Anyway, the range is all over the map, because the needs/skills are all over the map.
Some dinky website with a couple PHP pre-packaged modules slapped in?
A custom CMS?
An Enterprise application?
Surely nobody in their right mind is going to expect to pay the same rate for these needs, yet PHP is suitable for all of them.

I personally ask a few simple questions about the client, and if I’m even interested in doing the work (90% of the time, not) then charge anywhere from $25/hour to $100/hour, depending on the work/client.

You can easily find rates if you dig around a few job postings—but they range from $30K to $90K, full-time in the U.S.

And that’s probably the way it should be, given the huge disparity of the actual task to be done.

Personally, though, as I said, most of the times the job doesn’t suit me long before we get to discussing rates anyway.

The reason those laws exist instead of letting the market “sort this out” as you say is because your model assumes efficient markets and perfect information. If a guild forms and the guild grants certification or some criteria for membership then it begins to form a trust, a known market defect. If information is imperfect then we have a classic signaling problem which is the source of approach alluded to by Rob—in effect, he are using a higher rate as a way of “signaling” to customers his separation from the underbidders…a workable, but not very efficient, solution. This is because it presents, by itself, no barrier to the “95% craptastic hobbyists” as Tiffany put it. My guess is Rob depends heavily on referral and references to enforce his separation.

@Ivo: Excellent point about the low end. I have a feeling the same is true relatively speaking on the high end, even though I agree with a lot of Marco’s comments.

Java/J2EE has a very good set of signals. For instance, I know that if the engineer works for, or comes recommended by ThoughtWorks (for instance) that I can expect a certain level of abstract programming ability and Java coding skill. Maybe Marcus could beat 90% of them with PHP and one hand tied behind his back. You have no way of knowing in the PHP world.

By this I mean someone writes PHP on their resume, maybe all they’ve done was download PhpMyAdmin and edited the config.inc.php or hooked up two mysql() calls so that they could build a recipe database. These people are not even lying on their resume, but…

Heck, getting a working J2EE installation up and being able to trace an error when it barfs is itself a great signal! tiffany calls it “big marketing budgets for use by big business,” I use the euphamism “signaling.” We pretty much mean the same thing.

@Marco: Yes, I completely overlooked the demand aspect of “enterprise” Java. My bad. I did mention it in my riff:

http://terrychay.com/blog/article/php-coders.shtml
and I alluded to it by calling demand for PHP “moderate.” but I was too stupid to read Ben’s “Leave a comment” instructions and the link didn’t appear.

As far as worth…I think it depends very much so on geographical location, experience, and portfolio (when looking at an individual). I wouldnt get $90K around here, neither would a JAVA programmer, so its somewhat relative to the location.

http://fuzzyblog.com/archives/2006/06/21/whats-a-php-programmer-worth/

Thanks for the great quote Terry; always appreciated.

PHP is widely available on $5 a month hosting plans. So there’s a much larger pool of programmers, even if 95% of them are craptastic hobbyists who worked their way into full-time gigs (like myself) as opposed to full-blown engineers.

Java (and .NET too for that matter) has two big things going for it: it was created by a big company with a big marketing budget for use by big business. So it has the respect of being a harder-to-learn, enterprise language—one that can command higher salaries.

The problem is with the number of people associate the LAMP stack with cheap costs, everyone talks about how much money they saved by going the LAMP way. What gets lost in these threads is that the cost savings come in a licensing perspective, good programmers still cost the same. You don’t have expensive middleware, you don’t have expensive application servers etc.

The main reason why the PHP freelance market is so crappy, is that it’s full of people with big eyes. A lot of the people looking for PHP freelancers, are themselves working on projects on the side. So, they can only afford something for $300.

I’ve personally stopped looking for side gigs, some come my way by word of mouth, and I’ll evaluate them, otherwise it’s just not worth my time to look for such gigs. I’m guaranteed to be outbid, and none of my reasons for a higher rate will stick with the customer. I can provide professional references from some fortune 500 companies, and the answer is “we don’t have the budget for that”, or “we’ll redesign the code once the site starts making money”.

Currently, I have a full time contracting gig with a company that recognizes my skillset, and utilizes it. If/When this contract is over, I’ll be looking for a similar company that will get my undivided attention, in return for a competitive rate. Side gigs are a great way to get experience, but they are never mentioned on the resume, and after a while aren’t worth the effort.

The real problem is the fact that the majority of potential customers considers Java ‘better’, ‘harder’, ‘more solid’ and therefore worth more money. The ones that go for PHP are often the ones on a tight budget. Combine this with the perception of PHP being ‘less valuable’ and PHP programmers end up getting paid a lot less than their Java collegues. I know some Java programmers that earn so much money that any PHP programmer I know could only dream of ever earning that rate. It’s sad, but it’s how things are at the moment.