If this is your first visit, be sure to check out the FAQ by clicking the link above. You may have to register before you can post: click the register link above to proceed. To start viewing messages, select the forum that you want to visit from the selection below. |
|
|
Thread Tools | Display Modes |
#1
|
|||
|
|||
open design practices and homebuilts.
I googled "open source aircraft" today, and got quite a few hits.
It appears that open source practices and hard-engineering (engineering that ultimately transcends from digital into physical, unlike software) are still pretty far apart from each other. There are several sites that refer to themselves as "open source" aircraft development sites. Widening the search shows that there is obviously a broad desire to adopt some of the best practices from software engineering into other engineering disciplines. I've seen lots of "we're hip because we are open engineering" garbage floating around, and after much revision of the wheel, it is still not rolling any better or straighter. I'd like to point out to anyone who is interested, that the grain of advantage in Open Source that most applies to other disciplines, is: effective distributed revision control. This is what products like Bazaar, and Apache Subversion do, and it is the primary reason why tens of thousands of hours of prior work, can be consolidated into a single project in a matter of minutes. While software like Wiki's and web forums work well for indevidual projects with simple file sharing, they do not permit cross collaboration between unrelated projects. If aircraft design was like software, you'd be able to take a wing design from any airplane, copy it, make some revisions, and save it into the design of a totally different aircraft, all without impeding the development cycles of either the original, or the new design. (yes I know, you can do this with some very expensive CAD software packages, but if you pay that much for software, your not going to publish outside of the corporation) You would also retain the option of including any new features added at later dates to either design into the other in a trivial way. Note that this _is_ possible to a large degree in homebuilt aircraft design. But only in the case that good revision control practices are observed. IMHO there should be NO MORE "open source" aircraft projects, until somebody gets around to implementing a public, vendor-neutral RCS (revision control system) for aircraft development, so that _any_ designer can submit and begin new projects and contribute to those of friends and peers. This is something that perhaps the EAA could sponsor? |
#2
|
|||
|
|||
open design practices and homebuilts.
wrote IMHO there should be NO MORE "open source" aircraft projects, until somebody gets around to implementing a public, vendor-neutral RCS (revision control system) for aircraft development, so that _any_ designer can submit and begin new projects and contribute to those of friends and peers. This is something that perhaps the EAA could sponsor? I doubt there is enough interest to make it worth EAA's while, or anyone else. If you are designing your own airplane, you probably (or most do) want to do a clean sheet design. If you know enough to be attempting your own design, then you probably don't need to slow yourself down trying to adapt someone else's details. You just do it yourself. Plus, many that are designing and building their own are not capable, or don't desire to put it all down in a computer. What does that really gain them? A couple good shop sketches, and start building, is their most likely course of action. If they do find the need to vary the design to accommodate some needed change, why would they want to take the time to go put the change into the computer. Only someone who is planning to publish the design, or otherwise go into business would do that, and they sure are not going to be interested in publishing the changes, and giving the whole thing away for free! I won't call you and others like you cheap *******s (g) because I don't know you. That is kinda' tongue in cheek, but I think the basic idea conveyed is correct. You just will not find a lot of people giving away something as valuable as an airplane plan for free, when they could be selling it. If you want to buy plans, there are boatloads of people willing to sell them to you for a couple hundred per pop. Buy a few plans and change and scale them like you need them to be, for your needs. Sorry I'm not more optimistic, and I really don't intend to offend. That just is the way I see it. -- Jim in NC |
#3
|
|||
|
|||
open design practices and homebuilts.
On Aug 27, 9:02*pm, "Morgans" wrote:
wrote IMHO there should be NO MORE "open source" aircraft projects, until somebody gets around to implementing a public, vendor-neutral RCS (revision control system) *for aircraft development, so that _any_ designer can submit and begin new projects and contribute to those of friends and peers. This is something that perhaps the EAA could sponsor? I doubt there is enough interest to make it worth EAA's while, or anyone else. SNIP Plus, many that are designing and building their own are not capable, or don't desire to put it all down in a computer. *What does that really gain them? Peer review, CAM support (and everything it implies), known revision history, peer contributed support documentation, Preexisting algorithms for doing structural, aerodynamic and flutter analysis, language portability, a common persistent knowledge base, and diversification of liability to name a few. *A couple good shop sketches, and start building, is their most likely course of action. *If they do find the need to vary the design to accommodate some needed change, why would they want to take the time to go put the change into the computer. *Only someone who is planning to publish the design, or otherwise go into business would do that, There are quite a few folks who use CAD as a primary design tool. A cocktail napkin sketch is usually as far as I go before CAD. and they sure are not going to be interested in publishing the changes, and giving the whole thing away for free! In open source environments intellectual property is considered a marketing resource, not a product. That doesn't preclude it from positively effecting revenue. What you typically find is that the development cycle itself becomes a pre-release marketing campaign, and subsequently once production is in place you have customers (who are often also contributors) already standing in line. If you were going to make a business out of it, the appropriate place for revenue generation would be in support. So for example, you could give the plans away for free, and put your cell phone on a 1 900 number to answer builders questions. I won't call you and others like you cheap *******s (g) because I don't know you. *That is kinda' tongue in cheek, but I think the basic idea conveyed is correct. *You just will not find a lot of people giving away something as valuable as an airplane plan for free, when they could be selling it. For small runs, (like plans) copyrights generally aren't very valuable. The ratio of litiigation costs vs. revenue recovered just don't add up. So the value of a plan is really the exclusive posession of it (which diffuses as popularity increases, since the probability of public release increases). It is worth noting that reselling plans you've bought, unless they are licensed non-exclusively, may be just as much a violation of copyright as making copies and selling them. It isn't the paper you pay for when you buy plans, but the ideas on it, and your license will generally restrict you from transfering the ideas, even though you own the paper. There are designs that are out of copyright, but people still sell plans for them. This is the same thing as reselling copies of Open Source CDs. You are selling the media only, the intellectual property isn't yours to sell. If you want to buy plans, there are boatloads of people willing to sell them to you for a couple hundred per pop. Buy a few plans and change and scale them like you need them to be, for your needs. Sorry I'm not more optimistic, and I really don't intend to offend. *That just is the way I see it. -- Jim in NC No offense taken. Please consider my words with the same grain of salt. My interest here is not in saving a few bucks, but rather creating a means for storing all of this knowledge that is dying off. There aren't that many people who know how to do this stuff, and a big honking public repository isn't such a bad idea IMHO. If it can be done in a way that is consistent with the best available practices in digital publishing, even better. There seems to be this assumption around that open source means no revenues can be generated. That is anything but true. There are hundreds of open source products that have revenue generating support infrastructure behind them. Often open source products create markets for other products that would never have existed otherwise. (a large percentage of Internet enabled products for example) Also, there is such a thing as people who WANT to contribute to the public domain! Should they be denied the right to do so? Some people think so, which is why the GPL came into existence in the first place. The core questions a 1. Would it improve the quality of the available intellectual property? 2. Would it improve the popularity of homebuilding? 3. Would it positively effect net revenues for business's in homebuilding? While there would be some winners and some loosers, I believe the answer to all three questions, over the long term, is yes. Matt |
#4
|
|||
|
|||
open design practices and homebuilts.
" writes:
This is something that perhaps the EAA could sponsor? Bwhaahahahahah! That's a good one. The EAA can't even sponser a decent web site. They're so old-fashioned they yearn for the days of the Brothers Wright. -- Among creatures born into chaos, a majority will imagine an order, a minority will question the order, and the rest will be pronounced insane. --Robert Brault, www.robertbrault.com |
#5
|
|||
|
|||
open design practices and homebuilts.
Matt,
I'm sorry, but I beg to differ. To start with peer review. Who are the peers? How many airplanes have they designed, built, and flown? Business? Most of the plans only aircraft are just that - plans only. Having built and test flown the airplane, selling plans is not a terribly expensive endeavor to get into. But it's not a big money machine either. How many sets of plans do you thing are sold for each design out there? How many airplanes do you think are built every year - kits included? Even the simplest airplane takes scads of time and money to develop. Are the "open source" people going to contribute to that? The core questions a 1. Would it improve the quality of the available intellectual property? Not in the least. You may re-draw the plans in CAD. But what does that accomplish? You have pretty plans, but has anything been added in the way of detail? If so, where did that information come from? 2. Would it improve the popularity of homebuilding? No. 3. Would it positively effect net revenues for business's in homebuilding? No. There are only so many people who will ever build an airplane. Completion rates are amazingly low. ANd that's not because the plans are hard to read. While there would be some winners and some loosers, I believe the answer to all three questions, over the long term, is yes. I'm sorry, but I beg to differ... Richard |
#6
|
|||
|
|||
open design practices and homebuilts.
On Aug 28, 4:29*am, Bug Dout wrote:
" writes: This is something that perhaps the EAA could sponsor? Bwhaahahahahah! *That's a good one. The EAA can't even sponser a decent web site. *They're so old-fashioned they yearn for the days of the Brothers Wright. -- Among creatures born into chaos, a majority will imagine an order, a minority will question the order, and the rest will be pronounced insane. --Robert Brault,www.robertbrault.com Actually, I like the new oshkosh365 site. It looks like they are migrating to a new content management system. It looks pretty nice, and there will likely be a solid API available for extending it. If not EAA, how about the Smithsonian? From what I understand they've got miles of old drawing and plans, and I'd imagine transliterating the most popular drawings into CAD would at least offset some postage. |
#7
|
|||
|
|||
open design practices and homebuilts.
"Morgans" wrote in message
... wrote IMHO there should be NO MORE "open source" aircraft projects, until somebody gets around to implementing a public, vendor-neutral RCS (revision control system) for aircraft development, so that _any_ designer can submit and begin new projects and contribute to those of friends and peers. This is something that perhaps the EAA could sponsor? I doubt there is enough interest to make it worth EAA's while, or anyone else. If you are designing your own airplane, you probably (or most do) want to do a clean sheet design. If you know enough to be attempting your own design, then you probably don't need to slow yourself down trying to adapt someone else's details. You just do it yourself. Plus, many that are designing and building their own are not capable, or don't desire to put it all down in a computer. What does that really gain them? A couple good shop sketches, and start building, is their most likely course of action. If they do find the need to vary the design to accommodate some needed change, why would they want to take the time to go put the change into the computer. Only someone who is planning to publish the design, or otherwise go into business would do that, and they sure are not going to be interested in publishing the changes, and giving the whole thing away for free! I won't call you and others like you cheap *******s (g) because I don't know you. That is kinda' tongue in cheek, but I think the basic idea conveyed is correct. You just will not find a lot of people giving away something as valuable as an airplane plan for free, when they could be selling it. If you want to buy plans, there are boatloads of people willing to sell them to you for a couple hundred per pop. Buy a few plans and change and scale them like you need them to be, for your needs. Sorry I'm not more optimistic, and I really don't intend to offend. That just is the way I see it. -- Jim in NC Agreed, and also a little more that I learned the first time that I specified a small project to be formed from sheet aluminum. There's a mildly embarrassing story that I tell on myself from time to time that illustrates just a little of the problem inherent in drawing up nice, plans to produce something that could readily just be hammered together and used. Sometime in the freshman or sophomore year of mechanical engineering, I took mechanical drafting--and I dare say that I was much above average. It happened that, in class, we were also taught how we might expect sheet metal to accept bends in the course of forming on a brake--which were taught that the determining dimmension would be the center of the cross section of the material with the result that the material inward from the center would compress and the material outward of the center would stretch. Some years later, I needed to specify the dimensions for a custom chassis for an electronic device. I confidently drew up the plans, using just a little of the fruit of that earlier education, and when the parts arrived the drill holes were all misplaced--by (drum roll) about half of the thickness of the material. Of course, I immediately challenged the folks at the shop that did the work, and immediately learned that (another drum roll) I had been taught pure popycock! The moral of that sad little story is that cad drawings, or really any drawings, which are supposed to permit a physical product to be accurately reproduced, need to be drawn and/or reviewed by people who are familiar with the way that such items are manufactured in the real world. Therefore, if anyone is serious about creating fabrication drawings, I would strongly advise that they start out with a few drawings of inexpensive sample parts that can be used to evaluate some of the discrepancies that could occur; and then pass the drawings, without any special explanation, to someone experienced in fabricating from drawings. The result will determine whether the drawings conform to real world practices. That is not to disparage the Open Source idea. At least in the case of computer software, Open Source seems to result in at least as good a product as closed source proprietary and has the added feature that the program and everything created from it won't become useless in the event that the originating company ceases operations. Just my rather long $0.02 Peter |
#8
|
|||
|
|||
open design practices and homebuilts.
On Sep 1, 2:04*am, "Peter Dohm" wrote:
"Morgans" wrote in message ... wrote IMHO there should be NO MORE "open source" aircraft projects, until somebody gets around to implementing a public, vendor-neutral RCS (revision control system) *for aircraft development, so that _any_ designer can submit and begin new projects and contribute to those of friends and peers. This is something that perhaps the EAA could sponsor? I doubt there is enough interest to make it worth EAA's while, or anyone else. If you are designing your own airplane, you probably (or most do) want to do a clean sheet design. *If you know enough to be attempting your own design, then you probably don't need to slow yourself down trying to adapt someone else's *details. *You just do it yourself. Plus, many that are designing and building their own are not capable, or don't desire to put it all down in a computer. *What does that really gain them? *A couple good shop sketches, and start building, is their most likely course of action. *If they do find the need to vary the design to accommodate some needed change, why would they want to take the time to go put the change into the computer. *Only someone who is planning to publish the design, or otherwise go into business would do that, and they sure are not going to be interested in publishing the changes, and giving the whole thing away for free! I won't call you and others like you cheap *******s (g) because I don't know you. *That is kinda' tongue in cheek, but I think the basic idea conveyed is correct. *You just will not find a lot of people giving away something as valuable as an airplane plan for free, when they could be selling it. If you want to buy plans, there are boatloads of people willing to sell them to you for a couple hundred per pop. Buy a few plans and change and scale them like you need them to be, for your needs. Sorry I'm not more optimistic, and I really don't intend to offend. *That just is the way I see it. -- Jim in NC Agreed, and also a little more that I learned the first time that I specified a small project to be formed from sheet aluminum. There's a mildly embarrassing story that I tell on myself from time to time that illustrates just a little of the problem inherent in drawing up nice, plans to produce something that could readily just be hammered together and used. *Sometime in the freshman or sophomore year of mechanical engineering, I took mechanical drafting--and I dare say that I was much above average. It happened that, in class, we were also taught how we might expect sheet metal to accept bends in the course of forming on a brake--which were taught that the determining dimmension would be the center of the cross section of the material with the result that the material inward from the center would compress and the material outward of the center would stretch. *Some years later, I needed to specify the dimensions for a custom chassis for an electronic device. *I confidently drew up the plans, using just a little of the fruit of that earlier education, and when the parts arrived the drill holes were all misplaced--by (drum roll) about half of the thickness of the material. *Of course, I immediately challenged the folks at the shop that did the work, and immediately learned that (another drum roll) I had been taught pure popycock! The moral of that sad little story is that cad drawings, or really any drawings, which are supposed to permit a physical product to be accurately reproduced, need to be drawn and/or reviewed by people who are familiar with the way that such items are manufactured in the real world. *Therefore, if anyone is serious about creating fabrication drawings, I would strongly advise that they start out with a few drawings of inexpensive sample parts that can be used to evaluate some of the discrepancies that could occur; and then pass the drawings, without any special explanation, to someone experienced in fabricating from drawings. *The result will determine whether the drawings conform to real world practices. That is not to disparage the Open Source idea. *At least in the case of computer software, Open Source seems to result in at least as good a product as closed source proprietary and has the added feature that the program and everything created from it won't become useless in the event that the originating company ceases operations. Just my rather long $0.02 Peter Thanks for the input Peter, I understand your point, that there are caveats in fabrication that are often overlooked in the design phase. But that is common in the design of everything really. Experience mitigates this over time, which is part of the reason for a good revision control system (RCS). Even with experience, a large project is going to have some revision, and this may simply be a matter of diversification in fabrications techniques from shop to shop. As you mentioned above you were tought wrong. But perhaps there is a fabrication process where what your were taught does apply? (invariably, there are some cases where different shops would produce differing products based on the same CAD, if only for a lack of exactitutde on the designers part) If you were using a revision control system, you would fork (duplicate) the design of the effected part only, and would subsequently maintain two documents, each notated to indicate their preferred manufacturing method. In this way, all future development work has the option of selecting which branch of the design tree applies to their situation. This is the core purpose of an RCS, which is what the OP was about. The biggest benefit of Open Source techniques is good revision control. The primary benefit of good revision control, is that caveats and exceptions are _retained_ in the design, so problems aren't experienced more than once. Incidentally, I've found a few aviation projects in public repositories that are currentiy using "subversion" as an RCS. So what I'm talking about is going on already at a seed level. I'm using "fossil" for some non-aviation related stuff, and hope to eventually use it for a homebuilt project. If so, I'll likely be converting an old out-of-copyright design into CAD, and then using that as a basis for revision. The most likely effect being the revision of a wood wing into aluminum. I don't have the experience, but if it's CAD, I can get review from many experienced designers, and they need not be local. With revision control, the full evolution of the design work is available, which will keep the next guy from making the same mistakes I make along the way. And if the next guy should make a few changes in his revision thats fine. It doesn't prevent my revision from being reproduced in it's exact form. Thanks! Matt |
Thread Tools | |
Display Modes | |
|
|
Similar Threads | ||||
Thread | Thread Starter | Forum | Replies | Last Post |
Take-off practices and procedures... | EridanMan | Piloting | 9 | November 17th 06 04:34 PM |
Take-off practices and procedures... | EridanMan | Piloting | 4 | November 17th 06 12:45 PM |
Student practices landing with gear up | steve | Piloting | 53 | August 19th 06 04:46 PM |