My apologies, my point was that they didn't end up as computer scientist. I mistakenly dumped all non-"pure CS" jobs in the programming category. Silly me. I had in mind, when I posted, my fiancee's school (she went to CMU), where in the last batch of data they have, something like 80% of the students got hired for software engineer position. Half of the rest was like (going from memory) related jobs, a few like you in admin and whatsnot, then you have 1-2 that do actual CS (My numbers are wrong, because I didnt see it in a while, but you get the idea).
Indeed. The problem stem from the fact that people think (rightly so) that one should be taught generic concepts as to not be died to a certain environment. Which is great, but then they pushed it too far. Its possible to come out of some of the top CS schools (from which Microsoft, Google, IBM, you name it, hire a TON of people for programming jobs) without even knowing what a design pattern is.
Because of that, currently at my job Im quickly becoming the guru of software developement, even though I just have an associate degree and a few years experience, because I specialized myself in it. I work in a very large IT department for a fortune 500. So the concept is totally insane: I'm self taught! But its not that Im good: its everyone thats worse, since they all did pure CS degrees, and the seniors all got promoted to various analyst/project manager positions (but I was semi-recently hired, so that leaves...me).
Well, the awkward thing is, while for every definition of it, Computer Science is indeed != Programming, in the common vocabularies lately (including in schools!), it is. Its sad, and it annoyes me, but we can't do much about it. A lot of (even prestigious-ish) universities call their Software engineering courses "Computer Science", or have software engineering classes as part of the CS department. (Almost) all companies that are looking to hire programmers fetch computer science graduates, etc. There is a very very clear cut line between computer science and programming/software engineering (and even the later two have quite the difference, though are more related), when it comes to theory and all. In the practical world, aside for the minority who come out of high end schools and are lucky to find a job as a computer scientist (where they most likely wont do -any- programming), virtually all computer science graduates end up programmers.
And its easy to see why: very few schools offers undergraduate software engineering degrees. So if you want to be a programmer with a diploma, computer science is usualy the way to go. Awkward too, since that makes for a lot of CS majors who don't really use what they learned, and a lot of underqualified software engineers.
Well, no, I wasnt thinking about the current wiimote, but about the 5 pound one. Anything 5 pound that you swing around at high velocity would be dangerous (as opposed to the normal wiimote, where its only dangerous if you're retarded). And the TV was the least of my worries. More like someone's eye or head =P
Assuming the strap is way tougher than the current one (even the enhanced one, cuz you dont want anyone to hurt themselves...probably need a totally redesigned wiimote with rubber handles, etc), this would actualy be an amazing idea. I'd buy one:) Only thing that would worrie me is the risk of injuries and such, but I'm sure someone can design something safe. I wouldnt give it to a little kid though:)
Indeed. While making complex AIs and mechanics can be difficult and require brain power, graphics need pure -time-. Making a higher definition model isn't systematically harder (it is, but people have had the talents for a while), its just...long, and time consuming. You need the same level of skill for making the textures, the meshes, etc. You just have to make more of -everything-.
Thus, it becomes incredibly long and expensive... I remember when Rogue Squadron for the Gamecube came out back then... they had a small team mind you, but making the model for the larger star destroyers took 1/6th of the time allocated to make the entire game (of course, it was in paralelle so its not like it was slowing down the other parts of the game, but still).
I just can't begin to think how long a game like a FFXIII will take in raw man hours (everybody added together). It must totally insane.
No, the pages are fairly similar in themes (colors, backgrounds, etc). They different vastly in -layouts-. I'm talking about a typical business management transactional web site. You can get all the field widths, font formattings, everything relating to navigation, titles, headers, borders, what have you in 1-2 CSS files and thats fine. The issue comes with the forms for example.
One page you'll have a menu of elements ala Windows Control Panel, bunch of floated stuff. The next you'll have a master detail order entry. The next will be a bunch of radio buttons. The other a ton of cascading dropdowns. The next will have drag and dropped divs. The next...you get the idea? You have hundreds of screens. A lot can be put in the same package (a master detail is a master detail is a master detail...usualy), so you can share a lot... But (almost) each screen will have one or two screen specific functionality, as simple as they are, and it adds up. You don't want your users to load up the entire app's styles if they're only going to ever be using 1-2 screens. So the only solution so far that I've seen, is to either: style inline (rofl, better off using font tags and table layouts than using CSS inline...), or have a sheet per page... Awkward, but not much one can do beyond relaying the extra to the preprocessor/backend language, and then you might as well not be using CSS at all... So I've been doing it that way, 1 CSS per page, plus a main set shared by everything... It works, but it feels stupid.
Hahaha, sorry, I was trying to find a simple example. But while the difference between the two is simple, its the little "CSS quirks" as I like to call them, that gets the blank stares, not the concept itself. The way, for example, that the position: whatever attribute's behavior varies -wildly- depending on the parent container... or things like, according to the specs, a floated element needs a width attribute, but isn't all that precise as to what happens if you don't specify one, etc. Ever (at least when you were new to CSS), looked for a few hours for a weird behavior (that was not one of IE's bugs!) that seemed to defy logic, until you looked the specs up, just to find a paragraph of "If the parent is this way, the behavior is that way, otherwise its this wildly different result, and if you do XYZ, then its up to the browser's implementation, so don't do it"). That alienates rich client developers moving to the web realms, something fast...
Re:Ohhhh JOY Yet Another Mystery Novel (YAMN)
on
CSS: The Definitive Guide
·
· Score: 2, Interesting
Indeed. Honestly, CSS started from a good idea. CSS 1 handled, well, styles, like fonts, colors, etc. Thats good, and CSS is good for that. It actualy makes sense for it, the syntax is peachy good.
Its when someone decided "Woooo, lets divide the page in 2, the logical structure and the layout!" that it got to become hell (aka: CSS2. For this post, I'll talk as if CSS2 was implemented at 100% in all browsers, to avoid some bozo saying "its microsoft's fault if you hate CSS blah blah blah).
Since CSS came about several years ago, back in the days, it probably made sense. Today, it doesn't.
A fairly typical enterprise web application has (for example, pulling stuff out of my ass):
an RDBMS. a layer of stored procedures a data access layer a data transformation layer business objects a business logic encapsulation layer a business service layer a web service or remoting layer to distribute the app a controler layer for the UI the actual UI.
A lot of these gets split in more categories (validation, security, etc). Now, some bozos want to start splitting it up even more. Semantic markup, styling, and if you push it, transformation (XSLT), and more. It makes sense in a web site, or a simple business front ends, like mmost transactional ecommerce sites, that have like, 5 different pages that need to be maintained a lot. But when you have several hundreds pages with vastly different purposes...it just starts being pointless. The CSS1 part still has its place, to make a "theme" for the entire site. But positioning? You need 1 stylesheet -per- page, since each page is 90% different aside from the navigation. And most of the semantic part is generated from the layer right under the UI layer, so the structure vs display is -already- split. Its just redundant.
But since all the wackos that don't get that are extremely vocal, and have been pushing it like crazy, the browsers and the web in general are going that way. So for the being, we're stuck with it. Even worse, I recently got hired by a top fortune 500 company, and amazingly enough, I'm the only one who semi-advanced (relatively speaking) knowledge of web interface technologies, so I'm stuck dealing with all of it, because when trying to explain to the other programmers whats the difference between position: relative and position: absolute, I get blank stares. Oh well. At least thats job security...
Bullseye. And thats one of Microsoft's biggest problem. Pseudo-piracy, from people who don't even know they're pseudo-pirating. (I don't know a better term for "using software that requires per-user licensing, with only 5 calls in a department of 100 people").
An extremely common scenario, too, is small businesses buying a single MSDN subscription, and installing all the software on everyone's computers (including non-developers). So like, a few grands, for douzans upon douzans of installs of windows, office, sql server, windows server, visual studio, etc etc etc.
Wouldn't even be surprised. Some of the softwares that make you qualify for a Visual Studio upgrade (competing upgrades) are incredibly cheap, and are valid for upgrade versions of Visual Studio Pro and I think even Team System. So you're literally better off (money wise) buying one of said software, and then buying a visual studio upgrade, and you're legit and saved like 100$.
Hahaha, so true! I'm canadian, but my fiancee is american, and a lot of my friends are too. I tease them ALLLL the time when they come over and complain about the cold =P
Well, I mentionned the BG thing, because according to the trailers of Heros of Mana for the DS, they're doing something like that (its more tactic than RPG, but still). And a lot of things that work on the DS, works on the Wii, so if that game works when its released, well...
3 types of "epic" games I could see work on the Wii:
#1 Elder Scroll style (minus the uber graphics, obviously). That one goes without saying.
#2 Tales style. Think something like Tales of Symphonia or Tales of the Abyss, maybe with a view from behind the character you're controling. Could work amazingly well (maybe thats what the rumored "Sword of Legendia" will be, or close?).
#3 Baldurs Gate style. Since the Wiimote can mimic a mouse, that goes without saying.
If developers play their card rights, we could see something quite interesting in that department, a bit like what we see on the DS (as you mentionned).
I agree. While I am incredibly against piracy of all kinds, there's just a point where it even I, in a moment of hypocrisy, will be typing to someone just this (That I'm against Piracy), while watching some show I can't find here. I won't try to justify it. While I think the system is flawed, I beleive in the value of copyright, so I don't think its right to do so: but being the flawed human that I am, there is a point where I just -want- to see certain shows that haven't made it here (I'm canadian, so that actualy includes a few american shows, not just stuff from abroad).
If I could pay for it, I would, but I can't. If it can be obtained through import, I will do so, and it usualy can be. But there are a couple of exceptions, where my only bet is to take a plane and record the show myself in the foreign country, because its either not available to purchase even in the country of origin, or its all region locked and such. And thats just not happening, sorry.
I'm not european, but I can picture such things happening around there with things such as video games too (I know the topic is about movie, but its a good example anyway). When all your online friends are raving about some game or another, and you have to wait 6 months to buy it, for twice the price (and not even localized, so it can't be used as an excuse), there's just so much resisting one can do...
I understand your point, however there's always the other side of the coin. Ajax used well enough, or in the case of "Atlas", client side, reusable components (as ASP.NET Ajax is, ironically enough, only 1/10th ajax in its functionality, its much more than that), allow you to scale better: more stuff done on the client = less on the server.
Bandwidth, I don't know: the core libs of ASP.NET Ajax are quite small. If you can really do all the -useful- stuff in 1/10th of the code without sacreficing -anything-, including security, functionalities, and::gasps:: development time (time = money), I'm happy for you, and I wish everyone I got to work with were that good. Unfortunately, for many companies, its not realistic.
Its a....framework. As in a toolbox to make development easier. Extenders to the web service framework to support JSON (is that how its spelled?), a bunch of ASP.NET controls for callbacks instead of postbacks, a javascript librairy. Its a framework to ease the use of ajax, thus "an anticipated ajax framework". Do we have a different definition of what a framework is? Its not an Ajax Platform, its an Ajax Framework.
Yeah, Opera is a newcomer in the list of supported browsers. The developers of the various.NET technologies seem to have some autonomy (more or less) from Microsoft's monopoly machine, and they listen a fair bit to the community. Half the reason its not "IE-only".
I agree, but then again, other browsers got all these plugins without the activeX issues. ActiveX isn't bad, its just old. Back when it was made, the world was different. Thats why Microsoft is phasing it out in favor of other, better alternatives, and instead of a one size fit all, it has different tools for different jobs: for example, XAML/WPF,.NET assemblies, etc.
Its almost true, but not quite. Once you have an active x installed in a way thats usuable by a browser, then it can be used "at will". So let say I make a web site, with my own active x, and I'm the only user. I don't risk anything from visiting that web site, obviously. But let say someone else learns about it, and stick the tag to use the same control on their web site, but this time abuse a security hole in it... When I go on that web site, I'm going to get owned by my own Active X. There was recently an issue like that, where an Active X used in Visual Studio (not even a web site) had a security hole, so someone could hook it up through a web site, and BANG! (it didn't affect IE7 though, as it handles these things a little better).
If ActiveXs could only be used by the web site that installed them, what you say would be more true (not totally, since certain phishing methods can catch off guard even a tech savy person, if its well done enough...social engineering is a powerful thing), but unfortunately, its not completly the case, especialy in older IEs.
Yeah seriously. I used to work for a consulting firm, and, well, UI sells. As much as the google "keep in simple" approaches work like a charm... making fancy interactive GUIs make customers throw money at you. When we started using Telerik's controls is when we started having to refuse contracts for lack of time to do them all! It was sweeet.
Unfortunately, now I work for in house enterprise apps for a fortune 500 company. And they don't care about that. Can't even use ASP.NET Ajax because of architecture decisions, so I have to do it the hard way. Had to spend all day on some javascript to make a user friendly date picker without messing up the architecture. That was hell >.
Because of the way ASP.NET works, its a bit dumb to make an Ajax toolkit. You're better off making a control toolkit, as it can be better targeted for your customers. And its what people have been doing. Look at ComponentArt, Telerik, and others. Their offerings, while sharing some things with ASP.NET Ajax, actualy embrace and extend it, so their offerings are still worth the money...
yeah, the source for a lot of the community controls can be viewed and all. The Extenders are incredibly easy to make, because its most of the point (ASP.NET Ajax's name is misleading, as its main appeal is to be able to make reusable client-side code blocks, ajax is second in line, so I prefered when it was called Atlas...oh well!)
Slashdot Mirror
My apologies, my point was that they didn't end up as computer scientist. I mistakenly dumped all non-"pure CS" jobs in the programming category. Silly me. I had in mind, when I posted, my fiancee's school (she went to CMU), where in the last batch of data they have, something like 80% of the students got hired for software engineer position. Half of the rest was like (going from memory) related jobs, a few like you in admin and whatsnot, then you have 1-2 that do actual CS (My numbers are wrong, because I didnt see it in a while, but you get the idea).
Indeed. The problem stem from the fact that people think (rightly so) that one should be taught generic concepts as to not be died to a certain environment. Which is great, but then they pushed it too far. Its possible to come out of some of the top CS schools (from which Microsoft, Google, IBM, you name it, hire a TON of people for programming jobs) without even knowing what a design pattern is.
Because of that, currently at my job Im quickly becoming the guru of software developement, even though I just have an associate degree and a few years experience, because I specialized myself in it. I work in a very large IT department for a fortune 500. So the concept is totally insane: I'm self taught! But its not that Im good: its everyone thats worse, since they all did pure CS degrees, and the seniors all got promoted to various analyst/project manager positions (but I was semi-recently hired, so that leaves...me).
Well, the awkward thing is, while for every definition of it, Computer Science is indeed != Programming, in the common vocabularies lately (including in schools!), it is. Its sad, and it annoyes me, but we can't do much about it. A lot of (even prestigious-ish) universities call their Software engineering courses "Computer Science", or have software engineering classes as part of the CS department. (Almost) all companies that are looking to hire programmers fetch computer science graduates, etc. There is a very very clear cut line between computer science and programming/software engineering (and even the later two have quite the difference, though are more related), when it comes to theory and all. In the practical world, aside for the minority who come out of high end schools and are lucky to find a job as a computer scientist (where they most likely wont do -any- programming), virtually all computer science graduates end up programmers.
And its easy to see why: very few schools offers undergraduate software engineering degrees. So if you want to be a programmer with a diploma, computer science is usualy the way to go. Awkward too, since that makes for a lot of CS majors who don't really use what they learned, and a lot of underqualified software engineers.
Well, no, I wasnt thinking about the current wiimote, but about the 5 pound one. Anything 5 pound that you swing around at high velocity would be dangerous (as opposed to the normal wiimote, where its only dangerous if you're retarded). And the TV was the least of my worries. More like someone's eye or head =P
Assuming the strap is way tougher than the current one (even the enhanced one, cuz you dont want anyone to hurt themselves...probably need a totally redesigned wiimote with rubber handles, etc), this would actualy be an amazing idea. I'd buy one :) Only thing that would worrie me is the risk of injuries and such, but I'm sure someone can design something safe. I wouldnt give it to a little kid though :)
Indeed. While making complex AIs and mechanics can be difficult and require brain power, graphics need pure -time-. Making a higher definition model isn't systematically harder (it is, but people have had the talents for a while), its just...long, and time consuming. You need the same level of skill for making the textures, the meshes, etc. You just have to make more of -everything-.
Thus, it becomes incredibly long and expensive... I remember when Rogue Squadron for the Gamecube came out back then... they had a small team mind you, but making the model for the larger star destroyers took 1/6th of the time allocated to make the entire game (of course, it was in paralelle so its not like it was slowing down the other parts of the game, but still).
I just can't begin to think how long a game like a FFXIII will take in raw man hours (everybody added together). It must totally insane.
No, the pages are fairly similar in themes (colors, backgrounds, etc). They different vastly in -layouts-. I'm talking about a typical business management transactional web site. You can get all the field widths, font formattings, everything relating to navigation, titles, headers, borders, what have you in 1-2 CSS files and thats fine. The issue comes with the forms for example.
One page you'll have a menu of elements ala Windows Control Panel, bunch of floated stuff. The next you'll have a master detail order entry. The next will be a bunch of radio buttons. The other a ton of cascading dropdowns. The next will have drag and dropped divs. The next...you get the idea? You have hundreds of screens. A lot can be put in the same package (a master detail is a master detail is a master detail...usualy), so you can share a lot... But (almost) each screen will have one or two screen specific functionality, as simple as they are, and it adds up. You don't want your users to load up the entire app's styles if they're only going to ever be using 1-2 screens. So the only solution so far that I've seen, is to either: style inline (rofl, better off using font tags and table layouts than using CSS inline...), or have a sheet per page... Awkward, but not much one can do beyond relaying the extra to the preprocessor/backend language, and then you might as well not be using CSS at all... So I've been doing it that way, 1 CSS per page, plus a main set shared by everything... It works, but it feels stupid.
Hahaha, sorry, I was trying to find a simple example. But while the difference between the two is simple, its the little "CSS quirks" as I like to call them, that gets the blank stares, not the concept itself. The way, for example, that the position: whatever attribute's behavior varies -wildly- depending on the parent container... or things like, according to the specs, a floated element needs a width attribute, but isn't all that precise as to what happens if you don't specify one, etc. Ever (at least when you were new to CSS), looked for a few hours for a weird behavior (that was not one of IE's bugs!) that seemed to defy logic, until you looked the specs up, just to find a paragraph of "If the parent is this way, the behavior is that way, otherwise its this wildly different result, and if you do XYZ, then its up to the browser's implementation, so don't do it"). That alienates rich client developers moving to the web realms, something fast...
Indeed. Honestly, CSS started from a good idea. CSS 1 handled, well, styles, like fonts, colors, etc. Thats good, and CSS is good for that. It actualy makes sense for it, the syntax is peachy good.
Its when someone decided "Woooo, lets divide the page in 2, the logical structure and the layout!" that it got to become hell (aka: CSS2. For this post, I'll talk as if CSS2 was implemented at 100% in all browsers, to avoid some bozo saying "its microsoft's fault if you hate CSS blah blah blah).
Since CSS came about several years ago, back in the days, it probably made sense. Today, it doesn't.
A fairly typical enterprise web application has (for example, pulling stuff out of my ass):
an RDBMS.
a layer of stored procedures
a data access layer
a data transformation layer
business objects
a business logic encapsulation layer
a business service layer
a web service or remoting layer to distribute the app
a controler layer for the UI
the actual UI.
A lot of these gets split in more categories (validation, security, etc). Now, some bozos want to start splitting it up even more. Semantic markup, styling, and if you push it, transformation (XSLT), and more. It makes sense in a web site, or a simple business front ends, like mmost transactional ecommerce sites, that have like, 5 different pages that need to be maintained a lot. But when you have several hundreds pages with vastly different purposes...it just starts being pointless. The CSS1 part still has its place, to make a "theme" for the entire site. But positioning? You need 1 stylesheet -per- page, since each page is 90% different aside from the navigation. And most of the semantic part is generated from the layer right under the UI layer, so the structure vs display is -already- split. Its just redundant.
But since all the wackos that don't get that are extremely vocal, and have been pushing it like crazy, the browsers and the web in general are going that way. So for the being, we're stuck with it. Even worse, I recently got hired by a top fortune 500 company, and amazingly enough, I'm the only one who semi-advanced (relatively speaking) knowledge of web interface technologies, so I'm stuck dealing with all of it, because when trying to explain to the other programmers whats the difference between position: relative and position: absolute, I get blank stares. Oh well. At least thats job security...
Bullseye. And thats one of Microsoft's biggest problem. Pseudo-piracy, from people who don't even know they're pseudo-pirating. (I don't know a better term for "using software that requires per-user licensing, with only 5 calls in a department of 100 people").
An extremely common scenario, too, is small businesses buying a single MSDN subscription, and installing all the software on everyone's computers (including non-developers). So like, a few grands, for douzans upon douzans of installs of windows, office, sql server, windows server, visual studio, etc etc etc.
Wouldn't even be surprised. Some of the softwares that make you qualify for a Visual Studio upgrade (competing upgrades) are incredibly cheap, and are valid for upgrade versions of Visual Studio Pro and I think even Team System. So you're literally better off (money wise) buying one of said software, and then buying a visual studio upgrade, and you're legit and saved like 100$.
7 actualy. 12 if you count the seals and polar bears. Everyone else died from the cold. Yet somehow I still have bills to pay, grrr....
Well, I mentionned the BG thing, because according to the trailers of Heros of Mana for the DS, they're doing something like that (its more tactic than RPG, but still). And a lot of things that work on the DS, works on the Wii, so if that game works when its released, well...
3 types of "epic" games I could see work on the Wii:
#1 Elder Scroll style (minus the uber graphics, obviously). That one goes without saying.
#2 Tales style. Think something like Tales of Symphonia or Tales of the Abyss, maybe with a view from behind the character you're controling. Could work amazingly well (maybe thats what the rumored "Sword of Legendia" will be, or close?).
#3 Baldurs Gate style. Since the Wiimote can mimic a mouse, that goes without saying.
If developers play their card rights, we could see something quite interesting in that department, a bit like what we see on the DS (as you mentionned).
I agree. While I am incredibly against piracy of all kinds, there's just a point where it even I, in a moment of hypocrisy, will be typing to someone just this (That I'm against Piracy), while watching some show I can't find here. I won't try to justify it. While I think the system is flawed, I beleive in the value of copyright, so I don't think its right to do so: but being the flawed human that I am, there is a point where I just -want- to see certain shows that haven't made it here (I'm canadian, so that actualy includes a few american shows, not just stuff from abroad).
If I could pay for it, I would, but I can't. If it can be obtained through import, I will do so, and it usualy can be. But there are a couple of exceptions, where my only bet is to take a plane and record the show myself in the foreign country, because its either not available to purchase even in the country of origin, or its all region locked and such. And thats just not happening, sorry.
I'm not european, but I can picture such things happening around there with things such as video games too (I know the topic is about movie, but its a good example anyway). When all your online friends are raving about some game or another, and you have to wait 6 months to buy it, for twice the price (and not even localized, so it can't be used as an excuse), there's just so much resisting one can do...
I understand your point, however there's always the other side of the coin. Ajax used well enough, or in the case of "Atlas", client side, reusable components (as ASP.NET Ajax is, ironically enough, only 1/10th ajax in its functionality, its much more than that), allow you to scale better: more stuff done on the client = less on the server.
::gasps:: development time (time = money), I'm happy for you, and I wish everyone I got to work with were that good. Unfortunately, for many companies, its not realistic.
Bandwidth, I don't know: the core libs of ASP.NET Ajax are quite small. If you can really do all the -useful- stuff in 1/10th of the code without sacreficing -anything-, including security, functionalities, and
Its a....framework. As in a toolbox to make development easier. Extenders to the web service framework to support JSON (is that how its spelled?), a bunch of ASP.NET controls for callbacks instead of postbacks, a javascript librairy. Its a framework to ease the use of ajax, thus "an anticipated ajax framework". Do we have a different definition of what a framework is? Its not an Ajax Platform, its an Ajax Framework.
Yeah, Opera is a newcomer in the list of supported browsers. The developers of the various .NET technologies seem to have some autonomy (more or less) from Microsoft's monopoly machine, and they listen a fair bit to the community. Half the reason its not "IE-only".
I agree, but then again, other browsers got all these plugins without the activeX issues. ActiveX isn't bad, its just old. Back when it was made, the world was different. Thats why Microsoft is phasing it out in favor of other, better alternatives, and instead of a one size fit all, it has different tools for different jobs: for example, XAML/WPF, .NET assemblies, etc.
Its almost true, but not quite. Once you have an active x installed in a way thats usuable by a browser, then it can be used "at will". So let say I make a web site, with my own active x, and I'm the only user. I don't risk anything from visiting that web site, obviously. But let say someone else learns about it, and stick the tag to use the same control on their web site, but this time abuse a security hole in it... When I go on that web site, I'm going to get owned by my own Active X. There was recently an issue like that, where an Active X used in Visual Studio (not even a web site) had a security hole, so someone could hook it up through a web site, and BANG! (it didn't affect IE7 though, as it handles these things a little better).
If ActiveXs could only be used by the web site that installed them, what you say would be more true (not totally, since certain phishing methods can catch off guard even a tech savy person, if its well done enough...social engineering is a powerful thing), but unfortunately, its not completly the case, especialy in older IEs.
Yeah seriously. I used to work for a consulting firm, and, well, UI sells. As much as the google "keep in simple" approaches work like a charm... making fancy interactive GUIs make customers throw money at you. When we started using Telerik's controls is when we started having to refuse contracts for lack of time to do them all! It was sweeet.
Unfortunately, now I work for in house enterprise apps for a fortune 500 company. And they don't care about that. Can't even use ASP.NET Ajax because of architecture decisions, so I have to do it the hard way. Had to spend all day on some javascript to make a user friendly date picker without messing up the architecture. That was hell >.
Because of the way ASP.NET works, its a bit dumb to make an Ajax toolkit. You're better off making a control toolkit, as it can be better targeted for your customers. And its what people have been doing. Look at ComponentArt, Telerik, and others. Their offerings, while sharing some things with ASP.NET Ajax, actualy embrace and extend it, so their offerings are still worth the money...
IE, Firefox, Safari, Opera the supported browsers as of the release.
yeah, the source for a lot of the community controls can be viewed and all. The Extenders are incredibly easy to make, because its most of the point (ASP.NET Ajax's name is misleading, as its main appeal is to be able to make reusable client-side code blocks, ajax is second in line, so I prefered when it was called Atlas...oh well!)