I'd be absolutely terrified to use any software written by someone who believed the above. Transactions are absolutely crucial if you've got multiple processes/threads simultaneously accessing and writing to the database. This tends to be a common scenario for a database sitting behind a website.
Sigh. Have you READ the mySQL architecture documents? I suspect not. You just wrote it off without investigating it. [ Actually, i did too, until recently].
mySQL allegedly does IMPLICIT row-level locking. It explicitly is a multi-threaded database. It has a lot of implicit atomic operations. Here, I'll make it easy for you:
[I'd give a direct link,but I dont want to slashdot the already busy mirrors]
Go to
http://www.mysql.com/mirrors.html
pick the one closest to you.Click the HUGE "documentation" graphic.
Click on the very first link: "MySQL 3.23 Reference Manual (One page per chapter with index)."
I sugest you read the full section 1, to try to shed some of your bias.
Then read seciton 5.4.3, which explicitly addresses the question, "Why does MySQL not support transactions?'"
Oops. Okay, my mistake. I underestimated the problem sets,and didn't properly look at them before posting. There's quite a lot of work there.
Still, though: If someone is interested in working there, they will need to;
1. use they tools they want you to use
2. use them in the manner they want you to use them.
This is a much more practical way of doing so than finding out after a month that they dont like your work.
And you can do the problems sets at your own pace! So I find it odd that you said you "didn't have enough time" to make the code pretty. Was anyone pushing you to submit your work before a certain date?
durnit, if I didnt already just accept a job offer, I would probably be applying at arsdigica myself. sulk.
"extra lengths you ask your job applicants to go to" ?
Sheesh. The link you referenced barely counts as a skills test. (the "home study" page). If someone cannot "pass" that, I wouldn't want them working for me. I'm glad to know that the other Phil doesn't, either.
The onloy catch I see, is intheir mission statement: "... work like a slave"
C) The qualifications (and I took a *real* good look at them, I really want to go!) are a bit unusual - in that they require SAT scores. I miss by 50 points, but isn't that exactly the attitude that you're trying to escape - that you need a standardized test to determine intellegence, that you need cash to determine eligibility? Or am I reading too much into the program?
I didn't see anything like "lets give everyone a university education".
Presumably, he uses the SATs, because it is a SPECIFICALLY TASKED TEST. It is a test supposedly designed to demonstrate how someone's mental skills measure up to the demands of college.
The fact that many colleges are no mental challege at all, doesnt mean that the test itself is invalid. And it even gets usedfor other things. Like one of the three ways you can get admitted to MENSA.
As a comment on sysadmins in general: You can have someone who went to a "teach yourself Novell/MCSE/whatever" course, and passed the test. They finally managed to make something stick in their head. But if they dont have a high degree of native intelligence, they will never make a top-notch sysadmin.
SO, one reason Phil could be doing this, is to ensure that people going around with a "I passed through ArsDigica" aren't going to be in the former set. It would look bad on his institution.
No offense meant to you personally. They had to draw the line somewhere. And you could always retake the SAT:-)
Go read the mySQL FAQ. There is an explicit comparison to postgres. section 22.2
Summary: If it outperforms Postgres by "an order of magnitude" , then who the hell cares if mySQL doesn't fit your definition of a "real" database?
As they point out, for many many things, transactions are only 'needed' for lazy programmers. And adding transaction overhead when you dont need it, slows things down.
Since there are multiple transactional databases, they chose to go for speed instead.
I think the "Obfuscated one" should have wone. When you 'translate' it into clearer english, its the best poem.
Plus, the code is more directly readable as a poem, if you skip the punctuation. PLUS: it actually outputs something, which intrinsically is bound to the meaning of the poem! I think that makes it the most "poetic" of all.
Whereas the "winning" poem was most just a translation job of a poem,that works in english, but doesnt seem particularly 'elegant' in perl to me.
I don't understand this fascination with HDTV, it's pointless. The current NTSC/PAL standards are just fine.
I see HDTV as important, not for TV,but for "other uses".
FOr example, "webtv" almost caught on really well, but is failing for the same reason it seems good: It relies on the "monitor" a family already has, in their standard television set. And a standard television set is LOUSY for displaying text from web pages, etc, etc.
In maybe 4 years, when an HDTV screen becomes more reasonable, then a large percentage of the population will then have high resolution monitors in their house. Then the webtvs of this world will really have something to work with.
Yet another advantage is that some video card makers will soon have units that can handle the resolutions of HDTV, so you could play your favorite games at 720P or 1080I on a bigscreen. Sweet!
I got to play with a playstation 2 in japan, with teken III on a philips HDTV. Very nice.
So are there any widely used libraries that make C coding CGI easier?
I know that at one point ncsa came with binary utilities for cgi scripts. but I dont remember any C libraries for the same task. Methinks a library in the class of CGI.pm, could seriously replace perl for people who know what they are doing.
Make stored proceedures for your SQL, so it will run faster
from
Make everything a stored proceedure so it will run faster
You seem to be saying the second. But only the first is true. There are better ways of optimizing non-SQL logic, than to bog down your database server with it.
Well, actually, some people might debate whether or not my code is "open source". It ain't GPL. But it IS in the debian distro.
Anyways, my reasons for my multiple projects have varied.
My FIRST project started because I wanted a kanji quiz program for UNIX, and it just didn't exist. I kept building onto it, even though I wasnt using it, because
It was teaching me loads about X programming
People emailed me saying "Hey, this is COOL! (But could you add this too?). This was after I myself had stopped using it, because I was too busy to study kanji:-)
I'm a perfectionist
Now I keep adding to it because
I'm actually using it now, so find features lacking
People occasionally still email me with requests [BIG motivator!]
I'm a karma whore:-)
And then there's a small driver project that I wrote, that had different motivation priorities:
I'm a karma whore
I wanted driver-writing experience
But I never had much need for it myself. And I never had people email asking for features. So,it just sits there, with one or two downloads a week. Sure is funny seeing mac and windows browsers download source to a UNIX driver, though.
I base my "worst case" scenario on what other people have already pointed out: to take legal action, requires legal standards of proof. This doesn't meet those standards.
So a student randomly accused is not going to jail, nor can they even be expelled (from a public school) without justifiable reason.
Accusations are going to be made, and they will be taken seriously. These accusations will affect someone's life, and they need to be made and handled responsibly.
HOW are they going to affect someone's life? They're NOT going to go to jail or anything for this.
Worst case for someone who just had a "bad day": they have to sit through a counselling session or two. SO WHAT? Big deal.
If you're mouthing off stupid things like "I'm going to punch out that teacher", then you probably SHOULD go to an anger management session.
On the up side, some real bullys will have to sit through counselling sessions, too. That's a good thing.
What I find interesting, is that all the big commercial UNIX vendors are listed as "participants" (SCO, Sun, HP, IBM), but there are NO LINUX VENDORS participating.
O.K. State the obvious. Lets take a look at kindergarten play time. They boys play dominance games. Which boy is the strongest or fastest. The girls play cooperative games where there is no winner like house.
Oh, but you're so wrong. There IS a winner in "house":-) Or at least, after the first year or two of them playing.
Boys play on physical, concrete superiority.
Girls play mind games. The girl who rules the most minds, is the winner.
I have to agree. The "Internet as [I knew] it" is already dead.
Used to be, USENET was 50% signal (with the other 50% being the actual questions) And the net was a place where there were useful sites, focused on information, not animation
Now USENET is 60% spam, and 35% noise. Similarly, the vast majority of websites think the most important component of a website is to have a flash/shockwave/javascript animated clock. What? You wanted actual useful information? Better try somewhere else.
The trouble with "cool features" like views.. Is that then people start using them.
ANd then 4 months later, they suddenly complain that database response is lousy. [since people have this nasty tendancy to create views that span multiple tables]
Personally, I think that the only reason that "views" should exist, is to get column-level "security". But I hope that one day TRUE column-level security should be implemented instead, and then views should be dumped.
I'm using MySQL. But damn, the thing sucks for everything more than simple selects. Plus selects are sometimes NOT optimised at all.
That, I think, is the basic summary. You have to do your own optimization. If you want to get full auto-optimization, buy a commercial database, for goodnessake:-)
That being said, I've been playing with mysql a little bit. And if you pre-test your selects, and do some careful table balancing (sometimes sacrificing space by lots of redundant tables) it seems to perform fairly well.
For example, three-way joins are just brain-dead. But if you keep an "All data" table, and then multiple cut down copies of it, nicely indexed, so you can restrict all your selects to only a two-way join... Then it seems pretty good.
Slashdot Mirror
I probably should have replied to YOUR article, not the AC one. So, here's a reply to tickle your "user posting history" reply flags.
;-)]
[That's how I know to check back on replies to MY postings, anyways
Please check my reply to the stubborn AC with this same subject.
Or, just comment on mySQL FAQ section 5.4.3, in reply to this article.
Sigh.
Have you READ the mySQL architecture documents? I suspect not. You just wrote it off without investigating it. [ Actually, i did too, until recently].
mySQL allegedly does IMPLICIT row-level locking. It explicitly is a multi-threaded database. It has a lot of implicit atomic operations. Here, I'll make it easy for you:
[I'd give a direct link,but I dont want to slashdot the already busy mirrors]
Go to
http://www.mysql.com/mirrors.html
pick the one closest to you.Click the HUGE "documentation" graphic.
Click on the very first link:
"MySQL 3.23 Reference Manual (One page per chapter with index)."
I sugest you read the full section 1, to try to shed some of your bias.
Then read seciton 5.4.3, which explicitly addresses the question, "Why does MySQL not support transactions?'"
Oops. Okay, my mistake. I underestimated the problem sets,and didn't properly look at them before posting. There's quite a lot of work there.
Still, though: If someone is interested in working there, they will need to;
1. use they tools they want you to use
2. use them in the manner they want you to use them.
This is a much more practical way of doing so than finding out after a month that they dont like your work.
And you can do the problems sets at your own pace! So I find it odd that you said you "didn't have enough time" to make the code pretty. Was anyone pushing you to submit your work before a certain date?
durnit, if I didnt already just accept a job offer, I would probably be applying at arsdigica myself. sulk.
"extra lengths you ask your job applicants to go to" ?
... work like a slave"
Sheesh. The link you referenced barely counts as a skills test. (the "home study" page).
If someone cannot "pass" that, I wouldn't want them working for me. I'm glad to know that the other Phil doesn't, either.
The onloy catch I see, is intheir mission statement:
"
C) The qualifications (and I took a *real* good look at them, I really want to go!)
are a bit unusual - in that they require SAT scores. I miss by 50 points, but isn't that
exactly the attitude that you're trying to escape - that you need a standardized test
to determine intellegence, that you need cash to determine eligibility? Or am I
reading too much into the program?
I didn't see anything like "lets give everyone a university education".
Presumably, he uses the SATs, because it is a SPECIFICALLY TASKED TEST. It is a test supposedly designed to demonstrate how someone's mental skills measure up to the demands of college.
The fact that many colleges are no mental challege at all, doesnt mean that the test itself is invalid. And it even gets usedfor other things. Like one of the three ways you can get admitted to MENSA.
As a comment on sysadmins in general:
You can have someone who went to a "teach yourself Novell/MCSE/whatever" course, and passed the test. They finally managed to make something stick in their head. But if they dont have a high degree of native intelligence, they will never make a top-notch sysadmin.
SO, one reason Phil could be doing this, is to ensure that people going around with a "I passed through ArsDigica" aren't going to be in the former set. It would look bad on his institution.
No offense meant to you personally. They had to draw the line somewhere. And you could always retake the SAT
Go read the mySQL FAQ. There is an explicit comparison to postgres.
section 22.2
Summary: If it outperforms Postgres by "an order of magnitude" , then who the hell cares if mySQL doesn't fit your definition of a "real" database?
As they point out, for many many things, transactions are only 'needed' for lazy programmers. And adding transaction overhead when you dont need it, slows things down.
Since there are multiple transactional databases, they chose to go for speed instead.
Okay, "an important point", I can see, but "the main point"???
Sheeesh. Look no further for what is wrong with the american schooling system.
I think the "Obfuscated one" should have wone.
When you 'translate' it into clearer english, its the best poem.
Plus, the code is more directly readable as a poem, if you skip the punctuation.
PLUS: it actually outputs something, which intrinsically is bound to the meaning of the poem!
I think that makes it the most "poetic" of all.
Whereas the "winning" poem was most just a translation job of a poem,that works in english, but doesnt seem particularly 'elegant' in perl to me.
Excellent. Concise, and to the point.
Thanks.
I see HDTV as important, not for TV,but for "other uses".
FOr example, "webtv" almost caught on really well, but is failing for the same reason it seems good: It relies on the "monitor" a family already has, in their standard television set. And a standard television set is LOUSY for displaying text from web pages, etc, etc.
In maybe 4 years, when an HDTV screen becomes more reasonable, then a large percentage of the population will then have high resolution monitors in their house. Then the webtvs of this world will really have something to work with.
[And PS2 games will kick ass ;-)]
I got to play with a playstation 2 in japan, with teken III on a philips HDTV. Very nice.
Thanks, but that's C++.
I specifically asked for a C library.
So are there any widely used libraries that make C coding CGI easier?
I know that at one point ncsa came with binary utilities for cgi scripts. but I dont remember any C libraries for the same task.
Methinks a library in the class of CGI.pm, could seriously replace perl for people who know what they are doing.
You seem to be saying the second. But only the first is true. There are better ways of optimizing non-SQL logic, than to bog down your database server with it.
More correctly:
Sun HARDWARE is expensive.
Solaris is not.
($75 is expensive??)
Anyways, my reasons for my multiple projects have varied.
My FIRST project started because I wanted a kanji quiz program for UNIX, and it just didn't exist. I kept building onto it, even though I wasnt using it, because
It was teaching me loads about X programming
People emailed me saying "Hey, this is COOL! (But could you add this too?). This was after I myself had stopped using it, because I was too busy to study kanji :-)
I'm a perfectionist
Now I keep adding to it because
I'm actually using it now, so find features lacking
People occasionally still email me with requests [BIG motivator!]
I'm a karma whore :-)
And then there's a small driver project that I wrote, that had different motivation priorities:I'm a karma whore
I wanted driver-writing experience
But I never had much need for it myself. And I never had people email asking for features. So,it just sits there, with one or two downloads a week. Sure is funny seeing mac and windows browsers download source to a UNIX driver, though.
For the bored: my nick is my domain
I base my "worst case" scenario on what other people have already pointed out: to take legal action, requires legal standards of proof.
This doesn't meet those standards.
So a student randomly accused is not going to jail, nor can they even be expelled (from a public school) without justifiable reason.
HOW are they going to affect someone's life? They're NOT going to go to jail or anything for this.
Worst case for someone who just had a "bad day": they have to sit through a counselling session or two. SO WHAT? Big deal.
If you're mouthing off stupid things like "I'm going to punch out that teacher", then you probably SHOULD go to an anger management session.
On the up side, some real bullys will have to sit through counselling sessions, too. That's a good thing.
What I find interesting, is that all the big commercial UNIX vendors are listed as "participants" (SCO, Sun, HP, IBM), but there are NO LINUX VENDORS participating.
What's up with that?
What are you talking about??? Post a quotation where there is ANY intent to support the ability of non-individuals to influence the political process.
Oh, but you're so wrong. There IS a winner in "house" :-) Or at least, after the first year or two of them playing.
Boys play on physical, concrete superiority.
Girls play mind games. The girl who rules the most minds, is the winner.
I have to agree. The "Internet as [I knew] it" is already dead.
Used to be, USENET was 50% signal (with the other 50% being the actual questions)
And the net was a place where there were useful sites, focused on information, not animation
Now USENET is 60% spam, and 35% noise. Similarly, the vast majority of websites think the most important component of a website is to have a flash/shockwave/javascript animated clock. What? You wanted actual useful information? Better try somewhere else.
The trouble with "cool features" like views..
Is that then people start using them.
ANd then 4 months later, they suddenly complain that database response is lousy.
[since people have this nasty tendancy to create views that span multiple tables]
Personally, I think that the only reason that "views" should exist, is to get column-level "security". But I hope that one day TRUE column-level security should be implemented instead, and then views should be dumped.
I'm using MySQL. But damn, the thing sucks for everything more than simple selects. Plus selects are sometimes NOT optimised at all.
That, I think, is the basic summary. You have to do your own optimization. If you want to get full auto-optimization, buy a commercial database, for goodnessake
That being said, I've been playing with mysql a little bit. And if you pre-test your selects,
and do some careful table balancing (sometimes sacrificing space by lots of redundant tables)
it seems to perform fairly well.
For example, three-way joins are just brain-dead.
But if you keep an "All data" table, and then multiple cut down copies of it, nicely indexed, so you can restrict all your selects to only a two-way join... Then it seems pretty good.
Your "own projects"?
Are they public projects?
If the answerisNO, then you could always get "free" solaris, and then run "free" oracle x86 on it.
$85for media+shipping&handling, doncha know. Sun's larger "free" package.