Disclaimer, I’m not peddling new software but thinking over exactly what I want from DBs
I’m impressed with the marketing acumen of technology like CouchDB with Relax, or RESTful design principles. There’s a science and art to naming which immediately imparts a message. Google is BIG, Microsoft makes software for microcomputers, and
Apple Steve Jobs is all about sharp design.
Where’s LazyMiracleDB, is it you?
I want a persistence layer that I never have to think about. You can store my data in an ocean cooled center in Finland for all I care. The perfect state machine is a black box I throw data in and forget about until I, or someone else needs it. It’s always there, and it’s Freakin’ Fast, and if no one needs it, it dies (archived) after a grace period. It’s language agnostic. I don’t want to have to know^ anything about how it works at the application level, just tell me how to throw stuff in and get it in one line of code or a single click. And if my personal LazyMiracleDB needs more resources or another server it let’s me know with a polite text or email: “resources over 80% @ URL.node, please confirm dialing up another server (Y/N), I could use a coffee break, shall I order you one as well?”^^.
I know what some of you are thinking, use Couch, Redis, Mongo, FluidDB or some other smooth DB*. While these are impressive technologies I still have to worry about a bunch of other stuff, so here’s my worries and wish list combined. My DB pain points:
- How will my persistence layer interact with other services and databases. You’re never alone on a network
- security of information
- Reckless synchronization: I want to be able to voice to text edit source on my phone, and have it live on thousands of public facing servers instantly. Eat that QA tests!
- Ubiquitous. Availability across clusters and clients
- phenomenal visualization and editing of what’s inside. It has to work on any platform, think Dropbox (local files on my phone) combined with Chromium’s search bar & universal doc handler but specialized for network apps. Instant search for my DB
- and worst of all installation across platforms. Groovy new tech is notorious for custom builds and tricky porting, something I never want to think about before 5am again**
^= I really do want to know how lazymiracleDB works, the curse of insatiable curiosity
^^= if you don’t imbibe caffeine, you’re not allowed to use LazyMiracleDB. Exceptions to blogger/geniuses I follow on twitter, or anyone who buys me a case of Redline Power Rush. Yeah, I wrote that page. If me, Seth Godin, and the 5% charity all become filthy rich and the servers disintegrate under the referral traffic coming from this post, I’ll do the Macarena on a YouTube video proudly displayed on Google’s homepage (assuming Larry, Sergey and the old guy are cool with it).
*= found a great tutorial on Redis this AM. Instapapered it to my phone with slides and all, go Simon, Antirez and the big Redis community, there’s a crapton of clients for your favorite language.
**= I am blessed with Ubuntu as my primary dev machine. All of you, whoever you are that make packages and dev libs (apt-get, debian, yum…) or source and install scripts that just work, Thank You! It’s really been a pleasure working on Ubuntu over the past 4-5 months after years of windows only development.