Re: [gentoo-user] OT: extract an image from a .doc file?

[Stroller]

On 14 Dec 2009, at 14:43, Willie Wong wrote:
> ...
> (b) If the Big Wig is already happily letting the computer sign those
> documents for him, is it prohibitive to try the non-technological
> measure? E.g., ask the Big Wig to provide another image of his
> signature?

Oh, for sure.

I just didn't expect it to be this complicated. I expected to be able  
to open the document and pretty much to be able to click on the file  
to ascertain it's file size. I expected to be able to turn around  
quickly to the boss and say "you'd have saved all this file space if  
you used a 20kb version instead".

When I posted here I was kinda expecting someone to be able to suggest  
a 2- to 5-minute fix. I had no idea it would be this complicated, and  
now I'm mostly only interested because it has become an interesting  
problem.

> (c) If the image file is that big, it is probably because the
> original that got included in the doc file has a ridiculously high
> resolution (maybe they just scanned the signature in, cleaned it up a
> bit? My signature usually fits in a 1/2 inch by 2 inch block, if
> scanned at 24-bit color and 600 dpi, this makes almost a 1M raw
> image). I hope if the processing/storage/bandwidth tax is high
> enough, an "upstream" fix would not be ruled out directly.

Yeah, I think I have a copy of my signature here which was scanned at  
about that kinda resolution, stored as a bitmap & has a large  
filesize. When I discovered how badly it slowed down Word when  
actually trying to place it in a document it got replaced with a much  
smaller gif version. The improvement in performance that this  
eventuated was, to me, slightly unexpected - surely whatever the  
original format, both images must be stored in RAM in about the same  
way.

Stroller.