Thanks for the feedback. We do have scroll effects, but aren't intentionally trying to hijack the scroll behaviour. Are you able to provide some more info eg, browser/os - if you could provide any screenshots to me, that would be super useful: [email protected]
Thanks very much, yes it's optional, but validated - very confusing, we'll drop out the hard validation and make it a little more intuitive to enter a number
Huxley is a great tool for testing. I think there is a big problem to solve around visual regression testing (I also think visual "unit testing" could be an interesting concept).
Obvious.io is focused on website monitoring, if you think about New Relic monitoring your backend and perhaps Pingdom monitoring your webserver, then Obvious.io is about the customer - does your website look like it should for them? And does it look right consistently over time.
On a side note though, we do have an API in the works so technically if you wanted to plug obvious.io into your CI pipeline, you could do that quite easily. E.g. you want to check that your release into prod is visually the same as the one that's currently in UAT.
Thanks very much!
We started originally with monthly pricing but decided to experiment with per check pricing. I'd be keen for any feedback on what the subscription price should be and how many monthly checks should be included in that.
Thanks for the feedback around the scrolling - I assume this is on the content site? Can I ask what browser/os this occurs on?
Thanks for the feedback. This features us actually in testing right now and should be released shortly. The plan is to spider the site at a less frequent interval (say, once a day), and aggregate everything that's changes across all pages in a nice report. The hypothesis is that this would be useful for people with large content based sites and CMS's.
We're using a little bit of computer vision to visually compare websites over time. Just getting a little bit of media attention down here in Australia and globally via Twitter.
Our most interesting engineering problem at the moment is scale. We've home brewed a distribution engine to scale across multiple digitalocean regions without heavy weight messaging between them. There's no quorum to this 'cluster' if a check node dies then the remainder take on its load till another helper can come on board and help distribute the load once again.
We're working hard to build a solid product, please give it a try and let us know what you think!