The things you do to be informative and accessible

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Duothematic channel. Primary topic is virtual worlds/OpenSim, secondary topic is the Fediverse beyond Mastodon. Follow me from Mastodon or anywhere else in the Fediverse.

Ort:: The Hypergrid

Geschlecht:: Männlich

Homepage:: https://hub.netzgemeinde.eu/profile/jupiter_rowland

The things you do to be informative and accessible

Sat, 23 Sep 2023 12:24:02 +0200 zuletzt bearbeitet: Sat, 23 Sep 2023 12:23:43 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I may need half a day to describe three measly images; CW: long (almost 1,200 characters), alt-text meta, image description meta

Artikel ansehen

Fediverse OpenSim

show all 27 comments

Sat, 23 Sep 2023 13:22:10 +0200 zuletzt bearbeitet: Sat, 23 Sep 2023 13:21:51 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I think it just took me longer to identify three nebulae than it takes other people to write #AltText for 50 pictures.

Now I'm not sure if I'll find the right words to describe them at my usual level of detail...

Sat, 23 Sep 2023 14:45:19 +0200 zuletzt bearbeitet: Sat, 23 Sep 2023 14:45:00 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I've just finished describing an image within an image. It fits into 30x13 pixels in a picture that'll end up at 800x533 pixels.

Only this partial description is about 4,700 characters long now. And it's far from complete. I could go even further. Maybe I'll actually do that later.

I still haven't described the four nebulae yet.

#ImageDescription

Sat, 23 Sep 2023 14:55:37 +0200 zuletzt bearbeitet: Sat, 23 Sep 2023 14:55:18 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

This image is part of a teleporter. I've just used almost 7,000 characters only to describe the teleporter.

IMHO, this is fully justified because nothing like this exists in real life, and most of you have never seen anything like it.

#ImageDescription

Sat, 23 Sep 2023 16:20:03 +0200 zuletzt bearbeitet: Sat, 23 Sep 2023 16:19:43 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Five hours and 18,000 characters so far. Personal record. Maybe Fediverse record for an #ImageDescription. Most recently, about 1,150 characters for a robot that's not much more than a 6-pixel speck in the picture.

And I'm not even half done describing one image. Out of three planned images.

I could just mention what's in the image and be done with it. But that wouldn't help #Blind or #VisuallyImpaired users because they wouldn't know what it looks like. And nobody would even know what it is.

Sat, 23 Sep 2023 18:12:11 +0200

Char

charvor@dragonscave.space

I may need half a day to describe three measly images; CW: long (almost 1,200 characters), alt-text meta, image description meta

Artikel ansehen

Sat, 23 Sep 2023 19:28:11 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Eight hours, 26,000 characters. I've still got six spaceships to describe, not to mention the nebula. And not to mention what'll be in the other two pictures.

Strike "half a day". I can be lucky if I can get this #ImageDescription done before midnight and the other two before Monday.

At least I probably won't have to research anything anymore.

Sat, 23 Sep 2023 19:28:24 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

@Char Um, for what?

Sat, 23 Sep 2023 19:28:40 +0200

volkris

volkris@qoto.org

@jupiter_rowland

Well, keep in mind that sometimes a blind person might simply want to know what’s in the image, and if the description is too long they might not want to wade through it to find out.

Sat, 23 Sep 2023 19:29:54 +0200

Char

charvor@dragonscave.space

@jupiter_rowland Image descriptions. :)

Sat, 23 Sep 2023 20:03:52 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

@volkris There isn't much I can do about it.

If I only mention what is in the images, then #Blind or #VisuallyImpaired users might complain that I don't tell thim what it looks like.

This would definitely matter even if the images were real-life photographs. But they won't be real-life photographs. They will be taken in a virtual 3-D world. A very obscure one. As in, absolutely nothing in these pictures will be familiar for anyone. Except for a few dozen out of over 13 million Fediverse users. Another reason why I'll have to describe and explain everything in an #ImageDescription.

So the first of these three images will have to get descriptions of

the walkway in the foreground (done)
a cubic background for a display (done) including three real-life nebulae (mentioned, but not described yet)
the ground that connects the walkway with the cube (done)
the teleporter in front of the display (done) including descriptions and transcriptions of all labels and buttons (done) and a description of the currently selected preview image (done)
a sign including transcription (done)
three different robots (done)
a black-hole-like object (done)
ten different fictional spaceships (four are done)
the surrounding ground, water and sky

The second image will have to get descriptions of

the walkway in the foreground (covered with the first image)
a cubic background for a display (covered with the first image) including three real-life nebulae (two covered with the first image, one new)
the ground that connects the walkway with the cube (covered with the first image)
the teleporter in front of the display including descriptions and transcriptions of all labels and buttons (only the top label and the four buttons at the bottom are covered with the first image unless I deliberately manipulate the teleporter to show the same as in the first image)
a sign including transcription (only the general look is covered with the first image)
four robots, one of them on a display stand, one in packaging (only one is partially covered with the first image)
an info kiosk
an armed drone
four spaceships
the surrounding ground, water and sky (will be covered with the first image)

The third image will have to get descriptions of

the walkway in the foreground (covered with the first image)
a cubic background for a display (covered with the other images)
the ground that connects the walkway with the cube (covered with the first image)
the teleporter in front of the display including descriptions and transcriptions of all labels and buttons (only the top label and the four buttons at the bottom are covered with the first image unless I deliberately manipulate the teleporter to show the same as in the first image)
two signs including transcription (only the general look of one of them is covered with the first image)
an elaborate space station with numerous writings to transcribe
a small asteroid with a used spaceship dealership on it, including seven spaceships (only two of them are partially covered with the first image), one robot (partly covered with the first two images) and three signs to transcribe
the surrounding ground, water and sky (will be covered with the first image)

Also, I have to mention where these pictures were taken. This is important within the context of the post.

They were taken in a place called Sulfur at an event called OpenSimFest 2023.

Would you know right off the bat what either of them is? No? See, that's why I have to explain that. Without this explanation, nobody would even understand what I'm writing about.

In order to explain it, I have to explain what OpenSim is.

Also, I have to explain why these exhibits matter, who created the items in them, the whole shtick, because it's all important to understand these pictures.

These initial explanations alone take up over 4,200 words.

This is also why I don't write three fully separate, stand-alone image descriptions: I'd have to add this 4,200+-word lead-in to each of them, causing massive redundancy.

This, in turn, is why I can't go gradually into more detail: I have to do the full 4,200+-word lead-in before I can even mention what is in each picture individually. Otherwise, I'd have to do a short lead-in, then describe what is in the first picture (the lead-in will only be part of the description of the first picture), then jump back to a general explanation for all three pictures, then jump back to the first picture. That'd be way too confusing.

Everything concerning all three pictures has to be explained before I can start writing anything about any individual picture.

Sat, 23 Sep 2023 22:26:15 +0200 zuletzt bearbeitet: Sat, 23 Sep 2023 22:26:14 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Over 13 hours in. Almost 34,000 characters for one #ImageDescription. And I'm still not done yet.

I've only got one spaceship left to describe, and this can be fairly short because that vessel is hard to "paint with words" due to its unusual design. And then there will be a few words about the surrounding scenery which I usually add last.

But I still haven't written the descriptions of the four nebulae, three of which I need for this image description. I can describe spaceships halfway decently, but I don't think I can do the same at the same level of detail with nebulae. And I absolutely need descriptions for them in order not to let non-sighted users down and come over as some ableist swine.

Maybe I should #AskFedi to do that or even resort to #Alt4Me. (The nebulae in question would be IC 434 including the Horsehead Nebula as well as NGC 2023 and the other surroundings, the Crab Nebula, NGC 604 and, irrelevant for this first picture, the Lagoon Nebula.)

Also, I might actually split the post I've planned in two. I currently estimate the whole thing including a triple image description to go well beyond 60,000 characters. I'm not even sure if that's still within the character limit of the hubs where I have my channel. It would definitely exceed 50,000 characters.

The downside would be that the double image description in the second post would have to use that in the first post as a reference. I'm not sure if and how this will work for #Mastodon users, especially those with a #ScreenReader. The alternative would be to put the whole 4,200-character non-image-specific lead-in into both #ImageDescriptions.

Sat, 23 Sep 2023 23:57:36 +0200

Cheryl Furse

CherylFurse@opensimsocial.com

@jupiter_rowland

You will break record of Stephen King to describe a room on 30 pages lol

Sun, 24 Sep 2023 00:53:17 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 00:53:16 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I think the first #ImageDescription is completed. It took me eleven hours to complete, and it measures 38,035 characters, now including the nebulae.

However, I've come to the realisation that I can't make this a thread. I'll need a "content warning" on the follow-up comment which would be the continuation. However, you can't add a summary (which is a content warning on #"Mastodon") to a comment on #Hubzilla. But without it, some mobile users will have a wall of text of tens of at least 25,000 characters in their federated timelines with nothing rolling it up or covering it up.

So I either have to drop everything into one post which won't be done until Sunday evening. I expect it to grow beyond 60,000 characters with all #ImageDescriptions in it. Downside: I don't know the character limit for posts on this hub or on the one where I have my clone. If it's 60,000 characters or lower, I'm doomed.

Or I have to make it two separate posts with the second one referencing the first one. The second one will therefore include a link to the first one. Downside: The link will most likely open in a Web browser. Also, I know for a fact that Hubzilla articles don't work in a #ScreenReader, so I guess neither do Hubzilla posts which renders the image descriptions largely moot.

Or I have to make it two separate, unrelated posts with no reference to the first one. Downside: The image descriptions in the second post would grow much, much longer because I'd have to repeat a whole lot of stuff from the first post.

Mastodon

Sun, 24 Sep 2023 10:42:43 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 10:42:41 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

So I'm going to write the #ImageDescriptions for the other pictures now. Maybe I'll be done by evening. And I hope the whole thing will fit into one post.

I've taken an interior detail shot, but I won't post it because it has eye contact in it, and there is no practical way for me to hide sensitive images from Mastodon users.

Sun, 24 Sep 2023 13:09:39 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 13:09:37 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I've decided to give some basic information in the #AltText of each picture, aside from explaining which #ImageDescription belongs to it.

That's because I'm going to hide the post itself behind a summary (that's a content warning for those of you on ActivityPub-based projects like Mastodon) which will mention how long the post is. And many will shy away from opening that and reading the monster post. But they will still want at least some information about the images.

I'm fully aware that this will cause redundancy. I know that this is, technically speaking, bad style. But it's the best I can do. And I won't actually describe or explain any of the things mentioned in the alt-text. If you need that, go for the monstrous #ImageDescriptions.

Sun, 24 Sep 2023 13:12:05 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 13:12:03 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

By the way: I've used about a dozen websites including Stellarium Net and over 2,300 characters to describe the Lagoon Nebula. But I'm done describing nebulae.

#ImageDescription

Sun, 24 Sep 2023 15:36:37 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 15:36:35 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I might be halfway through the second #ImageDescription. The robots are done, the kiosk is done. What's still missing is a police drone and four spaceships.

I'm at over 12,600 characters now. This might go well beyond 20,000 characters in the end, especially seeing as none of the ships is actually simple in any way.

And then I'll have to write the description for the third image. I'm afraid the final post will exceed 80,000 characters.

Sun, 24 Sep 2023 17:15:17 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 17:15:15 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

It will go well beyond 20,000 characters. I'm at over 17,700 now, and I've got three more ships to write about. I've just used 3,300 characters to describe a ship that doesn't even have an interior.

#ImageDescription

Sun, 24 Sep 2023 17:43:46 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 17:43:44 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

Make that 25,000 characters at least. I'm having to deal with four kinds of cargo boxes inside a ship, all of which have writing on them.

Writing means #Transcription is necessary. Even if you can't read the text. Especially then, actually. And a transcription would be non-sense if I didn't also describe what each box looks like.

Another ship that'll go way beyond 3,000 characters.

#ImageDescription

Sun, 24 Sep 2023 18:54:28 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 18:54:26 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

I think the second #ImageDescription is done. It stands at 26,196 characters now which might change with further edits. In the meantime, the first one has grown to 38,749 characters.

I wonder if this hub has a character limit. #AltText for both described pictures included, I'm over 65,000 characters already. Chances are good that the final character count will surpass 90,000.

Sun, 24 Sep 2023 20:38:16 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 20:38:13 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

The third #ImageDescription is looking good so far. Two transcribed signs, one fully described space station, no complex individual ships to write about. I might actually stay below 80,000 characters. But I still don't know if I'll be able to post it here.

Sun, 24 Sep 2023 22:04:38 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 22:04:35 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

9,694 characters for the third #ImageDescription, even though this could only be achieved by skipping the description of two of the used little ships altogether. But they're all rather odd.

Now I can get the images ready and see if I can post this monster.

Sun, 24 Sep 2023 22:34:58 +0200 zuletzt bearbeitet: Sun, 24 Sep 2023 22:34:55 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

It's out.

A miracle has happened: hub.netzgemeinde.eu has a character limit of not under 80,000 characters. I was actually able to post this giant. 76,734 characters raw and excluding the three alt-texts, 77,834 characters including BBcode and linked URLs. Not counting the 297-character summary/content warning, though. It took me the whole weekend to research and write this, including spending more than 24 hours in-world on the same sim.

The first #ImageDescription has 38,650 characters, the second one has 26,213 characters, the third one has only 9,687 characters. This has to be a #Fediverse record. I guess it won't be broken anytime soon, and if it will, I hope it won't be me who'll break it.

Still, I regret nothing. Maybe not putting some more work into the last image description because it might leave a few detail questions unanswered. Otherwise, nope.

Here's to hoping it makes it to Mastodon instances in one piece.

Mon, 25 Sep 2023 09:28:09 +0200 zuletzt bearbeitet: Mon, 25 Sep 2023 09:28:05 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

And I'm already dissatisfied with the descriptions. Not so much because I've left two of the small ships in the third picture entirely undescribed.

Not because I could have put more effort into describing the preview picture on the teleporter in more details either. I mean, in order to do that, I could have just gone where that preview picture was taken. It would have been a walk of just a few minutes or a teleport of a few seconds. And then I could have looked around.

But, for example, I could have mentioned whether a certain piece of transcribed text is legible for sighted users or illegible or so small that the text itself has become invisible. At least I didn't transcribe text of which not even a bit is within view.

I could have mentioned if an element that I'm describing is partially or fully obscured.

In fact, I could have mentioned if an element that I'm describing is entirely invisible from the camera point-of-view because it is, for example, hidden deep within a spaceship or on its far side.

I could have put more effort into describing the six containers on that one freighter. I could at least have tried to transcribe those graffiti that are visible.

See, I didn't describe these pictures from looking at the pictures themselves. I described them by spending hours in the very virtual place where I've taken them and moving the camera around until I was done.

One could say I got carried away, and I didn't pay attention to whether something is fully obscured from the camera or not. I mean, some people may be curious about details of the ships or the space station that are not visible from the camera point-of-view. But technically speaking, they do not really belong into an #ImageDescription. And #Blind or #VisuallyImpaired people may assume that everything I've describes is actually in plain sight.

At least I didn't describe anything fully outside the pictures.

Or maybe I could have explained what the several destinations shown on the teleporter are. Or given a full list of available destinations as part of the technical description and explanation of the teleporter, even if only ten are actually shown on its surface. And explained all these destinations.

You never know what people might want to know. And after this week's Saturday, I won't be able to go back and research things to answer their questions because the event will be over, and the location will be shut down.

Mon, 25 Sep 2023 15:42:48 +0200 zuletzt bearbeitet: Mon, 25 Sep 2023 15:52:50 +0200

Scott M. Stolz

scott@authorship.studio

If your goal was to give them a full experience and background, I think you succeeded. Your descriptions would also help sighted people understand what they are seeing in the pictures.

For example, some people might not even know what type of ships they are looking at, even if they had 20/20 vision.

To someone who is not familiar with what these ships are, the description they would come up with would be something along the lines of "A bunch of CGI generated space ships, but I don't know anything about them."

Your descriptions filled in details that even sighted people may not have noticed.

Wed, 27 Sep 2023 13:04:41 +0200 zuletzt bearbeitet: Wed, 27 Sep 2023 13:04:31 +0200

Jupiter Rowland

jupiter_rowland@hub.netzgemeinde.eu

My goal was to be informative towards people who don't know what they're looking at, but who are curious about this "new and exciting" medium/technology that's 3-D virtual worlds. I dare say that if there were 13,000,000 #Fediverse users, at least 12,999,980 wouldn't know what these ships are.

None of them are any standard designs. None of them come from any known established sci-fi franchise. All are custom designs by their creator except for one that mimicks a few 1930s Buck Rogers toys.

I could have given them separate shots of each ship at full professional-level D-SLR resolution. They still wouldn't know what they're looking at. Thus, I think the descriptions were indeed justified.

Wed, 27 Sep 2023 17:23:25 +0200

Scott M. Stolz

scott@authorship.studio

@Jupiter Rowland

I think the descriptions were indeed justified.

I think so too. Because it provides valuable information that is not conveyed in the picture, and everyone can benefit from it.

In my opinion, alt text generally can be shorter because it only needs to convey what is in the picture from the perspective of someone new to the subject(s) displayed. If the average person doesn't know what something is, it might be better as an additional description everyone can see, rather than in the alt text. And the W3C concurs.

Unfortunately, Mastodon's character limits get in the way of that. I am not sure if that is a solvable problem with Mastodon.