Right now I am deep in a project which I refer to mentally as the Keyback Steno.  It is a keyboard like this one http://stenoknight.com/demo.html which has electromagnetically locking keys. As you can see in the photo of a traditional stenotype keyboard, the keys are on levers. If it were possible to hold one end of the lever with an electromagnet while the key was in its down possition, then I would be able to controll the keys release programatically. If all the keys were held down, than I could programatically command some of the keys to pop up. This would allow the keyboard to function not only as an input device, but also as a device for displaying text. Since one can type on a stenotype keyboard at 250 WPM than it is reasonable to assume that reading on such a keyboard could be accomplished at a similar or higher rate of speed. Theoretically speaking, this should allow me to create a low cost and useful haptic device which could compete with the overpriced braille displays used by blind people today.

Unfortunately, stenotype keyboards cost around $4000 a peice today. If I were to approach design, development and production with the same mindset as those companies which produce such costly stenotype keyboards, I would do visually impaired individuals no good. But when you examine the parts involved, it seems like an unreasonable price. One begins to think to themselves "What parts really are involved in making such a keyboard?" "I could do that cheeper."

Over the past months, I have spent a lot of time on youtube and I have learned a lot about manufacturing. I have learned that one of the key reasons that products like stenotype keyboards cost so much is that they are designed by people who do not understand the costs involved with the various manufacturing techniques. They use large injection molded plastic parts, which are reasonable to use only in large production volumes, when it would have been cheeper for them to use more expensive, higher quality (and therefore easier to work) materials such as sheet metal, wireforms, lasercut foam, and woods.

If I were to attack the problem of designing a keyback steno keyboard with the mindset of most companies in the area, I would be interested in building up a set of tools. The first thing I would do is I would purchase Solid Works in order to model my keyboard. I don't know how much SolidWorks costs because THEY DO NOT LIST A PRICE ON THEIR WEBSITE! I would need to purchase a new computer and that would cost me around $700-1500, the computers I have do not run SolidWorks. Howerver, I would tell myself that the purchase of SolidWorks and the computer was a cost saving, because it would save me money on prototyping. Afterall, with SolidWorks, one can model their designs, measure mechanical forces, ect all without every having to build a prototype.

I would also hire a number of engineers who were capable of telling me the mechanical properties of the materials I was using, so that we would be able to optimize the weight and cost of the materials. This would also be a cost savings, because the engineer would prevent unecesary rounds of prototyping and we would be able to design a product with overall lowered material cost.

I would also bring in outside consultants for designing the electromagnets, as they are far from standard, an outside computer programming consultancy to write firmware and drivers, as well as outside consultants to run me through CE certification.

Going along with the "good tools" philosophy, I would buy the 10 000 EUR (I had to ask, the price is not on their website) LPKF ProtoMat S43 http://www.lpkf.com/products/rapid-pcb-prototyping/circuit-board-plotter/protomat-s43.htm along with their 10 000 EUR mutlilayer press in order to prototype the circuit boards which will have my special electromagnets printed on them. I would also consider this to be a cost savings, because prototyping specialty multilayer circuit boards is famously expensive.

<crossed out>Finally, I would end it all of with a tens of thousands of dollar stratasys 3D printer in order to prototype the chassi and other parts.</crossed out> Finally, I would purchase a high powered laser cutting system (this is something that is surprisingly easy to find the price of, 40 thousand euros and up to cut metal), as well as a small manual sheet metal bend press (10 000 euros, the full CNC ones are 100 thousand euros and up and are actually more expensive than laser cutters). I would also need a quality drill press with a high powered spindle (500 euros). And for all my prototypes, I would need to source materials in low quantities, which would end up costing me many hundreds of euros in shipping fees and sample pack fees.

In the end, I would have invested hundreds of thousands of Euros in research and development, I would have outside investors to please, and there would be no way that my low cost haptic computer system would be anything like low cost, and it certainly would not be open source.

HOWEVER:

While it is nowere near the quality of SolidWorks, there is a program called FreeCAD which can replace it to some degree. FreeCAD is free and open source, and it exists because some people chose not to pay up for SolidWorks and decided to design an free open source cad project instead.

In place of an army of engineers, I have this recent invention called the internet. It's not perfect, but it's WAY cheeper! In the end, I almost certainly will pay someone with experience to look over my plans, but there is a huge leap between a short consultation and employees. The open nature of the hackerspace and the open source / free software comunity as a whole, is on my side however. I can talk about my project to anyone, and I have already received loads of valulable feedback, absolutely for free!

As for the software, there are already free open source libraries for stenography, and there is already a driver framework (brltty) which will alow me to complete everything without writting more than a few hundred lines of code. If I were developing my product like the companies described above, I would imediately reject this oportunity, however, as the libraries in question are GPL and it would mean I would lose controll of my "intelectual property".

The protomat router for making precice PCB prototypes is a more serious problem for me. I have thought about purchasing it. It is VERY difficult to achive the level of precision that they claim. However, there are a number of open souce router projects, and these projects did not come to exist due to people saying "well I can just buy the LPKF machine". They came to exist, becuase they looked at the LPKF machine and said "it's 10 000 euros, requires Windows, and it's not open source, I'll try to do it myself".  And if I go with one of these router projects, I will most likely be able to both cut out sheet metal parts and mill circuit boards on the same machine (though the curcuit boards will have a higher minimum trace size than on the LPKF... ).

As for the 3D printer, yes, you can buy cheep 3D printers today. You can do so BECAUSE the reprap project exists. Untill reprap came about, 3D printers were so expensive that they were out of reach even of most comercial labs. It is only thanks to the likes of Prusa and Ax that we can even have the discussion about whether the fabtotum should be bought!

--------------------------------------------

All this said, free open source software like FreeCAD is written on computers, and most of that silicon is closed as closed can be. It was only thanks to tools like the closed sorce definitely not DIY processors by Intel that FOSS like FreeCAD can exist. And Prusa printed the Prusa on a definitely not DIY printer himself.

I agree with you generally that we need quality tools. But I also strongly believe that throwing out the DIY mentality loses many of the positive aspects of the hackerspace. It can even lead to ruin of sorts. The ruin I am talking about, is when people start buying into comercial tools which are imperfect (and believe me, comercial tools are imperfect too) then they start developing those tools. This is what I call a captive comunity. A captive comunity, is when a comunity of well meaning developers begin building a comunity of free content on top of a close source/proprietary base. A classic captive comunity is that of EagleCad. A significant portion of EagleCad's value comes from the efforts of people in the EagleCad community and not from the firm itself. If those efforts had all been directed towards KiCad or some other open source alternative, the world would be a much better place. Open source effort built attop proprietary platforms helps those platforms for free while doing the DIY community a great harm.

Even just using a product, without actively working on it, helps that product, by increasing general knowlege of the product in a community and providing ample oportunities for yourself to provide free technical support to the users of that product. For example, on c-n-c.cz it is far easier to get free technical support using the closed source software Mach3 than the open source software LinuxCNC. By choosing the convenient an "quality" ready built software solution, the c-n-c.cz community has promoted a less DIY solution over a more DIY solution thus making it harder for the open source solution to gain foothold.

I agree with you generally, that brmlab could use some more good tools. However, I wory that if we invest too heavilly in proprietary soloutions, brmlab will lose its value. Brmlab shouldn't exist to build captive comunities for companies like EagleCad who wish to get free work done by enthusiests. Brmlab should exist for its own purposes and for the purposes of promoting the DIY and open source communities as a whole.

----------------------------

From the open source standpoint, fabtotum is a rather concerning project. They claim that they are open source, however they use a creative commons NON COMERCIAL license for everything they publish. This means, that their work cannot flow back to other open source projects. It also means that they cannot legally accept patches without contributors signing a dead tree contributors agreement. Though I bet they don't realise that aspect of their licensing decision.

-----------------------------

In conclusion, I think that brmlab SHOULD invest in more quality tools. We don't even have a full sized drill press, and throughout the reprap project we have been cutting threaded rod with a rotary sander for want of a table saw! However, I also agree with those who are sceptical of giving up our DIY attitude for more complex equipment.

There's also another thing to consider. I thought for a little bit, about maybe buying the LPKF router and putting it in BRMLAB. Afterall, I certainly wouldn't find use for it 24 / 7. But I quickly set asside the idea. There is no way in hell I'm ever going to bring a 10 000 euro peice of equipment to brmlab and leave it lying around. It would almost instantly get destroyed or stolen!

Tim

---------- Původní zpráva ----------
Od: Mario Lombardo <mario@alienscience.com>
Komu: Brmlab: Hackerspace Prague, (main discussion) <brmlab@brmlab.cz>
Datum: 10. 3. 2015 19:25:50
Předmět: [Brmlab] Fabtotum


Dear BRMlab,
I have expressed interest many times, both publicly and privately, that it is vital to have reliable and quality tools. At the time of the Fabtotum crowd funding campaign, a year or so ago, the 3D printer we had in the lab needed constant tuning, we didn't have an optical 3D scanner, nor a milling machine. A member even said it goes against the spirits of a  hackerspace to buy such a tool. Yet back then, in my experience with using the 3D printer it proved it wasn't a reliable tool nor a quality tool of its rank. Even so, almost a year later, we still don't have an optical scanner nor milling machine. Does our 3D equal in quality of the Fabtotum? If not, who will take the time to improve it or make a new and better one? Will we approve such an expense and time? As of now, we still have a down larger 3D printer, so what of it?
The downfall of poor quality tools is that they slow progress and stifle creativity. As hackers and makers, we have enough obstacles to overcome than let a tool deceive or encumber vital steps in a process or discovery--testing and measuring equipment at the top and prototyping equipment second--both vital nonetheless. 
My idea was to help us. b00lean expressed interest in such vision and so did a small minority, so a small pool of us gathered and helped purchase a future unit.
Sadly, much time has passed, even more than anticipated, yet the question remains, would BRMlab desire, care, and utilize such a tool? At this time, I would like to hear more voting and financial support than voiced we, the original party. For if not, then it can go with General Bytes because b00lean seems to like it. If any members are bashing the tool, I would be curious if they have firsthand experience? If not, part of the effort in acquiring the Fabtotum was that "it could" do all of the proposed feats it its capacity. We believed just as we might believe in something else, like purchasing the SDR and finally much later a member is able to make an incredible GSM device.
As a consensus, however BRMlab votes this evening, I will __abstain__ and hope for this tool or some tool some BRMlab collective deem worthy of our respect, care, and use. We need better tools, and I was hoping for this one. Just think about us and not your ego--to build one as a hack project. That has always been an option, but can you rely on what it makes? A good tool will deliver. A hack will mostly deliver an accomplishment or proof-of-concept that is itself. That's wonderful but wildly different from the mark of a tool. Both have merit, but whether both should occupy the same desktop?... Hmm.
I've made both, and the one thing that annoys me in making a tool, is that I've got to be bitching accurate in every way, letting no computation be an error or succumb to its duplicity tenfold. Otherwise, I'd keep a rubbish bin nearby my knee and simply push the craft to its demise... "Do over!"

On Monday, March 9, 2015, b00lean <b00lean@b00lean.net> wrote:

Ahoj,

Minuly tyden v pondeli dorazil Fabtotum ( http://www.fabtotum.com/ ), ktery jsem objednal a zaplatil 10.10.2013 (celkem  21 463,67 CZK )

Zatim jsem ho nerozbalil.

 

V rijnu 2013 jsme se dohodli, ze se na fabtotum slozime a moje firma doplati zbytek ( http://brmlab.cz/project/fabtotum )

K prispevku se prihlasili nasledujici:

 

mario 1000 CZK

pasky 1000 CZK

pborky 1000 CZK

JoHnY 1000 CZK

niekt0 1000 CZK

Ondrej Mikle 1000 CZK

Eliáš 1000 CZK

b00lean doplati zbytek.

 

Na zminovany ucet(na strance) dorazila pouze jedna platba ( 1 000 Kč - Ján Teluch - nevim nick asi JoHnY)

Nekteri ze zminovanych mi dali penize hotove (vubec uz nevim kdo ale asi jen 2 lide) nebo je vubec neposlali.

 

Od doby objednani Fabtotum a jeho doruceni se zmenilo hodne predevsim vsak to, ze jiz nejsem clenem brmlabu, nemam do neho pristup a je mi lito zaplatit za fabtotum tolik ze sveho a nechat ho v brmlabu k "opotrebeni/rozbiti" lidmi, kteri na nej neprispeli a tudiz k nemu ani nebudou pristupovat s patricnou odpovednosti.

 

Navrhuji nasledujici reseni:

A. Fabtotum si ponecham a vyplatim ty, kteri prispeli.

B. Brmlab odkoupi Fabtotum za zbytkovou cenu ze svych penez (zaplati zbytek me).

C. Najde se v radach brmlabu dostatek dalsich lidi, kteri na pristroj prispeji tak aby brmlab nemusel nic odkupovat, nebo brmlab doplati mene.

 

Prosim o vase konstruktivni reakce a komentare,

b00lean

 

 



--
>>sent from mobile<<
_______________________________________________
Brmlab mailing list
Brmlab@brmlab.cz
https://brmlab.cz/cgi-bin/mailman/listinfo/brmlab