Personally, I see a fair number of engineering talks, both in person, and online. Alot of the really great talks I've seen have come from great engineers that give talk themselves. Facebook, for example, has a great library of engineering talks, as do a handful of other companies... Last year I went to Velocity and got to see a bunch of great talks, and then I saw a "Bryan Cantrill" talk, and the nerd in me has found engineering talk nerdvana (A state of amusement, and attentiveness for more than 42 seconds at a go). If you haven't seen a talk by him, watch one, now! No really, now! 

Bryan, I love your talks, please make more of them, and not just because Jason drinks and agrees to pimp you out, because you sir, are more than just entertaining, and amazingly smart, but you are truely changing ideas and the future with each one, so thank you!

For those of you who've bumped into me the last few days will probably have come to hear that I'm going to part ways with Rackspace Cloud. While my time here has been great, it really has, the time for bigger, better, and more exciting things has come, and from Monday the 5th of December I'll be working for the Enterprise Cloud group within Sungard Availability Services. This job actually also comes with some old friends from my previous life as the Head of Technical Janitorial Services (HTJS) for Rackspace Private Cloud. These are the smartest, most driven folks I know, and while the choice ot leave Rackspace wasn't easy, the choice to join them was. 

I'll also be migrating to Denver, Colorado (and, I believe there'll be snow during my journey, yay!). So let me know if you're in the area, and come around to say hi :) So sorry about the lack of updates, and more updates to follow! (As well as a new Daikon release too!!)

UPDATE: Not sure why I didn't actually 'publish' this, but here goes... And yes, there was Snow, and Ice... which in a 430+ BHP rear wheel drive with summer tires (who knew they made winter tires??) is not really made for... at least, made for staying on the road. While I survived unscathed, as did my car, I'm certainly excited to have "Pirelli Scorpion - Snow and Ice" tires on my car now. 

So, just a handful of quick notes as I've not blogged in a little while. A few things have happened that've kept me pre-occupied the last few days... I officially sprouted a startup with a friend, and now co-founder, and I'm learning to write python. As a good project to get started with Python I also proceeded to sprout a CLI for ElasticSearch called 'Daikon' (after the radish used in Kimchi).

If you're interested in Daikon, or ElasticSearch, feel free to pop me a message, or go poke around in the code (it's OpenSource too...). My GitHub (which btw, rocks) page for daikon is : http://www.github.com/neogenix/daikon. Feel free to comment, or propose changes, or even make requests for features!

Other than that, I have arond 30 posts 'half' finished, so I should knock at least one of those out this week to keep the posts flowing.

I just stumbled upon a great article about James Dyson, and what makes different, and what made him the guy he is. The part that really got me was how he hires, and I think that's a very good way of choosing... certainly, you have to have an eye for the want, or drive within the folks you hire, but hiring people who don't know the right way to do things seems like the best way to go...

http://www.wired.co.uk/magazine/archive/2011/11/features/the-seventh-disruption-james-dyson

Feel free to comment below...

So as most of you know, my day job is working at one of the big boys in cloud infrastructure hosting. As such I've gotten to learn a TON about different cloud types, and etc... As part of my big drive to do a startup, I've gotten the run down of a few cloud providers, and have been trying to work out who gives the best bang for buck for my startup idea. In the meanwhile, I figured I'd do a post on different *major* types of clouds, and some examples of each. So without further ado, here're the major types of clouds :

Infrastructure as a Service or IaaS

This is the big one, and probably the one with the most examples right now. Essentially, you can look at this like an abstracted utility style version of what you have in your company data centre, and/or even your desktop type computing is moving into the IAAS model. Usually, this comprises of computing, memory, and storage in the cloud in different shapes, sizes, and/or forms.

Some examples of IaaS clouds are:

Rackspace Cloud (Cloud Servers, Cloud Files, Cloud Load Balancer), and then of course Amazon AWS (Amazon EC2, Amazon S3, Amazon EBS).

Note: This is also usually what gets melded into the form of 'Public', 'Private', and 'Hybrid' Clouds. Public being the above mentioned two, Private being a dedicated installation (usually), and then Hybrid being a mix of the two (again, usually).

Platform as a Service or PaaS

This is basically something along the lines of 'you load your app on our servers. We care about the servers, and the scaling, and you care about the app'. It's another level of abstraction above the IAAS model which allows you to focus on your application, and less on your infrastructure, while of course being a huge positive for the developer focused market, this isn't always the best option and in many cases creates vendor lock-in though a reliance on the underlying platform, and as such, the vendor. This is also one of the easiest ways to get locked into a platform, although over time the vendors are moving to a less 'one-sided' model on this front, and starting to adopt more open standards (such as SQL for example).

Some examples of PaaS clouds are :

Heroku, EngineYard App Cloud, Google App Engine.

Software as a Service or SaaS

This is the last line of Cloud Vendors. Essentially, this is where a company produces software that makes sense for you to allow them to run completely. The only thing you need to do in these cases is pay for them, possiby integrate with them, and then of course use them. This of course has the highest possibility of vendor lock-in with your data residing inside their application, but also this has the least outlay as you don't have to pay to have an app built, or keep it running, you simply use it. This is proving to be the most popular for buisness applications due to the instant on nature of the applications.

Some examples of SaaS clouds are :

Google Apps, ZenDesk, SalesForce

 

Comments, Questions, Ides? 

So like, lots of joysticks...

So, for those of you who know know me you'll have heard me say at some point or another that I want to start a startup, some have asked why, others just kinda assume. If you're wondering why, it's pretty easy to explain. I want to invest myself in something that I can help guide, shape, and mould and have true ownership in. As such, I've been reading a bunch of articles, and figured I should share some of them :

Joel.is: How to Start Your Startup in 4 Steps
URL: http://joel.is/post/5507881155/how-to-start-your-startup-in-4-steps?44977aa0

My biggest take away was that you should work on something you would use, something you're passionate about, something you'd buy, I didn't see this in any of the other articles, but I figured I must have missed it, because surely everyone believes that, right? I can't sell meat products if I'm a veggie, I can't sell cars if I don't know how to drive, and love driving for that matter, and startups are no different. Again, quick, small, simple, fast!

Joel.is: Fear of Not Shipping
URL: http://joel.is/post/3830271787/fear-of-not-shipping?1fcd2400

It's ok to push something that's not perfect, just get it out the door, and be willing to fix it fast. Now, there comes a point that your mistakes cost you too much, but initially, you're riding a coaster, and taking the chance is sometimes worth the risk!

BufferApp: Idea to paying customers in 7 weeks
URL: http://blog.bufferapp.com/idea-to-paying-customers-in-7-weeks-how-we-did-it

My take away from this article was pretty huge. Keep it simple, do it fast, and it's easy to get it out the door quickly at first, it will get better over time!

Paul Graham: Ideas for Startups
URL: http://www.paulgraham.com/ideas.html

Ideas are not worth millions of dollars, execution is! If you think you have a million dollar idea, someone else may have already thought of it, and they may already have executed on it, but if you execute on it in a million dollar way, the most insanely easy idea can be worth much more than that million dollar idea someone else thinks they have... Facebook is a great example of this!

Paul Graham: How to Start a Startup
URL: http://www.paulgraham.com/start.html

Understand your customers, don't do it alone, and have an idea... that's really what you need. The rest of the article covers funding and different insights around that, which are actually really great insights. Read it!

Paul Graham: Ramen Profitable
URL: http://www.paulgraham.com/ramenprofitable.html

This one is a great article to focus on. So many startup founders I've met are very much focused on selling as soon as they hit their 'mark'. While this is great, it's not always going to happen. Focusing on being Ramen Profitable is something that makes sense, and helps you focus on the important things, instead of the pay-day.

Paul Graham: The 18 Mistakes Startups Make
URL: http://www.paulgraham.com/startupmistakes.html

This was an epic article. The big take aways for me are that you shouldn't try do it alone, you should do it lean, and fast! It's a great article, and for those who don't know who Paul Graham is, you should google him. Oh, and location matters, if only for your ability to know people to help you... you can do it from anywhere I'm sure, but it's easier when you're surrounded by other startup minded folk at every turn.

StartupLawBlog: The Moonlighting Survival Guide
URL: http://www.startuplawblog.com/2011/08/05/the-moonlighting-survival-guide/

So one of the things that you should be mindful of is your current employer (most folks are going to start something while working for someone else) and their policy about moonlighting, and what your rights are. This is a good place to start, if only to get the creative juices flowing around possible stumbling blocks.

Chris Dixon: Developing New Startup Ideas
URL: http://cdixon.org/2010/03/14/developing-new-startup-ideas/

This article is around sharing your idea. Many of the folks I've spoken to over time are more about being protective abotu your idea than Chris is. It's a refreshing change, and in all honesty when I took one of my ideas and started to socialise it, I got a better feel, and more ideas on how to extend, and make it more reflective of what I wanted it to be. The more I talked about it, the more I worked further through my plan, and the more ideas I had.

Chris Dixon: Why Shouldn't You Keep Your Startup Idea Secret
URL: http://cdixon.org/2009/08/22/why-you-shouldnt-keep-your-startup-idea-secret/

Again, share, share, share :) It's not a secret, someone else has probably had it before, and you need help to execute it... Interesting NDA note on the bottom... keep it in mind!

BigBangTechnology: How To Poach A Startup Employee
URL: http://bigbangtechnology.com/post/how_to_poach_a_startup_employee

It's easy, just offer them something better than they currently have... what a statement, and how much impact it makes. Finding the right people may not be easy, but when you do, getting them isn't actually as hard as some folk make it out to be.

comments?

So, like many, I was confused by the new Facebook interfaces, but also like many, I have huge respect for the company as a whole, so I figured after seeing a TechCrunch article on how to enable it, that I should go ahead and give it a whirl. And, I have to say, it was pretty bad ass!

After I ran through the steps in the artcile, I got this message on the top of my profile

Boom, then I had it... what wonders awaited me I thought..

So of course the first thing I did was change my cover image. I'm pretty sure that the cover image is going to actually end up with me changing it once a week. The image selection it brought up was pretty good, made me think that my images were actually all bad quality... will need to work on that. Cat photos, here you come!

The first thing I discovered was that my birth, and family/relationship dates are all on the timeline, as well as huge events for me, things like the date I joined Facebook (October 12th makes it 5 years!!!! Longer than I've been at my day job... CRAZY!)

It was fast, intuitive, and the things I noticed was when stuff happened, more than ever before. I noticed when I became friends with people, what I did around them, etc... Pictures from trips, and a timeline that actually showed when what happened. It would've been great to have it pull geo information out of my photos though and automatically update places I've been, but I'm sure they'll get there.

The biggest thing I love is the 'view as' selection. Essentially, it lets you impersonate a user and see what your timeline looks like to them, based on your permissions for their groups, etc... It's really intuitive, and I have to say I love it.

I'm going to start a Movie BucketList... I'll post them once in a while with the movies that are coming that I want to see, trailer porn if you will :)

Johnny English: Reborn

John Carter

The Amazing Spiderman (Reboot)

Underworld Awakening

Immortals

The Dark Knight Rises (Teaser)

The Sitter

Killer Elite

The Hunger Games (Teaser)

Tucker and Dale vs. Evil!

Real Steel!

I used a clean installation of Ubuntu 11.04 to create a Chef server 0.10.x. Here're the steps I took to get it all installed and working. Let me know if you have any questions, or changes, in the comments section of the steps below.

Step 1 : Setup APT Configuration

First setup the apt soruces file for the OpsCode repo for 0.10.x and your respective build of Ubuntu (in this case 'Natty').

/etc/apt/sources.list.d/opscode.list

Then go ahead and install the OpsCode Keyring

Step 2 : Setup Chef, and Chef Server

Install chef, and chef-server through apt-get. This will install all required software, which includes the chef web-ui, and solr. There are hundreds of packages that this bad boy will install, it's ok, that's the right number of packages... Trust me though, it's worth it... no matter how dirty you're feeling.

During the installation, you'll be asked for the Chef Server URL (in the format of: http://fqdn:4000), the Chef AMQP password for Rabbit-MQ, and the Chef admin password for your Web-UI. If this installation fails the first time, see the notes below to repair the issues. Also, make sure that you re-run the installation once you've fixed the issues

Note: I found that the Jetty configuration by default is set to not start in the /etc/default/jetty file, this is pretty easy to fix by changing the NO_START line to 0, from it's default 1.

Note: I found that rabbitmq-server would not start unless the shortname was listed in /etc/hosts for the hostname in question of the server you're installing chef on. Also, make sure it's not started when trying to install chef.

Once installed, you'll see something that looks similar to this running :

Step 3 : Setup Knife

As your user (not 'root'), run the following commands.

It will lead to a menu as follows, fill er up!

Your configuration is now good to go!!!

Step 4 : You're Done

Yup, that's it, you're done!

Note: If you want to configure knife to run from another machine, it's slightly different, but almost the same. Read the Chef documentation on http://wiki.opscode.com for further information.