissionless images as Commons. That is, within the range of 10-30 %.
Probably. But by that reasoning, even if we were to blindly transfer every image from, say, en.wikipedia that has {{GFDL}} in it, 70-90% of the uploads through the bot would be acceptable for commons. Is that the case for "normal" uploads? I'd think that the upload rate of copyvios on commons through "normal" means will increase with the popularity of the project.
I'm not saying that all images on en.wikipedia that are tagged {{GFDL}} are good for commons. All I'm saying is that most of it is, and that the average copyvio rate will likely be lower that what we get through "normal" uploads. /Especially/ for en.wikipedia, where it is (relatively) easy to call something "fair use" instead of GFDL, and be done with it. Those won't even make it through CommonsHelper in the first place.
Cheers, Magnus
_______________________________________________ Commons-l mailing list Commons-l@lists.wikimedia.org http://lists.wikimedia.org/mailman/listinfo/commons-l
--------------------------------- For ideas on reducing your carbon footprint visit Yahoo! For Good this month.
Wednesday, 17 October 2007, Magnus Manske wrote:
Absolutely. And there's nothing to stop anyone from checking the uploads through the bot account. They are no different from "normal" uploads, except many (most?) of them will already have been checked on a wikipedia, as well as by the "transferer", and they all have an {{Information}} template.
Ignoring the comparison to "brand new uploads" for a moment to compare this to the manual moving of images to Commons, which I think is more relevant, it's plain to see that this would reduce the instances of * images changing names when being moved to Commons (makes it hard to identify transferred images) * images losing their Wikipedia histories (lots of uploaders fail to cite the images' original locations and/or descriptions) * images changing licenses (lots of users assume everything on Wikipedia is GFDL--this affects both images that should be transferred, but with a CC-license for example, and fair use images) * thumbnails and preview images being uploaded
I could give real-world examples of users committing all of the above mistakes en masse, thus causing a lot of extra work (more so than if they'd just uploaded plain copyvios), but I don't want to single anyone out here.
Yes, bots do wacky stuff sometimes, but people do it way more often. As long as the bot makes sure there's an {{information}} template with the source filled out and an acceptable license, I'd be more inclined to trust the bot to transfer the image properly than the random Wikipedia user who decides they need an image from English Wikipedia on their own language edition.
And like you said, the images still get screened by the posse patrolling the latest files on Commons, just like all new uploads.
On Wed, 17 Oct 2007 08:53:07 +0300, samuli@samulilintula.net wrote:
Do I detect a bit of a mocking attitude here? I don't see any reason for that. Why would we want to lose a chance to check images that are being uploaded on Commons? It is very good practice that we check if source is provided, if the source seems plausible, if permission is provided, if there is anything dubious about the image etc. We should try to check these things with fresh uploads and transfers from other projects.
If we are to presume something, I presume that local projects have at least as many copyvios or sourceless/permissionless images as Commons. That is, within the range of 10-30 %.
It seems that the community would rather see a higher success rate and is quite irate when a human review fails as badly as a bot review would: http://commons.wikimedia.org/wiki/Commons:Administrators%27_noticeboard/User...
Magnus Manske wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
Worked very nicely for me - got a couple dozen PD stamp images moved over with little ado. The most work came in when I decided to give some of the images more reasonable names, had to manually edit a bunch of en: refs to them. Auto-editing of image references would be a nice addition.
Stan
Stan Shebs wrote:
Magnus Manske wrote:
OK, state of the union:
- WatchFlickr now adds {{flickrreview}} again
- CommonsHelper has a "direct upload" switch (off by default)
- Direct upload by CommonsHelper will add {{BotMoveToCommons}} with
the appropriate source language, so things will end in subcategories of [[Category:Files moved to Commons requiring review]]
Worked very nicely for me - got a couple dozen PD stamp images moved over with little ado. The most work came in when I decided to give some of the images more reasonable names, had to manually edit a bunch of en: refs to them. Auto-editing of image references would be a nice addition.
Stan
It's probably doable with http://commons.wikimedia.org/wiki/User:Orgullobot/commands
http://commons.wikimedia.org/wiki/User:CommonsDelinker/commands does indeed support this. However, sysop access is required to edit the page. Might be a good idea to contact [[commons:User:Bryan]] as he controls CommonsDelinkerHelper, an admin bot editing that page.
Cheers! Siebrand
-----Oorspronkelijk bericht----- Van: commons-l-bounces@lists.wikimedia.org [mailto:commons-l-bounces@lists.wikimedia.org] Namens Platonides Verzonden: woensdag 17 oktober 2007 22:39 Aan: commons-l@lists.wikimedia.org Onderwerp: Re: [Commons-l] Direct transfer of properly licensed images from wikipedias to commons?
[..]
It's probably doable with http://commons.wikimedia.org/wiki/User:Orgullobot/commands