You should ask for title and years experience in the role. Without that a comparison is tough.
Obviously, how you structure the context defines what it means to you. If you share it with everyone from managers to newbie graduates your context will be different than if you just share it with people you consider to be at the same level as you.
You don't even need the app to discern between annual and dollar per hour, to the app it's just a number. Actually even an uniterpreted string would do.
Either allow configuration of currency, or a customised message or something like that.
I don't see this flying, where I work though. Even among the people I'm closest to, I don't think they'd take to this sort of information sharing amongst each other. Even pseudo-anonymously. I'm not sure if that's because they'd feel it's tacky or if it's because it has been drilled into everyone from corporate culture that you _never_ talk about it.
As an aside, I would actually much rather know not what the guys I've been working alongside for the past decade are making compared to me, but what new hires in the same positions as us are being hired at. Especially when you throw in all the hiring freezes and pay freezes over the last ten years.
I personally liked the other trick better: the one that gives you the average salary. Or maybe median is more appropriate. The point is that it'd give you an idea of whether you are making more or less than at least half your co-workers.
Instead, you could generate the histogram of the entire sample set, perhaps curve fit, so it's not obvious for small sample sizes, how many people contributed. Then you put a little red "You are here" dot on the histogram so they see how they compare. This robs the employer of any data they can use to see who partook in the experiment, while still giving the same information back to the employee.
The only thing that kind of bugs me with this app is that the amounts have to be expressed in USD, so i have to recalculate them back to my own country's currency when viewing the results, otherwise, nice app :).
The results are for your own use anyway -- it'll work just as well even if you don't bother to convert.
Of course, that'll skew whatever larger purpose the developer is hoping to have with the data, but if the project seems to have legs, that just provides a big incentive to add support for other currencies. :}
I like your implementation. To everyone asking about context, I believe this tool is intended to be used when you have already created such context, where coworkers want to know group salary information while still protecting their personal information from each other.
I had imagined the use would be where people were already in a group and wanted to share information anonymously. I had a brief moment before I fell asleep wondering how to prevent someone from just reloading the page as others submitted their salary information, thus being able to know what salary mapped to what user by the order they show up. I see that you set a minimum limit on the number of users in a pool, have you addressed this once the pool grows beyond that size?
Thing is, this salaryshare is a way to compare salaries with people that you already know, while Glassdoor is more of a general market analysis tool.
What I'm wondering now is how (and if) I can apply the idea of SalaryShare on Job4Dev. If you create a resume there, it already links the companies you listed as working for with the company database. Should I ask for "last compensation received", as well?
It sounds like an awfully rude thing to do.
The warning will confuse some people (such as yourself); that's why the dev asked if it was okay.
If you want to know what people are making, you don't ask them, you ask their employers.
There is a reason large HR firms poll employers and not employees, as I said.
It works by creating "salary pools". When you create a pool you get an unguessable URL (e.g, I created a pool at http://salaryshare.me/8a6e32dd8c9166905db6cfd588044bad). Share that URL with everyone that you want to participate, and after a minimum number of people have given in their salaries, the results are disclosed.
Apply any standard disclaimers that should apply for an application developed in less than 6 hours.