User talk:Victorgrigas

Welcome to the Commons, Victorgrigas!

First steps tutorial Our first steps help file and our FAQ will help you a lot after registration. They explain how to customize the interface (for example the language), how to upload files and our basic licensing policy. You don't need technical skills in order to contribute here. Be bold contributing here and assume good faith for the intentions of others. This is a wiki - it is really easy. Getting help More information is available at the Community Portal. You may ask questions at the Help desk, Village Pump or on IRC channel #wikimedia-commons. You can also contact an administrator on their talk page. If you have a specific copyright question, ask at Commons talk:Licensing.		Goodies, tips and tricks Put Babel boxes on your user page so others know what languages you can speak and indicate your Graphics abilities. All your uploads are stored in your personal Gallery Please sign your name on Talk pages by typing ~~~~ Use the CommonSense tool to find good categories for your files (then other people can find them too!) To link to an image page, write this: [[:Image:Foo.jpg]], it makes this: Image:Foo.jpg If you're copying files from another project, be sure to use the CommonsHelper Made a mistake? Did you want to rename or move a file? Simply upload the file again and mark the old one like this: {{bad name|Correct name}} For more information read the full Deletion guidelines
(P.S. Would you like to provide feedback on this message?)

Archives 1, 2, 3, 4

Medical videos

Hello, I have been following the conversation about medical videos at https://en.wikipedia.org/wiki/Wikipedia_talk:WikiProject_Medicine#Osmosis_concerns https://en.wikipedia.org/w/index.php?title=Wikipedia_talk:WikiProject_Medicine/Osmosis_RfC&oldid=833259249 with great interest, having been at the Kiwix/internet-in-a-box session at Wikimania.

Years ago I was working at a rural school in the Middle East when the fourth grade class came down with lice, as fourth-graders sometimes do, and I was enlisted to help scrape the labels off the medical shampoo bottles that were given to the students. Everything was fine until one of the students who had not been in class that day was given a bottle with the label still on, and concerned mothers started arriving at the school to have a chat. At that point, it might have been interesting to have had a video on the topic, since a relatively high percentage of women in that area were not literate.

I suspect that such videos might still be useful today in rural and refugee areas, and I wonder if Doc James and OsmoseIt have given any thought to the technology needed for subtitles and translation of public health and medical videos, or if text-to-speech artificial voice could be used to generate a current script in multiple languages. It seems that a technical solution to this problem might also address the same issues that are being raised about revision control and corrections to the scripts of the current videos. I don't know if such a thing exists or if the current technology in speech synthesis has anything to offer this project, but if there is any interest in this approach, I suspect Mr. Victorgrigas would know where to begin.

On a more personal note, the efforts of Wikiproject Medicine are much appreciated, it is truly a lifeline in areas without access to Western medicine. Avery Jensen (talk) 01:33, 30 March 2018 (UTC)

User:Avery Jensen thanks for the kind words. There have been ongoing efforts to translate the subtitles into other languages. With respect to voice dubbing or text to speech of the captions I do not think this has been done yet.

Unfortunately EN WP due to various concerns appear to be against using these videos going forwards and thus this collaboration with Osmosis will likely be rapping up or at least stopping for the time being.

Within respect to inline references within the scripts, that is possible but more work than Osmosis is willing to put in. Doc James (talk · contribs · email) 01:37, 30 March 2018 (UTC)

Doc James, I don't see that the enwiki is against the videos per se, they are just against the way they are currently being used: the commercial aspect, the lack of correction mechanism (which if you think about it, Offline Wikipedia has the same issue with updating), the inline citation/verifiability issue, the placement of the videos in the article, and especially the lack of the expected evaluation after the initial start of the project. None of these is insurmountable, in fact, the most difficult thing is to convince a partnering institution to have patience with our weird collaborative culture. And for now there does seem to be a consensus to keep the videos on Commons and link to them from the article. Or maybe it is time for a sub-project that is oriented more towards public or international health and does not depend so much on enwiki. Avery Jensen (talk) 02:09, 30 March 2018 (UTC)

Some of the request includes that they exactly follow our articles. Problem is we do not have a collaborator willing to do that.

We also flip text back and forth all the time. Many do not want video overviews as this editing cannot be done. Doc James (talk · contribs · email) 05:11, 30 March 2018 (UTC)

Well, I don't agree with that (the video following the article word for word), and that is where an evaluation at the end of the trial might have come in handy. Text is completely different from video; you can do more, or at least different things with video and that is why you use it. The criticism of the breastfeeding video I thought was important, as the phrasing was misleading, and this could impact on maternal/infant health. Likewise with the abortion video (that was not by Osmosis), the arguments against this were overwhelming. But the criticisms of the epilepsy video, hatted under the green, don't work for me. Why should the screen not show Na+ and Ca2+ ions or talk about neurotransmitters and receptors, as long as you can understand the concept. This is "roughly tuned input" and is helpful for the learner.

If you wanted to keep updating videos, it might be useful to establish some kind of priorities, and a schedule. For example if a breastfeeding film was getting a lot of page views or was critical in some other way, and had misleading language, you would want to put those changes at the top of the priority list, and try to get this updated within 6 months, or whatever your criteria. If it was just a question of some American slang terms creeping into the text, or agreeing on an open-source font that does not look like click bait (yes please!), this could be postponed to the next scheduled review unless there was some other reason, like translation, to bump the priority.

It seems like you need a different way to look at video; if the medical project community cannot work this into their current model, you need a new model. It seems like more and more education portions of WP are being outsourced.

Just a thought, Wikisource might be a useful place for scripts, as opposed to a read-only google doc. https://docs.google.com/document/d/1UvgeiOGmrUKaP32pra1NTOb2xsO_KaTnHdTIlWDM5jQ/edit Looks like this is already open source. Not sure which Commons file this is for. There is a way to connect Commons items with Wikisource. Fae would know who to ask, pinging Fæ. Avery Jensen (talk) 17:08, 30 March 2018 (UTC)

┌─────────────────────────────────┘
This is kind of sad really. The videos have been removed from the articles and it looks like there probably won't be any further curation of the current videos, unless the medical project does it. There is a list of issues with individual videos; the videos themselves are at Category:Videos from Osmosis. Some of the people involved in the RfC have written a very thoughtful essay about the role of video in articles at WP:Wikipedia is not YouTube. Unfortunately it's more about what not to do rather than a vision of what is possible.

I agree it would be nice for Wikimedians to understand how video is made, unfortunately much of Wikimedia involves a very long learning curve, either looking over someone's shoulder or googling for outside resources that may or may not be accurate. Even for those of us who have been involved in training and outreach, there is little likelihood anyone will send us to a conference or come around to give a workshop.

It would be nice to have MP3. How can we get MP3 and some kind of editor? Avery Jensen (talk) 04:46, 31 March 2018 (UTC)

Howdy - well, where do you want to have this conversation?

Specifically per this osmosis issue -

These videos are:

Stylized animations with burned-in English language text. Some are arguing about the facts in them. For the wiki-ethic all these things are problem. The style of the videos are not uniform to the rest of Wikipedia or Wikimedia projects nor could they become that because we don't have the raw video editing files, only the exported (and then converted to .webm) versions of them. We can't change anything in the videos, only maybe cut portions out if we can find someone who knows how to do that (like me). The English is, well English. it's only one language. And the facts in it can't be updated. Even if you got a robot to read updated text do you want to watch a video with a robot reading the text? Just have Siri read the Wikipedia article then. There are a lot of engineers out there who love to dream up stuff like that to get clicks but I'd strongly advise against embarking on anything like that. Humans need to be involved in making most of any video, because humans know what content ought to be there, just like text.

I think these Osmosis videos are in a way on the right track though. I haven't read all the threads involved, but I would love to find out who made them, and see if we could get ahold of the RAW editing files of one video, just to show everyone how this video was made. I imagine that you would have some motion graphics files, a .wav audio track, some .edl file (edit decision list this is basically the instructions to tell the computer what sequence in which to montage everything together), a script for the narrator and a few other assets like .jpegs for logos and stuff. Those things are the starting off point for video editing, collaborative or otherwise. If our community had those things to mess around with then you would start to see them say 'ah ha! I have fixed the factual errors in the script!' and 'ah ha! I have an open-source font we all like, or can at least tolerate' and 'ah ha! we've standardized a style of putting text in a video that works with left to right to top to sideways to upside down to inside out to any language.' Right now we have only the final product of a long assembly line of video production and we need access to the beginning of the conveyor belt to fix/wikify anything.

Now, I think that yes, medical videos that illustrate topics that 'make things plain' are good, but right now the wild west of video on Wikipedia might have a new sheriff in town?

Generally I think that if any video has style issues, for now (while there is no web-based collaborative video editing tool) videos could be put into a 'see also' section (as either links or thumbnails) or a gallery.

I also think that a serious video editing and production discussion needs to happen. I'm happy to be involved with this. The Wikimedia community needs to understand how video is made.

Here are my thoughts:

1.) Wikimedia’s place in the world of online video and audio:

We should give our communities the ability to appropriately illustrate sections of Wikipedia articles with video (or other media or projects) as they so decide. How? Develop a video-editing philosophy and corresponding rules that are appropriate to and supportive of Wikipedia articles similar to the rules of editing the text of Wikipedia. I think we should encourage our communities to collaboratively develop ideas, draft scripts, write storyboards, draft shotlists, production schedules, finance (through grants), shoot using our app (with their own hardware or rented hardware through grants), upload, log and edit, copy and migrate existing footage from other sources, caption, translate and transcribe - any of the steps needed to produce video and audio and make it accessible in all the world’s languages.

This includes the ability to:

Create independent documentaries Create tutorials Record ‘oral citations’ Fork Wikipedia and other project content to use as scripts to make ‘video versions’ of Wikipedia articles

We would also have the chance to visit other multimedia-centric projects focused around FLOSS media in line with the ‘backbone of free knowledge’ in the strategic plan:

Music and video streaming Free music apps New video uses (360 video, 3D and so forth)

2.) Creator gap:

Right now we're not attracting video or audio producers to our projects because of technical and philosophical issues, not because the general public has no or poor ability to create or edit quality content. As of the time of writing, billions of people have HD or 4K video cameras in their phones that can shoot HDR, stabilize, with microphones that can record high-quality audio and so forth. Sites use these now-common tools, and acquire large volumes of high quality content. Examples:

https://freesound.org/ (sound effects) https://www.youtube.com/ (video) https://vimeo.com/ (professional video) https://www.jamendo.com/ (music)

In terms of existing Wikimedians creating high quality content: once trimming, editing and logging features are built, longer footage or footage that is 95% bad can be picked apart and only the 5% useful material can be used. Video-shooting is standard on most DSLR cameras today. The commons community has not been uploading raw video that they shoot because of the technical issues involved with uploading that footage and the futility of having footage on commons that can’t be edited. If tools were enabled, the existing commons community would be capable of shooting video should they feel inclined.

Also, simply having tools like collaborative video editing around allows professional content producers (like news organizations, medical content producers, libraries, schools and so forth) to choose to 'dump' their raw footage onto Wikimedia Commons (or some other new project) expecting that our community can find use in editing it. Wikimedians can use the raw material.

3.) Usage scenario mapping:

Here are the technical considerations and costs of large data storage: https://phabricator.wikimedia.org/T178171

4.) Patents:

There are plenty of tools and tasks that can be achieved per video on Wikimedia without having a community debate about the use of patent-issues, and I’d argue that before getting support from the community about supporting the mp4 or any other patent encumbered anything, tools like video editing and trimming and audio tools involving the patent-free MP3 and the Mjpeg should be taken on first, to SHOW the value of video on the projects and start to build a community who uses such tools and understands their value before a hypothetical is used to argue for use of patents.

Victorgrigas (talk) 02:34, 30 March 2018 (UTC)

Sure you want to watch an osmosis video with a robot reading the text. For various reasons, like retention. Also if you can edit the text to remove errors, or make updates, also for ease of translation. I'm not sure how you would get Siri to read the captions though. If the medical community is passionate enough to evaluate the videos for accuracy, that would be the most critical. They seem to know exactly what they want -- even if they're a bit chaotic at the moment -- whether it's one important fact that's wrong, whether it's out of date and lacking basic new information, or whether it just doesn't follow the text of the WP article. Bots are not ideal, but people are only willing to do so much. Are people really willing to record a whole new video every time there's a medical advance? I got the idea that osmosis budget was just for the original videos and not for keeping them current. That Medium article I think misses the point of the children's videos, that they're pre-verbal...it's just a "one of these things is not like the others" sort of matching thing, the baby head laughs for the right answer. It's a way to indicate "yes" without using words.

How do you cut the videos? I see this was done earlier, and the modified version uploaded. I don't suppose it's possible to have segments and just swap out a segment when it gets outdated.

The other points

1). Collaborative is always nice, but realistically people don't have time to do that, or with photography in particular they are used to working alone. As it is, we can't even get bus fare to take pictures of authors at book festivals. Also there is the educational mission of Commons, although that tends to get interpreted pretty broadly.

2). I have original video of historical events I wouldn't mind uploading, but it does need cropping a little, and I don't know where to get the tools for that. Also, webm. Does Commons need cell phone videos? I recently saw a proposal for a cell phone photo contest, not sure what happened to it. I suspect this is the future.

3). Not sure if this means "limiting factor" or "this will work with a minimum of effort" or "WMF can afford more servers".

4). I suspect anything that is not open source will not gain traction. But why do we not have MP3 already and why were we not talking about this at Wikimania? I suspect User:Fuzheado might have some opinions about this. Avery Jensen (talk) 05:58, 30 March 2018 (UTC)