Dash Discontinued Q&A


#1

Recently we announced our decision to discontinue manufacturing the Dash. This announcement likely comes as a shock to Dash owners. We’d like to use this forum post to offer support and answer any questions. We appreciate everyone who has supported the Dash through the years. You are a great group an valued part of this community.

If you loved the Dash please let us know :slight_smile:. Your positive feedback would mean a lot to everyone on the team who had worked on the Dash.

Please refrain from creating additional posts about the announcement in order to keep all interested members informed of new comments.

If you need help porting Dash code to a different board please create a new post under the 3rd Party Hardware category. https://community.hologram.io/c/hardware/3rd-party-cellular-hardware

All new Dash technical support questions should be asked under the community GitHub repository as a Issue.
Hardware: https://github.com/HologramEducation/dash-hardware
System Firmware: https://github.com/HologramEducation/dash-system-firmware
Arduino IDE: https://github.com/HologramEducation/hologram-dash-arduino-integration


#3

#4

Damn, guess that’s why no news on the Cat-M1 version I signed up for notifications.

So, if we download the open source firmware we can set it up to talk to our own servers for the inevitable day when it’s EOL’d?

And, I know I’m just a small developer relatively new to Dash, but it is a sweet board and I really enjoyed using it! I was so looking forward to an even lower cell powered version.


#5

Sorry Jer. It’s really cool to hear how much you enjoyed your Dash.

The System firmware gives you access to all the same code we used to send messages to our server. As far as I know, you can change the server endpoints. Post any technical issues you encounter when trying this in one of the github repos above.


#6

Just wanna mention that the cloud isn’t going away and so your Dash will always be able to send and receive messages. It’s only the OTA update service that will be going away in a year.


#7

Thanks for the notice. I really enjoyed working with the Dash. I need to buy a replacement antenna, where can I get the nice antenna you guys have on the Dash? It works better than anything else I have tried.

Thanks :slight_smile:


#8

@themikemachine

I prevously asked Hologram about their antennas, the exact antennas are custom made for Hologram and they don’t resel the antenna’s only. From looking their antennas are very close to the Laird EPR8221A1-15UFL. The actual PCB trace / radiating element is identical but the differences are:

  • Laird does not have adhesive back, Hologram antenna does
  • Laird has a longer cable 6", Hologram has ~3"
  • Laird antenna is on a rigid PCB, Hologram antenna is flexible

All that being said I would expect you to have nearly identical performance from the Laird antenna minus like 1/2dB fro the cable loss.


#9

This is an incredibly disappointing, and I believe, very poor decision.

The Dash was an absolutely brilliant and well designed product. I spent a few months designing a one-off remote car starter using one and in the process came up with several other product ideas, a few of which I intended to investigate for commercial products.

So the Dash wasn’t profitable. I get it. You want to become an IoT network company, not a hardware company, I get it. But here is what you may not be considering:

The Dash was the gateway for many small (and large) developers to discover your network and explore it’s capabilities. Now, trying to do so with 3rd party hardware poses all kinds of challenges and many potential customers won’t even give Hologram a second look.

I’ve been developing Arduino based projects for several years and when needing cellular capability, Hologram and Particle rose as the top two contenders because of the integrated hardware and network offerings. I chose Hologram because of your network’s competitive pricing model with intent of developing several projects.

The biggest failing of the Dash was Hologram’s inability to properly promote and document the wonderful product they created. When compared to getting started with Particle’s offering, there exists a minimal amount of examples and documentation. Developing with the Dash was intimidating but after solving one problem after another it became apparent what a great product and robust network Hologram had created. However, I was only able to convince other Arduino makers of this fact by revealing what I had discovered, as your website did little to promote the great projects that could be built on your hardware and network.

Consider the case of Pan Am Airlines. They wanted to create a world class international route structure, and always considered the domestic feed too unprofitable and left it to other airlines. But they soon learned they needed to control that domestic feed to make their customers happy but by the time they tried to fix it they were in decline as their competitors built their international capabilities. Or maybe closer to the mark, Apple Computing. Their products are so loved because of tight integration between hardware and software. Now if I want to use Hologram’s network for upcoming projects, it’s going to be a major pain to figure out where to start.

Your biggest failure with the Dash was not appreciating and promoting what you had created. You are going to leave a major segment of the IoT market behind in abandoning it now, just as you are trying to position your network as the major force in IoT. You really need to consider the Dash’s profitably not in a vacuum, but in context to providing access to the network you are trying to build.

In short, if you want dominance in the IoT network game, you need to offer hardware to let people experience what you are offering.

  • List item

#10

True… and I agree.
I BELIEVED in Dash, in its community, and in Hologram.
We supported the community thinking it’s OPEN SOURCE spirit will be kept alive… forever… or beyond NBIoT at least.

In my opinion, it’s too early to kill DASH… and very disappointing for early adapters of your hardware… Your roadmap for the N2 series is also very much anticipated, because SODAQ and Nimbelink is not HOLOGRAM.

Eventhough I have not seen any “DASH production/lifespan until 20__” clause in Hologram just like how NXP and other manufacturers guarantee their product existence, I never though Hologram will give up too soon…

I can see in the community forum that HOLOGRAM SIM buyers are growing (using them in other hardware platforms)… and I thought that continuing/improving/being the best hardware platform will make HOLOGRAM as a 1-stop-shop for developers and R&D groups.


#11

@Reuben,

SIM + HARDWARE + CLOUD + OTA = Hologram Platform
You’re already giving competitive pricing for the SIM… which will draw more clients compared to your pricing before.
Business wise, I thought Hologram is going for a 1-stop-shop kind of Platform which ARDUINO (a hardware/firmware company) will not do or might take a while to engaged in.

Particle, Sodaq and NimbeLink is not always better than Hologram. And I think if you’ll kill the hardware part of your platform, the community will follow suit, and Hologram will not get the chance to improve their remaining products. Are your tech team willing/prepared to test your SIM and Cloud platform against the hardware of the companies mentioned above? Is Hologram’s platform trimmed down to become like Carriots and the like?


#12

You asked for feedback, so here it is: “You should dance with the one who brung ya.”

No, I won’t tell you how to run your business, but I want to tell you what this sudden and completely unexpected action means to your customers.

A number of us, myself included, have been with you since the Kickstarter days, when we patiently waited for you to launch the product, and even for you to “get the packaging right” and hand-solder the boards. Those were the days.

The Hologram Dash has been an outstanding hardware, software, and data communications system. The pricing has been competitive, and the integration of the various parts of the Hologram ecosystem has been outstanding.

Kudos are due: The team has done a great job both with ongoing developments of the hardware and software (front end and back end), and the support has been timely and appropriate from many members of the hologram team (special shout-out to Reuben).

Whatever happened recently in your company comes as a nasty shock to all of us who were looking forward to the next step in the Dash development (Cat M1), which you had announced earlier this year, then withdrawn without explanation.

Worse, and borderline inexcusable, you have canceled the current product with no notice, and without giving your customers “one last chance” to buy a number of Dash units to allow us to continue to support our own programs while investigating the transition to another platform.

It would’ve been far better for Hologram to say something along the lines of, “We will not be continuing hardware or software development on the Dash. We plan to continue to sell the product for the next X months, or until current supplies are exhausted. Your current products will continue to operate (except for OTA). We are also investigating continued Dash supply via a third party.”

The open source design, while noble from your perspective, does little for the user base who are looking to buy a well-designed development board – I want to use the boards as the turn-key tool they’re meant to be, not manufacture them.

For those of us who have units in the field, we are at a dead end if we ever want to duplicate those systems or implement additional units. For those of us who were heading toward commercialization for a specialized product, this is a major setback, that, as mentioned above, should have been handled better.

Many of us regret the Dash was clearly not the hit you had hoped for. I wish everyone on the Hologram team luck and success with future products, and I hope you will take this note as constructive criticism on how to continue to support your loyal customers, who, like you, are also working on our contributions to the Internet of Things.

Sincerely,

MichaelM


#13

I agree with @MichaelM.

My prototypes are also heading towards commercialization this July… I can’t plainly say to client that main component is ‘discontinued’ and I need another 3 months for re-work.

This is not to mention that 3 months was also dumped because U260/U270 was discontinued, and the transition to U201 is not ‘smooth’.

Another reason why we developers can’t reproduce your open-source hardware is the ‘pre-certification’ your board already has. It’s too costly for developers to individually acquire certificates and it is heaven-sent when we found DASH.


#14

This is obviously a terrible, horrible, really bad day for small developers. My last hope is that Hologram will come out with a library, similar to SIMCOM, that will support Arduino boards such as the MKR GSM 1400.

So, Hologram, can we soon expect to see a library to support uBlox modems? One that mimics your cloud API that we’ve gotten familary with?


#15

I could not agree more with all of the above posts. For us “currently small” developers, discontinuing the Dash is no different than Hologram discontinuing cellular connectivity. We respected and admired what you were doing for IoT, and we appreciated the opportunity you were providing us to jump in head first! The value hologram brought to us was a unified software, hardware, and infrastructure platform. The most disappointing aspect of this decision is the fact that Hologram made no effort to communicate with the current Dash user community. You gave no consideration to our plans for the product and its unique capabilities before making this decision. It would have taken minimal effort to solicit our input via the Community forum. This had to be a well thought out strategy and you neglected to make us aware of what may be the fate of the Dash as you worked through the possible options. So the question remaining for all of us is…Why stay with Hologram at all? Is there a possibility to partner with a hardware supplier or continue manufacturing the Dash without any further development? I understand you have to prioritize the allocation of resources, but completely eliminating the Dash forces most of us to completely start over.


#16

Hi all,
Some thoughts (not necessarily official):
First off, really appreciate the call-out @MichaelM. Was always happy to help you guys with stuff and I remember getting that box of Dash button pushers from you guys back in the day. Super helpful before we switched to more sane buttons on the 1.1.

As one of the engineers who’s worked on the project since the beginning, I’m really happy to hear how much you all liked working with the Dash. I spent hours at our initial CM getting test equipment setup and dealing with problems on the 1.0 and then hours more when we transitioned to a new (much better) CM for the 1.1 and 1.2. Was always excited to get new features out to you guys and see things grow. The feedback here makes me think we did accomplish our goal of making a really useful IoT platform.

Of course, now all that is bittersweet since we’re winding it down. I definitely know how you all feel. I feel a sense of mourning right now myself.

I think there’s some good feedback here in this thread and we’ll definitely be discussing it at meetings this week.

We absolutely don’t want people’s future products to be left in the lurch and so I think one goal here is to help transition to other platforms. Someone made a suggestion about libraries for other platforms and I know @benstr and others have been working on these already and can probably point you to some.

If you’re already way down the road with the Dash and were on the verge of a big order, keep in mind that this is open source hardware and so it’s very possible to spin up a bunch of boards yourself. We can’t really manage that process anymore, but I think we can provide some tips. We have good companies we can refer to. It won’t be as simple as getting them from our store but maybe not too much worse. Of course, the economics might not be great for you, but that’s something you’ll have to take into account.
Feel free to start a thread about that or send over an email if you want more information.


#17

While this is going to cost me a bunch of time I just spent over the last 5 months designing in the Dash to our product, the main disappointment is the manner this has been handled. When I saw Dash 1.2 was open sourced I could see where this was all heading, but what I expected was Hologram to have alternative suppliers already in place (prepared to make for stock) before dropping it. And then also a commitment to at least contribute to the project in a more limited basis.

We only have 35 SIMs in the field right now, but if I was sitting on 500+ in the field I would be pretty nervous right now. Making snap decisions like this just creates fear that things are going bad.

I have 3x Dash of which 2 are unused (were just about to go into the field) so if anyone is desperate for them I can likely sell them to an interested party. We are not quite at the stage to be able to get a contract manufacturer to make 100’s for us so will have to take our design in a different direction - the Nova may even be viable.

Such a pity the Dash has to die, 6 months ago we looked extensively at options and the Dash was the best fit from the perspective of price, maturity, OTA and openess. There still is not a high level solution I am prepared to shift to so will have to go lower level in our design.


#18

I love the dash because of the very low power. We could wake it up, check the text and then put it to sleep in milliseconds.

The other major part of our design is with the Onion 2 based on openwrt but the SDK doesn’t install on the Onion properly for the Nova even with a 1G swap and 15G overlay.

If you guys at Hologram are able to get an Onion 2 install working with the Nova and python SDK, I would be very grateful to know the details. We have potentially 100’s of units round the world.

Thanks,


#19

I appreciate all the feedback and echo everything @reuben wrote above. We’re taking your feedback and discussing internally to determine any ways we can help.

For instance, I have good news to share about remaining Dash inventory! Our original plan was to offer remaining stock from our store for a limited time following the announcement. Although, as the announcement approached last week it became evident we would not have the inventory and so we removed the option from the announcement.

Through some unforeseen circumstances, I’m happy to share we’ve been able to secure nearly 100 Dash v1.2 units for purchase. We’ve prioritized turning on the Dash store page while supplies last. You should see the store reflect this change within 24 hours. I’ll also post the link here once available.

This is the last of our inventory. Orders will be first come first serve and limited up to 10 units per customer. If the order volume is high there is a possibility of overselling. In the case of an oversell we will refund the last orders placed.

We hope this will help some of you who are looking for a buffer while you find a viable alternative.

On a side note, have you seen the new Arduino MKR boards which are Cat-M1/NB capable? They look great and work wonderfully on our network. https://blog.arduino.cc/2018/05/12/the-mkr-family-gets-bigger-with-two-new-iot-boards/


#20

Annnnd it is live. https://hologram.io/store/hologram-global-dash


#21

@benstr,

MKR boards’ shortcomings:

  1. PinOut is based on Arduino UNO’s format… I like Hologram’s pinout better, the best.
  2. MKR has M0 only… Hologram’s M0 + M4 is a beast…
    With Hologram, I can do post-data acquisition processing in the IoT device then send processed (smaller) data to the network. With MKR, I can’t do complex computations due to limited memory space (like bit-banging)… and thus raw data needs to be transmitted to server and have it processed there.
  3. MKR’s 1 LED vs. Hologram’s 1 White + 1 RGB. Right now, I’m using Hologram’s RGB LED to indicate status of Network connectivity, logic/process failure indicators, etc. Now I’m back to blinking white led at different frequencies and repetitions (like morse-code) to indicate the different status… (sigh)

I’m seeing a high-tech version of Arduino in Hologram… and DASH’s hardware upgrades are keep getting better and better from version 1… and now I’m back to Arduino hardware’s “consistent-to-Uno”'s philosophy…

Hope I can get a DASH equivalent out there… (goes back to treasure hunting once again).
Feature wise, DASH is for engineers… Arduino and the likes are for hobbyist… IMHO.

BTW, if I decide to manufacture the DASH board myself, how will the ‘board pre-certification’ work? Is the certificate granted to the board itself and not to Hologram? I understand that SARA’s modem are pre-certified already… but Hologram also mentioned that their board (not SARA) is pre-certified so I’m thinking there’s a separate one that Hologram acquired for DASH?