Announcement

Collapse
No announcement yet.

My 2 cents - jstuyts

Collapse
X
 
  • Filter
  • Time
  • Show
Clear All
new posts

    My 2 cents - jstuyts

    Note: I have not published a plug-in yet (and unfortunately won't be able to for a while because I am working on our new house), but I am not only a developer but also a paying customer.

    Instead of replying to each topic separately, I will give my opinions here. I share a lot of opinions with others, but I wanted to state them anyway so HomeSeer has a better insight into what their customers and developers are thinking.

    Choosing HomeSeer

    So, why did I decide to pay (a lot) for HomeSeer? It is because of 1 or more of the following reasons:
    • The quality of open-source and/or free systems was disappointing:
      • Heavy-weight
      • Slow
      • Unstable
      • Crappy user interface
    • Most functions of a specific Z-Wave device I have were unavailable or needed cumbersome configuration, while HomeSeer supports it completely out of the box.
    • I do not have the need to tinker. I want a mature, stable system. It has to run our house. If something does not work (because of a restart, update, etc.), I will get complaints. (If you have customers instead of family, I expect them to complain harder when things they paid for don't work.)
    But I also chose to buy devices that are not tied to HomeSeer: Z-Wave and relay boards, so I have the freedom to switch to another system (with a lot of effort) if I think my time and money is better spent elsewhere.

    Developer license

    I agree with most people here that having to pay for a developer license is a big hurdle. Especially because HomeSeer also takes 30% of your revenue. Building a plug-in is time-consuming, and the chances of getting a reasonable amount of money for your effort are (as far as I can tell) slim.

    I absolutely do not mind that HomeSeer gets 30%. Building and running a store, and supporting developers cost time and resources. So It is normal for them to want to have some of that cost covered. But if a lot of money is currently being made by plug-in sales, I think it would be better to lower their cut so their users can enjoy cheaper plug-ins. HomeSeer already is a pretty tough proposition: buy this advanced home automation system for hundreds of dollars, and then spend about the same amount of money on plug-ins to be able to use your expensive devices, and get your home working like you want to.

    Sidebar: I also think, users understandably want plug-ins to be free. I do not expect that they want others to work for them for free, but they simply expect that the manufacturer of their device provides a free plug-in for HomeSeer. As the user base of HomeSeer is quite small, having manufacturers write plug-ins is unlikely to happen. (This is similar to driver support for Linux.) Having an easier way to build plug-ins will hopefully convince more manufacturers to write plug-ins for their devices.

    Development experience

    Developers spend a lot of their (free) time improving HomeSeer by writing plug-ins for it. Unfortunately this takes more time than needed at this moment:
    • The documentation is lacking (but there are already plans to improve it in the future )
    • There are a number of bugs. I am building a simple plugin and already ran into multi-threading and UI problems.
    • The best practices are not clear:
      • I too chose to build a multiple-instance plug-in, and now it looks like I have to rewrite quite a bit of code. (I actually like the concept of multiple-instance plug-ins: if something strange happens with one device and the instance crashes, the other devices will continue working).
      • There are too many design decisions to make upfront. Decisions that are difficult to make because you do not have the experience to oversee their consequences. Providing examples that show you how to use every feature are not helpful here. Get a basic example plug-in working that will cover most use cases, so beginning developers can have something running quickly.
    User interface

    Administration and control are totally different things:
    • Administration requires an advanced user interface:
      • Lots of devices (I see people on this forum that have well over 500 devices)
        • The plug-in I am working on is for relay boards. The largest board that I have has 24 relays and 8 I/O ports, so, including the root device, I have to add 33 devices to HomeSeer. And that is for a very simple device. There are plug-ins that support a lot more advanced devices.
      • Lots of values per device
      • Lots of actions per device
      • Setting up advanced behavior involving multiple devices
    • Control and information requires a simple interface:
      • I want my house to do this specific thing now. The thing can be quite complicated, but starting it should be easy.
      • I want to know this specific thing about my house now. Again, the information can be quite complicated, but I should be able to find it easily.
      • The usual user interface for control and information are:
        • A mobile app. Customization = (The app has to be very fast. I have an app for led lights, but I prefer to walk over to the other room to get the physical remote attached to HomeSeer because it is a lot faster.)
        • A (responsive) website. Customization =
        • Physical buttons, sensors, lights, screens, etc.
    Splitting old and new devices in the user interface is undesirable. And by splitting I also mean having a full experience for one set and a crippled one for another set on the same page. A specific function (e.g. a device list) must work well for all devices. I would hate having to navigate between 2 similar pages just because I have old and new devices, and I think it looks very unprofessional: I do not want to have to think about the version of the device when I want to do something, and I suspect many other users do not want to either. The basic navigation for devices and plug-ins has to be the same. That the actual pages of plug-ins and devices look differently and/or use different technologies is fine.

    Compatibility

    Compatibility is an (almost absolute) must. The devices I buy for my house (hopefully) lasts for many years (decades maybe). A plug-in controlling those devices needs to have the same lifetime. If a number of my devices become useless when HomeSeer is upgraded, there is little reason for me to stick with HomeSeer.

    Onboarding new users

    I understand that 1 of the reasons for the changes in HS4 is the onboarding of new users. They are currently overwhelmed by the advanced pages of HomeSeer. But I think onboarding can be done without dumbing down the interface for advanced users. I also suspect that most beginning users become advanced users in no time.

    For example: If the addition of a device or plug-in results in multiple HomeSeer devices, simply show a link like why do I see more than 1 device? that links to a help page the first time. The help page can explain how HomeSeer works, and what the best way is to manage your devices (e.g. hide instead of delete them). Advanced users can simply dismiss the message once.

    That's all I have for now. Hopefully it is useful.

    #2
    Thanks for the feedback, and I believe we are addressing all of your concerns. We will have a UI that does not leave existing HS3 users disappointed.
    website | buy now | support | youtube

    Comment

    Working...
    X