5 ms·

> effectively, the service CF is offering to them is not hosting but cost reduction and insulation from the consequences of their speech. This is somewhat tong

> effectively, the service CF is offering to them is not hosting but cost reduction and insulation from the consequences of their speech. This is somewhat tongue-in-cheek on my part, to be clear, but this raises an interesting point about whether orchestrating a DDoS attack is a form of free speech. I'm inclined to say "yes" more than "no." (You could draw a parallel, for example, to counter-protestors who try to drown out Westboro Baptst Church picketers by holding up their own signs....)

andrewflnr 3y ago
You could try to draw a parallel, but it would fail. The most technically rigorous difference is that a protest and counter-protest are aimed at the same audience, whereas a DDoS is aimed at the first "speaker". Less rigorously but more practically, a "message" intended to cause technical malfunction is not "speech" and we all know it. It's even less "speech" than an extortion threat, which is already not protected. The "parallel" would be feeding the Westboro Baptists cognitohazards and crying "free speech" when they go insane and you get prosecuted.
- Dylan16807 3y ago
  It's much closer to trying to be so loud they're drowned out than that weird analogy. I'd say even "malfunction" is a stretch. DDoS isn't hacking.
  - andrewflnr 3y ago
    Hacking isn't the only way to cause a malfunction; a hammer tends to do it. You can't deny that one of the functions of a server is to reply to good-faith requests, and that a DDoS intentionally degrades that function. What is that if not causing malfunction? You're looking for nuance where there is none. (And yes, it's a weird analogy. It was the closest I could come up with on short notice, and I think that's illustrative of the dangers of analogic reasoning. FWIW yours fails on the same audience/targeting aspect as the first one.)
    - spondylosaurus 3y ago
      One of the functions of a mailbox is to receive/reply to good-faith messages, but 10,000 people can get together and decide to send postcards that say "YOU SUCK." to a single recipient. That person's mailman might not be pleased, but all the mail is conveyed through legitimate channels and legitimate means...
      - andrewflnr 3y ago
        That ellipsis is bearing a lot of weight. What do you think any of that proves?
    - Dylan16807 3y ago
      > FWIW yours fails on the same audience/targeting aspect as the first one. Drowning someone out by being much louder and drowning someone out by packet flood have the same audience: other people that would receive the message. > What is that if not causing malfunction? We can use a narrow definition of malfunction, or a wide definition. If we use a narrow definition, then making the server malfunction is always an attack on the speaker. But a DDoS does not cause a malfunction under a narrow definition. If we use a wide definition that includes DDoS, then only some ways of causing a malfunction are an attack on the speaker. Yelling at someone is perfectly good counter-protesting, even though it degrades their ability to make a speech. "causing malfunction -> attacking the speaker" and "DDoS -> causing malfunction" are not both valid at the same time for the same definition of malfunction.
      - andrewflnr 3y ago
        At this point I'm not convinced that you know either what point you're actually arguing or how a DDoS works.
        Dylan16807 3y ago
        You're too focused on specific technicalities that you picked. If you had a server on a shared medium or a half-duplex connection then DDoS would be exactly the same as shouting over someone. The shouting is aimed at the audience and gets in the way of the server's signal. Putting the server on a full-duplex connection changes the technical details but it doesn't have any meaningful effect on what happens. The purpose is not to break anything, the purpose is to cause so much extra signal that the server can't get its own signal out. Just like a counter-protest. The thing being intentionally degraded is the connection. Not the "speaker". Or if it's a higher level DDoS then it's usually the DDoS clients filling the queue. Which is also a pretty normal counter-protest thing, getting in the way and taking up all the close audience space.
Dylan16807 3y ago
If you get a thousand people together to perform a DDoS, I can see that. But usually a DDoS is done by one person. That's not analogous to counter-protesting.

andrewflnr 3y ago

You could try to draw a parallel, but it would fail. The most technically rigorous difference is that a protest and counter-protest are aimed at the same audience, whereas a DDoS is aimed at the first "speaker". Less rigorously but more practically, a "message" intended to cause technical malfunction is not "speech" and we all know it. It's even less "speech" than an extortion threat, which is already not protected. The "parallel" would be feeding the Westboro Baptists cognitohazards and crying "free speech" when they go insane and you get prosecuted.

Dylan16807 3y ago
It's much closer to trying to be so loud they're drowned out than that weird analogy. I'd say even "malfunction" is a stretch. DDoS isn't hacking.
- andrewflnr 3y ago
  Hacking isn't the only way to cause a malfunction; a hammer tends to do it. You can't deny that one of the functions of a server is to reply to good-faith requests, and that a DDoS intentionally degrades that function. What is that if not causing malfunction? You're looking for nuance where there is none. (And yes, it's a weird analogy. It was the closest I could come up with on short notice, and I think that's illustrative of the dangers of analogic reasoning. FWIW yours fails on the same audience/targeting aspect as the first one.)
  - spondylosaurus 3y ago
    One of the functions of a mailbox is to receive/reply to good-faith messages, but 10,000 people can get together and decide to send postcards that say "YOU SUCK." to a single recipient. That person's mailman might not be pleased, but all the mail is conveyed through legitimate channels and legitimate means...
    - andrewflnr 3y ago
      That ellipsis is bearing a lot of weight. What do you think any of that proves?
  - Dylan16807 3y ago
    > FWIW yours fails on the same audience/targeting aspect as the first one. Drowning someone out by being much louder and drowning someone out by packet flood have the same audience: other people that would receive the message. > What is that if not causing malfunction? We can use a narrow definition of malfunction, or a wide definition. If we use a narrow definition, then making the server malfunction is always an attack on the speaker. But a DDoS does not cause a malfunction under a narrow definition. If we use a wide definition that includes DDoS, then only some ways of causing a malfunction are an attack on the speaker. Yelling at someone is perfectly good counter-protesting, even though it degrades their ability to make a speech. "causing malfunction -> attacking the speaker" and "DDoS -> causing malfunction" are not both valid at the same time for the same definition of malfunction.
    - andrewflnr 3y ago
      At this point I'm not convinced that you know either what point you're actually arguing or how a DDoS works.
      - Dylan16807 3y ago
        You're too focused on specific technicalities that you picked. If you had a server on a shared medium or a half-duplex connection then DDoS would be exactly the same as shouting over someone. The shouting is aimed at the audience and gets in the way of the server's signal. Putting the server on a full-duplex connection changes the technical details but it doesn't have any meaningful effect on what happens. The purpose is not to break anything, the purpose is to cause so much extra signal that the server can't get its own signal out. Just like a counter-protest. The thing being intentionally degraded is the connection. Not the "speaker". Or if it's a higher level DDoS then it's usually the DDoS clients filling the queue. Which is also a pretty normal counter-protest thing, getting in the way and taking up all the close audience space.

Dylan16807 3y ago

If you get a thousand people together to perform a DDoS, I can see that. But usually a DDoS is done by one person. That's not analogous to counter-protesting.