The hell with that. I use trees everyday with Oracle8. There are also 2 methods for tree queries in PostgreSQL.
It's worth remembering that objects are not stored on disk/ram in tree format - it's a matter of query, or how you view the objects. In C++/Java, a tree is viewed by how a function follows pointers. In SQL, all you need is foreign keys, and connect by (which is sort of a 'ORDER BY:PREVIOUS.PK_ID =:CURRENT.PARENT_ID' - which is obviously not part of SQL92, but simple nonetheless.)
Plain and simple, MySQL is faster and has multi-user (at the DBA level) support. It means different applications only get access to the SQL tables they need. It means you can give users their own databases easily. It's incredibly fast. All it lacks is a little transactional auditability and data integrity - for most applications run out here on the net, this isn't an issue, and if it is, you can normally work your way around it.
You're not quite grasping the point here....
Firstly, "MySQL is fastest" - a currently debated issue. I will not step into that mire.
Next, "and has multi-user support" - everyone does. What's your point? Is it "different applications only get access to the SQL tables they need" - we have a words for that: schemas, tablespaces. Every DB out there has them.
"You can give users their own databases" - Again, different schemas. But in MySQL, the system catalogs are page-level locking! How much slower is this multi-user solution than a real RDBMS?
"and if it is, you can normally work your way around it" - you have a piss-poor understanding of concurrency. DB vendors & researchers & universities have been working for decades on this problem. Transaction integrity in a multi-user concurrent environment will not be written in CGI on a weekend.
Please check your hubris and understand why we demand ACID compliant DBs. This stuff is *hard*.
You have some whacked-out population samples. I know many educated people - I spent too many years in school, then went into tech. Some of the educated people I know smoke. Some of them don't. Not many of them started at 15 - we start at college, mostly. I've never met one who has 'suffered some stunting of their growth'. Some of us quit; some of us don't want to quit; I know almost no one who is _unable_ to quit.
I admire your post 2 levels up, where your motivation not to use is that your energy came directly, unaltered, from youth. But you seem to be spending too much of that energy telling others what not to do...
Grandia II, Shenmue, and Skies of Arcadia all come out in the next month or so. Sega should see at least 1 DC acquisition, followed by 3 games, giving them a respectable 1:3 software ratio in my segment.
It's worth noting that local governments allow for the least oversight, tend to have the least voter turnout and citizen participation, and take local biases as fact (whereas a larger body must face disagreeing opinions, eg, segregation vs integration in the 60's). Taken together, these 'qualities' of local governance create the greatest abuses of public land, civic money, and violations of human rights. The federal court system tends to be the normal redress for people shafted by local & state governments.
Obviously all governance cannot be done by the fed, nor should it - locality always has a place, that's the point of representative government. But by being a citizen of a large nation you must learn to share governance with those who disagree with you. You cannot stick your head in the sand like the anarchists and proclaim that "everything will be wonderful once power is vested in a goverment small enough to agree with me."
Secondly, "ramrod socialized medicine through"? Using a nasty verb doesn't convince everyone you're right. Some of us _want_ a single-payer medical system not controlled by HMOs.
does a revolution start with a change in action, or a chasnge in thought?
Is there a difference? Does not a true change in thought inspire action, nor new action bring new insights?
You certainly did write something about many-eyes(and thanks!), which is why I found your statement about Linus so startling.
Back to discussion: I'm afraid I cannot answer your chicken-and-egg problem. I would be happy to listen, though. Do you have an insight into your question?
In Windows, I believe there is less process-creation overhead than in UNIX (assumption based on the non-forked model of windows), so the distinction may not be as great.
Windows process creation time is substantially higher than Unix, which is why MT is so vital to the MS world - it's the only way for them.
Not to mention, perl wasn't designed for the web whereas ASP was.
Considering its usefulness, one might take the converse of your statement and claim that the web was designed for perl...
In the UNIX environment, however, our kernel MT model leaves much to be desired. I vaguely recall reports that showed how poorly Linux did in MT compared to other platforms (obviously Solaris, but I think we even lost to Windows).
In Linux, there is no hard distinction between processes and threads - they differ only in how much local environment vs. shared environment they have. Because of that transparency (that threads can be treated as processes), some of the POSIX thread model does not apply. Solaris is not like this, so the many-to-1 is needed b/c threads cannot be treated as processes. Moreover, I have no specs in front of me, but I recall Linux thread&process creation being substantially faster than Win32.
Hell, Linux isn't even POSIX compliant with MT.
Experience will teach you that POSIX isn't always right.
I've heard that Apache is comming(sic) out with an MT version,
You seem somewhat behind the times. Visit the apache homepage and learn more.
Also, if you want to see a good MT Unix-based webserver, look at Aolserver. Biggest complaint on it is no mod_perl equivalent, but has all the threading issues worked out that you seem interested in.
Cygnus was one of the first (and few) to build a profitable business on supporting open-source products; they certainly reshaped the model in that respect. I have something of a blind eye towards business model innovations, though -- they don't affect me as a source contributor, at least not directly.
Linus has done many wonderful things, but he actually has less claim on this mantle than Tiemann. Linus invented a kernel, not a business model or an ideology -- he supplied the movement's most important object lesson, but didn't invent the movement.
Pardon me? Linus validated the many-developer approach. The BSDs have been typical small-team shops; Cygnus seems more open but less people supply compiler patches (compilers are hard); FSF has typically been.. less than trusting of outside developers (eg, Xemacs). Linus threw the doors wide to chaos, there are perhaps more developers on the linux kernel than on any other open source project. Linus put the lie to the assertion that "geeks can't manage". Linus did not invent a business model, nor an ideology; he validated the massively distributed open source development model, by including more people and input than any other project. (Not to be ad homienum, but I wonder why business model == movement in your mind...) Show me a project larger; that stretched the implementation boundaries of open source more than the kernel.
Speaking of 'most important object lesson', I think Cygnus has a better shot at that -- without their work on gcc, we'd all be fucked, but I do not think they have significantly reshaped the development model.
I suppose that those motivated solely by greed (is there really such an animal?) have no interest in speech. But I wonder: an athlete spends many years training, often living on a subsistence income (at least before the medal roll in), begs and pleads for sponsors, knows that the most minor of injuries will destroy EVERYTHING; and they're in this for the money? Why didn't they go for the MBA track in school? Either they need a few accounting classes; (or maybe you do).
2. The sheer love of performing, if only for a brief moment, at the peak of human capacity. Geeks on/. ought to be able to
relate to this one. Gold medal or not, after dedicating years of your life to a sport, why should you care about possible
restraints on your reporting on it during the competition? That's not nearly big enough to bother risking your future eligibility over.
I suppose cowardice is a valid reason for some.
Keep those heads down! No risk!
`technology without humanity is worse than no technology at all.'
Or maybe they were just parodying Ganhdi's 7 social ills.
moving on....
*This is not to say that nuclear weapons were the wrong thing at the time but too many of the scientists involved weren't considering the outcome of their work.
Given that almost all of them had PhD's, I don't see how 'more schooling' is going to improve broad-thinking. You sound like an ivory-tower bigot -- life lessons are not found only, or even most often, inside school walls.
He even got so desparate he came to my house and banged on my door at 3 am screaming for help because he couldn't get it working (I couldn't either for that matter, but I wasnt' foolish enough to spend a month trying).
You need better friends; or at least smarter ones.
The HUMAN GENOME PROJECT announced its results on Monday, and across the world, scientists asked themselves "Can we play God? What hath we wrought? And where the hell is the data?" The URL given out at the press conference led to a "404 Helix Not Found", and in the days since the announcement, nothing else turned up. Eventually the data slid out late Thursday evening. Pretty disappointing, even when you realise that this is only a beta of an unsorted pile of source code, and they haven't even set up an anonymous CVS server for patches. In the end, eager researchers were left scrabbling around trying to find the final pieces in unofficial archives. Yep: there are, apparently, zero-day genomez sites.
They got a first class fab, the PCI bridge chip business, AND StrongARM. Too bad for them tho that the original StrongARM team up and quit enmass.
Intel got a.35 micron fab when the industry standard was.25, and Intel was moving to.22. That's not what I call first class. Intel had already made PCI chipsets for a while; acquiring DEC didn't vastly change that. And of course the StrongARM team quit enmass -- Intel is a 'hire-em-young-&-screw-em-while-theyre-naive' company, just lke microsoft (and you wonder why both companies have poor quality control). Last I heard a big chunk of the StrongARM guys went to Candence, or some other tools company. Bummer.
...and I don't remember signing any wavers. Although the earplugs came as part of the ticket price.
Now these guys are definitely bad, no question. The coolest thing, IMHO, is that they'll take a few weeks to set up a show, in which time they'll take volunteers from the community to help build their implements of destruction -- then they'll destroy it all. Fun!
Very insightful post; as a layman, you have excellent questions (and in science, the best thing to hear isn't "eureka", it's "that's funny..."). But let's explore a few things about epicycles:
Firstly, they were backed by religion. Everything orbited the Earth, if you recall, and according to both the Church and Aristotle, they orbited in circles (aristotle claimed the solar system was geometrically perfect - each body orbited the earth on a sphere, inside of which was inscribed a regular polygon). One could argue that force of habit is as binding as religion, but probably not win that argument, in tht force of habit does not possess axemen.
Secondly, until Newton's gravity, epicycles were more accurate, ie, they gave results consistent with observation. Once Newton's gravity was on the field, Kepler showed how to compute the correct ellipses, and the simpler argument was more correct.
So two points are worth remembering here - one, that epicycles were the best solution to a geocentric orbital model - a model which could not be discarded b/c of the catholic church; and two, more advanced (ie, complex) physics had to be invented to compute the simpler explanation.
Today there is no religious body powerful enough to squash disagreeable science. World views we have accepted a priori bind our perceptions, but we have learned to question all our instincts these days. Also, simpler explanations don't fall out of a system until we are mired inside a more complex explanation, and made familiar with the system we're trying to model. I recall what a professor once told me:
"You don't ever understand quantum mechanics, you just get used to it."
Sadly, the docs are not free. Now, I'm not really whining about shelling out the $40 or whatever for the 'interbase handbook' or whatnot; my complaints are these:
Timeliness. Programs CHANGE! Open projects change QUICKLY! You must have open docs, or at least docs not committed to paper, to revise them.
Sheer quantity of docs! Ever used Oracle? Oracle8: The Complete Handbook is not all you need! You inevitably end up buying the whole goddamn bookshelf! RBDMS's are complex beasts! With that in mind, closed docs means that such a project can't be called 'free' after all.
firstly, I can only assume you read suck.com daily (as do I, with my morning coffee). So you know that the cyncism typified by Filler sort of pervades the whole site. And while they have much to say on serious subjects (and do so with biting remarks and wit - yesterday's analysis of the GOP convention was killer); on social subjects they often laugh/bitch not at the filth of how we deal with life, the universe, and everything, but with our predictable, often puritanical, sometimes hilarious reactions to such.
So in this case I think the joke is about how self-important the software industry is, rather than crying wolf about corporate espionage. There are more serious activism journalism sites than suck; that's where I would look to see that kind of interpretation.
BTW: I see you go to umich. Are you at the Ann Arbor campus? North campus or Central? God I miss that town -- if only it had a tech economy of decent size.
does no one on this fscking web board understand humor? suck.com's article is parody (and mockery), not insightful reporting on some important-undercurrent-in-software-with-political- ramifications-that-will-doom-us-all.
That was one of the better posts I've seen on slashdot this month (good month for trolls).
Details.. How do we find people who want to learn? The idea of mentoring is wonderful - I always wanted to have a mentor, and I would love to give the experience to someone else. But where? Firstly, your suggested group, those who are about to give up, tend not to be all that visible. That's one of the definitions of 'giving up': dropping out of relevant social circles. Secondly, other groups? I didn't finish high school, electing instead to skip half of it and flee to college. So I'm maybe as disconnected and uninformed as to the average teen's lifestyle as anyone else here. How do we reach out?
The historical version of mentoring usually involved the 'apprentice' if you will ('kohai' if you won't) begging for the mentorship of a proven expert in a craft. These days, there are very many talented people in CS/IT, but most of what we do is hidden, by closed-source, PR, anything-written-for-the-backend isn't visible, etc,etc. Worse if you're a sysadmin, all here know that you (not me! I quit that racket) are invisible unless something goes wrong. So there's no public visibility, and thus very few guiding lights for youths to follow. I think I started rambling; but that's fine too - these are things I'd like to talk about.
Slashdot Mirror
The hell with that. I use trees everyday with Oracle8. There are also 2 methods for tree queries in PostgreSQL.
It's worth remembering that objects are not stored on disk/ram in tree format - it's a matter of query, or how you view the objects. In C++/Java, a tree is viewed by how a function follows pointers. In SQL, all you need is foreign keys, and connect by (which is sort of a 'ORDER BY :PREVIOUS.PK_ID = :CURRENT.PARENT_ID' - which is obviously not part of SQL92, but simple nonetheless.)
Firstly, "MySQL is fastest" - a currently debated issue. I will not step into that mire.
Next, "and has multi-user support" - everyone does. What's your point? Is it "different applications only get access to the SQL tables they need" - we have a words for that: schemas, tablespaces. Every DB out there has them.
"You can give users their own databases" - Again, different schemas. But in MySQL, the system catalogs are page-level locking! How much slower is this multi-user solution than a real RDBMS?
"and if it is, you can normally work your way around it" - you have a piss-poor understanding of concurrency. DB vendors & researchers & universities have been working for decades on this problem. Transaction integrity in a multi-user concurrent environment will not be written in CGI on a weekend.
Please check your hubris and understand why we demand ACID compliant DBs. This stuff is *hard*.
I admire your post 2 levels up, where your motivation not to use is that your energy came directly, unaltered, from youth. But you seem to be spending too much of that energy telling others what not to do...
That's what I'm saying...Grandia II, Shenmue, and Skies of Arcadia all come out in the next month or so. Sega should see at least 1 DC acquisition, followed by 3 games, giving them a respectable 1:3 software ratio in my segment.
I noticed!
...is your Trade Wars server speeding the fuck up! I mean, if I'm going to waste time on addictive door games, why not quickly?!
Obviously all governance cannot be done by the fed, nor should it - locality always has a place, that's the point of representative government. But by being a citizen of a large nation you must learn to share governance with those who disagree with you. You cannot stick your head in the sand like the anarchists and proclaim that "everything will be wonderful once power is vested in a goverment small enough to agree with me."
Secondly, "ramrod socialized medicine through"? Using a nasty verb doesn't convince everyone you're right. Some of us _want_ a single-payer medical system not controlled by HMOs.
Have a good weekend!
If you're willing to live in Sillycon Valley, we could really use some sysadmins (and have you worked/suffered with Oracle before?)...
You certainly did write something about many-eyes(and thanks!), which is why I found your statement about Linus so startling.
Back to discussion: I'm afraid I cannot answer your chicken-and-egg problem. I would be happy to listen, though. Do you have an insight into your question?
Also, if you want to see a good MT Unix-based webserver, look at Aolserver. Biggest complaint on it is no mod_perl equivalent, but has all the threading issues worked out that you seem interested in.
Cygnus was one of the first (and few) to build a profitable business on supporting open-source products; they certainly reshaped the model in that respect. I have something of a blind eye towards business model innovations, though -- they don't affect me as a source contributor, at least not directly.
Speaking of 'most important object lesson', I think Cygnus has a better shot at that -- without their work on gcc, we'd all be fucked, but I do not think they have significantly reshaped the development model.
Keep those heads down! No risk!
Or maybe they were just parodying Ganhdi's 7 social ills.
moving on....
Given that almost all of them had PhD's, I don't see how 'more schooling' is going to improve broad-thinking. You sound like an ivory-tower bigot -- life lessons are not found only, or even most often, inside school walls.
I get the feeling I'm going to be repeating that line fairly often. Who should I attribute it to? Do you have a more substansive bio than 'jms'.
You need better friends; or at least smarter ones.
Find more at ntk.net
Intel got a .35 micron fab when the industry standard was .25, and Intel was moving to .22. That's not what I call first class. Intel had already made PCI chipsets for a while; acquiring DEC didn't vastly change that. And of course the StrongARM team quit enmass -- Intel is a 'hire-em-young-&-screw-em-while-theyre-naive' company, just lke microsoft (and you wonder why both companies have poor quality control). Last I heard a big chunk of the StrongARM guys went to Candence, or some other tools company. Bummer.
Now these guys are definitely bad, no question. The coolest thing, IMHO, is that they'll take a few weeks to set up a show, in which time they'll take volunteers from the community to help build their implements of destruction -- then they'll destroy it all. Fun!
whoops
Firstly, they were backed by religion. Everything orbited the Earth, if you recall, and according to both the Church and Aristotle, they orbited in circles (aristotle claimed the solar system was geometrically perfect - each body orbited the earth on a sphere, inside of which was inscribed a regular polygon). One could argue that force of habit is as binding as religion, but probably not win that argument, in tht force of habit does not possess axemen.
Secondly, until Newton's gravity, epicycles were more accurate, ie, they gave results consistent with observation. Once Newton's gravity was on the field, Kepler showed how to compute the correct ellipses, and the simpler argument was more correct.
So two points are worth remembering here - one, that epicycles were the best solution to a geocentric orbital model - a model which could not be discarded b/c of the catholic church; and two, more advanced (ie, complex) physics had to be invented to compute the simpler explanation.
Today there is no religious body powerful enough to squash disagreeable science. World views we have accepted a priori bind our perceptions, but we have learned to question all our instincts these days. Also, simpler explanations don't fall out of a system until we are mired inside a more complex explanation, and made familiar with the system we're trying to model. I recall what a professor once told me:
Sadly, the docs are not free. Now, I'm not really whining about shelling out the $40 or whatever for the 'interbase handbook' or whatnot; my complaints are these:
Timeliness. Programs CHANGE! Open projects change QUICKLY! You must have open docs, or at least docs not committed to paper, to revise them.
Sheer quantity of docs! Ever used Oracle? Oracle8: The Complete Handbook is not all you need! You inevitably end up buying the whole goddamn bookshelf! RBDMS's are complex beasts! With that in mind, closed docs means that such a project can't be called 'free' after all.
I guess it's 'Raymond Open' not 'Stallman Free'.
So in this case I think the joke is about how self-important the software industry is, rather than crying wolf about corporate espionage. There are more serious activism journalism sites than suck; that's where I would look to see that kind of interpretation.
BTW: I see you go to umich. Are you at the Ann Arbor campus? North campus or Central? God I miss that town -- if only it had a tech economy of decent size.
does no one on this fscking web board understand humor? suck.com's article is parody (and mockery), not insightful reporting on some important-undercurrent-in-software-with-political- ramifications-that-will-doom-us-all.
That was one of the better posts I've seen on slashdot this month (good month for trolls).
Details..
How do we find people who want to learn? The idea of mentoring is wonderful - I always wanted to have a mentor, and I would love to give the experience to someone else.
But where? Firstly, your suggested group, those who are about to give up, tend not to be all that visible. That's one of the definitions of 'giving up': dropping out of relevant social circles.
Secondly, other groups? I didn't finish high school, electing instead to skip half of it and flee to college. So I'm maybe as disconnected and uninformed as to the average teen's lifestyle as anyone else here. How do we reach out?
The historical version of mentoring usually involved the 'apprentice' if you will ('kohai' if you won't) begging for the mentorship of a proven expert in a craft. These days, there are very many talented people in CS/IT, but most of what we do is hidden, by closed-source, PR, anything-written-for-the-backend isn't visible, etc,etc. Worse if you're a sysadmin, all here know that you (not me! I quit that racket) are invisible unless something goes wrong. So there's no public visibility, and thus very few guiding lights for youths to follow.
I think I started rambling; but that's fine too - these are things I'd like to talk about.
Seumas, more thoughts along this vein?