Adam Leeb of Freewrite explains why Freewrite is so expensive

[anon63581570]

Regarding price, the majority of the product costs come from three things, the screen, the mechanical keyboard and the ‘computer’ required to run the screen. The rest of it is small stuff that all adds up. Yes, we make money on the Freewrite but no the margins that are required to sell in retail, which is why we sell direct. Then there is the whole cost of development that needs to be amortized and I can tell you that we are still far from recouping our entire investment in the development. And we ran a very lean approach.

By using E Ink, we were basically forced into a microscopic ecosystem of components that are able to drive it. Hence the cost. The chip we use is the same that is used in the Kindle Paperwhite which is more expensive than the entire compute module at cost. And we aren’t buying at Amazon quantities either. It’s a tough one. Should we have dropped E Ink which would have given us a lot more options? Maybe but I really do love E Ink for this application!

Instead of pulling the Freewrite in a completely new direction, I would very much like to create a new product that responds to what you all want. Whether it is a Bluetooth mechanical keyboard with some additional writing features or a new dedicated writing tool, it’s still unclear what would have the most interest. Either of those things could be done. We have actually been working on some stuff internally around both of those areas but they are very much works in progress.

So not only is eink itself expensive but apparently, the components to allow driving of eink displays seem to be expensive (as of the design phase of the freewrite.) I’m not a fan of the freewrite’s design aesthetics but they are openly communicating so I’m trying to find ways to help as best I can, which lead me to posting on the pi subreddit asking if the pi zero or compute module could serve as the platform for a successor device. It’s a niche that is very not mainstream, but deserves to be filled.

[fuzzyfungus]

It’s a pity that (to the best of my knowledge) the displays used in the OLPC never really became available and are unlikely to be something you could get without a rather substantial minimum order. Those had excellent paper-like performance; but with exactly the same drive requirements and behavior as other LCDs, because they were LCDs(just transreflective rather than backlit, most of the time).

E-ink has weird drive requirements(fairly high voltages, and fiddly control of those if you want remotely adequate refresh rated and grayscale reproduction); so you can’t just drop one in where basically any cheapo SoC ever expects a DSI panel. It probably doesn’t help that the people who are buying e-ink in quantity are mostly concerned about page-at-a-time refreshes for display of static stuff, not interactive display use.

Not sure if anyone makes character-based displays of sufficient size anymore, those would also suit; but, again, not if they are a custom item just for you. The ones you normally see are the 4x20 ones which wouldn’t cut it.

More broadly, even if they can get the hardware they want, I’m not sure what their best option is in terms of balancing ‘focus on writing’ and ‘features that people are going to demand’. If memory serves, @beschizza 's rather scathing review had far too many mentions of ‘firmware’ for something that aspires to be a typewriter-like. On the other hand, you aren’t going to get document backup to the cloud and such without firmware (of nontrivial complexity and likely in need of periodic updates to cope with changes to the state of the world, even if it is bug free the first time). That is tricky. My experience has been that devices with a substantial amount of computer need to either hide it extremely elegantly or not at all: if you can’t poke at it it has to work like magic; if it doesn’t work like magic you at least need access to poke at it(see also, trying to figure out something like wonky Exchange calendar behavior on a cellphone vs. a full computer). I remember this being a consideration back in the alphasmart days as well: the ‘classic’ ones obviously ran some sort of firmware; but it was pretty close to invisible, albeit not terribly full featured. The Dana was clearly classic PalmOS(with its already non-legendary reliability augmented by running on a display/aspect ratio more or less unique to that one device, which gave software that made assumptions pause). The Dana was undoubtedly more capable; but the classics “faded into the background” much more easily: they powered on in an instant, ran for ages, and pretty much just worked, for what little they did. The Dana too often reminded you that you were actually babying a palm pilot.

I don’t really know how best they can approach that issue: the alphasmart’s level of stolidly doing nothing except emulating a keyboard would probably be a hard sell for people who don’t necessarily have a PC to transfer to on a routine basis(plus, if it is enough, they probably already have an alphasmart); but adding BT, wifi, interaction with cloud entities means more complexity, more potential wonky bugs(or unsupported services you can’t use/services orphaned when their API changes, devices that disagree with your understanding of BT HID, etc.)

It’s also tricky because, with their use of relatively exotic(by tablet standards) hardware and low volumes, if they try to go too full featured they pretty quickly bump into “maybe you just need a Chromebook with a block list?”; but cutting down too far will leave everyone whining about that missing something. Not sure how they should balance that; and that is probably ultimately the hardest problem. They’ll never be as cheap as the mass market stuff going custom; but they have a shot at being cheap enough through sheer incremental advance; they don’t have a shot at being conceptually sound through sheer incremental advance.

[anon63581570]

https://www.flickr.com/groups/alphasmart/discuss/72157690451506394/

One of the group users basically asked "Take your alphasmart, give it and $250 to a guy. What are you willing to spend that money on.’

My answer is one that I’ve posted both here and there as well as my own blog in the past.

Visual reference. Not shown are fold out legs much like with standard keyboards. Also keyboard would need a diff keyboard so that you can hotswap between files much like the neo.

I want it to do the following things:
Word Processing. Plug into the usb on something else either to transfer files out, hit a button so it ‘types’ out whatever’s open into the device it’s plugged into (that sees it as a keyboard and storage device,)
Able to read epub docs since ‘why not’.
An SD card slot (full sized SD card mind you.)

Optionally include:
Wifi so you can sync to a central server (include Windows, OSX, and Linux solutions so you can host the server on your own network in addition to google/dropbox/etc web synching) for document syncing not to a ‘cloud’ but to a local machine that would either be the teacher’s personal box, or for those at home a dedicated backup device so you have your copy of device, copy on SD card, and off-site copy. I cannot count how often I have lost work and wished a backup existed or finding backups I’d forgotten about years later.

The other thing is I want it user serviceable. Something I could go in and replace the battery on, pop the screen out if need be, pop keys off to clean the keyboard. I’m tired of sealed black boxes. The problem is sealed black boxes are easier to make, easier to sell people on the sex appeal of, and easier to convince people to trade up in two years when the battery that can’t be replaced gives out for the new shinier model.

Market it as a cheap durable alternative to Chromebooks. There is literally no software to need updating or get infected or the like. If you can make them cost $200 or lower they’ll have a market. I’m not sure you can do that, but it’d be nice. Epub support would be good for distributing texts to students or staff that you want easily read and notes taken on, but not editable on its own.

So I dunno I’m guessing, in the end, something that could sorta double as an e-reader but I’m unsure… This has less a book form-factor and more ‘keyboard’ form-factor. The keyboard NEEDS to be full-size though. I will not compromise on that.

However that image? that is what I am looking for.

I had ideas about being able to play music/podcasts/whatever maybe with dedicated media keys, but that’s a distraction innit? Could potentially be useful but eeehhh.

As for screen go with a pure black/white LCD for the sake of contrast and power usage. matt display facing for the sake of keeping glare to a minimum.

[beschizza]

I believe you’re remembering this from @jlw, not me: https://boingboing.net/2017/10/24/everything-ive-written-on-my.html

[jlw]

The Edsel is still sitting here.

[beschizza]

something similar to @singletona082’s suggestion: https://boingboing.net/2015/06/25/is-alphasmart-the-ultimate-wri.html

A standard display tech would be fine given e-ink’s ecosystem nightmares, if we’re talking a practical product rather than dreams and fancies