Pulled down Apache 2.1 via anonymous CVS.
I think I'll compile it for full debug, light it off, attach to it with gdb, and watch it handle an HTTP request.
1) Why? I want to understand the nitty gritty, and
2) I can. It's perfectly legal not to want to be a sheep all your life.
Not to disenfranchise the sheep, mind you. I totally grasp the need of some to be sheltered from the gnarly details. Fine. Enjoy. Bother me not.
Translation is really at the heart of software.
How do you a) recognize the information in whatever you define as reality, and
b) express that information in another language?
c) maintain that expressed information over time?
The KJV is a great example of a porting attempt that is either a monument to its new language or a dog what won't hunt, depending upon whom you talk to.
The best comment on the effort was offered in advance by one of the source authors, Qoheleth:
"...vanity of vanities; all is vanity. What profit hath a man of all his labour which taketh under the sun?" (Eccl1:2,3)
IMHO, there is more bass in the KJV mix in these 'wisdom' portions, for all more modern translations are easier on the eye.
I wasn't clear. Yes, PL/SQL draws heavily from ADA.
My attempted point was that using packages like DBMS_SQL does little to hide the user from the underlying C implementation.
Never did any Java stuff on Oracle; I presume their JDBC stuff is more modern.
Good points.
The primary target for the technique is something like a web page, where you have to get a bunch of <select> values which are not stable enough to simply cook into a script.
The exact amount of time increase is going to be situation-dependent. Clearly this is not something that should be done with every query, mainly lookup tables, or when you see real benefit from knowing in advance the number of rows for the result.
Uhhh, I'd call PL/SQL the Procedural Language extensions for SQL, which implies quite a bit more than implementation, or perhaps I'm having a pedantic moment.
I think your remark on frustrations is on the money. 80% of all programming languages are the same, the remainder being genuine feature differences and syntactic nuances.
Never did a whole lot in PL/SQL, though. The one or two non-trivial modules I did write seemed so close to the underlying C code (I'm pretty sure that's the source language) that you might as well write in C to begin with.;)
SELECT -1 as Sort_key, count( ID ), ''
FROM tbl_some_table
UNION ALL
SELECT 0, ID, Name
FROM tbl_some_table
Your guaranteed to get something, if only a zero in the first row.
You make one trip to the database, a cheap, yet noticable performance boost.
You know exactly how many rows are in the result.
You aren't too platform-specific.
You can chain a truckload of these (keeping a weather eye on the Sort_key values).
Are there any arguments against this strategy, beyond the usual "that's too advanced for the FNG coders we prefer to hire to maintain"?
Some serious hedz agree with you here
Won't kill you with detail here, but on pg.4 they demonstrate that int const turns out to be a cleaner approach in the presence of typedef.
Regular expressions look like the north end of a south-bound mule in any environment.
Blame not Perl for the fact that your predecessor neglected to type:
#Here is a wild, experimental regex that I thugged off the internet.
#It worked with my simple test file, but who knows if this dog will hunt on prodution.
It'll just beget a new academic field: Nerdiology.
Consider conferences on Geek Culture someday, where Prof. Bipperton Fusslebeak delivers a sad, acedmic commentary on contemporary culture: "An Analysis of the Correlation between Increased Use of Open Source Software, and Slashdot Posts Centered Around Deviant Sexual Behaviors in the Post-.Com Era".
Yeah, but the variable is just a label. The actual type in question is either int or int*.
However, you example clearly shows where 'the C++ way' (and I'm one of these who got into coding after C++ was fairly standardized, and so never have done any 'real' C) is not the preferred answer.
The preferred answer is: knowledge.
...computers are too hard to use and too hard to program.
Sure, hiding the details is a Good Thing.
I think there is a counterargument, though; y'all just can't express differential calculus easily in arithmetic terms.
The genius of MS is in targeting the lowest common denominator. Why does Joe User care about the difference between physical and logical partitions? Redmond has the a:\ b:\ c:\ market firmly in hand.
The utopian vision of 'source code for all my people' will never occur; they can't be bothered (as a whole) to vote, much less think. And we love the sheep nonetheless.
ASCII, UNICODE? English?
Very little research shows that standardization of all kinds promotes the general welfare, though some non-standard implementations are amusing.
It takes little analysis to show that widespread VCR availability actually helped Hollywood, sold pallets of blank tapes, etc.
Now, I want to know what intellectual barrier stands between all of the suits and that truth.
All I can guess is that the suits really only care about the 'tactical' cash they can make, and turn a blind eye to the 'strategic' thugging they give themselves by being so stupid.
OTOH, given the quality vacuum that is the US entertainment industry, it the difference between crap and DRM-wrapped-crap is beneath notice.
I'm happy about this. I bought Might and Magic 3 from Loki (which company died and was reborn as a mind-bending C++ library), and it's about the only game I play.
Don't see what's so hard about compiling a game for multiple targets, especially if you used Qt or something to hide all of the OS personality issues. Targeting Linux might just lure a purchase out of me, for all first-person and real-time games don't byte my naughty bits...
Re:It's really good that filesystems are maturing
on
Reiser4 Benchmarks
·
· Score: 1
As each verbal evolutionary contest ends in a single, 'highlander' word, the list of winners, paired with definitions, must be a dictionary...
I would see what the original article was about, but it is, of course,/.ed. Oops.
What I want to see is material on setting up a project in an intelligent way.
I've got Lakos' book, but there just isn't much out there on non-trivial stuff.
How about articles that show intelligent uses of CVS, Doxygen, and autotools?
What about using C++, boost.python, mod_python, and apache for that killer website?
What about doing work in platform-agnostic languages and components, e.g. Perl/Apache?
ANSI-SQL databases?
The fact that information is not always physical is slightly catastrophic for this metaphor.
You miss the point entirely.
The fact that Andrew Wiles deviant cases abound doesn't negate the value of the article's point, any more than a particular quantum phenomenon renders Newton's equations of motion useless.
While Newton is wrong when enough significant figures are included, these simplified equations do just fine for describing, say, an unladen swallow in flight.
Does mean that I've got to get some work done before meine kleine fraulein takes me off the market...
Slashdot Mirror
+1 Funny.
No, Geddy Lee is not Nancy Wilson, silly rabbit.
Pulled down Apache 2.1 via anonymous CVS.
I think I'll compile it for full debug, light it off, attach to it with gdb, and watch it handle an HTTP request.
1) Why? I want to understand the nitty gritty, and
2) I can. It's perfectly legal not to want to be a sheep all your life.
Not to disenfranchise the sheep, mind you. I totally grasp the need of some to be sheltered from the gnarly details. Fine. Enjoy. Bother me not.
Translation is really at the heart of software.
How do you
a) recognize the information in whatever you define as reality, and
b) express that information in another language?
c) maintain that expressed information over time?
The KJV is a great example of a porting attempt that is either a monument to its new language or a dog what won't hunt, depending upon whom you talk to.
The best comment on the effort was offered in advance by one of the source authors, Qoheleth:
"...vanity of vanities; all is vanity. What profit hath a man of all his labour which taketh under the sun?" (Eccl1:2,3)
IMHO, there is more bass in the KJV mix in these 'wisdom' portions, for all more modern translations are easier on the eye.
The feet of man who uses hypotheticals may no longer be aground.
Never argue with a drunkard, a woman, or a fool.
Proof by analogy is fraud.
I wasn't clear. Yes, PL/SQL draws heavily from ADA.
My attempted point was that using packages like DBMS_SQL does little to hide the user from the underlying C implementation.
Never did any Java stuff on Oracle; I presume their JDBC stuff is more modern.
Good points.
The primary target for the technique is something like a web page, where you have to get a bunch of <select> values which are not stable enough to simply cook into a script.
The exact amount of time increase is going to be situation-dependent. Clearly this is not something that should be done with every query, mainly lookup tables, or when you see real benefit from knowing in advance the number of rows for the result.
Uhhh, I'd call PL/SQL the Procedural Language extensions for SQL, which implies quite a bit more than implementation, or perhaps I'm having a pedantic moment. ;)
I think your remark on frustrations is on the money. 80% of all programming languages are the same, the remainder being genuine feature differences and syntactic nuances.
Never did a whole lot in PL/SQL, though. The one or two non-trivial modules I did write seemed so close to the underlying C code (I'm pretty sure that's the source language) that you might as well write in C to begin with.
Is doing queries like:
SELECT -1 as Sort_key, count( ID ), ''
FROM tbl_some_table
UNION ALL
SELECT 0, ID, Name
FROM tbl_some_table
Your guaranteed to get something, if only a zero in the first row.
You make one trip to the database, a cheap, yet noticable performance boost.
You know exactly how many rows are in the result.
You aren't too platform-specific.
You can chain a truckload of these (keeping a weather eye on the Sort_key values).
Are there any arguments against this strategy, beyond the usual "that's too advanced for the FNG coders we prefer to hire to maintain"?
Some serious hedz agree with you here
Won't kill you with detail here, but on pg.4 they demonstrate that int const turns out to be a cleaner approach in the presence of typedef.
Regular expressions look like the north end of a south-bound mule in any environment.
Blame not Perl for the fact that your predecessor neglected to type:
#Here is a wild, experimental regex that I thugged off the internet.
#It worked with my simple test file, but who knows if this dog will hunt on prodution.
It'll just beget a new academic field: Nerdiology.
Consider conferences on Geek Culture someday, where Prof. Bipperton Fusslebeak delivers a sad, acedmic commentary on contemporary culture:
"An Analysis of the Correlation between Increased Use of Open Source Software, and Slashdot Posts Centered Around Deviant Sexual Behaviors in the Post-.Com Era".
Yeah, but the variable is just a label. The actual type in question is either int or int*.
However, you example clearly shows where 'the C++ way' (and I'm one of these who got into coding after C++ was fairly standardized, and so never have done any 'real' C) is not the preferred answer.
The preferred answer is: knowledge.
I think there is a counterargument, though; y'all just can't express differential calculus easily in arithmetic terms.
The genius of MS is in targeting the lowest common denominator. Why does Joe User care about the difference between physical and logical partitions? Redmond has the a:\ b:\ c:\ market firmly in hand.
The utopian vision of 'source code for all my people' will never occur; they can't be bothered (as a whole) to vote, much less think. And we love the sheep nonetheless.
ASCII, UNICODE? English?
Very little research shows that standardization of all kinds promotes the general welfare, though some non-standard implementations are amusing.
It takes little analysis to show that widespread VCR availability actually helped Hollywood, sold pallets of blank tapes, etc.
Now, I want to know what intellectual barrier stands between all of the suits and that truth.
All I can guess is that the suits really only care about the 'tactical' cash they can make, and turn a blind eye to the 'strategic' thugging they give themselves by being so stupid.
OTOH, given the quality vacuum that is the US entertainment industry, it the difference between crap and DRM-wrapped-crap is beneath notice.
I'm happy about this. I bought Might and Magic 3 from Loki (which company died and was reborn as a mind-bending C++ library), and it's about the only game I play.
Don't see what's so hard about compiling a game for multiple targets, especially if you used Qt or something to hide all of the OS personality issues.
Targeting Linux might just lure a purchase out of me, for all first-person and real-time games don't byte my naughty bits...
I can see where your stated "faith that Moore's law is a long way from failing" is "kind of" in keeping with Moore's law...
As each verbal evolutionary contest ends in a single, 'highlander' word, the list of winners, paired with definitions, must be a dictionary... /.ed. Oops.
I would see what the original article was about, but it is, of course,
What I want to see is material on setting up a project in an intelligent way.
I've got Lakos' book, but there just isn't much out there on non-trivial stuff.
How about articles that show intelligent uses of CVS, Doxygen, and autotools?
What about using C++, boost.python, mod_python, and apache for that killer website?
VBScript, you mean.
In VB, you need:
Next i
What about doing work in platform-agnostic languages and components, e.g. Perl/Apache?
ANSI-SQL databases?
The fact that information is not always physical is slightly catastrophic for this metaphor.
{
c+=2;
}
Driving the c++ concept just a little further down the road to absurdity, we now open the door for:
c4 //a really explosive language
cIV //c++ is now taken intravenously, or is revealed to be a conspiracy run by Sid Meier.
All jokes aside, the future of c++ is boost. Boost is the only way c++ can hope to stay abreast of the scripting languages and Java.
You miss the point entirely.
The fact that Andrew Wiles deviant cases abound doesn't negate the value of the article's point, any more than a particular quantum phenomenon renders Newton's equations of motion useless.
While Newton is wrong when enough significant figures are included, these simplified equations do just fine for describing, say, an unladen swallow in flight.
Does mean that I've got to get some work done before meine kleine fraulein takes me off the market...
Consider that, if it sucks, you can say 'Fsck GNU/Hurd', or just FGH for short.
Nice little sequence there.
Preach it, brother.
The irony is crowned by the fact that RMS proclaims himself atheist.
But, like the Nazarene carpenter, and an Ayn Rand character, he has a gift for tweaking Big Brother and the Holding Company.