[thelist] In Defense of Fahrner Image Replacement

Egor Kloos studio at dutchcelt.nl
Sat Aug 9 07:37:45 CDT 2003

All this talk of FIR is starting to look like a flash mob.

> I admit I'm being very strict and stuffy and un-design-like, but I 
> came to the Web in 1998 when nobody had the faintest idea what to do 
> and most texts for websites had to be images "because browsers don't 
> support this font". In 1999 people relented, only all navigation 
> buttons had to be images and text was allowed to be text. It took 
> years for designers to understand that putting text in an image is 
> never a good idea.
I agree it took me a while to get my head around this, after having to 
correct or alter the text in a button for the tenth time it started 
making sense, well kinda. Most designers, especially the ones coming 
from print had to (re)learn how to work for the web. Leaving centuries 
of typography knowhow behind was and is a bitter pill to swallow.

> For page headers, well, OK...I suppose, every once in a great while. 
> But what's wrong with
Well I think this was one of the reasons for using the FIR 'solution' 
in the first place, Google doesn't read your images but will read the 
<H1> tag. This way you don't have to repeat the title and or 
description of the page to get Google to display this under the link in 
search results.
Using this for quotes would also be good for getting a search result. I 
couldn't see myself needing to use it that readily throughout a 
website, only for these exceptions. It may be superfluous of me to 
point out that it shouldn't be something you would design into a site. 
It's a coding trick  not design. It's a solution to a problem that 
'may' occur with a design.
My advise is to use the FIR solution only when it's really necessary, 
and as a result it might not be used at all.

> If you want to use an image it won't be 'semantic', period. If you 
> want to be 'semantic', don't use an image.
Well and image is an image, and it would seem to me, as Peter went on 
to say (correct me if I'm wrong), that this is semantic. No matter what 
W3C says, no matter how the image is used. I mean what else could an 
image be? XHTML 2 would like to see it as an <object> fair enough, but 
the argument still stands.
But a image is not text and it doesn't really need to try to be. 
Graphic design is not only about typography so all us crayon pushers 
should get a grip and design sites for our users and use the solutions 
needed to get the job done. This may include the a FIR like solution, 
but probably it won't.

> Agreed, it's not a hack. It's something to be avoided anyway.
You should never avoid a solution, even if it is this minor. Just like 
I would argue that one could use tables for layout purposes. God, aka 
Zeldman, might strike me down in furious anger for saying that. But for 
sites that are, for example, promotional in nature and have a shelf 
life of a couple of months it really doesn't matter. As long as your 
target audience can use it. It's that simple and your solutions, in the 
end, should also be that simple.

My apologies for being this long winded. I'm off to a flash mob.


DutchCelt Design
C. Egor Kloos


More information about the thelist mailing list