Why did I do it like this ? I'm lazy. I don't want to re-rip my collection. Ever. As disks get bigger, at some point ~7mb 256k mp3s are going to look tiny. On a good system ($2000 amp + speakers) mp3s DO NOT sound as good as the original CDs. My WAVs however, sound identical.
I've been using plain.wav files (exact same directory structure) and ignoring mp3's and other lossy and/or drm-laden formats for years; glad to see i'm not the only one, most of my friends thought i was being silly and wasteful. The trend to larger storage seemed pretty obvious many years ago, and committing to degraded information or usability just to keep things as compact as possible just felt short-sighted.
On the other hand, the older i get the more i think worrying about lossiness in audio is moot.:(
There is also an XML file with all the album info in it including the CDDB query ID in case I need to requery CDDB without needing the actual CD. Additionally both.m3u and.pls playlist files are auto created.
I wrote a custom (java) application that aggregates all the music info together along with various custom categories on a per-song level. I use that to edit, display and generate playlists by filtering and stochastically combining different categories. All data i keep in a basic.csv file so i can also manipulate it with simple text processing (awk/sed etc) and/or in spreadsheets as well. Entering that incrementally over the years has been painless, and is just about perfectly-tuned to my needs at this point. That also unfortunately means that the application interface (gui) is a bit of an idiosyncratic hack, so i haven't released it, but maybe someday...
Finally a album cover photo (from Amazons XML feed) is auto downloaded and placed into the folder as folder.jpg
That's the only part i haven't done. Couldn't really see the need, given that mostly when listening to music i'm either not physically near my computer, or if i am i am absorbed in something else. Not to mention finding album covers is not an entirely trivial effort (cd covers are easy to scan, but scanning takes time. Vinyl i've ripped is more of a pain---too large to scan easily, stuff on the internet is low-resolution, and seems to mainly exist on pretty crappy websites i'd rather avoid).
I have 4 or 5 user IDs across a multitude of systems in my company and can never remember the ones I use about once a month or so. Typically I end up having to request a password reset for those systems.
At my former employment i had at least as many, with the same problem, and much the same solution. Several of my coworkers kept the usual piece of paper in their desk with passwords, and many just kept text files on the system they used most often.
I complained at one point and was told i should just use the same password everywhere. Sadly, every system had different password requirements, expired at different times, and several had different allowable characters (one was case-insensitive, others had different non-alphanumeric symbols missing or required)---just keeping track of all the systems required a list. I used to get password expiration notices from systems i'd never even logged into.
A lot of co-workers just rotate through all 8 or 12 iterations of passwords and then restore their original password.
That was also a solution i used a few times out of frustration. Problem was that around iteration 7 or so i'd lose track and forget some subtle detail of iteration 6, and end up locked out of the system, requiring a reset anyway.
The end result was:
I had a paper list of several old, some current passwords in my desk drawer
I gave up on choosing good passwords; abcdef01, increment number as required worked for many, some required rotating the abcdef through a few systematic, obvious and easy-to-guess variations (month names, colours, slightly mangled worked well)
Most passwords would eventually require a reset, resulting in a new password to be sent to my manager, then to me, all in clear text through email
What they protecting so obsessively through password schemes i'll never know. Guess it worked though, i often couldn't get into the system i needed to get work done.
Well, aren't you feeling smug. Most people are just optimizing in different ways, and devoting their optimization effort/budget to things you may not value equally or are unaware of.
To show they are out of order they got a big red sticker at eye height. Now guess how many people see it?
Not surprising. I expect that most subway/tube/metro users follow the same path over and over; why expend effort looking for out-of-order doors when they are relatively rare? Anomalies have to be fairly large/extreme to be noticeable enough to shock someone out of their well-followed routine. Many program optimizations are based on a similar principle--optimize the common case at the expense of the uncommon.
You then got the people that stand behind them and then still go towards the broken door and press the button because obviously the person in front of them didn't press it right.
Or, perhaps, they were not clearly paying attention (being on auto-pilot) to the activities of the person in front of them, and thus on an effort/reward basis felt the small extra cost to verify the flaw was more worthwhile than just assuming the door was broken and moving on---an extra button press is a small effort compared with changing doors. Bad judgement in some cases, but it's not that unusual for people in front of you to change their minds, and sometimes they may fail to operate something correctly or have a lower threshold for effectively doing so, and some buttons are also finicky. People are also sometimes curious as to what the flaw might be---maybe it's even something they can correct, but the person in front of them judged it not worth the effort or out of their skill-set.
One person will get up, knuckles dragging across the floor, make their way against the traffic to the out of order door.
Resource 2 is overloaded, while resource 1 is apparently not being utilized. Sometimes the situations arise because most people opt for maximum convenience, rather than because resource 1 is unavailable. It's a small chance, but maybe worth an investigation if the queue for resource 2 is excessive, and/or one has a strong dislike for fighting over a limited resource.
I buy 100 shares, which initially dip. Now I've taken a small loss on something I expect to shoot way up within the next few days... Would I sell? Hell no! Now, at around a 10% dip I might get rather worried...
You really need to pay more attention to how this is supposed to work: buy low, sell high. It's really not that complex.
seriously, what if in the ocean the waterflow is spinning very hard itself under certain conditions, wouldn't that be a possible explanation for the disappearances in the Bermuda Triangle?
Yes, that's right. A suitably airplane-shaped hole would indeed allow an airplane to fall to the bottom of the ocean without getting wet, nicely and logically accounting for its sudden and complete disappearance. Similarly, holes isomorphic to boats and drowning people would account for those inexplicable losses.
Oh, wait, Bermuda triangle---you probably meant a triangular hole. No, sorry, that's just stupid.
I predict a sudden rise in the sales of noise-cancelling headphones after VoIP becomes commonplace on airplanes.
I stopped using those on airplanes. While they do reasonably well at eliminating the very loud, dull roar of the airplane they leave behind the more high-pitched human voices, including people talking and babies howling.
I guess it's all a matter of what one considers "noise" on an airplane. There's also the attendant feeling of greater ear-pressure, and the ache of wearing headphones continuosly for several hours.
The problem with these potential yak-fests by seatmates and by nearby or loud passengers is being unable to escape from them.
Indeed, and with airplanes being very noisy environments people will talk even louder. The key then is to force them to have/want to escape from you. The return of the BFR/boombox? Inexplicable bouts of Tourettes? Ah, to be able to fart at will...
what's the big deal? It's like SimEarth only you drive the bugs now? I don't get it.
The big deal is content-creation. In most multiplayer games that's the duty of the game provider, and at their expense. Here you do it, and they sell it! Everybody wins!
Individual inventors often don't have the resources to patent something as soon as they come up with it, and by the time they do, several companies will have filed several patents covering the same ground.
If it's a race to get to the patent office then perhaps the idea is just not that novel and shouldn't be patented in the first place.
Indeed; helping increase the volume of bad patents doesn't seem like a useful goal. The main problem with software (and other more abstract) patents is not the slow process for granting them, it's the fact that they keep granting idiotic patents that are very obvious to anyone even remotely skilled in the art. Searching and validating beforehand may be expensive, but is a lot cheaper than a court fight....at least in an overall sense.
On the other hand, if their intention is to fuel outrage in blogs and community websites (like/.), or provide even more material for people to make fun of them, this will be a great success.
I don't understand why nobody seems to sell capsule flights, particularly for long hauls.
Lots of reasons it turns out:
Some people who lie down for a long time will have some difficulty standing up after. That would slow down deplaning, especially in an emergency.
For equivalent/maximal space usage, capsules would have to be stacked. Then you need a certain agility to access upper ones, which limits who can have which capsule.
Oh good, your capsule is the one previously occupied for 7 hours by someone with serious b.o., perfume, or even vertigo issues. Cleaning a capsule is a lot more work than cleaning a seat.
Capsules take up more space. Think about it---your average coach seat does not have room for you to stand up, and is fairly minimal in surroundings in other dimensions too. A capsule that wouldn't be obscenely claustrophobic would definitely require more space. Keep in mind that a few inches of extra space is the major difference between business and coach prices.
It's hard to eat/drink laying down. While the whole point might be to facilitate sleeping, not everyone sleeps well on planes, so inevitably they'd be consuming food/drink.
People wouldn't stay in their capsules. They'd keep coming out to stretch etc; you'd need to provide some kind of mingling, common space, which would require more space per person.
On the other hand, you could say that combinations of keys count for sending signals. This assumes the user can depressed keys instantly but this means that for each key, we've doubled the amount of signals we can send. So, the smallest power of 2 above 146 is 256 or 2^8.
Why are we stuck with binary keys only? There's also:
Speed of key press
Depth of key press
Duration key is held
Right now most people use duration for repeat, but using it to map a key to shifted form might be ok...i also rather like the idea of hitting the keys very hard to generate caps. Either of those only save a shift key or two, but even that would free up shift for other uses....could also support more character formatting than just case---mode keys to make a character bold, italic, etc...
...big boys like Dell and HP, small guys like Ram's PC Shop. Guess what? Prices have fallen even against inflation.....In the automotive world, we have heavy regulations...Car prices have risen, faster than inflation.
Comparing PTTs with the PC or auto industries is highly misleading. PC prices are under continuous reduction because of the extreme rate of progress (perhaps now slowing down..?)---actions by dell/hp etc are more likely associated with driving out competition relatively early in the pc market development than an indication of long-term positive effects of deregulation. Vehicle manufacture is heavily regulated because of the great consequences to personal safety if not.
In the soda world, we have almost no regulations (except for some USDA/FDA ones).
Yes, and it's worked out really well. We basically have 2 dominant manufacturers with an almost indistinguishable product, and the good deals for soda you crow about merely underline your gross ignorance of the costs involved.
In the medicine world, we have excessive regulations, and prices have climbed beyond inflation.
Perhaps one day you'll live in a place where doctors are not regulated and medicine is correspondingly cheap. If you survive your first blood-letting let us know how it goes.
In the clothing world, we have few regulations (some tariffs on cotton and other materials). I can buy a nice, quality hoodie for US$10 at H&M. A few years back they were over US$50 at the mall.
You're making your troll too obvious. No one could be that confused.
Tell me again how regulations help and anarchy hurts?
In general they help because they allow you to maintain your ignorance without getting hurt or badly ripped off (YMMV). They hurt because those of us who've considered the issue in any depth are pained by the numerous shallow assessments offered as informed opinions on public fora.
In other words turf the problem to the end-user? This is an improvement?
It's a solution that involves neither os-bundling nor supplying all components in every piece of software.
This is a matter of enforcing a product dependency structure: either the user already has all the dependencies (built into the os), or all the dependencies come with the product, or they are at least informed of what dependencies they must resolve. From a consumer convenience standpoint the former two are preferable, but convenience has its price, either in producing software monocultures (os-bundled software) or in redundancy and unnecessary cost (i already own x, yet my new software comes with x).
Consumers are well capable of dealing with external dependencies, from "batteries not included" to needing a cd-player to play their new cds. Note how these systems are best enabled by open/accepted interface standards.
True, I *could* query the system using *open* standards and detect and use a pre-existing player. But, what if there isn't one? To cover that corner case (real developers cover corner cases, unlike OSS where they often blow them off) I would need to also provide a media player with my installer.
Umm. Put on the box under requirements: "Media player capable of playing the following formats..."?
If they're like most, they have a back room with a stack of overhead projection tablets that nobody uses anymore because they're 480x640.
Was it really the resolution that was the downfall of those devices? Presentations using lcd overhead projection, iirc, were self-limiting in that eventually the image would start to "melt" as the lcd panel became overheated by the projection bulb. It would eventually recover, but the heat/cool cycle was such that at least in my experience full-class presentations were not possible.
But i suppose it might still be good for watching the matrix, if you timed it right.
can someone please give a brief explanation of what the big deal is with lunar dust?
It's very very fine dust; think of how plaster dust manages to get everywhere, even clogging vacuums.
It's also apparently quite sharp (what with the general lack of erosion up there), and thus it manages to not only get everywhere, but also be irritating when it does. For the google-impaired.
so the factory calls them first. No answer -- it's an hour earlier in Ontario, so the warehouse office is closed. Guess who gets the order?
Oh my god, you're right---and that scenario must happen at least, what 1--2 times a year?! 10's, if not 20's of dollars could be lost!! Surely an excellent reason for millions of people to change their lifestyles!
Canadians take note: you risk the complete loss of the incredibly lucrative industry that relies on emergency, last-minute blivet purchasing by morons who don't understand time zones and can't wait a hour!
In Canada, currently the province of Saskatchewan doesn't observe it.
Saskatchewan does observe DST. However, people there do not have to reset their clocks becausethe province also changes time zones at the same time, with a net 0 effect.
In fact, that sounds like an excellent model for Ontario to follow too. Increase the length of DST to make the americans happy, but switch time zones for the DST extension so nothing really actually differs from the current system: everybody's happy!
But all the examples in the article aren't tested in practise. Maybe the escape analysis the author describes works as advertised. But without actually testing and analysing real code produced and without actual benchmarks the article doesn't proof a thing
The author is giving you a high-level, greatly edited view of some of the major optimization techniques in use today. The original academic/technical papers on which that is based demonstrate/measure their techniques on various benchmarks.
One thing the author seems to forget is that you would at least need a fallback mechanism when a method is overridden in a subclass
This is well-known. Code-patching, on-stack replacement, pre-existence etc are used in production jits to recover from that kind of thing. This paper has a decent summary of current approaches:
Slashdot Mirror
According to the article, yes it is more effective than a hammer.
What about a magnetic hammer?
Why did I do it like this ? I'm lazy. I don't want to re-rip my collection. Ever. As disks get bigger, at some point ~7mb 256k mp3s are going to look tiny. On a good system ($2000 amp + speakers) mp3s DO NOT sound as good as the original CDs. My WAVs however, sound identical.
.wav files (exact same directory structure) and ignoring mp3's and other lossy and/or drm-laden formats for years; glad to see i'm not the only one, most of my friends thought i was being silly and wasteful. The trend to larger storage seemed pretty obvious many years ago, and committing to degraded information or usability just to keep things as compact as possible just felt short-sighted.
:(
.m3u and .pls playlist files are auto created.
.csv file so i can also manipulate it with simple text processing (awk/sed etc) and/or in spreadsheets as well. Entering that incrementally over the years has been painless, and is just about perfectly-tuned to my needs at this point. That also unfortunately means that the application interface (gui) is a bit of an idiosyncratic hack, so i haven't released it, but maybe someday...
I've been using plain
On the other hand, the older i get the more i think worrying about lossiness in audio is moot.
There is also an XML file with all the album info in it including the CDDB query ID in case I need to requery CDDB without needing the actual CD. Additionally both
I wrote a custom (java) application that aggregates all the music info together along with various custom categories on a per-song level. I use that to edit, display and generate playlists by filtering and stochastically combining different categories. All data i keep in a basic
Finally a album cover photo (from Amazons XML feed) is auto downloaded and placed into the folder as folder.jpg
That's the only part i haven't done. Couldn't really see the need, given that mostly when listening to music i'm either not physically near my computer, or if i am i am absorbed in something else. Not to mention finding album covers is not an entirely trivial effort (cd covers are easy to scan, but scanning takes time. Vinyl i've ripped is more of a pain---too large to scan easily, stuff on the internet is low-resolution, and seems to mainly exist on pretty crappy websites i'd rather avoid).
At my former employment i had at least as many, with the same problem, and much the same solution. Several of my coworkers kept the usual piece of paper in their desk with passwords, and many just kept text files on the system they used most often.
I complained at one point and was told i should just use the same password everywhere. Sadly, every system had different password requirements, expired at different times, and several had different allowable characters (one was case-insensitive, others had different non-alphanumeric symbols missing or required)---just keeping track of all the systems required a list. I used to get password expiration notices from systems i'd never even logged into.
A lot of co-workers just rotate through all 8 or 12 iterations of passwords and then restore their original password.
That was also a solution i used a few times out of frustration. Problem was that around iteration 7 or so i'd lose track and forget some subtle detail of iteration 6, and end up locked out of the system, requiring a reset anyway.
The end result was:
What they protecting so obsessively through password schemes i'll never know. Guess it worked though, i often couldn't get into the system i needed to get work done.
People are idiots.
Well, aren't you feeling smug. Most people are just optimizing in different ways, and devoting their optimization effort/budget to things you may not value equally or are unaware of.
To show they are out of order they got a big red sticker at eye height. Now guess how many people see it?
Not surprising. I expect that most subway/tube/metro users follow the same path over and over; why expend effort looking for out-of-order doors when they are relatively rare? Anomalies have to be fairly large/extreme to be noticeable enough to shock someone out of their well-followed routine. Many program optimizations are based on a similar principle--optimize the common case at the expense of the uncommon.
You then got the people that stand behind them and then still go towards the broken door and press the button because obviously the person in front of them didn't press it right.
Or, perhaps, they were not clearly paying attention (being on auto-pilot) to the activities of the person in front of them, and thus on an effort/reward basis felt the small extra cost to verify the flaw was more worthwhile than just assuming the door was broken and moving on---an extra button press is a small effort compared with changing doors. Bad judgement in some cases, but it's not that unusual for people in front of you to change their minds, and sometimes they may fail to operate something correctly or have a lower threshold for effectively doing so, and some buttons are also finicky. People are also sometimes curious as to what the flaw might be---maybe it's even something they can correct, but the person in front of them judged it not worth the effort or out of their skill-set.
One person will get up, knuckles dragging across the floor, make their way against the traffic to the out of order door.
Resource 2 is overloaded, while resource 1 is apparently not being utilized. Sometimes the situations arise because most people opt for maximum convenience, rather than because resource 1 is unavailable. It's a small chance, but maybe worth an investigation if the queue for resource 2 is excessive, and/or one has a strong dislike for fighting over a limited resource.
I buy 100 shares, which initially dip. Now I've taken a small loss on something I expect to shoot way up within the next few days... Would I sell? Hell no! Now, at around a 10% dip I might get rather worried...
You really need to pay more attention to how this is supposed to work: buy low, sell high. It's really not that complex.
seriously, what if in the ocean the waterflow is spinning very hard itself under certain conditions, wouldn't that be a possible explanation for the disappearances in the Bermuda Triangle?
Yes, that's right. A suitably airplane-shaped hole would indeed allow an airplane to fall to the bottom of the ocean without getting wet, nicely and logically accounting for its sudden and complete disappearance. Similarly, holes isomorphic to boats and drowning people would account for those inexplicable losses.
Oh, wait, Bermuda triangle---you probably meant a triangular hole. No, sorry, that's just stupid.
I predict a sudden rise in the sales of noise-cancelling headphones after VoIP becomes commonplace on airplanes.
I stopped using those on airplanes. While they do reasonably well at eliminating the very loud, dull roar of the airplane they leave behind the more high-pitched human voices, including people talking and babies howling.
I guess it's all a matter of what one considers "noise" on an airplane. There's also the attendant feeling of greater ear-pressure, and the ache of wearing headphones continuosly for several hours.
The problem with these potential yak-fests by seatmates and by nearby or loud passengers is being unable to escape from them.
Indeed, and with airplanes being very noisy environments people will talk even louder. The key then is to force them to have/want to escape from you. The return of the BFR/boombox? Inexplicable bouts of Tourettes? Ah, to be able to fart at will...
what's the big deal? It's like SimEarth only you drive the bugs now? I don't get it.
The big deal is content-creation. In most multiplayer games that's the duty of the game provider, and at their expense. Here you do it, and they sell it! Everybody wins!
Individual inventors often don't have the resources to patent something as soon as they come up with it, and by the time they do, several companies will have filed several patents covering the same ground.
If it's a race to get to the patent office then perhaps the idea is just not that novel and shouldn't be patented in the first place.
We already have enough BS patents out there...
/.), or provide even more material for people to make fun of them, this will be a great success.
Indeed; helping increase the volume of bad patents doesn't seem like a useful goal. The main problem with software (and other more abstract) patents is not the slow process for granting them, it's the fact that they keep granting idiotic patents that are very obvious to anyone even remotely skilled in the art. Searching and validating beforehand may be expensive, but is a lot cheaper than a court fight....at least in an overall sense.
On the other hand, if their intention is to fuel outrage in blogs and community websites (like
Lots of reasons it turns out:
bee-hive...sugar cane...Otherwise, he was out of luck. Nothing sweet in the diet. Nothing at all.
No sugar beets? Stevia? Fruit? Berries? What a crappy place to live.
Why are we stuck with binary keys only? There's also:
Right now most people use duration for repeat, but using it to map a key to shifted form might be ok...i also rather like the idea of hitting the keys very hard to generate caps. Either of those only save a shift key or two, but even that would free up shift for other uses....could also support more character formatting than just case---mode keys to make a character bold, italic, etc...
...big boys like Dell and HP, small guys like Ram's PC Shop. Guess what? Prices have fallen even against inflation.....In the automotive world, we have heavy regulations...Car prices have risen, faster than inflation.
Comparing PTTs with the PC or auto industries is highly misleading. PC prices are under continuous reduction because of the extreme rate of progress (perhaps now slowing down..?)---actions by dell/hp etc are more likely associated with driving out competition relatively early in the pc market development than an indication of long-term positive effects of deregulation. Vehicle manufacture is heavily regulated because of the great consequences to personal safety if not.
In the soda world, we have almost no regulations (except for some USDA/FDA ones).
Yes, and it's worked out really well. We basically have 2 dominant manufacturers with an almost indistinguishable product, and the good deals for soda you crow about merely underline your gross ignorance of the costs involved.
In the medicine world, we have excessive regulations, and prices have climbed beyond inflation.
Perhaps one day you'll live in a place where doctors are not regulated and medicine is correspondingly cheap. If you survive your first blood-letting let us know how it goes.
In the clothing world, we have few regulations (some tariffs on cotton and other materials). I can buy a nice, quality hoodie for US$10 at H&M. A few years back they were over US$50 at the mall.
You're making your troll too obvious. No one could be that confused.
Tell me again how regulations help and anarchy hurts?
In general they help because they allow you to maintain your ignorance without getting hurt or badly ripped off (YMMV). They hurt because those of us who've considered the issue in any depth are pained by the numerous shallow assessments offered as informed opinions on public fora.
In other words turf the problem to the end-user? This is an improvement?
It's a solution that involves neither os-bundling nor supplying all components in every piece of software.
This is a matter of enforcing a product dependency structure: either the user already has all the dependencies (built into the os), or all the dependencies come with the product, or they are at least informed of what dependencies they must resolve. From a consumer convenience standpoint the former two are preferable, but convenience has its price, either in producing software monocultures (os-bundled software) or in redundancy and unnecessary cost (i already own x, yet my new software comes with x).
Consumers are well capable of dealing with external dependencies, from "batteries not included" to needing a cd-player to play their new cds. Note how these systems are best enabled by open/accepted interface standards.
True, I *could* query the system using *open* standards and detect and use a pre-existing player. But, what if there isn't one? To cover that corner case (real developers cover corner cases, unlike OSS where they often blow them off) I would need to also provide a media player with my installer.
..."?
Umm. Put on the box under requirements: "Media player capable of playing the following formats
If they're like most, they have a back room with a stack of overhead projection tablets that nobody uses anymore because they're 480x640.
Was it really the resolution that was the downfall of those devices? Presentations using lcd overhead projection, iirc, were self-limiting in that eventually the image would start to "melt" as the lcd panel became overheated by the projection bulb. It would eventually recover, but the heat/cool cycle was such that at least in my experience full-class presentations were not possible.
But i suppose it might still be good for watching the matrix, if you timed it right.
There are two ways this can make them money:
You missed one:
3) it's effectively a forced registration. Customer lists are an important asset for both internal uses and for sale to other companies.
can someone please give a brief explanation of what the big deal is with lunar dust?
It's very very fine dust; think of how plaster dust manages to get everywhere, even clogging vacuums.
It's also apparently quite sharp (what with the general lack of erosion up there), and thus it manages to not only get everywhere, but also be irritating when it does. For the google-impaired.
so the factory calls them first. No answer -- it's an hour earlier in Ontario, so the warehouse office is closed. Guess who gets the order?
Oh my god, you're right---and that scenario must happen at least, what 1--2 times a year?! 10's, if not 20's of dollars could be lost!! Surely an excellent reason for millions of people to change their lifestyles!
Canadians take note: you risk the complete loss of the incredibly lucrative industry that relies on emergency, last-minute blivet purchasing by morons who don't understand time zones and can't wait a hour!
In Canada, currently the province of Saskatchewan doesn't observe it.
Saskatchewan does observe DST. However, people there do not have to reset their clocks becausethe province also changes time zones at the same time, with a net 0 effect.
In fact, that sounds like an excellent model for Ontario to follow too. Increase the length of DST to make the americans happy, but switch time zones for the DST extension so nothing really actually differs from the current system: everybody's happy!
What if you happen to get your receipt ten seconds before the reset and don't make it to the restroom in time?
Allow the current code and last code to have access: everyone is guaranteed a minimum access time.
But all the examples in the article aren't tested in practise. Maybe the escape analysis the author describes works as advertised. But without actually testing and analysing real code produced and without actual benchmarks the article doesn't proof a thing
The author is giving you a high-level, greatly edited view of some of the major optimization techniques in use today. The original academic/technical papers on which that is based demonstrate/measure their techniques on various benchmarks.
One thing the author seems to forget is that you would at least need a fallback mechanism when a method is overridden in a subclass
This is well-known. Code-patching, on-stack replacement, pre-existence etc are used in production jits to recover from that kind of thing. This paper has a decent summary of current approaches:
Google Maps is good, but would I rather have it running on my machine? Damn right I would!
No, i don't think you would. You would probably find it a bit of a resource hog.