1
edit
Changes
→My views and Conclusions
==Summary of Talk==
This talk was about MongoDB, a NoSQL data store. The way this topic was presented was from a more technical side of things, more about educating us on what this is exactly. He didn't talk much about the philosophy or what MongoDB does for programmers, but what it is and how it functions. He first went into some detail about what NoSQL databases are. And what they essentially are, are database that don't use a structured query language. They also don't require a database structure, or table structures.
He then talked about how MongoDB came to be. It was originally part of a larger initiative, something specifically made for a company. But then the realized that what they had made had serious potential, and more to offer. because of this, they shifted focus to it. It wasn't originally an open source project, but in 2009, version 1.0 was released as open source under the AGPL licence.
He then went into a more technical talk about what MongoDB is, and what it does. Essentially, MongoDB is a high performance document database that uses JSON documents to store things. It is written in C++ and uses memory mapped files to gain speed. The way it stores related data is with other JSON documents and arrays embedded into a JSON document. The MongoDB project is very community open, and anyone is able to submit code revisions. He talks about how the database is extremely reliable, through the use of 2 techniques called sharding and making a replica set. Replica sets are other machines with the database on them, so if your main machine dies, the other one will automatically take its place. Sharding is a way of spreading your data across many different servers. It handles all the distribution and retrieving automatically.
He then does a demo to show the power, flexibility, and easiness of MongoDB. You don't have to do a lot of heavy lifting to create databases, because all you have to do is send data to it, and MongoDB handles all of the creation. It almost eliminates the need for a DB admin that is needed for SQL databases. The mongo shell lets you run JavaScript, and this enables people to pick it up rather quickly because they have already been coding this way. Because of MongoDb, there is now a term called the MEAN stack (a new type of LAMP stack), which is MongoDB, Express, AngularJS, and NodeJS. This allows you to run your application fully in JavaScript from top to bottom.
People have started to pick up MongoDb and use it as their main data store. However, there are some things that MongoDB can't do, such as transactions, for e-commerce things. He never talked about whether this would be possible to do in the future, or if there was any mention of it. He stressed about how reliable MongoDB is. During their first year of business, they had 100% uptime using MongoDB. He believes that it is the future, as it is currently the number 6 type of database being used.
==Speakers View on Open Source==
The speaker didn't directly say anything about if he liked the open source model, or anything about what he liked most about it, but I was able to piece together what it sounded like he said. He said he runs the MongoDB user group in Toronto, which right now has about 400 members. They have talks and meet-ups, and it sounds like it is quite a nice community. Based on this, it sounds like he really enjoys the community aspect of open source. Throughout the presentation, he kept talking about wonderful things the MongoDB community had to offer. He really wants the community to grow, and to spread the knowledge about MongoDB. However, it didn't sound like he was to enthralled about how the open source community doesn't document things.
=Comparison of Speakers Views on Open Source=
From what I could tell, Adrien (the OpenCL speaker) is all about open source, and Kevin (the MongoDb speaker) seems to really enjoy the community of open source. Adrien talked a good bit about how software patents are a terrible thing, and that programmers won't go much further if we are always competing against each other. Kevin didn't really go into much detail about open source. From what I could gather, it didn't seem like he was for it or against it.
=My views and Conclusions=
For me, I wasn't really much interested in open source going into FSOSS, and I still am on the same boat. I don't really have anything against it, I just don't like the way the model works. I believe that to have a high quality product, the best way to ensure that is to control as many of the variables as possible. In open source, while you don't have to accept peoples code revisions, you are relying on random people to fix certain things, and even if it works, the quality may not be up to the level you want. So all in all, I think open source is a fine thing, it's just not for me.
