[comntr]

I think most of the spammers can be deterred by a simple puzzle, like 23+47. If we want to raise the bar, we make the puzzle more and more complex. Obviously, the puzzle is returned as an svg picture where the letters are "rendered" with little squares. My point is that 99% of the spammers out there are lazy and won't be able to pass this little test vs someone who's written a thoughtful comment and can definitely add 23 to 47.

[darekkay]

> Obviously, the puzzle is returned as an svg picture where the letters are "rendered" with little squares.

This will make it impossible for blind people to post a comment. There are ways around it, but one has to be aware of this issue.

[smt88]

It sounds like you may not have spent any time researching anti-spam techniques.

Your initial "puzzle" can be solved with 8 characters of JS: eval("23+47").

Your SVG picture can be solved using off-the-shelf OCR like Tesseract.js.

Even very challenging reCAPTCHA reading tests are mostly solvable by spammers.

You'd be better of using something with thousands of expert person-hours behind it, like reCAPTCHA v2.

[comntr]

That's right, I haven't.

The puzzle will be sent as an SVG, obviously.

Em.. "off-the-shelf OCR" sounds neat, but anyone who knows such words isn't an average spammer. The goal of basic SVG puzzles is to block 99% of the spammers who just type dumb comments on keyboards. The rest 1% can be taken care of by human mods.

TBH, I don't like the reCAPTCHA-like solutions. They are just annoying from my personal experience and if they rely on any 3rd party service, I'll give them a hard pass for this reason alone. My approach is to use trivial SVG-style captchas with adjustable complexity, e.g. instead of asking "23+34", we can ask "log(32)/log(2)" and effectively filter out everyone except people familiar with math, or "md5(2615), first 7 hex digits" and let in only people familiar with cryptography. Forcing users to detect birds and crosswalks will just make them upset, IMHO.

[super-serial]

It's completely dependent on the traffic of the site if a spammer takes the time to break a custom captcha.

I work on a site with 10 million monthly pageviews and spammers register on a form that has recaptcha and email verification... and we tried hidden input fields and other tricks, but each day we have consistently had 5 new spam accounts. With SVG they can just take a screenshot of what a user sees and send that to OCR. Complex math will turn away as many legitimate users as spammers.

The only real way to stop spam is to use a 3rd party API to detect it, or use something like a karma system that builds up over time. I think we're at the point where simple solutions won't work well unless you have a small site.

[comntr]

That's true when we talk about 10M monthly pageviews, but I doubt that this little extension will reach such popularity levels. If this somehow happens, by that time there will be a way to enable 3rd party captchas for any page.

[georgyo]

You said SVG twice now, but I don't see what is obvious about it.

It's an XML document that should be easier to figure out than a raster image format such as jpg or png.

[comntr]

The catch is that the text will be represented as small geometric svg shapes, so the spammer will need to first render the svg to png and then run text recognition tools. But in that svg we can easily add some css animations that make sure the entire image is never rendered, so spammers will need to run the entire browser to take screenshots and will need to assemble the image from multiple frames.

[swiley]

Won’t it have to be converted to a raster image before it can be OCRd?

Granted all you need to do is render it to a canvas but that’s an extra step on top of everything you need for a raster image, I’m not sure it’s easier.

[comntr]

And just rendering to canvas may be very tricky if the captcha is animated with css, i.e. it moves a bit and different parts of it appear at different times.

[atoav]

It can in theory be solved, but the more important question is: will it be solved?

I used a bunch of one-word-answer questions for over a decade now for sucessful spam prevention — trivial for a determined attacker with the time and resources to circumvent (and similarily trivial for me to replace with something else).

This also means for a decade I didn’t ship my user data to google.

Unless you are a really juicy target fending off the bots is enough.

[comntr]

Can you elaborate on those one word answer questions?

[vntok]

What fruit starts with A and is found in red yellow or green? Please answer in 5 letters

[ubercow13]

Please no