Good points... well it is all about good sense.

Only the last point is possibly raising some very important questions and maybe deserves more discussion within the scientific community.

For example, many statistical softwares have graphic standards that are well below those ones of most modern scientific journals!

As also noticed by the author, digital imaging softwares (e.g. Adobe Photoshop) must be utilised in order to obtain clearer and higher resolution images. In several cases, rewriting of the automatically generated dots, axes, numbers etc. might be necessary in order to improve quality.

In fact, in suspect cases many high IF journals (e.g. Science) request that a copy of the original image is sent to the editorial board prior to publication.

However, I feel that this is still an extremely sensitive and potentially dangerous point: the borderline between a clearer perception of the information conveyed by the image and a biased highlight of the information one would like to transmit is clearly fuzzy.

Wouldn't it be a good idea to submit both versions of each image (before and after the editing) as a good practice? Well of course this would considerably slow down the reviewing process...

What do you think?

Best,
Gianluca