h a l f b a k e r yFlaky rehab
add, search, annotate, link, view, overview, recent, by name, random
news, help, about, links, report a problem
browse anonymously,
or get an account
and write.
register,
|
|
|
Similar to open-source software and other
media, the open-source car would be
engineered, designed, and cross-checked
by an army of contributors working for
free.
Amateur engineers with some spare time
and modeling software could each design
a valveseat, a bearing, or a rearview
mirror.
A few proffesionals would still be
needed to put everything together and
make sure it works, is safe, and to make
other important decisions.
Like open-source software, this system
would give the customer a cheaper
product along with the pride of knowing
that they contributed something to the
project.
OSCAR
http://www.theoscarproject.org/ One of the earliest and still active open source car projects [django, Aug 26 2006]
The Open Source Car: A Design Brief
http://www.makezine.com/01/car/ "The time is right for a true people's hybrid vehicle. The web is peppered with how-to sites for converting your old car into an electric vehicle, but why not develop SourceForge-style documentation for an open source hybrid?" [django, Aug 26 2006]
Open source everything
OpenHumanity Why stop with cars? [ixnaum, Aug 27 2006]
Open Source Car on the Horizon
http://hardware.sla...2030246&threshold=1 slashdot.org discussion [xaviergisz, Dec 09 2006]
[link]
|
|
hmm, I'd have thought the hardest thing about making a car would be the making of it, not the design. |
|
|
So whom do I go to when my brake cylinder cracks at 75MPH because the amatuer engineer who designed it forgot to take vibrational stress into account when he designed it because it was the first part he ever designed. |
|
|
Open source works well with software, Not so good with actual products that get built. |
|
|
Even though it's overbaked [see links], I give this a croissant simply because google indexes the Halfbakery posts so it's one added reference for this great idea. :-) |
|
|
Open source generally applies to something that is hidden from view(ie Source Code) With cars I can simply go buy one and look at it and I can then discern how it is built and then develop improvements and add ons(as is done by every aftermarket parts provider) Im not really sure that you can do open source with a car. |
|
|
I think in reality all cars already are Open source. The underlying concepts of the cars operation and requirements are availble to all. DIfferent companies choose to develop thier version of the product and if possible sell them to the public. Other developers work on improvement and share them with others for free(shareware) A good example would be all the people posting instructions to turn your diesel into a veggy burner. Tinkerers slave away in thier garage inventing new concepts that are reviewed by everyone else. How is what you are proposing different than what already happens. |
|
|
WHat if a manufacturer were to release plans for thier car? What could you possibly do with them, Build your own ferarri from scratch? |
|
|
**edit: [kumpf] has deleted several annos, with the result that my replies look like whiny rants. I'm going to leave them, though, as they make points pertinent to the discussion.** |
|
|
/held legally responsible for what they sell// |
|
|
If you're talking about developing a car from scratch, therein lies the weakness. |
|
|
Even if there were individuals capable of performing the vast amount of software modelling that goes into ensuring every single little piece of a car is strong enough and safe enough under all possible conditions, and just supposing they were able to afford the tens of thousands of pounds/dollars/WHY necessary to purchase the licences to run that software, no lawyers are going to allow their clients to make a car designed by amateurs. And most automotive professionals will have a clause in their contract stating that any job-related brain output belongs to their employer. |
|
|
Thus while I could happily design you a brake system, you'd be a fool to to trust me as I haven't done so for over 11 years. I've not kept up with the latest legislation, and I have no way of knowing if the components I design for you are able to withstand the forces the car will experience under heavy braking without making some and testing them. I can't afford that. Neither can you, and why should a car company do it when they have suppliers capable of doing the same thing under a contractual obligation to get it right? |
|
|
I could design you an engine mounting, or a suspension link bushing, but then I'd lose my job. |
|
|
There's a reason why there are so few car companies. Cars are incredibly dangerous and highly uncomfortable machines. It takes a vast pot of money and literally thousands of people working for hundreds of component suppliers to ensure every single aspect of their fit, form, and function meets varied legislation from around the world designed to ensure that a car starts off safe, remains that way, and is acceptable to customers with ever higher expectations of comfort, convenience, and reliability. |
|
|
As a weekend project for a one off it has appeal, but I can't see it happening for mass production. |
|
|
How do you even hope to get anything for around a thousand bucks without mass production? |
|
|
And do I have to sign a waiver too before I use the same roads you do? In a society so frightened of legal action that packets of nuts carry a warning that they "May Contain Nuts" and coffee cups warn us that "Contents May Be Hot", how are you going to get insurance? |
|
|
"Geiko.com, how may I help you? And is this car type approved, Sir? Oh, you "imagine it's designed quite well"? Fine sir, we can insure you for up to $10million third party liability, your annual premium will be...uh...$11 million." |
|
|
As a professional engineer who has spent his whole working life in the automotive industry, it really pisses me off that people think cars are easy just because they're everywhere. They are incredibly complex pieces of machinery. Would you entertain the same method of design for a public building, or a commercial aircraft? |
|
|
Although looking through the HB "Car" category, it looks like we're already most of the way there... |
|
|
Software is relatively cheap and easy to test - there's only one way to test a car, and that's to build one. That is VERY expensive unless you start using available components from existing vehicles. I can see it might work for adaptive vehicles, but my argument all along has only been that it won't work for starting a vehicle from scratch without massive financial backing from somewhere. |
|
|
Im pretty sure that Microsoft does not allow you to take its software and then call it your own software, It allows you to see what makes its software work so you can design items to work smoothly with it. This relationship already exisits in the aftermarket parts industry, Manufacturers cooperate with parts producers to develop add on components for thier vehicles. This is the esscence of an open source environment. |
|
|
It's just a matter of accumulating enough open infrastructure to support this. The design and testing software would have to be open source. Same with the control software for the robotic tools used to assemble this car. In the end everything is just "information". The rest are a bunch of raw natural resources which become "organized" into a shape of a car.
I believe open source cars are not coming yet (not enought of the underlying infrastructure is opened yet) but they are definitely coming. If a video card could be open sourced why not a car?
Lot of other things are going to be open sourced (take a look at my shameless plug - link) |
|
|
What happens to all of the people that currently get paid to design car parts when their jobs can be outsourced to bums surfing the internet and doing things for free? |
|
|
Hmm... What do all the people that design car parts for free in their spare time do for a living? Halfbaking doesn't exactly pay the bills... |
|
|
I'm not sure how useful the
software analogy really is. The
concept of "crashing" is somewhat
different for starters. |
|
|
This has been done, admittedly old school. International Harvester. |
|
|
Well, if this thread is anything to go by, then yes, the way to an open source car is to pool our collective ignorance, make some wildly naive assumptions, and ignore anyone who actually knows what they're talking about. Should be a breeze. |
|
|
There would definately be problem
implementing this plan. I admit that it
would be all but impossible for
amateurs to design a safe and reliable
car in their spare time. |
|
|
However, as long as the amateurs are
designing simpler, non-vital parts such
as trunk latches and rearview mirrors, I
don't see why this plan wouldn't work.
And, as design software becomes
cheaper and more accessible, more
people will be contribute more and
more ideas. |
|
|
I think the problem that [egbert] has somewhat forcefully stated is that, although you may be able to get amateurs to do your rear view mirrors, you aren't going to be able to get professionals who'll do your engine mounts free of charge. They're paid by car manufacturers who won't let their employees work on similar stuff for another organisation. |
|
|
Sorry to wade in again, but [phlish], that is a hopelessly naive understanding of mechanical engineering. |
|
|
A case in point; on one of the products my ex-employer used to make, the radius in a particular corner of a pressed (stamped) part was crucial to the overall function of the assembly as it controlled the position of a mating part. However, the original size of the radius was lost during assembly due to a mechanical deformation (crimping) process. Anyone reverse engineering that assembly would not have known about this feature and would have missed it out. |
|
|
This is a particular example of a very common situation, where small details, tolerances, surface finish, even a tighter than normal material spec, are all important features of a design. Miss them and the function, design life, and even safety of a part can all be compromised. |
|
|
I don't see why this idea is getting such
a bad reception. There are precedents
for amateur engineers and cars as a
general concept are fairly old-tech. |
|
|
The biggest problem is one that is
raised by egbert: that of staying the
right side of current safety and
environmental legislation. This might
find a solution by involving car
companies. If one of the aims of the
project were to create a simple, cheap
safe environmentally friendly car that
could be manufactured and sold in the
third world without any intellectual
copyright issues then they might come
on board. |
|
|
Maybe I'm too much of an idealist. |
|
|
I just dont think that this is "open Source" The definition of open source implies some base upon which one builds custom accesories, in which the "base" is not normally open for public scrutiny. This idea and the concept just dont fit the model. Now if we were talking about commitee based or community based car design in which everyone gets to have a say then sure, but this is definatly not an "open Sourcable" thing. |
|
|
//you can take it apart and figure out how its made // |
|
|
You can do with closed-source software - source code just makes it a lot easier. I see the analogy here being the specifications and engineering drawings. |
|
|
//Open source generally applies to something that is hidden from view// |
|
|
Engines not hidden ??It seems to me car manufacturers go out of their way these days to cover up the engine workings. |
|
|
//So whom do I go to when my brake cylinder cracks at 75MPH// |
|
|
Whoever you bought the car from. |
|
|
The mechanism by which the car was designed can be separated out from the liability for any malfunction in the unit you bought. |
|
|
//a car designed by amateurs// |
|
|
Just because the design is freely available, doesn't mean the designers work for free; or that they aren't qualified. |
|
|
Having said I agree with the points about actually making and testing the thing. |
|
|
But as a design-strategy; I don't see this as far-fetched at all. Most car companies currently share technology and components - this would simply be another process adding to this - a system of 'open-engineering' where people could spot improvements and flaws. |
|
|
Standard safety laws would apply just as much to a vehicle produced this way as a vehicle produced behind closed doors. |
|
|
This could be empowering for developing countries; it could be empowering for the environment lobby - who could plainly see (given access to the designs and the relevent expertise) where engineering inefficiencies and chances of improvements. |
|
|
"Hey Honey - I can't get the car to start... I need to take the kids to school" |
|
|
"A schoolrun? The car isn't set up for that - you'll have to re-compile the engine..." |
|
|
I think perhaps [raspberry_retart] is using the term 'open source' in a somewhat metaphoric sense. It is somewhat akin to my own thinking on the subject, if the intention is to release the design of cars from fiercely-guarded corporate control and instead to involve the creativity of a large number of people. Personally I hold this to be socio-politically desirable. |
|
|
In the absence of the closed 'base' [jhomrighaus] refers to, might I suggest the age-old phenomenon of tradition? In some sense this is already happening in the street rod industry. Translated into fashionable language one might see the design of a stock 1932 Ford three-window coupe as a set of component interface specifications with, of course, certain other physical requirements. It may be represented as a sort of tree or net, from which one may take out branches or sections of branches, and design something to fill the resulting space. Tradition shows what has worked thus far, and therefore what might work in somewhat different ways. The key to the traditional model is that the interface set (or base) is not a corporate product but something that exists culturally in a community. This might be strengthened if it is published in some neutral form, preferably but not exclusively on paper. |
|
|
Another factor is that designs that are more like Meccano and less like jigsaw puzzles lend themselves better to tradition-based development - which answers some of [egbert]'s concerns. In that sense the Model T and Model A are preferable to the '32. |
|
|
I think that parts of the light-aircraft industry and much of the sailing-boat industry work something like this. |
|
|
services to open chains off reruns to less inferior
value sets, which dominant, lee sencitive, or
eeven a -over-robust say concret blocked shit
head, kommunist, cell chess player party/firm
earth eating wirm bacteria, chief's, of destruction,
a lighter, atsmosphere, is a singlularity, of
impossible, calculations,, So, I have now, kareless
ness been, in a houmorless way, been killed, say
'again',.. |
|
|
returning depressents over head 'pyramid' thinking
of multiple planet's perfecting
multidicipline,/spline, treads, could be planned or
thought OFF, this way,,, :-) ,,,.. |
|
|
According to the Oscar website, it started in 1999
and so far they are up to version 0.2, which seems
to be a motorised recumbent trike with chairs in
it, and which also seems to exist only on paper (or
screen). |
|
|
I think "open source" is probably the worst way to
go about designing anything. If you have stringent
controls and standards, then it is instead merely
the worst way to go about designing anything with
more than one component. |
|
|
And I don't see how this is meant to produce a
cheaper car than the equivalent type of car
produced in the normal way. The manufacturing
costs will be at best the same, and more likely far
greater because the guy designing the rear spoiler
didn't talk to the guy designing the boot lid, and
therefore didn't realize that they could be nicely
integrated into one panel instead of two. |
|
|
I want my car designed by a trained and closely-
knit team of designers who know what they're
doing, and who are under pressure to design
something that can be made in large numbers on a
normal production line. I also want it to be
designed by Jaguar, but that is merely a
gallendacious aside. |
|
|
//in reality all cars already are Open source.// A
beautiful idea. Not completely true, though,
according to [egbert]. If industrial products of the
complexity of cars were truly open source, then
Shirley the USSR would have had an easier time
(less cost, better results) imitating Western
technlogy? |
|
|
Moreover: Could any modern engine run
without its computer running proprietary
software? |
|
|
Not to mention: open source depends on users to
report bugs, which are fixed in the next release.
Even granting that it asymptotically approaches a
bug-free state, it does not do so unless many
people agree to use the earlier versions. It's easy
enough to mess around with buggy software
without risk, and lots of people enjoy doing so.
However, testing a buggy car -- where failure
could kill you, without special precautions -- is
probably a more expensive and technically
demanding proposition, and certainly one which
fewer people would undertake. |
|
| |