Hi from Ergofip!

[olivierpons]

Hi,

I'm Olivier Pons. Typist since I'm 8. Geek (not my choice but life is life...).
I'm sorry for my english which is perfectible.
Here's what I've done:
- bought the Ergodox
- bought the Infinity Keyboard
- bought the Infinity Ergodox
- bought a lot of other keyboards, including "Ornata Chroma" to test the Razer configurator program
Since I wasn't happy for a lot of things, I've decided to build one.
I've spent more than 3 months trying to find someone who believes in this project: I finally found Damien Vidal, passionate about electronics and embedded systems. Then I've spent almost one year trying to convince him to finish the RGB prototype. He did it 6 months ago.
Then I've spent more than 3 months to convince a sales and communication director, working in a huge company, to work with me.
Then I've spent 2 weeks and 3 week end to make the movie with Jean-Romain from Bonjour Studios.
And last but not least: I've spent one month, every night, trying to get back my wife and my two children, because this were very hard times for the family.

And here's the result:
A 100% customizable split international keyboard with mechanical keys and RGB lighting.
- inspired by the Ergodox.
- Fully open-source (GPL v3)
- Totally re-written from scratch and still GLP-licenced
- Incredibly simple but powerful online international configurator

We've made an incredible good product.
I'm very proud of it and Damien, an electronics engineer specializing in the industrialization process, has already been involved in the mass-production of many products so we are 100% sure to go mass production if we get the funds.

A lot of people say that it's the Ergodox with RGB so I'll make this clear:
- It's totally re-written from scratch because you have no choice if you want to handle RGB leds
- It's still GLP-licenced
- It has 8 keys for the thumbs
- It's international and so is the configurator.
- You can configure that keyboard even if you have no programming knowledge (that's a huge difference)
- It's not a kit: it's a keyboard, pre-built.
- 2 years guarantee.

These were some of the differences with the Ergodox, but there are more.

Ergofip has actually only 2 things in common with the Ergodox:
- split keyboard
- 80% of the design

[reaper]

Welcome and thanks for the info. 

[Findecanor]

To be honest, I can't really understand what all the symbols on the keyboard mean.
Monkey, smiley face, Grandma (?), Darth Vader (?)... and a few that are unintelligible to me.

Are there any parts that are compatible with the mainline ErgoDox, for those who want to customize it?
Are there notches in a plate for those who want to open Cherry MX switches?
Any possibility of modifying it to use larger thumbkeys?

So, if the firmware is Open Source, is the hardware design licensed under a Open Hardware license?

Edit: I would like to raise a middle finger to whoever designed your web site. It is seriously annoying and difficult to navigate and it makes you look like a sham!

[dantan]

I'm inclined to think very positively of this effort, but surprisingly with 3 days to go it hasn't even netted 5% of its 30000 Euro goal.

[olivierpons]

Yes. Fail. $349 is very expensive for a keyboard. People needed more videos, more "tangible" proofs.
Anyway I've learned a lot.

Since then, I have a lot of contacts, 16k+ on LinkedIn especially in the gaming, and video industry.
But videos and tests are important. My keyboard is a prototype and I can't afford to make ~4-5 prototypes for people to test (I guess that's what it takes for a keyboard to become famous).

One year and a half redesigning everything but the shape, testing, managing people, making film and so on, for nothing.

If you read the thread, one guy here is probably happy of this fail (even though I still find the attitude of being happy of someone else's troubles despicable) ....

What's left? The configurator I've done is very easy to use and I'm very proud of it. The website is nice and I'm proud, too.

So my Ergofip adventure stops here (even thoug yesterday two of my students wanted to buy one because after a lot of search they realized I wasn't lying about the fact that this keyboard is actually the only one in the world like that).

[out of topic]
Here's my next step: buy the Infinity Ergodox again (yes, again, because the first one has contact failure which makes the right LCD not working properly. I did not send it back, but I don't use it, $300 thrown away (it's one of the many reasons why I've started the Ergofip adventure: the quality)), and change a lot of the program so I can show everything I want on each layer.
Here's my point of view: if you buy an Ergodox, it's cool. What could make me want another one? The LCD, right? So the LCD should be 100% customizable, otherwise there's no point of buying the Infinity Ergodox. On my Ergodox I have 6 differents layers: a "classical" layer for writing, a layer when I photoshop, another one in "QWERTY" mode, another one for trackmania, another one which has only 6 active keys for my son and so on. I call them 6 differents "fixed" layers: they are layers you stay forever.

On the Infinity Ergodox, here's the way it works:
- you have the layer 0 on which you're supposed to stay "forever" and access only other ones "temporarily"
- if I say that it's because there's only one "fixed" picture on the layer 0 that is displayed and the LCD is separated in 4 parts that show the numbers of the other layers when you show them.

Example: you are on layer "0", it has a nice picture where you can read "Olivier Pons's Ergodox" <- amazing.
You press "SHIFT" -> the picture disappear, and the number "1" appears on the first "part" of the LCD. Only that.
With "SHIFT" pressed, you press "CTRL" to go to another layer -> the number "2" appears on the second "part" of the LCD. So you see "12" written on the LCD.

Seriously, who wants that?

I want something so much simpler: I just want to display a whole picture on the LCD to show me on which layer I am!

So the only last thing I'll do about keyboard (until I'm motivated again) is to order another Infinity Ergodox, analyse the whole open-source program, fork it and change a lot of the LCD - side to make it work the way I want.

[/out of topic]

[Findecanor]

If you read the thread, one guy here is probably happy of this fail (even though I still find the attitude of being happy of someone else's troubles despicable) ....

If you mean me, then you have completely misunderstood me.
I was just saying that your website was crap. Then take that as constructive criticism to do something about it so that you would appear more professional to prospective customers!

But I can't really tell if you mean *this* thread or some other thread on a crowdfunding site or elsewhere, because there aren't any links left in this thread.

[thurst0n]

>> I've spent one month, every night, trying to get back my wife and my two children, because this were very hard times for the family.

Can we please get a follow up on this part of the post?

[olivierpons]

@thurstOn Thanks for asking! Those hard times are gone.
Now I'm trying to make money to get a better life because the Ergofip cost me around $1000 / month for 16 months (I count all the days where I didn't work for money but for the Ergofip), and I'm broke... and disenchanted.

[dantan]

@thurstOn Thanks for asking! Those hard times are gone.
Now I'm trying to make money to get a better life because the Ergofip cost me around $1000 / month for 16 months (I count all the days where I didn't work for money but for the Ergofip), and I'm broke... and disenchanted.

I was still ready to keep an open mind about the Ergoflip, but this statement was the straw that broke the camel's back.

Doesn't matter how honest your project was. If you basically took 80% from the Ergodox and all you did was design the PCB to handle LEDs, get new keycaps from Maxkeyboard, and code a better configurator, that cannot be $16000 spent in development. For that kind of money you would have many prototypes, photos of you in China at the factory, quotes from factories, 3D mockups of layouts that are better than Ergodox, samples of better keycaps that could be bought if you met MOQ, etc.

If you need to run up $16000 in costs over 16 months for so little done, you're not just undiplomatic when you didn't credit the ergodox and infinity ergodox. You're a downright bad businessman and bad project manager. That crowdfunding project would have been doomed to failure in your hands.

BTW I am not posting on that Datahand II thread anymore because it doesn't realistically help get the Datahand II made. But as this thread increases in length and more conversations take place, it will help everyone see how problemmatic the OP is.

[nmur]

i don't see any actual improvement over any of the existing ergodox offerings, besides the intense RGB LEDs, and the benefit there is really subjective. despite that, this is still the most expensive. you can get an ergodox EZ bundle with wrist rests and tenting kit (tenting for an ergodox is almost necessary for any real ergonomic benefit), which is ready to ship right now from a company with great customer service. '

the ergodox market is honestly saturated when you consider its degree of niche. something that's 5% different from the current ergodoxes just can't target any significant market

[olivierpons]

Doesn't matter how honest your project was. If you basically took 80% from the Ergodox and all you did was design the PCB to handle LEDs, get new keycaps from Maxkeyboard, and code a better configurator, that cannot be $16000 spent in development.

We took the PCB 3D schema, removed everything but 10-16 points (I'm prettty sure you don't know what I'm talking about because, when I read all your your posts in the community, you've never done something valuable with your hand)

I probably guess (guessing and not verifying seems what could describe you the best) that you've never done programming in your life.
And maybe if you've typed more than "main() {}", you've never done web development, otherwise you would not speak like you do: just try to do DragNDrop without using HTML5 facilities and you'll see the top of the iceberg (here, too, I guess you dont know what ECMA6 is, and dont know about HTML5, and JavaScript at all).

Before making totally false assumptions, please read for once in your life what is said about the Ergodox on the website. It has been there for 8 months:
https://www.ergofip.com/en/

"Ergofip is an open-source (GPL v3) keyboard initiated by Olivier Pons, after being inspired by the ErgoDox project."
"To be able to implement RGB Leds, USB 3.0 and the HUB behind, everything had to be made from scratch.
Even though some people may think the Ergofip looks like the ErgoDox, it's technically very different."

For that kind of money you would have many prototypes, photos of you in China at the factory, quotes from factories, 3D mockups of layouts that are better than Ergodox, samples of better keycaps that could be bought if you met MOQ, etc.is.

You already know the result with $600000 and after more than one year here: http://blog.keyboard.io/?

If you need to run up $16000 in costs over 16 months for so little done,

I've stopped read there. You're so stubborn and voluntary blind that it's just a loss of time speaking with you, and of course like all the questions I've asked, you wont reply to:
- give me a link to the keyoard.io configurator
- give me a link to the keyoard.io working prototype

With $16000 I've done far more than you'll ever could in your whole life.

You'll never answer to my questions because you know the answers and you know it would put you in an uncomfortable position.  )

Dont answer here too please, the readers will make up their minds reading your insulting and aggressive posts, the way you talk is enough to imagine all what you say is totally false and/or racist. If you hate French or an adept of the KKK then go somewhere else, this place is not for you: this community is very positive and I like to speak with a lot of guys here: I've learned a lot!

[olivierpons]

i don't see any actual improvement over any of the existing ergodox offerings, besides the intense RGB LEDs, and the benefit there is really subjective. despite that, this is still the most expensive. you can get an ergodox EZ bundle with wrist rests and tenting kit (tenting for an ergodox is almost necessary for any real ergonomic benefit), which is ready to ship right now from a company with great customer service. '

the ergodox market is honestly saturated when you consider its degree of niche. something that's 5% different from the current ergodoxes just can't target any significant market

Here are the improvements:
- it's very easy to configure. We've made a video, and unless you've not seen it, you dont realize how easy it is to make macros and configure the LEDs so you dont see the benefits
- you're right: it's the most expensive keyboard: the Ergodoz-ez is available for $325 with 20 leds or so, and ours with 78 leds for $349
- the ergodox market is honestly saturated when you consider its degree of niche. something that's 5% different from the current ergodoxes just can't target any significant market: you're right too: 5% of $30000 = 1/20. It clearly shows you're 19/20 right.

Note: I'm not trying to sell it because this project is over.

The actual improvements are (in real life): easy to configure when you need many layers, incredibly easy to do macros (you cant do macros unless you're a seasoned developper right now with the other keyboards), 8 keys for the thumbs, and the RGB are easy to configure so you always know on the layer you are.

Each single improvement is very important for me because that's exactly what was missing for the Ergodox (not talking about the ErgodoxEz which is a simple copy/paste of the Ergodox), and with the Infinity Ergodox, it's still impossible to configure easily the screen (which is the only reason for me to buy it).

[algernon]

I didn't want to comment here, but after reading the last few posts, I'm afraid I must. Just so I won't be dismissed as some random dude without a clue: I've spent significant time working on keyboard firmware (QMK, UHK's firmware, and Kaleidoscope, which powers the Keyboardio Model 01 among other things). When I say significant, I mean thousands of lines of code. The vast majority of Kaleidoscope plugins are my work. There's very little I didn't do with firmware yet, and even less that I can't. As someone with a burning passion for firmware stuff, I can't in good conscience abstain from commenting about a few things said in this thread.

While there are plenty of stuff I could comment on, I'll stick to firmware & software things only. For the most part, anyway.

To be able to implement RGB Leds, USB 3.0 and the HUB behind, everything had to be made from scratch.

RGB LEDs can be handled with existing firmware, or at worst, existing libraries that interface well with existing firmware. Been there, done that on two different keyboards. It's not rocket science, and you don't need to rewrite anything for that. It is doable with QMK, obviously doable with Kaleidoscope, and so on. The hardware side may be a bit more difficult than the firmware side, but there are plenty of keyboards with per-key RGBs, and a number of options to make it work (an MCU with more pins, or a separate LED controller and I2C between the two, etc), all of which have been tried, and plenty of information available.

The USB3.0 parts may have warranted bigger changes, yes. But from scratch? I don't think so.

- it's very easy to configure. We've made a video, and unless you've not seen it, you dont realize how easy it is to make macros and configure the LEDs so you dont see the benefits

I have not seen the video, but I do know how easy it is to make macros on my Keyboardio: I press a "RECORD MACRO" key (or combo), tap the key I want to assign the macro to, tap the keys for the macro, hit "RECORD MACRO" again. Done. Okay, this code is fairly new, because I didn't get around to it earlier, but it's *simple*. Key remapping and LED configuration can be similarly done, without an external software.

Easy configuration is not unique: UHK has a GUI Agent, Kinesis has its tool as well (granted, the Kinesis tool is not open source), and Keyboardio is working on one too (and the GUI is designed in a way that it will work for all keyboards supported by Kaleidoscope, with little glue code required). The hard part here is the firmware part, that allows updating all these, without having to re-flash, but that's not unique either, the Animus firmware (used for the Diverge and other unikeyboard products) has supported this for a while, as far as I recall. So does the UHK. And HEX-generating configurators have existed for quite a while, and some of them are pretty reasonable too.

Again, you don't need to change anything in the hardware to make this work. It can all be done in firmware & a configurator app. Building a product around the configurator would be silly.

incredibly easy to do macros (you cant do macros unless you're a seasoned developper right now with the other keyboards)

There are many, many people who made their own keymaps AND macros with QMK, even though they are nowhere near being seasoned developers. I've helped some of them with trickier ones, but the vast majority of them did so by looking at documentation and other examples. It's *not* hard. Can it be made easier? Yes. It's a great goal to make it easier, but asserting that one needs to be a seasoned developer to create macros with all other keyboards is false. Even if you only count keyboards with open source firmware. If you don't, the assertion is an even bigger lie, I'm sorry.

8 keys for the thumbs

One of the big complaints about the ErgoDox is about the thumb cluster. Putting more keys there just makes it worse. More is not always better. I have reasonably sized hands, and I can comfortably use two keys on each side, three if I stretch a bit. Of the eight thumb keys on the ergoflip, I could easily reach four, but their arrangement is so that if I tent the keyboard, and use the posture I'm used to on the ErgoDox EZ I have, I'd have a hard time not hitting both keys. Besides, as far as I can tell, the ErgoDox PCBs supported a 80 key variant for a good while now, with a similar thumb cluster setup as the Ergofip.

So, this isn't really an improvement either, because the ErgoDox already supported this. The firmware too (QMK at least, possibly others as well).

still impossible to configure easily the screen (which is the only reason for me to buy it)

The easiest route here is to work on a way to make it easier to configure the screen. That's a bit of firmware work, kinda cheap, too.

As a summary, so far it seems that the major complaints are:

- No easy configurator that supports macros and LED coloring: this does not need new hardware. Most of it is outside of the keyboard too, and the firmware side can be pretty darn lean. You just need a reasonable GUI.
- Not enough thumb keys: ErgoDox PCBs support 8 thumb keys / side. So do the firmware, and at least some of the existing configurators.
- Not easy to configure the screens on the Infinity ErgoDox: firmware & configurator changes, both are open source, aren't they?
- No per-key RGBs: This is the biggest one, which does require change in the hardware. But the keyboardio has this, so it's not unique to the Ergofip, either.
- No USB3.0 HUB stuff: This I can't comment about, as it is something I don't care about, and as such, haven't researched how hard it would be.

Pretty much all of this, bar the USB3.0 HUB, is supported by keyboards that came before the Ergofip, and out of this five complaints, only the last two would require any change to the ErgoDox hardware.

[olivierpons]

First of all I'd like to thank you so much for such a constructive answer.

RGB LEDs can be handled with existing firmware, or at worst, existing libraries that interface well with existing firmware. Been there, done that on two different keyboards. It's not rocket science, [snap]The hardware side may be a bit more difficult than the firmware side, [snap]

The USB3.0 parts may have warranted bigger changes, yes. But from scratch? I don't think so.

You've been in both hardware and software: thus you can imagine how many lines we should change using existing stuff, and you just said that USB3.0 parts may have warranted bigger changes. So hardware + software modifications = a choice we all have to do: either try to improve/upgrade to fit our needs, or rewrite: the amount of work is the same (you seem very good so you know that).

I have not seen the video

That's very honest from you. Type "youtube ergofip", it's 4mn length. Not a big deal. I honestly dont understand why people make comment about something just by relying on 2-3 pictures. Weird.

on my Keyboardio: I press a "RECORD MACRO" key (or combo), tap the key I want to assign the macro to, tap the keys for the macro, hit "RECORD MACRO" again. Done. Okay, this code is fairly new

"Okay, this code is fairly new". So one year and a half - when I've started my project - it didn't exist? So my project - at the time - could be legit?

but it's *simple*.

100% agree.

Key remapping and LED configuration can be similarly done, without an external software.

I don't know how you could easily do LED configuration by layers so simple. Seriously, just give a try to the configurator! Maybe you'll change your mind on this side.

Easy configuration is not unique: UHK has a GUI Agent, Kinesis has its tool as well (granted, the Kinesis tool is not open source), and Keyboardio is working on one too

They are not online = on the web. This is one of the big advantages I've worked on: (1) you go to the online configurator, you log-in. (2) you have all your configurations online, ready to share. (3) you modify it, click "generate" and you upload the configuration file (4) you click on "reset" on the Ergofip, it's detected as a harddrive, your dragndrop your configuration, the Ergofip reboots and tada! It's done.

So yes, I do 100% agree: there are thousands of configurators. Even the Razer configuration for my Chroma Ornata is very powerful but you have to install it on the PC (not Linux compatible), and it works only on that PC. It's not online. And even though it's very powerful, it's not easy to grasp.

(and the GUI is designed in a way that it will work for all keyboards supported by Kaleidoscope, with little glue code required)

"with little glue code required" -> maybe this is easy, and not a big deal for you, but it's always very hard for beginners. I've spent 4 days with a writer who bought the Ergodox, explaining how to do this and that. You're right: it's easy, but the problem is that it's not easy for everyone. It's easy for 1/100000th of the population.

Building a product around the configurator would be silly.

I'm a developper, and you can see with my 15k+ contacts on LinkedIn, and the 3 latests books I've been asked to review by Packtlib on MeteorJS and Django that I'm not on the electronic side. I wanted a nice configurator, and to know on which layer you are visually. The Infinity Ergodox solves the second problem, but there's no editor for the macros, and the way the LCD works is a non-sense to me (I've already explained this), and would require a lot of rewriting of the firmware. When I got the Infinity Ergodox (8 months late, I was one of the first buyers) it had a contact failure on the LCD and I was so disappointed that I started looking for an electronical engineer.

There are many, many people who made their own keymaps AND macros with QMK, even though they are nowhere near being seasoned developers.

This doesn't mean it's easy! Maybe if there are 1000 people who did manage, there are 100000 who did not buy the keyboard because of this! Look at the ErgodoxEz: that's exactly the proof of that: maybe there are 1000 people who like to solder, but 10000 would like to buy a pre-soldered keyboard, and that's why the ErgodoxEz is a success.

I've helped some of them with trickier ones

...see? 

It's *not* hard. Can it be made easier? Yes.

That's just what I wanted to read: it can be made easier. Thank you for that.

It's a great goal to make it easier, but asserting that one needs to be a seasoned developer to create macros with all other keyboards is false.

Ok you're right so I should apologize about this: I've not tested all the hardware and firmware of all the keyboard of the world, and out of the 20 different configurators I've tested, no one was "easy" (in the sense "easy for everyone including newbies") to use. So it's not all the keyboard. I should have said that in the first place.

One of the big complaints about the ErgoDox is about the thumb cluster. Putting more keys there just makes it worse.

I totally disagree on this: I do have the Ergodox, the Infinity Ergodox, and the Ergofip and a 1u key is far easier to reach and tap than a 2u.

I can comfortably use two keys on each side, three if I stretch a bit

That's what I wanted to read, and I thank you again for being so honest: "if I stretch a bit". You don't have to stretch to reach the 1u key on the bottom of the Ergofip.

Of the eight thumb keys on the ergoflip

It's the Ergofip Nevermind. You're right on the thing I didn't explain: in all case, all positions, the "top" keys of the thumbs are hard to reach. So my reflexion was "If they're hard to reach, you can split them so you have more keys - top key still hard to reach - and middle + bottom easier to reach". Maybe I've never properly expressed myself for this, and I'm sorry about that.

The easiest route here is to work on a way to make it easier to configure the screen. That's a bit of firmware work, kinda cheap, too.

Again, for you it's easy, it's like a walk in the park by a sunny day. For 99,99% of other ones it's not. I'm sorry, if you think you're like everyone but you're not: you're good at keyboards and it seems you dont realize how bad at hardware and firmware "everyday people" is.

As a conclusion I'd like to thank you very much for being honest and:
- You need a reasonable GUI (even though you think it's easy, it's not that easy (otherwise you'd have tons of "easy" GUI))
- per-key RGBs: it does require change in the hardware (but it's not unique)
- USB3: it does require change in the hardware (but it's not unique)

What made (made, not "makes") the Ergofip unique is that it would have combined all those features together.
Maybe I've never mentionned it, but I didn't talk about the micro SD card reader and the HUB behind, because it was not working but it was implemented in the design, only the electronic "routes" were problematic (sorry for my english) and it wasn't working. Just a few days of debugging to make it work + another prototype, but this costs money and Damien Vidal didn't want to add one cent more into the Ergofip (and that's why he's 80% responsible of the failure: we couldn't do anymore video or make 100% finalized prototype to send to youtubers for a test, even though I did have planned that one year and a half ago...).

I'm still convinced that there's so much room for improvement that there's a place for another "ergodox-like" keyboard. But I've stopped the adventure here.

[algernon]

I have not seen the video

That's very honest from you. Type "youtube ergofip", it's 4mn length. Not a big deal. I honestly dont understand why people make comment about something just by relying on 2-3 pictures. Weird.

I have now watched the video, there was very little in it that wasn't available elsewhere. A few details about the configurator, and that's about it.

on my Keyboardio: I press a "RECORD MACRO" key (or combo), tap the key I want to assign the macro to, tap the keys for the macro, hit "RECORD MACRO" again. Done. Okay, this code is fairly new

"Okay, this code is fairly new". So one year and a half - when I've started my project - it didn't exist? So my project - at the time - could be legit?

Similar solutions did exist, adding the same feature to QMK for example would have been reasonably straightforward too. A dynamic macros feature already exists in QMK (though it works a bit differently, as far as I remember), and while that's newer too, it shows that building on existing software is often easier than starting from scratch.

Key remapping and LED configuration can be similarly done, without an external software.

I don't know how you could easily do LED configuration by layers so simple. Seriously, just give a try to the configurator! Maybe you'll change your mind on this side.

Watched the video, the color configuration there is, indeed, easy. Best is to have both, so you can reconfigure LEDs without software, if need be, but use a tool when that's the better option.

Easy configuration is not unique: UHK has a GUI Agent, Kinesis has its tool as well (granted, the Kinesis tool is not open source), and Keyboardio is working on one too

They are not online = on the web. This is one of the big advantages I've worked on: (1) you go to the online configurator, you log-in. (2) you have all your configurations online, ready to share. (3) you modify it, click "generate" and you upload the configuration file (4) you click on "reset" on the Ergofip, it's detected as a harddrive, your dragndrop your configuration, the Ergofip reboots and tada! It's done.

Well, the Kinesis tool comes with the keyboard. You don't have to go anywhere, you don't have to install it, either.

The ErgoDox EZ configurator, qmk.sized.io work similarly to yours: they generate a firmware online, and you re-flash it. The reason Kinesis, UHK, and Kaledioscope are going for a non-online version is because re-flashing sucks from an end-user point of view. Not to mention, having to log in somewhere means that I don't have the data locally, unless I export it - meaning, I rely on a service that can go away anytime, to be able to configure my keyboard. No thanks. I've been bitten by similar things before, I'm not trusting my configuration to a single point of failure (instead, I put it in git and mirror it everywhere).

Thus, for both non-techie end-users (who will hate flashing), and tech-savvy people (who prefer the source), the online configuration is no good. There are people somewhere inbetween, for them, a mixed architecture can work: the same app that would run locally, now runs online, and instead of uploading the configuration, it generates a HEX. With an Electron-based app (on which both UHK's agent, and Kaleidoscope's Chrysalis is built on), this is within reach.

Did they exist a year and a half ago? Kinesis and qmk.sized.io did.

So yes, I do 100% agree: there are thousands of configurators. Even the Razer configuration for my Chroma Ornata is very powerful but you have to install it on the PC (not Linux compatible), and it works only on that PC. It's not online. And even though it's very powerful, it's not easy to grasp.

I disagree here. I find UHK's agent a lot easier to use than qmk.sized.io (which bears a lot of resemblance to the app I've seen in the ErgoFip video). Yeah, I have to "install" it, as in, download and double click it. I can do that. It's easier than registering on a website. Runs on Linux, Mac, and Windows too.

(and the GUI is designed in a way that it will work for all keyboards supported by Kaleidoscope, with little glue code required)

"with little glue code required" -> maybe this is easy, and not a big deal for you, but it's always very hard for beginners. I've spent 4 days with a writer who bought the Ergodox, explaining how to do this and that. You're right: it's easy, but the problem is that it's not easy for everyone. It's easy for 1/100000th of the population.

The glue code would need to be written once, by whoever ports the GUI to a keyboard. End users don't see that. From that point on, you just click around, assign keys, colors, and whatnot. Granted, it did not exist a year and a half ago, but again, a lot of others one could start from, did.

Building a product around the configurator would be silly.

I'm a developper, and you can see with my 15k+ contacts on LinkedIn, and the 3 latests books I've been asked to review by Packtlib on MeteorJS and Django that I'm not on the electronic side. I wanted a nice configurator, and to know on which layer you are visually. The Infinity Ergodox solves the second problem, but there's no editor for the macros, and the way the LCD works is a non-sense to me (I've already explained this), and would require a lot of rewriting of the firmware. When I got the Infinity Ergodox (8 months late, I was one of the first buyers) it had a contact failure on the LCD and I was so disappointed that I started looking for an electronical engineer.

As a fellow developer, when I encounter a problem, I try to fix it first, in the existing software (if it is open source, which it is in case of the ErgoDox). I agree about the LCDs, mind you, and a visual indicator of the current layer is also a good thing. But I still maintain that doing this with either QMK, or even the Infinity's firmware, would have yielded better results in the long run.

There are many, many people who made their own keymaps AND macros with QMK, even though they are nowhere near being seasoned developers.

This doesn't mean it's easy! Maybe if there are 1000 people who did manage, there are 100000 who did not buy the keyboard because of this! Look at the ErgodoxEz: that's exactly the proof of that: maybe there are 1000 people who like to solder, but 10000 would like to buy a pre-soldered keyboard, and that's why the ErgodoxEz is a success.

There are a lot of people who used either the Massdrop configurator, qmk.sized.io, or the ErgoDox EZ configurator. There may have been users who didn't buy an ErgoDox, because of the shortcomings of the configurator, but I doubt that would be a significant number (seeing the number of people who do use the existing configurators).

I've helped some of them with trickier ones

...see? 

There are things you just can't do with an online configurator, and you have to write custom code. The fact that this can be done, with a little help, by people with very little programming experience, is telling.

One of the big complaints about the ErgoDox is about the thumb cluster. Putting more keys there just makes it worse.

I totally disagree on this: I do have the Ergodox, the Infinity Ergodox, and the Ergofip and a 1u key is far easier to reach and tap than a 2u.

Lets agree to disagree here then =)

I can comfortably use two keys on each side, three if I stretch a bit

That's what I wanted to read, and I thank you again for being so honest: "if I stretch a bit". You don't have to stretch to reach the 1u key on the bottom of the Ergofip.

I think you misunderstood: I can reach the 2u keys fine, and the bottom most 1u with a very tiny stretch. Replacing the two 2u keys on the thumb cluster with 4 1u ones would make things worse for me, because my thumbs rest around the middle of the 2u keys. To reach the bottom 1u, I'd have to bend them, to reach the top 1u, straigthen, while taking care not to press the bottom one.

This appears to be a similar setup to the ErgoDox80. Some people swear by it, I find it annoying. But this is personal preference, really. :]

Of the eight thumb keys on the ergoflip

It's the Ergofip

Sorry about that! Brain did an autocorrect mistake, it seems =)

The easiest route here is to work on a way to make it easier to configure the screen. That's a bit of firmware work, kinda cheap, too.

Again, for you it's easy, it's like a walk in the park by a sunny day. For 99,99% of other ones it's not. I'm sorry, if you think you're like everyone but you're not: you're good at keyboards and it seems you dont realize how bad at hardware and firmware "everyday people" is.

I'm not saying everyone should do it. I'm saying that someone building a product (ie, you) should, so your users don't have to. If you have something that is almost good, but not quite there, I believe it is easier to extend it to do what you want, than to start from scratch.

Maybe I've never mentionned it, but I didn't talk about the micro SD card reader and the HUB behind, because it was not working but it was implemented in the design, only the electronic "routes" were problematic (sorry for my english) and it wasn't working.

Now, a micro SD reader is something that would have been terribly useful, indeed. So much storage for layouts, LED themes, and whatnot! (And one could ship the offline variant of the configurator on there, too

[olivierpons]

I have now watched the video, there was very little in it that wasn't available elsewhere. A few details about the configurator, and that's about it.

That's what I wanted to read! There are "few" but this doesn't mean they are not important...

...it shows that building on existing software is often easier than starting from scratch.

It's like developping new apps: it all depends on the quality of the product, and the quality of the developpers. Damien Vidal is very good. And makes very high quality products. I've relied heavily on his qualities (though I've been very disappointed by his involvment that broke the whole project).

Watched the video, the color configuration there is, indeed, easy. Best is to have both, so you can reconfigure LEDs without software, if need be, but use a tool when that's the better option.

100% agree and very good idea (even though I dont know you could reconfigure LEDs without software... I cant imagine... if you can I'd be very interested).

Easy configuration is not unique: UHK has a GUI Agent, Kinesis has its tool as well (granted, the Kinesis tool is not open source), and Keyboardio is working on one too. Well, the Kinesis tool comes with the keyboard. You don't have to go anywhere, you don't have to install it, either.

As I told you I've not tested all the configurator and all the keyboards... I dont know how the Kinesis tool work I'll check it out ASAP, thanks for the suggestion!

Having to log in somewhere means that I don't have the data locally, unless I export it - meaning, I rely on a service that can go away anytime, to be able to configure my keyboard. No thanks. I've been bitten by similar things before, I'm not trusting my configuration to a single point of failure (instead, I put it in git and mirror it everywhere).

Ok I get your point of view. I do the same principle for facebook. But billions of people do it for facebook... Maybe you don't like the principle, and so do I, but a lot of people like it. Moreover, it's a one-click login with gmail button or facebook button.

Thus, for both non-techie end-users (who will hate flashing), and tech-savvy people (who prefer the source), the online configuration is no good.

This is the point I disagree, and I've read carefully your arguments. They are really interesting, but I've decided to make the Ergofip after having talked to more than 500 people (all my students + conferences I give), and they all said it was better online, and (believe me or not) it was for almot only one thing: to be able to see their configurations on their mobile phone... dont ask me why, I've stopped trying to understand people and I'm just relying on actual facts.

There are people somewhere inbetween, for them, a mixed architecture can work: the same app that would run locally, now runs online, and instead of uploading the configuration, it generates a HEX. With an Electron-based app (on which both UHK's agent, and Kaleidoscope's Chrysalis is built on), this is within reach.

This is very interesting, may I ask you links for this I'd like to know more about those technologies.

Did they exist a year and a half ago? Kinesis and qmk.sized.io did.

They all didn't have all the features grouped, maybe some, but no all.

I find UHK's agent a lot easier to use than qmk.sized.io (which bears a lot of resemblance to the app I've seen in the ErgoFip video). Yeah, I have to "install" it, as in, download and double click it. I can do that. It's easier than registering on a website.

Easier than one click? (OAuth2). I seriously doubt it... And I'm just relying on the study I've made, and I hate to tell all my story, all my life, because I think my life aint worth a penny, but there you go: when I say it took me one year and a half to make this and some people dont believe me it's because I dont want to loose time and explain in great details all the things I've done to come to this point. I've rewritten 4 times the configurator, taking in account all note, all remarks, until my 8 years old daughter understood how to use it, and my challenge was to make my 6years old son use my configurator: the principle of drag'n drop + colors were so fun that my son played with it a whole afternoon. So when I say "easy", I seriously mean "easy"... I've never found another keyboard app that a 6 years old kid could play with. If you know another one I'm your man! I like so much to learn! And that's maybe a part of the failure of the Ergofip: I'm the kind of guy who hate to explain the failures. Why loosing time to explain all my 3 attemps whereas something working and easy is enough? Maybe I'm wrong there too....

The glue code would need to be written once, by whoever ports the GUI to a keyboard. End users don't see that. From that point on, you just click around, assign keys, colors, and whatnot. Granted, it did not exist a year and a half ago, but again, a lot of others one could start from, did.

It's the story of the iPhone... it's so easy to use... everybody could have done it, right? Click, scroll, easy... but no one had made it before... once it's done you can say "everybody could have done that with very few work". I think it's not that simple...

As a fellow developer, when I encounter a problem, I try to fix it first, in the existing software (if it is open source, which it is in case of the ErgoDox). I agree about the LCDs, mind you, and a visual indicator of the current layer is also a good thing. But I still maintain that doing this with either QMK, or even the Infinity's firmware, would have yielded better results in the long run.

You may be right, and that's the only thing left to me in the keyboard field: try to understand exactly how everything is written, and add my own "layer" to this so everyone can easily configure macros and the picture displayed on the Ergodox LCD (a different picture on the whole keyboard, for *each* layer).

I think you misunderstood: I can reach the 2u keys fine, and the bottom most 1u with a very tiny stretch. Replacing the two 2u keys on the thumb cluster with 4 1u ones would make things worse for me, because my thumbs rest around the middle of the 2u keys. To reach the bottom 1u, I'd have to bend them, to reach the top 1u, straigthen, while taking care not to press the bottom one.

I've bought the plank keyboard, I'll measure if I'm faster or not... with no 2u keys and all keys very close to each other... in 3 months (according to massdrop).

This appears to be a similar setup to the ErgoDox80. Some people swear by it, I find it annoying. But this is personal preference, really. :]

I swear by it

I'm not saying everyone should do it. I'm saying that someone building a product (ie, you) should, so your users don't have to. If you have something that is almost good, but not quite there, I believe it is easier to extend it to do what you want, than to start from scratch.

The problem was all about RGB's and Damien told me everything had to be rewritten. It's not my side, I believed him. Regarding the software, I won't comment what has been done because I dont know how hard this is to build everything, but from what I've made (see my liveedu videos), I had very few more work to finish the job...

Now, a micro SD reader is something that would have been terribly useful, indeed. So much storage for layouts, LED themes, and whatnot! (And one could ship the offline variant of the configurator on there, too

Yes. I never talk about my attempts, or my failures, and I seldom talk about things that are not finished. I just like when the job is done, and especially when it's well done... that's maybe a problem with nowaday's people who like to telling their own lives (and I dont give  a flying f... of all the lives of almost everybody but my family, it sound like a behavior problem!). I'll never ever make a blog like keyboardio does. Never ever. So I will always fail, always!