RIP Tidbyt

General Discussion
4

so - deep breath & reflecting on my journey in life - here is a more considered reaction from me.

many of us have been where they are. built a product that was cool/fun/exciting & then a larger entity came along and said - we want to buy “you” (not your product, but you). It’s great for them, we might get some support for the devices / apps - but I’ve seen it a million times - one company buys another for their team’s brainpower & the old products get shutdown. I love that they were able to advance themselves as entrepreneurs - it’s a goal I have every time I start a new venture & I love to see others succeed. Will run my devices as long as I can & add as many features as I can. But congrats to the team & their personal success!

2 Likes
5

Why do I feel like it won’t be long until my Tidbyt becomes a paperweight?

3 Likes
6

Ouch. You’d think they could have bothered to tell us, their customers. :frowning:

2 Likes
7

I’m bummed, but no hate for the Tidbyt team. Congrats on the acqui-hire.

  • What are the chances of piecing together a self-hosted private Tidbyt server?
  • Could the Tidbyt team open source the manufacturing process? I know approximately zero things about hardware, so I don’t know if that’s actually a thing one can do. If it is possible, I’d love to have the flexibility the order the components and assemble the devices myself.
4 Likes
8

It’s part of a venture’s life arc to be acquired, sure. It’s a bummer and a mystery as to how long this lovely machine I use in my classroom will be functional, but I get it.
Still, I went to the Tidbyt site from the Modal site having read the news and the first thing that popped up was this offer of free shipping. That’s bad form. If you’re shutting down manufacturing and distribution, cut the marketing, please.

Screenshot 2024-11-12 at 9.09.28 AM
Screenshot 2024-11-12 at 9.09.28 AM1318×1334 99.7 KB

2 Likes
9

Based on the last few days of radio silence, it seems that this is purely an acquihire exercise. A bunch of people scored new jobs and stopped caring. Just noncommittal statements and zero follow-up.

The company is good as dead, but to avoid further bad press after the disastrous Gen 2 launch, the subscription service and the widget spinoff thing, they didn’t want to say goodbye. It’s easier to let everything fade out of existence than making any formal statements of abandonment.

How is Tidbyt supposed to pay the bills when it isn’t selling anything?

Why spend any time or money pivoting the platform to be 100% local/community driven, when they have new jobs to do? Modal aren’t going to help, the posts suggest they took the people, not the product. The shut down of hardware production was likely a necessary evil for them, before they could commit to the new roles.

They simply set the stock levels to 0, uploaded a blog post and walked away. No updating their customers directly. A vague EOL notice will be posted one day, after the server use volumes drop below a certain threshold. Less people to anger before the servers are decommissioned.

If we’re lucky, they roll out the final stretch of what’s left over from their prior promises, like Home Assistant integration and fixing the multiple bugs in the apps. Maybe, just maybe, they’ll do the right thing and create user-friendly self hosted services after pushing firmware and app updates to properly tie up loose ends and sunset the service.

After realising that their entire concept was doomed earlier this year, I rolled out self-hosting and fixed the problems that they wouldn’t fix. My Tidbyts are mostly fine for now. In its current form, self-hosting is too complicated for most and won’t prevent thousands of devices becoming e-waste unless the founders do the right thing.

But that’ a big ask. Rohan and Mats have already demonstrated that they weaponize poor communication. Will we ever get clarity with the timeline of things to come? Maybe, but given the definition of “clear” has been established with release of Gen 2, they might already think that the blog post is the clearest thing ever written, based on their intended design.

Maybe we should have read the post standing 3 feet or greater.

7 Likes
10

So there is a way to self-hosting right now? Any links to get me heading in that direction?

3 Likes
11

This forum, the subreddit, the discord and most importantly, the official github already has everything you need. As you had to ask in this format, its probably safe to assume that it’ll likely be out of your current skillset so it’ll be an uphill battle, but poke through the developers topic as a starting point.

I went down the pixbyt route.

1 Like
12

Appreciate the response.

1 Like
13

Does pixbyt ultimately rely on tidbyt API for pushing images to the device? Everything in the doc makes it look like you need the “device id” from the mobile app so you are still reliant on their cloud.

5 Likes
14

not just an EOL for hardware, but community repo pull requests haven’t been touched in two weeks, just before the acquihire was announced. i just submitted a new app yesterday; i wonder if it’ll ever make it in given how clearly the team has just… walked away.

4 Likes
15

Very disappointing, right after they said that “nothing will change”.
I have two apps there waiting for approval, one of them is just a clock… barely needs review. :frowning:

A suggestion for @rohan, maybe you guys could select a couple of senior community developers and give them permission to approve PRs? Just tell them what to look for when reviewing apps and let them take the load off your backs! I’m sure this would help keep things going and less dependent on you guys.

6 Likes
16

given how little regard the tidbyt team clearly has for its users, i wouldn’t hold my breath. for now, i think i’m going build my app in on more sensible platform (starlark was a big red flag in retrospect), rendering webp/png from a cloudflare browser rendering worker, and then pushing it using their API until it’s inevitably unplugged; i’ve seen this movie before.

2 Likes
17

I would definitely volunteer to be one of them if they do this!

2 Likes
18

if anyone else is looking to continue building apps without relying on community repo maintenance or hosting a starlark server, i just ported MealViewer Next Lunch from a community app to a Cloudflare Worker app. It runs every hour, rendering the screen in a browser, taking a webp screenshot, and then using the Tidbyt API to push that to my device. Feel free to use it as a template.

2 Likes
19

Have you heard anything back yet? I hope they reviewed your apps by now :frowning:

20

I had two apps awaiting review, one of them was a simple clock that did not use any external APIs. I asked @matslina if he could approve at least this one and he did! He also approved all the other similar clock apps that were pending.

I’m still waiting for the Bluesky user counter app to be approved after two weeks. :slightly_frowning_face:

21

It would be great to at least open source the Software on the device so we can modify it to publish directly do the tidbyt device without the overall infrastructure. Thanks a lot.

4 Likes
22

Last year I created a Python script that serves as a Python-based server emulation for managing and displaying WebP images on Tidbyt devices based on certain conditions. The updated (refactore) version allows users to specify device names (helpful if you own multiple devices) and control the logging level through command-line arguments. The script is structured to continuously manage and display different apps based on time conditions and app settings.

Use Cases for the Emulation

  • Handle the management for a Tidbyt digital display device
  • Use .webp files generated from .star files to render animations
  • Ability to display WebP’s for specific durations
  • Ability to turn off rendering certain WebP’s based on nighttime when the display should be off or show darker versions
  • Ability to queue the order of the Tidbyt apps
  • Ability to continuously cycle through the selected apps
  • Already have the next WebP ready to be sent to Tidbyt when the current WebP finishes displaying

The main goal of the script is to allow users to control the display of WebP images on a Tidbyt device based on time settings (daytime vs nighttime) and pre-set schedules. It also offers a facility to customize display content and duration for each app. It utilizes command-line arguments to configure the device name, day start/end times, and logging level, making it flexible for different environments and use cases.

The script with its readme file are located on my GitHub repository: GitHub - jekeybl/PythonTidbytServer: Python emulation of the Tidbyt server

4 Likes
23

This is awesome!
The only problem with this approach is that it still relies on pixlet push and this in turn relies on the Tidbyt Servers. In the event of the servers getting shutdown this will also stop working.

Some folks on Discord are experimenting with custom firmwares and alternatives that would allow the device to fetch the images from a simple http server.

1 Like