sigmaml

> OOP is too verbose by it's essence compared to closures.

Could you comment on Smalltalk's blocks?

I heard a variant thereof. ``Research is what I'm doing when I don't know what I'm doing. Science is what I'm doing once I figure that out!"

My experience is rather different.

I travel to rural areas in the South Indian state of Andhra Pradesh every month. The connection quality is as poor as its availability is flaky. I find that GMail, Google Drive and the Google search page always load. Business correspondence and collaborative editing have not been problems at all. I have not tried Google+, though. I shall try it the next time.

On the other hand, my biggest problems have been with GitHub. Not once in the last 18 months have the GitHub Web pages loaded. That makes reviewing and merging pull requests very painful. I reported this to GitHub long ago, but have not seen any improvement yet.

Two points.

1. Google fought then-giants such as AltaVista, Excite and Yahoo in the process of establishing itself as the leader in search space. The incumbents had economies of scale and (relatively) massive amounts of data to crunch through. Yet a Google happened!

2. ``Google's very existence makes it harder for similar services to exist." History has proved time and again that such a premise is false. When it does become false, it is usually because of `dissimilar' disruptions in the landscape.

I read HN as much for news links as for the ensuing intelligent discussion and pointers. Should you take them away, to me, the appeal of HN drops significantly.

It depends a lot on the service that you provide. I provide compute services. Most of my servers are CPU-bound most of the time.

It is very inconvenient on a large page like this, should you need to quickly move up/down to read a related comment. If you know it, dragging the scrollbar, on the other hand, is quick to land you in the approximate target region.

It is not clear what you refer to when you say ``buy".

By the way, you have drifted away from discussing the technological aspects of the Web. This line of argument doesn't help establish why HTML/JavaScript/CSS are better.

That is ELISA (with an `S'), if I understand what you mean correctly.

I worked on a similar project; the US consultant insisted that each method should be less than 25 lines. The code involved several decision trees. This rigid limit made the code quite inconvenient to follow and enhance. After much sweat and swearing on both sides, the limit was increased to 40 lines!

I came across AGG very recently. Yes, several parts of the code are designed and implemented very well. Unfortunate that Maxim died.

Author of the blog post here. Thank you all for your insightful comments and critiques. A few points.

-> Several people have suggested that JVM's mature GC may be using other cores, resulting in the > 370% CPU utilisation. I admit that I do not understand the internals of OpenJDK's GC. However, I would like to point out that 100% of the roughly 1,200,000,000 statistics results instances created during one run are not released until the program terminates. They are all stored in statistics results containers until the very end of the run. However, see below.

- From a processing perspective, other than the manipulation of containers and bit sets, almost all of the remaining processing is numeric. It is full of integers and doubles getting added, subtracted, multiplied, divided, square roots calculated, etc., as part of various statistical computations.

- Where I needed sets was in preparing the several million working sets over which processing iterations occur. Working sets are prepared based on user's query criteria, several thresholds for filtering elements under a variety of constraints, etc. From a garbage perspective, these working sets are the biggest garbage generated during a run.

- Go's numeric performance is certainly very good. I have had good experiences with it in other projects.

- It is true that I could not identify the causes to which the performance differences between the Go version and the Java one could be attributed. Was I being naive? Am I incompetent? Possible. As I concluded, this behaviour may be obvious to more informed people, but it surprised _me_, and, that is what I admitted.

Author of the blog post here. Some of the comments here suggest that I was probably new to Go. Not quite. I was reasonably active in the Go community (and on golang-nuts) in the first two years of Go's public life. Also, when discussions of generics were taking place informally in the mailing list, I was one of the first to have actually written at least a half-serious `proposal' for generics (you can read it at http://oneofmanyworlds.blogspot.com/2011/11/draft-2-of-propo...).

As for achieving `generic' code through static generation using an external tool, I had had considerable success with it in some earlier projects. Except that the number of types is much higher in this project, and things could not be as simply - or clearly - generated statically.

I cannot publish the code because of the client's IP involved -- not because I fear criticism; see, I was ready to face _this_ criticism :-). Those of you who read my original post in golang-nuts may have already read that, of course!

> Rather than storing data structures directly in the hashmap, I would store them in an array and map the keys to array indexes.

Author of the blog post here. That is one of the reasons I shifted to using bit sets: specific element types can be elided.

> just don't force your paying client to test it out with you or you might be one less client shortly.

Author of the blog post here. I agree, in retrospect!

Requiring the user to touch the trackpad (with two fingers, at that) just to know the relative position in the page is an inconvenience, at least to some.

Also, isn't the scroll bar really useful when you need to scroll up and down the page quickly by an arbitrary amount?