Any compressed format is unreadable unless you know how to read it. So with a JPEG image, we’re used to looking at the decompressed image (which is a field of color, same format as any image, compressed or otherwise). That’s “the JPEG” to us. But if you think about it, the JPEG is “actually” the binary file which our computer knows how to display as an image.
With this in mind, the analogy to lossy compression becomes apt once again. The cryptic sentence “Duncan is a teacher and a writer” is like the compressed file, in that it does not resemble the original thing being described, but it does generate a good image if you know how to unpack it.
However, I don’t think just saying “lossy compression” will convey this to someone else automatically, so I don’t think it’s a great argument for cutting “sazen” and just saying “lossy compression”.
In fact, I think there are other important disanalogies.
If I say “Duncan is X feet Y inches tall” (fill in your actual measurements), this is a “very lossy compression” in that it chooses to preserve a very narrow range of information about Duncan, but it’s not Sazen.
A lossy compression in the computer sense is making an earnest attempt to preserve the entire image, sound, video, etc. However, humans do this other thing, where we convey partial information, but in a way that’s not necessarily vague or context-heavy at all. EG, precise scientific measurements (like height, width, temperature). “Lossy” doesn’t provide a very good abstraction for what’s going on here—measurements are “lossy compression” in a sense, but the description doesn’t seem very useful to capture what’s going on there.
JPEG has a standardized compression scheme. Many images use this scheme, and the same tool decompresses all of them. The tool does not require prior knowledge of what any of the images look like when decompressed.
Unpacking a sazen does not rely on knowledge of a particular compression scheme, it relies on knowledge of the referent. This seems pretty different to me.
Furthermore, the fact that you can’t read JPEGs without the right tool doesn’t seem to me like it has much to do with the compression. Raw, uncompressed bitmap files are also pretty unreadable to humans without the appropriate tools. And you can achieve lossy compression of a picture without changing its format, e.g. by reducing its resolution.
Re: lossy compression.
Any compressed format is unreadable unless you know how to read it. So with a JPEG image, we’re used to looking at the decompressed image (which is a field of color, same format as any image, compressed or otherwise). That’s “the JPEG” to us. But if you think about it, the JPEG is “actually” the binary file which our computer knows how to display as an image.
With this in mind, the analogy to lossy compression becomes apt once again. The cryptic sentence “Duncan is a teacher and a writer” is like the compressed file, in that it does not resemble the original thing being described, but it does generate a good image if you know how to unpack it.
However, I don’t think just saying “lossy compression” will convey this to someone else automatically, so I don’t think it’s a great argument for cutting “sazen” and just saying “lossy compression”.
In fact, I think there are other important disanalogies.
If I say “Duncan is X feet Y inches tall” (fill in your actual measurements), this is a “very lossy compression” in that it chooses to preserve a very narrow range of information about Duncan, but it’s not Sazen.
A lossy compression in the computer sense is making an earnest attempt to preserve the entire image, sound, video, etc. However, humans do this other thing, where we convey partial information, but in a way that’s not necessarily vague or context-heavy at all. EG, precise scientific measurements (like height, width, temperature). “Lossy” doesn’t provide a very good abstraction for what’s going on here—measurements are “lossy compression” in a sense, but the description doesn’t seem very useful to capture what’s going on there.
JPEG has a standardized compression scheme. Many images use this scheme, and the same tool decompresses all of them. The tool does not require prior knowledge of what any of the images look like when decompressed.
Unpacking a sazen does not rely on knowledge of a particular compression scheme, it relies on knowledge of the referent. This seems pretty different to me.
Furthermore, the fact that you can’t read JPEGs without the right tool doesn’t seem to me like it has much to do with the compression. Raw, uncompressed bitmap files are also pretty unreadable to humans without the appropriate tools. And you can achieve lossy compression of a picture without changing its format, e.g. by reducing its resolution.
Yep, makes sense to me.