Just to follow up-- http://emis.impa.br/EMIS/journ... constructing "r-ary" (i've always heard them called n-ary) maximally balanced gray codes. A fairly recent set of findings (2007) which would guarantee you optimality.
Means that every wheel has to spin close to 2000 times for a complete brute force of a 5 digit code (1000 on average; this comes from 100000 combinations / 5 shared equally over all wheels / 10 per revolution), or 250 times in the brute force of a 4 digit code (125 on average). Probably still not possible for a 1s attack, but much better than the bounds you suggest.
> Through a physical interface? I find that unlikely. At least one of those luggage combination dials would have to make 10,000 revolutions in the process No.
(F9 1.1, more but I don't know how much for F9FT...)
But... When we're talking about impact, we're going to just focus on the direct amount of non-oxidizer fuel used in the first stage, but not the massive logistics chains involved, the cryogenics, etc? I agree the impacts aren't *too bad* but I think you're low by an order of magnitude and maybe two. The cost and environmental impact of rocket launches is not in the kerosene used.
This is exactly it. I've been a subsystem maintainer. It's not really a fun place to be. I have to admit sometimes I was deliberately a jerk to be heard.
There are times where being blunt and even, to an extent, personal can be justified. It certainly often makes what you want to have happen now, happen. But you end up paying the penalty for it later-- when people are worried about embarrassment and don't mention their concerns; when you hear the same attitude in kind; when you lose valued contributors. So IMO better save that "ammo" for the true existential concerns and not make it business as usual for everyone's sake.
It's draining to be on the receiving end of the abrasive behavior, but I came to learn it's draining just to have to exhibit it yourself.
Really IMO population isn't the place to start. Figure out how many people need to get into a commute target in how much time. Then figure out if the station density for that place is practical.
Probably the worst thing is the asymmetry in station loading--- you need lots of relatively low-use stations to minimize walking distance, but you need lots more stations than that in the densest areas to deal with queue lengths during peak demand (and they will tend to only be full in one direction at a time).
Assuming uniform loading for capacity factors is erroneous. People take more trips at 8AM than at 3AM or 2PM. Adding more station density out at the edges of San Jose doesn't help capacity in the downtown at rush, too.
So, simple logic... if the number of prisoners is reasonable, allowing them to vote makes no difference on election outcomes. And if the number of prisoners is unreasonable, holy shit, we're disenfranchising a big set of society-- not letting them have any influence on the laws that have been used against them.
IMO "not letting prisoners/felons/etc" vote is a huge fuck-up/back door to democracy. All you need to do to erode the political influence of a class is criminalize things associated with that class.
Look dude, it's clear you're a true believer-- I don't know what degree to which the things you're saying are deliberate spin, and what aren't. I am neither long nor short TSLA (am investigating perhaps buying convertible notes & hedging with shorting the stock, but that would be a mostly-long position).
1. David Morton wasn't brought in to take TSLA private-- or shouldn't have been, as that's not his expertise. Sure, he did a transaction like that as a junior financial executive-- 16 years ago, but has been a CFO of a public company for the past 3 years and now went to Anaplan which is working on going public. It's believed that he was being positioned to take the reins as TSLA's CFO before his departure. I can sure understand shitting a brick and polishing your resume if you go somewhere to be a public CFO, and your first day, the chairman/CEO tweets about privatization without talking to you.
2. It's never good, with this degree of scrutiny, to do the kind of stupid shit that has gone down the past month.
3. TSLA needs more capital. It is difficult to fund significant growth from operations, and practically impossible with TSLA's degree of debt loading. While it may be possible to survive Q1's debt payment requirements (or even rally the stock and convert the notes), that doesn't leave extensive funds for capital expansion.
That said, with TSLA's prior capitalization it should have been easy to get funding from a small (in share) equity round-- if stupid shit stops capital markets should open back up.
4. TSLA's valuation is predicated, even now, on making it through this challenging time and significantly growing and maintaining market share. This is never good if you have many market entrants who are willing to sell product at a loss (for regulatory compliance reasons). This is never good if other participants are willing, nay practically required, to enter the market and sell below cost and have greater access to subsidies and favorable trade terms than you.
TSLA has a better product and has an initial lead in share and a positive brand reputation. In a growing market, that's a wonderful position to be in. But success is predicated on strong execution, access to capital, and not doing stupid stuff.
> It seems that their range was too short, so they sort of fudged the numbers a bit to make it sound better?
"New European Drive Cycle" -- aka the measurement standard used by the European Union. It is not surprising that a European automaker would choose to report NEDC numbers-- though of course these are being replaced with WLTP (but WLTP gives a lower result than NEDC, so you don't want to report a WLTP number when your competition has reported NEDC).
Model X 75D has a NEDC of 259 miles, vs. 280 miles for the EQC-- so it's a bit better than the 75D and a fair bit worse than the 100D.
> It's always the person in back's fault in a rear end collision. ALWAYS.
Not necessarily in a merge. I agree it's probably the human driver's fault, but if you dive in front of someone it is not necessarily physically possible for them to stop.
I don't see why it's bad for the car to blink the turn signal to notify other people what's happening if the driver has failed to. It seems to deeply offend you, though.
This is the car "saving you" from violating this law. It's OK if it tries to only do it when it's pretty sure you forgot and not other times when it thinks you chose not to signal on purpose or that signalling may not be appropriate.
> Yeah, I was confused about that argument about fractured filesystems (among other things) in the original post. How and why does that matter to them, unless they're doing something at a lower level than userspace?
If you want to know about filesystem changes, odds are you are using the inotify API. Sadly, the inotify API does not behave exactly identically between different filesystems (and between different application access methods). It has also drifted in behavior across kernel versions.
Even though Unix has pretty good abstractions, at some point different behavior under the hood starts to show through this.
It's mostly a huge testing burden. There's definitely ways to structure your code that it unconditionally works, but it's not completely obvious and it needs to be (repeatedly) verified.
There's the "always checksum" -c option. It makes rsync more expensive in local resources -- it has to read every file instead of relying on datestamps --- but much more reliable.
> As Tesla has repeatedly pointed out, a profitable company does not need equity rounds
This is false. If you want to grow, you need capital. If you want to grow quickly, you need a lot of capital: more capital than can be generated from operations, and borrowing against capital equipment and receivables only provides very limited leverage. Also, by using purely debt for leverage, with limited cash balance-- there's a massive exposure to any small downtick in the business.
If TSLA does indeed reach sustained profitability, they still need massive increases in scale to justify their valuation: 20% CAGR or more sustained for the next decade; this will take significant capital.
Profitability is nice, in that it gives you choices in how to proceed and this is option value. But profitability does not, in itself, escape the need for equity raises.
Hey, I'm pretty skeptical of TSLA. But A) general assembly of something that is subsequently gonna go on dirty, dirty roads doesn't belong in a clean room, (hell, we don't build circuit board assemblies in clean rooms) and B) Sprung buildings are really cool and are not what I'd call a tent.:P
You're looking in the wrong place. 15 USC 78(u)(a).1 authorizes investigations and assignment of civil penalties associated with security *REGULATIONS* established under the securities acts, along with blue sky laws and other mechanisms. Criminal liability can in turn attach from violation of the regulations, per 78ff. So you need to leap from USC to the extensive CFR to know the established regulations (though such regulations are void to the extent that they exceed the SEC's statutory rulemaking authority and conflict with actual established law).
You've got it backwards. This isn't to detect criminals, but rather to determine whether specific people are accredited personnel entitled to special access. False negatives are OK (you can figure out if it's really them / why it's not working), but false positives are bad.
> and generally loss leaders (aka "compliance cars").
Don't you find that scary for TSLA? Tesla has great product, but a bunch of competitors willing to sell product below cost generally isn't good for your ability to sustain margins, especially if the gaps between products are going to narrow. (And gaps between innovative products and everything else *always* narrow).
> Right, the important thing is to buy a reputable product like a Thinkpad.
Even then-- if you buy the Thinkpad X1 ultrabook, you end up with a socketed NVME drive.... but if you buy the Thinkpad X1 convertible laptop, you end up with soldered-down storage.
Slashdot Mirror
Just to follow up-- http://emis.impa.br/EMIS/journ... constructing "r-ary" (i've always heard them called n-ary) maximally balanced gray codes. A fairly recent set of findings (2007) which would guarantee you optimality.
Means that every wheel has to spin close to 2000 times for a complete brute force of a 5 digit code (1000 on average; this comes from 100000 combinations / 5 shared equally over all wheels / 10 per revolution), or 250 times in the brute force of a 4 digit code (125 on average). Probably still not possible for a 1s attack, but much better than the bounds you suggest.
> Through a physical interface? I find that unlikely. At least one of those luggage combination dials would have to make 10,000 revolutions in the process
No.
https://en.wikipedia.org/wiki/...
I come up with something more like 90 cars:
> 474600 kilograms / (840 kilograms/m^3) / (8 liters/100 kilometers) / 22000 kilometers / 3.56 = 90.17
(F9 1.1, more but I don't know how much for F9FT...)
But... When we're talking about impact, we're going to just focus on the direct amount of non-oxidizer fuel used in the first stage, but not the massive logistics chains involved, the cryogenics, etc? I agree the impacts aren't *too bad* but I think you're low by an order of magnitude and maybe two. The cost and environmental impact of rocket launches is not in the kerosene used.
This is exactly it. I've been a subsystem maintainer. It's not really a fun place to be. I have to admit sometimes I was deliberately a jerk to be heard.
There are times where being blunt and even, to an extent, personal can be justified. It certainly often makes what you want to have happen now, happen. But you end up paying the penalty for it later-- when people are worried about embarrassment and don't mention their concerns; when you hear the same attitude in kind; when you lose valued contributors. So IMO better save that "ammo" for the true existential concerns and not make it business as usual for everyone's sake.
It's draining to be on the receiving end of the abrasive behavior, but I came to learn it's draining just to have to exhibit it yourself.
Really IMO population isn't the place to start. Figure out how many people need to get into a commute target in how much time. Then figure out if the station density for that place is practical.
Probably the worst thing is the asymmetry in station loading--- you need lots of relatively low-use stations to minimize walking distance, but you need lots more stations than that in the densest areas to deal with queue lengths during peak demand (and they will tend to only be full in one direction at a time).
Assuming uniform loading for capacity factors is erroneous. People take more trips at 8AM than at 3AM or 2PM. Adding more station density out at the edges of San Jose doesn't help capacity in the downtown at rush, too.
Same thing China says about dissidents.
It's the kind of thing you see in an encyclopedia.
e.g. from Encyclopedia Britannica
> Hillary Clinton, in full Hillary Rodham Clinton, née Hillary Diane Rodham, (born October 26, 1947, Chicago, Illinois, U.S.)
This lets you know what she was originally named.
So, simple logic... if the number of prisoners is reasonable, allowing them to vote makes no difference on election outcomes. And if the number of prisoners is unreasonable, holy shit, we're disenfranchising a big set of society-- not letting them have any influence on the laws that have been used against them.
IMO "not letting prisoners/felons/etc" vote is a huge fuck-up/back door to democracy. All you need to do to erode the political influence of a class is criminalize things associated with that class.
Look dude, it's clear you're a true believer-- I don't know what degree to which the things you're saying are deliberate spin, and what aren't. I am neither long nor short TSLA (am investigating perhaps buying convertible notes & hedging with shorting the stock, but that would be a mostly-long position).
1. David Morton wasn't brought in to take TSLA private-- or shouldn't have been, as that's not his expertise. Sure, he did a transaction like that as a junior financial executive-- 16 years ago, but has been a CFO of a public company for the past 3 years and now went to Anaplan which is working on going public. It's believed that he was being positioned to take the reins as TSLA's CFO before his departure. I can sure understand shitting a brick and polishing your resume if you go somewhere to be a public CFO, and your first day, the chairman/CEO tweets about privatization without talking to you.
2. It's never good, with this degree of scrutiny, to do the kind of stupid shit that has gone down the past month.
3. TSLA needs more capital. It is difficult to fund significant growth from operations, and practically impossible with TSLA's degree of debt loading. While it may be possible to survive Q1's debt payment requirements (or even rally the stock and convert the notes), that doesn't leave extensive funds for capital expansion.
That said, with TSLA's prior capitalization it should have been easy to get funding from a small (in share) equity round-- if stupid shit stops capital markets should open back up.
4. TSLA's valuation is predicated, even now, on making it through this challenging time and significantly growing and maintaining market share. This is never good if you have many market entrants who are willing to sell product at a loss (for regulatory compliance reasons). This is never good if other participants are willing, nay practically required, to enter the market and sell below cost and have greater access to subsidies and favorable trade terms than you.
TSLA has a better product and has an initial lead in share and a positive brand reputation. In a growing market, that's a wonderful position to be in. But success is predicated on strong execution, access to capital, and not doing stupid stuff.
Model X 75D has a NEDC-measured range of 259 miles, vs. 280 miles for the EQC-- so it's a bit better than the 75D and a fair bit worse than the 100D.
> It seems that their range was too short, so they sort of fudged the numbers a bit to make it sound better?
"New European Drive Cycle" -- aka the measurement standard used by the European Union. It is not surprising that a European automaker would choose to report NEDC numbers-- though of course these are being replaced with WLTP (but WLTP gives a lower result than NEDC, so you don't want to report a WLTP number when your competition has reported NEDC).
Model X 75D has a NEDC of 259 miles, vs. 280 miles for the EQC-- so it's a bit better than the 75D and a fair bit worse than the 100D.
> It's always the person in back's fault in a rear end collision. ALWAYS.
Not necessarily in a merge. I agree it's probably the human driver's fault, but if you dive in front of someone it is not necessarily physically possible for them to stop.
Turn signal effort is an enforced exaction, not a voluntary contribution.
I don't see why it's bad for the car to blink the turn signal to notify other people what's happening if the driver has failed to. It seems to deeply offend you, though.
This is the car "saving you" from violating this law. It's OK if it tries to only do it when it's pretty sure you forgot and not other times when it thinks you chose not to signal on purpose or that signalling may not be appropriate.
> Yeah, I was confused about that argument about fractured filesystems (among other things) in the original post. How and why does that matter to them, unless they're doing something at a lower level than userspace?
If you want to know about filesystem changes, odds are you are using the inotify API. Sadly, the inotify API does not behave exactly identically between different filesystems (and between different application access methods). It has also drifted in behavior across kernel versions.
Even though Unix has pretty good abstractions, at some point different behavior under the hood starts to show through this.
It's mostly a huge testing burden. There's definitely ways to structure your code that it unconditionally works, but it's not completely obvious and it needs to be (repeatedly) verified.
There's the "always checksum" -c option. It makes rsync more expensive in local resources -- it has to read every file instead of relying on datestamps --- but much more reliable.
> As Tesla has repeatedly pointed out, a profitable company does not need equity rounds
This is false. If you want to grow, you need capital. If you want to grow quickly, you need a lot of capital: more capital than can be generated from operations, and borrowing against capital equipment and receivables only provides very limited leverage. Also, by using purely debt for leverage, with limited cash balance-- there's a massive exposure to any small downtick in the business.
If TSLA does indeed reach sustained profitability, they still need massive increases in scale to justify their valuation: 20% CAGR or more sustained for the next decade; this will take significant capital.
Profitability is nice, in that it gives you choices in how to proceed and this is option value. But profitability does not, in itself, escape the need for equity raises.
Hey, I'm pretty skeptical of TSLA. But A) general assembly of something that is subsequently gonna go on dirty, dirty roads doesn't belong in a clean room, (hell, we don't build circuit board assemblies in clean rooms) and B) Sprung buildings are really cool and are not what I'd call a tent. :P
You're looking in the wrong place. 15 USC 78(u)(a).1 authorizes investigations and assignment of civil penalties associated with security *REGULATIONS* established under the securities acts, along with blue sky laws and other mechanisms. Criminal liability can in turn attach from violation of the regulations, per 78ff. So you need to leap from USC to the extensive CFR to know the established regulations (though such regulations are void to the extent that they exceed the SEC's statutory rulemaking authority and conflict with actual established law).
You've got it backwards. This isn't to detect criminals, but rather to determine whether specific people are accredited personnel entitled to special access. False negatives are OK (you can figure out if it's really them / why it's not working), but false positives are bad.
> and generally loss leaders (aka "compliance cars").
Don't you find that scary for TSLA? Tesla has great product, but a bunch of competitors willing to sell product below cost generally isn't good for your ability to sustain margins, especially if the gaps between products are going to narrow. (And gaps between innovative products and everything else *always* narrow).
> Right, the important thing is to buy a reputable product like a Thinkpad.
Even then-- if you buy the Thinkpad X1 ultrabook, you end up with a socketed NVME drive.... but if you buy the Thinkpad X1 convertible laptop, you end up with soldered-down storage.
> is only 4 times more than Tesla's real-world miles on active mode and 2 times more than Tesla in shadow mode.
> take them over 130 years to match Tesla's active AP miles and double that to match its shadow mode miles.
Is one of these reversed?