Posted by Gary Mortimer on November 29, 2009 at 12:34am
Chris might pull this post down fast when he sees it!!I am getting increasingly angry at the attitude of UNAV towards Ardupilot, most people know I don't use an Ardupilot, I'm not clever enough to put one together, but I like the project.http://www.rcgroups.com/forums/showthread.php?t=1030817&page=19I will let you guys read and decide.
In complex system development more human interactions is actually a good thing and I can say for a fact that it is actually encouraged in most environments. I work in an environment at the moment that does this and does it well. By using / having access to more people you inevitably pick up issues / problems much quicker. In-fact, the collaboration often takes place with people that have no background in the system being developed. This has the advantage of gaining a different perspective or range of perspectives.
What you talk about is one such model or method of complex system design. Just remember there are many of them, each with their advantages and disadvantages. Unfortunately it is counterintuitive to take a single approach for every circumstance.
Guys, I thought this was such an important debate, it deserved its own post, so I've created it here, starting with my own thoughts on the subject.
PS: Krzysztof. The code has matured tremendously over the past year. Please check out the 2.5 beta, which will be out in a week or so. I think you'll find it "worth re-using" ;-)
Thanks Bosak, I too did not understand few things you mentioned but tried to answer my perspective. It is like 4 different blind people touching the same elephant describe it differently , Ask 4 normal person( with 20/20 eye sight) looking out through 4 different windows of same house to describe the world outside, you will find 4 different perspectives too. I and other am sure value your perspective but may not agree to same belief. If there is any thing I can help you technically with altimeter problem then pls feel free to ask. BTW I don't have any AP currently and so no real world UAV expereince other then my will and wish to learn, have fun and help if I can. However I do have some real flying exp in UL, ML and LSA aircrafts.
I am not making anyone Richer for now :)) Cheers
'10 times more developers '
In complex system development, quantity of human interactions is the last thing you need. The community has many advantages like instant data gathering. But the number of loosely coupled/remote ppl participating in the project part-time is the last thing I would cite as an advantage. A community is the best in defining what has to be done, but then, there is the dirty job of keeping the gained grounds in technology domain (regression, testing, storage, change tracking). This is another very weak spot of ad-hoc community, no matter how devoted they are, nobody works full-time.
About Ardupilot code, I looked at it twice the last year, maybe 2h in total.
It is very good, but not excellent enough to be worth re-using and adapting for a typical small company.
For me what is excellent here is the occasional inspiration, but this comes in verbal or T3 contest form, not in the code.
Ithink we should all just accept that we are here for different reasons. Some, like me are here to have some fun and learn some new skills. Others however, are here for more commercial orientated reasons. One thing that does annoy me is people that are here for commercial reasons and have a wealth of knowledge, but never contribute any of that knowledge. I wish I had more knowledge to share, but alas my background is international relations / policy / politics. As i said before, this is a hobby for me. Something I can do to take my mind of everything else around me and relax, when things actually work that is!!! :)
There are no exceptions for the poor attitude of UNAV towards Chris. Critiquing a product is not a personal attack and to take it as such just highlights that the staff at UNAV are insecure about their product.
Open source and inexpensive hardware is always going to be a threat for the big dollar guys. If they aren't the only game in town, it's hard to justify the big bucks. So what makes up the difference in price? Profit and time. If the open source and inexpensive projects attract the right attention then in time it will be just as good as the big dollar guy's product.
The down side is ArduPilot's source code is out there for public scrutiny. The big dollar guy's stuff isn't.... so they can make lots of claims that they never have to back up.
So I'd suggest that ArduPilot just keep moving forward. I suspect DIYDroves has about 10 times more developers currently working on the project than any of the big dollar guys....
Morli, I would like to explain, but I am not sure I understand all of the phrases. Not that I wouldn't like to, but the altimeter I am working on needs some workarounds and no community on earth is going to solve this. In most cases your arguing is well thought and has many valid points...
Why are we having to justify the existence of our hobby or even DIYDrones?
This is another example of the corporate world attacking the worth of open-source, which is based on the resistance to such things as "bloatware".
To quote Opensource.org: "The promise of open source is better quality, higher reliability, more flexibility, lower cost, and an end to predatory vendor lock-in."
Corporate America does not have a copywrite on quality and innovation.
@ Bosak , Since you used my text as reference and I have few hours to kill, so here we go, :)) (, BTW I don't have spell checker in this system and so few typos here and there :)), and I don't have any gripe against any one) for the record
>>>There is one threat for companies coming from amateurs.
This is the scenario: claim?! you can do complex system! for pennies,undermine the positions of those who really did the effort and can count it, then just say it is not feasible withing budget and raise the price again.<< How many times has a Diydrones or any amateur /hobbiest company for that matter raised its price claiming " if you need ""complex"" system , pay us few K$ more More" ??!! No one I guess is here expecting Ardu or any hobby AP product to be any thing more than learn,fun,share,win race,laugh,feel satisfied to have atchived some thing more than a standard RC flying, In any case who gave any rights to anyone telling me that if I want to learn, use , have fun with UAV or technology , I need to go to this or that company alone , pay few thousands $ to just to figure out if the technology/product is suitable or no, meets my expectations or not, etc etc , still may only learn to use/fly uav , not the technology or inards of UAV????????? If so then every airline manufactrer should /will have gripe with aeromodellers, model plane/helli flyer who should actually buy a real plane from Professional plane maker to fly and have fun
Same as marketing that you can make Ardupilot for 25, 50, 100USD.<< I rather spend few 100$ & learn , have fun and even accept the disappointment that it does not meet my expectations or the product is no good!!! than spend few K to have the same experince.
This is like making the others look like the monsters of greed, while declaring the work of software developer is worth NULL, while not necessarily leading to the creation of competitive product in the long term. << who is "others" & Since when did Monopoly create any better and competitive product any where in the world?? & Since when did "software developer" alone matters in UAV and being the only " professional" in UAV industry whos time/effort is nullified because hobbiest is undermining his effort? If so then what would airline industry , pilots gripe about when may be in next 100 or 200 years when passenger+ cargo transport might go UAV too? BTW IMHO Hobbiest is also professional from same or another field but wants to have fun and satisfaction in learning, sharing with other similar minded people.That is why it is called Open source. Unhappy of being software developer? choose another profession buddy. Knowledge is not exlusive domain where only privy few "Professionals" get paid "High" to develop!! what ever.. while every one else needs to wait and spend few K to learn about it!!
All this has one effect: creates artificially crafted virtual reference point for price, << who sets the value /price of product?? and what is real reference point for cost of any technology/product/knowledge ?
that is used to dramatically weaken the public perception or the makers of complex electronics, << who is public?? and why is it complex??
without even making the makers of cheap solutions incredibly rich.<< There!, you have you answer, make it cheap and sell few thousand APs and become rich!
At the end there is no progress, as the advance of open source is usually quickly swamped by absence of professional regression testing and all the things nobody would do for free. <<< You mean to say every one of hobbiest here is wasting his time buying hobby AP,building, writing their experince,sharing tips, tricks, issues, etc etc ?!! Therefore, if the development of open source project is not really matched in pace with the pack of companies supporting it (for its own economic reasons), usually wanders into complexity black hole. << ?? :0
On the other hand, it is hard to contribute to a project whose final, non-regressing advantage is low price.
So there is a thin line between claiming progress and declaring naive target price (which is equivalent to making false effort estimation). <<How is price for a effort estimated?
If the hobbyists cannot estimate true effort of the projects they are starting, this is even more true for potential clients of bigger companies.<<< ??!! , Estimate? Why would a hobbiest estimate his project?! I thought hobby is for fun, take a break and some thing to take his mind off from daily chores.. in any case if it is big bite for a hobbiest to chew then hobbiest abandons his project and doesnot waste sleep and time over it, he moves on.>>
Therefore the next thing after imposing export ban is to create perception that the work on autopilots is almost worthless, to make sure that US will start losing the UAV race one day.<< Isn't this( open source forum) one way to defeat such ban or monopoly??why is US or any single country or entity alone privy to UAV technology and why is it a race?!! In any case what is your gripe? You are manufacterer of Complex Product who gets paid for " professional" development/ed product which is not winning any " race" any way. Am I to assume that since any professionally developed complex machine isn't winning any race, you are here to see why is it so?!! .
Did I miss some thing coz I did not see Ardu pilot in topper/winner list of T3 race?!!
Last but not least(for now) , how many of arduAP or any similar product is flying complex missions by mil or gov( who alone can afford few 100K$) any where in the world that is so much concern to a commercial company/ies well developed product by teams of very well paid professionals costing realistic price is loosing it marker too??? cheers
'I think Krzysztof has a lot to gain by having UAVs go mainstream through Ardupilot.'
That's right and this is why I like contests even if I am far from winning them.
Comments
In complex system development more human interactions is actually a good thing and I can say for a fact that it is actually encouraged in most environments. I work in an environment at the moment that does this and does it well. By using / having access to more people you inevitably pick up issues / problems much quicker. In-fact, the collaboration often takes place with people that have no background in the system being developed. This has the advantage of gaining a different perspective or range of perspectives.
What you talk about is one such model or method of complex system design. Just remember there are many of them, each with their advantages and disadvantages. Unfortunately it is counterintuitive to take a single approach for every circumstance.
PS: Krzysztof. The code has matured tremendously over the past year. Please check out the 2.5 beta, which will be out in a week or so. I think you'll find it "worth re-using" ;-)
I am not making anyone Richer for now :)) Cheers
In complex system development, quantity of human interactions is the last thing you need. The community has many advantages like instant data gathering. But the number of loosely coupled/remote ppl participating in the project part-time is the last thing I would cite as an advantage. A community is the best in defining what has to be done, but then, there is the dirty job of keeping the gained grounds in technology domain (regression, testing, storage, change tracking). This is another very weak spot of ad-hoc community, no matter how devoted they are, nobody works full-time.
About Ardupilot code, I looked at it twice the last year, maybe 2h in total.
It is very good, but not excellent enough to be worth re-using and adapting for a typical small company.
For me what is excellent here is the occasional inspiration, but this comes in verbal or T3 contest form, not in the code.
There are no exceptions for the poor attitude of UNAV towards Chris. Critiquing a product is not a personal attack and to take it as such just highlights that the staff at UNAV are insecure about their product.
Keep up the great work Chris!
The down side is ArduPilot's source code is out there for public scrutiny. The big dollar guy's stuff isn't.... so they can make lots of claims that they never have to back up.
So I'd suggest that ArduPilot just keep moving forward. I suspect DIYDroves has about 10 times more developers currently working on the project than any of the big dollar guys....
This is another example of the corporate world attacking the worth of open-source, which is based on the resistance to such things as "bloatware".
To quote Opensource.org: "The promise of open source is better quality, higher reliability, more flexibility, lower cost, and an end to predatory vendor lock-in."
Corporate America does not have a copywrite on quality and innovation.
>>>There is one threat for companies coming from amateurs.
This is the scenario: claim?! you can do complex system! for pennies,undermine the positions of those who really did the effort and can count it, then just say it is not feasible withing budget and raise the price again.<< How many times has a Diydrones or any amateur /hobbiest company for that matter raised its price claiming " if you need ""complex"" system , pay us few K$ more More" ??!! No one I guess is here expecting Ardu or any hobby AP product to be any thing more than learn,fun,share,win race,laugh,feel satisfied to have atchived some thing more than a standard RC flying, In any case who gave any rights to anyone telling me that if I want to learn, use , have fun with UAV or technology , I need to go to this or that company alone , pay few thousands $ to just to figure out if the technology/product is suitable or no, meets my expectations or not, etc etc , still may only learn to use/fly uav , not the technology or inards of UAV????????? If so then every airline manufactrer should /will have gripe with aeromodellers, model plane/helli flyer who should actually buy a real plane from Professional plane maker to fly and have fun
Same as marketing that you can make Ardupilot for 25, 50, 100USD.<< I rather spend few 100$ & learn , have fun and even accept the disappointment that it does not meet my expectations or the product is no good!!! than spend few K to have the same experince.
This is like making the others look like the monsters of greed, while declaring the work of software developer is worth NULL, while not necessarily leading to the creation of competitive product in the long term. << who is "others" & Since when did Monopoly create any better and competitive product any where in the world?? & Since when did "software developer" alone matters in UAV and being the only " professional" in UAV industry whos time/effort is nullified because hobbiest is undermining his effort? If so then what would airline industry , pilots gripe about when may be in next 100 or 200 years when passenger+ cargo transport might go UAV too? BTW IMHO Hobbiest is also professional from same or another field but wants to have fun and satisfaction in learning, sharing with other similar minded people.That is why it is called Open source. Unhappy of being software developer? choose another profession buddy.
Knowledge is not exlusive domain where only privy few "Professionals" get paid "High" to develop!! what ever.. while every one else needs to wait and spend few K to learn about it!!
All this has one effect: creates artificially crafted virtual reference point for price, << who sets the value /price of product?? and what is real reference point for cost of any technology/product/knowledge ?
that is used to dramatically weaken the public perception or the makers of complex electronics, << who is public?? and why is it complex??
without even making the makers of cheap solutions incredibly rich.<< There!, you have you answer, make it cheap and sell few thousand APs and become rich!
At the end there is no progress, as the advance of open source is usually quickly swamped by absence of professional regression testing and all the things nobody would do for free. <<< You mean to say every one of hobbiest here is wasting his time buying hobby AP,building, writing their experince,sharing tips, tricks, issues, etc etc ?!! Therefore, if the development of open source project is not really matched in pace with the pack of companies supporting it (for its own economic reasons), usually wanders into complexity black hole. << ?? :0
On the other hand, it is hard to contribute to a project whose final, non-regressing advantage is low price.
So there is a thin line between claiming progress and declaring naive target price (which is equivalent to making false effort estimation). <<How is price for a effort estimated?
If the hobbyists cannot estimate true effort of the projects they are starting, this is even more true for potential clients of bigger companies.<<< ??!! , Estimate? Why would a hobbiest estimate his project?! I thought hobby is for fun, take a break and some thing to take his mind off from daily chores.. in any case if it is big bite for a hobbiest to chew then hobbiest abandons his project and doesnot waste sleep and time over it, he moves on.>>
Therefore the next thing after imposing export ban is to create perception that the work on autopilots is almost worthless, to make sure that US will start losing the UAV race one day.<< Isn't this( open source forum) one way to defeat such ban or monopoly??why is US or any single country or entity alone privy to UAV technology and why is it a race?!! In any case what is your gripe? You are manufacterer of Complex Product who gets paid for " professional" development/ed product which is not winning any " race" any way. Am I to assume that since any professionally developed complex machine isn't winning any race, you are here to see why is it so?!! .
Did I miss some thing coz I did not see Ardu pilot in topper/winner list of T3 race?!!
Last but not least(for now) , how many of arduAP or any similar product is flying complex missions by mil or gov( who alone can afford few 100K$) any where in the world that is so much concern to a commercial company/ies well developed product by teams of very well paid professionals costing realistic price is loosing it marker too??? cheers
That's right and this is why I like contests even if I am far from winning them.
-
1
-
2
-
3
-
4
of 4 Next