Unfortunately not. The parent is completely correct that using the term "module" or "modularity" in Java 9 is completely misleading. It has absolutely nothing to do with any meaning of "module" in any other programming language.
I'm actually astonished Java made it this far without such a feature! There is no language / system on the planet that has such a feature. Except Java 9.
class path system Which works exactly like the LD_LIBRARY_PATH for any compiled language on Unix or the equivalent on Windows.
Because higher frequencies mean higher temperatures. That is nonsense. Modern CPUs need more power, hence they produce more heat.
Remember when CPUs didn't need active cooling? Yes, my 6502, got about 110C hot. But the transistors still worked at that temperature. Modern CPUs have so small transistors that they can not work at that temperature anymore as the electrons would simply jump several transistors far through the substrate.
I don't know how quickly stuff will change, but we have to fronts: first new main memory technologies, like magnetic based RAMs. And secondly, different operation systems to exploit that. Better/Different operating systems is probably the slow part.
Magnetic based RAM (Memristor, see: https://en.wikipedia.org/wiki/... magnetic based main memory that holds its memory when switched off) will blur the distinction between RAM and "hard drive" or any external memory.
Bottom line that means with a new operation system, probably more object oriented, you would no longer need to hold "files" on a "hard drive" but could keep the object a life in memory.
For that you would need OS support in case multiple "editors" are working on that "file". Bottom line a program would probably no longer work on multiple files, but would spawn a new instance/process for each file (what e.g. modern web browsers already do) and such a "file" would be mapped into the address space of the process.
In the long run we had something like text files, binary files (with structure like a GiF, but no code "attached") and "object oriented" files. Consider them like streamed Java objects containing also the class files for the relevant objects.
I basically imagine a "graph data base" system whee nodes are files and files can be executable, just like right now, but everything in main memory, never paged out, because main memory and backing storage is the same thing.
This thing would be a revolution for computers, external file systems only used for backup or data transfer.
On top of that we have processors that have fast reprogrammable FPGA like structures included. In the very long run we also have reprogrammable RAM with data flow processing capacity in the RAM directly. But I guess we are a long way to go to get this mainstream. Perhaps in GPUs or "neural net co processors".
A big contra point to my ideas above is optical computing and probably quantum computing. The later is in its infancy, but I would not wonder if there is a sudden big progress.
Optical computing we have since a while, but not scaled down to compete with anything we do right now. But: there are ideas of holographic RAMs that like the Memsistors keep their memory when switched off (and much much longer and more reliable). There are prototypes where data lines on the boards are replaced with optical fibres. Super expensive but it makes some access paths quicker, because it avoids interferences. There are some IBM research things (and probably HP too, but I forgot) where they are able to connect ordinary transistors with optic ones and have hybrid chips.
Anyway, I think it is time to rethink how the "whole thing" is working. It will basically still be a von Neumann architecture but the coupling of processor, most of the main RAM, and storage will be completely different. And that requires rethinking the OS.
iOS more or less works like that already, from an outside point of view. Inside it is in ordinary Unix, ofc. with a conventional file system.
I don't care what the wikipedia article says and I'm to lazy to read it.
Silicium cells _don't_ contain heavy metals. They don't contain _any metal at all_.
The only metals are outside in the _panel_ and _wiring_
So: super easy to recycle. And if you had physics in school: you damn would know that.
Then we have cells with a mirror on the back side, which again is on the outside! That is in most cases silver, and silver is counted as a precious metal, not as a heavy metal. Even chemically it is a heavy metal. Perhaps you meant that...
So: all metals in a panel can be super easy recycled, they are not a threat on a land fill. And bottom line I assume that the whole cell gets recycled as the silicium is besides the fact that it is doted, still super pure!
We actually do that in Germany on an "experimental scale" and feed the hydrogen in a very low percentage into the ordinary gas grid.
The idea is you basically "sell" H2 to the gas company, and later "buy" CH4 for your gas turbines. Bottom line most often that Gas company and electric company are owned by the same power company.
In my opinion the article is actually pretty good. Because you don't grasp that the following is wrong.
CAISO needs to quickly sell the excess to avoid overloading the electricity grid, which can cause blackouts.
The power is actually transported to Arizona via "the electric grid"... and usually you simply disconnect the power plant if you can not get rid of the power... no thread of overloading or black out (facepalm).
Oversupply causes prices to fall, even below zero. And this is absolute nonsense.
The price goes below zero because the power company has only 3 choices: a) disconnect a part of the solar power (probably with contracts forcing to pay for the not used power anyway), which might make it complicated to reconnect it (getting it in phase etc. or having "special protocols") b) powering down a fossile plant, with the problem that they already know that they have to power it back up close to nightfall, which might cause costs (more costs than "selling" the excess power for a negative price) c) giving the power away, for a negative price, which is bottom line a lower loss than the costs in a) or b)
There is no "market force" that drives power prices into the negative, that idea is absurd.
So: the article is complete bullshit, but you find it informative.
The more solar power you install, the more fast-reaction generating capacity you need. That is nonsense. You need the same amount you already have.
Power plant operators know when which plant is producing what amount of energy (clouds, no clouds, sunrise, wind etc.) and use ordinary load following plants to follow that curve. The remaining peaks are more or less the same like ordinary grids with no solar power.
From a grid operators point of view, there is no difference if a cooling house is "suddenly" starting its cooling and draws "suddenly" 5MW, or if a solar plant is "suddenly" covered by clouds and drops by 5MW production. The only difference is: the solar plant is not "suddenly" covered by clouds, operators have prognosis's for that, and: actually the cooling house usually is announcing its power up times to the grid operator, too.
That is not "process", that is the end product. I understood the parent meant "process" as in making silicon pure and doting it later and making the "wafers".
Not sure how much soldering they do in our times. I assumed they simply have a copper alloy evaporated on the surface.
Anyway, as panels will get recycled the solder tin (funny that we say tin in german, but as you are right, we used to use lead, too) will be recycled, too. Probably only as a block of metal.
And? Subject, predicate, object: still the same order in most european languages. Nitpicking about an exception like and adjective makes not really sense, or does it?
Then there's the thing where German sometimes outrageously at the end the verb puts. Not sure if I get it, but in english you mostly have the same freedom to rearrange words for emphasizing as in german.
In other words: if a german sentence has the verb at the end then it is mostly for dramatic/lyric or emphasizing. You don't need to learn that. You can basically convert most english sentences word by word into proper german, without changing any order. In my previous sentence however you would shift "convert" behind "english" when you translate it, the same for changing, it would indeed end up at the end of the sentence as "zu aendern".
But working out the reason for not having decimal time is a good exercise toward understanding why C being too coarse for conveniently expressing temps that humans care about most is a decided shortcoming Considering that 7.5 billion minus roughly 400.000.000 americans use C, that is a pointless exercise. I already explained to you: if you grow up with a unit of measurement, you prefer that over others. Beyond that there is no particular benefit of one over the other, unless you compare complete systems, where SI clearly wins.
The reason why we have 24 hours is super simple. Most old number systems used a mixed 20 and 60 based system. And during old times the day had 12 double hours... which later got split up into smaller hours, hence we have 24 now. And why do we have complex number systems instead of ten based systems? Well: by a cool accident the year has more or less exactly 360 days. That is the reason why a circle is measured with 360 degrees. The sun, or a particular star, is rising every night "one degree off" versus the night before. From a 360 "number system" we split down to a 60 based one and have minutes with 60 seconds and hours with 60 minutes. Of course you could define the length of a second arbitrarily and come up with a 10h day. However by a cool accident most adults have a heart beat of about 60 per minute, when healthy.
You probably were right with the "convenience" of Fahrenheit if it would start at the freezing point of water with zero and had 300F at the boiling point of water. Now I have to memorize the freezing point, I guess it was 17F? And memorizing such arbitrary stuff is completely pointless (for me). I guess it is even a stupid question asked in school and you get looked down uppon if you can not memorize it and get a bad grade in "nature science"?
With the current definition it feels very cumbersome for anyone who is using C. (But as I said: that is only because we are used to use C)
Your system of trucks carrying LNG to homes does not exist. Of course it exists. I see one every few days, facepalm.
Did you even bother to consult the references I provided? That PDF? Yes, of course. And did you check your numbers?
Compressing costs about 2% of the energy stored in the gas, so if you compress lets say 200m^3 into 1m^3, you lose the energy of about 4m^3.
If you cool it down to -160C, then that is about 5% of the energy stored in the gas, so you lose about 10m^3 worth of energy.
Clearly written in the pdf, so your numbers are way off.
There is no source of LNG to an individual home. I grew up in a rural area. Many had liquid gas stores, but granted that was not LNG but Propan/Butan.
This thread sprang from the post: "It might also be easier to buffer. Just pump it into a tank at your side, if there's an interruption in supply, you still have this reserve." The proposal was for a single home to "pump" natural gas into some kind of a pressure vessel in the home. Well, and I just proposed to use liquid LNG instead of a pressurized tank... on the other hand if you want to connect to the ordinary gas net, the obviously you are right and a pressurized tank is the simpler solution.
Well in my resume, I have about 30 programming languages listed, that I actually have used in commercial software development.
As long as there are no new languages with new concepts, or better syntax for old concepts, I don't see anything interesting on the horizon.
I mean: I could learn CoffeeScript... but for what? I only use JavaScrip in an HyperCard clone called NovoCard (for iPads)... I don't really like web development.
If I was to really learn something it would probably a library like AngularJS or ProcessingJS.
I still have not done real SmallTalk though... that would be fun:D
However if you have suggestions, I'm full ear (german saying, sounds stupid in english)
Strange, you must be using a different version of Prolog than I used to because, in the version I used, it was pretty hard to write a program looking like a pascal one.
With "Pascal" I mean: the rules more or less worked like procedures and not like rules.
In the version I used, every problem was pretty much solved with recursion. Well, that indicates that you probably did not really use logic/rule based programming, just like me;D
Probably you missed that basically all had electric power and could have AC if they wanted. Heating is indeed an interesting thing. Basically in no country I have ever been where it is HOT in summer, has any heating in winter. No idea why. Considering that floor heating was invented by the romans, you could assume that countries around the mediterranean where aware that it is cold in winter.
Sorry, as I said before, it is a matter of how you grew up. 1/4 of "a thing expressed in 16th steps" is simply 1/4. 1/4 of "a thing expressed in 10th steps" is simply 1/4.
No one who is cooking cares about the actual units and if they are "easily dividable" (an american myth, probably an explanation why that nation is going downhill).
And luckily, except when you only need a single one, no one is using 'spoons' anymore in cooking recipes.
Consider the difficulty in implementing decimal time, where a day was 10 hours instead of 24. That was a political problem, and honestly makes no sense anyway. what is next? A ten month year? With what... 3 ten month weeks and a 'break week'? It can't be helped but the planet has a 365 days year and some odd 28 days months. Going completely artificial in a time thing makes no sense at all.
I for my part like it that the sun is rising roughly 6:00 in the morning and setting 18:00 in the evening. And the times 'expand' in summer and 'shrink' in winter. A very convenient system. Around 12:00 the sun is straight over my head... more or less, as timezones are quite big now.
Slashdot Mirror
Unfortunately not.
The parent is completely correct that using the term "module" or "modularity" in Java 9 is completely misleading.
It has absolutely nothing to do with any meaning of "module" in any other programming language.
I'm actually astonished Java made it this far without such a feature!
There is no language / system on the planet that has such a feature. Except Java 9.
class path system
Which works exactly like the LD_LIBRARY_PATH for any compiled language on Unix or the equivalent on Windows.
The most successful and most widely used (programming) language on the planet is a disaster?
In what world do you live?
Because higher frequencies mean higher temperatures.
That is nonsense. Modern CPUs need more power, hence they produce more heat.
Remember when CPUs didn't need active cooling?
Yes, my 6502, got about 110C hot. But the transistors still worked at that temperature. Modern CPUs have so small transistors that they can not work at that temperature anymore as the electrons would simply jump several transistors far through the substrate.
I don't know how quickly stuff will change, but we have to fronts: first new main memory technologies, like magnetic based RAMs. And secondly, different operation systems to exploit that. Better/Different operating systems is probably the slow part.
Magnetic based RAM (Memristor, see: https://en.wikipedia.org/wiki/... magnetic based main memory that holds its memory when switched off) will blur the distinction between RAM and "hard drive" or any external memory.
Bottom line that means with a new operation system, probably more object oriented, you would no longer need to hold "files" on a "hard drive" but could keep the object a life in memory.
For that you would need OS support in case multiple "editors" are working on that "file". Bottom line a program would probably no longer work on multiple files, but would spawn a new instance/process for each file (what e.g. modern web browsers already do) and such a "file" would be mapped into the address space of the process.
In the long run we had something like text files, binary files (with structure like a GiF, but no code "attached") and "object oriented" files. Consider them like streamed Java objects containing also the class files for the relevant objects.
I basically imagine a "graph data base" system whee nodes are files and files can be executable, just like right now, but everything in main memory, never paged out, because main memory and backing storage is the same thing.
This thing would be a revolution for computers, external file systems only used for backup or data transfer.
On top of that we have processors that have fast reprogrammable FPGA like structures included. In the very long run we also have reprogrammable RAM with data flow processing capacity in the RAM directly. But I guess we are a long way to go to get this mainstream. Perhaps in GPUs or "neural net co processors".
A big contra point to my ideas above is optical computing and probably quantum computing. The later is in its infancy, but I would not wonder if there is a sudden big progress.
Optical computing we have since a while, but not scaled down to compete with anything we do right now. But: there are ideas of holographic RAMs that like the Memsistors keep their memory when switched off (and much much longer and more reliable).
There are prototypes where data lines on the boards are replaced with optical fibres. Super expensive but it makes some access paths quicker, because it avoids interferences.
There are some IBM research things (and probably HP too, but I forgot) where they are able to connect ordinary transistors with optic ones and have hybrid chips.
Anyway, I think it is time to rethink how the "whole thing" is working. It will basically still be a von Neumann architecture but the coupling of processor, most of the main RAM, and storage will be completely different. And that requires rethinking the OS.
iOS more or less works like that already, from an outside point of view. Inside it is in ordinary Unix, ofc. with a conventional file system.
I don't care what the wikipedia article says and I'm to lazy to read it.
Silicium cells _don't_ contain heavy metals. They don't contain _any metal at all_.
The only metals are outside in the _panel_ and _wiring_
So: super easy to recycle. And if you had physics in school: you damn would know that.
Then we have cells with a mirror on the back side, which again is on the outside! That is in most cases silver, and silver is counted as a precious metal, not as a heavy metal. Even chemically it is a heavy metal. Perhaps you meant that ...
So: all metals in a panel can be super easy recycled, they are not a threat on a land fill. And bottom line I assume that the whole cell gets recycled as the silicium is besides the fact that it is doted, still super pure!
We actually do that in Germany on an "experimental scale" and feed the hydrogen in a very low percentage into the ordinary gas grid.
The idea is you basically "sell" H2 to the gas company, and later "buy" CH4 for your gas turbines.
Bottom line most often that Gas company and electric company are owned by the same power company.
But you avoid the "storage problem" of H2.
The power company in Arizona, that bought the power and got money on top, profited.
Wow, that was easy again.
In my opinion the article is actually pretty good.
Because you don't grasp that the following is wrong.
CAISO needs to quickly sell the excess to avoid overloading the electricity grid, which can cause blackouts.
The power is actually transported to Arizona via "the electric grid" ... and usually you simply disconnect the power plant if you can not get rid of the power ... no thread of overloading or black out (facepalm).
Oversupply causes prices to fall, even below zero.
And this is absolute nonsense.
The price goes below zero because the power company has only 3 choices:
a) disconnect a part of the solar power (probably with contracts forcing to pay for the not used power anyway), which might make it complicated to reconnect it (getting it in phase etc. or having "special protocols")
b) powering down a fossile plant, with the problem that they already know that they have to power it back up close to nightfall, which might cause costs (more costs than "selling" the excess power for a negative price)
c) giving the power away, for a negative price, which is bottom line a lower loss than the costs in a) or b)
There is no "market force" that drives power prices into the negative, that idea is absurd.
So: the article is complete bullshit, but you find it informative.
The more solar power you install, the more fast-reaction generating capacity you need.
That is nonsense.
You need the same amount you already have.
Power plant operators know when which plant is producing what amount of energy (clouds, no clouds, sunrise, wind etc.) and use ordinary load following plants to follow that curve. The remaining peaks are more or less the same like ordinary grids with no solar power.
From a grid operators point of view, there is no difference if a cooling house is "suddenly" starting its cooling and draws "suddenly" 5MW, or if a solar plant is "suddenly" covered by clouds and drops by 5MW production. The only difference is: the solar plant is not "suddenly" covered by clouds, operators have prognosis's for that, and: actually the cooling house usually is announcing its power up times to the grid operator, too.
Why are the reporters always writing such a nonsense?
You feed power into the grid: it needs to be consumed. Or you can not feed it in.
And: I guess the solar power was teleported to Arizona, to prevent "overloading a wire"?
My parent said "no one attacked the US". :D
I did not answer to an "taking over" post, at least I don't remember
That is not "process", that is the end product. I understood the parent meant "process" as in making silicon pure and doting it later and making the "wafers".
Not sure how much soldering they do in our times. I assumed they simply have a copper alloy evaporated on the surface.
Anyway, as panels will get recycled the solder tin (funny that we say tin in german, but as you are right, we used to use lead, too) will be recycled, too. Probably only as a block of metal.
And?
Subject, predicate, object: still the same order in most european languages.
Nitpicking about an exception like and adjective makes not really sense, or does it?
Then there's the thing where German sometimes outrageously at the end the verb puts.
Not sure if I get it, but in english you mostly have the same freedom to rearrange words for emphasizing as in german.
In other words: if a german sentence has the verb at the end then it is mostly for dramatic/lyric or emphasizing. You don't need to learn that. You can basically convert most english sentences word by word into proper german, without changing any order. In my previous sentence however you would shift "convert" behind "english" when you translate it, the same for changing, it would indeed end up at the end of the sentence as "zu aendern".
But working out the reason for not having decimal time is a good exercise toward understanding why C being too coarse for conveniently expressing temps that humans care about most is a decided shortcoming
Considering that 7.5 billion minus roughly 400.000.000 americans use C, that is a pointless exercise.
I already explained to you: if you grow up with a unit of measurement, you prefer that over others. Beyond that there is no particular benefit of one over the other, unless you compare complete systems, where SI clearly wins.
The reason why we have 24 hours is super simple. Most old number systems used a mixed 20 and 60 based system. And during old times the day had 12 double hours ... which later got split up into smaller hours, hence we have 24 now. And why do we have complex number systems instead of ten based systems? Well: by a cool accident the year has more or less exactly 360 days. That is the reason why a circle is measured with 360 degrees. The sun, or a particular star, is rising every night "one degree off" versus the night before. From a 360 "number system" we split down to a 60 based one and have minutes with 60 seconds and hours with 60 minutes. Of course you could define the length of a second arbitrarily and come up with a 10h day. However by a cool accident most adults have a heart beat of about 60 per minute, when healthy.
You probably were right with the "convenience" of Fahrenheit if it would start at the freezing point of water with zero and had 300F at the boiling point of water. Now I have to memorize the freezing point, I guess it was 17F? And memorizing such arbitrary stuff is completely pointless (for me). I guess it is even a stupid question asked in school and you get looked down uppon if you can not memorize it and get a bad grade in "nature science"?
With the current definition it feels very cumbersome for anyone who is using C. (But as I said: that is only because we are used to use C)
Your system of trucks carrying LNG to homes does not exist.
Of course it exists.
I see one every few days, facepalm.
Did you even bother to consult the references I provided?
That PDF? Yes, of course. And did you check your numbers?
Compressing costs about 2% of the energy stored in the gas, so if you compress lets say 200m^3 into 1m^3, you lose the energy of about 4m^3.
If you cool it down to -160C, then that is about 5% of the energy stored in the gas, so you lose about 10m^3 worth of energy.
Clearly written in the pdf, so your numbers are way off.
There is no source of LNG to an individual home.
I grew up in a rural area. Many had liquid gas stores, but granted that was not LNG but Propan/Butan.
This thread sprang from the post: "It might also be easier to buffer. Just pump it into a tank at your side, if there's an interruption in supply, you still have this reserve." The proposal was for a single home to "pump" natural gas into some kind of a pressure vessel in the home. ... on the other hand if you want to connect to the ordinary gas net, the obviously you are right and a pressurized tank is the simpler solution.
Well, and I just proposed to use liquid LNG instead of a pressurized tank
Well in my resume, I have about 30 programming languages listed, that I actually have used in commercial software development.
As long as there are no new languages with new concepts, or better syntax for old concepts, I don't see anything interesting on the horizon.
I mean: I could learn CoffeeScript ... but for what? I only use JavaScrip in an HyperCard clone called NovoCard (for iPads) ... I don't really like web development.
If I was to really learn something it would probably a library like AngularJS or ProcessingJS.
I still have not done real SmallTalk though ... that would be fun :D
However if you have suggestions, I'm full ear (german saying, sounds stupid in english)
Strange, you must be using a different version of Prolog than I used to because, in the version I used, it was pretty hard to write a program looking like a pascal one.
With "Pascal" I mean: the rules more or less worked like procedures and not like rules.
In the version I used, every problem was pretty much solved with recursion. ;D
Well, that indicates that you probably did not really use logic/rule based programming, just like me
My almonds come from greece, turkey, spain, france and other mediterranean sites.
We did not talk about CANDU reactors to that point.
That was a different post of yours.
And perhaps you want to check the GP.
I would say your argument about CANDU reactors is pointless, too.
As their fuel does not need 'reprocessing', facepalm.
Probably you missed that basically all had electric power and could have AC if they wanted.
Heating is indeed an interesting thing. Basically in no country I have ever been where it is HOT in summer, has any heating in winter. No idea why. Considering that floor heating was invented by the romans, you could assume that countries around the mediterranean where aware that it is cold in winter.
Everyone does.
Same with cold, btw.
Sorry, as I said before, it is a matter of how you grew up.
1/4 of "a thing expressed in 16th steps" is simply 1/4.
1/4 of "a thing expressed in 10th steps" is simply 1/4.
No one who is cooking cares about the actual units and if they are "easily dividable" (an american myth, probably an explanation why that nation is going downhill).
And luckily, except when you only need a single one, no one is using 'spoons' anymore in cooking recipes.
Consider the difficulty in implementing decimal time, where a day was 10 hours instead of 24. ... 3 ten month weeks and a 'break week'?
That was a political problem, and honestly makes no sense anyway.
what is next? A ten month year? With what
It can't be helped but the planet has a 365 days year and some odd 28 days months. Going completely artificial in a time thing makes no sense at all.
I for my part like it that the sun is rising roughly 6:00 in the morning and setting 18:00 in the evening. And the times 'expand' in summer and 'shrink' in winter. A very convenient system. Around 12:00 the sun is straight over my head ... more or less, as timezones are quite big now.
We never talked about 'took over'.
And a close to 400 sunk ships (with two lost uboats) hardly count as hit and run.