larryllix Posted March 31, 2020 Posted March 31, 2020 Just a comment about "sterile". Windows 10 has adopted this sterile look with all it's rectangular square cornered boxes and detail now. Plus it all went to shades of grey. Cripes it looks like Windows 2.0 before even Win 3.1 came out. I don't know what happened but this just looks like a step back into the 1980s. Get some rounded corners on those boxes for a softer (Win 7) look. It all sounds good to me but I see the extra power making it appear too complicated for the "one click, remote control, afraid of tech crowd". However, I am sure it can be done in this hidden "everything can be clicked" age. All the best with this. It is needed in some crowds for sure.
madcodger Posted March 31, 2020 Author Posted March 31, 2020 Just a quick note to say that I'm thrilled to see this discussion moving forward. I plan to dive into more of this in the next few days (a bit busy as we close out the quarter at work). Nice to be working with such smart folks. I feel like a kid at their parents' workplace. Every fifth word sorta/kinda makes sense. But I didn't start this thread to be the smartest person in the room... I started it to attract those folks. Thanks for showing up! 1
Michel Kohanim Posted March 31, 2020 Posted March 31, 2020 @larryllix, I have no clue what you are suggesting with Polisy and subscriptions and X10 ... @kewashi, As I mentioned before, we are investigating Vue for the Admin Console. We will not use homegrown javascript because we'll eventually end up where we are with Java. Need to use a supported framework that can be distributed amongst multiple developers. Also, and as we discussed, I worry about monetization for you: 1. If you go this route of UI sales for ISY customers, then we'll definitely have issues with a rules engine inside the UI because the customers will call us for support and not you. The number of tech support calls we get for MobiLinc and Agave is probably 25% of all our calls. Adding more complexity increases more calls and we just don't have the extra resources to train the staff or direct customers to you 2. If we go the route of us contracting you to develop the UI for us (based on HousePanel), then - as mentioned above - we will need to use a platform. And, currently, the platform is Vue 3. We can make a package out of HousePanel so that it's easily installed and runs as daemon. You don't need node servers for that With kind regards, Michel 1
kewashi Posted March 31, 2020 Posted March 31, 2020 8 hours ago, Michel Kohanim said: 1. If you go this route of UI sales for ISY customers, then we'll definitely have issues with a rules engine inside the UI because the customers will call us for support and not you. The number of tech support calls we get for MobiLinc and Agave is probably 25% of all our calls. Adding more complexity increases more calls and we just don't have the extra resources to train the staff or direct customers to you 2. If we go the route of us contracting you to develop the UI for us (based on HousePanel), then - as mentioned above - we will need to use a platform. And, currently, the platform is Vue 3. We can make a package out of HousePanel so that it's easily installed and runs as daemon. You don't need node servers for that Thanks @Michel Kohanim this all makes perfect sense to me and is very helpful in directing me in moving forward. On point 1, if we go the route of UI sales it is easy enough to remove the Rules engine capability. Although, you might try it to see if you like it and want to support it. Up to you. I'm not in love with it. I added it because it wasn't hard and I had the need to provide multi-hub architecture rule support. It is easy enough to neuter it. If all you have is an ISY hub or multiple ISY hubs, then I think you can do everything you want from an ISY program. For hobbyists with multiple hub types, they could still use a pre-release hobby version with no support. On point 2 - If we opt for the UI development, then there is a ton of work to convert HousePanel from jQuery to Vue. I'm elbow deep in evaluating that as we speak to see what it would involve so I can get you a thoughtful proposal. I should mention that there isn't anything magical about using Vue that would prevent you from getting stuck in a new rabbit hole of custom js code. While Vue is a standard platform the developer is still free to build all sorts of custom crap around it. I promise I wouldn't do that to you. Vue by itself doesn't do enough to standardize around. How you use Vue will dictate what you are standardizing upon. I think the same applies to "how" one uses lower level libraries like jQuery too. Okay I'll get off my soap box - just advising you to not think Vue is a panacea that can protect you against absolesence. The best way to future proof is to have a software architecture that is flexible and built on web standards. This is why I designed HousePanel around CSS - that is the key standard at play here, not jQuery or Vue. Thanks for pointing out #3. Very useful to know. 2
Michel Kohanim Posted April 1, 2020 Posted April 1, 2020 @kewashi, 1. Using a rules engine that's part of the UI is not something we would ever promote. Of course, we don't mind at all if users want to use it and I am sure many will like it (just like HASS.io) as it bridges some gaps. A better approach would be for you to make node servers out of the integrations you have made. This way, there's a higher likelihood of monetizing each rather than dumping everything into the app 2. I understand. Our UDAjax as CSS based but not maintainable. Having a framework makes things a little easier especially when libraries and dependencies are upgraded With kind regards, Michel
kewashi Posted April 2, 2020 Posted April 2, 2020 1 - Got it @Michel Kohanim ... The plan will not include a UI based rule engine 2 - I have a full week free next week to work on the plan so expect to have it to you soon. 1
kewashi Posted April 5, 2020 Posted April 5, 2020 Important BugFix Version 2.230 posted tonight of HousePanel. If you are actively evaluating it, please update to this new version. No need to start over, just download and replace the changed files and restart your Node.js server.
Mustang65 Posted April 9, 2020 Posted April 9, 2020 On 3/30/2020 at 10:45 PM, larryllix said: Just a comment about "sterile". Windows 10 has adopted this sterile look with all it's rectangular square cornered boxes and detail now. Plus it all went to shades of grey. Cripes it looks like Windows 2.0 before even Win 3.1 came out. I don't know what happened but this just looks like a step back into the 1980s. For those that are not familiar with how Windows 286/386 arrived. 8 disks each. 2
asbril Posted April 9, 2020 Posted April 9, 2020 1 hour ago, Mustang65 said: For those that are not familiar with how Windows 286/386 arrived. 8 disks each. At least they were not the square floppies
Mustang65 Posted April 9, 2020 Posted April 9, 2020 Just now, asbril said: At least they were not the square floppies I have them on the 5 1/4 inch floppies also. Probably in the back of my old storage cabinet. Starting to dump this stuff now. 2
simplextech Posted April 9, 2020 Posted April 9, 2020 38 minutes ago, Mustang65 said: I have them on the 5 1/4 inch floppies also. Probably in the back of my old storage cabinet. Starting to dump this stuff now. Here is you really must have it... https://winworldpc.com/product/windows-20/20 and here's MS DOS to go with it https://winworldpc.com/product/ms-dos/1x
kewashi Posted April 9, 2020 Posted April 9, 2020 Too funny... I have a full room dedicated to old tech including many similar throwbacks - including an original iPod 1.0, flip phones, VCR players, BlackBerries, WinBook webcam, Turbo Pascal install disk (amazingly it came on one floppy), and so much more. It is a sobering reminder of just how fast tech moves - because it wasn't too long ago this stuff was all pretty hot. I'm still kicking myself for selling my Commodore 64. ------- Now, on a more modern topic, I pushed an update to HousePanel to GitHub tonight with the usual collection of bug fixes 1
Brian H Posted April 9, 2020 Posted April 9, 2020 (edited) My first computer was a user built Heathkit H8. Cassette tape storage, 4K of ram and you had to input the boot loader with a keypad. Eventually got the update rom to allow it to boot without having to load the boot loader manually and a 8" single sided floppy disk drive. HDOS or CPM OS. Edited April 9, 2020 by Brian H 1
HABit Posted April 9, 2020 Posted April 9, 2020 Amazing! I remember those days too. First computer was an Ohio Scientific, same cassette boot loader and eight data, 16 address switches on the front with a load switch. Still remember the jump code was a C3 to get the thing running after a reset. Moving is a great catharsis for throwing out hoarded electronics/computer - er, treasures. I let go of all my 3.5 floppies when I realized that I had no computers with a 3.5 drive to even load them. I still hung onto a fully working early Pentium with Windows 95 so periodically can play Mech Warrior. 1
mapeter Posted April 9, 2020 Posted April 9, 2020 8 hours ago, larryllix said: weaving those little wires though the tiny iron doughnut My first experience programming was on a Data General Nova 800 with a whopping 16K of core memory. Toggled in the bootstrap from the front panel and then loaded the paper tape. 1
asbril Posted April 9, 2020 Posted April 9, 2020 3 minutes ago, mapeter said: My first experience programming was on a Data General Nova 800 with a whopping 16K of core memory. Toggled in the bootstrap from the front panel and then loaded the paper tape. My evolution was Texas Instruments 16 K; Commodore 64K and then IBM PC Jr (with cordless keyboard)......
auger66 Posted April 9, 2020 Posted April 9, 2020 On 3/25/2020 at 11:02 AM, lilyoyo1 said: . . . Out of all the complainers, how many have even personally reach out to one of the developers here to see what can be done and for how much? Who has reach out to udi saying with your permission in going to take these steps to raise money? If any has, then kudos to them. Should it happen, I'll even donate. We'll see how many others follow suit This isn't exactly what you are talking about, but me sort of. I contacted the eKeypad developer last year to make an simple interface for an iPad in a rental unit. I've fooled around with HomeSeer and iRule extensively in the past and don't have the interest to do that again. I would rather just pay someone and tweak it as necessary. The developer was very difficult to get a reply out of. I finally did with someone's help and worked out the terms. I spent some time opening a port, sending him graphics and what I wanted. Never heard from him again, and finally gave up trying, which is too bad because I think it would have worked perfectly for what I wanted. Michel, you might want to rethink featuring them on your home page. Jayson was nice enough to talk to the one time, but he clearly isn't into eKeypad anymore. 1
kewashi Posted April 11, 2020 Posted April 11, 2020 (edited) Hey @auger66I'm not quite sure I know what you are talking about, but this thread is about HousePanel and its quest to provide a new modern GUI for ISY hubs. We took a dirt road down a nostalgia street of old geeks who remember the good ol days, but I will bring us back on topic here soon enough. I'm working on a version of HousePanel that should be pretty amazing for ISY users. It is available now as a beta preview on my GitHub site found here: https://github.com/kewashi/hpserver I'm still actively developing it and tweaking it. Made a number of big updates tonight. Also working on a transition from jQuery to Vue so that it will become a more standard back end with support. Stay tuned for more. In the meantime if you want to try HousePanel, go for it - costs nothing during this beta phase. And I promise I won't ignore you. I'm actively engaged in this forum - at least on this thread. Just don't expect a polished pro product yet. That will be coming later. Edited April 11, 2020 by kewashi
lilyoyo1 Posted April 11, 2020 Posted April 11, 2020 On 4/9/2020 at 6:44 PM, auger66 said: This isn't exactly what you are talking about, but me sort of. I contacted the eKeypad developer last year to make an simple interface for an iPad in a rental unit. I've fooled around with HomeSeer and iRule extensively in the past and don't have the interest to do that again. I would rather just pay someone and tweak it as necessary. The developer was very difficult to get a reply out of. I finally did with someone's help and worked out the terms. I spent some time opening a port, sending him graphics and what I wanted. Never heard from him again, and finally gave up trying, which is too bad because I think it would have worked perfectly for what I wanted. Michel, you might want to rethink featuring them on your home page. Jayson was nice enough to talk to the one time, but he clearly isn't into eKeypad anymore. A developer doing a personal project for an individual isn't choose to bring the same as developing a system for the isy in general for every single person to use. I can't speak on your specific situation with a developer but what this discussion is about is like comparing apples to Oranges. Yours is a business deal that didn't work out vs someone advocating and speaking for a group as a collective. Not knowing both sides and the details of your interaction, I can't speak on it. I'm sure clients that I've walked away from would speak bad about me at the same time not giving both sides of the story (and vice versus). Because of that it would be improper for Udi to get in the middle of things and not support a developer that had a product that works with their system
TrojanHorse Posted April 11, 2020 Posted April 11, 2020 I think the point is that UDI has ekeypad on their website as a viable option. It’s not a good one IMO. It’s abandonware and so is agave as far as I can tell. That’s in part why I’ve thought UDI should have *something* of an app / simple remote interface that’s theirs and they control. Sent from my iPhone using Tapatalk 1
kewashi Posted April 11, 2020 Posted April 11, 2020 (edited) Never mind... I see you weren't talking about my project... anyway, it will require patience and time. Hopefully not too long. Edited April 11, 2020 by kewashi misinterpreted prior note
madcodger Posted April 11, 2020 Author Posted April 11, 2020 (edited) 10 hours ago, kewashi said: Hey @auger66I'm not quite sure I know what you are talking about, but this thread is about HousePanel and its quest to provide a new modern GUI for ISY hubs. We took a dirt road down a nostalgia street of old geeks who remember the good ol days, but I will bring us back on topic here soon enough. NO, @kewashi, THAT IS NOT WHAT THIS THREAD IS ABOUT. I started it, and you have been an active contributor, but don't try to hijack it. Start your own if you wish. This thread was started to gauge the need/desire for a better user interface for the ISY. There appears to be some interest, which is great. But ANY developer is open to contribute, and discuss their ideas here. HousePanel may or may not fit the bill (personally, I'm not yet interested as it requires too much DIY for my tastes). If you want a HousePanel thread, go start one, or revive one. But please don't think this thread is all about your product. Contributions and ideas are welcome. Hijacking is not. Edited April 11, 2020 by madcodger 2
lilyoyo1 Posted April 11, 2020 Posted April 11, 2020 9 hours ago, TrojanHorse said: I think the point is that UDI has ekeypad on their website as a viable option. It’s not a good one IMO. It’s abandonware and so is agave as far as I can tell. That’s in part why I’ve thought UDI should have *something* of an app / simple remote interface that’s theirs and they control. Sent from my iPhone using Tapatalk Keypad has had updates so it can't be said that they are abandonaste. It's definitely not a popular app so it could seen that way. The same for Agave. @James Petersonis still active on these forums. While Agave hasn't pushed out new updates or features (which is disheartening for people who paid good money for it), he has been clear on the forum his intentions and reasons.
Javi Posted April 11, 2020 Posted April 11, 2020 (edited) I have started an Android App after thinking about this problem for some time and have had a few rewrites as my assumptions about the ISY API and reality have had differences. I would like to make this for iOS also however I do not have much free time and am more familiar with Java (native Android), so this will be in the distant future. Development for a small screen is much different than development for a large screen so finding a way to incorporate all the features of the ISY into a small screen while also keeping the UI clean will be a challenge. Once I have a database and have developed methods in Java it will be easier for me to start on another platform. There are already a couple of apps for the ISY which (in my opinion) have development driven by subscriptions and/or app sales revenue. I believe @madcodger started this topic because these apps are are not satisfying. If the revenue model has not provided a satisfying experience then creating another app with this model may produce competition, but (in my opinion) it may also produce even less incentive/revenue for development. I know UDI is/has provided incentives to produce an app even though many users may not have noticed. However if the developer is not already invested in the ISY this may produce another unsatisfying product or abandon ware (UDI mobile anyone?). From my experience anything provided for free (or open source) has resulted in too may feature request and overwhelming support request for a single developer. Once I have charged a couple of dollars these request become manageable. So even if I am not motivated by revenue (although it would be a nice) charging for something will save me time. The other potential problem with open source is that another (malicious) developer may copy the app which can result in me receiving support request for something I can not control or even worse the users credential being compromised. With that in mind I will open source some of the code for a basic app and have already done this. The “ISY Socket Log” is open sourced to help other developers create there own apps with out having to reinvent the wheel, although I am behind in updates. I am invested in the ISY ecosystem but my family and work must come before my hobbies. So this process will be slow. Users can help any developer by becoming a tester and understanding that this process will be slow especially for developers which are not motivated by revenue. All apps must support both 4.x and 5.x as I believe most users have a 4.x system and will not fix/upgrade something which is not broken. I have a 4.x system myself which may take me more than a day to upgrade, but it is not broken, so even though I am capable of upgrading I will not be doing so anytime soon. The 5.x firmware is much easier to develop for than the 4.x. So even though I think all apps must support both firmware versions there may only need to be basic 4.x support and informing users to upgrade their firmware if they want additional features is acceptable. The area I see most important in regards to competition is the user interface. With the 5.x firmware the zwave/insteon/node_server control problems are fixed (aside from a few bugs), the only problem developers need to solve is how to present these items to the user on a mobile interface. This will be the most difficult thing to solve. Things I am (trying to) implement(ing): App should be able to connect to multiple ISYs simultaneously. App should have multiple easy to access screens/fragments. This should include a screen which mirrors the ISY and other(s) which allow customization. Users should be able to change Icons/Text for a device. The Icons/text should be customizable to the state of the device. Users should be able to combine devices into a single (list) item. Users should be able to create folders (rooms/locations) that are unique to the app. Users should be able to duplicate items/nodes. App should be able to use both the local and remote connections to the ISY. This should be handled by the app and the user should not have to specify if they would like to use the local or remote connection. Limitations: Push notifications require an external server which has ongoing cost. When I am at the point of implementation I will see if it is possible to the UDI portal cloud node servers for this purpose. If it is not possible then this would have to be a subscription. This should only be used for notifications as it is not a good use of resources to update devices with push notifications. Depending on the platform (iOS/Android) there may be services which can be run to keep widgets updated. However if the app is fast the user may realize they only needed this feature because the app they were using was not fast enough. When I am at the point of implementation I will have more information on this, and my include options such as requesting status for widgets when the screen is turned on. What other features would you like to see? Edited April 11, 2020 by Javi 1
lilyoyo1 Posted April 11, 2020 Posted April 11, 2020 (edited) 3 hours ago, madcodger said: HousePanel may or may not fit the bill (personally, I'm not yet interested as it requires too much DIY for my tastes). This is part of the reason some developers including udi isn't jumping to develop an interface. The people wanting something all have different ideas and desires in regards to what they are looking for. Due to this, the time and effort they spend to develop and maintain something may not bear the financial rewards to make it worthwhile. Not only that, it simply won't be what people are looking for causing the cycle to repeat itself. Edited April 11, 2020 by lilyoyo1 4
Recommended Posts