Alan Kasindorf just introduced me to a great blog by Eric Day, Oddments.
If you are into learning about alternatives to MySQL like Drizzle, or how to scale writes to a database using Gearman, then I wholeheartedly recommend his blog.
I really like the samples of code he puts up that acts as a very useful, and direct tutorial to new technologies.
This week a web project that I’m working on started to fall behind. On the key things that made this happen was adding a set of feature requests last minute.
I understand it’s important to deliver value, but when new sets of features are added to a project, it causes the project to slip.
What should you do?
Learn to say, “No.”
It’s hard to do, but take the design of the iPhone for example. Instead of saying yes to more features, Steve Jobs took the approach of saying “No,” in order to create something light-weight, elegant and a work of genius.
Less is more.
Ever see characters any of these characters:
1) ^M
2) <feff>
3) A black diamond with a white question mark in it.
4) What<92>s going on?
???
Does your version control system tell everything has changed when it doesn’t?
Does your web app break because one of the above characters?
If you answered yes to any of these questions, you are not alone!
If you’ve dealt with this and figured out it please definitely shared your findings in the comments below.
I believe it is *the* number one issue in working in a hybrid environment, e.g. with Windows, Linux, and OS X.
In the next blog posts, I’ll talk about how I’ve struggled with these issues with other developers.
Being on the bleeding edge of technology just isn’t as profitable as making money. A good example right now is the beer industry. BUD is making tons off of our down economic times. But I bet you if you look at BUD, you might find a non-networked FoxPro application, or something from 10 years ago. You might find that there are copying processes that you would’ve replaced with rsync ages ago, or user rights managements issues still solved with point and click.
I can’t say for sure, because I’m not there. If the clothing that most people wear in Fortune 500 companies is any indication, their tech is 10 years behind the times.
How do you communicate to them?
Kathy Sierra style and use a deck. Create a power slide presentation for free at Google Docs.
If you’re at a big company, a coder that turns his passion and goals into something others share uses a deck. By my count, you should spend an hour of your day making a deck to get a point across.
ScribeFire is a Firefox plug-in that allows you to post to your blog. In fact, I didn’t have to go through the WordPress interface to publish this, and used ScribeFire’s interface instead. It’s a must have plug-in if you’re a blogger.
Technorati Tags: scribefire, wordpress, plugins
It’s not that difficult to store data into your Amazon S3 account and share that data with 50,000 or so people on the cheap.
Armed with just your Access Key ID and Secret Access Key, you can use these two cool storage clients.
S3 Webmaster
For Windows, you can download the S3 Webmaster for free, and you just simply drag and drop your files and S3 will automatically set the permissions for public download.
S3Hub
For Macs, there is S3Hub, which works whether or not you have an S3 account. If you don’t have an S3 account you can view other public S3 buckets. If you do have an account you can easily upload and share the assets you stored.
Transmit
If your data storage needs run the gamut from sftp, webdav, iDisk, automated upload workflows, and S3, you’ll want to plunk down some cash like I did and get Transmit. At $17.95 it’s a bargain. I use it’s webdav and S3 features almost daily.
How fast can you get on the web? With newLISP it’s about as fast as typing:
How fast can you create a backdoor with newLISP?
If you telnet into port 1234 in localhost, you’ll see something that looks like this:
Trying 127.0.0.1…
Connected to localhost.
Escape character is ‘^]’.
newLISP v.10.0.0 on OSX IPv4 UTF-8.
>
This opens up a lot of possibilities for distributed computing.
For example, you can set up a newLISP server that’s ready to respond to a newLISP client with this command:
Your newLISP client can have code that sends a computing problem to be solved to the server:
Or let’s say you had a farm of newLISP servers:
(set ‘result (net-eval ‘(
(“192.168.1.100” 4711 {(+ 3 4)})
(“192.168.1.101” 4711 {(+ 5 6)})
(“192.168.1.102” 4711 {(+ 7 8)})
(“192.168.1.103” 4711 {(+ 9 10)})
(“192.168.1.104” 4711 {(+ 11 12)})
) 1000))
(println “result: ” result)
(exit)
If the above example reminds you of Gearman, you get +12 points.
If you work with ACL, this may be the most important video you watch ever. Zed shows how ACL is not Turing complete, which explains all the problems you’ve been having with control lists.
Zed Shaw – The ACL is Dead from CUSEC on Vimeo.
Pre-requisites:
Now we’re ready to build an EC2 LAMP Server.
You’ll find that a lot of ec2 stuff happens in the .ec2 directory.
To list the possible servers that you can set up run:
I ran
and got 1477 possible servers. Some are Windows.
Let’s say we see this listing:
If we want to start up the ubuntu server listed above we just type:
And then we run this command:
We should see either “pending” or the actual instance running with its FQDN listed in the 4th column. An example FQDN is this:
ec2-173-33-159-95.compute-1.amazonaws.com
And if we go to:
http://ec2-173-33-159-95.compute-1.amazonaws.com/
we should see a webserver.
And if we ssh:
we’ll get the root prompt:
#~
That’s basically it. Now you can go in and mess around with server settings.
In the next blog post, we’ll look at how to save your custom server settings and set up using S3.
I was really inspired by the story of Pownce., and how they used S3 for storing media assets.
I really liked how easy it was to share media with Pownce. Alas, it’s no more and shut down on December 15th of last year.
What’s a guy to do? Well, using a combination of Transmit, a file upload and synchronization client, and S3 I can hack my own file sharing. I just upload the files I want to share and send the links to friends.
Here’s a link to an S3 media asset.
Once I felt comfortable about S3, I took the dive into EC2.
Pros: You can have a server of your choice, in my case LAMP, running in less than 5 minutes.
Cons: If the server crashes, all your data is gone.
Solution: Use ESB for database storage and S3 to back up files and custom executables.
How much does it cost?
For 1 LAMP server with 1.7GiB RAM running with a virtual 1.7Ghz CPU, and about 160GiB of storage will run you about 10 cents an hour, which is not bad considering that a colo will charge you around $100 per 1U or 12.5 cents per hour, and you’re stuck with the hardware you have.
The savings are even greater if you’re just setting up servers for prototyping and tearing them down. My last EC2 bills averaged at about $35 per month.
The next few blog posts will go into more detail about how to set up your own EC2 virtual server, and the pitfalls I’ve run into.
