Who should do rgb-CMYK conversion - designers or printers?

Question

I have been having a very interesting discussion on a previous thread in response to problems with the colour conversion from rgb to CMYK using InDesign and the resulting unsatisfactory colours in the final magazine delivered by my printers.
This has raised a number of issues and led me to further research. I would like to air these for a wider debate hence this post.
Perceptive readers will note from my spelling of colour, that I am English and indeed I work out of the UK. I am a historian, writer, photographer and editor. These days it is as a full time freelance but for a long time I was part time when I helped my partner publish a specialist sports magazine.
We started it a long time ago, in the days when you sent a typesetter galleys of type and photos and agreed on a layout. Our typesetters migrated to Pagemaker and we went with them. They and we were PC based and we still are, which is a bit of an anomaly in the design world.
Creating the text in Word was easy enough but the images remained a problem before digital cameras. We had a scanner but it was a flat bed scanner and created rgb images. The printers needed they said CMYK images then only available using cylindrical professional quality scanners. So there was a period when we paid the printers to scan the images for us, from the original photos, paying per print.
One of our small amateur publishing team was a well known professional photographer, so we started out with some high quality images but even so the scanned results, as they appeared in print, were patchy. I particularly remember one feature, covering a major international event where we had been supplied with high quality transparencies by a top class sports photographer and we duly passed on to the printers to scan. The results were clearly out of focus and the photographer was enraged and said he would not work for us again.
Of course we got proofs but they were low resolution in general and when we queried the quality of any image, we were invariably told that the proof did not reflect the final quality image as it would be printed. Trouble was it often did.
We ourselves continued to strip out costs from the magazine and eventually one of our team went on a series of Pagemaker courses and got some hands on experience working next to our former typesetter who was coming up for retirement. We took over the design and found a sympathetic new printer, down the road, also familiar with Pagemaker where we could pop over to look at the proofs, and get a second scan if the first was not OK. All went happily along for quite a few years. The sports magazine got sold and then I got a contract to edit a magazine dealing with the historic environment. By now digital photography had come along but we simply supplied hard copy prints (we use a lot of historical images) or digital images to the printers, with the Pagemaker files and they did the conversion. This routine stayed in place for a few more years and then the owner of the printers sold up and retired.
We found a new printers', a short run magazine specialist who agreed to accept our Pagemaker files (by now 6.5) although they were primarily Quark and Apple based. They did have a handful of PC clients and kept one PC for proofing their work. We continued to send the new printers Pagemaker files for printing and included the source images in case they needed to be redone. It quickly became apparent that the new printers worked down to a price rather than up to a quality and a fast turn around was the main aim. Conversations with their chief designer were perfunctory but because we had a lot of Pagemaker expertise we did not need a lot of support. We were under pressure to move to InDesign and it was probably time anyway.
Around two years ago matters can to a head with a font problem. This turned out to be a known issue with Pagemaker but resolving it caused problems with the printers, who said we had to upgrade to InDesign and it would go away. We were also assured that if we did so, and supplied them with a .pdf from InDesign, we could get a guaranteed result. We were told not to send them any more source images for comparison but to provide CMYK images. How we converted them was up to us.
Our designer went on several InDesign courses but they dealt with the design process and differences from Pagemaker.  Getting to a .pdf was the target, not balancing colour thereafter. That was considered the role of the printer and his workflow press press processes.
When we needed to export our first .pdf from InDesign, the chief designer called up our designer and talked him through it. The importance of compatible profiles and presets was never raised. We have never been given any written information by the printers on .pdf presets or profiles we might need to set up at our end. As explained in my earlier thread, we are now required to submit all our images already converted to CMYK and this has caused and continues to cause, problems with the colour balance, in general the final result has too much magenta in it. Matters came to a head with the latest issue.
We are in talks with the printers to resolve this but it seems the chief designer has now left and not been replaced. We have asked if they can do the colour conversion to CMYK for us and the answer has been a definitive "No". Approaches to other printers has met witha standard salesman's response that all we need to provide is a .pdf from InDesign and they will do the rest. All require CMYK images.
It seems to me in all this saga, that the printing community has been busy de-skilling and downsizing and putting responsibilities back onto the end user wwho may not be experienced enough in the area of prepress and colour, to understand the issues involved.
So the question then arises, why can't the printers do the rgb to CMYK conversion anymore? Our printers maintain they are not able to accept any work which requires them to do this. But reading around the subject it seems there are printers who are moving to handling the rgb conversions for their customers "late  binding rgb workflow" I think it is called. This is a link to a very interesting article about it in Print Media Management (I assume I can put links here). http://www.printmediamag.co.uk/technical-articles/205.aspx
What does anyone else think about this as a concept?

Answer 1

Just after the turn of the century I was working for a large-format output service. ID had just been released, but nobody used it, and Quark 4 was state of the art. Color management was a "new" concept to everyone I knew.
One of our vendors sold us paper with the promise that they could provide profiles for every stock to match our HP plotter (not that I think we would have necessarily, at that point, understood how to use them). They never managed to do that, and we never managed to match color from the same file on glossy and matte papers. Nor were we able to reliably match color that we saw on screen or in a customer's inkjet or laser proof without making test strips and using five or six times the paper in trial and error adjustments as it took to make a single poster the customer ordered. It was ridiculous, and expensive, and it convinced me that I needed to learn at least the basics of color management. Shortly after I left that job and went freelance.
I switched to InDesign partly because it actually handled color management better than Quark in those days (and may still -- I've never looked back and stopped buying upgrades at 6.5), and I read as much as I could (Real World Color Management from Peachpit Press is an EXCELLENT and easy to understand primer). I also found a local printer for offset where the prepress manager understands CM and profiling, and was willing to talk (we're now good friends), and I also found another large format service where the staff was CM literate. I use a colorimeter to calibrate and profile my monitor, I use the output profiles they recommend and the export settings they ask for, and what I see in print from these vendors "matches my screen as close as dammit," to use the OP's words.
The owner of the offset printery said something I found flattering, but a little shocking, the last time I went for a press check -- I supply the best color in files that need the least amount of adjustment on press of anyone they print for. My attitude is if the print is off at make-ready, it means my numbers are off, and that was my fault. I usually complain the reds aren't bright enough, then let the pressman convince me that if I tweak the reds too much I'll lose the blues and to have confidence that things will fall into place when the ink dries because his measure ink density numbers are where they should be. He's always right, by the way.
Is this shop unusual? Maybe a little for my neck of the woods, but there are plenty of them out there with the same dedication to keeping up with technology at the pre-press end, and the knowledge and experience with putting ink on paper and a commitment to quality and service, so you should be able to find someone almost anywhere. And size is not necessarily an indicator -- this printer is a mom-and-pop shop with two presses, a folder, and a half-dozen employees. Dad has been in the print business for about 40 years and knows only what's rubbed off about color management, but he knows about presses and ink and paper and what will work. The oldest son is the prepress tech, and he's a complete geek, and either one of them will talk to anyone, and wish more of their clients would take the time to ask questions about the printing process and file preparation, and how they can improve the quality of both what comes in and what goes back out.