Bracken King

Developed novel information theoretic approximations
Formulated a framework for approximating high-dimensional information theoretic statistics using automatically inferred relational models. Validated the metrics on synthetic data, demonstrating significantly enhanced accuracy when faced with limited quantities of data.

Analyzed diverse biological data sets
Applied the approximation framework to a variety of data sets: selected compact gene sets for cancer classification; selected sets of phospho-protein data for enhanced statistical models; selected representative sets of experimental conditions to improve the efficiency of idiosyncratic drug hepatotoxicity studies; identified non-redundant cluster assignments to improve hypothesis generation.

Extended framework to molecular systems
Adapted the information theoretic approximations to analyze configurational space of small molecules and protein sidechains using molecular dynamics or rotameric enumeration. Demonstrated improved convergence of configurational entropy calculations compared to existing methods. Analyzed the conformational freedom of residue sidechains in MD simulations of HIV protease bound to drug candidates.

Performed protein sequence-structure-function studies
Contributed to the design of an enzyme mutant library by combining sequence data and computational structure-based design methods to identify residue positions likely to impact, but not destroy, activity when mutated. Developed a novel information theoretic phrasing for incorporating phylogeny into sequence-based residue coupling analyses.

Designed modified lanthanide binding tags
Developed novel techniques for computationally exploring the energy landscape of protein backbones. Identified candidate amino-acid positions responsible for the lack of crystallographic order of an existing lanthanide binding peptide. Applied computational protein design techniques to suggest single and double mutants expected to have improved order in the crystal structures.

Identified putative biomarkers to predict drug response of human tumors
Developed a system to analyze genetic, mRNA/protein expression, and phenotypic data in a consistent framework for biomarker identification. Generated putative sets of biomarkers of tumor cell-line response to treatment with various drugs.

Mentor for Undergraduate Researchers
Formulated and directed independent research projects for three undergraduates, applying information theory to: gene selection for cancer classification, identification of residue coupling from MSAs, and transcriptional network inference.

Guest Lecturer, Biomolecular Kinetics & Cellular Dynamics
Prepared and delivered a 1.5 hr lecture on diffusion-limited enzyme kinetics. Reviewed theoretical developments, Brownian dynamics simulations, and finite element methods for the calculation of diffusion-limited rate constants.

Teaching Assistant, Thermodynamics of Biological Systems
Prepared and presented twice-weekly recitation sections for undergraduates. Organized and held monthly review sessions. Helped develop and grade homework and exam problems.

For reasons that aren't worth going in to, Tyler and I have recently begun the process of setting up a server through Rackspace's cloud services. I've never used Rackspace before, and all I really knew was that (1) they're expensive (though this doesn't actually seem to be true, at least for their cloud options) and (2) they have great customer support. From our limited experience, this latter point seems spot on, even though we have yet to communicate directly with anyone at Rackspace.

Now you might argue (understandably) that this is a silly statement to make given our lack of actual experience with anyone at Rackspace, but if you've ever set up a web server, take a look through this getting started guide, and you'll see what I mean. If you haven't ever set up a web server, I'll just say (1) count yourself as lucky, and (2) it's generally much more difficult than it should be for a number of reasons, not the least of which is the generally awful documentation available at various web hosts.

It took me a while to figure out what exactly about the help section made it so usable, but Tyler recognized it pretty quickly (from our chat transcripts):

I really like how it doesn't bother with options. It's like "here's what you want to do. Don't do it any other way." That's how advice should be. If I know enough to decide for myself, I wouldn't be reading the tutorial in the first place.

While I agree that this is part of the success, there's a second aspect that is similarly important. Even though they only tell you one way to do things (basically), they fully explain the reasons for doing it that way. That combination is a surprisingly rare one, and I think the principle is worth keeping in mind if you're ever in a position to put together materials for training, be it online or in person.

Most help/advice falls to one extreme or the other: you either get an exact prescription of what to do without any rationale, or you get a full description of many different options without any real structure. With the former, you can very quickly run in to problems if anything at all unexpected happens (like if some options change after a couple years); the "help" may be rendered completely useless. With the latter, you need to be an expert to derive much utility from the many options, leaving even a moderately-advance user without much help.

Even giving a clear path while also explaining all the options can be a tough route to follow, as the "clear path" can be easily obfuscated by all of the options, particularly if decisions aren't independent of each other (e.g., the options in step 2 impact the options in step 5). The beauty of the single, well-reasoned path is that it provides help to both beginner and advanced users. The simple narrative is easier to follow for those who are still learning, but the added depth in each part can help give more advanced users a foothold on how to approach more detailed configuration.

Even if you're not writing documentation for software or a website, the chances are pretty good that at some point you'll be trying to explain something that you understand well to a customer who doesn't. Sometimes one of the extremes (directives without explanation, or exhaustive list of options) is appropriate, but a well annotated narrative is likely to be a great option in many cases.

Most of us receive dozens, or even hundreds of emails each day. Staying organized can be a serious challenge, and the more unread emails you have piling up in your inbox, the more likely it is that something important will slip between the cracks. Google released a new feature this week to help solve this problem.

Gmail's new "Priority Inbox" feature automatically flags your emails as "Important" or "Not Important" to help you make sure you see the emails that really matter to you. When you look at your Priority Inbox, the first thing you see is a list of your important emails. When you open an important email, you can either deal with it right away, or star it if you need to come back to it later. The second thing you see in your inbox are all your starred items. Then below all that you see a list of "Everything else" which includes unimportant emails and important emails that you already read but didn't star. This makes it really easy to make sure that you've at least handled everything important even when your inbox is a mess.

You may be wondering how it works. How could Google possibly know what emails are important to you? Well, it's not a perfect system, but they collect tons of data about which emails you and other users regularly read, respond to, forward etc. and they use this data to make educated guesses about which future emails you'll be interested in. You also have the option to manually mark messages as important or unimportant and Google's algorithm will learn your preferences over time.

If you're interested in trying the priority inbox, you'll need to enable it in your account by clicking here. It's not for everyone though. I love the idea, but I almost never have more than 10 emails in my inbox anyway (I'm a believer in Inbox Zero) so it doesn't really help me that much (and it doesn't work with multiple inboxes). But I think anyone that struggles with email overload should definitely give it a shot.

If you do try it, let me know what you think in the comments. It's a pretty radical departure from the conventional inbox and I'm interested to see how people respond.

We've talked previously about the usefulness of application shortcuts in Chrome, and about how to enable them on a Mac. As part of the hack to get things working on the Mac, we had to give each application shortcut it's own profile. It turns out that giving an application shortcut it's own profile can be useful on it's own, and it's easy to do on any OS. Here's the why and the how:

Why
The main reason to give an application it's own profile is to decouple it from your main browsing session. This is particularly useful if you have multiple accounts at the same site. For the example here, I use Google Docs at work and at home, and I wanted to be able to use either without having to worry about switching accounts. When you set things up like this, you may have to do a little extra work to install whatever extensions and user scripts you want for each profile, but it also means you can completely customize each webapp.

How
Giving an application shortcut it's own profile is very simple. If you haven't already, start by creating an application shortcut for your site by navigating to the website, and choosing "create application shortcut..." from Chrome's menu. In my case, I created one for docs.google.com. Depending on your OS, you'll have a few different options about where to save it. For Windows 7, I like to pin them to the taskbar, but feel free to create it wherever you prefer.

After creating the shortcut, you now need to edit it to tell it to use the new profile. To do so in Windows, right click on the pinned icon, and then right click again on the shortcut (Google Docs WORK, in the screenshot), then click on Properties. Doing so will bring up a new window:

To change the profile, you need to append some text to the end of the "target" field, as shown in the screenshot. The text you need to append is:
          --user-data-dir="[path_to_profile]"
Where [path_to_profile] is the file path to wherever you want to store the new profile. Chrome will create the folder if it doesn't exist, but you'll probably want to keep them all together (for this example, I've created a C:\Profiles directory for this purpose, and have made profiles called Home and Work underneath that).

And that's it. The next time you launch the application shortcut, it'll use the new profile. If you want to install extensions and the like, just open a new tab with ctrl-N after opening the shortcut. The procedure is pretty much identical for Mac and Linux, but if you run across any problems, just let me know.

If you own a smartphone, you generally have two different options when you want to read web content. You can either open up your phone's web browser and read the content online (like you do on your computer), or you can download an app from the content provider and read the content from the app. The content itself is generally the same, but the presentation can be dramatically different.

Even though there are very good reasons to stay away from apps, the fact of the matter is that most iPhone/iPad apps provide a much smoother experience than mobile websites. I still personally prefer to read content from my phone online (Google Reader has a great mobile site), but I can understand why people prefer using apps, or iPad magazines.

I'm not writing this post to convince you to stop using apps on your mobile devices. You should use what you like. However, I do want to convince you that there's no real reason for most apps to be so much better. Everyone talks about how much nicer they look (which is true), but it's important that everyone understands that there's no real reason for the differences in design. When you develop an iPhone app, you can decide what every single pixel on the screen should look like. When you develop a mobile website, you can decide what every single pixel should look like. You can literally make a website look pixel-for-pixel the exact same as an iphone app.

You might be confused. Why did I say that iPhone apps are more attractive than mobile websites if they both have the exact same capabilities? I'm a little baffled by this myself. For some reason, when companies make iPhone apps, they put a lot of effort into the design. They don't seem to care about the design on mobile apps. That's just the way things are.

The problem is that mobile apps are setting back technology as a whole. We spent the past 10 years moving away from closed, platform-specific desktop applications (Microsoft Office for example) in favor of great web apps which are much more accessible and run on any device. These web apps still have a long way to go, but they get better every day. However, Steve Jobs has single-handedly convinced everyone to abandon all that progress and go back to the 1990's. People no longer want open software. They no longer want platform agnosticism. They want their apps to look shiny, and if that means selling out to Apple, so be it.

Well I like shiny things too. I get it. I understand why people are paying for apps instead of using free mobile websites. I just want you to understand that there's no real technical reason why most downloadable apps are better than their web-based counterparts. The next time you want to read an article from Wired Magazine and you decide to pay $4.99 for the app instead of reading it for free online, just keep in mind that you're doing it because wired decided to make a crappy website, not because apps are fundamentally better.

One day mobile apps will start moving to the web the same way desktop apps have been for the past decade. When that happens, I hope you don't fight it because you think that apps on your phone are better for some reason. Just keep an open mind.

Businesses are slaves to inertia. Most people work in a certain way only because that's how the people before them worked. When we write business emails, we try to copy the tone we're used to hearing from business emails we receive. When we name our companies, we try to make them sound like other companies that came before us (a pharmaceutical product called veroxitran, a web 2.0 startup called disrupt.ly, etc). Sometimes following the status-quo makes sense because it's a great way to avoid wasting time on minutia, but sometimes the status-quo can lead us astray.

One example of a misguided standard business practice is sending automated emails to your customers from an email address like no-reply@fakecompany.com. It's pretty common to order a product online and receive a confirmation email that says in bold letters, "Do not respond to this email. Your response will not be read." What kind of message does that send to your customers? Do we really need to be that hostile and unavailable?

There's probably a very good reason for some companies to use no-reply email address. I'm sure that if companies like Amazon or Facebook tried to read all the responses to their automated emails (including those super annoying "out of office" replies or any other automated replies), they would be overwhelmed. There would be no way to deal with all those emails.

The problem is, small companies see these emails from big companies and they decide that they should do things the same way. I couldn't disagree more. Most small businesses only exist because of their superior customer service, and copying the big boys just dooms you to the same forgettable and impersonal service that they suffer from.

Facebook probably sends out millions of automated emails a day. Your small business probably sends out a few dozen. Is it really that big of a problem for you if people respond to those emails? Do you really hate your customers so much that you don't want them to be able to get in touch with you?

I'm probably preaching to the choir here, but if your company sends out impersonal no-reply emails to your customers, I suggest that you strongly consider making the switch. Just because an email is automated doesn't mean it can't be personal. Every single automated email at LAS comes from an address that forwards straight to me. We probably send about 100 of these emails per day (welcome emails, reminders, etc) and I can assure you that I've never been bothered by the responses.

Shouldn't we make it as easy as possible for our customers to get ahold of us?

I recently met someone who was in charge of the SEO of a relatively large internet company. I mentioned LAS to him and spent a little time talking about the site before he basically said "SEO probably isn't worth it for you guys." Now we spend a lot of time talking about SEO on this blog, and we obviously think it's important, so I was a bit surprised to hear this response from someone in the industry. Eventually we figured out that what he means by SEO, and what I mean are fairly far apart from each other. The overall goal is the same (improving the standing of your site in the eyes of search engines), but the extent  of effort put into it is quite a bit different.

You've probably at some point heard about the 80-20 rule that describes a remarkably common phenomenon in which a small portion of causes (e.g., 20%) accounts for a large amount of the effect (e.g., 80%). The specific 80/20 ratio isn't particularly magical, but the basic point shows up all over the place from wealth disparities (20% of people have 80% of the wealth -- I think it's actually much more extreme than that, just an example) to customer service (20% of customers require 80% of the support). In the context of SEO, the idea would be that 20% of the effort might yield 80% of the SEO results (ignore, if you will, the fact 100% of "effort" and "SEO results" aren't really meaningful statistics). While 80-20 originally referred to a specific distribution (following a scale-free, or power law behavior), it's commonly used as a way to describe any situation that looks something like the following:

When we talk about SEO on this blog, we're really talking about that first 20% of effort. At a big company that has a dedicated SEO staff (like the guy I was talking to), the 20% is a given; they're focused on pushing further and further up that curve to the right. So in that sense, this guy was absolutely right. It's not worth it for LAS (or your company, in all likelihood) to be worried about that last 20% of results.

If you look through our SEO archives, you'll see titles like SEO Checklist: 6 Easy Things You Should Be Doing and Get Started With SEO By Writing a Blog. The 6 easy things from the checklist post might be 5% of effort that yields 20% of the results, lets say. Sites like Website Grader can help push you a little further. Then there's registering your business on sites like Google places or Yelp, and your blog on Technorati or Google News that pushes you a bit further up the curve. Starting and maintaining a blog fall a little closer to taking a decent amount of effort, while still generating results. And that's as far up the curve as we've really gotten on this blog, and probably about as far as we're likely to get.

For a small company, you're not going to be competing with the big guys at the top of the SEO ranks. You're better off getting to that 80% by going after long tail keywords and the like, and focus on maxing out your effort in other areas where you really can distinguish yourself, like having great customer service, or great content; things that people can see but search engines can't.

A while ago I wrote a post comparing two different systems for managing blog comments. Disqus and IntenseDebate are the two big players in this field, which is why my earlier post only considered those two as options. My conclusion was that IntenseDebate was better, but only slightly.

Well, I recently changed my mind. Now I'm of the opinion that you shouldn't use Disqus or IntenseDebate. In my opinion, you're generally better off using the native commenting system such as the built-in Wordpress comment system (or a homemade system in my case). Let's take a look at the pros and cons.
 

What's better about Disqus/IntenseDebate?

Despite my newfound opposition to both of these products, they do offer some great advantages which I want to mention. First, both of them offer tons of features that you won't find in native commenting. They have all kinds of bells and whistles such as reputation points, Twitter reactions, and much more.

Also, both systems are widely used (Disqus especially) so by using them on your blog, you're creating a more consistent experience for your readers. If someone already has a Disqus account, that means they can easily comment on your blog.
 

Why I still wouldn't use Disqus or IntenseDebate

Despite all the great features, I can't recommend either of these products. I've used Disqus on three different blogs, and IntenseDebate on one. I ran into major bugs with both of them. The problems with Disqus were much more severe (which is why I originally picked IntenseDebate for this blog), but comments are just too important to accept any problems at all.  Commenting on a blog is incredibly basic, and easy to implement (I built a commenting system for a friend's blog from scratch last night and it's working fine). There's just no reason to accept a buggy commenting system.

I should say that both companies responded to my bug reports promptly and they were very nice, but neither could solve my problems. Nice customer service is great, but it's meaningless if nothing is resolved.

Another issue I have with both Disqus and IntenseDebate is how much longer pages take to load. Because they aren't built directly into the blogging software (they're hosted on an external server), the comments sections have to load after the rest of the page. By using a built-in comment system, you can significantly speed up page loads.

And then there's the SEO aspect. Because third-party comment systems don't load with the original page, that must mean that search engines can't crawl the comments. This isn't a big deal because comments rarely contain keywords that show up in searches, but I've seen it happen before. Why not let search engines index all that great user generated content?

Customization is another concern. When the commenting system is installed directly into your blogging software, that generally means that you can make changes to the code. Disqus and IntenseDebate both offer tons of customization options, but if you want something unique, you're out of luck.

To wrap things up, I just want to point out that comments are incredibly important to most blogs, but they don't need to be very sophisticated. All the bells and whistles offered by Disqus and IntenseDebate are cool, but completely unnecessary. If I have to choose between an awesome system that occasionally doesn't work and a clunky system that is rock solid, I'm going clunky every time.

Note: You may notice that we're still using IntenseDebate on this blog. That's just because I haven't gotten around to switching us to my homemade commenting system yet, but I've started using my own system for new blogs that I set up. I plan on updating this blog eventually.

Tyler and I live on opposite coasts, so as you might imagine, we spend a fair amount of time communicating over various channels including email, IM/chat, telephone, and video chat. With rare exceptions, mostly related to graphical discussions, I don't think our work is made any harder by using these channels as opposed to meeting face to face (and may be helped to some extent).

Given the many ways to discuss ideas, projects, and the like, it seems worthwhile to think about the strengths and weaknesses of various options. The main differences between the media, at least in my eyes, come along four dimensions:

1. Permanence/records
2. Scheduling flexibility
3. Topic flexibility
4. Multitasking compatibility

So here's how the various options stack up along those axes, in order of most used to least:

Email
The benefits of email are pretty well laid out, but it's worth reiterating them. Email is king at the first two points. Assuming one archives messages, every email communication is cataloged, providing a searchable comprehensive record. Most text chat options have similar permanence, but I personally find searching through chat logs significantly more difficult. Email also provides extreme scheduling flexibility relative to the other options which require both people to be present at the same time. For Tyler and myself, who are separated by 3 time zones (plus a few more hours in terms of our effective work schedules), such flexibility is key. Multitasking is clearly pretty good as well, although the scheduling flexibility also means crafting an email can take more time than it should if you're not careful. Email's main deficiency is topic flexibility, as it's difficult to shift gears quickly in an email thread. 

IM/text chat
Tyler and I are basically always chatting online when we're both working, as a way to run quick questions by each other or to flesh out some ideas. As mentioned above, assuming you have decent chat logs, the permanence scores well, though the difficulty of retracing a chat marks it down a touch. Scheduling flexibility is ok, only in as much as chat is the hands down multitasking winner in my opinion, and is thus accessible at almost any time. There are very few circumstances that prevent a quick chat conversation, even compared to email. Topic flexibility is also a win for chat, as the permanent but instant format let's one switch topics with ease, without abruptly abandoning previous topics.

Voice/video chat
If you've been keeping track, I've already doled out top prize for the four categories to email (permanence, scheduling) and chat (topic flexibility, multitasking), which doesn't leave much for voice and video calls. Keeping records of these calls is always difficult (and we usually use email if there are important points to remember). The scheduling flexibility is not good, particularly for video. And the topic flexibility is ok, but can be trickier in these media than something like chat, where it's easy to switch back to the main topic after a tangent by looking through the history.

The real win for these types of media, however, comes from their complete failure at multitasking. When on the phone and even moreso on video, your attention is necessarily focused on the conversation. Ty and I have started having video calls on a roughly weekly basis, and they have really provided a great complement to our other forms of communication. Given this, I think the voice call is pretty much dead for Tyler and myself, but the video call will live on as the ultimate conversation unitasker.

I sum up
Anyway, that's my take on things, as an analysis of what works for us, but I definitely think it informs the strengths and weaknesses of various options. For Tyler and myself, in-person meetings aren't a real option, but it's worth thinking about where they'd fall in the above categories. In almost all respects, they're really just an even more effective unitasker than video or voice. It's hard to think of them as such, and there are certainly other advantages, but for a quick chat or a detailed response, I'd imagine IM and email, respectively, are generally better options. Either that, or this is all just sour grapes.

If you write a blog about news and current events, there's an easy way to increase your reach on Google. You can submit your blog to Google as a news source so that it will show up on Google News. Not only that, but Google shows recent news results at the bottom of the first page of normal searches. This can be a great way to (temporarily) show up on the first page of competitive searches.

Not every blog can get on Google News. This blog for example wouldn't get accepted to Google News because we tend to focus on tips and suggestions rather than current events. They also only accept blogs with more than one author. You can read all about the requirements and learn how to submit your site for review in their Getting Started guide. It's definitely a great way to boost your traffic assuming you can get approved.

Last week I talked about how a wiki can be a great organizational tool for your team. In case you don't feel like reading my original post, here's a brief explanation of what a wiki is: A wiki is basically just a website that can be edited by anyone on your team or at your company. It's used to store information about all the different things you need to keep track of such as documentation, files, to-do lists, etc.

As I mentioned in that post, there are a bunch of different tools to help you set up a wiki, but my favorite is Google Sites. In this post, I'll go over how I set up my wikis using Google Sites.

The first thing we need to do is make a new "site". Log in to Google Sites at sites.google.com (or sites.google.com/a/YOURDOMAIN.com if you are using Google Apps with your domain). Click the "Create new site" button.

When creating the site, you'll need to choose a name (I'd go with "Wiki") and you can change the URL if you want to. By default the wiki will be shared with everyone on your domain (if you're using Google Apps) which is probably what you want, but you might want to click "More Options" and make sure everything is set up the way you want. When everything is ready, click the "Create site" button.

The first screen you'll see is the home page for the wiki. Now that the site is created, you really just need to know how to a few things: Create new pages, organize your pages, and create a dashboard to make the wiki more accessible.

Creating new pages

Your wiki will be made up of a bunch of different web pages. You'll want to make a different page for every different item you want to keep track of. Here are some examples of the types of pages you might want to make (some of these should probably be broken into several pages):

• An introduction for new employees
• A list of files that employees might want to print such as time sheets, vacation requests, reimbursement forms, etc.
• Documentation about how your operations work such as information about vendors you purchase from or documentation about your I.T. procedures
• Scripts and training materials for specific departments such as sales

The idea is that all the knowledge at your company should be stored in the pages of your wiki so that if one of your employees gets hit by a bus (or if they're just on vacation), you won't lose access to any critical information.

So how do you make a new page? Just click the "Create a page" button at the top right of the screen. You'll be asked to enter a new name, and you can also choose the type of page you want. Most of the time you'll want to use a "Web Page" because that's basically just an empty text document. the "File Cabinet" page type can be useful if you want to make a page to upload and organize external files, but you should generally enter as much information as possible directly into the wiki rather than uploading lots of other files because the search function won't be able to search the contents of files you upload. You can also make a "List" page which can act as a public to-do list. I'd stay away from "Announcements" and "Start Pages" for now.

Once you've picked the page type and named the page, click "Create Page". Assuming you went with the "Web Page" option, you'll see a basically blank page with the title you chose. This is where you should enter all the information that belongs on the page. Enter everything you want to show up on the page and click "Save" in the top right corner. Whenever you visit this page in the future you can click "Edit" in the top right corner to make changes, so don't worry about getting everything perfect from the start.

Now that you know how to make a basic page, I suggest you spend some time entering as many different pages as you can think of. Once you have a decent number of pages made, you're ready for the next step...

Organizing your pages

There are two different ways to find information on your wiki. The first is to search using the search bar which is on every page. That's pretty self-explanatory so I won't waste your time talking about it. The other way is to browse to find the page you want, but this requires you to set up a structure for your pages so that you can find everything.

To make browsing easy, you'll need to add links from the home page to the other pages you created, but you probably won't be able to put all the links on one page. You'll need to break the pages into categories. Each category will need it's own new page which will link to all the pages in that category.

Note: there are different ways to set this up in Google Sites. They have a built-in page structure which you can try out, but I suggest manually building a directory as I describe here even if you also use the built-in structure.

For example, let's say you want to break your wiki into the following categories: Internal Operations, Sales Training, and Project Management. First, you'd need to make a page for each one of those categories. Once those pages are made you should add links to them on your home page so that they'll be easy to find. Then from the category pages you can link to the individual pages in that category, or you can break things down even further by making sub-categories. Here's an example of what the "Operations" category page might look like:

As you can see, this page is just a list of links to other pages. By breaking down all your content into categories and using category pages like this to link to them, you can set up an easy way to browse all the content on the wiki.

That covers all the basic functionality you need from a wiki, but there's one other thing I'd suggest.

Creating a dashboard

If you're storing a lot of information in the wiki, it might be helpful to turn your homepage into a dashboard with a summary of important info. If you want to try this out, go to the home page and click on the edit button.

You'll see a menu at the top of the page with a "Layout" option. Click on that to give your page multiple columns. You can obviously lay the page out however you want, but I generally use two columns for the dashboard. You should already have the links to your category pages already entered, so now you just need to pick other things to add. You can add components to this page by clicking the "Insert" menu item. There are a lot of options, but some of the most handy things to include on your dashboard are calendars, google docs, a list of recently updated files, or recent list items.

I'll let you play around with the endless options, but to give you some inspiration, here's the dashboard from one of my old wikis:

Hopefully this tutorial helps you get your wiki started. I promise that once you get everything set up, you'll fall in love with how easy it is to access all your information. You should also encourage other members of your team to log in and keep it up to date.

Let me know in the comments if you have any questions.

Note: This post is meant for people that build, or are interested in building software. If you don't care how software is made, you might want to skip this one.

Ok, so this title is a little inflammatory. Before getting into this, I should say that I am a huge fan of open-source software. This site is built on PHP, MySQL, and Apache. We also use all kinds of open-source libraries like JQuery. This post is not an argument against using open-source software, it's an argument against building new software based off an existing open-source project.

Let me explain what I mean. When starting a new software project, you basically have two choices. You can build the entire thing yourself from scratch, or you can take code from an existing open-source project and modify it to fit your needs. For example, there are a number of web applications based on the popular open-source blogging software Wordpress. Because Wordpress is open-source, developers can easily download the source code and use it as a starting point for new applications (rather than starting from scratch).

Why would you start with an open-source project?
There are plenty of good reasons to base your project off something like Wordpress. They already have many common elements built that every website needs. Whether you need the basics like a login system or even a full blown CMS, Wordpress definitely offers a solid starting point. You can skip past weeks of tedious development by starting with a working code base.

So if open-source software can save you so much time, what's wrong with it?

Problem #1 - It's harder to make changes
The reason that open-source software is easy to start with is because other developers have already written ridiculous amounts of code. This code may be very well written, but it's also probably completely alien to you and your developers. It also probably includes a lot of stuff that you don't need. It's easy to go into something like Wordpress and make minor changes, but it can be incredibly challenging to try to fundamentally alter how a large open-source project works.

So if you're building serious software (and not just a website about your cat), you may actually save time by starting from scratch. If you start with a clean slate, that means that all the code will make sense to you, and it means there won't be anything unnecessary complicating things.

Problem #2 - It stifles creativity
This is much more important than the first problem. One of the most difficult parts of the software design process is when you start with a blank canvas. It's also one of the most important parts. When you don't have anything built, that gives you ultimate creative freedom. Your project has no form or direction, which puts you in a position to create something truly inspired. When you're designing something from scratch, every decision you make will be based on what you think is right rather than standards set by someone else.

When you start with someone else's project, you're boxed into a corner before you even get started. Sure you have full access to the code so you can make whatever changes you want, but you're a lot less likely to get creative. Wordpress already has a login page, and you'll use that as your starting point. That means that whatever you end up with for your login page will be some altered version of what the Wordpress designers came up with. You never have an opportunity to truly create something for yourself. There's almost no way to design great software when you skip the "blank canvas" stage.

I'm sure many people will disagree with my opinion on this topic. It's certainly easier and safer to start with an open-source project, but I think that doing so dooms you to mediocrity (unless you rewrite tons of code in which case there was no point).

Just to clarifying, I'm totally fine with using open-source libraries and frameworks (JQuery, Rails, etc.). I just don't think that your core product should be a modified version of someone else's core project (normally). Also, I have no problem using Wordpress or Joomla to build a basic website or blog. My argument only applies to people trying to build real applications with functionality that goes way beyond the built-in features of any open-source project.

As I mentioned in passing in a previous post, I recently finished up some schooling. In the process of finishing, I've been stripped of a number of heavily subsidized student software licenses, and have been looking into replacing them with full licenses for my new job in the real world. As happens frequently with such software, there's been a certain amount of sticker shock as I look into the full retail price.

In particular, one of my most used software packages (MATLAB), which runs around $50 for a student license, costs $2000 in the private sector (plus another few thousand for various upgrades that I'll need). At first, I was pretty shocked at this price (and still am to a large extent). Despite the fact that the company will be paying for the software, I was still hesitant to ask them to purchase such expensive software.

So all of that intro is really just a way to lead into the main point of this post: how much is your software worth? Without devolving too much into my poor grasp of economics, there are two different ways to think about it:
(1) the worth of software is determined by what you're able to do with it that you couldn't without it. From this perspective, I'd say almost all software is a tremendous bargain. From the example above, having MATLAB available will unquestionably enhance my productivity by at least a factor of two (and in all likelihood, much more). From that perspective, a few thousand dollars is a steal given the productivity boost relative to, say, my salary. The same type of gross productivity gains almost certainly exist for the software you rely on in your business.

Now, there's obviously a problem with the above analysis. If I were to apply the same reasoning to, for example, food and water, I'd determine that they are also worth a great deal (as I'd be unable to survive without them, which would in all likelihood negatively impact my productivity). That brings us to

(2) The software is worth what it costs to produce. In particular, for most products, it's the average cost of producing one more copy of the software (or, maybe, the average cost of making one piece, including the first). Much like food or water, this is the reason that many software products are so cheap. Making one additional copy of a piece of software is virtually free, and assuming that many copies are sold, the same applies to the average cost.

As anyone who had taken an Econ class (or has started to take one before switching out) knows, the price of a product is set by the intersection of supply (as dictated by the "worth" in in (2)) and demand (as determined by (1)). When there is a difference between the two perspectives, there is value to be had. From the buyers perspective, they pay less than the software is worth to them, and for the seller, they earn more than the product cost to make.

For any reasonable economic situation, both sides end up benefiting, because otherwise the transaction wouldn't occur. Software, however, represents a particularly extreme case. As mentioned in (1), most software provides extreme value to the consumer in terms of productivity gains. Imagine life without email or the Internet, and the overall value of software to the user is pretty clear. As mentioned in (2), the marginal cost of an additional copy of software is almost nothing. As such, virtually any payment represents a huge benefit to the seller.

The result of all this is the extremely wide range in prices of available software. If even two competing products exist in a particular area, the price is quickly driven to basically nothing (consider webmail as an example), demonstrating how little an email product costs to produce. In contrast, if a single dominant product exists, such as Photoshop (or MATLAB, for me), the price approaches the value to the consumer, independent of production costs.

So the conclusion of all this rambling is as follows: software is basically free to produce, and as a result, it's usually cheap. But that doesn't really reflect it's worth, which is much better indicated by the monopoly-type situations in place for dominant forces like Photoshop, and specialized software like MATLAB. The next time you're pricing software, it obviously pays to shop around, but the chances are good that you're getting a great deal regardless.

If you have your own website or blog, then you know how hard it can be to monetize your traffic. Last month I wrote a post about different ways to make money with a blog, and I want to dive deeper into one of those methods: the Amazon affiliate account.

Amazon offers a program called Amazon Associates. It's basically a way for you to market products sold on Amazon and take a small commission on all the sales you generate. This can be a nice alternative (or addition) to advertising on your site because it's much more focused and targeted.

Here's an example of how it might work. Let's say you periodically review books on your website. Normally you would include a link to Amazon or some other online bookstore so that your readers can buy it if they're interested. With Amazon Associates, you'd log into your Amazon account and generate a special link to the product page for that book. Whenever someone clicks on that link, Amazon knows that they came from your site, and so they pay you a small commission on the sale. The experience is exactly the same for your users, but you get a small reward for referring potential customers to Amazon.

Obviously you can do this with more than just books. If you write a tech blog, link to electronics on Amazon. Music sites can link to MP3s. Virtually all websites talk about products from other companies in one way or another, so it shouldn't be too hard to see where Amazon can fit in. (Note: In case you're wondering, no, we don't use affiliate links on this blog. The Less Annoying Blog isn't meant to make money)

So first thing's first. How much money can you make? The rates vary based on what types of products you sell (MP3s pay better than electronics for example) and how much you sell. The lowest you can make is 4% of the total sale amount, and it can go as high as 15% for some items. You can see the full fee structure here, but it looks like most websites with moderate traffic would be making about 6% commissions.

Now let's take a look at some of the different tools Amazon offers to help you market their products. There are four things in particular that I find interesting.

Links
I explained how links work in my example earlier. Basically you can just make a custom link to any product on Amazon. You include that link somewhere on your website, and then you'll get credit for any sales from the link.

Banners
Banners are sort of like pre-made advertisements that you can easily include on your site. Because the banners have graphics in them, you can't generate your own which means they don't actually point at specific products. Instead, you pick a general category for the banner (like "electronics" or "Health and Beauty") and Amazon gives you a list of options to choose from. Then you just copy and paste some HTML code and you're done. This can be a great alternative to ad networks like Google Adsense because you have a lot more control over what your users see.

Widgets
Amazon offers all kinds of widgets to include on your site. The widgets are big and graphical like the banner ads, but they're also interactive. For example, you can include a "My favorites" widget so that your visitors can see products you recommend. Other interesting widgets include "Search", "Deals", "MP3 Clips", and "Wish List".

aStore
This feature is really cool. You can build your own customizable store populated with products from Amazon. This allows you to sell appropriate items to your audience, and it's incredibly easy to set up. If you have real products to sell you might be better off using something like Shopify, but this is by far the easiest option I've seen if you're just reselling products.

So if you're looking to make some extra money with your website, definitely take a look at Amazon Associates. Signing up only takes a few minutes (here's the link), and you can start building links, banners, widgets and more right away.
 

One of the biggest challenges businesses of all sizes face is the organization and storage of information. It's tough enough to keep track of your own thoughts, and it gets significantly harder when you work with other people. And it's not just enough to store everything on a hard drive somewhere. You need to actually be able to find the information you're looking for in the future or else there's not much point in storing the info in the first place.

That's where a wiki a can help you. Most people have heard of some wiki sites such as Wikipedia and WikiLeaks, but it seems like most people don't really understand what a wiki actually is. A Wiki is just a website where the users can control the content. Think of it as an online version of the bulletin board at your office. Anyone in your office can add, remove or edit content from the board, and then everyone else can see it. But unlike the bulletin board, a wiki has virtually unlimited storage capacity, and search is built in so you can easily find whatever you're looking for.

Setting up a wiki will give everyone you work with a clear place to store documentation, files, procedures, and anything else that you can think of. Here's an screenshot of the homepage of a wiki my friends and I used while working on a project (Anooly.com) a few years ago:

Hopefully that image gives you a good idea of how great a wiki can be for collaboration. From document sharing to a group to-do list, a wiki can handle just about anything.

So now you should be sold on the idea of using a wiki for your business, but you're probably wondering what you need to do to set one up. There are dozens of decent wiki platforms out there, but I personally like to use Google Sites. If you're more technical and you want to host the wiki on your own servers, I'd recommend MediaWiki (the same platform Wikipedia uses) or DokuWiki. I'll be writing a full tutorial on how to create a wiki in Google Sites at some point, but for the time being, here are a few things you should look for when deciding what software is right for you:

• Privacy/Security - Most wiki software let's you block access to the outside world so that you can store private information. Make sure you can create accounts for your team and that no one else will be able to view your wiki.
• File storage - You'll be entering most information directly into the wiki itself, but you will also probably need to upload and store existing files. Before deciding on software, make sure it can support your file upload needs.
• Different page formats - What will you be doing with your wiki? Do you need to-do lists? A calendar? Forms? Some people prefer only the most basic content editing features, but maybe you need something more sophisticated.
• Search - As I mentioned, there's no point in entering information if you won't be able to find it again. You should definitely test out the search capabilities before deciding which product to use.

Well that's it for now. My full wiki tutorial will be ready in the next week or two, but hopefully you won't let that slow you down. Go out there and start a wiki for yourself.

The main focus of this blog is to provide tech tips for small businesses, and it seems like about 97% of our posts end up being about Google. This is partially because I'm a bit of a fan, but it's also because Google products touch on pretty much all aspects of the small business universe including advertising, location discovery, and productivity tools.

Well, if you enjoy the content on this blog, that means you'll probably like Google's new project: The Google Small Business Blog. It just launched earlier this week so there's not a ton of content up there yet, but it looks like it will be a great resource for anyone that relies on Google products (and really, who doesn't?). I expect to see a lot of tips on using Google Apps (gmail, calendar, docs), Adwords (online advertising), Places (listing your business on Google maps), Youtube, and more.

So I encourage you to add Google's new blog to your list of regular reading. If you don't feel like checking another website every day, maybe it's time to start using an RSS reader to make it easier to stay on top of the blogs you follow.

One of the best features of Chrome is the ability to create "application shortcuts" that let you launch web sites as stand alone applications. Unfortunately, application shortcuts are disabled in Chrome for Macs. While other options exist to create site specific browsers (SSB) based on Firefox (Prism) or Safari (Fluid), I'm a big fan of Chrome, so I wrote a little script that lets you create your own Chrome application shortcuts on a Mac. The script basically automates a solution that I found when looking around online, at the end of this post.

What are Application Shortcuts?

We haven't talked much about site-specific-browsers or application shortcuts (Chrome's version of SSBs), but they can be a great way to make your web applications feel more like desktop apps. SSBs basically provide a single purpose application that launches a single web site with it's own window, process, and icon. This allows you to manage your web applications along side your desktop apps. In the case of the solution shown here, it also lets you manage separate profiles for each of your web apps, letting you easily handle multiple logins, or have different sets of extension for each app.

How to use the script

If you've ever run a shell script before, it should be pretty obvious how this script works. Either way, here's a quick run down about how to create an Application Shortcut using the script.

Download the script
To get started creating your own Chrome application shortcuts on your Mac, download the script from here, and save it to your Desktop (or wherever you'd like). To run the script, you'll need to open up Terminal.app, which you can find through Spotlight or in your Applications folder. Once you've launched the terminal, change to the directory where you saved the script (e.g., type 'cd ~/Desktop' if you saved the script on your Desktop). Make sure that the script is executable  with the command 'chmod +x makeApp.sh'.

Prepare the app
The next step is to decide what site you want to use to create an Application Shortcut. For this example, I'm going to create an app for Google Analytics. To create an Application Shortcut, you need three things:

1. A name for the app; for now, the script doesn't support names that include spaces (sorry). For the example, I'm using 'Analytics'
2. The full url of the app, including http://. The address for GA is 'http://www.google.com/analytics/'
3. An icon as a png or jpg. If you don't include an icon, you'll just get a default icon. If you want to use a proper .icns Mac icon, just add it to the application after creating it. I found this analytics icon in the Fluid Icons Flickr group. I saved the png to my Desktop with the name 'analyticsIcon.png'

With these three things in hand, you're ready to create the application shortcut. Run the script by typing './makeApp.sh' into the Terminal window. The script will prompt you to enter the three pieces of information from above. For the url, you can drag from the address bar in your web browser and drop in the terminal to transfer the url. For the icon, you can do the same thing by dragging the image file into the terminal. After answering those three questions, your web app will be created and ready to run. I should also point out that the script assumes that Chrome is installed in '/Applications/Google Chrome.app'; if you have Chrome installed elsewhere, it should be pretty obvious how to change the script if you open it up with a text editor. 

Run the app
Once you've created your application shortcut, you can launch it through Spotlight, or in the Applications folder of your home directory. Once you've launched the app, you can set the icon to stay in the dock even after closing the app. If the site you've created the app for requires a log in, you'll need to log in (and save the password if you'd like). This is necessary because these application shortcuts each have their own Google Chrome profile. While this means that bookmarks, passwords, and extensions are not shared with your main instance of Chrome. It also means you can stayed logged in to your account without affecting your main browsing session. If you want to add extensions to an app shortcut, you can create a new window (Apple-N) to get a full Chrome window in the correct profile. It's also worth noting that the behavior of these shortcuts isn't perfect after closing the last window without quitting out of the app. If you do so, and then click the icon again, it will launch your Chrome homepage rather than the correct app. To avoid this problem, just always quite the app (Apple-Q) when closing the last window. Aside from these drawbacks, your application shortcuts should behave just like other apps.

Once you've gone through this procedure once, you can repeat it to create apps for all your commonly used sites that might be better served with their own icon in the dock and in your application switching. The separate profiles also means that you could have, for example, multiple versions of Gmail that automatically stayed logged in to different accounts. If you run across any problems with the scripts, feel free to let me know in the comments and I'll try to help out.

Product Name: SheerSEO
Website: www.sheerseo.com
Price: Plans differ, but the best value seems to be $10/month (after a 90 day free trial)
Overall Grade: B-
Recommendation: This probably isn't the right product for most people, but it's worth using the trial to find out for sure.

Overview
SheerSEO is a very basic tool for staying on top of the SEO basics for your site. It tracks your site's search engine rankings, what other sites are linking back to you, and a few other important inbound marketing metrics. SheerSEO doesn't actually help you action, it just gives you information and lets you use decide what actions to take.

The Good
The best thing about SheerSEO is the price. For some reason, most software that helps with SEO is ridiculously overpriced. SheerSEO isn't exactly cheap (it ranges from $7 to $200/month) but it's an absolute steal relative to the other SEO software out there. They also give you a 90 day free trial which is awesome. I think that's the longest free trial I've ever experienced.

SheerSEO also does a decent job of focusing on only the most important SEO metrics. Other tools definitely offer a lot more functionality, but SheerSEO covers the essentials and all the features are easily accessible.

The Bad
My main complaint about SheerSEO is the design. It seems like a lot of effort was put into collecting useful data, but the mechanism for displaying the data is underwhelming. The best pages are clunky and unattractive, and the worst pages are useless and confusing. For example, the image to the right shows an entire page. Seriously, there are basically 12 pages linked to from the navigation bar, and one of them just shows the number of backlinks (and a whole lot of whitespace). There are a few other pages that are similarly sparse. The information SheerSEO provides is pretty useful, but they need to put some thought into how they display everything (a dashboard would be nice).

The writing on the site could also use a lot of work. From what I can tell, the company is based in Israel so it may have been written by a non-native English speaker. Either way, they should hire someone to re-write all the text. The entire application only has a dozen or so paragraphs, so it wouldn't be much work at all. As an example of the confusing writing, there is a page called "Guarded Backlinks". I don't know what guarded backlinks are so I read the instructions and this was the first sentence:

The "guarded backlinks" is a feature which enables you to guard your site's valuable backlinks.

So yeah, that wasn't too helpful. Most of the time I could figure out what was going on, but the product would be much more usable if they spent an hour re-writing all the text.

Finally, as I mentioned above, SheerSEO doesn't really help you take action. It doesn't recommend new keywords or suggest improvements you could make. Basically, they just give you some info and you have to figure out what to do with it. I still find the software really useful, but I don't think it would be good for SEO beginners.

Conclusion
I'm very conflicted about this one. SheerSEO is definitely useful, and the price is right, but I'm not really sure who it's meant for. It's not designed well enough to appeal to beginners, and it doesn't offer as many features as you'd want if you're serious about SEO. So I think it's a great tool for people that know and care about SEO but don't consider it a major priority. If you just want an affordable way to monitor your keyword rankings and backlinks, SheerSEO is probably a good choice. If you're still learning about SEO or you consider it an important part of your job, you should probably keep looking.

But remember, there's a 90 day free trial, so you might as well sign up (you don't need to give a credit card or anything) and try it for yourself.

Last week Bracken touched on the usefulness of planning at a start-up. It's pretty common these days to suggest that planning isn't very useful. The consensus among most lean start-ups seems to be that you should spend as little time as possible in meetings reviewing strategies and projections, and you should instead spend that time actually building your company.

This tactic is understandably controversial. After all, if you don't spend much time planning out your business, aren't you sure to make mistakes? Businesses don't randomly become successful (except for Twitter). Success is generally the result of great ideas, great strategy, and great execution (again, this makes you wonder why Twitter is so successful). So how can you really argue that planning isn't important?

This conversation is happening all over the world. New school businesses criticize planning while old school businesses defend it. But I think that both sides are speaking different languages and that's why no one can agree. No one can really deny that planning is absolutely essential to the success of any business. The real issue is: What is the most effective way to approach planning?

10 years ago, there was only one way to plan out your business strategy. You'd sit in a conference room with your business partners and look at excel reports, powerpoint presentations, and endless printed out pages of information. You'd research the current market, come up with projections for the future, and use these numbers to predict how your company might perform. After days (or weeks or months) of this, you'd come up with some clear strategy and then you'd get to work executing it.

That strategy might have been necessary 10 years ago, because most businesses needed significant investment to get off the ground. A mistake early in the process would have been incredibly expensive, so it was worth spending as much time as possible up-front to minimize mistakes. Well, things have changed. There are many types of businesses that can be started with almost no investment other than time. Software in particular is an industry where you really only need a computer to get started (which you probably already have). As long as you can pay your electric bill, your company is still alive. This means that the cost of failure is insignificant. You can start a business today, and close down a week from now, and you'd probably just be out about a hundred dollars and some time.

The ability to accept failure without any major consequences significantly changes how a company should approach planning. It's not that planning is any less important for these companies, it's just that they can plan much more effectively by combining the planning and execution stages into one. These companies can spend a couple hours brainstorming ideas, pick a direction, and then get started. Every few hours they can review what they're doing and adapt based on their experiences. Traditional companies have to base their planning on assumptions (a.k.a. guesses). No matter how much time you spend, you can't know for sure what challenges your business will actually face until you actually face them. By iteratively planning and executing in small increments, you can make sure that you only spend time on challenges that actually matter.

So to summarize, I don't think anyone can deny that planning is essential. But I encourage everyone to consider that planning doesn't have to happen all at once before you get started. The more real-life experience you have, the more relevant information will be available to you. The more information you have, the better your plans will be.

Over at Less Annoying Software, we recently published a white paper about how customer management software can help small businesses and individuals improve relationships with their customers (where "published" in this case just means posting a PDF online). The paper also discusses some of the problems that small businesses run across when using CRM software (which has largely been designed for larger enterprises), as well as important features and considerations when choosing, implementing, and utilizing such software. I've written a brief summary of the paper below, but if your interest is piqued, download the white paper here (direct PDF link).

What is CRM
Despite the fact (or perhaps because of it) that our product at Less Annoying Software is a web-based customer management tool, we haven't talked much about CRM software on this blog. Customer relationship management (CRM) is defined as a widely adopted strategy within a company for managing, documenting, and optimizing all interactions between the company and its customers. Over the past 15 years or so, CRM software has been increasingly adopted in order to improve customer service and sales efficiency, particularly by larger enterprises. Small businesses have recently begun to jump on board as well, at a rapid rate.

The problems with CRM
While CRM can unquestionably be an outstanding tool for small businesses, problems arise when using software that was originally designed for much larger companies. Despite the clear benefits of well-implemented CRM systems, around half of all CRM implementations fail to live up to expectations. The majority of these of these failures are a result of companies and employees not fully understanding or adopting the system (as I mentioned last week).

Customer management software
While these types of failure rates may be a necessary evil at large enterprises, small businesses actually have better options. The poor adoption rates are largely a function of the complex nature of CRM software, but much of that complexity is tied up in features that aren't important for small businesses. Customer management software is a pared-down version of CRM software that focuses on the documentation and organization of customer interactions, with some basic reporting. This type of software can generally provide most if not all of the benefits of full-blown CRM for small businesses, while reducing the cost complexity, and implementation times.

Anyway, that gives you a flavor for what's in the white paper. If you're thirsting for more, feel free to check out the full thing (PDF link).