Does there exist a set of data that measures internet traffic that is specific to image downloading and uploading?

Question

I am currently writing a report, I would like to make a claim which I am struggling to back up:

The volume of image data transferred digitally has increased as the internet has become more popular and fast download speeds has become more easily accessible.

Of course I don't actually know this claim to be true. However, I think it's a reasonable claim given that the number of gadgets people own seems to be increasing (at least where I am from), as does the global population and there are developing countries working on high speed internet infrastructure.

Does anybody know of any data that could help me back up my claim? For example data where volume of image data being transferred is measured over at least, lets say 10 years would be great!

Answer 1

I think you're unlikely to find this as true "open data," since there are substantial privacy concerns to internet traffic monitoring, and the volume of data involved to do proper analysis is considerable.

That said, there is an organization, the Center for Applied Internet Data Analysis (CAIDA), which makes datasets available to researchers under oversight. A brief look at their data overview doesn't reveal any sets which are obviously content-oriented, but they may be able to advise.

Beyond getting the data yourself, you may be able to find a citation in media or academic publishing which substantiates your hypothesis. This GigaOm story, "How the core of the internet has changed from data to content" might lead you down the right path.

Ultimately, the amount of still image data looks like it will be a drop in the bucket, if the Cisco study covered in this Recode article is correct: they claim that video traffic is currently 78% of internet traffic, and headed for 84% by 2018. If you still want to get numbers about image data, perhaps digging deeper into the Cisco study will yield results.

Answer 2

(sorry, this was getting too long for a comment ... I know it doesn't directly answer the question)

My gut says that it's a safe statement to make (after having been a web server administrator for the last ~20 years) ... but you have to consider a few things :

• We can claim an increase just due to the increase in the number of people using the internet, even if everything else is held constant.
• We can claim an increase due to the larger file sizes, even if the total number of images being transferred was constant or declining.
• What is an image?
  
  • Is an animated GIF or similar an image, a type of movie, or something else entirely?
  • Are movies a series of images plus an audio track, or are they something that's not an image?
  • Are PDFs always images, never images, sometimes images (eg, if the PDF is of a scanned document), or fractionally images (the proportion of the contents that were inserted as images)?
  • Do the PDF rules hold true for other document types (MS Word, MS PowerPoint, Adobe Illustrator, Adobe InDesign, etc.)
  • Does line art qualify as images, or only bitmaps/rasters?
  • Are all readings from a CCD array an image? (eg, telescope images vs. spectrometer readings) Or only when we place it into a format that a typical person on the Internet can use it?

I'm likely being pedantic here, but it's likely that someone could just as easily show that images are decreasing if they go with a really narrow definition of 'image', and look at proportional volume per person, and show that the increase in movies, streaming audio and moving large software distribution to the internet has led to a decrease in the share of bandwidth consumed by images.