Hacker News new | comments | show | ask | jobs | submit | lutorm's comments login

Dragon already lands on water, so there's no technical reason it couldn't do it after a launch vehicle failure, if the event was survivable. Now we know that such a scenario exists, but that was not obvious before.

My understanding is Dragon 1 does not float fully loaded. It can get 3,10 kg ISS used and dispose of 3,10 kg of waste, but only to recover ~2,500 kg. https://en.m.wikipedia.org/wiki/Dragon_(spacecraft)

Make that "3,310 kg ISS" not "3,10 kg"

I believe the difference between up and down cargo is because the trunk is dropped and burns up on entry, only the pressurized capsule itself lands.

As pointed out in the article, this is basically a direct consequence of the commercial crew program:

"The contracts call for orders to take place prior to certification to support the lead time necessary for missions in late 2017, provided the contractors meet readiness conditions."

There is plenty of time for NASA to back out if either of the companies don't meet the milestones.

That's 1.2^3, or 70% more... not even close. Also remember that making the rocket bigger also increases the dry mass, which further increases the fuel requirements.

The difference in ISP between a liquid-fueled rocket (say 450s for LH2/LO2, 350s for RP1/LO2) and solid propellant (250s) makes a huge difference to your prop fraction. To get, say 8km/s delta-v, just plug it into the rocket equation:

LH2/LO2: 6.1 initial/final mass. RP1/LO2: 10.3 solids: 26

Staging can help this a bit, but the basic fact remains. It's very difficult to get useful payload to orbit with solids.

That's not correct. The maximum acceleration of a Dragon mission has been 5G, so if you abort at that time you need 5G just to get your butt off the launch vehicle. You can't abort sideways.

The abort system for the Dragon capsule provides 4.5G of acceleration (human dragon space flight will not reach 5G, don't mix in cargo launches, irrc SpaceX said that the nominal max acc. for human flight will be below 3G similar to an STS launch) and it does kinda abort side ways, as soon as it clears the main rocket it tilts and gets the heck out of the way because it's quit important to be as far away from the rocket as possible in case it explodes (also to prevent any chance of rocket derby catching with the capsule during decent) and out-running it upwards isn't a possible in the long run.

If the point is to abort, you can't rely on the engines being throttled down to the accel limit -- you have to be able to successfully get out of any reasonable scenario and the engines running away does not seem unreasonable. (I don't know exactly what the abort requirements are, but the point is just that it's not easy.)

Also, during an abort scenario, the aero drag is pushing Dragon back onto the launch vehicle, so you actually need more thrust to overcome that, too. (Although that obviously depends very strongly on the dynamic pressure at which you abort.)

Although if you count F9-4 as a failure, you should also count the STS-51-F Abort-to-Orbit as a failure, which brings it to 3/135.

Probably STS-93 too.

These two bring up an interesting philosophical point. Are they successes of failures? To me, it sounds like engineering done right. "Something bad" happened, and yet the vehicle survived and even completed its mission.

Well then it's only 1 spacex failure.

People aren't arguing the merits of success they just want to use the same rules to talk about nasa and spacex

That's a hope though, not a fact...

That's an estimate based on engineering analysis and statistics, just like any estimate of reliability.

It's not an estimate based on statistics, because there are no statistics for the system you are talking about. It is a prediction based on engineering analysis, and the prediction for space shuttle reliability was quite a bit better than the actual record turned out to be.

I want SpaceX to succeed as much as anyone (probably more since I work there), but one should not delude oneself into thinking that predictions are fact.

You know Congress is not subject to the Freedom of Information Act, right? They just happened to exclude themselves from that one.

Yes, but I'm saying they shouldn't be excluded :)

How could they possibly think that jamming public frequency bands would not be illegal?

https://www.fcc.gov/encyclopedia/jammer-enforcement clearly states "Federal law prohibits the operation, marketing, or sale of any type of jamming equipment, including devices that interfere with ... wireless networking services (Wi-Fi)."


I'd fully support criminal charges. If I jammed public spectrum in order to profit, I'd be in prison.


They're getting fined minutes of revenue, that's practically the same as prison!


Approximately 24.67 minutes for what it's worth. Marriott made > $12 billion last year and was fined $600k. I'm curious how much revenue the wifi fees bring in and whether that fine even makes a dent in income from that alone.


The article says they were charging meetings up to $30,000 for internet access alone. I'd guess it was probably profitable.


... revenue is absolutely meaningless in this conversation. You should be looking at profit. It doesn't matter what Marriott charges, it matters what they make, and saying they "made > $12 billion" is totally false: they made at the high end something like $2 billion (which is still high enough that you could have used those numbers as an honest ratio instead of distorting the discussion by looking at revenue).


I'd say revenue's probably the better number. Profit varies by quite a number of factors, but revenue is a reasonably reliable yardstick for the size of the business.

For example, consider Amazon, who didn't turn a profit for decades:


Also, maybe you could tone down the drama? Even if you were right, which you aren't, the comment you're replying to was clearly well meant.


Nobody uses profit when determining the appropriate fines for an individual. A court doesn't look at an offender making $100,000/year and say, well, they're only saving $5/month, so we'll only fine them $60.

If we're talking about the discrepancy between punishments for individuals and punishments for companies, then we should absolutely be looking at revenues, not profits. Or alternately, punishments for individuals should look at savings, not income.


> If I jammed public spectrum in order to profit, I'd be in prison.

Is this true? Can someone cite examples of this happening?


Here's a fellow who spent 15 months in prison for jamming ham radio, so not even for profit:


This guy got seven years, but he was interfering with public safety radios operated by police and the like, so that's not the same:


Looking at other convictions, it seems that the more usual punishment is arrest, probation, and a massive fine. Even compared to that, fining a big hotel $600,000 is a joke.


I want the board to go to jail, not some low level janitor. Or do we only go as high as the paper trail takes us? This lets companies erect firewalls and appear innocent like the vw CEO...


The alternative is imprisoning people without evidence. I'll take guilty people going free anyday.


Or, tweaking the law along the lines that corporate executives are vicariously liable for the job related misdeeds of their underlings, if it can be shown they did or should have known about the goings on and did nothing to stop it.

Suddenly, there will be a lot more care about ethical behavior..


I don't agree. People should not go to prison for crimes committed by others. We should still have to show each individual is guilty, including the executives. But I don't think that would solve the problem anyway. Owners and shareholders will still be immune. They will simply hire figurehead executives to sign things and speak at events.

These people are financially motivated, so the punishment should be financial. Most criminals don't think they are going to be caught anyway, so prison isn't a disincentive. Financial punishment may not be a disincentive either (but I suspect for most shareholders it would), but it has the effect of reducing the efficacy of bad actors, leading to better outcomes overall.

The real problem is the fines are usually far outweighed by the profits from acting in this manner. Make the punishment an audit at the end of which they must pay the costs of the independent audit (whether done by private third party or government) as well as a multiplier on whatever profits were determined to stem from the elicit activity.


Yes, fines might be effective if they were bigger.

Many of the people I found while searching got fines that probably equated to something like a year's salary. Fine these companies an amount comparable to a year's revenue and I bet things would change in a hurry.


I didn't say anything about prison, but the term used was carefully chosen in that in some capacities, you are liable for the actions of other people.

While I agree with you about prison not working, fining a multibillion dollar company the equivalent of a few day's profits aren't good enough. The fines need to be much more strident, along the lines of administrative dissolution or something else that would motivate the top brass via angry shareholders.


At what point are we willing to entertain a fine large enough to bankrupt an enterprise?

Should we consider the economic impact to the neighborhood, city, state, or nation when levying fines?

Do you think VW shareholders are angry? I think not. They know management did what it had to do. I'm just mad that they threw the programmers under the bus.


It doesn't seem like we consider the economic impact of fines to individuals. The government looks to be more than happy to bankrupt people as part of punishment. Why should corporations get special treatment. "Too big to fail"?


Owners and board members of companies should be liable not just for the action of their employees but also the actions of their subcontractors. Otherwise, we will have situations like the one John Oliver mocked in the North Dakota petroleum industry.



Well the odd thing is, in prison where you are not permitted a cell phone if you "had" one the authorities are not permitted to jam it.


They could take it since you're a prisoner.



This is novel. You scrolled to the bottom without reading the top. The top of that page is very clear:


Federal law prohibits the operation, marketing, or sale of any type of jamming equipment, including devices that interfere with cellular and Personal Communication Services (PCS), police radar, Global Positioning Systems (GPS), and wireless networking services (Wi-Fi)"

Then it references The Communications Act of 1934:

"The Communications Act of 1934

Section 301 - requires persons operating or using radio transmitters to be licensed or authorized under the Commission’s rules (47 U.S.C. § 301)"

But wait, there's more! 47 U.S.C. 302b. reads:

"(b) Restrictions No person shall manufacture, import, sell, offer for sale, or ship devices or home electronic equipment and systems, or use devices, which fail to comply with regulations promulgated pursuant to this section"

(Emphasis mine).

Have a read of the Communications Act before you get into trouble.


The FCC doesn't even have authority to prosecute criminally.

Well, there's a novel legal theory.

Try running an unlicensed broadcast transmitter, and let us know how that goes for you.


Excuse me, aren't you supposed to be a lawyer? Please read item number 8 here and stop spreading misinformation that can get others in serious trouble:


"What are the penalties for using a jammer? Can I go to prison?

Yes. The unlawful use of a jammer is a criminal offense and can result in various sanctions, including a jail sentence."


They're a large business. Regulatory compliance is a question of whether they can get away with it, and no individual will be prosecuted.


The AP needs to know your network exists to send deauth packets - if you lower the power on your device so that your network is only accessible to you, you'll end up fine. So the argument for the other side would basically be that you're not operating a private device if it's broadcasting so their customers may attempt to connect to it. I think many companies' security groups would have a problem with you saying that someone needs to be able to run a publicly accessible access point in their building.

I think the complication comes here from the fact that they were charging for internet and then preventing people from using their own. If there were hotspots trying to mimic the network, I don't think there would be any problem with deauthing them since that's a security risk but it seems like in this case it was just a money grab.

I guess the question becomes for other companies not doing it for the money grab, if you are offering wifi, do you have a duty to protect your customers from rogue hotspots or is it their problem?


I would hope that those security groups you mention would know that, under the law, they do not own the airwaves in their buildings, and therefore it doesn't matter what their opinion is on whether somebody "needs" to run a hotspot. All that matters is that a hotspot is an authorized device and intentionally interfering with the operation of an authorized device is against the law.


That is not a distinction the law makes. You are not allowed to operate a GPS jammer for your home safety, either.

There are exceptions, prisons can operate cellphone jammers for example. I don't think there are any for private enterprises.


> There are exceptions, prisons can operate cellphone jammers for example.

It is not legal for prisons to jam cell phones.

"The Communications Act prohibits non-Federal entities from using cell jammers. The FCC cannot waive this statutory prohibition absent a change in the law by Congress." [1]

There was an attempt to add an exception for prisons, but it was never passed (Safe Prisons Communications Act of 2009[2]).

[1] http://www2.fcc.gov/pshs/docs/summits/Combating-Contraband-C... [2] https://www.govtrack.us/congress/bills/111/s251


Could the federal government set up a contracting arm that sells jamming services to state facilities? Or does non-Federal entity mean something else?


Look up the Communications Act of 1934, specifically 47 USC 302a:

"(c) Exceptions

The provisions of this section shall not be applicable to carriers transporting such devices or home electronic equipment and systems without trading in them, to devices or home electronic equipment and systems manufactured solely for export, to the manufacture, assembly, or installation of devices or home electronic equipment and systems for its own use by a public utility engaged in providing electric service, or to devices or home electronic equipment and systems for use by the Government of the United States or any agency thereof. Devices and home electronic equipment and systems for use by the Government of the United States or any agency thereof shall be developed, procured, or otherwise acquired, including offshore procurement, under United States Government criteria, standards, or specifications designed to achieve the objectives of reducing interference to radio reception and to home electronic equipment and systems, taking into account the unique needs of national defense and security."

(Emphasis mine)

That's a very clear exception.


Hmmm... I'm a bit of a goose and misread that. The bit I got wrong is:

"under United States Government criteria, standards, or specifications designed to achieve the objectives of reducing interference to radio reception and to home electronic equipment and systems, taking into account the unique needs of national defense and security."

Sorry about that.


Why would you need to? I'm not sure I understand the point.

You're not indiscriminately preventing wifi from working. If your network is broadcasting with enough power to reach the AP, and therefore customers besides yourself, they would be preventing that.

If someone sets up a wifi access point outside your building with the same name as your wifi, you should be able to prevent people from accidentally connecting to that rogue access point. That's why basically every enterprise AP comes with that ability. I can't imagine if this were actually illegal the FCC would allow the sale of systems with the ability to do that.

The distinction here is that it was done as a money grab and not for security reasons.


Why are you talking about "with the same name as your WiFi"? That might change things (the people setting up the rogue network would be deliberately interfering with your network, after all), but as far as I can see that's not what anybody is doing here.


Because I'm not trying to defend what they did -- they were wrong and just trying to grab money.

I am saying that using deauth packets to prevent wifi from operating does not seem to be outright illegal and there are valid reasons to be able to do it. Considering basically every high end AP has the ability to target rogue APs and the FCC takes a dim view on people selling jammers, it seems like the context of the application of this determines whether you are jamming or not and not the use of deauth packets by itself.


Many AP's have the ability to operate outside of their proper bands and above the stated power limits too. Existence of a feature doesn't make its usage legal.

The legal principle here is remarkably clear cut.

Are you willfully interfering with another user of the band? Yes? You're breaking the law.

The method used to accomplish the interference isn't germane to the question of whether you're interfering or not.


I'm aware that nearly every enterprise AP has the ability to detect rogue access points. I am not aware of any with the ability to attack them. Also, as far as I know, none of them can target devices that are not actually attached to their network.


It's a common feature. IIRC, in Cisco WLC land the feature is referred to as active rogue containment:


As someone who has experienced the fallout firsthand from a neighboring tenant in a downtown Seattle office building who used it, I'd challenge the idea that there are valid reasons to do this.

If you control the physical network - don't allow rogue APs on your network.

If you control the client and care about them connecting to access points that aren't under your control, then manage that instead.


There is a debate here about whether syntactically valid 802.11 frames count as "jamming".


From the horse's mouth: https://transition.fcc.gov/eb/jammerenforcement/jamfaq.pdf

Section 333 of the Communications Act : “No person shall willfully or maliciously interfere with or cause interference to any radio communications of any station licensed or authorized by or under [the Communications] Act or operated by the United States Government.” 47 U.S.C. § 333.

If you're willfully screwing with other people's transmissions, you're breaking the law. It's that simple. The method being used doesn't really enter into it.


Wifi operates in unlicensed spectrum and transmitters and receivers are required to handle interference in this space. Is it still jamming if it's not a licensed station doing the broadcasting?


...I suppose it does say "authorized" there, and 2.4 and 5Ghz transmitters are authorized.


It always feels funny to me to refer to the ISM bands (the spectrum itself) as "unlicensed".

- Devices operating there have to be approved

- There are rules about how the devices work

- There are rules about how the _operator_ is allowed to operate the approved devices ("Operation is subject to the following conditions" sticker on most equipment).

(all of this under FCC part 15 rules)


In other bands, the operator in control of the transmitter generally needs to file paperwork with the FCC to get a license (in some systems like cellphones, the base station is in control so the subscribers don't need licenses, just approved equipment).


Understood you're playing devil's advocate, but the response would be: Why don't I stand outside with a WiFi antenna pointed at your house, doing the same thing, causing your WiFi devices to drop connections over and over, and then you tell me that I'm not jamming you.


Even further, denial of service attacks can be performed using totally valid and normal HTTP or HTTPS requests. In that case, it's the volume and the intent that matter.


"unlicensed" devices are still authorized by the FCC.

So your devil's advocate is unnecessary.


Yes its still jamming. Even though the devices are unlicensed, you are still not allowed to intentionally cause interference with other devices. You can think of it as the requirement to use the ISM bands is you have to be nice to other users of the band.

The requirement on receiving interference means that you don't get to gripe to the FCC about other users of the band that are otherwise being nice.


For added irony, the hospitality industry claims that allowing hotspots will "cause interference with their networks".


That's not even implicit, but an explicit part of the use of the 2.4 and 5GHz spectrums. "There may be interference. Wear a helmet."


Just playing devils advocate but it doesn't sound like they're creating radio interference. It sounds like they're interfering at the protocol layer, which would be another matter (likely also against regulations)


You don't have to be "creating radio interference". You have to be "interfering with radio communications".

Intentionally preventing a radio communication from occurring is interfering with it.


Ultimately if you follow the network layers it gets to physical media. Without physical media nothing works. Ergo, just layering over the top of the physical layer means you still interfere with wifi comms.


How? The law as written refers to "willfully or maliciously interfere with" authorized equipment. They purchased equipment and configured it to do just that, clearly on purpose. The only possible outs here are if the equipment is a hoax and doesn't actually work, or if they managed to purchase and configure this equipment completely by accident.


Would the people who claim that those syntactically valid 802.11 frames in question aren't "jamming" hold the same opinion if it were the exhibitors, attendees, or other punters who brought in the disruptive equipment? Or are they simply being pedantic about the imprecise usage of the word "jamming"?


Bonus points: "Jamming" is not used in the law.


No there's not.


Jamming is not a relevant term.

Interference is, and it is illegal.


So that blurb on the FCC site isn't the law. The fines are issued under 47 USC 333, but on its face that only prohibits interfering with any "station licensed or authorized" under that chapter: https://www.law.cornell.edu/uscode/text/47/333. Is a WiFi hotspot a "station?" It's not licensed either. Though it is "authorized" to operate in the unlicensed band.

Using deauth packets, versus actual jamming, sounds like it wouldn't even violate unlicensed band requirements. Last I checked nothing in the FCC regs requires you to respect 802.11 or other protocols while operating in the unlicensed band. You're supposed to abide by power limits and things like that, which the hotels were doing.


Yes, a WiFi hotspot is a station: https://www.law.cornell.edu/uscode/text/47/153.

  The term “radio station” or “station” means a station
  equipped to engage in radio communication or radio 
  transmission of energy.


That begs the question of whether it is a "station."


> Is a WiFi hotspot a "station?" It's not licensed either. Though it is "authorized" to operate in the unlicensed band.

It's a very interesting question, but there is a very good argument that it is.

Have a look at pages 3-6 of this for a taste: http://apps.fcc.gov/ecfs/document/view?id=60001011985


This sort of "commodification", for lack of a better word, is visible in many areas, I think. I was thinking about this lately in relation to tools. It's great that you can now get tools for unbelievably low prices compared to, say, 50 years ago. This is a big win in many instances, because a shitty tool is often better than no tool at all, and it opens up access for many who would never be able to afford the old tools.

The flip side, however, is that it's devalued quality. It's remarkably difficult to actually find high-quality stuff these days. Even what used to be high-end brands have been bought up by some conglomerate that is now selling cheap Chinese versions under the old names.

There's still a market for quality, of course. You don't use tools from Harbor Freight when building rockets, but you'll never see those in any store you visit and they're likely to be priced far outside the reach of a normal person. It's like the middle ground has been lost, most stuff is cheap and low quality and then there is this small high end of really expensive stuff.


HN is always full of news about startups that burn millions of dollars in order to, perhaps, eventually, have enough of a product and enough exposure to become a profitable business.

How is what the winner of this bid did any different from what all these startups do every day? At least they staked their own money to make a name for themselves.

Isn't the term for this a "loss leader"?



Guidelines | FAQ | Support | API | Security | Lists | Bookmarklet | DMCA | Apply to YC | Contact