I've been a fan of Dan Appleman for about as long as I've been a professional programmer. He is one of my heroes. Unfortunately, Dan only blogs rarely, so I was heartened to see a spate of recent blog updates from him. One of the entries asks a question I've often wondered myself: can you really rent a coder?
but I’d be highly suspicious of any professional developer who can’t find a stable, long term relationship with a client eventually.
Rent A Coder is exactly how I found stable, long term clients. I havn’t been back in over 5 years now, havn’t needed to, but they certainly got the ball rolling for me.
It’s a great starting place, for both buyers and sellers.
How is this different than offline development experience ?
Surely you can demonstrate better value to your employer than the random, anonymous programmers on Elance, guru.com, RentACoder, or oDesk – right?
Only if my employer can display better judgment than sheer greed.
Unfortunately, Dan only blogs rarely… just like Jeff? 
@ejunker
That’s because Kevin (and digg PR) are prone to making up shit. A. It wasn’t done through elance, and B. They paid me a lot more than $200 (try 70 times more plus equity). The only hint of truth to it is that I did work on elance up to about a year before, and some of it for Kevin.
third
I used RentACoder for two personal projects of mine. Both were websites that I wanted to build, that I just didn’t have time to build myself (new baby, full time job). I must say for this, it worked great, and very affordable, but there were some caveats.
I had to be on top of the coders to handle edge cases and security vulnerabilities. I had to review the code myself and make sure things were done right. I have the ability to do that; I think your average manager type thinking of these sites as factories that you can hand a spec to and get code back is going to be quite disappointed.
But as basically a coding assistant, they work great. That’s my $.02.
Christo: we have to cut Jeff some slack for a while. There’s a new baby in the house.
Don’t forget that on a lot of these sites the services are being offered from all over the world (and a lot of them lower priced that mainland US). There are good developers available who can provide adequate services at a comparatively low price point. Of course, understanding of requirements and deliverables is a risk that must be managed. So for quick basic code cutting it is a viable alternative, but as always, caveat emptor.
I considered trying to do some work on a similar rent-a-coder site just for some basic experience while I was in college. I never ended up pursuing it however. I suspect a lot of people on these sites would be college students or similar. I also agree that the quality of work would be akin to a college assignment.
I used to use Rent a Coder frequently between roughly 2003-2006. It become more difficult to find a reasonable project to bid on, but my success rate was easily 75% or more.
You may be bidding against people from countries who can do it for a lot less money, but if you sell yourself to the buyer effectively (instead of just replying like some of the i can do this 4 $50 bids), and the buyer knows what they are doing, appreciates effort and good ethic, you can easily win with a bid 10x the average low bid.
I made a lot of money from it with this technique, but stopped using it after it was easier to find work elsewhere and an incident with a dodgy buyer who completely shafted me with vague specs. The project ended up going into arbitration and because of the vague specs I lost (but not without a lengthy fight) and gave me a rating of 3 which ruined my average from all 10s.
Would I use it again? Hell no. It’s tedious to find any decent work, and if you get into arbitration (where it seems RAC automatically favours the buyer), you’re in the shit.
To be fair, with Bugzilla I design features with people who I’ve never seen, who live thousands of miles away, several times a week.
However, I agree that it’s hard to imagine getting a quality product out of any of those rent-a-coder sites. Development is not like buying some commodity–if a developer costs more, he’s probably actually going to produce a better product (though not always, given cost-of-living differences in various nations).
When people tell me that they’re going to offshore development to some cheap company, I usually make an impassioned attempt to convince them to do otherwise for quality reasons. It’d be the same for the rent-a-coder things.
-Max
I may be wasting my typing skills here because perhaps I should read the 44 comments before mine, but a quick scan tells me to continue, so here goes.
I used Rentacoder to find someone who could build an interactive site for me. Before I turned to Rentacoder I looked at the portfolios of some people whose sites I liked, but design is one thing and functionality is another, and I was in no position to judge whether the sites that looked good were well coded.
And the functionality I wanted was an amalgam of things I had seen on various sites but not everything on one site, and there were some things I couldn’t see on any site I looked at.
So I saw no reason not to invite bids on Rentacoder, and every reason to give it chance, and I spent time asking the bidders follow-up questions to get a feel for their responsiveness, and for how well they had read the spec.
And there is a grading system in Rentacoder that acts as some kind of guide if you take the time to look at what the graders asked the coders to do and looked at how capable the graders themselves appeared to be.
So I chose someone who it seemed knew what he was talking about and responded like he wanted to do the work.
And it took longer than expected, and there were twists and turns along the way, but he remained good humoured, and so did I - and I tested and tested and tested (I never want to be a professional tester of website breakability) and after each change, I tested again. And I set up usability tests and we tweaked some more.
And now we are just about ready to launch, and my heart is in my mouth because I don’t know 100% for double certain sure that it won’t break under the strain of every combination of everything that people can throw at the site.
But I know that the man who did the job is very capable at fixing problems, and I don’t know that I would have been any better off taking a chance on a coder I had found some other way.
That’s my perspective.
I use RentaCoder all the time and have had around 80 succesfull projects, however you do need to be very cautious
1, I only use it for graphic design (as I can’t do that)
2, be really specific about what you want
3, when I give feedback to the designer I don’t do it just in words. I take screen shots and circle the bit I’m talking about to make my requirements really clear
4, I regularly choose 2 coders (i pay both) and I choose the pursue the one I like best.
5, If I find a good designer I keep giving them all my work and I pay them big bonuses, sometimes twice the bidding price. Hey they should get paid for good work
The only time a project went wrong was all my fault as I hadn’t been clear in my explenation. I still paid the coder the $600, threw the work away and did it my self (it wasn’t design)
I suspect I spent more time writing the spec and managing the programmers than I would if I had done the work myself
Interesting. If it encourages you to write specs, it might pay off in terms of quality the long run.
Your analysis is pretty much spot-on. On this subject, I have been quite extensively on the buyer side, both as an individual, as counsel to buyers, and for my employers. The recurring observation seems to be that the cost of conveying what is to be done can easily become overbearing: the initial specifications have to be written by someone with a technical background, with enough detail to avoid grey areas while spending less time on it than on actual development.
Then, there’s the question of bad programming. A non-technical buyer on any of these sites might get a package that breaks on the next browser update, or displays a vulnerability, or doesn’t scale past ten users, or doesn’t play nice with unicode. There are so many ways in which a piece of software can fail that simply cannot be imagined by a non-technical person and can manage to say hidden for weeks.
Yet, it works. Several of my employers have used some form of off-shore delegation when in-house technical people had a fairly good idea of what had to be done (but not the specialized knowledge of how to do it with a particular language/framework/server), those in-house people who had that specialized knowledge but were too busy could still pitch in warnings and general remarks, as well as quickly survey the result, and the off-shore developers had the manpower and skills to do what was expected. Of course, this required a fairly detailed specification, but this is what formal test cases are for.
I second (and extend) Daniel’s comments above … I used RAC to jumpstart my freelance business 2005-2007. Simply by being able to communicate in clear English, I found I could draft proposals that the buyers on the site often told me were miles better than the competition, and I could generally get approved on my going rate at the time (which admittedly was somewhat lower than I charge now).
Another trick is to ask lots of questions before bidding on anything … you’ll find out if your prospective buyer is the sort of person who appreciates a thoughtful approach, and will be responsive to any questions/issues that arise during the project. Many buyers won’t write back, or offer curt, unhelpful replies, and that’s a great sign to skip those projects!
I’ve now moved on, but the experience I got with RAC allowed me to build a portfolio and in a couple of cases led to direct referrals for new business.
I’ve worked with sites like this in two contexts:
The former almost always ends up being a disaster. We all know the key to working well on large projects is face-to-face collaboration. And sites like RentACoder just can’t cut it.
However, if you can distill the task down to a few sentences, a paragraph at most, I’ve found that option 2 above works out great. It’s just finding the right balance.
I’ve been poking fun at those ridiculous projects to clone eBay, Amazon, or YouTube for years. What are these people thinking? It cost millions of dollars to develop those sites and they expect to reproduce that for $500?