Your description makes it sound like it should work. But it doesn't. If I switch to Windows Explorer in Windows, I can't Ctrl+Tab to the next Explorer window. It just doesn't work. I can Alt-Tab my way across the different Explorer windows though, even though they're all the same application.
On OS X, I can CMD-Tab until Finder is selected. This is much faster, because it's by application and not the individual window. When I get to Finder, I can then hit CMD-` until I reach the Finder window I want. The difference in time to find the correct window is:
Windows: O(W) Mac OS X: O(A + AW - 1)
Where W is the total number of windows open, A is the total number of unique applications open, and AW is the number of windows per unique application. O(W) is worst case. i.e. We may have to switch through every window in the system before finding the correct one. O(A + AW - 1) is far better, because you tend to only have a few applications open, but many windows per app. The worst case in this instance is that there is one window per application.
Using my current XP desktop as a common case, I find that I have 10 applications with an average of 3 windows per application. That gives an O(W) worst case of about 10*3 or 30 switches. Using O(A + AW - 1), however, I get a worst case of 10 + 3 - 1, or 12 switches. The OS X method is 60% faster!
IPF ships with Solaris 10 and Solaris 8/9 had the Sun Firewall on one of the CDs.
My machine still runs Solaris 8, and I haven't played enough with 10 to know. Thanks for the update.:-)
I don't remember a CD for Sun Firewall, but it's possible they added it later. Sun tends to update their software distributions once per quarter, so it can be difficult at times to say for certain what a given OS version has bundled.
IPtables is very convoluted, powerfull but screwy. PF (OpenBSD/FreeBSD) is the best IMNSHO.
Amen to that. My Sun Box at home is sitting behind a FreeBSD firewall/NAT router. The thing is amazingly simple to keep running.:-)
Alt-Tab (its actually command tab) didn't come into its own on OSX until 10.3 (IIRC, could be 10.2)
10.2. I'm terribly lazy about upgrading.;-)
when they introduced the bezel style application switcher.
You mean a popup window of sorts? In 10.2, the dock would highlight the app. Works fine. Including the Shift for backwards. What tends to frustrate users is the misunderstanding of how Cmd+Tabbing works. They think that it's supposed to work like Windows with every window selected. Instead, it switches between applications, then you hit CMD+` to switch between the windows of the application. Microsoft Windows has nothing quite like this little feature. Oh, and if it has been removed from 10.4, I'm gonna be pretty pissed.
but there are some inconsistencies that drive users crazy (like dialogue boxes that don't have full keyboard access).
Indeed. One would expect the arrows/tab and enter keys to work, but often they don't. That is one of my pet peeves.
I think you need a better example, windows has since 3.0 (maybe longer) the standard where alt-command works on the global application space while ctrl-command works on the application space.
As I just told another user, it doesn't work. There may be a "standard" at the application level, but there's nothing at the OS level that makes Ctrl-Tab work. In addition, Alt-Tab tabs through EVERY window, which can be quite tedious when you have a lot of them up.:-)
And where exactly do you *buy* a supported Fedora release from Redhat?
I think he was referring to the RedHat 9 debacle where RedHat officially discontinued their Desktop line, and sent them all to the Fedora project. This forced anyone who wanted a supported Linux desktop to purchase RedHat Enterprise or find another vendor.
I don't think the situations are comparable, but it's at least fair to understand the reference properly.
Ctl-Tab - cycles through currrent applications windows/documents
I think you may be confused. I just tried that on my Windows XP machine. It worked in Mozilla (switching each tab), but it failed miserably for every other program I tried.
Alt-Tab - cycles through running applications
Alt-Tab cycles through every open Window, not every open Application. Again, I think you're confused.
Depends on the architecture - Solaris on Sparc makes sense, why anyone would choose Solaris over Linux on an x86/amd64 platform is beyond me. (I support both at work, and aim to replace as many Solaris boxes as possible with Linux.)
Depends. If it's an actual Sun-built box, then it's just as good as a Sparc box. If it's Solaris/x86 on a random Dell or something, then I agree completely.
Fair enough. You do have to go with commercial solutions on Solaris.
- proper package management (not that red hat provides that, but compared pkgadd anything seems good)
This has got to be a personal preference. While Sun may send out patches in huge (and annoying) bundles, at least you can always get the right package without any troubles. I can't count the number of times I've been stuck in RedHat RPM HELL. No, no. I need to bury the memory. It's too painful. Too... THE HORROR!
- real command line utils (GNU), ie:
ACK! You're not one of those guys who installs GNU utilities on the Solaris server, thus breaking everyone's scripts are you? Gah! I hate it when people do that!;-)
Seriously, many people use GNU utilities on Solaris without complaint. I personally prefer the old-style Unix utilities, but maybe that's just because I'm used to them.
- color ls
I absolutely HATE color ls. It looks so... toyish and loud. I much prefer the Solaris standard of using underlines, brightness, bolding, and italics to represent the different attributes of the file. It works so well on Solaris because even the console is Black text on Pure White graphics mode.:-)
- tar with gzip, bzip support (no need for piping)
Oh, like piping is going to kill you. Why, in my day...:-P
Unix, not Linux. GroupThink(TM) says that Unix is inferior to the great and wise Linux. (Excuse me while I hurl here.)
Most well trained admins prefer a Unix server box over a Windows server box. Unfortunately, Windows servers were dictated by managers and MSCEs (i.e. People who thought Microsoft was "cool", but really didn't know what the hell they were doing.) I'm willing to bet that part of the reason that Linux has caught on in the server arena is because it's close enough to a true Unix, yet "cool" enough at the moment to get accepted.:-)
My point still stands, though. Mac OS X has a lot of little "nice" things that add up to a completely "nice" Desktop. That's why people like them. It's not just the Jobs Reality Distortion Field. Or if it is, he's managed to make it permanent this time.;-)
Heh. Good point. I've never quite been able to break the habbit of calling it "Alt-Tab" even though Alt has nothing to do with it. It works out, though, because Windows users seem to understand the terminology better.:-)
Sun developing for linux is ultimately counterproductive to its own long term future.
That's not entirely true. Sun's strategy has always been to sell hardware and complete solutions. It really doesn't matter to them if they're selling Linux or Solaris. In fact, long before JDS they provided an option to preload RedHat on many of their systems. Why anyone would chose RedHat over Solaris for a server system is beyond me, but a lot of customers were demanding it.
All the JDS is lacking is a true follow through. This half-assed release-it-and-then-drop-it strategy is guaranteed failure.
The big question is what happened to those half-million to million-plus units that were supposed to ship in China in 2004?
What about those of us here in the US who *paid* for JDS and were promised major upgrades every quarter? We saw the JDS 1.0 -> 2.0 upgrade, then it stopped while Sun worked on JDS/Solaris.
Sun needs to learn that the only way they're going to make inroads into the desktop market is if they follow through. Rome wasn't built in a day, and neither was the popularity of Java or Solaris. If Sun would take the time to listen to their customers and implement the features they are demanding, then they'd have a very good chance at success. *sigh*
Why are so many Slashdot users in love with Apple?
Why are so many users in love with Google? Why does everyone prefer Unix servers over Windows servers?
The answer is that it Just Works(TM). Apple has made Mac OS X an extremely pleasent environment to use with little things like Alt-TAB through applications, then Alt-` through an application's windows. These little things add up into a much nicer user environment.
Same thing with the Desktop search. Google, Jeeves, MS Indexed Find, and other search technologies just don't compare to Spotlight. Spotlight digs into the file and generates all the meta-data automatically. From that meta-data, it then generates indexes that make the search lightening fast. And it does it all without compromising system integrity, system security, or finding "ghost" files.
Honestly, if you haven't used an Apple, consider getting one or borrowing one. After just a little bit of everyday use, I think you'll find what everyone is raving about.:-)
Why don't you use the.responseXML attribute, which gives you a DOM object,
Primarily because of browser differences. The IFrame solution seems to be more portable than responseXML. The other half of it is that I'm still doing development and have temporarily eschewed XMLHttpRequest because I'm not running a server. (It's all internal to itself.) Using an IFrame allows me to perform architecture tests on new applications straight from the files on disk.:-)
instead of "shunting" it into an IFRAME, thus losing all hope of having more than one request at a time?
That is one annoying point to the IFrame solution. However, Multithreading in a browser can be a difficult thing to deal with, so it's something I'm trying to avoid for now. Currently I use a linked list of items to load so that they all get loaded in sequence. It works quite well for the moment, and can easily be switched over to XMLHttpRequest when I'm ready.
Desktop search is the voice recognition of the new century. It will sort of work, but never well enough to make it worth relying upon.
A nice prediction, except for one problem: Apple users are already using Desktop Search. It's here now, it works, and it's much loved by users. Same thing with Google Web Apps. GMail, GMaps, and Google Search are all here today, all much loved by users, and all wiping the deck with competitors.
Voice Rec was one of those things that we always saw coming, but never saw the reality of. (Although it has gotten into niche applications like voice dialing.) The threats to Microsoft, OTOH, are already banging at the gates (ha ha) and are threatening Microsoft's bottom line. Unless Apple's and Google's growth were to abruptly stop tomorrow, even conservative projections don't look good for Microsoft.
Saying things like that are a calculated gamble, words like that can send stock prices down, so there has to be a reason for it. "Honesty" aside, it is business.
If that's true, then the gamble requires that Microsoft have something up their sleeve to help them have a "fighting comeback" in the marketplace. The problem is that Microsoft has never been very good about keeping their mouth shut about future developments. Which means that the only thing in their pipeline right now is Longhorn. Now just about every feature that could actually let Microsoft compete is getting stripped out of Longhorn, thus leaving them with nothing more than a few whiz-bang features.
Ballmer may really believe that Longhorn is going to take the world by storm, but my gut feeling is that Microsoft is doomed to irrelevency. Longhorn will be more of the same, with no acknowlegement of the paradigm shifts Apple is pushing onto the desktop and Google is pushing into Internet apps. The result will be that Microsoft will begin losing their desktop dominance to Apple and their Internet dominance to Google/FireFox, which will leave Microsoft in the position of having to become a cross-platform application provider, again.
"Take for instance the Siebel database. Now I've never used that interface. But I'd love to go to it and say 'who is the account manager for the Commonwealth Bank of Australia?'," Ballmer told the partners.
I can say one thing for sure. He's DEFINITELY never used the Siebel interface!;-)
This article honestly sounds like Ballmer was getting a bit beat up by Microsoft's partners and shareholders. They've basically gotten him to admit that.NET is.NOT, Microsoft can't even search its own desktop (Quote: "It's important for people who search a corporate network,"), and that SQL Server development has ground to a halt (ceding victory to Oracle). He then goes on to make a set of pathetic promises ("In the next six months, we'll catch Google in terms of relevancy," and, 'This may be addressed in the next release [of SQL Server] in 18 months, Ballmer said, but conceded he "really didn't know",' and, "Government has really been pushing for stronger interoperability. We can't support open source, but we can support interoperability,") and say that Microsoft will never give up the fight.
I'm sorry, but Ballmer has effectively admitted that Microsoft is now irrelevent. He's trying to grip at pavement by muttering about interop and standards compliance. This is an amazingly similar situation to the introduction of Netscape Navigator. Microsoft almost missed the boat then, but managed to throw enough resources, money, and outright theft behind capturing the browser market. Microsoft's best attempts today only come out as a pathetic whimper. No super-search engine, no desktop search, nothing. If Ballmer was smart, he'd get his boys to activate the existing Databasse File System in NTFS, then use it to push Google and Apple away from the Desktop. Once solid in that area, they should tie it into their online search engine, thus using their desktop monopoly against their competitors.
On the bright side, I am quite glad that Microsoft isn't that good anymore. At the very least, they have to watch where they step with the justice department looking over their shoulders.:-)
Either you didn't try hard enough, or you didn't know enough
Geez. Who stuck a bee in your bonnet?
It's quite a hack to have an IFrame for this and all that. XmlHttpRequest is cleaner because you can read headers , set headers and even check on status code on return.
Do you have any idea what the subject is even about, or do you just like to rant? The parent poster was talking about parsing XML directly and how slow it was. (You can get an XML string via XMLHttpRequest, BTW.) I responded that I don't worry about the speed because I don't do it in JavaScript. I shunt the data to a hidden IFrame (which can still be done with XMLHttpRequest if you understand document.open(), which I'm guessing you don't from your senseless rant), let the browser parse it for me, then I walk the W3C compliant DOM tree. There's practically nothing faster or easier to do.
I've been using XmlHttpRequest extensively recently - it's a surprise nobody noticed it until google brought out GMAIL. Now it's the latest Buzzword that you NEED on your resume to get a job as a web-dev (which I'm not).
I was doing it before it was cool. What's your point? Just because you just found XMLHttpRequest and think it's God's gift to web developers, doesn't mean that it's the ONE TOOL for everything. Good Web apps are a lot more complex and difficult to get going. For example, have you ever shunted a JavaScript program into an existing program in a pluggable and user runnable fashion? I have. In fact, I wrapped the entire program in XML Meta-Data so that I could provide Icons, Associations, and Security Features before the program was ever loaded into memory. My program then sits as a text string in a psuedo-filesystem until it's actually ready to execute. Then I pull some wizardry to run the program without touching the HTML DOM. Would you like to hazard I guess as to how it's done, oh great and powerful jackass^W web wizard?
Has anyone actually tried parsing XML using javascript??
Yes and no. I usually let the browser do the parsing in an IFrame, then I just walk the DOM. It's much easier than taking a string of XML and trying to break it down like a good parser should. It's far too tempting to cheat like hell and do stuff like recursively find the outer tags.
Surprisingly, there's only one flip phone I've ever wanted to get. It was this Nokia model that was like a "Candy Bar" phone on the outside, but would flip open to reveal a full keyboard and joystick. Plus it had a built in FM radio (although it was really the keyboard that had me interested). Unfortunately, I need a tri-mode phone because of the areas I travel through, so the phone I *really* wanted wasn't an option.:-(
In the middle of the presentation there's a slide titled "Changing the Game in Candy Bars" with some cool phones in the background. [...] Candy bars, I would guess, are a fairly stable commodity. A Mars bar last year is going to be the same as this year. Eye candy, sure, but not candy bars.
That has got to be one of the dumbest attempts at setting marketting goals that I have ever heard. Candy Bars are consumables. You buy it, eat it, then you don't have it any more. You don't eat your cellphone. (Unless that's what they're trying to fix.;-)) A cell phone is a device with an expected lifespan, value, and deprecation schedule. There's no way you can compare the two.
BTW, I love all the "Wow!"s and "Whoa!"s in the background. It just further emphasizes how little content is actually in this presentation.;-)
Re:Pilots are *ALWAYS* better than the show
on
P2P and TV
·
· Score: 1
Remember the original Star Trek series (or were you even born then)? There was a 2-part special with Captain Christopher Pike (a basket case with only his head functional) and the crew of the Enterprise being shanghaid to a "forbidden planet". Apparently, *THAT* used significant chunks of the Star Trek pilot.
I wasn't there for it, but I know the episodes you're referring to. I'm actually surprised you haven't seen the unaltered pilot. CBS airred it back in the early 90's (or was it late 80's?) as a "Lost Episode". Of note is that the ship made no "whoosh"ing sound, parts where in color vs. b&w, Spock was not the first officer, and Majel Roddenberry (Nurse Chapel, Luxawanna Troi, Computer Voice) played Number One.
The most important thing to understand about the pilot episode was that it was just too forward thinking for the networks. A thoughtful captain? A female first officer? A hardcore Sci-Fi plot? The network just wasn't ready for such depth and demanded that it be revamped into more of an action/adventure show. Roddenberry complied, but he still kept many of the background commentary elements that made Star Trek so unique.
You'll note that Captain Picard turned out to be much more of a Captain Pike type of character. By the time TNG aired, the world had changed enough to allow for a more thoughtful show. It took TNG a little while to grip pavement (the TNG characters started very raw) but once it got going, we got such amazing episodes as Best of Both Worlds, Darmok, Inner Light, Yesterday's Enterprise, etc.
After a while it got old, and I stopped watching, as did many others.
Spock's Brain had nothing to do with it? No, the real reason you stopped watching was the same reason why the quality of Star Trek was dropping. The network was ticked that Gene was getting away with such controversial content, and replaced him with a different producer. That producer was responsable for making many of the more "empty" episodes at the series' end. A few months later, NBC looked at the new targetted audience numbers and started kicking themselves for removing Roddenberry and killing Star Trek.
To your original point, Pilots are often worse than the full show because they are very raw and without development. Star Trek was the exception because the changes were actually away from what everyone really wanted.
If only/. editors could develop a method for fusing dups.
I'll have you know that this is NOT a dupe! The first article was in the Science section. This article is in the Hardware section! It must mean that Nuclear Fusion will be coming to power our computer hardware soon! Yippee! </sarcasm>;-)
Your description makes it sound like it should work. But it doesn't. If I switch to Windows Explorer in Windows, I can't Ctrl+Tab to the next Explorer window. It just doesn't work. I can Alt-Tab my way across the different Explorer windows though, even though they're all the same application.
On OS X, I can CMD-Tab until Finder is selected. This is much faster, because it's by application and not the individual window. When I get to Finder, I can then hit CMD-` until I reach the Finder window I want. The difference in time to find the correct window is:
Windows: O(W)
Mac OS X: O(A + AW - 1)
Where W is the total number of windows open, A is the total number of unique applications open, and AW is the number of windows per unique application. O(W) is worst case. i.e. We may have to switch through every window in the system before finding the correct one. O(A + AW - 1) is far better, because you tend to only have a few applications open, but many windows per app. The worst case in this instance is that there is one window per application.
Using my current XP desktop as a common case, I find that I have 10 applications with an average of 3 windows per application. That gives an O(W) worst case of about 10*3 or 30 switches. Using O(A + AW - 1), however, I get a worst case of 10 + 3 - 1, or 12 switches. The OS X method is 60% faster!
Is that a bit clearer?
IPF ships with Solaris 10 and Solaris 8/9 had the Sun Firewall on one of the CDs.
:-)
:-)
My machine still runs Solaris 8, and I haven't played enough with 10 to know. Thanks for the update.
I don't remember a CD for Sun Firewall, but it's possible they added it later. Sun tends to update their software distributions once per quarter, so it can be difficult at times to say for certain what a given OS version has bundled.
IPtables is very convoluted, powerfull but screwy. PF (OpenBSD/FreeBSD) is the best IMNSHO.
Amen to that. My Sun Box at home is sitting behind a FreeBSD firewall/NAT router. The thing is amazingly simple to keep running.
Alt-Tab (its actually command tab) didn't come into its own on OSX until 10.3 (IIRC, could be 10.2)
;-)
10.2. I'm terribly lazy about upgrading.
when they introduced the bezel style application switcher.
You mean a popup window of sorts? In 10.2, the dock would highlight the app. Works fine. Including the Shift for backwards. What tends to frustrate users is the misunderstanding of how Cmd+Tabbing works. They think that it's supposed to work like Windows with every window selected. Instead, it switches between applications, then you hit CMD+` to switch between the windows of the application. Microsoft Windows has nothing quite like this little feature. Oh, and if it has been removed from 10.4, I'm gonna be pretty pissed.
but there are some inconsistencies that drive users crazy (like dialogue boxes that don't have full keyboard access).
Indeed. One would expect the arrows/tab and enter keys to work, but often they don't. That is one of my pet peeves.
I think you need a better example, windows has since 3.0 (maybe longer) the standard where alt-command works on the global application space while ctrl-command works on the application space.
:-)
As I just told another user, it doesn't work. There may be a "standard" at the application level, but there's nothing at the OS level that makes Ctrl-Tab work. In addition, Alt-Tab tabs through EVERY window, which can be quite tedious when you have a lot of them up.
And where exactly do you *buy* a supported Fedora release from Redhat?
I think he was referring to the RedHat 9 debacle where RedHat officially discontinued their Desktop line, and sent them all to the Fedora project. This forced anyone who wanted a supported Linux desktop to purchase RedHat Enterprise or find another vendor.
I don't think the situations are comparable, but it's at least fair to understand the reference properly.
Ctl-Tab - cycles through currrent applications windows/documents
I think you may be confused. I just tried that on my Windows XP machine. It worked in Mozilla (switching each tab), but it failed miserably for every other program I tried.
Alt-Tab - cycles through running applications
Alt-Tab cycles through every open Window, not every open Application. Again, I think you're confused.
Depends on the architecture - Solaris on Sparc makes sense, why anyone would choose Solaris over Linux on an x86/amd64 platform is beyond me. (I support both at work, and aim to replace as many Solaris boxes as possible with Linux.)
Depends. If it's an actual Sun-built box, then it's just as good as a Sparc box. If it's Solaris/x86 on a random Dell or something, then I agree completely.
- iptables
;-)
:-)
:-P
Fair enough. You do have to go with commercial solutions on Solaris.
- proper package management (not that red hat provides that, but compared pkgadd anything seems good)
This has got to be a personal preference. While Sun may send out patches in huge (and annoying) bundles, at least you can always get the right package without any troubles. I can't count the number of times I've been stuck in RedHat RPM HELL. No, no. I need to bury the memory. It's too painful. Too... THE HORROR!
- real command line utils (GNU), ie:
ACK! You're not one of those guys who installs GNU utilities on the Solaris server, thus breaking everyone's scripts are you? Gah! I hate it when people do that!
Seriously, many people use GNU utilities on Solaris without complaint. I personally prefer the old-style Unix utilities, but maybe that's just because I'm used to them.
- color ls
I absolutely HATE color ls. It looks so... toyish and loud. I much prefer the Solaris standard of using underlines, brightness, bolding, and italics to represent the different attributes of the file. It works so well on Solaris because even the console is Black text on Pure White graphics mode.
- tar with gzip, bzip support (no need for piping)
Oh, like piping is going to kill you. Why, in my day...
What groupthink colored world are you living in?
:-)
;-)
Unix, not Linux. GroupThink(TM) says that Unix is inferior to the great and wise Linux. (Excuse me while I hurl here.)
Most well trained admins prefer a Unix server box over a Windows server box. Unfortunately, Windows servers were dictated by managers and MSCEs (i.e. People who thought Microsoft was "cool", but really didn't know what the hell they were doing.) I'm willing to bet that part of the reason that Linux has caught on in the server arena is because it's close enough to a true Unix, yet "cool" enough at the moment to get accepted.
My point still stands, though. Mac OS X has a lot of little "nice" things that add up to a completely "nice" Desktop. That's why people like them. It's not just the Jobs Reality Distortion Field. Or if it is, he's managed to make it permanent this time.
Answer: Yes, actually. You can't view the thread in its entirity if you don't dive below.
:-)
Explanation accepted, though. Sorry I got on your case.
Heh. Good point. I've never quite been able to break the habbit of calling it "Alt-Tab" even though Alt has nothing to do with it. It works out, though, because Windows users seem to understand the terminology better. :-)
Sun developing for linux is ultimately counterproductive to its own long term future.
That's not entirely true. Sun's strategy has always been to sell hardware and complete solutions. It really doesn't matter to them if they're selling Linux or Solaris. In fact, long before JDS they provided an option to preload RedHat on many of their systems. Why anyone would chose RedHat over Solaris for a server system is beyond me, but a lot of customers were demanding it.
All the JDS is lacking is a true follow through. This half-assed release-it-and-then-drop-it strategy is guaranteed failure.
The big question is what happened to those half-million to million-plus units that were supposed to ship in China in 2004?
What about those of us here in the US who *paid* for JDS and were promised major upgrades every quarter? We saw the JDS 1.0 -> 2.0 upgrade, then it stopped while Sun worked on JDS/Solaris.
Sun needs to learn that the only way they're going to make inroads into the desktop market is if they follow through. Rome wasn't built in a day, and neither was the popularity of Java or Solaris. If Sun would take the time to listen to their customers and implement the features they are demanding, then they'd have a very good chance at success. *sigh*
Why are so many Slashdot users in love with Apple?
:-)
Why are so many users in love with Google? Why does everyone prefer Unix servers over Windows servers?
The answer is that it Just Works(TM). Apple has made Mac OS X an extremely pleasent environment to use with little things like Alt-TAB through applications, then Alt-` through an application's windows. These little things add up into a much nicer user environment.
Same thing with the Desktop search. Google, Jeeves, MS Indexed Find, and other search technologies just don't compare to Spotlight. Spotlight digs into the file and generates all the meta-data automatically. From that meta-data, it then generates indexes that make the search lightening fast. And it does it all without compromising system integrity, system security, or finding "ghost" files.
Honestly, if you haven't used an Apple, consider getting one or borrowing one. After just a little bit of everyday use, I think you'll find what everyone is raving about.
Why don't you use the .responseXML attribute, which gives you a DOM object,
:-)
Primarily because of browser differences. The IFrame solution seems to be more portable than responseXML. The other half of it is that I'm still doing development and have temporarily eschewed XMLHttpRequest because I'm not running a server. (It's all internal to itself.) Using an IFrame allows me to perform architecture tests on new applications straight from the files on disk.
instead of "shunting" it into an IFRAME, thus losing all hope of having more than one request at a time?
That is one annoying point to the IFrame solution. However, Multithreading in a browser can be a difficult thing to deal with, so it's something I'm trying to avoid for now. Currently I use a linked list of items to load so that they all get loaded in sequence. It works quite well for the moment, and can easily be switched over to XMLHttpRequest when I'm ready.
Desktop search is the voice recognition of the new century. It will sort of work, but never well enough to make it worth relying upon.
A nice prediction, except for one problem: Apple users are already using Desktop Search. It's here now, it works, and it's much loved by users. Same thing with Google Web Apps. GMail, GMaps, and Google Search are all here today, all much loved by users, and all wiping the deck with competitors.
Voice Rec was one of those things that we always saw coming, but never saw the reality of. (Although it has gotten into niche applications like voice dialing.) The threats to Microsoft, OTOH, are already banging at the gates (ha ha) and are threatening Microsoft's bottom line. Unless Apple's and Google's growth were to abruptly stop tomorrow, even conservative projections don't look good for Microsoft.
Saying things like that are a calculated gamble, words like that can send stock prices down, so there has to be a reason for it. "Honesty" aside, it is business.
;-)
If that's true, then the gamble requires that Microsoft have something up their sleeve to help them have a "fighting comeback" in the marketplace. The problem is that Microsoft has never been very good about keeping their mouth shut about future developments. Which means that the only thing in their pipeline right now is Longhorn. Now just about every feature that could actually let Microsoft compete is getting stripped out of Longhorn, thus leaving them with nothing more than a few whiz-bang features.
Ballmer may really believe that Longhorn is going to take the world by storm, but my gut feeling is that Microsoft is doomed to irrelevency. Longhorn will be more of the same, with no acknowlegement of the paradigm shifts Apple is pushing onto the desktop and Google is pushing into Internet apps. The result will be that Microsoft will begin losing their desktop dominance to Apple and their Internet dominance to Google/FireFox, which will leave Microsoft in the position of having to become a cross-platform application provider, again.
Personally, I think that's a good thing.
"Take for instance the Siebel database. Now I've never used that interface. But I'd love to go to it and say 'who is the account manager for the Commonwealth Bank of Australia?'," Ballmer told the partners.
;-)
.NET is .NOT, Microsoft can't even search its own desktop (Quote: "It's important for people who search a corporate network,"), and that SQL Server development has ground to a halt (ceding victory to Oracle). He then goes on to make a set of pathetic promises ("In the next six months, we'll catch Google in terms of relevancy," and, 'This may be addressed in the next release [of SQL Server] in 18 months, Ballmer said, but conceded he "really didn't know",' and, "Government has really been pushing for stronger interoperability. We can't support open source, but we can support interoperability,") and say that Microsoft will never give up the fight.
:-)
I can say one thing for sure. He's DEFINITELY never used the Siebel interface!
This article honestly sounds like Ballmer was getting a bit beat up by Microsoft's partners and shareholders. They've basically gotten him to admit that
I'm sorry, but Ballmer has effectively admitted that Microsoft is now irrelevent. He's trying to grip at pavement by muttering about interop and standards compliance. This is an amazingly similar situation to the introduction of Netscape Navigator. Microsoft almost missed the boat then, but managed to throw enough resources, money, and outright theft behind capturing the browser market. Microsoft's best attempts today only come out as a pathetic whimper. No super-search engine, no desktop search, nothing. If Ballmer was smart, he'd get his boys to activate the existing Databasse File System in NTFS, then use it to push Google and Apple away from the Desktop. Once solid in that area, they should tie it into their online search engine, thus using their desktop monopoly against their competitors.
On the bright side, I am quite glad that Microsoft isn't that good anymore. At the very least, they have to watch where they step with the justice department looking over their shoulders.
Either you didn't try hard enough, or you didn't know enough
Geez. Who stuck a bee in your bonnet?
It's quite a hack to have an IFrame for this and all that. XmlHttpRequest is cleaner because you can read headers , set headers and even check on status code on return.
Do you have any idea what the subject is even about, or do you just like to rant? The parent poster was talking about parsing XML directly and how slow it was. (You can get an XML string via XMLHttpRequest, BTW.) I responded that I don't worry about the speed because I don't do it in JavaScript. I shunt the data to a hidden IFrame (which can still be done with XMLHttpRequest if you understand document.open(), which I'm guessing you don't from your senseless rant), let the browser parse it for me, then I walk the W3C compliant DOM tree. There's practically nothing faster or easier to do.
I've been using XmlHttpRequest extensively recently - it's a surprise nobody noticed it until google brought out GMAIL. Now it's the latest Buzzword that you NEED on your resume to get a job as a web-dev (which I'm not).
I was doing it before it was cool. What's your point? Just because you just found XMLHttpRequest and think it's God's gift to web developers, doesn't mean that it's the ONE TOOL for everything. Good Web apps are a lot more complex and difficult to get going. For example, have you ever shunted a JavaScript program into an existing program in a pluggable and user runnable fashion? I have. In fact, I wrapped the entire program in XML Meta-Data so that I could provide Icons, Associations, and Security Features before the program was ever loaded into memory. My program then sits as a text string in a psuedo-filesystem until it's actually ready to execute. Then I pull some wizardry to run the program without touching the HTML DOM. Would you like to hazard I guess as to how it's done, oh great and powerful jackass^W web wizard?
Has anyone actually tried parsing XML using javascript??
Yes and no. I usually let the browser do the parsing in an IFrame, then I just walk the DOM. It's much easier than taking a string of XML and trying to break it down like a good parser should. It's far too tempting to cheat like hell and do stuff like recursively find the outer tags.
Now you tell me. ;-)
:-(
Surprisingly, there's only one flip phone I've ever wanted to get. It was this Nokia model that was like a "Candy Bar" phone on the outside, but would flip open to reveal a full keyboard and joystick. Plus it had a built in FM radio (although it was really the keyboard that had me interested). Unfortunately, I need a tri-mode phone because of the areas I travel through, so the phone I *really* wanted wasn't an option.
Ah, here it is.
In the middle of the presentation there's a slide titled "Changing the Game in Candy Bars" with some cool phones in the background. [...]
;-)) A cell phone is a device with an expected lifespan, value, and deprecation schedule. There's no way you can compare the two.
;-)
Candy bars, I would guess, are a fairly stable commodity. A Mars bar last year is going to be the same as this year. Eye candy, sure, but not candy bars.
That has got to be one of the dumbest attempts at setting marketting goals that I have ever heard. Candy Bars are consumables. You buy it, eat it, then you don't have it any more. You don't eat your cellphone. (Unless that's what they're trying to fix.
BTW, I love all the "Wow!"s and "Whoa!"s in the background. It just further emphasizes how little content is actually in this presentation.
Remember the original Star Trek series (or were you even born then)? There was a 2-part special with Captain Christopher Pike (a basket case with only his head functional) and the crew of the Enterprise being shanghaid to a "forbidden planet". Apparently, *THAT* used significant chunks of the Star Trek pilot.
I wasn't there for it, but I know the episodes you're referring to. I'm actually surprised you haven't seen the unaltered pilot. CBS airred it back in the early 90's (or was it late 80's?) as a "Lost Episode". Of note is that the ship made no "whoosh"ing sound, parts where in color vs. b&w, Spock was not the first officer, and Majel Roddenberry (Nurse Chapel, Luxawanna Troi, Computer Voice) played Number One.
The most important thing to understand about the pilot episode was that it was just too forward thinking for the networks. A thoughtful captain? A female first officer? A hardcore Sci-Fi plot? The network just wasn't ready for such depth and demanded that it be revamped into more of an action/adventure show. Roddenberry complied, but he still kept many of the background commentary elements that made Star Trek so unique.
You'll note that Captain Picard turned out to be much more of a Captain Pike type of character. By the time TNG aired, the world had changed enough to allow for a more thoughtful show. It took TNG a little while to grip pavement (the TNG characters started very raw) but once it got going, we got such amazing episodes as Best of Both Worlds, Darmok, Inner Light, Yesterday's Enterprise, etc.
After a while it got old, and I stopped watching, as did many others.
Spock's Brain had nothing to do with it? No, the real reason you stopped watching was the same reason why the quality of Star Trek was dropping. The network was ticked that Gene was getting away with such controversial content, and replaced him with a different producer. That producer was responsable for making many of the more "empty" episodes at the series' end. A few months later, NBC looked at the new targetted audience numbers and started kicking themselves for removing Roddenberry and killing Star Trek.
To your original point, Pilots are often worse than the full show because they are very raw and without development. Star Trek was the exception because the changes were actually away from what everyone really wanted.
FYI, there's more info here. Wikipedia claims that part of the cost is caused by making a device that has to stand up to constant, everyday use.
If only /. editors could develop a method for fusing dups.
;-)
I'll have you know that this is NOT a dupe! The first article was in the Science section. This article is in the Hardware section! It must mean that Nuclear Fusion will be coming to power our computer hardware soon! Yippee! </sarcasm>