Fwd: Exciting Job Opportunity to work on webRTC atEricsson

From: "Dr.D " 
I got past that one ... Now I had to Google.. Riak  to find out it is a
NoSql database..  
That is something I have yet to try to understand.  a "No" Server Query
Language Database..   
If you don't have a Server Query Language to talk to it; then do you just
Think the data into place..  ?


=============================================================== From: Stephen Haywood ------------------------------------------------------ Alot of the noSQL stuff uses key:value pairs to store data so there is no such thing as "SELECT * FROM table," because there are no tables. -- Stephen Haywood Information Security Consultant CISSP, GSEC, OSCP T: @averagesecguy W: averagesecurityguy.info

=============================================================== From: Dan Lyke ------------------------------------------------------ Well for a while MongoDB's big selling point was its write speed, which basically came from the fact that it didn't guarantee writes, it just queued stuff up in memory and hopefully got around to writing it at some point. You can see where I'm going here: Why bother with data in the first place? Managing it just slows things down... (IMHO: "NoSQL" databases are like the people who tell you about their novel new database structure in which everything fits in to a single table that has three columns... And it's so flexible, because you can add fields easily, and everything's a string, and...) Dan

=============================================================== From: Aaron welch ------------------------------------------------------ There is a place for both. SQL excels at previously structured data, where NOSQL excels when you know your query but not the structure of your data. This allows you to keep adding new data to the structure without refactoring the application. There is also a difference in wether your data is read or write specific and at what level. If you do a massive amount of reads, building a big mySQL box gets alot more expensive than just scaling out another node or partitioning the data onto a new machine. -AW

=============================================================== From: Eric Wolf ------------------------------------------------------ Coming from a disaster response perspective... NoSQL excels in an environment when you cannot predict how people will use your system (updates or queries) and you want to invoke Moore's Law to save lives. Let's just fire up a larger EC2 instance... Ironically, the ability to handle massive update rates is more important than ensuring write completion. Disaster responders will double-check their work if they can get any response from the database. No response is the only fail. "Big Data" is another area where NoSQL excels. You don't know the exact nature of your data or how to extract meaning until you have some critical amount of data. -Eric -=--=---=----=----=---=--=-=--=---=----=---=--=-=- Eric B. Wolf 720-334-7734

=============================================================== From: Sean Brewer ------------------------------------------------------ Different NoSQL databases have their own methods for querying, creating records, etc. With say, couchdb. You don't query it with SQL. You usually grab data with a HTTP request to route that's a pre-defined view, which are map-reduce queries written in javascript. Sometimes other languages, but the default is JS. You don't *have *to have a view, but it's faster that way. Creation of data is simply a POST request with your record (for CouchDB, it's JSON), to your database.