Funding Espruino

It's been suggested a few times recently that I should have a way for people to give donations on espruino.com.

I used to have one when Espruino started, and I think it got used once in the whole year so I took it off. Recently things have changed quite a bit though. Currently around 1/3 of all users are not using the Espruino or Pico boards, and that number is growing all the time - especially with things like the ESP8266 port.

While I could sell programmed ESP8266's, at the moment I'd rather not have to support them (and realistically most people will just buy a NodeMCU clone off AliExpress which I could never compete with).

Folks like Marijn Haverbeke (of CodeMirror/Tern/etc fame) have a whole donations thing, and companies like Ubuntu have a click-through screen every time you download (which you're basically only going to be doing if you don't have an Espruino board).

Any thoughts on the best way to approach this?

A PayPal button is probably easiest, but then there are things like Patreon as well... has anyone done this before? any advice?

Back in 2011 I wrote one of the first cookies disclosure scripts. That has a "buy me a beer" button. It works reasonably well, I think partly because it sort of quantifies the amount you hope for as a donation. I don't get lots of donations these days - it's old hat and there are better solutions - but when I do I get just enough to cover a beer - maybe two. I think without that people would wonder how much to donate and would worry about it being too little/insulting. And we all know that where there's a decision to be made doing nothing is often easier!

Do you have any ideas for new projects? How are the boys over at smart.js (https://smartjs.io/) funding their projects?

It seems to me (as an outsider in the IoT world) that there are many burgeoning companies that are setting themselves up as IoT and Cloud based technologies. While I personally "like" tinkering with home automation thingies ... I suspect that the winner will be some industry broad standard with which "appliances" will embrace and come pre-shipped/enabled ... as opposed to smart plugs and the like.

From where I sit, I'm not sure that a revenue stream on Espruino boards is going to necessarily grow into the future. I base that on the notion that the cost of boards is tumbling and the availability of environments for programmers to write code upon is increasing. Combine that with the moore's law of technology increase and "secret sauce" of an embedded custom written JavaScript may be displaced by "commodity" JavaScript runtimes once the base board has RAM and CPU to accommodate. For example, in my Java world, the latest Java 8 has a first class JavaScript engine built into the JAVA JVM ... so one can write JavaScript AND Java running on a JVM ... and since those JVMs are being ported all over the place ... it drags a common JavaScript environment with it.

In my work life, I tend to think about what can I do to delight customers NEXT ... so the challenge is always to look ahead 6 months, 1 year and 5 years and attempt to predict what they might want based upon all the input I hear. Many times though I don't LIKE what I hear because it is divergent from what we thought the direction was going to be or our competition are stronger in those areas. It is then that we have to be smarter and go to more heroic lengths ... and sometimes that takes us in directions we hadn't anticipated.

At those times, I get a large cup of coffee and re-read "Who moved my cheese?".

@Ollie thanks! I think for individuals that's a really good idea...

Thing is, it's becoming apparent that companies are starting to put Espruino in products that they're selling - for them, one beer doesn't really seem that fair. I guess there could be a prod for a beer, and an option for more.

Marijn (linked above) sells a support contract where he says he'll look at any reported issues within the week for a certain monthly fee - although usually I try and do that anyway :)

How are the boys over at smart.js (smartjs.io/) funding their projects?

Like most IoT companies, someone gave them a big pot of cash... Something I'm trying to avoid.

This shouldn't really be a discussion about the future of Espruino, but personally I think it's got a good future. Sure, there are new boards, IDEs, etc cropping up weekly at the moment, but most haven't had (and won't have) the attention and support that Espruino has.

... and that shows - when you get a proper board, plug it in, and things just work.

Blimey no, I wasn't suggesting you ask for a beer @Gordon! Just that you try quantify what might constitute a donation if you decide to put the button back! I think it makes it easier for people.

Re corporate, well that's a different matter - FWIW there are a lot of corporates using my script (some that are direct competitors to me) that have never donated a bean! But this is open source, where we hope the benefits of altruism outweigh the bad bits. But you can't depend on donations there.

So for corporate then:-

1. Commercial licensing
   
2. Paid support contracts
   
3. Enterprise versions
   
   

Re 3. they seem to come in two flavours - with private features that the open source community never see, or as seems to be a trend, Enterprise gets your latest version - you stay totally open source but retard the open source free version - so for Espruino "free" might be 1.75 while Enterprise "paid for" is able to access v1.81

and that shows - when you get a proper board, plug it in, and things just work.

The plug and play "ronseal" aspect of the Pico and original Espruino board is a massive plus - I guess its the price point of the cheaper boards such as the ESP8266 that attract. Your commercials are your own business but what you seem to be up against is cheap and increasingly powerful microcontrollers :/

I'd be open to either one-time or (probably better for all involved - smaller up-front number for donor, and it's a more stable stream of income for Gordon) monthly donation of some sort. I can also see potential in Gordon offering paid support deals (though his support is so vigorous now anyway, that I don't know if that would be enough of a draw - unless this was targeted at commercial customers who are using Espruino in products).

What draws people to the ESP8266 is that it's wifi, and it's cheap - it's very appealing for someone who wants small projects. Once they're using the ESP8266 as a wifi adapter, it immediately begs the question - can I just lose the other microcontroller and use just the ESP?

Thanks - in a way it's a bit difficult asking here, because everyone who is answering is already contributing a huge amount, and I'd feel bad taking your money :)

It's actually more the companies that are using Espruino on other devices that then want free support. But then that's a bit difficult as well - those people will just neglect to mention that they're using some other board when they ask questions on the forum.

Delaying builds would work, but I don't want to screw over 'normal' Espruino users (and for instance @DrAzzy provides nightly builds himself) - so it's a hard balancing act. Paid support could be a good one - specifically some companies e-mail direct because they don't want to post publicly - I could offer private support as a service and just direct people to the forum otherwise.

Espruino is also getting used more in teaching... That's great, but if someone's running a workshop with people paying £100 a pop to use $10 Nucleo boards with Espruino, it'd be nice if they at least considered making a contribution...

The sad thing with ESP8266 is that looking at it, I'm not sure it's worth me selling them. I can't charge that much more for them, and once I've programmed them, packed them up, 5% of them have got lost in the post and I have to support the rest of them, it won't actually be worth my time!

... anyway, I'm getting sidetracked. How would you actually take payment? PayPal seems like the obvious option, but there are a whole bunch of other things (like Patreon) that specialise in this kind of thing and I wonder if they're somehow better?

Well the Paypal fees hurt a bit - especially if you're receiving a significant amount - not referring to donations here, something else. No experience with Patreon, doesn't look very "enterprisey". Another alternative might be a do via a direct debit bureau? Gives you plenty of control should license or support contract terms need to be varied and pricing altered.

One other thought - what about commercialising ease of access/programming - a cloud based web ide - much as you have now. Your teachers will use that - just because it's plug and play and requires no set up - if you can get it to that point - think Office 365 style?

But more than that use-case, it would also allow options to restrict/open-up functionality based on the type of subscription to the online IDE.

For example:-
Free - no code minification, maybe limit access to modules to those a hobbyist might need.
Enterprise - minification, all the modules, paid for modules, maybe pre-built code solutions.

You could go further too. Perhaps allow third party developed modules and code solutions in a "marketplace" section of the IDE that could just be purchased and loaded up by anyone. Espruino takes a % of the third party developer' sales.

An online IDE could be a hub that ties the whole Espruino ecosystem together.

Bit blue sky, and certainly for the long-term, but there are plenty of ways to take the skin off a cat!

Hmm... good point about Patreon... PayPal doesn't look that great either - but I guess businesses are using it more now. There is also a kind of 'PayPal direct debit' thing you can do too, which might be easier to administer than doing it direct. As @DrAzzy said, having a monthly contribution might work out well for a lot of people.

PayPal fees are a pain, but as far as I can tell are actually not that much worse than other credit card handling companies. I think Stripe takes around 5% (could be wrong though).

I really don't want to restrict very much because arbitrary restrictions in Toolchains was something that drove me to do Espruino in the first place. The Espruino boards themselves do have more features (TV out for example) that are too painful to support on other boards, and maybe I should be a bit more aggressive about stripping features out - I spend far too much time trying to make sure things still fit into the Olimexino board!

... but having said that, if I strip it down too much then people will just put up their own builds :)

A marketplace for modules sounds interesting - I wonder if it would convince more people to contribute?

I think for now, probably a donation, and maybe a line after the

 _____                 _ 
|   __|___ ___ ___ _ _|_|___ ___ 
|   __|_ -| . |  _| | | |   | . |
|_____|___|  _|_| |___| |_|_|___|
          |_| http://espruino.com

screen on the non-official boards.

Stripe is what most people seem to use these days for straight payments - has a native JavaScript API and fees are neither that high nor full of hidden charges.

Personally, a paypal button would appeal most to me. For most things you pay for online, they'll have a nice easy paypal button that's almost automatic to click before even thinking about the other payment options.

A couple of things that might help, based on the conversations above... People will usually prefer to use something official and simple, especially if they're not as savvy with compiling or dealing with binaries. If you focus on getting your boards set with the best features, especially exclusive things like the bootloader, tv, IDE support, and market these advantages, then a large subset of buyers would rather buy the board that it all comes built into, not the one that they'll have to work on a bunch to get some features working, or use a third party binary. Not all of us are comfortable jumping intoa world of github and makefiles!

Also, I've got a Nucleo board at home (a pico too!), but I shouldn't expect you to tiptoe around breaking anything on it while you're releasing new builds. As long as I've got a build that works on it that I can revert to, I don't see any reason why you shouldn't be aggressive with stripping out features or breaking things for my third party board while you're improving the espruino boards. And even if someone does put the effort into doing their own builds, I'd still prefer to use official ones and that'd be a good incentive to by an official board.

Also, if people are frequently using espruino in bulk for classes our products, it might help to market your bulk discounts more. Maybe even offer a specific pack of 20 or something, with a reminder of the advantages of using an official board.

Thanks! Yep, I do offer bulk discounts (even via Tindie), but I should advertise them more as you say... Also I'll try and reintroduce a starter kit soon, and perhaps a 'hack day kit' with a kit of parts for tinkering.

Yeah - it might be good to try to talk with some people who've used Espruino for teaching - if you can find them - and see what they'd want in a kit. Education is definitely a good place for Espruino. The learning curve is definitely a lot less steep than other microcontroller platforms, yet it's powerful enough to do real stuff. The console itself makes just such a difference in lowering the barrier to entry. And it's javascript, which is an incredibly useful language...

Hi Gordon! I just ordered another Pico and an adapter board via Tindie, so I thought it a good moment to comment on the funding issue. I am probably not representative of the main contributors here (~70 retired from a UNIC/C environment), but am fascinated by the idea of having Espruino for embedded projects. I am buying the extra Pico to support Espruino development work. For me something more like the form factor of the original Espruino would be better, but I wanted the ram and flash of the Pico. I am hoping the adapter board will help me move beyond a protoboard with the Pico, but the 0.05" connectors are a big hurdle for me. Perhaps selling a "pinned" board with the smaller connector pinned as well would be possible?

I suspect getting funding for a software-only product is difficult, so maybe a relationship with the Nucleo folks (I really like Espruino on the Nucleo F411RE) would be an option? I know they have mbed, but a few minutes playing with Espruino should convince them that it would be worth supporting development for their platform.

As far as size goes, I think small beyond a certain point is not valuable unless the price can become trivial (ESP8266). On the other hand, wireless communication is very desirable (not necessarily wifi due to power requirements).

Hope these observations are useful, and keep up the good work.
Duane

Hi Duane, thanks!

Yes, the 0.05" is a bit of a pain... at least with the adaptor board you can only solder them once and convert them to 0.1" though! The problem with soldering headers is you then have to have solder 0.05" stuff to connect them to as well!

ST do seem to be quite interested in Espruino - I got invited to display at their stall at Maker Faire Rome, and they provided STM32s for the first batch of Espruino Picos in return for me porting to the Nucleo! I'm not sure I'll ever get any support payments from them as at the moment they basically have all they want for free, but it would be great if I did. After all, they pay quite a lot to mbed to be able to mbed-enable their boards.

I really want to create a smaller, cheaper Espruino board, but you're right, I don't think it would add much. Also, at some point I have to sell the board so cheap that I make so little that I can't afford to provide the amount of support I'd like.

I'm really interested in the nRF51/52 at the moment and I'll be supporting it at least a bit for the micro:bit, so I'm thinking I might do a dev board for that at some point in the future. Hopefully it could really add something as it definitely plays to the 'low power' side of Espruino.

If not ST, how about other board manufacturers? I'm thinking specifically of kickstarters for wifi capable stm32 boards, like the wifimcu, or the currently running Redbear Duo one. A lot of these kickstarters are trying to launch new hardware, but then market all about their software. I'm sure that some would be up for commissioning official espruino compatibility (especially if already using a compatible chip, but they just need peripheral support and a badge saying it's official).

I hadn't heard of the Redbear Duo before, but I see:

JavaScript - Yes. If you know JavaScript already, you are ready to go! We are porting an open source JavaScript interpreter for microcontroller to the Duo

I'd be interested to see which one - I guess it could be Espruino already, given it's an ST chip? Although there's enough RAM that it might not be.

RedBear contacted me just over a year ago about using Espruino:

Just want to inform you that we are porting your code to an ARM M4 platform and if successful, we will sell a board using the ported codes.

So I guess this could be it. At the time I asked if they'd contribute something to help support the project (in return for me listing the board on the site and helping to support their users), and it went quiet soon after.

And I just realised - the copyright case I'm fighting at the moment is with Beijing Red Fox (a translation from the legal firm) from Sept. 12, 2014 - one week before I was contacted by Red Bear Labs. It could well just be coincidence at the moment though, especially as Red Bear seem to be based in Hong Kong - I've asked the legal firm for confirmation.

... so basically, I have tried to get some manufacturers to support it and will continue to do so. The problem is really that very few people value the actual software (or the documentation/support) - and because it's Open Source many companies think: "Why should I pay for this if my competitors can use it for free?" (that's an actual email I got).

About all I've got that I can convince them with is the fact that I own the copyright and control the website - so can choose whether I advertise their product or not.

Of course the other side of this is say I did manage to get a $1/board royalty in return for providing support on the website. I don't think that would actually pay for the time I spend - I'd have to start providing significantly less support per user if that were the case.

Hi Gordon,

I saw espruino in a presentation about IoT recently and ordered it as nice "toy" for coding nights where one can have visual success with even basic javascript knowledge and a "bare pico".
In the past, I used a USB-traffic light as some visual bonus, but this had a limited API and no "on board" intelligence.
So I'm not the typical user of your products because I do not make any money with it.

Located in Frankfurt/Germany, it's hard to get a pico/espruino via internet.

To order a pinned-only pico, there is a swiss link (billed in SFR, customs problems for EU-citizens).
There is a german vendor for unpinned-only PICOs with a nice bundle offer.
But to buy an espruino, I have to contact another vendor.

Your Tindie bills in USD and there is a long shipment distance from UK to Germany; perhaps some amendments like billing in EUR and pay by credit card would help to order directly from you (to increase your profit, as you do an excellent support job, so you should get the €s).

Greetings, Uli

P.S: On the other hand, I consider to contribute whenever you launch another interesting iteration (probably with some IBAN-id/credit card option cause I'm not a friend of paypal).

Hi Uli,

Thanks! Yes, getting stuff locally is a complete pain. I'm actually talking to Watterott in Germany about getting Picos stocked at the moment, so hopefully that will be an option for you in a month or so.

I've been pushing Tindie to support different currencies too - but I'm not really sure when that will happen :( Being in the UK and charging in US$ is a real pain!

@Gordon - wow, sounds like some pretty significant challenges to getting funding through third party hardware, and that's not even considering the messy stuff like deliverables and legalities. I suppose you'd need to land a spot in a seriously large quantity or high margin project to make it worth the time and trouble.

Yes, if I could get it into something with large-ish volumes that'd be better.

Support from the actual chip manufacturers would be perfect (like mbed does) - then I can focus on getting Espruino used as much as possible without worrying about licensing.

Having heard back from the lawyers, the company that registered the Espruino trademark for China is http://www.redfoxchina.com/. Apart from the 'red' bit and the timing, it seems totally unrelated to RedBear.

My guess is that around the Pico KickStarter there was a bit of a buzz about Espruino, and several people decided to do things with it - Red Fox China decided they could make a bit of money by grabbing the trademark.