I saw a comment on Reddit the other day that made me think of political strategy.  The post referred to the fact that the proposed “Buffet Rule” would not actually generate that much money. My response was, of course not. The goal at this point is not to raise money. The goal at this point is to start moving the needle with response to public opinion on taxing the rich.

It is interesting how much public opinion has changed with respect to the tax rate for the highest income in America. Setting aside questions about how effective such taxes are, it is remarkable how there seems to have been a consensus that taxing them is bad. In this post I want to present a theory that large issues like behave as though… well, i can’t think of a good physical analogy but allow me to explain.

I still believe quite strongly that much of politics is governed by group dynamics, in particular people’s need to identify with a group. I recently heard of some interesting research that talked about how what I call group dynamics effects cheating. They set up a math test where one is paid $0.50 for each question they answer correctly. The catch is that 2 minutes into the test someone who isplanted by the experimenters stands up and says “I’m finished”. The person running the experiment asks him how many problems he finished, the plant lies and says all 20, collects his money, and leaves. The real participants now know that cheating is possible and lucrative. Turns out that after seeing this a large number of people also cheat.

Thats not the interesting part though. Things get awesome when they dress the planted person in a competing college’s sweat shirt. In this case, when the person cheating is in a different group, cheating goes down.

What?

Yes.

So to recap, if someone in your group cheats, cheating becomes less bad. If someone not in your group cheats, cheating becomes worse. Turns out morality is a somewhat fluid thing, not that we should be surprised. So the question is, how does this apply to politics?

The usefulness of the Occupy movement has not been, really, any particular achievement with regards to policy. The lasting achievement of the Occupy movement has been to create two new groups in the eye of the public: the 1% and the 99%. Such a division is not new, a viewing of old political cartoons shows that it was quite common to depict rich people as having bodies that looked like bags of money. The effect is the same, to create a group of people defined by their money, an amount of money one could not hope to achieve in their lifetime.

Prior to the 1% it was simply the rich vs. the poor. The problem with rich vs. poor is that, well, no one wants to be poor. Even if they are, they will self-identify with the rich. This led to the norming of many of the evils historically associated with the rich: greed, political manipulation, and exploitation. They were not necessarily good… just, acceptable just like cheating in the test became more acceptable after one person did it.

But with the introduction of the 1% now suddenly it is possible to frame alot of these discussions in such a way that the person cheating is not part of your group anymore. Now the greed, political manipulation, and exploitation become unacceptable activities, well worth punishing the fuck out of people for.

It is striking out much of politics is actually just this sort of framing. You should try to paint the picture such that as much of your constituency is included as possible. As long as they can identify with what you are talking about they will generally find good stuff great, mediocre stuff good, bad stuff mediocre, and horrible stuff bad. Of course the opposite is true too

We elect representative governments because we need people to make the hard decisions that we wont make for ourselves. We need them to make those choices that benefit us in thousands of ways that we can’t easily count or deposit in the bank.

But what good is a representative government when all the Representatives do is repeat the angriest words of the people in your constituency? And when those people are so angry as to refuse to compromise over a routine legislative action…

People get hurt.

All this to say: Republicans… grow some balls, do your job and do whats right. You are pandering to a minority constituency that in all probability has no idea what is actually in its best interest.

You see, so long as developed countries have such overwhelmingly superior air power they have some measure of control over other less developed nations. If you look at past wars we have lost impressively few aircraft and it is not for lack of effort on the behalf of our enemy. In fact, the story of an american plane downed in the Bosnian peace keeping mission is practically a legend.

As the Iraq and Afghanistan wars have shown us, if we have to fight on the ground we are much more vulnerable. So much so in fact that it could be possible for a small 3rd world country to win such a war against us. All that is necessary is to inflict enough casualties that the american public loses it’s appetite for fighting and we will pull out.

But you can’t keep control over a country if you are constantly afraid of the skies. Trains and trucks used to restock your fighters will have a small chance of surviving. Nor can even relatively small groups of people move around without being blown up. The key then, if you are a small poor nation fighting a large one is to bring your enemy to the ground.

That is currently very expensive and difficult. Only fairly large missiles can reach high flying planes and these are pretty expensive. It also take relatively high technology to overcome jamming and other modern anti-missile methods like flares and whatnot. Anything reasonably cheap is also going to involve a radar transmitter that will broadcast your location to the enemy. All in all this means that fighting a long and protracted war is near impossible.

All that is about to change. I think. Witness the new class of lasers being developed by the Navy. An anti-aircraft weapon would not need to be as strong, aircraft have far thinner skins and move far slower then missiles. Additionally, the weapon fires instantaneously and is potentially impossible to track. The optical technology need to focus the beam on a plane 20 km away already exists so now it is just a matter of powering the device. Throw in some ultracapacitors and a remote charging station and you have a virtually silent indefensible weapon. And, since all you need is electricity, you can fight for years at only 5o bucks a fire.

No more No Fly Zones. At least, no more no fly zones enforced by aircraft.

Why meditation? Well, for years I have been reading misc. news reports about the benefits of meditation. A few months ago I read a particular article about a study done in Boston where participants underwent an 8 week meditation program after which the study found significant changes in brain structure. This was significant, to me in any case, for several reasons. The first is that nearly every scientific study on meditation I have seen was largely qualitative. Those that did involve something quantitative like an MRI brain scan were largely performed on long time participants of meditation. These studies, while cool, are always subject to caveats. Secondly, these changes occurred over a time period of 8 weeks… a breathtakingly short period of time. It is only recently that we have come to fully accept that the brain is plastic into adult hood and here we have an example of a “simple” exercise that can healthy change the brain in 2 months. The final impressive thing about this study is that the participants were simply given instructions and a series of tapes and told to follow them at home.

So, if you add all that up you have dramatic healthy changes in brain structure over a short period caused by largely self guided behavior. Incredible. Needless to say it piqued my interest and I started looking into it alittle further. Once I started looking I heard this story on the radio. In my mind it was possible that the first study I mentioned above cheery-picked their participants, the story of meditation at the prison mentioned above addressed this. Here we have people who in most respects are a population of people who one would expect have few of the traits expected of a meditation practitioner.  And yet, they do it, voluntarily. In fact there is a 1 year waiting list on the program. Plus, this isn’t the self guided 8 week program mentioned above. No, these inmates are put into a gym for 10 days and do nothing but practice meditation for 17 hours a day. Incredible.

I didn’t start with the 8 week program mentioned above. I actually found this great resource called Mindfulness in Plain English. Despite how straight forward and encouraging this book is, what they suggest is still rather intimidating. You start, for 20 minutes a day, just sitting and watching your breath. Thats it. I tried this for a week and, frankly, got intimidated.

So, I found the tapes and book mentioned in the 8 week study and decided to commit to the program. The first two weeks you practice what is called a body scan. You lay down and, guided by a tape, slowly scan over your body simply paying attention to it. Every time your awareness is drawn somewhere else, be it a distracting noise or thought, you simply notice it and return your attention to the body part. Thats it.

The next two weeks involve alternating between practicing hatha yoga, again guided by the tapes, and the body scan used during weeks 1 and 2. In addition to this you are asked to be aware of one pleasant event each day, jotting down whether or not you were aware of it as it was happening, how you were feeling and what you were thinking, as well as what it means to write it down. The 4th week you do the same but for unpleasant events.

Weeks 5 and 6 involve yoga and sitting meditation (That stuff that intimidated me earlier!)

Week 7 is up to you.

And week 8 you practice along with the tapes again, this time being aware of how your practice has evolved and changed.

So, I just finished week 2 of this program and I thought I would do a little check in.

Practicing the body scan has been harder then I anticipated, I often space out towards then end of the time period and I have a tough time concentrating. Also, I’m almost always tired so it is hard not to fall asleep! Despite this I feel like I have noticed changes in me. The most obvious is that I am much more aware of tension in body. I find that I notice tension creeping into my shoulders and neck as I work or walk and almost immediately work to relax the tension. I used to do this before, but I would only notice the tension after it started to hurt and I was not nearly as effective at relaxing.

A more subtle change is that I feel more competent at noticing when I am getting anxious and dealing with it. This is a rather recent developement, but within the last few days I feel like when I catch myself getting anxious early I can do the following: stop, focus on my breathing for little bit, relax my gut like I mentioned above, and be mindful of the anxiety. That last part is really the meat (I think) of what this whole meditation thing is about.

This is because everything I have described so far has been mindful meditation. Instead of simply practicing concentration, mindful meditation encourages awareness as well as concentration. In a way you are exercising your ability to control your awareness. This is not as easy as it seems, how many times a day do you catch yourself day dreaming… how many times do you not catch yourself! It is more then just day dreaming though, for me, for example, it has to do with anxiety. Once I start getting anxious it starts off a chain reaction where I start obsessing about things that make me even more anxious. Just thinking about those things as I write is starting off the chain.

Control, then, is important because it is how you break the chain. If you become anxious you focus your attention on the anxiety, not the obsessive train of thoughts that follows. Once you are aware of your anxiety, you can accept that you are anxious and move on to a new moment. Its ok to be anxious, it happens and usually for a reason. But there is no reason to become dis-proportionally anxious.

That is how I think it happens. Like I said, it has only happened a few times but it is encouraging.

I hope to do an update after the next two week block.

The problem is this: while I believe I have a way to determine when people reach a consensus and when they are getting closer to a consensus, I don’t have a way of helping people communicate. If people are trying to reach a consensus they need to be able to talk and exchange ideas. This becomes difficult though using text for even a relatively small number of people. Effectively using most online tools requires alot of self discipline. Witness the decline of popular forums. Dedicated users can craft some incredible communities but the addition of new people will often cause the collapse of useful dialog.

While I won’t deny that dedicated people can reach a consensus through terrible tools they are not the audience that we are targeting. We want to provide a tool that is capable of being used by anyone regardless of how long they have worked together or how well they work together. This is a fundamentally different problem. What we have to do is somehow mimic the techniques that effective teams learn but on a huge scale.

So what are these techniques? Well effective communicators have a high signal to noise ratio. This is necessary because people just don’t have the time or patience or presence of mind to be able to sort through a huge amount of messages to determine the relevant ones. For the most part, this is the central challenge: how can we sort the wheat from the chaff while still maintaining continuity in the discussion.

I have the inkling of an idea as to an approach to take towards this. Twitter. Well, not exactly Twitter but something like it.

What fascinates me about twitter is that people are actually capable of carrying out a discussion on it. Sure, it isn’t often a good discussion but it is a discussion. What is great is that this discussion occurs despite the absence of any sort of continuity like you see in a forum or in an email chain.

The continuity is achieved through hyper-linking and general knowledge of the topic being discussed. This is pretty effective considering twitter posts are so small that one can’t summarize or attempt to restate the previous message or argument.

I bet, too, that if you analyzed Twitter networks you would find that there are definite communities which interact through a limited number of linking profiles. For example, someone who follows Sarah Palin’s profile (in a serious sense and not merely to keep an eye on crazy) would not likely follow Rachel Maddow.

Which leads me to my idea: pass around messages of a limited size between users but choose which messages go where in a smart way.

Like I said in my last post, it is easy to say “Smart way” but difficult to codify what we mean. In general though, here are the guidlines:

• Messages are attached to solutions. People will only ever see messages related to solutions they have seen.
• Messages are not sent around to change people’s opinion or whatnot. Messages are sent around based on their effectiveness.
• The effectiveness of messages will be judged in two ways: how it affects consensus on the solution it is attached to, how people perceive the message.
• We will judge how it affects consensus by looking at the position of  people before and after they read the message. Do they all approach the middle or become more entrenched in their previous positions. This could mean they all agree a solution sucks or that it rocks.
• The second way will be the productiveness of the message. People can indicate “Yes/No” if a message was productive or not. This might help filter out stupid messages like “This solution is something the Nazi’s would have used!” more quickly then if we just used the effectiveness of the message.

Ok, so thats the basic idea as of now. Still pretty nebulous but the details will come as I work on them. Just a summary though for my own purposes:

How this will work is people will sign into the website because they are part of some group that needs to make a decision. This can be a choice between actions, crafting a group statement, or whatever. After they log in people will be able to submit solutions. These solutions will be sent to other random users and those users will have the option to submit messages relating to the solution in order to give extra information to other users. They will also rate the solution. At any time users can submit new solutions or messages to solutions that they have read in the past provided those solutions are still active.

What goes on behind the scenes is that we are using an algorithm similar to what netflix uses to determine how close to a consensus people are on a given solution. Solutions that seem to achieve consensus better will be sent to more people and the opposite will be true for solutions that do not achieve consensus. Messages will be sent in a very similar way.

In this way users do not need to read every solution or every message.

Alright, time to practice lecture. Peace.

The basic idea is this: it seems like no one cares about reaching a consensus any more. Rather, the goal is to get 51% of a given population to agree with you and you are done. The problem is that it is very hard to reach a consensus, especially as the group that is attempting to make a decision gets bigger. To address this problem I am going to create a website that facilitates the process of reaching a consensus assuming that the number of people participating is rather large.

What are the steps of consensus making? There is no real consensus (haha) on how to go about it but in general: everyone is encouraged to express their point of view, specific areas of disagreement are described and then discussed, and the final decision is owned by the whole group. The last part is very important because is signifies that everyone goes in expecting to reach a consensus. Crucially, there is also almost always a facilitator that performs functions like summarizing discussion and group opinion, controlling discussion to make sure it is productive (no one speaks to much, disagreeable people are called out, … ext…).

There are several things about what I described above that make it difficult to use consensus as a decision making process in both large groups and varied settings. The first is expectations. People generally don’t expect to have to reach a consensus and instead try to win a decision making process. The second is the facilitator. This is a highly skilled and important position and relatively few people have the ability or practice to perform it well. What I would like to do is create a website that helps solve both of these problems.

Why create a website when there are so many ways to communicate as a group online? The real reason is that none of them work well. Forums become crowded and users have a difficult time ensuring that they see every post. Reddit does a better job in it’s comment section but, like forums, when the number of comments gets above a certain number it is very easy to get lost in the crowd. Email lists, bulletin boards, facebook, and many others all suffer from the same problems.

So what do I want to create? Lets first talk about how I would solve the problems inherent in reaching a consensus online.

The first problem of expectation is partially solved by creating a website with the sole expectation of creating a consensus. People know that what they are getting into is not the typical majority wins situation. Participants should also get feedback as to how well their participation is helping the group reach a consensus. Do their comments and suggestions help draw the group together or cause it to be more divided? Are people just ignoring their posts because they suck? Are they not saying anything at all or too much? Are they being to stubborn? Crucially, these are all things that a facilitator would monitor and tell the individual.

So, how do we get a facilitator? One option would be to have someone elected to be a moderator. The problem with this the quality of the process is only as good as the quality of the moderator. Plus, this does not scale well with large groups. So we need to develop some kind of algorithm that can facilitate the decision making process… What do we need this algorithm to do:

• Recognize differences/similarities in opinion
• …. thats about it.

Everything can be inferred from this one metric. If there is a minimum level of differences in opinion then we have a consensus. If a persons actions reduce differences, this is good and should be encouraged. Does one person not move towards a consensus as all? This is bad and should  be addressed. So if everything comes down to recognizing differences in opinion… how do we go about doing this vital step?

Actually, someone has done it for us. Netflix lives and dies by its ability to recognize similarities in opinion and has developed a relatively robust algorithm to recognize and more crucially, predict these similarities.

What if, instead of movies we have solutions. Each user rates solutions and we identify regions of opinion. In a way, we would expect solutions to fall into general categories like Romantic Comedie, Horror, and Action movies. You will note that not everyone needs to see every solution. The netflix algorithm allows us to anticipate what an individual would think of a possible solution even if they haven’t seen it.

Once we have identified regions of opinion we need to start working to bring everyone together. This is where it gets tricky. We need to be able to do several things at this point. One is to modify solutions and the other is to facilitate discussion. Modifying solutions is relatively easy. We can just create a new solution that is “linked” to the old one.

Discussion though implies some kind of back and forth. What is most difficult is that at this point we have not assumed that every individual has even seen the solution that someone wants to discuss. So, what if people issue statements. Statements can reference other statements and *have* to be reference solutions but are written as self-sufficient statements. What is key about these is that once we have enough data we can start maximizing the usefulness of these statements. For example, say someone writes an epic statement that causes everyone to agree with the statement it is linked to. We can recognize this and try to make sure that as many people see it as possible.

How do we recognize a statement that helps people reach a consensus? Well, we can compare the votes related to the statement+solution combo to the previous solution. If there is greater consensus we run with it. One problem that I am still working out is how to deal with statements. I feel like each statement should be tied to a solution, and that one just re-votes on a solution. We can then use changes in votes to evaluate the effectiveness of solutions. This prevents people from “liking” a statement.

It is important to remember too that consensus that an idea is bad is also important. In that case we still circulate a bad solution+statements until a certain confidence is reached that it is bad. Then we kill it.

Ok, that is about it for now. I will have to address alot of this in detail before I actually start coding. If I had to guess I’m going to have to have some individual posts on:

• The Netflix Algorithm and how to modify it for our purposes
• Given the netflix algorithm and the details worked out above, how to evaluate a consensus and progress towards a consensus.
• Details of a solution and how people will vote for it, how to decide to who and when a solution is sent to a user.
• Details of a statement. When to delete statements, which statements to send on, how voting works with statements vis a vi solutions.

One I have worked out these details I suppose I can start thinking about how to implement this as a website.

The reason for this is that my website got hacked about 3 weeks ago causing all of my wordpress, gallery2, and mediawiki websites to redirect to a malware website which automatically downloaded junk and tried to convince you to run it in order to clean your computer of malware…. Pretty funny actually but ultimately very annoying. Especially since I am hoping that lots of people will use www.micro-fab-wiki.org .

I am not sure what the real source of the infection was but the end result was that every *.php file in the directory that is controlled by the ftp username I use had a string inserted at the top. This string adds some javascript to every .html file created that loads a script from a server. This script then redirects the browser to the malicious website. It does this to all .php files, even those that don’t create .html files. This ended up being a huge problem as you will see.

The first thing I did to fix it was upgrade all of my instillations of wordpress, gallery2, and mediawiki. This ended up being a mistake in the case of gallery2 because it used .php files and those files were, as I mentioned before, “infected”. This corrupted the upgrade process and, while I still had all of my photos, I lost all of the comments, descriptions, and everything else stored in the database.

After I upgraded everything I started looking more closely at all the files and noticed that all the .php files were changed. I wrote a python script that fixed all the files:

import os, os.path, stat, time

badCode = 'Bad bit of php' + '\n'

def processDirectory ( args, dirname, filenames ):                              
 print 'Directory',dirname                                                   
 for filename in filenames:
 dummyString = filename.split('.')
 if dummyString[-1] == 'php':

 dummyFile = open(dirname + "/" + filename,'r')
 steve = dummyFile.readlines()
 dummyFile.close()
 if steve[0] ==  badCode:
 steve.pop(0)
 dummyFile = open(dirname + "/" + filename,'w')
 dummyFile.writelines(steve)

 dummyFile.close()

top_level_dir = "/home/craig/Desktop"                                                    
os.path.walk(top_level_dir, processDirectory, None )  

This assumes that the malicious code, stored as “badCode”, is the first line of the .php file.

This worked but the source of the infection remained and reinfected all the .php files. So I made new instillations of each website on their own subdomains. This way I could create a separate user for each and effectively quarantine them. This way if one got hacked I wouldn’t lose all of them like I did this time.

What I found out while researching this is that the space shuttle actually slows down after it loses its solid rocket bosters because it is to heavy for its main engines (those three big cones sticking out the back). !!! is what you should be saying. Putting a shuttle into orbit is all about speeding it up fast enough speed, so why would you possibly slow down if you didn’t have too?

Here is where things get interesting. Solid rocket boosters burn from the inside out along the entire interiour. Increasing the length of the booster then is a way of increasing the thrust. Increasing the diameter would increase, then, the length of time that the booster burns for. So then it would seem that an easy fix to the problem would be to make the solid rocket booster wider. Then the solid rocket booster would burn long enough that when they are jetisoned the shuttle will be light enough it won’t slow down.

But, of course, that isn’t the case so why? Well, it turns out that the solid rocket booste segments are manufactured in Utah of all places and have to be shipped to Florida by rail. The rail lines pass through the mountains using tunnels. These tunnels were bored long ago and are only just wide enough for trains of standard gauge. The solid rocket booster segments then must fit through these tunnels, effectively limiting their size. Since their size is limited, the duration over which they can burn is limited.

!!! Seriously !!! Well, maybe. I don’t really know, but the pieces all fit together. The best part of all this is though that the width of rail road tracks was set over 2000 years ago in Rome. Roman chariots would wear grooves into their roads and anyone who wanted to use those roads would have to make sure their axels were the same width or risk breaking them. Even after the Roman empire fell, anyone wishing to use the roads still had to make their axels the same width. Old habits die hard and our rail road tracks are evidence of this.

So: Rome set a standard width for their chariots, rail roads adopted this out of conviniece, a company in Utah wins a contract to build solid rocket boosters for the new space shuttle, NASA engineers discover a limit to the size of the solid rocket boosters, and the space shuttle has to slow down for a bit durring it’s accent towards space.

PS Mickey asked a really good question, “Why not carry less fuel so the space shuttle is light enough when the solid rocket boosters are let go?” The reason is that while you can’t make the boosters last longer, you can make them stronger. So, you make it so that the shuttle is too heavy for 10 seconds or so. Durring this time, you still gain height, which is good, even though you are slowing down. Plus, you are getting as much fuel as possible going as fast as possible.

Stackoverflow isn’t terribly unique, reddit.com  and digg.com are both examples of websites that give users a score for how well they perform. Instead of relying on moderators like many on line communities, these websites use algorithms and input from the community to determine how well a user is doing. Many times users get extra priveliges or status for having done well.

Like any system though with well defined rules it has become a game to many users to literally game the system. The administrators can control the algorithms behind calculating a users score but they can’t control the input provided by the community and this is where some users find sport. One learns the algorithm, learns the “personality” of the community, and then manipulates the community to improve their score.

This isn’t entirely bad. Most of the algorithms on these websites are designed well enough that even behavior associated with “gaming” the system is not completely negative. Neither is it typically good though. Instead of encouraging growth or improvement there is more of a race to the mean which can often drown out the few truely altruistic users that are trying to help make the place better.

Which brings me back to the original paragraph and the stackoverflow power user… His blog entry was entirely about how to game the system and get as many reputation points as possible. Remembering that the purpose of stackoverflow is to answer programming questions, he would submit valid answers but would manipulate the timing of his submission, the formating, and questions that he answered so as to maximize his score.

These users are not doing anything wrong, they are just playing the game the that the site programmers created. But this leads me to some questions I have been asking myself lately. Can we create a framework that encourages an open egalitarian online community to grow and work towards a semi-common goal?

The problem, I think, is to much carrot and not enough stick. In the past, way back in the day, you get the idea, most online communities had moderators. These moderators had the job of banning the accounts of people who missbehaved too much and more often simply acting as a threat. This is terribly labor intensive, unreliable, and subject to the whims of the person moderating. A good moderator can create a great community, but he can quit or get bored.

Partly in response to this and the growing popularity of crowd-sourcing websites have starting allowing users to moderate. YouTube is an example of this, users can now vote on comments and theoretically the best comments rise to the top. It is a great idea, no longer is simply being the first to post enough, your comment has to appeal to other users to be seen. Unfortunately what usually rises to the top is some kind of humor, not discussion. Again, the system is doing its job, humor rarely offends enough to be down voted, is quick to “consume”, and makes the viewer happy. Having such comments rise to the top does maximize enjoyment.

Fortunately or unfortunately thats what most voting systems do, they maximize enjoyment. And, since most voting systems are binary, they maximize quantity of enjoyment, not quality. That is true of simple systems, but even more complex ones like reddit or stackoverflow fall prey to the same effect. These more complex systems use the voting to create a carrot of sorts by allowing user’s to accumulate points.

The carrot helps, but I believe that we need a return to the stick. Part of the problem is that these are profit making websites (or at least they are trying to be) and can’t afford to turn away users. So, I am proposing several ideas:

1. Your ability to participate in the community is directly related to the communities acceptance of you.
2. While a public score is hard to avoid, tie the user’s return from the site to their score. Do this in a continuous fashion, with diminishing returns as scores get higher.
3. As a corollary to #2, make sure users who participate in the site get some kind of return on their investment. And, make sure that return is more then just “status in the community”.
4. Give users the opportunity to provide more then just a yes/no input into the system.

1, 2, and 3 are really all related and can probably be summerized better but in the end they represent something unique, I think.

My response was no, it is not a good idea.

And then someone asked, what if there is a head wind?  Great question actually, my response is here. Basically, so long as Vair > 0.224 Vcar, you come out ahead. Otherwise it will cost you energy to have the wind turbine there.

I am posting some pdf’s with the work I did to figure this out. It is really rough, but some one might find it interesting. I used Fundamentals of Thermal-Fluid Sciences by Cengel and Turner as a reference.

Page 1, Page 2, Page 3, and Page 4.