I’m building this site as a side project, and I’m trying to find ways to speed up the development process. I’ve created github issues organized based on the bullets above.
Interview / Pay
I hate interviews that require hours of work on pointless programming tasks. So I’m going to do it like this:
I’ll look at whatever source you can show me. I’d prefer it be in Elixir/Phoenix, but if you don’t have that, then show me something else you’re proud of. Tell me what rate you expect.
If interested, I’ll schedule a call (hangouts/skype/etc). We’ll do a phone screen, and then talk about the project. (~30 min)
If that goes well, we’ll choose an ‘interview’ task that is one of the issues I’m working on. We’ll discuss that task so that you can estimate it. Then we’ll agree to a rate, max number of hours, and acceptance criteria (which WILL include automated tests). You will be paid for the hard part of the interview, provided you can deliver useable work. We can communicate via Slack during development to make sure you don’t get stuck.
Once you complete that issue, we’ll know whether we want to work together on the rest.
Word of warning, this is never entirely reliable. At best I get told that my IP is actually 200 miles away, at worst I get gibberish. I’ve yet to see a site that actually guesses where I really am. IP’s are not static in their location in the world and assuming they are is asking for a lot of hurt…
As an example, your site says I live in California… but I live in NM, which is 1,123 miles away according to Google…
If you want actual location guessing, use the HTML5 web location API for that.
Even if it seems to guess my location correctly (Hamburg, Germany) i do get a “Page not found”, it would be much nicer if you’d gave a little bit more text and give me the opportunity to choose some alternate locations which are either nearby or I might relocate to.
Depends on the browser, chrome uses information other than IP, and I’d imagine firefox would too but I’ve not tested it. Using Chrome with the web location API tests do report my correct location though. Regardless, IP geolocation is a failure and it quite bugs me when sites use them since it always seems to be exceedingly wrong.
Yeah, I can understand the frustration. My concern is that the user experience for asking a user for access to the location info isn’t very good with browsers. But it’s easy enough to add that I’ll get to it as soon as I’ve got some spare time. (Or rather, I’ll probably find someone from the board to put that in for me)
As for Chrome getting it right, I’m curious how they pull that off. I suspect it’s a combination of their own IP geolocation db, and perhaps some knowledge of you based on your logged in account? Interesting (and a bit scary!)
Either way, sorry to have bugged you with the bad location! Thanks for weighing in, as always!
Just for an alternate data point, it guessed my location correctly in sarasota, fl. So… percentages I guess.
I think the best experience is load the site with the best guess in location via ip… then have a ‘get the content near me automatically’ button or something to where I can click it, then get hit with that html5 popup that asks my permission. More steps, but I think it ultimately would increase acceptance rate.
Without all locations having content, though, it’s tricky. Wonder about comparing the user location with the locations where you have content and finding the closest match. Then saying - you’re here, and the content closest to you is here.
I’m logged in in chrome and they pull my location from my phone. Geolocating based on IP will never be reliable. When you browse through mobile site will get IP of mobile provider center usually located in capital.
And by showing “Internal server error” instead of a static page telling why there is no content for me, you just shoot yourself in the foot. ¯\_(ツ)_/¯ And if your target audience is your local market ie. US, you should write so, there’s no shame in this.