Hardware Is Now Open (sourced) For Business 42
Posted
by
samzenpus
from the free-and-open-model dept.
from the free-and-open-model dept.
ptorrone writes "CNBC has an interesting article about the growing trend of hardware companies going open-source. 'The open-source hardware movement is migrating from the garage to the marketplace. Companies that follow an open-source philosophy make their physical designs and software code available to the public. By doing so, these companies engage a wave of makers, hobbyists and designers who don't just want to buy products, but have a hand in developing them.'"
Hey look, it's Commodore and company! (Score:2, Informative)
Seriously though, Commodore, (Apple?), and lots of others used to provide complete wiring schematics either with their hardware, or as a seperately available book up to what, the mid to late 80s?
If the PC and related technology hadn't taken off, would anybody even be talking about 'open hardware' nowadays, or would that just have continued to be the assumed norm?
Just some food for thought.
Re: (Score:1)
Re:Hey look, it's Commodore and company! (Score:5, Insightful)
Oh yes in the 1960s all "lab" grade electronics like power supplies, oscilloscopes, whatevers, came with thick manuals with schematics, part lists and theory of operation as well as the basic user manual.
That's still the norm for many Amateur Radio products. What is missing from many products supplied by the big name manufacturers these days is source code for the embedded MCUs.
Many radios and test equipment used to be available in kit form too. But that has gone away since the advent of surface mount technology. Most Amateurs don't have the equipment, patience or eyesight to do SMT at home. Besides, pick and place robots will assemble a circuit board in minutes, reducing labor cost to a few cents per board. So, instead of saving a bunch of money on hand-built hardware as it used to, it actually costs more to offer kits than it does fully assembled boards. The technical support costs for kits is pretty high.
Re: (Score:3)
Re: (Score:3)
The problem is gonna be the patents, especially for mobile GPUs. Anybody who has looked into that stuff can tell you pretty much any way to make a screen render has a patent or hundred and they are all held by a handful of players who sure as hell don't want you playing in their sandbox. Wireless is just as bad, with pretty much every way to send a signal patented up the wazoo.
So while you might pull this off in countries without the IP bullshit but in the corporate states of America its not likely.
Different era. (Score:1)
Back then computers were just for us geeks.So of course they'd include schematics.
Today, computing is an appliance (thanks St. Jobs!) and a prestige/luxury item.
Ardino and Rasberry fill thos eniches now.
Re: (Score:1)
The PC? The full assembly source to the IBM PC BIOS was in the Technical Reference Manual.
Re: (Score:1)
And I have the service manuals for the Mac 128K/512Ke/Plus
However I am wondering how many "hobbyists" have the ability to work on multilayer boards, re solder BGAs etc etc etc.
Arduino etc are at the level of the old development boards, they are not "NEW", they are just current implementation.
Hell I have a UK101 which was a kit set project version of the Ohio Challenger 1.
God I wish young people would bother to find out about the 70's and 80's which were probably the most interesting years for computers,
hund
Re: (Score:2)
And here I am pretty proud of having the SMs and PMs for Selectrics, Mag Cards, Displaywriters, and IBM Electronics. I may never actually fix another one. Darn.
Re: (Score:1)
I work for a contract SMT manufacturing company.
We are not shit scared of BGAs, what we are shit scared of is QFP packages. We will not accept assembly jobs with BGA parts we can't place (the pad pitch is too small), as reworking them is a pain, but in terms of yield, BGA is much better than QFP, the reason being that for a similar number of contacts, the pitch of a BGA is much larger than a QFP because the BGA can spread them out over the package area, where-as the QFP has to have them at fine pitch along
Re: (Score:3)
IBM released schematics for PC, as well. There is a difference between releasing schematics, and 'open hardware'!
A circuit could be patented, the firmware (and source) may not be supplied or is otherwise encumbered, board layouts not supplied, etc. OSHW projects usually have all of this... everything you need to make it, unencumbered from any restrictive licensing.
Since the PC schematics were readily available, all the clones had to do was make a functionally identical BIOS (as the firmware was copyrighted)
Nothing really new here ... (Score:2)
Such things are good, but lets not pretend this is something new.
Re:Nothing really new here ... (Score:4, Interesting)
Yep. Many years ago, I was trying to fix a TV from the 70s. Full schematic glued to the inside of the set.
And to a certain extent, hardware has always been open source anyway. A motivated engineer can remove and identify components one by one and follow the wire traces on the circuit board. It's easier to reverse engineer a circuit board than a piece of software. Still, it's a lot easier if they give you the schematic up front. So I'd like to give a big shout out to SparkFun (www.sparkfun.com) electronics, who have made my life a lot easier.
Re: (Score:2)
Many years ago, I was trying to fix a TV from the 70s. Full schematic glued to the inside of the set.
Same here, but with a radio from the 30s or 40s. I guess the difference is that those devices didn't require a code listing for the firmware. A simple circuit diagram wouldn't be that useful these days.
Also, in the past if a device broke then you would have it repaired. They were made to last. These days you just throw it away and get a new one.
Re: (Score:1)
Releasing without all the "licensing crap" is precisely how we got to the point where openness and freedom are so rare that it feels new again.
Sorry but this is catagorically untrue. Innovation that is down outside the bounds of license falls into prior art. It's that simple. You don't even need creative commons licensing it's so simple.
Re: (Score:2)
That ignores the fact that RMS produced his license as a practical matter to quiet the cries of his contributors. Detractors like to paint RMS as some sort of communist/anarchist crusader but his license was really about preventing abuse and keeping order.
The CC licenses are just an extension of that.
Not everyone will play nice and you need a mechanism to deal with those people.
In the footsteps of Arduino (Score:5, Insightful)
If these companies are trying to occupy the same marketplace as the Arduino, i think it's too late. Otherwise it's definitely a good move.
In any case IMO what really allowed the Arduino to take off was not much the fact that it was open source, but rather the fact that it had readable documentation, which anyone could actually follow and make things work.
I am still amazed at the extent to which, to this day, the documentation for many Arduino-wannabe boards (e.g TI MSP 430, Chipkit 32, and others) really sucks.
Re: (Score:2)
You can take an Arduino schematic and run it on a 28-pin DIP, 5$USD ATmega328P on a single-sided PCB you make yourself at home or even a protoboard. I don't see most hobbyists doing that with anything related to the Raspberry Pi.
Re: (Score:2)
4X the cost? That seems rather high. I have purchased my last Arduino in the $20 range and if you are really cheap you can get the knockoffs on ebay for a lot less. The RasPi and Arduino were never meant to do the same things. Case in point, I have witnessed student teams tackle the same problem with real time sensing using the Pi and the Arduino. The Pi team ran a full linux distro and built their applications using python. The Arduino team used the Arduino. The systems ran on battery power. As you can ima
Re:In the footsteps of Arduino (Score:4, Insightful)
Indeed, "open-source hardware" just sounds stupid, if only because hardware doesn't have what programmers call "source". What we need is free hardware: the one with 4 freedoms RMS keeps talking about. Free hardware implies free and readable specs and documentation, since that's the only way to assure that users can use it and study it. It also implies a free and readable description of the manufacturing process, so that anyone skilled in the trade can make exact or modified copies.
Re: (Score:1)
...where-as the AVR does not have a debug protocol or JTAG, you have to replace the chip with an emulator. Have fun soldering all those wires from your emulator to your application board!
What about debugWire? Not full-blown JTAG, but it is an in-circuit debugging interface (uses the RESET pin). I believe it's now on all but the simplest of devices.
Natural Progression (Score:1)
This is about products, not components (Score:4, Interesting)
What's new here is the trend. Companies saw the RepRap project spawn a bunch of companies with a lot of compatibility from the start. Non-RepRap companies are seeing this as a threat to the investment they made using traditional methods (closed design, proprietary supplies and software).
Business people understand the IBM PC clone model. You had a market leader that everyone copied. The old-school thinking was they failed to protect their intellectual property, and lost market share to competitors who copied their design. In other words, they believe IBM could have kept nearly all the marked had they done a better job of keeping it closed, and bought Microsoft while they could.
RepRap and projects like it have upended that thinking. Arduino is seen as a component, not a product, by these people. But 3D printing is getting a lot of press, and business people are starting to take notice. When you create a 10 year plan, and can achieve a huge reduction in R&D spending, along with a reduction of risk, they take notice.
One of the concerns is the believe the a mature market only has room for two main competitors. That means you have a lot of losers. An open source machine makes it much more likely that your company will end up as one of the two majors, and that is a huge reduction of risk. This is becoming a hot topic among many executives. Many are somewhat scared and unsure what to do - if anything.
Re: (Score:2)
The old-school thinking was they failed to protect their intellectual property, and lost market share to competitors who copied their design
I don't think anyone significant believed that. Even IBM had long acknowledged that they thought the PC clones were the reason for the amount of volume they got. They knew long ago that thanks to that they got a decent chunk of a massive market instead of 100% of a pathetic small market (also, getting a decent chunk of change from all the clones that had to license a lot of patents, it's not like IBM let them 'steal', they licensed the relevant patents). IBM didn't plan that in the beginning, but they re
Re: (Score:2)
I don't disagree. I'm not talking about IBM themselves, but executives outside the IT industry who don't understand the nuances of the PC market, and the whole concept of the FOSS movement. Those folks think Apple is doing it right with a closed garden, and were surprised that Android phones outsold Apple, Nokia and Microsoft.
My dad can't understand why people would buy a phone from that hippy (Jobs @ Apple), when you can get a phone from solid companies like Nokia and Microsoft. He thinks Google can't k
Business People: a hard time focussing on mone (Score:3)
It's a common stereotype that the problem with business people is that all they care about is money.
If only that was the case.
The reality is we all tend to have some model of how things should be paid for and what makes our company different from another.
But, we always need to step back and look at it objectively.
Open Source is not some enemy of revenue on its own.
The old telecom companies (Bells, ATTs...) used to have all kinds of open source products. They knew their revenue was from having a monopoly position over communication.
This is very similar to Google today. They saw that they could be very friendly to open source as their revenue model was service/ad based. I'm sure there are bean counters at Google, but they're not simplistic bean counters who simply say people are using X amount of Google service, so they need to pay Y dollars.
I don't quite know the model for hardware companies. But perhaps just name recognition is enough. Sure with open hardware, anyone could make a copy, but most people, me included, would still pay for the name, to ensure it is done 'right'. I know I could buy a $20 router, but I end up with the Cisco/LinkSys/DLink...
Perhaps enough of a market develops that large companies start paying to support projects while reaping the manufacturing benefits.
I always see the negatives....... (Score:2)
Open Source Company/Business:
- Get others to do your work for you, claim glory.
- Promote the product for free, by getting others to do it for you.
- Create a license that controls the sale of all products that use the original source.
- No responsibility or legal worries for the company, blame the other guy.
All i see is the company benefiting, mostly?
I'am all for open source. But when a business comes into it, is it "really" open source?
These damned kids ... (Score:2)