Print

Please login or register.
1 Hour 1 Day 1 Week 1 Month Forever
Login with username, password and session length

[methylatedspirit]

Question: what is this? And why is it full of what appears to be a jumble of error messages and Pride and Prejudice quotations?

I think what's happened here is that it's scraping all these blogs, and it's just extracting all the text, including the error messages. It's not actually static, either; the content changes if you refresh. But it's all Pride and Prejudice in between all the errors?

Then if you scroll down, you see a comments section, and you can type in comments. It's mostly spam, but I've managed to get "peepee poopoo" into the comments, so it is working. And it seems that by you linking it, all the spambots are finding this odd anomaly of a website and using it to spam the heck out of the comments. To echo your question, what is this? Is this a blog? An experiment? Some SEO scheme?

If you do a Whois on the website, it apparently belongs to some place called APNIC (Asia Pacific Network Information Centre). I don't think it's run by them; APNIC's apparently just a regional Internet address registry, so I imagine it's an address that APNIC sold/gave/assigned to the random entity who runs this thing.

[Naturegirl1999]

What does SEO stand for?
Also, Eschar, how did you come across this?

[methylatedspirit]

What does SEO stand for?

Search Engine Optimization. It's a collection of techniques used to get websites higher on search engine results. In this case, I think the technique they're using is "keyword stuffing", just trying to stuff every keyword into a webpage so that the search engines see it. It doesn't really work anymore, since search engines are significantly smarter now.

But then again, unless if there's an actual web domain associated with it, I can't tell why you'd do this, considering that search engines usually prefer websites with actual domains (like example.com), not just an IP address. Is this just someone's web design and scripting experiment gone public?

[methylatedspirit]

Looking at the HTML source, I think something's gone utterly awry on that page. Here's an excerpt:

Code: [Select]

<p> "I have received a letter this morning that has astonished me
<a href="/loginflat/enc/man.sh">Shadow Security Scanner performed a vulnerability assessment</a>
 "No; it would have been strange if they had; but I make no doubt they
 <a href="/loginsuper/webmail./editor/artikelinfo.php">your password is</a>
 Elizabeth's mind was now relieved from a very heavy weight; and, after
 <a href="/loginerror/components/com_forum/vsadmin/mwchat/libs/info.php">HTTP_FROM=googlebot</a>
 be greater than at home. He heard her attentively, and then said:
 <a href="/login1/cgi-bin/1220/gallery.php">AutoCreate=TRUE password=*</a>
 There is no danger of Wickham's marrying Mary King. There's for you! She 

So it's Pride and Prejudice (at least passages of such), but then there's this garbage in between, which I think is server-related log files and HTML from somewhere. It lends credence to the idea that it's scraping other web pages, since I have no reason why it has any of these things in it. The more I look at this, the less sense it makes.

[Eschar]

I don't actually remember how I found this. Through Shodan.io, I think.

If it was just this, I'd wonder if it were some project to attract spammy comments like a honeypot of some sort, by copying the same blogs those spammers look for.
Problem is, this isn't the only one, I stumbled across a bunch more today though I don't remember the search terms I used (I'll dig them out of my search history tomorrow.) If it's an "anomaly", there's a lot of them. All of them have login widgets at the top.

When I first found it I don't think there were many or any comments. However, there were numerous spammy comments on it even before I linked it.

Also, the comments don't seem to be permanent. I couldn't find methyl's, for example.

As methyl notes it doesn't seem to make a lot of sense.

[methylatedspirit]

Shodan says it's owned by "Taiwan Star Telecom Corporation Limited" on both the Organization and ISP fields. But if you look it up, TST Corp. Ltd. is a mobile network provider (Wikipedia's outdated; it apparently offers 5G on its own website). What business is it doing hosting this nonsense on its own servers? I certainly never heard of any mobile providers letting you set up your own website.

My guess is the same as Eschar's: Honeypot.

[Eschar]

Did a regular google search for the site's IP and found it mentioned in a file in a GitHub repository, "thornhill-corp/populate".

[Eric Blank]

I remember that too. I wonder how many cameras you can still access

[Eschar]

I remember that too. I wonder how many cameras you can still access

Numerous. Shodan.io and insecam.org are quite useful for finding them.

[KittyTac]

What makes a MUD be one? I play Marosia which is kinda like a MUD in many ways but yet it's not considered one usually.

[methylatedspirit]

How do you mix colors together? Rather, given some representation of two colors, what operation do you perform on the components to model the mixing of colors? Looking this question up gives me some color theory stuff along with this basic text program that outputs a secondary color given two primary colors. I want a general solution, not specific cases.

What would be the easiest color space to do this in? I'd prefer just being able to deal with all the components separately, without doing some operations that cross the components over.

Spoiler: Some background on the question (click to show/hide)

I generalized that blue-pink gradient maker to allow for 4 colors (starting x-axis, ending x-axis, starting y-axis, ending y-axis). It does this by merely adding the RGB components together. A thing I noticed is that if you didn't choose some very distant colors (in RGB terms), you'd end up with mostly white.

This one is (#FF0000, #00FF00, #0000FF, #FFFF00) (red, green, blue, yellow), internally halved to avoid saturation of RGB values. I think I may have rotated this thing wrong, but my point should still stand.


I'd say that's pretty good.

Now look at this, (#00ffff, #00ff21, #ff0000, #ff00dc) (cyan, light green, red, magenta):

Swing and a miss.

Clearly, there has to be something wrong with how I'm adding these colors together. I'm just naively adding the RGB values together, and hoping for the best, so it's extremely possible that that's how it's all going wrong.

I tried redoing everything in CMYK... and I think something's gone horribly wrong. Here's (#FF0000, #00FF00, #0000FF, #FFFF00) (red, green, blue, yellow) in that thing:


Fixing this code is a story for the Coding thread, but I think it's clear that something's very wrong. So now I'm thinking if my color addition logic is entirely wrong in the first place.

[scriver]

Isn't there this whole opposite land thingy in colour theory where because (iirc) what we see is the lack of colours rather than the presence of the colours we perceive you have to mix two other colours than the colours you want to mix to get the colours you want to get

[WealthyRadish]

A place to start is bilinear interpolation.

https://en.wikipedia.org/wiki/Bilinear_interpolation

[methylatedspirit]

Isn't there this whole opposite land thingy in colour theory where because (iirc) what we see is the lack of colours rather than the presence of the colours we perceive you have to mix two other colours than the colours you want to mix to get the colours you want to get

(Brain slightly drunk due to medication)
Depends on which color space you're using. RGB (Red, Green, Blue) starts with black, and adds R, G and B components to make colors. Full R, G and B equals white. CMYK (Cyan, Magenta, Yellow, Key (i.e. Black)) starts with white, and adds C, M,Y, and K components to produce colors. Full C, M, Y and K equals black. Technically, you don't even need K, but it's hard for printers to generate an ideal black using just CMY.

The terminology is that RGB is an additive color space, while CMYK is a subtractive color space. This is why you get RGB in screens, and CMYK in printers; it's reflective of the "default" color of the medium. Color mixing is different between these two because RGB adds colors, while the other subtracts colors.

Our vision is subtractive. Sunlight starts off as white, and objects reflect some wavelengths, while absorbing others. We see the reflected wavelengths as colors. This is pretty much the definition of subtractive; color is being removed to generate other colors. Doesn't stop us from working with additive color spaces, since (assuming everything is ideal) an additive color space and a subtractive color space should be able to represent the same color, but in different ways that ultimately do not matter to the end user. They should be equivalent.

Then again, there's the whole thing about color gamut, and how sRGB (i.e. the standard color space for pretty much everything you see on a computer unless you go out of your way to use something else) has a narrower gamut than CMYK, hence fewer colors can be faithfully represented in sRGB, but that's a story for someone else to tell.

[methylatedspirit]

A place to start is bilinear interpolation.

How, precisely, do you do bilinear interpolation between two colors? That needs at least 5 points to work with. They're two points in 3D space, so all I can think of is averaging the sum of each component individually. The mixed color is the midpoint of those two points, essentially. Filtering would smooth things out, but I don't think it would fix the fundamental issue of the fact that my color mixing method is likely faulty.

Surely there's some averaging function, that when given two points (representing colors), it produces an output such that it models the mixing of colors.