IIUC the 30% performance hit was only against a synthetic workload, not anything that one would expect. The more normal hit was closer to 3%. Still not a pleasant change, but less disastrous. Additionally, someone said that if you were still using hard disks rather than SSDs, that you wouldn't notice, because it was hidden by disk latency. Perhaps they're right.
So basically, the patch is really annoying, and slows down your system, but it's necessary, and not crippling. Unless you are on an older Intel CPU with SSDs rather than disks, and you are running close to the limits of your system. Unless you are running a very unusual workload, that it is expected nobody is running except for tests.
Again, this is just my understanding, and is based on a summary of the above posts.
It's not clear to me that Spectre is the attack being talked about. If not, then it is disingenuous, or lying, to try to also blame other manufacturers. There was also a note on the linked page saying that Spectre is a more difficult to execute. They didn't say (at least in terms that I understood) how much more difficult, or whether local access was needed.
If the attack being discussed was what that page called "Meltdown" they didn't say that any chips other than those from Intel were affected.
Which eye? The eye has evolved about 10 times (that I've heard of) and there's no particular problem about explaining any one of them. The problem, actually, is explaining why it evolved so slowly. It's as if it was not nearly as advantageous as we assume.
I refer you to "Climbing Mount Improbable" by Richard Dawkins, specifically the chapter "The fortyfold path to enlightenment".
And being cold outside (as well as excessive hot spells) is to be expected in the mid-latitudes during global warming because that causes the jet stream to slow down due to the decrease in temperature difference between the poles and the equator, which leads to the jet stream slowing and occasionally stalling so that an short extreme weather event turns into a longer one.
Why don't you pick a real problem? They *do* exist, but those are just popular because they're easy to spout, not because they're difficult to explain.
But that's not what the "Quoran" said. It said not to drink fermented grapes. So vodka should be perfectly fine...although I'm a bit dubious about the details. For some reason I think the vodka needs to be made from potatoes, and that beer isn't acceptable. Maybe I need to look up that quotation again some day.
Actually, no. The cold weather in mid latitudes is expected to happen during global warming. What happens is the jet stream slows, often stalling, because the poles are warmer so the temperature difference between the equator and the pole is less so there is less energy to drive the winds. This means that both cold snaps and hot spells are likely to stop and linger in one place instead of just moving on through, and THAT means that the cold at the ground can get more intense when it's cold, and the hot spells can get more intense when it's hot.
It's all part of the same engine. It wasn't *just* drama that made Al Gore's film feature a cold spell caused by the slowing of ocean circulation back before anyone had changed the term "global warming" to "climate change". It's what you expect to happen in the mid latitudes. (Of course, you also get more intense hot spells in summer, but those probably weren't deemed as photogenic.)
OTOH, please note that so far the ocean currents haven't slowed, just the jet stream. But the ocean currents *have* showed signs of weakening. Possibly because Greenland is dumping loads of fresh water on top of the heavier salt water. This will *probably* have more effect on Europe than on the US, though.
That said, I am not a climatologist. And I don't run climate models. These are opinions I've picked up by reading popular science magazines like Scientific American...but they are my extrapolations from many of those articles.
You do need to remember that before he linked up with Hitler Mussolini was considered one of the good guys. And he only linked up with Hitler because nobody else was standing up against him.
That said, he was a dictator. But he wasn't a particularly vile one before he came under the Nazi influence. His main intent was to "Make Italy great again!". If Britain or France had stood up to Hitler before the Czechoslovakia election, he would probably have been counted among the allies.
OTOH, modern Italy would never have been born. And Mussolini did have militaristic desires. (Look up the invasion of Ethiopia.) He just wasn't very good at them.
So when you say "Look how well that worked out before", you need to look at the larger picture than just WWII, because that wasn't inevitable. It was caused because France and England wanted German fascism to combat Russian communism...but Hitler had bigger plans.
I think you are being blinded by your desires to political reality.
For Trump to be impeached the House, which is dominated by Republicans, would need to vote a bill of impeachment. For the impeachment to be executed the Senate, which is also dominated by Republicans, would need to vote for a conviction.
Despite what it says in the constitution about grounds, political reality is that the House can impeach on any grounds that it finds suitable, and the Senate can convict if it feels like it.
So far, no president has actually had the sentence of impeachment executed. It would be interesting, though not pleasant, to consider what might happen if the Senate voted to confirm the impeachment, and the President contested the validity of the vote. There is no explicit provision for enforcing the sentence. It those working for the executive branch were to be personally loyal to the President despite the vote of impeachment there doesn't appear to be any way to make the impeachment stick. It would come down to who the army was loyal to.
Sorry, that hypothesis doesn't fly. It may be harder to debug C or assembler than C++, but most other languages provide more usable debugging facilities than does C++, and most of them are easier to write unit tests for. The unit testing for C++ is basically an add-on. Even assert statements in C++ are crippled, unless you use an extension. (Assert statements should include an optional message that is printed with the error, and which can dump variables of interest in a formatted way.)
So C++ is more difficult to debug. There *are* more external tools that you can use to do the debugging, but that's because they are needed.
The first problem I had with Ada was that strings of different lengths were of different types. That could be solved with bounded or unbounded strings, but then you couldn't compare against literals. There were other problems, and there were always ways around them, but I had to used unchecked conversion too often for things that were perfectly safe. The entire thing was a mess due to over-concern with simple type conversions. Dynamically allocated storage is also a mess, though, given the original ideas, quite reasonable. Follow an Ada program through a simple addition of a node to a tree and you should see what I mean.
Another part of the problem was that the optional annexes couldn't be used as they usually weren't implemented. It's great to not have to load them if you don't need them, it's not so great if they aren't available when you need them.
There are lots of other minor design flaws. In a way you can forgive them, because this was being designed at about the same time as C++ was being written, and, unfortunately, it was designed by a committee. So the kitchen sink got thrown in. But they did design everything in a way that enabled maximal security...even where it didn't make any sense. (Why can't I compare two literal strings of different lengths? They've probably changed that by now, though.)
I *HAD* the goddamn coverage I needed. I just needed to stand there bleeding while I tried to convince some damn office worker several states away that I had an injury that needed the coverage, and he appeared to be under pressure to decide that I didn't need it without knowing anything about the case. The insurance did, eventually, pay, but by god I'm not going to do anything to support that kind of system when I have a reasonable choice otherwise.
Did you ever try negotiating payment with a broken arm?
Even with good insurance I once had to negotiate coverage with a broken thumb, because the insurance didn't want to pay. Ever since then I haven't been happy to trust insurance companies, or any other group that has an incentive to not cover your medical problem. And I do recognize that such is essential. Your proposed solution seems to mean the in every medical situation you need to negotiate coverage, which is just about the worst possible approach.
Sorry, but a personal claim by "Anonymous Coward" isn't very convincing. Make a reasoned argument, or cite an external reference, and I'll take you seriously, but for personal history arguments to carry any weight the person needs to have a track record.
The suggestion to "research it on your own" is reasonable. Of course, you run across the problem of what sources to trust.
A different comment said the study quoted didn't say what the article implied.
From other sources I have heard that the older varieties of stent tended to become quickly clogged with blood clots, but that the current versions incorporate a treated surface that prevents that from happening. Anecdotally, I have friends who have had stents, and their conditions have improved markedly...from unable to walk and incoherent to able to walk up a hill and nearly as reasonable as before their problem. Well, that's been a decade or so, and he's failing again. But I'd hardly say that the stent wasn't effective, and I'm not sure that's the problem now anyway. (He's currently in his mid-to-late 80's.)
I would have said "unwieldy". But it already was unwieldy.
It's probably true that if I invested sufficient time C++ would be a complete language. But even were I an expert it would take me a long time to program anything, because there are too many fiddly details.
Let me modify that. C++ is certainly a complete language, but so is C and so is assembler. None of them make it easy to turn well defined ideas into code. I prefer C++ over C because of a very few choices made long ago, e.g. the inclusion of references. Most of the changes since that are not helpful. They don't add functionality, and they don't reduce the number of additional libraries that are needed.
That said, I understand that they want to avoid breaking changes. But they need changes that require breaking things. This means they need to create a minimally different new language. Unfortunately, most such attempts fail. There are multiple different populations of users with different needs, and for one language to serve them all well is impossible. E.g., my use case would be served by a key-value database being built into the language. This can currently be handled by an attached library, but because the library isn't a part of the language, everything needs to be converted to strings before being saved and converted from strings while being restored. But there's no good method of conversion. (It's actually a simple problem in my case, but the general problem isn't soluble by a separate library.) I also need a kind of simplified Actor variation of concurrency. (I don't need the full Actor model.) This is all *doable*, but C++ sure doesn't make it easy, even though the ideas are well-defined.
It looks moderately interesting, but I'm not interested in a language that doesn't allow concurrent execution, and I didn't notice anything about how you handle Unicode. (utf8 is fine, even utf32 internally, but forget about utf16...except where required by foreign function calls.)
The "scan the stack backwards" might solve the problem that caused polish notation to be disliked, but the explanation needs to be clearer.
Also, you need to do something to make the name more searchable (though it sure is an improvement over some language names). I kept getting a Canadian radio station, and I *never* search for radio stations.
All that said, it looks interesting, but I currently need a language that's complete and workable. Good luck with your recruiting.
There was a collection of artistic works by an "AI" printed a few decades ago called "The Policeman's Beard is Half-constructed". It *had* been edited to include only the works that seemed interesting, but I can guarantee that this is also normal for human poets and authors.
The poems didn't emphasize rhyme or even assonance, but this is also common among modern poems.
Personally, I think that a current effort by a major AI would be much better, but it would definitely depend on the critical skills of the trainer. There are reasons that Kipling hasn't really been eclipsed by any more modern poet, and part of it is the way he uses rhyme and assonance, and another part is the emotive handling. Check out the ballad of Boh Da Thone http://www.kiplingsociety.co.u.... Robert Service https://en.wikipedia.org/wiki/... is also good. But note that they don't just use poetic tricks, they also tell a human story. This is going to be difficult for an AI. The poetic tricks are reasonably simple, but understanding the kind of story that people like to hear is more difficult.
The part that didn't sound incoherent sounded wrong. The volume of a sphere grows more quickly with the radius than the surface area. Of course, it was a metaphor, and one can imagine rules such that the "thickness of the rind" would increase fast enough that the volume of the rind would grow faster than the core...but they sure weren't mentioned.
Not that I think the metaphor is in any way useful. It's not as if any one person knows all the stuff "in the core".
If you say so. I never knowingly saw any image of it. I was operating off an article about how annoyed the artist was with the misappropriation of his work.
If it's an actual AI, the bias is not caused by the programming, but rather by the training data set. As to whether it would be more accurate, that's a separate problem from either consistent or fair. My assertion was that it could be more consistent, but that this wouldn't imply it would be fairer. Accurate is a yet more difficult problem when dealing with people, and it's one that human level intelligences have yet to solve.
To be fair, "fairness" is only a solved problem in the opinion of the person doing the judging. But accuracy is even harder, as most people realize that they don't really know what someone else meant, and sometimes realize that they didn't know what they, themselves, meant.
I'm not saying that Facebook is making a good faith effort to solve the problem. I've never looked at it, so I have no idea. I'm saying that the problem as stated is a hard problem. It's easy to, say, ban certain particular words, but that doesn't accomplish very much.
I'm not sure that the problem as stated could be addressed by anything much short of a human equivalent AI, and even that would only allow some particular set of standards to be applied uniformly. It sure couldn't guarantee that the standards were fair.
As an example consider the text "You with a donkey's member!" This is apparently a violently abusive comment, but that depends on context that isn't present. I'm sure I could come up with a context where that would be encouragement. And every single word in that sentence is perfectly harmless. Or what about "Pepe the frog"? That was intended to be a humorous children's cartoon character....but it didn't stay that way, much to the annoyance of the creator.
That said, the evidence seems to support the assertion that Facebook encourages hateful posts, and is more reluctant to censor nazi-ish posts than those with an opposing message. Again, I have no direct evidence for this as I never visit that site, and am relying on material published by others.
The problem is, the Democrats, while not quite as blatantly awful, are nothing to cheer. I support my Representative, and I don't consider one of my Senators better dead. They are all Democrats, and I only feel moderately represented by one of them. That the Republican choices would have been worse is scant reason for liking the situation...and in the case of one of the Senators I'm not sure even that's true.
In this case it often is. The crime is tax evasion. Hotels frequently need to pay a tax on rooms rented, and adhere to various safety regulations that Airbnb ignores. They've been ruled an illegal business in a couple of places because of this.
It stood for extra large, and it was a Mac modified to hold more memory. But that's decades ago, and it might have been somebody's custom modification rather than an official version. It was, however, available at the same place I bought the official Mac versions. (It's even possible that somebody just thought "fat Mac" was inelegant.) I used so many different versions it's hard to keep track of all of them.
I never did any of the modifications, and the repairs were done at the same place that sold the Macs and the custom modified Macs...so I may not have kept them straight. Or, it being decades ago, I may just have scrambled my memories.
Slashdot Mirror
IIUC the 30% performance hit was only against a synthetic workload, not anything that one would expect. The more normal hit was closer to 3%. Still not a pleasant change, but less disastrous. Additionally, someone said that if you were still using hard disks rather than SSDs, that you wouldn't notice, because it was hidden by disk latency. Perhaps they're right.
So basically, the patch is really annoying, and slows down your system, but it's necessary, and not crippling. Unless you are on an older Intel CPU with SSDs rather than disks, and you are running close to the limits of your system. Unless you are running a very unusual workload, that it is expected nobody is running except for tests.
Again, this is just my understanding, and is based on a summary of the above posts.
Not 100% fucked, it sounds like only 3%. Some estimates say 30%, but aren't those on artificial workloads?
OTOH, others above have said that if you're still running on hard disks (as I am) you won't even notice.
It's not clear to me that Spectre is the attack being talked about. If not, then it is disingenuous, or lying, to try to also blame other manufacturers. There was also a note on the linked page saying that Spectre is a more difficult to execute. They didn't say (at least in terms that I understood) how much more difficult, or whether local access was needed.
If the attack being discussed was what that page called "Meltdown" they didn't say that any chips other than those from Intel were affected.
Which eye? The eye has evolved about 10 times (that I've heard of) and there's no particular problem about explaining any one of them. The problem, actually, is explaining why it evolved so slowly. It's as if it was not nearly as advantageous as we assume.
I refer you to "Climbing Mount Improbable" by Richard Dawkins, specifically the chapter "The fortyfold path to enlightenment".
And being cold outside (as well as excessive hot spells) is to be expected in the mid-latitudes during global warming because that causes the jet stream to slow down due to the decrease in temperature difference between the poles and the equator, which leads to the jet stream slowing and occasionally stalling so that an short extreme weather event turns into a longer one.
Why don't you pick a real problem? They *do* exist, but those are just popular because they're easy to spout, not because they're difficult to explain.
But that's not what the "Quoran" said. It said not to drink fermented grapes. So vodka should be perfectly fine...although I'm a bit dubious about the details. For some reason I think the vodka needs to be made from potatoes, and that beer isn't acceptable. Maybe I need to look up that quotation again some day.
Actually, no. The cold weather in mid latitudes is expected to happen during global warming. What happens is the jet stream slows, often stalling, because the poles are warmer so the temperature difference between the equator and the pole is less so there is less energy to drive the winds. This means that both cold snaps and hot spells are likely to stop and linger in one place instead of just moving on through, and THAT means that the cold at the ground can get more intense when it's cold, and the hot spells can get more intense when it's hot.
It's all part of the same engine. It wasn't *just* drama that made Al Gore's film feature a cold spell caused by the slowing of ocean circulation back before anyone had changed the term "global warming" to "climate change". It's what you expect to happen in the mid latitudes. (Of course, you also get more intense hot spells in summer, but those probably weren't deemed as photogenic.)
OTOH, please note that so far the ocean currents haven't slowed, just the jet stream. But the ocean currents *have* showed signs of weakening. Possibly because Greenland is dumping loads of fresh water on top of the heavier salt water. This will *probably* have more effect on Europe than on the US, though.
That said, I am not a climatologist. And I don't run climate models. These are opinions I've picked up by reading popular science magazines like Scientific American...but they are my extrapolations from many of those articles.
You do need to remember that before he linked up with Hitler Mussolini was considered one of the good guys. And he only linked up with Hitler because nobody else was standing up against him.
That said, he was a dictator. But he wasn't a particularly vile one before he came under the Nazi influence. His main intent was to "Make Italy great again!". If Britain or France had stood up to Hitler before the Czechoslovakia election, he would probably have been counted among the allies.
OTOH, modern Italy would never have been born. And Mussolini did have militaristic desires. (Look up the invasion of Ethiopia.) He just wasn't very good at them.
So when you say "Look how well that worked out before", you need to look at the larger picture than just WWII, because that wasn't inevitable. It was caused because France and England wanted German fascism to combat Russian communism...but Hitler had bigger plans.
I think you are being blinded by your desires to political reality.
For Trump to be impeached the House, which is dominated by Republicans, would need to vote a bill of impeachment. For the impeachment to be executed the Senate, which is also dominated by Republicans, would need to vote for a conviction.
Despite what it says in the constitution about grounds, political reality is that the House can impeach on any grounds that it finds suitable, and the Senate can convict if it feels like it.
So far, no president has actually had the sentence of impeachment executed. It would be interesting, though not pleasant, to consider what might happen if the Senate voted to confirm the impeachment, and the President contested the validity of the vote. There is no explicit provision for enforcing the sentence. It those working for the executive branch were to be personally loyal to the President despite the vote of impeachment there doesn't appear to be any way to make the impeachment stick. It would come down to who the army was loyal to.
Sorry, that hypothesis doesn't fly. It may be harder to debug C or assembler than C++, but most other languages provide more usable debugging facilities than does C++, and most of them are easier to write unit tests for. The unit testing for C++ is basically an add-on. Even assert statements in C++ are crippled, unless you use an extension. (Assert statements should include an optional message that is printed with the error, and which can dump variables of interest in a formatted way.)
So C++ is more difficult to debug. There *are* more external tools that you can use to do the debugging, but that's because they are needed.
The first problem I had with Ada was that strings of different lengths were of different types. That could be solved with bounded or unbounded strings, but then you couldn't compare against literals. There were other problems, and there were always ways around them, but I had to used unchecked conversion too often for things that were perfectly safe. The entire thing was a mess due to over-concern with simple type conversions. Dynamically allocated storage is also a mess, though, given the original ideas, quite reasonable. Follow an Ada program through a simple addition of a node to a tree and you should see what I mean.
Another part of the problem was that the optional annexes couldn't be used as they usually weren't implemented. It's great to not have to load them if you don't need them, it's not so great if they aren't available when you need them.
There are lots of other minor design flaws. In a way you can forgive them, because this was being designed at about the same time as C++ was being written, and, unfortunately, it was designed by a committee. So the kitchen sink got thrown in. But they did design everything in a way that enabled maximal security...even where it didn't make any sense. (Why can't I compare two literal strings of different lengths? They've probably changed that by now, though.)
I *HAD* the goddamn coverage I needed. I just needed to stand there bleeding while I tried to convince some damn office worker several states away that I had an injury that needed the coverage, and he appeared to be under pressure to decide that I didn't need it without knowing anything about the case. The insurance did, eventually, pay, but by god I'm not going to do anything to support that kind of system when I have a reasonable choice otherwise.
Did you ever try negotiating payment with a broken arm?
Even with good insurance I once had to negotiate coverage with a broken thumb, because the insurance didn't want to pay. Ever since then I haven't been happy to trust insurance companies, or any other group that has an incentive to not cover your medical problem. And I do recognize that such is essential. Your proposed solution seems to mean the in every medical situation you need to negotiate coverage, which is just about the worst possible approach.
Sorry, but a personal claim by "Anonymous Coward" isn't very convincing. Make a reasoned argument, or cite an external reference, and I'll take you seriously, but for personal history arguments to carry any weight the person needs to have a track record.
The suggestion to "research it on your own" is reasonable. Of course, you run across the problem of what sources to trust.
A different comment said the study quoted didn't say what the article implied.
From other sources I have heard that the older varieties of stent tended to become quickly clogged with blood clots, but that the current versions incorporate a treated surface that prevents that from happening. Anecdotally, I have friends who have had stents, and their conditions have improved markedly...from unable to walk and incoherent to able to walk up a hill and nearly as reasonable as before their problem. Well, that's been a decade or so, and he's failing again. But I'd hardly say that the stent wasn't effective, and I'm not sure that's the problem now anyway. (He's currently in his mid-to-late 80's.)
I would have said "unwieldy". But it already was unwieldy.
It's probably true that if I invested sufficient time C++ would be a complete language. But even were I an expert it would take me a long time to program anything, because there are too many fiddly details.
Let me modify that. C++ is certainly a complete language, but so is C and so is assembler. None of them make it easy to turn well defined ideas into code. I prefer C++ over C because of a very few choices made long ago, e.g. the inclusion of references. Most of the changes since that are not helpful. They don't add functionality, and they don't reduce the number of additional libraries that are needed.
That said, I understand that they want to avoid breaking changes. But they need changes that require breaking things. This means they need to create a minimally different new language. Unfortunately, most such attempts fail. There are multiple different populations of users with different needs, and for one language to serve them all well is impossible. E.g., my use case would be served by a key-value database being built into the language. This can currently be handled by an attached library, but because the library isn't a part of the language, everything needs to be converted to strings before being saved and converted from strings while being restored. But there's no good method of conversion. (It's actually a simple problem in my case, but the general problem isn't soluble by a separate library.) I also need a kind of simplified Actor variation of concurrency. (I don't need the full Actor model.) This is all *doable*, but C++ sure doesn't make it easy, even though the ideas are well-defined.
It looks moderately interesting, but I'm not interested in a language that doesn't allow concurrent execution, and I didn't notice anything about how you handle Unicode. (utf8 is fine, even utf32 internally, but forget about utf16...except where required by foreign function calls.)
The "scan the stack backwards" might solve the problem that caused polish notation to be disliked, but the explanation needs to be clearer.
Also, you need to do something to make the name more searchable (though it sure is an improvement over some language names). I kept getting a Canadian radio station, and I *never* search for radio stations.
All that said, it looks interesting, but I currently need a language that's complete and workable. Good luck with your recruiting.
There was a collection of artistic works by an "AI" printed a few decades ago called "The Policeman's Beard is Half-constructed". It *had* been edited to include only the works that seemed interesting, but I can guarantee that this is also normal for human poets and authors.
The poems didn't emphasize rhyme or even assonance, but this is also common among modern poems.
Personally, I think that a current effort by a major AI would be much better, but it would definitely depend on the critical skills of the trainer. There are reasons that Kipling hasn't really been eclipsed by any more modern poet, and part of it is the way he uses rhyme and assonance, and another part is the emotive handling. Check out the ballad of Boh Da Thone http://www.kiplingsociety.co.u.... Robert Service https://en.wikipedia.org/wiki/... is also good. But note that they don't just use poetic tricks, they also tell a human story. This is going to be difficult for an AI. The poetic tricks are reasonably simple, but understanding the kind of story that people like to hear is more difficult.
The part that didn't sound incoherent sounded wrong. The volume of a sphere grows more quickly with the radius than the surface area. Of course, it was a metaphor, and one can imagine rules such that the "thickness of the rind" would increase fast enough that the volume of the rind would grow faster than the core...but they sure weren't mentioned.
Not that I think the metaphor is in any way useful. It's not as if any one person knows all the stuff "in the core".
If you say so. I never knowingly saw any image of it. I was operating off an article about how annoyed the artist was with the misappropriation of his work.
If it's an actual AI, the bias is not caused by the programming, but rather by the training data set. As to whether it would be more accurate, that's a separate problem from either consistent or fair. My assertion was that it could be more consistent, but that this wouldn't imply it would be fairer. Accurate is a yet more difficult problem when dealing with people, and it's one that human level intelligences have yet to solve.
To be fair, "fairness" is only a solved problem in the opinion of the person doing the judging. But accuracy is even harder, as most people realize that they don't really know what someone else meant, and sometimes realize that they didn't know what they, themselves, meant.
Thank you. An excellent example of re-framing the context.
I'm not saying that Facebook is making a good faith effort to solve the problem. I've never looked at it, so I have no idea. I'm saying that the problem as stated is a hard problem. It's easy to, say, ban certain particular words, but that doesn't accomplish very much.
I'm not sure that the problem as stated could be addressed by anything much short of a human equivalent AI, and even that would only allow some particular set of standards to be applied uniformly. It sure couldn't guarantee that the standards were fair.
As an example consider the text "You with a donkey's member!" This is apparently a violently abusive comment, but that depends on context that isn't present. I'm sure I could come up with a context where that would be encouragement. And every single word in that sentence is perfectly harmless. Or what about "Pepe the frog"? That was intended to be a humorous children's cartoon character....but it didn't stay that way, much to the annoyance of the creator.
That said, the evidence seems to support the assertion that Facebook encourages hateful posts, and is more reluctant to censor nazi-ish posts than those with an opposing message. Again, I have no direct evidence for this as I never visit that site, and am relying on material published by others.
The problem is, the Democrats, while not quite as blatantly awful, are nothing to cheer. I support my Representative, and I don't consider one of my Senators better dead. They are all Democrats, and I only feel moderately represented by one of them. That the Republican choices would have been worse is scant reason for liking the situation...and in the case of one of the Senators I'm not sure even that's true.
In this case it often is. The crime is tax evasion. Hotels frequently need to pay a tax on rooms rented, and adhere to various safety regulations that Airbnb ignores. They've been ruled an illegal business in a couple of places because of this.
It stood for extra large, and it was a Mac modified to hold more memory. But that's decades ago, and it might have been somebody's custom modification rather than an official version. It was, however, available at the same place I bought the official Mac versions. (It's even possible that somebody just thought "fat Mac" was inelegant.) I used so many different versions it's hard to keep track of all of them.
I never did any of the modifications, and the repairs were done at the same place that sold the Macs and the custom modified Macs...so I may not have kept them straight. Or, it being decades ago, I may just have scrambled my memories.