I've not heard of Berg before, just having a quick look on their front page, shows its hardware to connect an embedded system to wifi to their cloud.
Isn't this a big worry if you're using anyone else's cloud service for your own embedded hardware.
Interestingly other platforms exist that perform similar functionality using the same WIFI chip Ti CC3000 (such as Spark Core). There seems to be a number of new chips from china which look to potentially give Ti, a run for its money, such as the ESP8266.
For those curious about the ESP8266, it's a 5$ Chinese wifi module which initially had limited/no English documentation. But now people are working on translating and testing the module[1]. Alternatives based on the TI chip from Adafruit and Sparkfun, for example, are about 35$ (with the chip by itself costing around 15$).
You can get modules even cheaper than $5 on certain sites, I think mine cost around $3 from Taobao, which makes it very tempting (the chip itself is even cheaper still).
Is it? My impression was that the ESP8266 can do maybe 56k level throughput, suitable for sensors but not for streaming meaningful amounts of data.
It is really really cheap though, which is great in itself and enables new applications.
Also the CC3000 is quite buggy, try using the newer CC3100/3200 if possible, pre-production silicon is already available.
I'm still waiting for mine to arrive, but I'm wondering if it _may_ be possible to get faster data rates without using the AT commands, and running code on the ESP8266 itself.
Ah yes that would be the holy grail, most BLE or Wifi chips have an MCU that could be used if their dev. environment weren't the software equivalent of unobtainium, require 500$ dev boards and / or be totally obscure.
If you find out anything, make sure to post it somewhere!
Note that "their cloud" (what they called Berg Cloud) was actually primarily a local wireless access solution, not a cloud in the cloud computing sense. It was supposed to make configuration and connection of the Little Printer and other future Berg devices easier. As other commenters have pointed out, you can have an LP-type device that connects directly to a normal wifi network (e.g. Adafruit's) and the content doesn't in principle need to come from a remote web service either: it could be provided by a script running on a local machine.
The quality of presentation and the user experience provided by Berg (not just this product but other things the same guys have worked on) is great, but they show their roots in the design school. The products feel to me more like the product of brainstorming at an ad agency than a mature response to necessity. London's design scene is full of this kind of "creative" activity, and it's super enjoyable to watch but probably a form of decline or decadence in terms of real manufacturing and craft. The pace of change is so great now in tech that designers tend to get left behind or reinvent the wheel.
Imagine Twitter redesigned by someone both technically knowledgeable _and_ UX aware. (No race conditions, URL shortening, or crippled API functions.) the reason we don't have this is a cultural gap between the hip designers and the technically informed.
No need to apologize... There is a remote element to the Little Printer service as I understand it, but they created their own concept when they invented to Berg Cloud.
Agreed - I think a lot of the activity there is a result of the RCA design program, and it's much needed to inject inspiration into the design-tech-culture space. But they're a distance from viable products, and (godblessem) more enamored with their own journey with craft than in getting costs down.
We just need some more intermingling and we'll get to where this is all pointing.
Sad to see that they're closing down, but understandably a nieche area of business.
The Little Printer was such a cute piece of tech, and I'm sure that they could've done a lot if not this fate was upon them.
Yep, we're opening up the relevant repos, and writing up documentation about how to deploy them, and I'm also arranging for a patch to the Bridge unit which allows you to repoint it at a different control URL.
Also, if I have time, I'm going to make this a shink-wrapped AMI to make deployment a little easier. There's sadly not a great deal of time left for this, though, so we'll see what I can get done!
I did a small exploratory project using it, and my takeaway was that it was a solution in search of a problem, at least the Cloud aspect, which did absolutely nothing useful, other than sparing you from setting up your own endpoints, but in exchange your had to deal with a really limited (and limiting) API and a more complex infrastructure.
I'd suggest "cloud-slaved". You're a slave to someone else's "cloud". They can turn you off at any time. They can even take away features from your purchased device. They can start charging for a previously free service. They can shove ads into your device. They can raise your price.
They can turn your device off. (The median life of cloud services is only a few years.) The cloud owner is the master. You are their slave.
I wondered when this might happen. You can't really rely on any of these devices with a 'cloud' based backend to work a couple of years after you've bought them (As anyone with a Nabaztag gathering dust in a corner will tell you).
It's really easy to get up and running and interfacing with it is simple. I have mine print out reminders for when to plant seeds for the garden and it also notifies me when automated downloads via FlexGet have finished.
Beyond the issues of relinquishing control over a service to a third party, I think another disturbing aspect is all the waste this causes: perfectly good electronics get thrown out just because some company goes out of business. The fact that these "cloud connected" devices are often proprietary and locked-down (for "security reasons") makes it worse.
Except the email I just got says if they can't find a way to keep the service running, they will open source it and open source the firmware on the cloud device.
This is possible, but rare. If your GPS navigator requires "cloud maps", it will be useless the moment its provider goes out of business or turns off the v1 servers in favour of v2 servers. This as already been discussed in the Ars Technica's article about Android and the problem of old phones with "broken" apps.
(Well, the reality is that it will be useless for 3 years, after which somebody will finally crack the DRM and put a OSM-based replacement on it that has 90% of the features of the previous software, but it nowhere as polished as it.)
> This is possible, but rare. If your GPS navigator requires "cloud maps", it will be useless the moment its provider goes out of business or turns off the v1 servers in favour of v2 servers.
I have a 10 year old GPS receiver in a box, with no way to update with maps from Garmin, not because Garmin went out of business but because they just choose not to support the device anymore.
Do I want to spend hours trying to figure out how to update a 10 year old device? Or would I rather spend $100-$200 on a brand new one with the latest maps?
Obsolescence isn't a bad thing, its how we move forward. We just need a clean process to recycle the waste that process generates.
The issue is that a GPS navigator is basically a solved problem, and the only part that requires updating are the maps.
Do I want to spend hours trying to figure out how to update a 10 year old device? Or would I rather spend $100-$200 on a brand new one with the latest maps?
The ideal situation would be a device that you buy once, but it's also one you could use with whatever map data sources you want (OSM, commercial, etc.) - based on an open format.
(I know GPS is also dependent on the satellites being available, but since it's government-owned and critical to many parts of the infrastructure, it's likely to stay around for the forseeable future.)
> The ideal situation would be a device that you buy once, but it's also one you could use with whatever map data sources you want (OSM, commercial, etc.) - based on an open format.
Like an iPhone/iPad/Android phone/Android tablet? Tomorrow's tech will be available to me soon, and much cheaper than today's tech. Support = people's time = expensive.
Chumbys still work (probably because bunnie seems benevolent). Even when its servers were turned off for a while, they still worked; you just couldn't choose a different clock face.
Genuinely curious, but why print reminders like that when you can use a calendar? Or using something like Pushover for notifications when a download has finished.
The reminders are only printed once a week on Saturday mornings and use maybe 5-10cm of paper. It's just nice to have a physical reminder to do things sometimes, something you can't ignore. Plus I get enough alerts on my phone already.
Like people have said about the Berg printer, it's really a solution looking for a problem. I just wanted to make a little networked printer and build a nice case for it (I'm doing it in the style of a piece of equipment from one of the Alien films). There's no real reason to have it, but I do, so I'm making use of it.
Great point. I think if you're launching a product like this you need to have a fallback strategy established from the beginning, or maybe just go open fromt he start so that the ecosystem isn't dependent on a single piece of locked-down infrastructure.
One of those things I always convinced myself I really wanted, but the fear of the cloud service going down put me off dropping the cash. £200 for a device that's useless if the business went under was not my cup of tea.
I'd like to think that charging £200 for a device that has little better use case than being cute and quirky played a part as well.
Especially considering that you can buy a professional POS thermal receipt printer for the same price for any viable projects requiring printing in that general form factor.
They were a leading speculative design studio, not another manufacturer of printers. Coming from a different place, more similar to how sci-fi stokes the fires. I'm shocked and sad that they went under given that they had been doing consulting for years. I wonder how taking VC last year affected this.
But not shocked that Little Printer or the Berg Cloud was commercially unsuccessful. Perhaps they just bet a little too much thinking that their excellent design and particular approach would translate to enough value for enough people.
A mindset, and growing recognition that startups are failure prone or targets of acqui-hire-and-shutdown. Disposable product-services are only good for those betting on the next (first/only?) quarterly earnings report.
I'm not surprised. I was really excited when I heard about this device and although it seemed like a bit of a novelty, I really wanted to try it out. Then I saw the price - $260. I understand you are paying for added value and well thought out design, but $260 was an insane price for a niche product like this
Square or similar should pick this stuff up. The personality the Little Printer exhibited was a refreshing (and for me, important) distinction. Not saying that they should put it out there 1:1, but there's a lot to learn in how this thing was made, branded and communicated.
For me, this wouldn't then just be a receipt printer -- you could just as well have it push the "opposite" direction. It's essentially a straight tunnel to a specific business or person from its maker. "Here's your daily summary", "Sara from London just asked about shipment costs". Push notifications to phones does that too, but sans this - for the lack of a better word - bond. For the right business, it feels like a very interesting point of difference.
There is a good usecase for devices like these: companies that place orders remotely with a large variety of suppliers. For instance, if you have a website where you allow people to order take-out food you could place one of your printers at their premises. No need to rely on their infrastructure at all, especially not if the device has it's own cell modem on board.
I spent a couple months working at a company that had almost that exact business model. We just used normal, cheap, thermal receipt printers attached to a Raspberry Pi.
I hadn't heard of this before but it doesn't seem dissimilar to the adafruit internet of things printer. No "cloud" to speak of - you consume web services via scripts run on the raspberry pi.
I'm not sure why we're surprised to see functionality disappear from online-tethered services when it's been happening in other arenas for years now - most visibly, online gaming. The oldest example I can come up with offhand (because it's the Dreamcast's birthday) is Phantasy Star Online. That game was fun, but really only as fun as it could be played online, and the servers have been off for a long time now. EA has started turning off servers for old games in the same way. This shutdown is a shame, but hardly unexpected, especially given how niche Little Printer was.
PSO was playable just fine offline and split-screen however. And if anyone is interested SCHTSERV[1] is a PSO private server that is still up. They have gamecube, dreamcast, and PC clients.
Yea I've been there, the local co-op was a ton of fun really. The only downside was that the framerate drops a bit if you have more than 3(2?) players.
Thanks!, I've got both gamecube and dreamcast with PSO and always wanted to be able to play them online for the full experience but didn't own them early enough to do so.
I think the biggest problem is that they assumed that people would like to have this disposable bite-sized information internet spews at us printed on physical medium instead of being displayed in some corner of a screen somewhere where it belongs. The kind of stuff that will be irrelevant after you've read it once. Not economic. Cute thing though. I give you that.
While I enjoy a spot of reverse engineering as much as the next hacker, it's probably easier just to ask them for the protocols in this case; try http://twitter.com/nickludlam
Crap -- I have a Little Printer still in the box that I've been meaning to play with. Really hope they open source the protocol so I don't end up with a Little Paperweight.
This is what really hit me one day. I had things in the box unopened from companies that no longer existed. That really put me off buying cute 'toys' unless I could commit to using them and exploring them in 90 days.
Berg has for a while explored speculative product experiences in the vein of Dunne & Raby (1) - though being design consultants, Berg's work tends to be optimistic. D&R embrace the dystopian aspects of future products, and I think they'd enjoy this conversation that's cropped up around what happens when connected objects lose their parents. (Especially when the objects are personified!) Hopefully Berg appreciates their inadvertent final contribution as well.
Does the Internet of Things need a federated model to be as trustworthy and successful as Internet classic?
Isn't this a big worry if you're using anyone else's cloud service for your own embedded hardware.
Interestingly other platforms exist that perform similar functionality using the same WIFI chip Ti CC3000 (such as Spark Core). There seems to be a number of new chips from china which look to potentially give Ti, a run for its money, such as the ESP8266.
reply