LinkedIn instituted a while back a “skills” feature, in which others can endorse you for a skill, and they end up ranked on your profile by how many endorsements there are. These are based on skills you add yourself, if any, or things plucked out of thin air your profile, and they are forced upon suggested to your connections when they visit LinkedIn. This gets a little weird, because people will endorse for a skill whether they know it to be applicable or not. For instance, I keep getting endorsed for Visio. Now, I have played with at least two versions of Visio, ever so slightly, but I never had a big reason to use it for real. I may have even helped someone else figure out a problem with it. I’ve done that many times over the years; helped with an unfamiliar software or such, even sight unseen. That just means I’m That Good at extrapolating, problem-solving, and guiding others. It doesn’t suggest any special skill at Visio, or whichever other thing it might be.
I try to endorse people only for what I am sure they know, having seen how odd it gets. I’ve found lately that there’s an alphabet soup of “skills” that I just don’t recognize at all, either because they are new or because they are nothing I’ve encountered.
Today I signed up on Dice. This was more challenging than it ought have been, overall, but at least it’s a general jobs site and not onerous merely to apply to, or register to be allowed to apply to, one company. Dice pulled the skills thing, too, rendering the signup and profile creation vastly longer than it would have been (only to crash because in the background it logged me off before I could save). The twenty skills it listed were pulled with no particular logic from the text of my resume. At least, I think they all were sourced, not fictional.
About half of those were nothing you might want to tout among twenty top skills. Even the ones that seemed reasonable to keep needed the last year used and length of experience set. Most of them assumed a most recent use of 2013. One assumed 1994 because of how it picked it up off the resume, but the reality would be 2007. I must say I was a little disappointed that the longest duration of experience I could select was >10 years, since I had ones that were 25, and another that was 21. On the other hand, it was gratifying that most were over 10 years. Through this, I felt vaguely uncomfortable with the list, and whether it was lacking, or misrepresenting me, even once fixed.
It also left me wondering just what I ought to be doing to modify my resume to make it optimally computer-parsable, so the modern silcon gatekeepers would better represent me and allow me into human hands.
I’m thinking I need to start by making a skills list. Not just ones that can be gleaned from my list-friendly experience posts and my resume, but also skills that aren’t software, hardware or technical. That is touched on in the LinkedIn skills, but arguably needs some analysis and refinement. Between the need for resumes to be parsed by machine, and the fact I am sure to encounter the skills list factor again, it seems a priority.