3 ms·
You are attempting to create an authority on a programmer's reputation/clout. A good way to spread this, and a feature I'd like to see, would be a simple API th
by mikegreenberg 15y ago
You are attempting to create an authority on a programmer's reputation/clout. A good way to spread this, and a feature I'd like to see, would be a simple API that gets the key stats (maybe not all) from your service. Maybe something that could generate a badge the programmer could include on their site. You're doing the work of pulling in data from SO, Github, Bitbucket, but don't offer much back to the user (aside from some self-affirmations that they can only access on your site). Exposing this data for others to use in interesting ways might be a good way to generate the authority you're attempting to create. (Not to mention the instant social-proof of programmers advertising their generated score on other sites.)
That said, I don't personally find that these services give very accurate depictions of a user's value. An arbitrary score means little without a control or reference point. (What would Linus Torvalds score?) Further, services like these typically rely on some secret-sauce to generate their numbers... and obfuscating this process lends itself to further skepticism of that value of your score. This is not an attack on your product specifically. I feel this equally about Klout, which I've ranted about elsewhere.
The problem services like yours should attempt to solve (IMHO) is indicating what qualities your score represents. This needs to be as clear and simple as possible. The best utility your service could offer will be in the following use case: A stranger, who is completely unaware of your scoring system, requests and consumes a report you generated for a user quickly; and to walk away with a strong sense of how that programmer relates to the metrics they care most about. Agreed, this is no small feat. But it's what services like BBB, Consumer Reports, and the like have spent years establishing their authority and clarifying their signals.
Hope this was food for thought. (Take it with a grain of salt. What the hell do I know anyway.)