3 ms·

I've started working on exactly this problem. I don't think that sticking everything to a specific vendor is the right solution, and hoping that everybody is go

I've started working on exactly this problem. I don't think that sticking everything to a specific vendor is the right solution, and hoping that everybody is going to adopt the same standard is just not going to happen. I'm building a simple open source API that will connect all your devices together regardless of whether they have a fancy API or not. [1] I just started working on it so it's definitely far from done, but I think that this might be a step in the right direction. [1] http://github.com/hjem/hjem http://github.com/hjem/hjem

wpietri 11y ago
It always concerns me when a programmer reads a customer problem and offers a technical solution. It could be that an open-source API really will solve the problems he mentioned, but if you want adoption, you'll need to get better at explaining how they customer's life is going to get better. If I say, "X is too much hassle" and someone replies with "Good news, I put a bunch of PHP up on Github", I'm not seeing less hassle. I'm seeing more hassle. Now I don't have to just worry about a half-dozen vendors, their possibly-janky products, and their willingness to spend 30 years supporting something that fell out of fashion. Now I can also worry about trying to glue all that together with somebody's hobby code that, if the average open source project is any guide, will be swiftly abandoned.
- bitdivision 11y ago
  Well I don't think OP was trying to help everyone immediately, but he is at least making a start. The beta API someone threw up on github obviously isn't going to help the average consumer, but it might be the start of a community of enthusiasts. Maybe one day those enthusiasts build a robust system with wider adoption and a company behind it. If you want something that 'just works', then github is probably not the first place you should look.
  - wpietri 11y ago
    > If you want something that 'just works', then github is probably not the first place you should look. Yes, and that's my point. The expressed problem was that somebody renovating a home looked at a bunch of the options and said that they were too much hassle, too much risk. When somebody says, "This technology doesn't look like something I want to deal with for 20 years," it seems to miss the point to add, "Hey, here's some source code to something even less likely to last 20 years." I'm not saying it's bad to try write code for fun. I too have a GitHub repository with some dubious home automation code. [1] But as a developer, when a user talks about a problem, I think it's vital to truly listen to the problem and then only talk about a solution if it actually solves something for them. [1] Basically, redshift for my house: https://github.com/wpietri/sunrise https://github.com/wpietri/sunrise

wpietri 11y ago

It always concerns me when a programmer reads a customer problem and offers a technical solution. It could be that an open-source API really will solve the problems he mentioned, but if you want adoption, you'll need to get better at explaining how they customer's life is going to get better. If I say, "X is too much hassle" and someone replies with "Good news, I put a bunch of PHP up on Github", I'm not seeing less hassle. I'm seeing more hassle. Now I don't have to just worry about a half-dozen vendors, their possibly-janky products, and their willingness to spend 30 years supporting something that fell out of fashion. Now I can also worry about trying to glue all that together with somebody's hobby code that, if the average open source project is any guide, will be swiftly abandoned.

bitdivision 11y ago
Well I don't think OP was trying to help everyone immediately, but he is at least making a start. The beta API someone threw up on github obviously isn't going to help the average consumer, but it might be the start of a community of enthusiasts. Maybe one day those enthusiasts build a robust system with wider adoption and a company behind it. If you want something that 'just works', then github is probably not the first place you should look.
- wpietri 11y ago
  > If you want something that 'just works', then github is probably not the first place you should look. Yes, and that's my point. The expressed problem was that somebody renovating a home looked at a bunch of the options and said that they were too much hassle, too much risk. When somebody says, "This technology doesn't look like something I want to deal with for 20 years," it seems to miss the point to add, "Hey, here's some source code to something even less likely to last 20 years." I'm not saying it's bad to try write code for fun. I too have a GitHub repository with some dubious home automation code. [1] But as a developer, when a user talks about a problem, I think it's vital to truly listen to the problem and then only talk about a solution if it actually solves something for them. [1] Basically, redshift for my house: https://github.com/wpietri/sunrise https://github.com/wpietri/sunrise