Lesson video

Hello, my name's Mr. Davidson, and you've made a great choice to learn about metadata with me today.

Today's lesson is called metadata from the unit, Representation of text, images, and sound.

In today's lesson, you are going to be able to describe what metadata is and how it's used to provide data about images and sound.

Today we have one keyword, which is metadata, and that's data used to describe other data.

We have three learning cycles today.

Let's start with the first, describe metadata.

We can collect data for anything that we can describe in the real world.

Now, consider this example.

A pair of trainers for sale in the shop.

What data could we capture to describe those pair of trainers? There are so many possibilities, and much more than the ones that I've listed here.

But with our pair of trainers for sale in a shop, we might need to know about the size of those trainers, so a number value that describes them in terms of the size for the actual trainers themselves, the colour that the trainers are, knowing that that colour might change for different pairs of trainers.

There might be a brand for the trainers.

So the company that makes those particular trainers, we'd need to record who that was.

And with them being sold in a shop, we're going to need to know how much we're going to charge for them, so the price could be data about those trainers as well.

It could be that part of a bigger collection, so that could also be some data that we store about those trainers.

We have so much flexibility in terms of the data that we can store.

We firstly have to define what that is and what values the data can take.

So let's think about how that example applies to computer science.

Izzy has created a word processing file on her computer.

Like when we described the trainers previously, the file can be described in various different ways as well.

This description and properties of the file are also data.

Note that this is in addition to the data of the file itself.

In some operating systems, you can right-click and inspect the data that's held about a file.

In this case, I can see lots of data that's been recorded all about the file itself.

Let's think about this for a moment.

Which of the following could be data used to describe a property of a presentation file? Could be bit depth, the number of slides, or the number of pages? What do you think? Correct.

For a presentation file, it's likely that the number of slides will be data recorded to describe that presentation file itself.

This data that we use to describe other data is referred to as metadata.

This is in addition to content data, which is the original artefact that's been created.

So when we create a text file, or an audio recording, or an image, that's stored as a binary sequence and processed as per any other data.

Metadata is then added to that content data to provide extra information about the properties of the original artefact.

It's extra data on top of the actual source that we need and use.

Metadata is really useful.

If we record enough data about our data, it makes it easier to catalogue it, and as a result, makes it easier to find.

Consider that when we search for things, most searching is performed using text.

So if we have textual data about images, audio, and video, it's going to aid the search process.

It's really helpful as well when we've got data that doesn't include text to provide that textual information on top.

Let's check you've understood that.

Fill in the gaps for these sentences.

Well done.

The first sentence should read, "Metadata is data about data." And for the second sentence, "The metadata of an artefact describes the properties of that particular item of data." Let's get on to task A.

Firstly, what I want you to do is to describe what metadata is.

Once you've done that, the second part I want you to consider Sam's problem.

Yesterday, Sam created a spreadsheet to record and then make a graph for a science experiment.

Unfortunately, Sam now can't remember where the file was saved.

I want you to explain why the last modified metadata of the file might help Sam locate it.

Pause the video and have a go at those tasks now.

Well done.

There were some really good answers there.

Let's check it against what I put.

Firstly, you had to describe what metadata is.

Remember that we said that metadata is data about data.

We use metadata to describe the properties of some content data.

Ultimately, metadata is textual data that is used to catalogue data.

In part 2, Sam couldn't find the file he saved.

Fortunately, the last modified metadata of the file does help Sam locate where that is.

Sam can search for the file using the metadata that is included.

In this case, the last modified metadata will have been updated when the file was last saved, and Sam will be able to search for that date to find the file.

Without that metadata, he wouldn't be able to search using the date at all.

So we can see how useful metadata can be.

Let's get onto the second part of today's lesson where we're going to calculate the metadata of images.

Firstly, we're going to take a specific example of bitmap images.

Metadata for bitmap images identifies fundamental image properties.

The two main properties of bitmap images are the image's resolution and its colour depth.

It's very likely that we'll have other metadata stored about the image itself, so perhaps physical measurements the image should be or a title for the image that would be displayed when the image is opened.

It's also likely as well that if the image was taken using a camera, extra metadata about the camera itself might be included.

Again, if we were able to inspect that metadata, we might have lots of other metadata that would be included because it was taken using the camera.

It might record the model of the camera, any technical data such as aperture measurement, there may be some GPS coordinates stored with it to allow us to know which location the image was taken in.

There may be some physical measurements.

So in this case, how close the item being photographed was to the camera.

And perhaps any zoom value that was recorded.

Consider that whilst metadata can be recorded about lots of different things, it's not necessarily going to have the same metadata as other files.

So in this question, I want you to consider which of the following is not likely to be metadata for a bitmap image? GPS coordinates, sample rate, or zoom? What do you think? That's correct.

Sample rate is related to audio recording, so it's not likely to be metadata recorded as part of a bitmap image.

Let's consider then that metadata is additional data along with the content data.

It's therefore going to affect the overall file size of an image.

In the case of a bitmap image, we would have our original image data, and the metadata would be stored in addition to the original image data.

Overall, the file size would be increased because of the metadata.

Metadata isn't fixed though, so the amount of metadata included will vary and we can only approximate it.

It may well be when we're calculating file sizes that we have to include a percentage.

Because we don't know how much that is going to be, a 10% increase will account for any metadata that we'd store in the file on top of the original image data.

Let's consider that as an example.

The file size of an image would still be calculated using the image properties, so our resolution and colour depth.

In this case, our resolution is 600-by-400 pixels and our colour depth is 16 bits.

Just the content data of the image would be the multiplication of those three factors, 600 multiplied by 400 multiplied by 16.

The file size of the content data would therefore be 3,840,000 bits.

If in our calculation we need to consider how much metadata would be stored as well, we would increase the content data by 10%, so we would multiply our previous answer by 1.

10, giving us a file size of the content data and the metadata of the image together as 4,224,000 bits.

So let's just check you remember.

What is the typical percentage increase of an image file when metadata is included? Is it 2%, 5%, or 10%? That's correct.

Typically, for metadata, we say that the metadata will be an extra 10% on top of the content data.

You are gonna have a go now at calculating the metadata of images on top of the original file size.

Remember that the size of an image file is calculated with the formula, resolution multiplied by colour depth.

And colour depth is expressed as bits.

What I want you to do is calculate the file size of the four examples below.

I want you though to include an extra 10% of metadata on top of that file size.

Pause the video and have a go.

We'll check your answers in a minute.

Well done.

There was lots of careful calculation you had to perform there.

For the first example, we had an image with a resolution of 1,000-by-600 pixels and a colour depth of four bits.

Multiplying those values together gave us 2,400,000 bits, and with our extra 10% on top, that takes it up to 2,640,000 bits.

For the second scenario, an image with a resolution of 3,000-by-1,000 pixels.

And a colour depth of 10 bits gives us originally 30 million bits.

But with the extra 10% on top, it gives us 33 million bits.

In third example, an image with a pixel count of 10 megapixels and a colour depth of eight bits gives us 80 million bits.

Increasing that by 10%, therefore, gives us 88 million bits.

And finally, an image with a pixel count of eight megapixels and a colour depth of five bits gives us content data of 40 million bits with the extra 10% on top.

That's 44 million bits.

Let's get onto the last part of the lesson today where we're going to describe how sound file metadata is used.

Let's consider a different example for our metadata this time.

we're going to use audio files and consider what metadata is going to be recorded for audio files.

Metadata for audio is likely to identify fundamental properties of digitised sound.

So we'll definitely be recording the sample rate, the bit depth, and the duration as a part of the metadata.

There's likely to be additional metadata that's recorded about that content data.

So in the case of audio, other metadata could include the year the track was recorded or perhaps who the copyright holder is.

If we think of music tracks and the audio has been recorded as part of a music release, extra metadata about the media collection might also be included.

Again, we could inspect the properties of that metadata using our operating system.

In this case, the track that has been recorded includes the artist's name, any album that that track is associated with.

And if it is on an album, there's gonna be a position on the album, so the track number.

It could record the genre of music that that track is.

And perhaps, it might have to record the age rating for a track.

So we've seen some examples of what could be included.

But for this question, I just want to check which of the following is not likely to be metadata for a sound file? Could it be duration, sample rate, or colour depth? That's correct.

Colour depth is likely to be metadata for an image, but not metadata for a sound file.

We saw before that metadata is useful for searching for certain properties, but metadata is also useful to provide information that software using the data can draw upon.

For our example of our audio track, the metadata can be used to display information about the track as it's playing.

We might see the track being played in a music player, and there's often visual elements that illustrate what the track is about.

In this case, the music player has picked up the metadata from the file.

It knows what the track number is, it knows what the album is, and it knows who the artist is as well.

You are gonna consider this in a bit more detail now.

For our last task, what I want you to consider is that some media player software has been designed to show an image of the artist playing.

The images are stored in a database that's available freely on a website.

What I want you to do is describe how the software could use the metadata to display an image for a particular artist.

For the second part of task C.

In the same software, a parent can restrict which songs can be played based on the age rating.

I want you to explain why metadata is useful in this case.

I want you to pause the video and have a go at those two tasks.

Think carefully about how useful metadata is for various different types of content.

Well done.

You worked really well with that.

In the first part of the task, we had to describe how the software could use the metadata to display an image for a particular artist.

So firstly, think about that when the player loads the track, it could inspect the metadata.

It would actually be able to read all the metadata associated with that file.

It could then find the artist's name provided in the file.

And the app could be programmed to search the database for the artist and load the image matching that artist's name.

It may well be that the metadata wasn't intended for that purpose, but by providing it, we can provide extra functionality and extra usage as a part of that file.

For the second part, in the same software, a parent could restrict which songs could be played based on the age rating.

You had to explain why metadata was useful in this case.

Remember that metadata can include many different things, and one of those might have been an appropriate age category to identify which age the audio is suitable for.

The software could be programmed to check that metadata against a setting in the audio software.

If it detects that a track doesn't match the age rating that it's restricted to, it could block those tracks and stop 'em from playing in the media player.

Well done.

You worked really well today.

Let's just recap what we've learned.

We saw that metadata is data that describes the features of other data.

And we found out that metadata provides extra information about the properties of an original artefact, and that metadata is expressed as textural data that can be searched.

That was really useful as well when we have image and sound files, because the metadata there provides a range of information that can be used in a variety of ways.