So we missed the show at Commons

[Ming]

When Google shut down Panoramio, someone started up a bot to scrape it into Commons. Amazingly, this was halted with a relatively small helping of dramah.

[Kumioko]

The commons project is in such need of editors that they can't hope to catch up. I edited there myself doing hundreds of thousands of edits in a pretty short time until I was unilaterally banned by Michael Maggs. I would probably still be editing thousands of edits a week had it not been for him. Personally I have no sympathy at all for the WMF sites anymore. They do not want editors, they want followers and they deserve what they get, which is the eventual collapse of the sites!

[Ming]

Ming has done a very little commons work, but (a) Ming doesn't have a decent camera these days, and (b) Ming's ideas are so out of step with commons that Ming just doesn't want to ahve anything to do with it.

[Poetlister]

On the whole, I'd say that one of the main uses of Commons is to make Wikipedia look good in comparison.

[Kumioko]

On the whole, I'd say that one of the main uses of Commons is to make Wikipedia look good in comparison.

Lol, well I would also add that the community, or lack there of on Commons is what Wikipedia is heading too. So all the problems we see on Commons will be compounded on Wikipedia as the community continues to decline.

[Renée Bagslint]

Commons is ahead of Wikipedia in one important respect -- it has no pretence whatsoever of being a reliable source or having any encyclopaedic value whatsoever. I'm not referring to the carefully curated collections of pornography, distateful though they may be, I'm referring to the fact that there is no reason whatsoever to believe the attribution. Anyone can upload a picture of anything and call it anything, and it will then be used on the so-called encylopaedias without further ado. I presume that there must be quite a number of hoax and/or fake pictures on Commons. Is there a thread for those somewhere?

[Kumioko]

and it has an entire category tree for Porn, including child porn!

[Poetlister]

Commons is ahead of Wikipedia in one important respect -- it has no pretence whatsoever of being a reliable source or having any encyclopaedic value whatsoever.

Wikipedia does not claim to be a reliable source; one Wikipedia article cannot be used to confirm a point in another article. And Commons claims to delete photos with no educational value, though of course the porn collection demonstrates that this is not always done.

[Anroth]

Commons is ahead of Wikipedia in one important respect -- it has no pretence whatsoever of being a reliable source or having any encyclopaedic value whatsoever.

Wikipedia does not claim to be a reliable source; one Wikipedia article cannot be used to confirm a point in another article. And Commons claims to delete photos with no educational value, though of course the porn collection demonstrates that this is not always done.

Ignoring Ronnies inability to understand that WP does not claim to be a reliable source (and has a number of places where this is pointed out explicitly in plain English)

Commons has actually made a dent in the porn - slowly to be sure - but a lot of low quality crap is steadily being removed. Granted this may have to do with a number of bannings of people who liked to claim everything was in scope...

Granted as a % of the total there is still a lot there. But at least it will be good quality photos

[Kumioko]

For what it's worth I was actively chiseling away at that myself when I was unfairly and pointlessly banned from Commonsn by Michael Maggs so he could score points with James Alexander and a couple others that didn't like me. Fortunately Michael has resigned from most of his positions of influence and Commons is better off for that.

[BrillLyle]

There's a very active project to improve Commons called Structured Data

It might be a huge improvement with regards to new uploads, especially by GLAM initiatives.

https://commons.wikimedia.org/wiki/Comm ... tured_data

[Renée Bagslint]

Structured Data is an effort to get lots of volunteers to tag a huge library of images. Who benefits from having all that work done? Step forward any company that is interested in image recognition or computer vision. How nice for them to get all that done for them for free, when they could easily have afforded to pay for it. Still, you get what you pay for, so bad luck if it all turns out to be nonsense. No doubt your driverless car is going to be jolly good at recognising any genital organs or obscure sexual practices that it happens to encounter in the road.

[BrillLyle]

Structured Data is an effort to get lots of volunteers to tag a huge library of images. Who benefits from having all that work done? Step forward any company that is interested in image recognition or computer vision. How nice for them to get all that done for them for free, when they could easily have afforded to pay for it. Still, you get what you pay for, so bad luck if it all turns out to be nonsense. No doubt your driverless car is going to be jolly good at recognising any genital organs or obscure sexual practices that it happens to encounter in the road.

Hi Renée,

I think maybe there is a misunderstanding of the project. The Structured Commons project will improve new image donations by institutional partners who want to incorporate rich metadata and Wikidata in the image donation process.

It is grant funded (or at least has been in the past -- https://commons.wikimedia.org/wiki/Comm ... loan_Grant) and is paid for by WMF via paying developer salaries.

https://commons.wikimedia.org/wiki/Comm ... data/About

I am not thrilled with some of the "community" folks who will have a huge say in how this works, but I've been attending the IRC meetings and I think this will significantly improve the Commons. I don't know about how it will fix things retroactively which is a huge problem but going forward image donations should be better once the project is implemented.

- Erika aka BrillLyle

[Renée Bagslint]

From Commons:Structured data:

The project Structured Data on Wikimedia Commons (2017–19) converts the free media files on Wikimedia Commons to a structured and machine-readable format, so that they become easier to view, search, edit, organize and re-use. To achieve that, the Commons backend is migrated to Wikibase, the same technology as used for Wikidata.

and from Commons:Structured data/About

It is expected that, after 3 years, approximately 5 million media files on Wikimedia Commons will contain some structured metadata - probably more. Depending on community processes, there will also probably be a 'long tail' period of several additional years until all files on Commons are described with some structured data.

To me, that says that there is an expectation that volunteers will retrospectively tag all the existing corpus.

[BrillLyle]

To me, that says that there is an expectation that volunteers will retrospectively tag all the existing corpus.

Yes, I agree that there is a huge assumption that volunteers will donate free digital labor to clean up the corpus.

The mechanism of these technical projects, as it has been explained to me, is:

Developers <----------> Technical users who provide tools, implementation, etc <----------> Editors

I am definitely in the far right category.

I think the weakness of this system (as it has been described to me) is:

• Developers creating things from scratch, not using existing systems (like library and info science structures) that are both functional and proven
• Developers having no experience with end-users' needs. They are typically not end-users
• If I hear "that's a community issue" when grave concerns are expressed I will scream. Not an okay answer or perspective IMO
• The number of people who do the tools/implementation seems to be quite small. And WMF doesn't pay them typically, where they pay the Developers. Problems ensue when free digital labor at this higher technical level is not paid. Expertise should be paid (I believe in paid editors as well)

I don't know, handwringing here. I don't know enough about the nuts and bolts here probably to comment.

I am participating as an active En Wiki editor in hopes of providing feedback that focuses on end-user needs. And to keep an eye on something I care about -- where Pigs is active too I am always a bit concerned. He tends to do things I only find out about much later, things that have negative impact or have been snuck under the radar. Ugh on that (and him).....

- Erika aka ~~~~
I am such a dumbass I always confuse Wikibase with DBpedia.