He's just so convinced that he's right.

However, it may be good to CYA and have the company sign a waiver saying that you advised against following the programmer's advice, but that they agree with the him, so they will not hold you responsible if/when they get hit with the dupe content penalty.

It depends on how and what is duplicated if you ask me. The trick is figuring out which you have ;)

There are over 60 sites that will all have the exact same content for most of their pages with the exception of a few unique pages.

But in this case, when the SEs see 60 sites with the exact same content coming online I'd think something more than a simple duplicate content filter is bound to be tripped. You'd better convince the client to convince the programmer to do that bit of extra work.

Then let him sign an agreement making him liable for any duplicate content penalty (and resulting loss of revenue by the company) if he's wrong.

A good idea, but it would be impossible to measure loss of revenue or even if penalties were issued as a direct result of duplicate content.

Perhaps his hesitance is because he thinks he won't get any extra money from doing this work because the job was taking on a flat fee or something? Is he refusing to do it or are you just trying to see his side of it?

If he won't do it you will need to convince his boss that this is required for SEO and if they don't accept that then not much you can do but hope for a day when they come to you and you can "I told you so" them

I'd think something more than a simple duplicate content filter is bound to be tripped

Just to play devil's advocate: not necessarily. Lazy example: Wikipedia.

Not that I don't advocate fixing duplicate content: IMO it has more harmful effects than just damage to search engine performance, and quite often seems to arise due to programming oversights.

At the same time, some circumstances where duplicate content exists would take more time and resources to fix than potential benefits gained.

IMO it's good to try to keep the developer on-side too. You want them to make the changes you suggest, after all. They may even feel a little hurt and indignant at your suggestion: you're telling them how to do their job. Try to talk 'em round (real-world counter-examples can work wonders) ;)

Lazy example: Wikipedia.

Sixty sites with the "exact" same content -- yeah, assuming page structure is the same -- is bound to raise red flags somewhere along the line. Also assuming that since the same programmer is doing the work that the nav, thus URL structure is going to be identical -- that's just trouble. It was easily sussed out and penalized by the SEs years ago, I have no reason to believe they can't suss it out today.

but pages aren't exact dupes -- different headers, different nav, different footers...mix this around enough and you no longer have duplicate page

Personally I've found exact dupes to be less problematic on the whole (probably easier to interpret?), although I thought the OP meant 'the exact same content' in different templates.

It also depends on what the client is expecting: 60 sites all performing well for the same words and phrases is obviously out of the question ;)

If you are indeed using robots.txt to keep the spiders out of the duped content, then what's the problem?

or are you just trying to see his side of it?

60 sites all performing well

The programmer is pushing back saying that duplicate content isn't an issue the client should be concerned about

Have you advised the programmer that there is a duplicate-content specification that they need be concerned about?

Who is the programmer working for, BTW?

And why the need for a programmer in the first place, for a static site?

He wants them all to live in the root!

An awfully strange idea coming from a programmer. Must not be a very good or experienced one. Of course, don't tell that to your client, but be informed by it.

(OK, I re-read and answered my own question above - I guess the programmer works for the client)

Even back in the dinosaur days of the 70's, programmers were taught "structured programming", (since replaced with more impressive-sounding paradigms (twenty cents!) like Design Patterns...) which means simply breaking a problem up into more simple components, and making use of hierarchy. It's a fundamental Computer Science 101 kinda thing.

Most programmers would automatically extend this methodology to the layout of data - in this case, HTML pages.

So, I'm kinda surprised and shocked that a programmer would opt to "put all the HTML documents in the root".

It doesn't make sense on so many levels that it's scary.