I have read your examples. I was already well aware of the implications of what I was saying. So your examples don't change my mind at all, and it's not because I'm being stubborn.
Are you saying the police shouldn't save "bad people" from others? Or if they are in a building on fire, the firemen shouldn't save them?[1] And that "bad people" should not be able to borrow books from a public library? And that "bad people" shouldn't be allowed to vote[2]? Or have freedom of speech?
You really think freedom of speech and the right to vote is worth less than a "medical insurance package worth up to USD500K"?
This is part of the civilization thing I was talking about. When you live in a civilized society with rule of law, everyone gets certain protections, certain freedoms, even free access to various resources.
Whether or not you, I or anyone else likes it or not.
Even a murderer on deathrow is not supposed to be killed by anyone else except the State.
If you still don't get it, you do not understand what civilization is and why it really is a good thing for us.
I'm not being stubborn. I understand very well the consequences. I prefer living in a civilized world. If it means some of my taxes go to keeping some "bad person" alive (hospital etc), that's fine with me.
My taxes already go towards keeping a bunch of "bad people" (e.g. certain politicians and their cronies) filthy rich. Which to me is even worse - that's not necessary for a civilized society - in fact those are symptoms of a corrupt society.
[1] Yes if they know, who to save, maybe they should save someone else first;), but firemen normally don't have time for such things - they just do what they can.
BTW having a law which says firemen should only save the good people first, would be uncivilized and impede them in their work.
Should they save a stinking murderer who has killed only one person, and who might at most kill only 5 more other people in his lifetime? Or should they save a well-heeled politician who has lied and thus caused the death of tens of thousands of people for the benefit of himself and his friends (and not the nation)?
To me they shouldn't have to think about such stuff. I'm fine if a fireman's reasons for saving one _first_ and not the other was: he looked lighter:).
[2] I am well aware that in the USA it is common for convicted felons to not be able to vote. This to me is wrong. Even prisoners should be allowed to vote (and do it freely without coercion or compulsion). This is part of reintegrating them back to society.
Yes, MSO2007 has issues (I don't know why MS Word uses CPU even if supposed to be idling, and I've had trouble pasting ) and the UI change is "meh", but in my opinion Open Office is very significantly worse[1]. Perhaps the latest version is much better now, but after so many years of disappointment, it's hard to believe it has improved dramatically. I'll still use it if I have to, but it is far from a good substitute.
In contrast, I've tried Kingsoft Office and they seem to be a better MSO replacement option. My bro says the spreadsheet doesn't support the matrix multiplication function (which Excel has), but he bought the Kingsoft Office anyway just to test it out, said it was cheap enough - I told him he could just download the eval copy. I think he just likes to support the underdog.
They don't have a Microsoft Outlook substitute though. Outlook sucks (CPU hog, crashes, hangs, search needs improvement) but the stuff it does (works with Exchange, calendaring, etc) is necessary in many organizations.
It's been around for years (reported in 2005) and still not fixed (implications of bug: you cannot step through a find-and-replace within a selection - it has to be for the whole document).
I just tested it on 3.2 and the bug is still there!
In contrast when I search for bugs in Word 2007 I see the usual crash bugs (I doubt OOo is immune) and stuff like:
"For example, in Word 2007, if you use a nonbreaking hyphen to join two words, the new combined word is flagged as misspelled (which is fine), but then you cannot "Ignore All" instances of the combined word when doing a spell-check. Wen doing a spell-check, you must ignore each individual instance of the combined word throughout the document. As you can imagine, if you have a proper name that includes a nonbreaking hyphen on every page of a 200 page contract, individually ignoring each instance of the combined word during spell-check can get very annoying very quickly. Thanks for any replies!"
I also have been very annoyed with Word 2007 by one bug - sometimes when I try to paste something into certain word documents, somehow all the formatting isn't copied over when it's supposed to (even when I choose "keep source formatting" - some behind the scenes weirdness, it works in other word docs). For example the table and contents are copied and pasted fine, but the font is wrong!
Smaller game companies don't need to worry so much about DDoS, overloading, and paying a premium for big pipes, if they didn't use stuff like Ubisoft's stupid authentication scheme.
If they want to use such schemes and have it work, they should pay for the pipes, systems, services etc required to do it properly.
It's very expensive? Well, it's supposedly going to make them more money right?
Forgive me if I don't feel a single drop of sympathy for them, and in fact feel a fair bit of schadenfreude.
I suspect if their organization isn't that incompetent someone might already have told them, the bosses just didn't want to listen.
> Does that mean nobody from my highschool is capable of using MS Word, Excel, Photoship, or Access? No it doesn't. > they taught people how to use computers in general. > How to look for stuff in the menus, and told us that we should read the help files if we got stuck
FWIW it took me a while to figure out that you had to click on the MS Office 2007 logo to do certain stuff, e.g. "save as" to save a document as a different file:).
Pressing F1 to help, doesn't help with that. Maybe my memory is bad/"nostalgic", but I think the previous microsoft office help was better.
Yes. The same for everybody. And why not? (It's not like they get the money upfront in their bank accounts - it's like insurance).
Everyone gets an equal chance to show how deserving or how undeserving they are. That's called being civilized. Everyone gets given a base level of "civility".
And that way they have fewer and fewer excuses for their behaviour if we ever have to justify punishing them for whatever they do, and even the form of punishment.
No: "I desperately needed the money to pay for medical bills".
If you think 500K is too high for everyone, then make it something lower.
I've already been doing something like that. Except autorun.inf is a folder with dummy stuff in it.
So they have to change the permissions (it's NTFS) AND recursively delete stuff:).
That said, since any computer has rights to write to lots of other places on the drive, doing that folder thing is probably overkill and doesn't really help that much more in the big picture.
But yeah, it has saved my drives from infection. These autorun malware are really rather common. It's almost like it's back to the old days of floppy viruses.
Yeah it was probably made in China, and typically nobody cares about QC/QA in the factory (or part of the QA is making sure the malware is installed;) ).
I found malware on a supposedly new PNY usb drive about a year ago. Perhaps it was a repackaged item.
Anyway, didn't affect the machine I plugged it into since auto-run was disabled (like it should be).
> Can you name anything that can be made in Lisp or other functional language that cannot be done more in depth in C/C++ or any other mid/low level imperative language?
In theory there's nothing you can't do in Lisp or other language that you can't do in any other imperative language. After all you can write a Lisp interpreter/compiler in the imperative language...
In fact I have a nagging suspicion that some Java programs are actually Lisp interpreters in disguise and the Lisp program just happens to be the XML configuration file (go look at some of those XML config files, and compare the structure/syntax with Lisp;) ).
In practice, lots of things require fewer lines of code when written in a higher level language - and the assumption was you are not using other people's libraries/code. So if you want to rewrite most stuff from scratch, it seems better to pick an expressive language (fewer lines of code) that at the same time doesn't totally suck in performance when compared to C. There are a fair number of candidates, not just Lisp. The Computer Science bunch can probably provide a comprehensive list.
And yes it can still be done in C - Linus, GNU et all have proven it:).
> Like what? Python, Ruby, Perl? Sometimes its really good to know how things work inside the language even if you're not going to use them.
Anything with lots of decent prefab that you can use for your work, especially standard or defacto standard. Shoddy prefab should be avoided.
So that could be Perl, Python, Ruby, Java, even.Net. Right tools for the job and all that.
Whereas the other scenario was more of the best way of making your own tools and buildings starting from the molecule/atom level.
> Every day for a programmer is full of moments where using the wrong library call may introduce bugs.
True, which brings us to the next advantage of using standard (or defacto standard) libraries. An experienced programmer who has taken over the code can more easily know whether the library is being used in the wrong way or not.
As you mentioned, people could near-instantly go "shouldn't be using Array.Sort that way". Whereas if it's a custom array sort, people would have to read more lines of code.
When you write most of your own libraries, someone else taking over the maintenance/development has to figure out whether the bugs are in your library or not (stuff may be called something similar as to the standard lib but you could have done something different from the standard library - otherwise why would you be writing your own right?).
Lastly, you often can't avoid library calls - you have to do use them to interact with the O/S or GUI or whatever. Unfortunately sometimes the documentation is ambiguous, not clear, or even plain wrong. But still documentation is more often nonexistent for "The Previous Coder in the Company" libraries/code ( not necessarily the coder's fault - just an emergent effect in many companies ).
> When I was getting my CS degree I took classes on formal methods for proving that your software is correct. It's not a clear-cut thing.
And there's also the huge assumption that the requirements are correct:). In the real world, your software might do exactly what the requirements say. But the requirements could be wrong.
Then all that verification becomes a big waste of time and money.
Car analogy: all you are proving with verification is the steering tyres will 100% turn with the steering wheel - they will never turn the wrong direction. But formal verification doesn't prove that you are turning the steering wheel in the right direction in the first place.
With a lot of bugs, the code is working as designed. The design just happens to be wrong.
> Actually, you are wrong. What you are suggesting is a lifetime cap.
1) I did say it could be an amount/period too - I'm no economist or actuarist, they can figure out the details. 2) The lifetime cap could be adjusted depending on the country's economic status.
The advantage of a cap (even if it can change) is that it sets some expectations up front. Think of it as a personal budget for spending.
> On the other hand, if you have a global budget for health, because of socialised medicine, then the problem is simply one of doing stats and updating the costs every year.
That's like not giving departments a budget but allowing them to draw directly what they want/need from the entire company's budget. That's not good. Yes you can still control spending, but it's usually better to just allocate separate budgets first.
Having no explicit limits up front would also mean voters expectations might be higher than what the country could afford in the long term. And when that happens politicians may find it easier to spend the money of future generations, for votes today. They're not going to be around 30 years later, or at least not going to be blamed by most for the disaster.
> The problem is with the newer generation of Java/C# who: can't write their own algorithms thus inevitably depend on libraries
The "whiz-bang" algorithms are all in the libraries already. So why reinvent the wheel? And if they aren't, it's not usually your job to put them in.
Furthermore most of the programming work in the real world rarely requires programmers to write new "classical style"[1] algorithms (e.g. fastest, memory efficient etc).
[1] In theory all programs involve algorithms even if written by noobs, but I doubt that's the meaning of algorithm you wanted when you said "can't write their own algorithms".
There's not much point figuring out the theoretical best possible algorithm to use for Client A's business requirements. Especially when Client A changes the requirements every week.
Or say you're writing a RADIUS or DNS server, is there really going to be an algorithm for that, the way you have "quicksort" for sorting? Most of that "classic" algorithm stuff will be in the libraries and OS you use. Whereas what you as a programmer do mainly is figure out stuff like RFC compliance (many times RFCs are ambiguous - so stuff can be compliant and in the real world still not work well or at all, and sometimes compliance to one part is stupid, so you might want to make "stupid" optional), workarounds for noncompliant clients/servers, exception handling, how to avoid being exploited by hackers, what to do if there's a DoS, data structures, what to do for logging, how configuration should best be done (including stuff like whether to allow updating the configuration and still not drop any sessions, and how to do it), scalability.
Yes is so sad that many of us use "prefab" for building stuff.
Because: a) we're not good at making our own bricks, doors, tiles, nails, etc. b) we think it's a waste of time to do so when people are already making entire prefabricated components that we can use, and the Customer/Boss wants the thing done yesterday (despite not being able to describe exactly what it is they want - and if they could they wouldn't need programmers;) ).
> It's not a political problem here because 70% or so of the people are covered by employer plans that more or less support the entire wasteful system. Again, the 30% are screwed.
Actually a significant proportion of those 70% are screwed too, they just don't know it yet... Having your healthcare tied to your employer is not a good thing ( http://www.health.com/health/money-article/0,,20223203_1,00.html ). Overall you have to deal with all sorts of crap (much not present in other systems), which is not good if you are actually very sick.
A lot of US people yell at the Europeans/Canadians and say "YOUR HEALTHCARE ISN'T FREE OR CHEAP, OTHER PEOPLE ARE PAYING FOR IT". The thing is, those other people are paying less on average than they would be if they were in the US system. Just look at the statistics - per capita spending on health care.
If you are very rich, the US healthcare is good for you, since you can really get the best and afford it.
If the majority who are not very rich want to stick with their current broken system where people are clogging up ER rooms just to get treatment (and thus indirectly increasing costs and decreasing quality of healthcare, and don't forget that taxpayers are still paying for those non-emergency ER patients), then perhaps they should get what they deserve. Representative democracy and all that...
> I know I would be much happier if I didn't have to put up with the stress of making the decision
Some may not believe it, but that can actually be true.
It's part of human psychology. When you make decisions, you often wonder whether you made the right decision etc. And that reduces your happiness and satisfaction.
I think the big question is not so much whether you think your life is priceless.
If you're going to pay for all the treatments etc yourself, hardly anyone else will care.
It's how much should other people pay for you, to keep you alive for X years longer.
And that includes total strangers - assuming your country has some sort of health care system.
Some may say ZERO, but in my opinion, if we want to pretend that we are living in this human construct called civilized society, then yes we SHOULD pay more than zero to help keep others (even total strangers) alive. Not doing so would be uncivilized.
But beyond a certain point it becomes unreasonable and unfair to expect others to keep paying for you. And I think this should be related to how rich the country is and how rich the average person in that country is. There just won't be enough money/resources to go around.
The issue is as technology improves, there will be more and more advanced treatments, and many of those will be more and more expensive. The billionaires of this world might be able to pay for replacement body parts (say a whole leg) to be custom made for them (in fact maybe 1000 could be made and the best one used), but you can't provide this to everyone who wants it. If they want others to pay for it they'll have to make do with a cheaper prosthetic. Of course if that "organ printing" stuff becomes affordable, then OK.
To me a fair healthcare system would give everyone a quota, say USD500K/lifetime (or maybe something else per time period, let the economists and actuarists work it out). Once you've used that all up, no more automatic help from strangers. Sure it's unfair to let you die/suffer, but past that point, it is even more unfair to make others keep paying for you. What makes you so much more important than them?
So beyond that point you have to pay for yourself or convince others to help pay for you. Perhaps others can donate some of their quota to you (subject to regulations and approval - otherwise there'll be too much swindling).
FWIW, I don't think it's fair to have other people to pay USD600K to keep me alive for a few more years or decades. I live in a 3rd world country and earn a 3rd world wage, so USD600K is a lot here, maybe if I live in a rich country it'll be different.
Nor should they pay USD600K even if it'll keep me alive indefinitely. Given a long enough time, the odds are I'll become a nuisance (since I am imperfect);).
The advantage of using libraries is it means you write less code.
The less code you write, the fewer bugs you create, and less code you are directly responsible for fixing and documenting.
Yes the libraries won't be perfect, but in general they should be less crap than your code (especially if used and fixed by many others).
People who like "writing everything" themselves should use stuff like Lisp- programming languages that are powerful because they allow a programmer to personally write all sorts of stuff.
The rest of us should use languages that are powerful because they allow the programmer to NOT have to personally write all sorts of stuff:).
The "real programmers" can sneer at us, but we'll have completed the project way before they have finished writing the BIOS, bootloader, operating system, libraries and editor so that they can actually start writing the "real program"...
In many of the possible futures you will have DRM in the brain.
After all once they start adding "videographic"+"audiographic" memory as an option, or even mere photographic memory, the Media Monopolizers will start requiring that either you get crippled in order to watch a movie, or you use brain augmentation that has DRM in it.
Then it'll be not a penny for your thoughts, it'll be USD0.99 to recall "their" stuff.
Good luck trying to share experiences with your friends that happen to also included copyrighted content (background music in shopping mall etc).
>Enterprise loads such as databases do many many seeks and tend to have long queues as many clients request the data. Size and throughput are less important for these loads than seek time (though still critical).
Did you even read the link?
"I saw 4KB random write speed drop from 50MB/s down to 45MB/s. Sequential write speed remained similarly untouched. But now I've gone and ruined the surprise."
That's for random writes. 4KB random writes at 45MB/sec is 11520 writes per second.
A 15000rpm drive doing 4KB random writes (noncached/buffered) will only manage about 250 IOPS ( assuming 4 millisecond seek times). Or about 1MBps.
That's 45 times slower. You'll need a lot of spindles to match that.
The only issues I see with SSD are whether reliability is really up to scratch, whether you can hotswap them, and perhaps capacity (if you somehow can't use a tiered storage scheme).
Should just get them to continue teaching, but to video cameras. Then after a reviewing process, put those that meet sufficient standards on youtube or wherever.
If they are accused of incompetence at least you would also have recordings to prove whether they are or aren't;).
All teachers need is a high enough salary to live a decent lifestyle (I know it's relative, but I'm sure you can figure out what I mean). Most good teachers have no illusions about becoming millionaires through teaching. They're not stupid after all. Being super rich is not their goal in life.
Good teachers enjoy teaching. Most don't like dealing with loads of admin crap, or politicking.
So you spend some of the money and resources not on high salaries, but on getting most of that crap out of the way.
Where high salaries can come in handy for teachers are: subsidized/free education for their own children[1], and housing loans/allowances (and in the USA, medical/health stuff).
I suggest that it may be cheaper to provide them that than to directly provide them higher salaries.
For example: instead of paying all teachers high enough salaries so that all their children can go to university, do masters, PhD etc, you just commit to paying for any of their children that want to (and meet the grade/entry requirements), and take a gamble that not all their children will want to do so, and not all would want to go to the most expensive universities[2] (and meet the entry requirements). And so I bet you end up paying less overall.
[1] It would be sad and ironic if teachers cannot afford to provide good education for their own children. And I'm sure most good teachers place significant value on education.
[2] and only the approved ones, otherwise people will be setting up "online super expensive university courses"...
Slashdot Mirror
Only if they have been convicted of behaving badly enough to require it.
I have read your examples. I was already well aware of the implications of what I was saying. So your examples don't change my mind at all, and it's not because I'm being stubborn.
;), but firemen normally don't have time for such things - they just do what they can.
:).
Are you saying the police shouldn't save "bad people" from others? Or if they are in a building on fire, the firemen shouldn't save them?[1]
And that "bad people" should not be able to borrow books from a public library?
And that "bad people" shouldn't be allowed to vote[2]? Or have freedom of speech?
You really think freedom of speech and the right to vote is worth less than a "medical insurance package worth up to USD500K"?
This is part of the civilization thing I was talking about. When you live in a civilized society with rule of law, everyone gets certain protections, certain freedoms, even free access to various resources.
Whether or not you, I or anyone else likes it or not.
Even a murderer on deathrow is not supposed to be killed by anyone else except the State.
If you still don't get it, you do not understand what civilization is and why it really is a good thing for us.
I'm not being stubborn. I understand very well the consequences. I prefer living in a civilized world. If it means some of my taxes go to keeping some "bad person" alive (hospital etc), that's fine with me.
My taxes already go towards keeping a bunch of "bad people" (e.g. certain politicians and their cronies) filthy rich. Which to me is even worse - that's not necessary for a civilized society - in fact those are symptoms of a corrupt society.
[1] Yes if they know, who to save, maybe they should save someone else first
BTW having a law which says firemen should only save the good people first, would be uncivilized and impede them in their work.
Should they save a stinking murderer who has killed only one person, and who might at most kill only 5 more other people in his lifetime? Or should they save a well-heeled politician who has lied and thus caused the death of tens of thousands of people for the benefit of himself and his friends (and not the nation)?
To me they shouldn't have to think about such stuff. I'm fine if a fireman's reasons for saving one _first_ and not the other was: he looked lighter
[2] I am well aware that in the USA it is common for convicted felons to not be able to vote. This to me is wrong. Even prisoners should be allowed to vote (and do it freely without coercion or compulsion). This is part of reintegrating them back to society.
Yes, MSO2007 has issues (I don't know why MS Word uses CPU even if supposed to be idling, and I've had trouble pasting ) and the UI change is "meh", but in my opinion Open Office is very significantly worse[1]. Perhaps the latest version is much better now, but after so many years of disappointment, it's hard to believe it has improved dramatically. I'll still use it if I have to, but it is far from a good substitute.
In contrast, I've tried Kingsoft Office and they seem to be a better MSO replacement option. My bro says the spreadsheet doesn't support the matrix multiplication function (which Excel has), but he bought the Kingsoft Office anyway just to test it out, said it was cheap enough - I told him he could just download the eval copy. I think he just likes to support the underdog.
They don't have a Microsoft Outlook substitute though. Outlook sucks (CPU hog, crashes, hangs, search needs improvement) but the stuff it does (works with Exchange, calendaring, etc) is necessary in many organizations.
[1] See: http://tinyurl.com/ykvya22
Lots of the bugs there look familiar to me: formatting not being saved, bullets not behaving correctly.
And stuff like this (I just picked one of the bugs from the above list which caught my eye):
http://qa.openoffice.org/issues/show_bug.cgi?id=56449
http://qa.openoffice.org/issues/show_bug.cgi?id=74707
It's been around for years (reported in 2005) and still not fixed (implications of bug: you cannot step through a find-and-replace within a selection - it has to be for the whole document).
I just tested it on 3.2 and the bug is still there!
In contrast when I search for bugs in Word 2007 I see the usual crash bugs (I doubt OOo is immune) and stuff like:
"For example, in Word 2007, if you use a nonbreaking hyphen to join
two words, the new combined word is flagged as misspelled (which is fine),
but then you cannot "Ignore All" instances of the combined word when doing a
spell-check. Wen doing a spell-check, you must ignore each individual
instance of the combined word throughout the document. As you can imagine,
if you have a proper name that includes a nonbreaking hyphen on every page of
a 200 page contract, individually ignoring each instance of the combined word
during spell-check can get very annoying very quickly. Thanks for any
replies!"
I also have been very annoyed with Word 2007 by one bug - sometimes when I try to paste something into certain word documents, somehow all the formatting isn't copied over when it's supposed to (even when I choose "keep source formatting" - some behind the scenes weirdness, it works in other word docs). For example the table and contents are copied and pasted fine, but the font is wrong!
Smaller game companies don't need to worry so much about DDoS, overloading, and paying a premium for big pipes, if they didn't use stuff like Ubisoft's stupid authentication scheme.
If they want to use such schemes and have it work, they should pay for the pipes, systems, services etc required to do it properly.
It's very expensive? Well, it's supposedly going to make them more money right?
Forgive me if I don't feel a single drop of sympathy for them, and in fact feel a fair bit of schadenfreude.
I suspect if their organization isn't that incompetent someone might already have told them, the bosses just didn't want to listen.
> Does that mean nobody from my highschool is capable of using MS Word, Excel, Photoship, or Access? No it doesn't.
:).
> they taught people how to use computers in general.
> How to look for stuff in the menus, and told us that we should read the help files if we got stuck
FWIW it took me a while to figure out that you had to click on the MS Office 2007 logo to do certain stuff, e.g. "save as" to save a document as a different file
Pressing F1 to help, doesn't help with that. Maybe my memory is bad/"nostalgic", but I think the previous microsoft office help was better.
Even if it's on, over a slower internet connection, you briefly see IE in the leftmost spot. This is suboptimal for choice.
But anyway, it does seem to be working - people are trying out other browsers.
It may also show how trusting and confident people are about downloading new software, installing it and running it on their systems...
Yes. The same for everybody. And why not? (It's not like they get the money upfront in their bank accounts - it's like insurance).
Everyone gets an equal chance to show how deserving or how undeserving they are. That's called being civilized. Everyone gets given a base level of "civility".
And that way they have fewer and fewer excuses for their behaviour if we ever have to justify punishing them for whatever they do, and even the form of punishment.
No: "I desperately needed the money to pay for medical bills".
If you think 500K is too high for everyone, then make it something lower.
I've already been doing something like that. Except autorun.inf is a folder with dummy stuff in it.
:).
So they have to change the permissions (it's NTFS) AND recursively delete stuff
That said, since any computer has rights to write to lots of other places on the drive, doing that folder thing is probably overkill and doesn't really help that much more in the big picture.
But yeah, it has saved my drives from infection. These autorun malware are really rather common. It's almost like it's back to the old days of floppy viruses.
Yeah it was probably made in China, and typically nobody cares about QC/QA in the factory (or part of the QA is making sure the malware is installed ;) ).
I found malware on a supposedly new PNY usb drive about a year ago. Perhaps it was a repackaged item.
Anyway, didn't affect the machine I plugged it into since auto-run was disabled (like it should be).
Dvorak's (and most layouts) are biased towards English though[1].
:). { } () * and so on are not in the middle row. :)
They don't help so much for stuff like C, Java, perl
[1] And so is slashdot
Filter error, please use fewer junk characters.
> Can you name anything that can be made in Lisp or other functional language that cannot be done more in depth in C/C++ or any other mid/low level imperative language?
;) ).
:).
.Net. Right tools for the job and all that.
In theory there's nothing you can't do in Lisp or other language that you can't do in any other imperative language. After all you can write a Lisp interpreter/compiler in the imperative language...
In fact I have a nagging suspicion that some Java programs are actually Lisp interpreters in disguise and the Lisp program just happens to be the XML configuration file (go look at some of those XML config files, and compare the structure/syntax with Lisp
In practice, lots of things require fewer lines of code when written in a higher level language - and the assumption was you are not using other people's libraries/code. So if you want to rewrite most stuff from scratch, it seems better to pick an expressive language (fewer lines of code) that at the same time doesn't totally suck in performance when compared to C. There are a fair number of candidates, not just Lisp. The Computer Science bunch can probably provide a comprehensive list.
And yes it can still be done in C - Linus, GNU et all have proven it
> Like what? Python, Ruby, Perl? Sometimes its really good to know how things work inside the language even if you're not going to use them.
Anything with lots of decent prefab that you can use for your work, especially standard or defacto standard. Shoddy prefab should be avoided.
So that could be Perl, Python, Ruby, Java, even
Whereas the other scenario was more of the best way of making your own tools and buildings starting from the molecule/atom level.
> Every day for a programmer is full of moments where using the wrong library call may introduce bugs.
True, which brings us to the next advantage of using standard (or defacto standard) libraries. An experienced programmer who has taken over the code can more easily know whether the library is being used in the wrong way or not.
As you mentioned, people could near-instantly go "shouldn't be using Array.Sort that way". Whereas if it's a custom array sort, people would have to read more lines of code.
When you write most of your own libraries, someone else taking over the maintenance/development has to figure out whether the bugs are in your library or not (stuff may be called something similar as to the standard lib but you could have done something different from the standard library - otherwise why would you be writing your own right?).
Lastly, you often can't avoid library calls - you have to do use them to interact with the O/S or GUI or whatever. Unfortunately sometimes the documentation is ambiguous, not clear, or even plain wrong. But still documentation is more often nonexistent for "The Previous Coder in the Company" libraries/code ( not necessarily the coder's fault - just an emergent effect in many companies ).
> When I was getting my CS degree I took classes on formal methods for proving that your software is correct. It's not a clear-cut thing.
:). In the real world, your software might do exactly what the requirements say. But the requirements could be wrong.
And there's also the huge assumption that the requirements are correct
Then all that verification becomes a big waste of time and money.
Car analogy: all you are proving with verification is the steering tyres will 100% turn with the steering wheel - they will never turn the wrong direction.
But formal verification doesn't prove that you are turning the steering wheel in the right direction in the first place.
With a lot of bugs, the code is working as designed. The design just happens to be wrong.
> Actually, you are wrong. What you are suggesting is a lifetime cap.
1) I did say it could be an amount/period too - I'm no economist or actuarist, they can figure out the details.
2) The lifetime cap could be adjusted depending on the country's economic status.
The advantage of a cap (even if it can change) is that it sets some expectations up front. Think of it as a personal budget for spending.
> On the other hand, if you have a global budget for health, because of socialised medicine, then the problem is simply one of doing stats and updating the costs every year.
That's like not giving departments a budget but allowing them to draw directly what they want/need from the entire company's budget. That's not good. Yes you can still control spending, but it's usually better to just allocate separate budgets first.
Having no explicit limits up front would also mean voters expectations might be higher than what the country could afford in the long term. And when that happens politicians may find it easier to spend the money of future generations, for votes today. They're not going to be around 30 years later, or at least not going to be blamed by most for the disaster.
> The problem is with the newer generation of Java/C# who: can't write their own algorithms thus inevitably depend on libraries
;) ).
The "whiz-bang" algorithms are all in the libraries already. So why reinvent the wheel? And if they aren't, it's not usually your job to put them in.
Furthermore most of the programming work in the real world rarely requires programmers to write new "classical style"[1] algorithms (e.g. fastest, memory efficient etc).
[1] In theory all programs involve algorithms even if written by noobs, but I doubt that's the meaning of algorithm you wanted when you said "can't write their own algorithms".
There's not much point figuring out the theoretical best possible algorithm to use for Client A's business requirements. Especially when Client A changes the requirements every week.
Or say you're writing a RADIUS or DNS server, is there really going to be an algorithm for that, the way you have "quicksort" for sorting? Most of that "classic" algorithm stuff will be in the libraries and OS you use. Whereas what you as a programmer do mainly is figure out stuff like RFC compliance (many times RFCs are ambiguous - so stuff can be compliant and in the real world still not work well or at all, and sometimes compliance to one part is stupid, so you might want to make "stupid" optional), workarounds for noncompliant clients/servers, exception handling, how to avoid being exploited by hackers, what to do if there's a DoS, data structures, what to do for logging, how configuration should best be done (including stuff like whether to allow updating the configuration and still not drop any sessions, and how to do it), scalability.
Yes is so sad that many of us use "prefab" for building stuff.
Because:
a) we're not good at making our own bricks, doors, tiles, nails, etc.
b) we think it's a waste of time to do so when people are already making entire prefabricated components that we can use, and the Customer/Boss wants the thing done yesterday (despite not being able to describe exactly what it is they want - and if they could they wouldn't need programmers
> It's not a political problem here because 70% or so of the people are covered by employer plans that more or less support the entire wasteful system. Again, the 30% are screwed.
Actually a significant proportion of those 70% are screwed too, they just don't know it yet... Having your healthcare tied to your employer is not a good thing ( http://www.health.com/health/money-article/0,,20223203_1,00.html ). Overall you have to deal with all sorts of crap (much not present in other systems), which is not good if you are actually very sick.
A lot of US people yell at the Europeans/Canadians and say "YOUR HEALTHCARE ISN'T FREE OR CHEAP, OTHER PEOPLE ARE PAYING FOR IT". The thing is, those other people are paying less on average than they would be if they were in the US system. Just look at the statistics - per capita spending on health care.
If you are very rich, the US healthcare is good for you, since you can really get the best and afford it.
If the majority who are not very rich want to stick with their current broken system where people are clogging up ER rooms just to get treatment (and thus indirectly increasing costs and decreasing quality of healthcare, and don't forget that taxpayers are still paying for those non-emergency ER patients), then perhaps they should get what they deserve. Representative democracy and all that...
> I know I would be much happier if I didn't have to put up with the stress of making the decision
Some may not believe it, but that can actually be true.
It's part of human psychology. When you make decisions, you often wonder whether you made the right decision etc. And that reduces your happiness and satisfaction.
Watch the following:
Barry Schwartz: The paradox of choice
http://www.youtube.com/watch?v=VO6XEQIsCoM
(esp see 3:41 onwards in this context)
Dan Gilbert: Why are we happy? Why aren't we happy?
http://www.youtube.com/watch?v=LTO_dZUvbJA
I think the big question is not so much whether you think your life is priceless.
;).
If you're going to pay for all the treatments etc yourself, hardly anyone else will care.
It's how much should other people pay for you, to keep you alive for X years longer.
And that includes total strangers - assuming your country has some sort of health care system.
Some may say ZERO, but in my opinion, if we want to pretend that we are living in this human construct called civilized society, then yes we SHOULD pay more than zero to help keep others (even total strangers) alive. Not doing so would be uncivilized.
But beyond a certain point it becomes unreasonable and unfair to expect others to keep paying for you. And I think this should be related to how rich the country is and how rich the average person in that country is. There just won't be enough money/resources to go around.
The issue is as technology improves, there will be more and more advanced treatments, and many of those will be more and more expensive. The billionaires of this world might be able to pay for replacement body parts (say a whole leg) to be custom made for them (in fact maybe 1000 could be made and the best one used), but you can't provide this to everyone who wants it. If they want others to pay for it they'll have to make do with a cheaper prosthetic. Of course if that "organ printing" stuff becomes affordable, then OK.
To me a fair healthcare system would give everyone a quota, say USD500K/lifetime (or maybe something else per time period, let the economists and actuarists work it out). Once you've used that all up, no more automatic help from strangers. Sure it's unfair to let you die/suffer, but past that point, it is even more unfair to make others keep paying for you. What makes you so much more important than them?
So beyond that point you have to pay for yourself or convince others to help pay for you. Perhaps others can donate some of their quota to you (subject to regulations and approval - otherwise there'll be too much swindling).
FWIW, I don't think it's fair to have other people to pay USD600K to keep me alive for a few more years or decades. I live in a 3rd world country and earn a 3rd world wage, so USD600K is a lot here, maybe if I live in a rich country it'll be different.
Nor should they pay USD600K even if it'll keep me alive indefinitely. Given a long enough time, the odds are I'll become a nuisance (since I am imperfect)
As another writer of crappy code:
:).
The advantage of using libraries is it means you write less code.
The less code you write, the fewer bugs you create, and less code you are directly responsible for fixing and documenting.
Yes the libraries won't be perfect, but in general they should be less crap than your code (especially if used and fixed by many others).
People who like "writing everything" themselves should use stuff like Lisp- programming languages that are powerful because they allow a programmer to personally write all sorts of stuff.
The rest of us should use languages that are powerful because they allow the programmer to NOT have to personally write all sorts of stuff
The "real programmers" can sneer at us, but we'll have completed the project way before they have finished writing the BIOS, bootloader, operating system, libraries and editor so that they can actually start writing the "real program"...
> Perhaps I just have DRM on the brain.
In many of the possible futures you will have DRM in the brain.
After all once they start adding "videographic"+"audiographic" memory as an option, or even mere photographic memory, the Media Monopolizers will start requiring that either you get crippled in order to watch a movie, or you use brain augmentation that has DRM in it.
Then it'll be not a penny for your thoughts, it'll be USD0.99 to recall "their" stuff.
Good luck trying to share experiences with your friends that happen to also included copyrighted content (background music in shopping mall etc).
>Enterprise loads such as databases do many many seeks and tend to have long queues as many clients request the data. Size and throughput are less important for these loads than seek time (though still critical).
Did you even read the link?
"I saw 4KB random write speed drop from 50MB/s down to 45MB/s. Sequential write speed remained similarly untouched. But now I've gone and ruined the surprise."
That's for random writes. 4KB random writes at 45MB/sec is 11520 writes per second.
A 15000rpm drive doing 4KB random writes (noncached/buffered) will only manage about 250 IOPS ( assuming 4 millisecond seek times). Or about 1MBps.
That's 45 times slower. You'll need a lot of spindles to match that.
The only issues I see with SSD are whether reliability is really up to scratch, whether you can hotswap them, and perhaps capacity (if you somehow can't use a tiered storage scheme).
> They can't fire them for an accusation and they can't let them teach if the allegations ultimately turn out to be true.
But such teachers can still teach:
http://slashdot.org/comments.pl?sid=1573766&cid=31382768
Not wonderful, but better than doing nothing.
Seems like a huge waste of time and resources.
;).
;).
Should just get them to continue teaching, but to video cameras. Then after a reviewing process, put those that meet sufficient standards on youtube or wherever.
If they are accused of incompetence at least you would also have recordings to prove whether they are or aren't
Same if they molest the cameras
A fair number of the desktop stuff can take sustained writes for quite a long while- e.g. the entire disk or more.
http://benchmarkreviews.com/index.php?option=com_content&task=view&id=454&Itemid=60&limit=1&limitstart=10
If that's not enough, some of the desktop benchmarks/tests, involve writing to the entire disk first, and then seeing how far the performance drops.
e.g.
http://www.anandtech.com/printarticle.aspx?i=3702
See: "New vs. Used Performance - Hardly an Issue"
They're not cheap, but they sure are cheaper than USD100K.
All teachers need is a high enough salary to live a decent lifestyle (I know it's relative, but I'm sure you can figure out what I mean). Most good teachers have no illusions about becoming millionaires through teaching. They're not stupid after all. Being super rich is not their goal in life.
Good teachers enjoy teaching. Most don't like dealing with loads of admin crap, or politicking.
So you spend some of the money and resources not on high salaries, but on getting most of that crap out of the way.
Where high salaries can come in handy for teachers are: subsidized/free education for their own children[1], and housing loans/allowances (and in the USA, medical/health stuff).
I suggest that it may be cheaper to provide them that than to directly provide them higher salaries.
For example: instead of paying all teachers high enough salaries so that all their children can go to university, do masters, PhD etc, you just commit to paying for any of their children that want to (and meet the grade/entry requirements), and take a gamble that not all their children will want to do so, and not all would want to go to the most expensive universities[2] (and meet the entry requirements). And so I bet you end up paying less overall.
[1] It would be sad and ironic if teachers cannot afford to provide good education for their own children. And I'm sure most good teachers place significant value on education.
[2] and only the approved ones, otherwise people will be setting up "online super expensive university courses"...