It seems that recently there were a group of cancer researchers who used Amazon’s EC2 infrastructure to run some tests related to drugs that could be used to treat cancer. You can find the original article here. There were a few takeaways from this article for me. First and foremost was “OMG $4,829 per hour is a lot of money.” Although compared to the cost of doing these calculations with hardware they owned instead of rented would have cost significantly more. The second takeaway was the realization that this is truly an awesome time to be involved in the tech industry. Virtualization technology is such that tens of thousands of cores can be spun up without ever talking to a human being. Then those virtual machines can be spun down, and you just pay for the time they were working.

How amazing is that?

I can imagine that there are people in the industry, professors and the like, who would have given an arm or leg to have access to these kinds of resources on a whim in the 60′s or 70′s. Just… amazing.

Today’s topic, unlike those of past Hexplanation Humpdays, is much less technical in nature. Instead, I wanted to take some time today to discuss something that, quite frankly, bugs me a lot: software patents. I’ve written numerous times on lawsuits that stem from these patents, and taken many an occasion to openly mock the ludicrous nature of these software patents on Twitter. Yet, I have realized that I haven’t seen many good non-technical explanations of this topic. So, naturally, I’ve decided to try my hand at explaining it and see where I land.

Hold on to your butts.

Writing a Book

If you are a programmer, or know a programmer, then you’re probably familiar with the turn of phrase “writing a program.” If you don’t meet the criteria above, then allow me to introduce you to this particular phrase that means exactly what you would think. Someone is sitting at a keyboard hammering out words into a work of text that the computer can interpret as instructions. Not very much unlike an author might do while writing a novel.

Just as a book is composed of words that make up sentences and so on, programs work in much the same way. You bring a lot of little building blocks, insignificant in their own right, into one cohesive whole.

Like authors, software developers have long had the privilege of obtaining a copyright on a particular work to maintain and protect their intellectual property. Much like when you write some text of poetry or prose, copyright immediately attaches (although it’s difficult, if not impossible, to take legal action if you never register the work with the USPTO). This protects you from a situation where someone may copy your exact work and claim it as their own for whatever reason.

This type of protection is beneficial to the system as a whole, and does a lot of good for dealing with those one-off situations where some troll decides to steal something that doesn’t belong to them.

Give to Caesar What is Caesar’s

There are those in the industry who are not satisfied with this level of protection because it still allows others to clone behavior in your application, and so long as they didn’t outright steal your source code then they could do what they want. So, if you came up with a super-clever new idea for an algorithm then proceeded to implement that algorithm and copyright that implementation, you couldn’t stop someone from implementing an algorithm that gets the same results, or perhaps even better results, with different code.

This is, in some opinions, a problem that needed to be addressed with the use of software patents.

You see, by treating certain pieces of a program as a unique invention, the patent holder gains significantly more latitude to prevent the cloning of such behavior in other applications. Patents are much more restrictive, and lucrative to possess, than copyrights.

So, why is this so bad? Returning to our author metaphor, let’s say that Bob writes an awesome novel titled “The Awesome Novel.” This novel is going to be highly lucrative, so far as Bob can figure, and is worthy of registering for copyright. Now, Bob can prevent others from outright copying or plagiarizing his awesome novel. However, this doesn’t quite satisfy Bob. He’s got one sentence in the entire novel that he’s really proud of. So, he decides to get a patent for that sentence as a unique invention and collect royalties every time someone uses that sentence in a book for the next twenty years. Great job Bob!

While this can’t actually happen with novels, it’s the best analogy I can think of for what is happening with software right now, and it’s absolutely insane.

The Cost of Software Patents

If I were someone who was first hearing about this issue, my first reaction would probably be, “This doesn’t really sound like something that affects my day-to-day life.” The truth of the matter is that in the mobile sector, where the fiercest patents are being litigated, is probably the sector of technology that affects us most.

For example, yesterday Cult of Mac reported on Twitter filing a patent for the pull-down-to-refresh interaction that first appeared in Tweetie (which later became Twitter for Mac). Apple has long been going after Android for their use of the slide-to-unlock interaction to unlock a smartphone, which will bring changes to how Android works. Would your experience on your mobile device really be the same if these design decisions had not become as pervasive as they are today?

The myriad of legal battles going on right now because of software patents is maddening. As is the true cost of those patents. Whole companies, so called patent trolls, are sitting on a library of patents with what seems to be no apparent intention of ever implementing any of the inventions described in the patents themselves. It would seem that these trolls wish to make their dime by suing the pants off of anyone who tries to do something remotely similar to their “invention” and go home happy.

As one comment on the Tech Crunch article about this Twitter patent suggests, this behavior would have crippled innovation in the early days of the computer industry:

What would the computer industry look like now, if, during the Homebrew computer days, before the industry really knew what it was doing, everyone patented every basic new computing invention *and enforced them*. Everything from the cursor, to grid fixed point text, to “directory on floppy disk”. Oh, and when GUIs arrived, patents on bezeled widget borders, patents on anything that worked like a button, patents on sliders, and scroll bars, and mouse pointers, and window-resize, and start menus, and on and on. -Ray Cromwell

So, there you have it, then ten-thousand foot view of software patents. Love it, hate it, or find a mistake I made? Leave me some comment love below and let me know. Until next time, my friends.

Yesterday, I spent a large part of my day sitting in a coffee shop writing, not entirely unlike I’m doing right now. What I realized is that I genuinely enjoy the experience of blogging from my iPad. It really is the ideal setup for my style of blogging. When I’m reading and sorting through the news article, I can lean back and hold it like a book. When I’m ready to write I can bust out the keyboard and hammer out the letters and punctuation at my normal not-incredibly-impressive-but-still-faster-than-using-the-on-screen-keypad pace.

If you’re reading this, you probably already know that I tend to blog about on-goings in the Tech Industry or something that catches my fancy. Lately, however, I’ve been notoriously quiet because I’ve been dealing with the transition into full time employment. I’m still within that first year after college, and as much as I enjoy no longer having exams I’m forced to admit that finding room for all the “extras” I’m used to being able to do on the side is still a challenge. No longer do I have ample amounts of time sitting in a boring lecture class where I can zone out or time in the afternoons where I can key up a blog post while I’m cooking dinner and not miss the time. No longer is my computer a device primarily used for my entertainment with the occasional coursework mixed in. No, the situation is very much reversed these days.

I don’t think any of these reasons are the single reason I’m no longer blogging regularly. I suspect it’s more of a “death by a thousand paper cuts” scenario between these and about five million other reasons. However, I’m hoping that the solution might be the flexibility my iPad offers me in terms of mobility and the increased throughput the bluetooth keyboard affords me. So what exactly is the current state of using an iPad and a keyboard these days? I know there have been various articles on this in the past – but as usual I feel compelled to provide my own spin on it.

First Impressions

Of course, one of the best things about having a keyboard to use with the iPad is being able to type using your normal key patterns. Most things work exactly as you would expect, and even many of the macro keys at the top of the keyboard (i.e. volume, brightness) work exactly as they do on a normal machine. You can even make accents over characters exactly as you can on the desktop applications.

As you would expect, typing is speedy. As I’m typing this I don’t discern any lag or delay. I also notice as I’m typing this that I’m not dealing with autocorrect quite as frequently, as my typing with a physical keyboard is much more accurate than it is with a touch keyboard. I also have the ability to very easily type characters with accents, if I so desire. (¿Cómo estás?) Additonally, using Option-Delete has the expected effect of backspacing whole words at a time, which is so much faster than using the select bar handles to select what I want to backspace when I need to nuke an entire sentence.

Now that we’ve established that this setup works pretty well on the basics, let’s see what happens when we try to do some more advanced things, such as key combinations and interacting with remote servers over SSH.

Pedal to the Metal

After using this for a few minutes some limitations of using the keyboard with the iPad became immediately obvious. None of them are really debilitating, or even devalue the presece of the keyboard when I’m ready to write something, but they do diverge from the normal unity of experience that I expected from Apple products.

What, specifically am I talking about? The absense of keyboard shortcuts. While some global ones work as you’d expect (Command-C and Command-V), other application-specific ones do not.

It may sound like a really simple qualm (I am, after all, using a keyboard with a device that really isn’t intended for a keyboard – I should be happy that it works at all), but there’s a lot to be said for the fact that in Apple Mail or Sparrow on my desktop, I can hammer out an email and then mash Command-Option-D to have the email dispached. On the iPad, however, no such option is made available to me. I’m required to reach over to the screen and mash the Send button. I am also unable to use Command-R to reply to the email I’m currently reading.

This state of affairs is fairly universal across the entire platform. Tabbing through things doesn’t work quite like you’d expect, and you don’t have the option of using the up/down keys to page through a list if you would like to do so. In Blogsy, I can’t use Command-B or Command-I to bold or italicize text. A minor inconvenience, but one that I notice nonetheless.

That said, I’m happy to report that in other areas the keyboard + iPad combo exceeded by expectations. Especially in the realm of interacting with a remote server using Panic’s Prompt SSH Client. While using Prompt I am able to successfully control pretty much any application on the remote server I could need (GNU screen, vim, you name it) and it all work seamlessly as if I was typing straight into iTerm on my Mac. Honestly, being able to maintain my blog’s server or respond to any emergencies at OpenStudy from my iPad with no fuss will pay for itself, I expect.

Moving Forward

I highly doubt that I will ever do any part of my full time job solely on the iPad. While I could certainly use vim running on a remote machine as my development environment, there are a number of niceties to having a full computer that I do have to appreciate. Not to mention the fact that I doubt I could stomach the amount of money I’d be required to pay every month to make compiling things with scalac a good experience on a remote server!

That said, it is a delight to see what a good experience interacting with my iPad using a keyboard is for most routine operations. Despite the shortcomings I’ve found, I still think it’ll be a pretty excellent way to write and communicate.

That said, I’m interested to know, have you or someone you know tried to use a iPad for for some or all of your workflow? Be it for your job or you hobbies, like I’m doing, I want to know. Likewise, if you’re brave enough to withstand a little mockery from your coworkers for turning your iPad into a more “laptop” like device and decide to give it a shot yourself, let me know how it turned out for you.

Today’s lesson: Javascript doesn’t do well when you combine array with the plus operator. Awhile back I was working on a bug that required me to iterate over an array of users on a particular question. Well, we have two arrays – one for people who are replying and one for people who aren’t. So, I did something like this.

var allUsers = usersViewing + usersReplying
$.each(allUsers, function(i, user) {
  user.resetStatus();
})

I was kindly provided with the following error message in the Javascript console. This isn’t the exact error, but the gist of it was:

Object [ has no method 'resetStatus'

After some digging I found a fun law of Javascript and the + operator with arrays.

["this is a string"] + [] = "this is a string"

WAT.

Sure enough, Array.concat() saved the day, but still – who thought that behavior would be an excellent idea?

I’m sure the anal retentive among you are now wondering why I used such a ridiculous title for this blog post. Well, today I was greeted with an interesting editorial by Anne Trubek in Wired that made the argument that our infatuation with correct spelling is antiquated and should be left behind. (In my mind I picture some of my readers visibly cringing at that thought. I did.) As one might expect, I do not agree with Ms. Trubek on this issue.

She argues, “Standardized spelling enables readers to understand writing, to aid in communication and ensure clarity. Period. There is no additional reason, other than snobbery, for spelling rules.”

She goes on to claim that text messaging, and presumably other digital forms of communication as well, are closer to speaking than writing and that enforcing spelling rules in these mediums is a practice should be abandoned. I can’t speak for Trubek, but when I press the little buttons with letters on them corresponding letters appear on the screen. Seems an awful lot like writing (or typing) to me. Which, if I’m not mistaken, would make that a written form of communication?

If we can all agree that I’m writing, then what is the real point here? Certainly, Trubek, you’ve come up with some ingenious logic to support why it makes sense to sacrifice this clarity in the English language. Surely there’s some research or some formidable argument that would give a skeptic such as myself pause for thought on this issue, right?

Trubek writes, “We need a new set of tools that recognize more variations instead of rigidly enforcing outdated dogma. Let’s make our own rules. It’s not like the English language has many good ones anyway.”

Oh, so you just don’t like English spelling rules. Consider me underwhelmed at your argument.

Correct spelling and punctuation may be cumbersome on a mobile device, and honestly I generally don’t care if someone uses abbreviations in a mobile setting. However, spelling is anything but “outdated dogma” that we should throw to the wayside, and writing, real writing, isn’t going anywhere at any point in the foreseeable future. Finally, I personally intend to continue spelling out words in my text messages. Does saying that make me a bit of a snob? Well, even if it does, at least I’m a snob who will be understood.

Finally finally, because this is entirely relevant:

So, from a 10,000 point of view it will help me to explain that, while most people know what a processor is, many consumers don’t realize how many different varieties of processors exist. Most desktop machines use a family of processor commonly known as “x86″. PCs and Macs alike use this variety of processor, and over the course of Microsoft’s history they have made the largest profit selling software to run on x86 machines.

However x86′s dominance as the de facto champion of all computing platforms has recently been challenged. Mobile devices, such as smartphones and tablets, often favor alternatives that consume much less power than x86 units. One family of mobile-friendly processors that has emerged over the past decade is the ARM family of processors. With an increasing number of casual Internet users transitioning to using these mobile devices more frequently than their full size computers, it seems as though x86 could soon be relegated to only being used by computing professionals and hardcore gamers. Granted, that is still a long way off, but that’s my personal opinion on the way the market is headed. Take it for what you will.

Now, typically, software manufacturers that sell directly to consumers won’t care too much about on what platform their software is going to be executed. In fact, before Mac switched to Intel chips, it was not uncommon for manufacturers to produce a version of their software to run on the old Mac processors (PowerPC, if you care to know the name). Given the emergence of ARM, and the general diversification of the consumer device offerings, it isn’t surprising to hear that Microsoft is planning on producing ARM versions of their popular Windows Operating System. (In fact, they might do so already to be honest – I just don’t know.)

What is surprising is that it seems, according to ArsTechica, Microsoft is going to make it harder to install an Operating System other than Windows on ARM machines that ship with Windows pre-installed. They plan to do this by making their feature called SecureBoot mandatory for these machines. This strikes me as very, very peculiar. What raises my eyebrow even further is that they mandate that non-ARM machines must allow users to turn SecureBoot off.

I struggle to imagine the reasoning for this. Windows, as it exists now, wouldn’t make much sense as a mobile Operating System. There have for a long time been a mobile variants of the OS that they have used for mobile phones, but to the best of my knowledge they have never used full-on Windows in a mobile device.  

There has been talk of ARM based economy PCs hitting the market, which is all well and good, but I’m beginning to suspect that – in spite of these new low power PCs coming out – Microsoft actually is planning to do something big with Windows on mobile because the people who buy economy PCs aren’t likely to hurt Microsoft’s market presence or revenue in the desktop arena. It’s not like they have an equivalent of Apple’s App Store that gives them continuous revenue. 

However, if they were going to give a model like that a shot for their “big boy” operating system, they would be smart to do it in the Mobile arena first. And if they did do that, it would be on ARM chips, and a large number of people not using their Operating System would significantly hurt the viability of such a continuous revenue business model.

Additionally, this is a marked change from how Microsoft has operated in the past. It’s a very Apple-esque move in many ways.

Microsoft has talked before about trying to give a constant feel throughout their products. Indeed, the XBox dashboard now looks like an HD version of a Windows phone in some respects. It is reasonable to suspect that the next step in that endeavour is unifying their mobile and desktop Operating Systems, and attempting to build a continuous revenue model to capitalize on their users after they purchase the original product.

Whether or not that will actually happen is something only time will tell. For now, we just have to sit and wait. As always leave me some comment love below. I’d love to hear what you think.

However, these sites really only work for people who get some sort of gratification from achieving those points or rewards. I am not typically one of those people. Don’t get me wrong, I do find it rewarding, but it’s not likely to make me feel rewarded enough to return to the site. (I’m also not much of a gamer in general. Skyrim has been sitting on my shelf, unopened, for over a week.) Yet, there is one thing that is always particularly effective at motivating people like me: money.

It would seem that there are new companies starting to pick up on this trend. 21habit, a site started by Microsoft and Amazon engineers, is designed to help you kick start life changes by guiding you through a 21 day challenge where you must report regularly about your success or failure. The service offers the traditional, no penalty, method that many Gamification structures offer. Yet for those who need some extra motivation, they’ve also provided a “committed” mode. Essentially, if you use the committed mode you wager $21 that you will complete your goal each of the next 21 days. For each day you succeed, you earn back $1. For each day you fail, $1 is donated to charity.

Another new-to-the-arena company named GymPact is doing something similar. Members of the service participate in a pact to go to the gym a certain number of days each week. On weeks you do not succeed, you loose some money. On weeks you do succeed, you get paid. GymPact uses the money from the pool of people who failed to meet their goal to pay those who succeeded. Overall, it is an interesting concept – but as a business model it sucks because it only works as long as there are people using the service who consistently fail. If everyone succeeds, then everyone… well, looses. I don’t see money motivated people like myself being willing to “take one for the team”, but I can’t deny that the concept is still interesting. 

I suspect we’re going to see a lot more of the money-powered gamification sites pop up. Overall, sites like Fitocracy and others have had a hard time attracting users who just aren’t motivated by some fictional currency. Naturally, the next logical step is to create sites that use a tangible currency to motivate its users. 

So what are your thoughts? Do these sites sound like something that would appeal to you? Leave me some comment love and let me know!

Further Reading:

• TechCrunch article on 21habit
• TechCrunch article on GymPact

Quite simply the service enables you set up triggers between multiple services in a dead simple fashion. Currently it supports a handful of services, including Twitter, Facebook, WordPress (Hosted and Self-deployed), Flickr, Craigslist, and more. As you can see in the screenshot above I’ve set mine up to auto-syndicate my blog posts to my Twitter and Facebook accounts. (This blog post will actually be the first to test these new triggers.)

Setting up new triggers in IFTTT is done in a Wizard format that guides you through the process of setting up your triggers and actions.

Overall, I’m pretty excited about this. I wish they had a few more services (such as Gravatar), but overall it is a pretty promising selection. They are still in “beta” status, so I expect that everything I’m missing will come in a future release. If you’ve got some time to check it out, then please give it a spin and let me know what you think. I expect a majority of my readers will have already heard about IFTTT, but I’m sure there are a few who come here from links on my Facebook feed who haven’t yet.

You can get to IFTTT by pointing your web browser to ifttt.com (or clicking that handy link). I’m just now scratching the surface of what I can do with this little gem of internet automation, so I’ll be sure to keep you guys apprised of anything cool that I come up with in later blog posts. As always, I’d love some comment love with your thoughts on it. Does IFTTT sound like something you would use? Or, are you already using it?

Essentially, whenever we’re working on a change that is going to be more than one or two lines we create a new branch on our git repository. We do any number of commits to our new branch, and then push our changes. Since we use Github for our repository hosting, we can then open a Pull Request to let everyone else know the code is ready to be reviewed. Then someone else will review the code, and check the code out to test it for any bugs.

Admittedly, I was apprehensive about this process at first. Mostly because I had never participated in a team where code review was done well before I came to OpenStudy. With most of my recent employers I had a lot of autonomy from other team members or supervisors. As long as my clients were happy, everyone else was happy. Usually code review only happened if someone wasn’t happy. Of course, I didn’t expect that autonomy to remain the same in the transition to a product-oriented company. It’s just not realistic, and it shouldn’t be. So I braced myself for some transitional pains when we started this workflow.

I found that the transition to the code review process at OpenStudy wasn’t nearly as painful as I expected. In fact, our code review process has been invaluable over the past month in my learning process, finding bugs in the code before the changes got merged into master, and otherwise making sure that the code we eventually published to our server was relatively polished. Nothing is worse than preparing a release, getting excited to push it to the server, and then realizing something is horribly wrong as soon as you get it live. Of course, I’ve never actually had that happen (cough cough).

So, the process of code review has been throughly beneficial to me. Although I can’t speak for Antonio and Matt, I suspect they have similar sentiments. I encourage you to strongly consider implementing a similar process in your development team as well. If you do decide to do so, I have a few ideas that you may find helpful. These are some things that I believe we’ve really done right at OpenStudy, and that have made the process more valuable for everyone involved.

First, everyone participates. Anyone who is actively coding needs to participate in the code review process, regardless of their position within the team. The truth is that being around longer, or having a more senior title, doesn’t make you immune to human error or mean your ideas will always be the best, yet I’ve heard many a horror story from friends who work under a managing programmer that doesn’t understand that. Unfortunately, my friends end up cleaning up the mess left behind by their boss more often than not.

Additionally, code review is one of those things that works best when everyone buys into it. The more buy-in that your team members have in the process, the more beneficial it will be for them, the product, and the whole company. The best way to get a team to buy into code review is for everyone to participate. So, even if you are that guy who is capable of taming two lions while writing perfectly formed assembly, you should still allow your code to be reviewed by your colleagues.

Second, we use GitHub. I was one of those people who honestly didn’t see a whole lot of value in GitHub over hosting my own private Git server. I’ve never been in a position to use their pull requests before, but it didn’t really sell me on them based on my initial assessment. However, at OpenStudy we make extensive use of what I’ve come to believe is GitHub’s golden egg: the line commenting feature. It’s possible for any one of us to single out an individual line in a file and ask a specific question about it, or make a specific suggestion. The benefit is twofold: the topic of conversation is immediately visible for anyone reading the pull request (no fumbling through a file looking for the contents of the specific line), and conversations about distinct issues or lines stay separated.

Overall, this reduces the amount of time we spend trying to communicate a specific problem or question and increases the amount of time we actually spend addressing the issue.

Finally, we encourage open communication. The idea behind open communication during the code review process is that all participants in the review converse back and forth on the suggestions that are made. This is what sets “code review” apart from “code approval” and enables the developers to learn from each other, instead of blindly issuing a list of corrections. Developing a good product is a team effort, and no line of code stands alone. Of course if your team ends up at an impasse in the conversation someone needs to be able to pull rank to keep the forward momentum going, but that should always be a last resort.

Code review may not work for every team, but it works pretty well for us. So, leave me some comment love below. I’m sorry it’s been so long since I’ve posted on here – but it’s been a really crazy month. I hope you and your family have a safe and excellent New Years, and here’s to hoping that I get back into the rhythm of regular posting in 2012.

This bill is especially bad because Section 102 of the bill requires that, after being notified of infringing content, service providers are required to take steps to prevent access to the content (link) and search engines are supposed to make the links magically disappear from their results (link). Similar provisions exist in the bill for payment processors (link).

Do I support online piracy? No. I believe people deserve to get paid for what they do. However, I also believe that under no circumstances should the government be allowed to interfere with the business of non-co-conspirator third parties with just a court order. The precedent this sets is dangerous, and scary.

Look into this yourself. Check out the OpenCongress profile for the bill. Read the text. Write your congressional representatives. Call them tomorrow, Wednesday November 16th.

We must prevent this bill from being passed or, I fear, give up our protected rights under the First Amendment in the digital world.