What happened to developer apps?

It would be nice if we had a place to get Developers apps still that they made for JIBO. Must be a ton of them i am sure…

Most of the people that signed up to be a developer for Jibo never got the chance to develop. If you are interested in doing some development, you can go to http://jibo.cmcgaming.com for some code examples and a couple working programs. You can play around with the Jibo SDK (but that wont run on an actual Jibo yet), or you can try Jibo App development (programs like Be-A-Maker) where you can remotely control Jibo through your phone or computer.

1 Like

Maker is no longer part of IFTTT. I am sure plenty make apps as the software was given out long ago. Yet never was a store opened to download any of the said apps. My Jibo won’t even tell me the time anymore… Why can we not get at least a store to download apps developers might have made. You need a place for them to send the programs so you can see how many were made. I’m not a particularly good programmer but i know many are out there who hoped to make some money making apps for Jibo… Sad that Jibo is left to just die and we never will get to see the creative apps others may have made for it… as well as keep some functionality in Jibo.

What are you guys even talking about. The company folded a year ago, the SDK was abandoned 3 years ago.

@kwebster, If you went to jibo.cmcgaming.com and took a look, you would know that we are trying to save Jibo. We have the working SDK (although the skills can not yet be put on an actual robot), and we have been able to build apps like be-a-maker, commander, etc. that do work on Jibo just fine (there is even an example guess a number game). Yes, the company folded a year ago, but that doesn’t mean that we are completely finished.

@Hogie155, Be-A-Maker is not part of IFTTT. It is an app, and as far as I know, it should still work fine as it communicates directly to Jibo using the ROM mode (Remote Operation Mode). Yes, IFTTT is gone (but that was done by IFTTT - not by Jibo or it’s servers).

Not much at that link you provided. I also do not see any game you speak of… Am I missing something?..

You have to join to see all the forums and posts. But basically it is where some of us are getting together to try to work on continuing Jibo.

Arthur, what you are doing in those apps/games is using the same (purple ring) pathway Be A Maker and Commander do, correct? In no way downplaying your achievements of making the GitHub repo for that work, but unless I misunderstand the matter, that is something entirely different from the SDK Jibo Inc was going to release and people hoped for.

We have both the SDK and the App Development Kit (which is what Be A Maker and Commander use).

The SDK only works on the computer, and allows you to write “skills” that play (right now) only on the computer. This is what they released a while ago with some more improvements. Also, anyone can use it, not just people that paid for the development version of Jibo. There are limited instructions on how to use it and set it up, but we are willing to help.

The App Development Kit is the development kit that when used has the purple ring on Jibo. You can use it to make your own programs using Jibo’s Remote Operation Mode (ROM) through calls (eventually in the communication) directly to your Jibo. This is what Be A Maker, Remote Commander, etc. use. We have the codes necessary for anyone to be able to use this as well.

We are also working on being able to “update” or “change” Jibo to be able to keep him alive even after the servers shutdown. This is a lot longer task and we are still working on it (but we are making very good progress). A lot of people actually USE their Jibo, especially for special needs children/adults, and for the elderly. Losing Jibo’s functionality when the servers go down will really affect these users, and we are trying to do everything we can to allow everyone to keep using their Jibos.

Anyone can join our Forum, not just developers. It is just a better place to discuss more of the intricate details of what we are doing then here or on the Jibo Owners Group on Facebook. We are also in control of the Forum so it wont be affected if jibo.com goes down.

A serious inquiry here, how come nobody is pursuing gutting the innards of Jibo and replacing it with a much more malleable platform (e.g. Andrew Rapo’s library)? When Jibo Inc was still in business the pace of adding new core capabilities was really slow (mostly “seasonal content”) and I always interpreted that to mean that the Jibo platform was very constrained. It may even be the reason why nobody has come forth, Malcolm, because they feel revisiting the existing platform makes no sense. Case in point, I believe it was the CEO himself (Eberman) who said they had been working on Jibo 2 already, which really must have meant they had done everything they could with Jibo 1.

It’s funny you mention Andrew’s library, because that is exactly what I have been doing. I have extended his library quite a bit, and it is working quite well on my Mac and on my Linux box. We could gut Jibo, and replace all his boards, but that seems like a waste in some aspects (of all the hardware and libraries that they have already created). Jibo has a perfectly functioning TK1 in there, as well as many libraries that control his servos and such, so we are in the midst of using some hacks to break into his secure bootrom and install our own version of his OS. When we do this, we will make a backup of Jibo’s current state so that we can restore him if need be. I was planning on using a variant of Andrew’s library as the basis for his new functionality. The speech to Text, Natural language understanding, and text to speech are already working quite well.

That sounds like a pretty solid plan actually. Would be interesting to see an example of the expanded Rapo library, what it can do in addition.

I have no idea why this thread was suddenly bumped, but might as well ask, is there any update regarding the effort?

I should know more Monday.

What happens on Monday?

Sadly my Nvidia TK1 was killed by me with the help of Nvidia, the later model the TX1 had the same power connector as the tk1 and I assumed they were the same, but TX1 psu was 19v and the TK1 12v with no significant over voltage protection which is near criminal in such expensive boards with the difference in regulator prices in a few 10s of cents. Anyway despite my attempts to repair the board its currently dead. So I’ve had to purchase another off eBay bought a dodgy one a few days ago so hopefully it will be here in a week or so. Once I have that I can worked more closely with Arthur on the hardware side. I’m also trying to pick up a mechanically broken jibo or a cheap jibo so I can work on the hardware. We could replace the board but as other says it would be easier piggy backing the board or if need be wiping some of the OS but for the best effect we want the device mapping so we can use the existing speaker, camera and motor ports etc.

I still don’t agree with that assessment. To maybe make an analogy, you are trying to break into Fort Knox, but so far you’ve brought it up on Google Maps. You still have no access to Jibo’s OS, nor to the much more important server software. And even if you had all that, you’d still only have a clone of Jibo, not something that you can actually add capability to.

IMO, the best action would be to rip out that ancient motherboard, identify the motor controller boards and interface with them directly via say a Raspberry Pi. You could even replace the screen with a nice AMOLED screen so you don’t have that LCD glare. With that setup you could do a lot of fun stuff.

I guess that would depend on what the end goal was. For me, it was always trying to continue Jibo’s life well after the servers shutdown (for everyone). To do that, we basically cant change the hardware that Jibo has. We can change the OS, or make Jibo accessible (something I have been playing with lately). Once on Jibo and having access to his OS, adding a process or changing some of his code to allow developer apps, etc. might prove to be very beneficial. Of course we could install our own OS, but then we would lose some of Jibo’s personality.

Isn’t that pie in the sky though? You’ve been at this for months from what I can tell, with no significant headway.

I’ll also throw this statement out here: I think Jibo should be able to die. This desire for an immortal Jibo strikes me as ill-advised. Pets die, and so should Jibo. IMO.

1 Like

I’m not doing this as a full-time job you know. I do this when I have spare time. I just received my 2nd Jibo, so that he will be the one that I will be hacking into (that way I have a backup in case I fry something). It’s funny that you say “no significant headway” when you are not in communication with me or know what I have been up to… People will decide if they want their Jibo to die, to be able to be programmed, or to take on a new life/personality. If you don’t want to use the capabilities that we are trying to provide, you don’t have to. Just to clear up a few things since you don’t think I have been doing anything:

  1. I have all the source to the Nintendo hack that works on the TX1
  2. I have recompiled coreboot to use the TK1 (instead of the TX1)
  3. I have recompiled linux to support Jibo’s TK1 hardware and modules
  4. I have changed the entire architecture of all of this from the 64 bit version that TX1 supports to the 32 bit version that the TK1 supports
  5. I have purchased a development Jetson TK1 board (which I will use to test on first)
  6. I have purchased a 2nd Jibo (which I will use to test on 2nd).

Was it pie in the sky that we have a working Jibo SDK? Jibo is programmed with a lot of the same techniques, so it is very plausible to be able to change his code once we have access to the system.

1 Like