You can see the SGMLness shining through in the current XML committee strategies - the underlying notion that XML, like SGML, must be a world unto itself with all imaginable functionality fulfilled through an associated standard (like SGML had the truly loathesome DSSSL).
So what you get is a design-by-committee concept of a programming language expressed as XML...notwithstanding the fact that truly good portable cross-platform programming languages already exist - perl, java, php, etc, ETC, ETC. Of course these languages, unlike XSLT, actually implement most of the programming concepts of the 1960s, like object-orientation. Added to which, all of these languages, even in the worst case, provide performance that XSLT will never match, and a higher degree of readability that XML is capable of (try reading any mildly complex XSL template without getting a headache).
I could go on and on...unfortunately the XML standards bodies have run amok, defining standards for any silly design idea their committees can come up with...its SGML all over again.
but Microsoft's declaration of RTF as a cross-platform document spec
They made no such claim.
For what its worth, RTF has already been consigned to the dustbin of history, and MS itself is pushing XML for the type of applications RTF was once used for.
Okay, so what standards have they subverted? HTML is an obvious answer, but then again, who hasn't subverted HTML? I believe Netscape veered from the standard before Microsoft did.
I don't believe that Microsoft has tinkered with any of the other standards I mentioned. ASCII and UTF are supported by Microsoft in their original forms. HTTP obviously. XML was co-authored by Microsoft, so their support for the standard is implicit.
I understand that MS development tools have not fully embraced the ANSI/ISO C/C++ revisions, but then again, neither has gcc.
On the whole, it doesn't really appear that Microsoft is any better or any worse than their competitors when it comes to the key protocol standards.
I'm fairly sure that site gets hundreds of thousands of hits a day.
Spread out over three or four machines, you could use almost any technology to do run a site that gets under a million hits. Hell, Tcl on one Pentium II could meet the demands for a hundred thousand hits a day.
Its interesting to see the debates that go on when various page generation technologies get discussed here. I think the most insightful point I notice is that not many people get the chance to work on a truly high volume web site to consider the scaling issues.
For most of the slashcode sites out there, and most others that might be discussed, people need to realize that any old technology will do. Even the most braindead architecture will adequately push out pages for a site that gets around 100k hits a day or less - for such a site, any which way you want to generate pages is certainly adequate.
Things get interesting when you start climbing above the ten million hits a day range, when a lot of these technologies really start to fall apart. As intuition would bear out, the simpler technologies do better under greater loads.
It would be interesting to get better simulated load testing for these technologies - I would love to see how PHP, JSP, and mod_perl stack up to old standbys like SSI when the hits peak over 100 million a day or more. Such testing would be beneficial to folks who have to have the functionality of a full programming language, and ultimately need an answer to the question: "how many servers do I need to keep page rendering times under three seconds given this technology?"
This "Java is actually faster than C" stuff has got to go - no one really believes it, the numbers don't bear it out, and its wildly counterintuitive.
As for your qualms about the JRE download issue, get used to it - Sun and Microsoft are never going to see eye to eye, and Microsoft is always going to make it at least a token effort to run Java programs on windows. One could argue that there is no excuse for not including Sun's 1.3 JDK on linux distros - I don't know who is to blame for this, Sun or the distro vendors. One could offer that RedHat has done an end-run around this issue with gcj, but this technology is still somewhat flaky and arguably if you aren't compiling to the JVM, you should probably be using C++ anyway.
Its interesting to note that after five years, Java still hasn't "made it" - Microsoft stalling tactics and general bad attitude may have stalled Java adoption long enough to keep it from becoming a de facto standard.
Let's start off with the assumption that Monica Lewinsky never happened, just to give you the benefit of the doubt.
Clinton's may have presided over the most succesful economic expansion in history, but he did not cause it. Closer to the truth is that policies set up in the eighties lead to both the recession of the early nineties and the resulting boom. If you disagree, name one action by Clinton that you think directly contributed to a strong economy.
Clinton had no foreign policy whatsoever, and it showed. Somalia?? Haiti?? The Gulf?? North Korea?? Has Clinton been able to deal with these situations at all? NO. Lets not even talk about Kosovo - a complete fiasco that Clinton dragged the entire US and EU into, following his botched agenda for creating peace.
Health care? Gays in the military? How many campaign promises did Clinton wuss out on? I'm amazed that gay support groups still support the Democrats after being sold down the river by the Clintons.
Waco. Elian. Filegate. Three great examples of a government ruled by a husband and wife who dine on power.
Sorry, but as an old TeX-head, I can tell you with satisfaction that writing one document in it does not make you literate in its varied commands. TeX does have a ridiculously high learning curve, and added to which it is only a display markup language - it does nothing to infer context and meaning in the content itself, which we've all learned by now is something you want to preserve.
Re:the first mark-down language
on
Perl and .NET
·
· Score: 2
what I know of SAX is parsing. who cares about parsing? it's something that should be simple, which it's not with XML, hence the need to used canned parsers.
I don't understand this - SAX and DOM parsers are canned by various vendors already.
Semantics is what is interesting, but after parsing XML, DOM does not provide rich or useful semantics for manipulating hierarchical data.
You should look at the Sun project to bind elements directly to objects. This is presume is your holy grail - binding arbitrarily complex data types to fully maleable objects directly.
Re:the first mark-down language
on
Perl and .NET
·
· Score: 2
the wacky syntax of DTDs
DTDs are extinct
coupled with the meaningless complexity of the DOMs
qualify this statement. How is the DOM overly complex? Are you aware of SAX, literally a Simple API for XML??
This would have happened anyway.
on
Perl and .NET
·
· Score: 2
It would only be a matter of time beofre before tunnled arbitrary proxies over whatever port they liked.
How long did you think blocking a port would sustain security? Ultimately, all traffic coming across the wire will have to be scanned, likely by highly adaptive heuristics that will attempt to discern the intention of the bits coming across. In this sense, SOAP may be useful - at least with a standard for encapsulating RPC over HTTP, we can strengthen this heuristics pre-emtoively, as the standard is being advanced.
More Idiotic Slashdot Devil's Advocacy
on
Perl and .NET
·
· Score: 2
I'm so tired of listening to the wannabes in here coming off with this devil's advocate BS.
Are you honestly telling me we should drop all the progress made with XML and think of some newfangled LISP mechanism??? Ever heard of DSSSL? Know anyone who uses it???? There's your answer.
As for HTTP, sure its showing age, but you show me a protocol that has become as succesful as quickly as HTTP.
No one cares how whiz bang your technology is if no one adopts it - hence CORBA and LISP.
Has anyone really formed a fundamental bond with a virtual community in the least three years?
Thankfully most of the web media companies have given up any notion of creating viable markets based on the perceived value of being on the same web site as other people.
Of course, numbers still matter when it comes to web auctions, but I think it would be a stretch to call Ebay users a community.
Charisma ? You want to tell me that Mahathma Gandhi, Albert Einstein, Johann Wolfgang von Goethe, George Bernard Shaw, Pablo
Picasso, Salvadore Dali, Steven Hawking, Frank Zappa, Jimi Hendrix have no Charisma ?
I don't know about the rest, but Einstein was no child prodigy.
Added to which, you need to take the "prodigy" stories with a grain of salt. Everyone thinks their child is of above-average intelligence. I obviously have no experience with the child in question, but the presumption that this child is working at "university level" at the age of nine is presumptuous at best, and most likely is more a function of parental ego than actual intelligence.
I've run into a few of these so called "gifted" children over the years, and not one of them has gone on to have an exceptional set of accomplishments as an adult.
The reasons is simple - socialization. Being a successful adult is as much a function of charisma as a function of intelligence.
Added to which, such children are typically treated as freakshow material by their peers, which will ultimately limit their endeavors.
My best advice is maybe bump the kid up a grade or two, keep him stimulated on the side, but don't let him be removed from his normal peer group, and don't let him avoid "mundane" tasks like physical labor.
Turing completeness is not a lofty goal for an instruction/programming system. You'll find Turing completeness in the strangest places - theoretically I can produce any computable result with TeX macros.
There are MORE PEOPLE who will be PAYING CUSTOMERS
Not so far. In fact, Be's deadness is so complete that its doubtful that a free, GPL'd BeOS could even attract mindshare at this point.
Be is simply the next Amiga. People will be playing with Be ten years from now. At least none of the Amiga advocates try to convince you that its going to take over any market - they're a happy clan of hobbyists. I suggest Be advocates adopt this mentality if they want to salvage any dignity.
Even in the "embedded" space where people do not see the technology or its branding, vendors will still go after the technology with the most momentum. I think one can safely say that if Be ever really makes a showing in the embedded market (has anyone adopted Be for a real product??), Be will always have at best 10% of the penetration of linux, their relative merits nothwithstanding.
I'm sorry, but its curtains for Be. Its mindshare is nonexistant, it has no direction, and it will soon be out of cash. Be essentially went out of business the day Apple chose NeXT instead of Be.
Paul - no one elected you policeman of the internet, and now it is clear that some users are being punished for your own idea of what is right and wrong.
You're using your own influence to strong-arm people into adherring to your filters - its that simple. Get off your high horse and try building a constructive way of dealing with spam that does not punish innocent users.
Search is a fickle mistress. I don't know of any search engine that has retained its cool factor for more than a year, and people I have spoken to at Google seem to have an understanding how tenuous their position is.
Lycos, Alta Vista, Inktomi, etc. all have had the cool factor and then lost it just as quickly. Whats next - Infrasearch? FAST?
Governments are already spooked about cyberterrorism - they won't hesistate for a second to shut down any ISP that implements this standard. Even then, the ISPs won't go for this either, as it makes DOS/other attacks even more palatable to crackers.
Basically the only elements of society who want this are the ones furthest away from the decision making process.
Whether your big book of ideals say anonymity is required for you to enjoy complete privacy is irrelevant - the powers that be are never going to allow this to be implemented at the carrier/backbone level, and frankly in the world we live in, this may be a wise choice.
Before you get into a tizzy, for 99% of us, the most intriguing thing we do online is buy things, and this is already tracked through our credit card numbers, so issues of IP tracing are irrelevant.
Slashdot Mirror
You can see the SGMLness shining through in the current XML committee strategies - the underlying notion that XML, like SGML, must be a world unto itself with all imaginable functionality fulfilled through an associated standard (like SGML had the truly loathesome DSSSL).
So what you get is a design-by-committee concept of a programming language expressed as XML...notwithstanding the fact that truly good portable cross-platform programming languages already exist - perl, java, php, etc, ETC, ETC. Of course these languages, unlike XSLT, actually implement most of the programming concepts of the 1960s, like object-orientation. Added to which, all of these languages, even in the worst case, provide performance that XSLT will never match, and a higher degree of readability that XML is capable of (try reading any mildly complex XSL template without getting a headache).
I could go on and on...unfortunately the XML standards bodies have run amok, defining standards for any silly design idea their committees can come up with...its SGML all over again.
They made no such claim.
For what its worth, RTF has already been consigned to the dustbin of history, and MS itself is pushing XML for the type of applications RTF was once used for.
I don't believe that Microsoft has tinkered with any of the other standards I mentioned. ASCII and UTF are supported by Microsoft in their original forms. HTTP obviously. XML was co-authored by Microsoft, so their support for the standard is implicit.
I understand that MS development tools have not fully embraced the ANSI/ISO C/C++ revisions, but then again, neither has gcc.
On the whole, it doesn't really appear that Microsoft is any better or any worse than their competitors when it comes to the key protocol standards.
From what I can see, they support all the important standards. What standards of note would people see Microsoft operating systems support?
Spread out over three or four machines, you could use almost any technology to do run a site that gets under a million hits. Hell, Tcl on one Pentium II could meet the demands for a hundred thousand hits a day.
For most of the slashcode sites out there, and most others that might be discussed, people need to realize that any old technology will do. Even the most braindead architecture will adequately push out pages for a site that gets around 100k hits a day or less - for such a site, any which way you want to generate pages is certainly adequate.
Things get interesting when you start climbing above the ten million hits a day range, when a lot of these technologies really start to fall apart. As intuition would bear out, the simpler technologies do better under greater loads.
It would be interesting to get better simulated load testing for these technologies - I would love to see how PHP, JSP, and mod_perl stack up to old standbys like SSI when the hits peak over 100 million a day or more. Such testing would be beneficial to folks who have to have the functionality of a full programming language, and ultimately need an answer to the question: "how many servers do I need to keep page rendering times under three seconds given this technology?"
As for your qualms about the JRE download issue, get used to it - Sun and Microsoft are never going to see eye to eye, and Microsoft is always going to make it at least a token effort to run Java programs on windows. One could argue that there is no excuse for not including Sun's 1.3 JDK on linux distros - I don't know who is to blame for this, Sun or the distro vendors. One could offer that RedHat has done an end-run around this issue with gcj, but this technology is still somewhat flaky and arguably if you aren't compiling to the JVM, you should probably be using C++ anyway.
Its interesting to note that after five years, Java still hasn't "made it" - Microsoft stalling tactics and general bad attitude may have stalled Java adoption long enough to keep it from becoming a de facto standard.
I await your replies.
Sorry, but as an old TeX-head, I can tell you with satisfaction that writing one document in it does not make you literate in its varied commands. TeX does have a ridiculously high learning curve, and added to which it is only a display markup language - it does nothing to infer context and meaning in the content itself, which we've all learned by now is something you want to preserve.
I don't understand this - SAX and DOM parsers are canned by various vendors already.
Semantics is what is interesting, but after parsing XML, DOM does not provide rich or useful semantics for manipulating hierarchical data.
You should look at the Sun project to bind elements directly to objects. This is presume is your holy grail - binding arbitrarily complex data types to fully maleable objects directly.
DTDs are extinct
coupled with the meaningless complexity of the DOMs
qualify this statement. How is the DOM overly complex? Are you aware of SAX, literally a Simple API for XML??
How long did you think blocking a port would sustain security? Ultimately, all traffic coming across the wire will have to be scanned, likely by highly adaptive heuristics that will attempt to discern the intention of the bits coming across. In this sense, SOAP may be useful - at least with a standard for encapsulating RPC over HTTP, we can strengthen this heuristics pre-emtoively, as the standard is being advanced.
Are you honestly telling me we should drop all the progress made with XML and think of some newfangled LISP mechanism??? Ever heard of DSSSL? Know anyone who uses it???? There's your answer.
As for HTTP, sure its showing age, but you show me a protocol that has become as succesful as quickly as HTTP.
No one cares how whiz bang your technology is if no one adopts it - hence CORBA and LISP.
Thankfully most of the web media companies have given up any notion of creating viable markets based on the perceived value of being on the same web site as other people.
Of course, numbers still matter when it comes to web auctions, but I think it would be a stretch to call Ebay users a community.
I don't know about the rest, but Einstein was no child prodigy.
Added to which, you need to take the "prodigy" stories with a grain of salt. Everyone thinks their child is of above-average intelligence. I obviously have no experience with the child in question, but the presumption that this child is working at "university level" at the age of nine is presumptuous at best, and most likely is more a function of parental ego than actual intelligence.
Didn't anyone see Magnolia??
The reasons is simple - socialization. Being a successful adult is as much a function of charisma as a function of intelligence.
Added to which, such children are typically treated as freakshow material by their peers, which will ultimately limit their endeavors.
My best advice is maybe bump the kid up a grade or two, keep him stimulated on the side, but don't let him be removed from his normal peer group, and don't let him avoid "mundane" tasks like physical labor.
Turing completeness is not a lofty goal for an instruction/programming system. You'll find Turing completeness in the strangest places - theoretically I can produce any computable result with TeX macros.
Not so far. In fact, Be's deadness is so complete that its doubtful that a free, GPL'd BeOS could even attract mindshare at this point.
Be is simply the next Amiga. People will be playing with Be ten years from now. At least none of the Amiga advocates try to convince you that its going to take over any market - they're a happy clan of hobbyists. I suggest Be advocates adopt this mentality if they want to salvage any dignity.
I'm sorry, but its curtains for Be. Its mindshare is nonexistant, it has no direction, and it will soon be out of cash. Be essentially went out of business the day Apple chose NeXT instead of Be.
You're using your own influence to strong-arm people into adherring to your filters - its that simple. Get off your high horse and try building a constructive way of dealing with spam that does not punish innocent users.
Lycos, Alta Vista, Inktomi, etc. all have had the cool factor and then lost it just as quickly. Whats next - Infrasearch? FAST?
That may be a related issue, but anon.penet.fi was shut down in response to specific cases involving child pornography. This is well documented.
Basically the only elements of society who want this are the ones furthest away from the decision making process.
Before you get into a tizzy, for 99% of us, the most intriguing thing we do online is buy things, and this is already tracked through our credit card numbers, so issues of IP tracing are irrelevant.
Unfortuntely, you have no privacy, deal with it.
The Grand Negas has taught you well.