mschoebel

Tab-delimited CSV with the raw-data:

https://geekregator.com/files/HackerNews-RawData2017.zip

The timestamps are in UTC+2. I keep the clock in the Geiger counter on daylight savings time. Too lazy to change it. :)

I have the GammaScout Alert: https://www.gamma-scout.com/EN/Home.php

It stores the data in its internal memory which I then transfer via its USB port to my computer once a week.

I'm in Germany and I have a Geiger counter running 24/7. I just looked at the data for January and February and the only thing that I notice is a VERY slightly higher reading on February 4th with 0.1727 microSievert/hour. Average for January was 0.1674, lowest was 0.1631, highest was 0.1703. So February 4th was less than 6% higher than the lowest value from January.

The difference was so small that I had just attributed it to normal fluctuations when I first saw it. Whatever caused this, so far it looks like it was a very small event.

I could probably provide a CSV-file with the raw data if anyone is interested. My Geiger counter stores a value every 5 minutes.

Have you looked at https://uimovement.com/tag/search/ and https://uimovement.com/tag/filter/

Still unacceptable as it would still cause my users to expose their IP address to someone else's server.

Let's see if I got that right: Specification says I must load the AMP script from cdn.ampproject.org?

If yes, then there is no effing way that I will ever use this. I will NOT use something that forces me to load scripts from a host that I have no control over. Does nobody see what a HUGE security risk that is???

Umm... FAQ says:

"What licence does Ultibo use? The GNU LGPL version 2.1, the same licence used by many other popular open source products."

FWIW... I just logged in to my Yahoo Account and removed the security questions. Just to be sure. I had already changed my password a few months ago when first rumors of this came up. I'm pretty sure that the option to remove the security questions wasn't there back then.

I probably should also add that I'm using FreePascal - not Delphi - and that the servers are a mix of Windows and Linux.

See: https://deusu.org

The backend is written in Pascal.

With the exception of the blog. That is written in Node.js. I did that as a way to learn Node.js. But as soon as I find some time I'll rewrite that in Pascal too.

Pascal

Thanks for the answer.

I suggest not to use AWS if you know that you'll need a server 24/7. Old-school hosters which offer dedicated servers are much cheaper for that use-case.

There are several offers here in Europe where you can get an i7-6700, 64gb RAM and 1tb SSD for less than €60/month. AWS would cost you at least 3-4x as much. You'll lose the flexibility of AWS, but save a ton of cash.

Do you have a rough estimate of how many servers you will need for Elasticsearch for the 1.7bn URLs of the latest CommonCrawl?

Have you ever in real life hear someone say "If 80kg is less than your weight, then you are overweight."

Probably not. We say: "If your weight is more than 80kg then you are overweight."

When talking we are used to first mention "x", and after that the numerical value we compare it to. No matter if it's less than or larger than what we want to compare it to.

If you swap these around, then whoever listens to you - or reads your code - will take a few seconds longer to understand what you mean.

That webpage has a last-modified date from 2009...

Never call yourself a hacker. To non-tech people "hacker" has the meaning of "illegally hacks into computers".

All the functions in that code return numbers.

> Or use strict mode would not allow that.

It was in strict mode. It allows that.

Great. So I could have just left that 100-line function as it was and been spared that 30-minute bug-hunt for the missing "+".

Of course I wouldn't have written that blog-article then, and wouldn't have gotten onto the HN homepage... Oh, well... :)

Thankfully the bug DID make my tests go red.

Some things just won't work without JavaScript. Or would be way, WAY too hard to implement without JS.

But a website should at least appear without JavaScript. I've seen lots of websites that look empty or horribly distorted when JavaScript is disabled. That should not happen.

I just posted a follow-up blogpost, comparing Node 0.11.15 and io.js 1.0.3 which were both released yesterday.

In that post I also benchmarked the various fixes for the typed-array slowdown you mentioned. BTW --nouse-osr makes all three tests run faster.

http://geekregator.com/2015-01-21-node_js_0_11_15_and_io_js_...

Ok, I hadn't tested with both before. Keeping the array alive and using --nouse-osr makes io.js only 2.3% slower than my original measurement for Node 0.10.35. Median of 5058ms.

And Node 0.10.35 shows basically the same results as before. I see less than 1% difference. Maybe just random fluctuation. Even if not. 1% is irrelevant.

Very interesting. After reading your comment, I tried allocating another Uint8Array and keeping it allocated throughout the entire test as a workaround for the issue you mentioned. Time for Node.js was unchanged, but io.js was down to about 5.5s now. Almost the same time as Node. Only about 10% slower.

The same happens when I use the --nouse-osr parameter that you mentioned.

I thought about that. But it would have diminished the point I was trying to make: Always test with different versions as performance may differ by a LOT.