Thursday, August 30, 2012

Fourty-Seven: short, aiming for 50

Getting close to 50 rants, and I still haven't done much.

I downloaded freeCAD and a project management software from sourceforge. I got a little bit stuck with studies but that's just me being lazy. I still watched all the lectures, but haven't got down to working on further assignments yet.  :)

I just wanted to look at software out there check out their designs and capabilities, it hopefully will help clear up my mind about the design of "freespaceships". 

First project I am planning to kill some time in is a robotic explorer of Earth. :) I would love to have a UAV of my design check out the Antarctic so I think I might go in that direction. I have always been interested in flight, I love uavs, I love planets, and the closest one I could explore more is this one right under me. :)

Though that realisation came to me during a Life /the crime drama series/ inspired zen moment. I love space related stuff and I want to go up there to explore other planets and solar systems and such. But really I am already in space, on one of the planets that are in space. :) So in a sense I don't need a spaceship or tons of technology to explore space. I am just limited to this planet and my close surroundings for now. With my limited budget I could do some activities in this field as a hobby while I am trying to develop and/or become part of the "open source space programs" movement.

At any rate I started thinking about and looking up UAVs and trying to figure out a way I could cheaply make and deploy one that can go to the Arctic or the Antarctic and snap some photos. Maybe later on return samples, or do continuous exploration and sampling for longer periods of time.

The key similarities between a project to send hardware to the arctic with sending hardware to the moon or other planets, that makes it a good stepping stone for noobs like me:
-Still pretty harsh environment. The problem isn't trivial.
-Large distance. Though not as quite as large as distance between planets, but large enough to make remote operation a hassle.
-Exoticness. It's a place very few man has gone before. :) And I don't know about hobby level robotic explorers but I think it's a field still quite pristine.

So this is what I have on my mind at the moment. I am thinking of balloons, blimps, winged aircraft, boats, rovers, and various hybrids of them all. I am thinking of various capabilities, like withstanding strong winds, frost, low temperatures, changing modes between floating and flying, or flying and rolling/roving. I am thinking of autonomy and teleoperation, communication, energy source, costs, design procedures, construction etc...

I am thinking of how to taylor the project so that noobs like myself could carry it out with some time, money, and smart ideas. :)

So yeah, that's what's going on at the moment.

There is also another interesting development in my industry. They are about to test remote controlled aircraft for aerial spraying of weeds. I have been ranting to my coworkers about robotics and how awesome it would be to have drones do some of our work, and they are already trying to make use of remote controlled vehicles. The next step is semy or fully autonomous drones and I am out of a job. :) This is one of the incentives for me to get my act together and get some engineering done. The world is changing and those who don't change with it are left behind.

I am so excited about the future prospects of UAVs I can't wait to build one. :)

 

Monday, August 27, 2012

Fourty-Six: mistakes, learning

Mistakes:
Today I just watched a TED talk about talking about your ambitions or goals to other people. Apparently it reduces the likelihood of your success.

Keep your goals to yourself!

It actually explains partially why I become so much calmer since March. You can see in the blog activity how there is a large peak in the amount of posts right at the beginning and then it all declined.

It could be attributed to the fact that starting up a space program is not easy, and so when I faced difficulties like actually designing a system I had in my mind, the enthusiasm dropped and with that activity dropped.

It could be attributed to simply realising that blogging about something isn't going to make it happen. Which is why after trying some other things I went back to try and learn programming so I could try and design the system I would enjoy to play or be part of for the rest of my life that would also pump out tangible results in terms of getting a whole lot of people off the planet.

But it also could be that by simply talking about these ambitions my brain got tricked into believing that it achieved the actual goal, it got some level of reward of success, and now it's back to being a normal brain that doesn't rant at the Internet every day about going to the moon. :)

In reality it is probably a combination of all of these and other factors.

Anyway, so there is my mistake. I talked about freespaceships.

Learning:
There is so much to learn, not enough time for it. Or at least that is my excuse for not doing enough of it. :)

So my machine learning course started last week, the third week of Algorithms is starting tomorrow, and Gamification is starting on Monday. I am going to have serious issues with keeping up without good time management. Luckily though it's all free, so the only thing I am losing is actual knowledge I don't acquire by not keeping up with the course.

Still kicking though, so I haven't given up.

I did that thing when I write part of a blog then just save and post it later. I wrote the above bits Saturday night, now I am continuing. :)

So yesterday I watched the Octave videos for machine learning cause I already watched the linear regression ones weeks ago. Played around with Octave and matrices a bit, and I think I am getting quite excited about using stuff learned in the course for projects that I have in mind. I need to re-watch the first couple of lectures, but I think I understand the concepts to an okayish level. :)

Algorithms is going alright. I managed to watch the second weeks lectures this weekend. First programming assignment I managed to get to 84% or so. I could fix it up more but it's hard to find time for programming. I am a week behind with it all, this week I need to do the second one, hopefully on Tuesday and Wednesday I can tackle it to some satisfactory level.

Then Gamification started today and I jumped straight in and watched all the videos. Another subject I can't wait to make use of.

I don't know how it all will work out, but I am having a lot of fun learning new things, hopefully soon I can actually start doing something useful.

More mistakes:
So apart from "freespaceships", I am hoping to use my new knowledge for "environmental robotics".

I am thinking about how robotics/software/AI could be used to increase efficiencies in the environmental industry. Specifically in monitoring, management, and rehabilitation of our natural resources to healthier states. River systems, national parks, farming landscapes, coastlines, mountain ranges, forests, deserts, swamps etc...

We already use a ton of technology, especially in monitoring. Like satellite and aerial photos, machines, computer systems etc. but seeing how all the other industries have made great use of robotics for increasing productivity and improving efficiency, I am hoping/anticipating that the same thing will creep into the industry I work in.

I wouldn't mind being part of the process, but also I think if I want to work in this industry long term, it doesn't hurt to be prepared for what's to come.

I saw a TED talk about Open Source Ecology, a project of open source farm equipment design that can be built from scratch. I got really excited because I can clearly see in my mind how both "freespaceships" /or it's equivalent/ and also "environmental robotics" could be done exactly the same way. Simply building a database of designs that anybody can freely implement or modify.

Then in comes utilisation of AI, Gamification, Crowdsourcing etc...and we have ourselves a whole new world where tea parties on the moon are not only possible, but are also last century and not really cool anymore. :)

Anyway, I just wanted to post something again to make sure that any potential space programmer out there will see I am still here and still active. Still working towards the insane goal of setting my foot on the moon as part of an exodus that was made possible by large amounts of people cooperating for the sake of reaching a goal, and not some tangent completely irrelevant purpose. /like stuffing pockets with paper/

Open Source Space Programming FTW! :)




Sunday, August 19, 2012

Fourty-Five: networked thinking

My first week at coursera went kind of ok. It was hard to find time to watch the lectures and work on assignment, but I managed to do half of the assignment and I think it won't take long to finish it off. But overall I enjoyed it a lot so far. I loved the feeling of accomplishment when my lame code actually worked as it was supposed to. :)

Then I was also thinking about "freespaceships" the network that thinks up free spaceships. :)

I mean if we individual humans are technically computers. Probably all nerds would agree. And if our computers are computers. And if our network of computers the internet itself is thought of as a computer, then really "freespaceship" is a software incorporating software running on actual computers, software running on networks of computers, software running on human brains, and software running on all of these at the same time. If thoughts are "software", then really "freespaceships" is about thinking up free spaceships.

It sounds pretty incoherent, but this place is for expressing my incoherent ramblings so NI!

Anyway here is some attempt at more coherence:
So I was thinking of a layered network. It would have a network of social interactions, with nodes of individuals or groups like hackerspaces, or organisations.
Then there is a network of actual solutions. In my head I imagined all hardware dreamed up and designed placed in a network where parts are linked together, and so it can be visualized where each component is used, or could be used. Also with building a database of engineering solutions it might be possible to make it easier to come up with new designs.
Then there is a network of resource allocation which is responsible for moving stuff around. Actual hardware, raw materials, components, money, maybe even people.

Now I am wondering how to put all that into one open source client. Something simple, clear, easily modifiable and upgradable.

I think google+ has a lot of potential as serving as the social networking core. The different design of how one can organise their connections allows for interesting structures. Being able to separate connections into different groups and sharing things selectively could potentially help with clearing up the clutter of random stuff as long as people stick to the topics of the circles they share stuff "into".

It will be interesting to see how it all works out in a couple of years.

It also seems to give great opportunity for spamming. Like what I am doing with this post right now. :P I am sharing this with all the new people who added me cause of Algorithms prt 1. Maybe I can infect some more people with the idea of an "open source crowd done space program".

I need more enthusiastic programmers or programmer wannabees like myself. After all this problem is largely a software development problem as I described before. It is software running on our brains + on our computers + on the computer that is made of our brains linked to our computers linked to each other. This brain computer hive thingy we are all part of but aren't quite aware of much.

So viva le hive mind revolution! :P WE ARE THE BORG! RESISTANCE IS IMPOSSIBLE!

:)

freespaceships FTW!

I wonder if there is going to be a spike in the site visit stats after I post this. If there is, it shows google+  and some extra amount of new people in your circles helps promoting silly ideas.

If there is no spike, then I will know need more people in my circles.

Other stuff:
my Machine Learning course is starting this week too, I hope I will meet some more cool people through that course. Also I can't wait to make use of the AI algorithms and hopefully in a year or so I can start building freespaceships properly with real software engineering and coding done. :P WOOHOO.

Saturday, August 11, 2012

FOURTY-FOUR: impossible, open thinking, study

 //I have been sitting on this rant for a while now because I keep disagreeing with myself and what I have written. I decided to post it anyway, along with the current update of what's going on with "freespaceships".

It is the general opinion about a mass collaborative space program that it is impossible. Well judging from the responses and the general lack of responses I got from people including the "crowd" I spammed on space.com.

There are legal  issues with open designing hardware, there are economical issues of gathering funds and resources to carry out construction, maintenance, and operation of hardware, there are technological issues of actually doing things in orbit or on the surface of objects in space.

Truckload of problems.

So how can one person solve all those  problems? Which is what I am trying to do.

The short answer is: There is no way a single person could do all of that.

The long answer is: Follow the example of problem solving know hows in history or in our current times and just simply break it all up into tiny little pieces of problems and solve them one at a time.

The problem is, breaking it all up into tiny pieces is in itself a problem.

I cannot see the actual problem field because I am not an expert in any of the broad fields or sub fields that are encompassed by legal, economical, and technological fields.

 So two obvious solutions are to become an expert in all the fields, or to outsource the whole thing to experts out there, but neither can be accomplished without serious time and money investment which I cannot afford. //actually I can afford some of it, this is why I enrolled in some coursera courses, see how much of an "expert" I can become in the following years//

To become an expert I would need to devote time that I am spending on trying to support myself and my wife by doing actual work. To outsource the problem to experts I would need to devote time to find the experts, find  out the specific problems, and convince the experts to solve them somehow. This is usually accomplished by offering money, which would require me to become substantially more wealthy than I am now.

I could keep going with my open thinking effort, the point is I have been running in circles in my head because of simple issues that I hit right at the point when I try to break down the large problem into small pieces.

From out of the mess I have been crystallising some key ideas that might be of some use: /they are mostly repeats of stuff I have read or have already written/
I don't need to design the actual collaborative system that runs the space program, I need to figure out the algorithm that grows the system from a simple seed.
I need to figure out ways to simplify the problem to chunks I can handle myself.
I need to build a simple system that has within it the capacity to grow and redesign itself as it draws in more involvement from other people.
I need to keep it simple, and I need to keep the problem chunks simple all the way along.


Other stuff needs ranting:
form dictates function dictates form dictates function dictates form dictates function dictates form....

Never give up, never surrender!

Now back to freespaceships:
The system isn't really about designing, manufacturing, and maintaining space hardware, it is about organising human collaboration. We already have several systems out there, I am only aware of a very tiny fraction.

 But the system I have in my head /which is very vague and is more like a shadow of a system/ hasn't arisen yet. I have this conviction that it could potentially be part of how we colonize space. Because a society that grows to inhabit its solar system would need systems that reliably organise and coordinate the collaboration of millions if not billions of it's members WITHOUT the current issues of resource allocation, and destructive internal conflicts.

One of the key issues is the waste of human capital. We have billions of human brains potentially all capable of solving high level problems like ISRU but they are mostly busy with trying to stay alive in turbulent political/environmental/economical niches where survival is hard. They cannot even afford to ponder about the possibilities of life on another planet, they have the pressing issue life in their corner of the world being extremely tough.

Anyway I am getting sidetracked.

Simple system, self changing, self monitoring, potential for growth.

I got stuck with our wiki because it takes too much effort for one person to grow it into useful size and it seems to be pretty useless. I want something smarter than a wiki. :) So I got back on track with learning JAVA instead.

I need a database with a nearly automatic entry method and update method. So basically I need to be able to just simply dump links into the thing, and let it sort out the data I am looking for, make the entry for the database and revisit the site to update itself, and also make fancy stat charts about the sites and also itself. Well it would be better if it was a smart bot that did all the searching, organising and presenting data by itself.

This sounds like a duplication of already existing systems out there, and it probably is. But since I have no access to those systems, I will have to try and reinvent the wheel so I can have a go at rolling it around. //unless I find what I am looking for while working on the problem, my guess is I would only need to find the right software out there and just implement it to this problem.//

Other notes about "freespaceships" //note I am using "freespaceships" as a title for something that will most certainly won't be called as such//:
So I envisage a problem pool, a resource pool, and product pool.

We have the problems like buying a screw, or how to stabilise a satellite, how to get more people play our game, how to abide by laws of all sorts of different countries.

We have a resource pool that contains all of us, all of our shared thoughts, shared designs, shared knowledge, shared finance, shared tools etc...Anything we can share that can help us reach our goal is a resource we have.

We will have products. Like a space port, space station, off world research facility, holiday resorts, factories, things we sell to people to get their money to fund our operation.

Problem pool we all go to, dip our hands in, grab a chunk we can manage, we solve it. I guess we might need a solution pool to put it back into. If we fail at solving we attach notes and our work to the problem and we put it back into the pool. At any given time each problem should be visible to all and also who is working or has worked on the chunk.

Solution pool would have solutions to problems. Probably solving problem chunks would involve looking into the solution pool to find solutions before trying to come up with our own.

Resource pool we would go to to get tools, or money, or people to help us solve the problem.

Product pool is where we put the solutions that then we can use to acquire resources to put into the resource pool.

//That above was the rant waiting to be published, I did minor edits as I re-read it.

Now the current update:

So I watched some more TED, more inspirations, I found a great project I was unaware of, and now I am hooked.

It is coursera , a website that offers a whole heap of online courses for free.

I was hooked straight away because it solves the problem of me being a dunce. Hopefully.
I enrolled to a couple of courses that I thought would be extremely relevant for my attempts of trying to solve the "mass collaboration problem" for my hypothesised "freespaceships" project.

The courses I am taking:
-Algorithms prt1: this will help me learn more advanced JAVA than the basic tutorial I am doing. I already set up both Eclipse and DRJAVA to be able to do the exercises, and I am ready to have a crack at this. They also have a neat online book with heaps of resources for noobs to un-noobise themselves.
-Machine Learning: awesome course, I already watched 3 weeks worth of videos on it trying to get a head start and also trying to see where I would potentially get really stuck cause of my lack of "smarts". I can't wait to implement the algorithms there for something more relevant for "freespaceships", or my other pet project "Sugarglider Environmental Robotics". I want to make smart drones to do all my current work while I am floating around in space. :)
-Gamification: this is essential for a fun and entertaining system to engage people and keep them involved. I love games. I would love to play one that can get me and a couple of million other people to the moon in 30 years if played right.
-Social Network Analysis: again an obvious course because I am trying to grow a social network. A pretty large one too. Hopefully this course will help me understand networks better. More knowledge means less dumb ideas. Hopefully. :)
-Networked Life: Another course on network, but this one seems to be more general.
-Artificial Intelligence Planning: this is a course on using AI for planning. Covers a wide range of different topics and approaches. This is also absolutely essential for a space program of any sort. :)

So I am absolutely not discouraged yet by the difficulty of creating "freespaceships". I am actually enjoying the whole process so far.

On another note I also had some more chats with J. Simmons at Mach30. I got an account at ODE so "freespaceships" is officially connected to MACH30 now. Hopefully freespaceships will become something useful in the future and be an integrated part of the global network of maker spaces, individuals, organisations that will get us to the moon to have a cup of tea.

Wife is forcing me to play some munchkin with her and a friend so I have to go now. :D

Signing out!