tylerhobbs

Great comment! Yes, there are many other approaches that are more true to reality, to the point where they are fairly thorough simulations of physics. My approach is somewhat like a five minute figure drawing sketch compared to the equivalent of hyper-realistic 100 hour paintings. Neither is necessarily a better approach, but both are good for different goals. What I like about my approach is that it is very easy for me to understand and manipulate. I can take advantage of this to make new effects that are less like reality but still intriguing in their own way.

Author here. Thank you! Yes, I think you have the basic gist of it. Split each line, then move an endpoint. There's nothing too complicated about the algorithm, the trick is really around fine tuning all of the possible parameters. If it's not obvious, my approach is purely going after aesthetic effects.

If you read the quoted portion of his comment, you'll see he's talking about the officers being sentenced, not the offense happening in the first place.

I'm assuming both you and your spouse are earning money, in my numbers. Additionally, you would presumably make investments with the money that you're saving, so you don't quite have to save the whole $1.5m straight up. I'm also not sure why you used a 2% withdrawal rate -- typically 4 to 5% is regarded as a conservative rate, which would be $60k to $75k per year.

I'm not suggesting that doing this is a walk in the park, you definitely have to prioritize around it (which may include living in a city that isn't insanely expensive), but it's doable.

3% of $1.5m is only $45k. If you and your spouse can live off of $45k while you're still earning, you can save ~$100k per year and accumulate $1.5m by the time you're 40, easily. See Mr. Money Moustache for details on how this isn't as unreasonable to do as you would think: http://www.mrmoneymustache.com/

> And apartment blocks had shops on the ground floor.

This is one of my favorite features of the Mueller area: the apartments do have shops on the ground floor. I actually live in a live-work unit in one of the apartments. My wife and I live upstairs, and she runs a hair salon on the ground floor below our unit. The other shops nearby include a small convenience/grocery store, a chiropractor, a shared-work space, restaurants, a small gym, etc.

The shops on the ground floor really improve the quality of the area, so I'm glad that Austin is finally pushing for more of this.

Thank you! (I'm the author of the post.) I am indeed writing my own code; I'm just using simple functions from Processing.

Thank you very much for the video link. It occurred to me while working on this that video game artists and animators probably do a ton of this, and have much more sophisticated models. Indeed, the result from that video is really, really good. I'm coming from a painting and abstract background. This was my first real exercise in studying natural patterns and textures through algorithmic artwork. I mostly intend to apply it to abstract artwork, but the more realistic approach shown in the video is a different, interesting angle on the same subject.

Hi, I'm the author of the post. I don't normally release my code, but in this case I did post a gist of the main function to Twitter: https://gist.github.com/thobbs/e879b7dad8f3733d58d1

I'm using Quil, which is a Clojure wrapper for Processing. The source is quite dense, so it's hard to understand without following the post. It's mostly about tweaking probability distributions along gradients.

This is true. Most companies either start using Cassandra when their existing database infrastructure is struggling with a particular workload or they're planning a new feature that would be prohibitively difficult or expensive with an RDBMS.

The one exception to this might be a need for higher availability, especially if expanding to multiple datacenters is part of the plan.

Cassandra isn't designed to replace relational databases for all use cases. There are certain use cases where Cassandra excels, certain use cases where an RDBMS excels, and plenty of ground in between where you're making tradeoffs.

CQL was designed to resemble SQL in order to present an API that is reasonably familiar for most developers. It's quite limited when compared to SQL because it focuses on the areas where Cassandra excels.

(Disclosure: I'm a Cassandra committer.)

"Cassandra [...] isn't real-time in the sense that you can do immediate, fast queries about the relationships across those records as they are hitting the system."

That depends on how you are using Cassandra. Typically, you are expected to know your query patterns up front, and so you will lay your data out accordingly when ingesting. When done properly, this allows for ~1ms queries that return completely up-to-date results.

Cassandra uses a technique known as consistent hashing to allow each node to independently determine what nodes are replicas for a given row. The process really just involves hashing the row key and then comparing the result to the token of each node in the cluster. If the hash falls in between a node's token and the token of the previous node in the ring, then that node is a replica for that row. There's not really any way that nodes can disagree on this.

Given that nodes cannot disagree on the set of replicas for a row, quorums must (by definition, and the pigeon hole principal) overlap by at least one replica.

You would need to define "consistency" in order to have a more reasonable discussion about what each system provides, but Cassandra certainly isn't only eventually consistent. You can choose the number of replicas that must respond in order to consider the read/write operation a success per operation, which allows you to have quorum-based strong consistency guarantees.

There are more details on the options here: http://www.datastax.com/docs/1.2/dml/data_consistency

Some of these are a bit dated, but it's still a reasonably accurate list of companies that have publicly disclosed their usage of Cassandra: http://www.datastax.com/cassandrausers.

Submitter here. For those who are short on time, these are a few of my favorites:

1. For those unfamiliar with Cassandra, Patrick McFadin's talk (Building a Cassandra Based Application from Scratch) can give you a good feel for how Cassandra tends to be used

2. Greg Ulrich's talk (Servers Fail, Who Cares?) covers some great ways that Netflix uses Cassandra for high availability

3. Scaling MongoDB with Cassandra is a fun lightning talk

A nice alternative is to do:

  " Allow easy navigation to relative files
  cmap %/ %:p:h/

This basically replaces everything before the slash with the directory of the file you're editing. That way, you don't have to necessarily cd, but can easily open files near the one you're currently editing. You can just do ":e %/foo.txt" to open foo.txt in the same directory.

We have been using Clojure for a product for about 9 months now, and in my experience, I have no trouble identifying what different blocks of code do. It may depend on proper syntax highlighting, but I think one of the keys is that there is so little boilerplate in Clojure that all of the code is important. You're not skimming over getters and setters, assignment of constructor parameters to the object, or some of the other things your eyes tend to automatically skip when reading, say, Java or Python. It tends to be pure logic or pure data, with very little else in the way.

S3 isn't where the SSDs will go, it's the EC2 instances. Many, many, people would pay those numbers for local, dedicated SSDs when their instance is running a database.

There's a very good Google Tech Talk by Geoff Hinton (who has worked closely with Dahl on a lot of this research and developed some of the key algorithms in this field) that explains how to build deep belief networks using layers of RBMs: http://www.youtube.com/watch?v=AyzOUbkUf3M

That video focuses on handwritten digit recognition, but it's great for understanding the basics. There's a second Google Tech Talk video from a few years later that talks directly about phoneme recognition as well: http://www.youtube.com/watch?v=VdIURAu1-aU

That is an excellent point, and well expressed. I do see how anecdotes and stories can be used in a very positive way to shift opinion; there are definitely some grey regions. (How do you know you're on the right sight?)

Great comment. I am most certainly not claiming that you shouldn't question research findings on other grounds, such as sampling bias or statistical significance, as you have done here; those types of points are extremely valuable.

K-means is only grouping all of the pixels in to K disjoint clusters, not actually altering the value of the pixels. Taking an entire cluster of pixels and shifting their values in the RGB space could produce some very interesting images. For example, you could easily make all of the almost-white pixels slightly blue. There would be no loss of variety among the pixels, they would just all have their blue values bumped up.

I don't know what you qualify as 'hip', but beyond the initial three or four first people in a startup, many startups are more than willing to start hiring remote workers.

This is especially true when you're looking for workers with very specific skill sets. It's generally easy to find a generic Java developer locally, but if you're looking for, say Cassandra experts (like we are at DataStax, with an office in Austin), then you're probably willing to hire someone anywhere in the world if they're good enough.

"Cassandra requires all nodes to be restarted to apply a schema change, which can be quite a big deal."

That's no longer true. In 0.7, keyspaces and column families may be created, altered, or dropped live.