Startup or Bust

Cucumber Tests with Delayed Jobs

2011-01-27T03:42:30Z

Here’s a little step I use when testing scenarios that involve background jobs:

When /^delayed_job runs all pending jobs$/ do
  Delayed::Job.all.map{|j| j.payload_object.perform}
end

Is there a better way? For example, can you disable Delayed Job in certain environments?

Starting to Resurrect Startup or Bust

2011-01-20T01:37:04Z

We’ll see how it goes, but after a three year hiatus I’m itching to get back to blogging. In case anyone’s interested, all of the old posts are archived here: startuporbust.com/the-early-days.

Wowza Media Server Installation on a Mac

2011-01-19T21:02:44Z

For the first tutorial, we’re going to get our hands dirty with the Wowza Media Server Pro10 by installing the example video recording application on a Mac. If you caught the introduction to the Wowza Media Server you know you’re going to a few things for the tutorials, but for this one you only need the Wowza Media Server Pro10 developer license and your web cam. For your reference, make sure you grab the Wowza Media Server Pro User Guide.

Fortunately, Wowza has made installation super easy and have even provided an array of example applications to get you started. To install the server, just download the Mac OS disk image file, double click on the set up file, and run through the instructions!

Now let’s get the server up and running. We’re going to startup in standalone mode rather than service mode.

Running the server standalone is best for developing Wowza Media Server Pro custom applications since the server can be started and stopped quickly and server log messages can be seen immediately in the console window. Running the server as a system service is most often used for server deployment where there server needs to continue to run even after you log off the machine.

Starting up in standalone mode is done with the following commands in Terminal:

cd /Library/WowzaMediaServerPro/bin
./startup.sh

If this is your first time starting the server up, it should ask you for your serial number that was emailed to you when you “bought” your free developer license.

Alright, so we have our server running, now the fun begins!

Navigate to /Library/WowzaMediaServerPro/examples/VideoRecording and run the install.command file.
This will install the example video record application in /Library/WowzaMediaServerPro/applications.
Now open up /Library/WowzaMediaServerPro/examplesVideoRecording/client/videorecording.html in your browser and you’re ready to go!

Click record and stop when you desire. Then watch it play! If you don’t see the camera view, right-click on the record button and select settings and choose your webcam under the camera icon.

Also of note is that the file you just recorded was saved to /Library/WowzaMediaServerPro/content as videorecording.flv. You can edit this location in the /Library/WowzaMediaServerPro/conf/Application.xml file. (see page 21 in the user guide)

That’s it! Pretty simple right? This will provide the basis for installation on EC2 in the next tutorial where we’ll get into a little more detail.

Wowza Flash Media Server Pro Introduction

2011-01-19T21:02:44Z

Question: So we all know there are a ton of startups out there using Flash as their means to deliver their media content, but what does it take to implement your own video app?

Answer: You need a flash server and up until now, you only had one option ready for production use: Purchase Adobe’s Flash Interactive Media Server for almost $5000. (To be fair, there is a trimmed down version without video recording available for $995, but that doesn’t fit my needs.)

In my case, I’ve yearned to implement flash streaming in some of my projects – namely Echonote – but five grand is a pretty penny for bootstrapped projects.

Thankfully, there’s a new kid on the block: Wowza Media Server Pro which is now available for $995 with features that top Adobe’s offerings. Not only that, they offer Wowza Media Server Pro for Amazon EC2 which opens up the playing field even further. For roughly $100 a month(not including upload/download charges), you can have your very own Flash Media Server running on an EC2 instance.

Over the next couple of posts, we’re going to be looking at installing a video recording application with the Wowza Media Server Pro10 developer license on a Mac to get our feet wet and then onto running the app on EC2 with persistent storage on S3.

If you feel like getting a head start, make sure you have the following items set up. We’ll be using all of them!

Webcam
Amazon EC2 account
Amazon S3 account
Wowza Media Server Pro10 developer license
Wowza Media Server Pro EC2 license
S3 plugin for Firefox (optional)

Lazy Sunday Update

2011-01-19T21:02:44Z

Yes I’m still here! Thanks to those of you who have sent in notes to check up on me. Things are humming along just fine though work on Echonote had slowed considerably(and obviously Startup or Bust) until recently.

Thankfully, I’ve found more time to put into Echonote, Startup or Bust and a few other endeavors over the last couple of weeks and am looking forward to getting back into the swing of things.

On a different note, Philly is booming with great stuff going on all the time; from PHP meetups to Philly on Rails to Agile Philly and many others. So if you’re in the Philadelphia area, tear yourself away from your computer and get out to meet some really great people!

Here’s what’s going on this week:

Tues, Feb 26: Philadlphia Computer-Human Interaction Meeting – “Taxonomy is User Experience” by Dave Cooksey
Sat, Mar 1: IndyHall WordPress Workshop #1 (IndayHall Calendar)

Did I miss something? Let me know in the comments!.

Lessons in Usability | Credit or Debit?

2011-01-19T21:02:44Z

It sounds simple; swipe, sign and your quickly on to enjoy your morning pick me up. Unfortunately, it seems as though most credit card pin pads were not designed with the customer in mind. What can we learn from the confusion often created by the pin pad interface?

“Could you try swiping again? It didn’t go though.”

That always happens; my card is practically falling apart it’s so old

“Oh I see, can you please flip the card over? The magnetic strip needs to face to the left; like it shows on the pin pad.”

Ok hotshot, it’s swiped.

“Credit or debit?”

“Credit.”

“Great, just hit Cancel then Yes.”

Ooh. Where is that cancel button? Why am I pressing Cancel anyway?

“Cancel is that one on the bottom.”

Okay, I found Cancel. Now I just have to hit the Yes button. It must be that bright green one right?

“Yes is the yellow button that is two buttons above the green button, ~~dumbass~~ sir.”

Two buttons above the green? This one or that one? The words are a little too small to read….

I’m sure you’ve had a similar experience at some point or another; it’s nearly a daily occurrence for myself. So back to the original question: What can we learn from these hard-headed pin pads and apply to our online applications?

1. Your users cannot read your mind.
It is very easy to become too comfortable with your product during development that you start to believe that since you know your product inside and out, then everyone else does also. Make it a point to consciously review this during development and regularly run simple usability tests. (my favorite: have mom or grandma run through the application. If they get it, you’re onto something.)

2. Clearly and properly labeled actions are essential.
In this case, the button labels were too small to see and the image directing the user which way to hold the magnetic field was not clear. In addition, requiring Cancel to be pressed to choose a Credit transaction is not ideal, to say the very least. Again, make sure your labels are making it through to your users.

3. A simple interface is not always easy to use.
A card pin pad is a relatively simple device: you’ve got your numerical buttons along with the action buttons. The simple task of purchasing a coffee with this simple device becomes difficult when the interface is convoluted. Choose your interface wisely and be careful not to oversimplify things. If you can’t budge on keeping things simple, tooltips and on page descriptions are great ways to clue your users in to more complicated processes.

An Echonote Example

Every time a user on Echonote initiates a new note, they select everyone they want to send the note to. Users can also save groups so that they don’t have to manually select each person every time they write a new note. To make group management usable and unobtrusive, I wanted to allow users to save and remove groups right at the time they were writing the note instead of having a separate and cumbersome group management page. Using groups had to painless.

Below you can see the Save a New Group field with the original description, “Save your selected friends to a new group.“

Apparently, what I thought was a perfectly fine description was not connecting with the users. A good portion of them had ignored this feature simply because they were confused as to what exactly to do with it.

To make a better connection, the description has been changed to, “Type a name below to automatically save everyone receiving this note into a new group to use next time around.”

It’s a little wordy, but it does help to break down the group saving process by explaining what exactly the user has to do and the benefit doing so.

Moving Forward
As you continue on with your development, keep your users in mind and involve them as early in the process as possible; it’s invaluable. Hopefully we’ll all eventually live in a pin-pad syndrome free world.

Side Note: RightScale’s Scalable Redundant Servers

2011-01-19T21:02:44Z

Spending way too much timing dreaming up ways to take advantage of RightScale’s scalable redundant server offerings built on Amazon’s S3 and EC2.

Side Note: Redfin on their Financial Model

2011-01-19T21:02:44Z

Great insight from Redfin’s CEO, Glenn Kelman, into modeling a startup’s financials. It is amazing what the costs per employee can add up to.

Looking Ahead With a New Blog Design

2011-01-19T21:02:44Z

It’s been a long time coming, but I have finally finished Startup or Bust’s redesign. Instead of using a modified free template this time I journeyed out and started from scratch. As I hope you can tell, my main goal in the redesign was ridding the site of everything extraneous and keeping it as simple as possible to improve legibility and usability.

I originally thought I’d farm out the task to someone more qualified to do the job, but thanks to Colin Devroe, I stumbled across the excellent Blueprint CSS framework. This framework really gave me the confidence to tackle CSS for the first time without having to wrestle too much with cross-browser issues. One of the greatest things about the framework is that provides an awesome base for grid design with vertical rhythm.

Looking Ahead

With the redesign out of the way, it’ll be back to the regularly scheduled programming here at SoB. Well except for the fact that I wandered a bit off topic in the past and I’d like to change that. From here on out I’ll be focusing primarily on building online startups with of course sharing my own experiences with Echonote.

Speaking of Echonote(still a bit rough around the edges, so please excuse the mess), we’ve been doing some small scale user testing over the last few weeks which I’ll be sharing in the next week or so. A few important fixes and I’ll be on to testing with a larger user base.

Looking forward to sharing more in the near future as things move forward with Echonote!

Quick Update

2011-01-19T21:02:44Z

So apparently the Blogathon a couple of weeks ago completely “blogged” me out; you’ve probably noticed that it’s been pretty quiet here since then. I’ll be getting back into more of a regular schedule here on SoB in the next few days. Until then, a couple of updates:

Thanks again to all who participated and/or donated in the Blogathon – the experience was extremely rewarding.
Thanks to the one and only, Alex Hillman, whose top donation won him the $50 iTunes gift card. Enjoy!
Echonote is a bit behind schedule, but progressing nicely. I planned on releasing to a small user base this week but of course have run into the unexpected hiccups along the way. Just a few more tweaks on the back end, then over to the designer to tidy up the front end and we’re ready for the first launch. This will happen in approximately two weeks.
Soon after Echonote is out into the wild, I hope to give SoB a nice makeover with a new look and content refocus.

Exciting things to come as Echonote approaches launch!