If using a different Web browser to access a server causes it to crash, you have more serious things to worry about, like finding another vendor that doesn't write software that takes down your server when it's accessed in a perfectly reasonable manner.
Using 32-bit unsigned integers gives a maximum of 4GB of addressable space. It had been 2GB until their recent firmware update. Also, there are no disks larger than 1TB currently on the market, so 4TB is also a practical limit.
Re:RAID5 is stupid, RAID 10 or no RAID
on
What NAS To Buy?
·
· Score: 4, Interesting
Um, no, try again. RAID-6 is n+2 redundancy, not n+3. RAID-10 is n+2 on a good day but you are really only guaranteed n+1, since if both mirrored disks fail then you are screwed.
The issue you describe in IE has probably nothing to do with the nesting and everything to do with hasLayout. Also, if you've got more than 2 or 3 <div> inside each-other, you should re-evaluate what you're doing and probably use a more appropriate element (ul, ol, dl, p, h1-h6, etc).
I strongly disagree. Unless you want to end up having to do like this article suggests and make a bunch of changes every time a new browser comes out (and maybe you do -- I bet it's good for business), it makes much more sense to test against something that you know adheres to defined standards and then doing minor fixups for IE at the end.
If you're writing and testing against IE, and you write a line of CSS that doesn't do what you expect and change it to make it work, but the reason it wasn't working isn't because you wrote it wrong but because IE calculates some dimensions incorrectly (read: hasLayout), then when you get around to testing it in everything else (and by "everything else" I mean "Firefox", since this is what the IE-first crowd seems to think means "everything else") it's broken. Now compound this issue 20 times, because there are 20 distinct things in the CSS that cause IE to fuck up. Maybe there are also some combinations of things that trigger a bug. So now, instead of writing hacks to work around IE's brokenness, you are writing hacks and sending different code conditionally to "work around" the browsers that are rendering it properly. Suddenly, when IE8 comes around and fixes the bugs you're relying on in IE, you've got a broken Web site again. It's just a bad idea, and getting things working across all the other browsers, frankly, takes a mere fraction of the time it does to get things touched up properly for IE.
Uh. min-width is in IE7. I'm surprised you don't know that if you're testing "primary in Internet Explorer".
Not to be high and mighty, but you really really really should develop against a standards-compliant browser *first*, which means any one of Konqueror, Safari, Opera, or Firefox, and then hack IE once you're all done using conditional comments. Since all the browser vendors other than Microsoft do a good job of adhering to standards at this point, by testing against one of those browsers you can pretty well guarantee you will be functioning in the rest of them. It makes much more sense than to test on the outlier (IE) and then try to fiddle with it until it works in everything else.
I'm quite confident that none of my sites will need to be updated for IE8 as long as Microsoft are doing their jobs, because the sites are written to conform to standards and only use conditional comments with special CSS for browsers <= IE7. That means that when IE8 rolls around, it will get served the same standards-compliant code as everything else and (for once) will not break on it.
"Deal with"? I personally couldn't live without having a global hotkey to control background apps (like Amarok) and a compose key. Can you explain why having fewer keys is better in your opinion? I'm very curious to understand, since I use mine constantly, and the only thing I can figure is that people are so blinded with anti-Microsoft rage that there's a little Windows logo on their keyboard that they'd rather go without the added functionality than have to see a silly logo.
All modern disks ship with some unused spare sectors that are used to remap onto failed sectors. This occurs all inside the drive's firmware, so even though the computer thinks it's addressing the same sector, in actuality the drive is pulling data from the remapped spare. The firmware is smart enough to only remap sectors when you try to write to a bad one, though, because if it decided to remap a bad sector that had data on it that you needed, you'd not be able to get back that data even if the disk was eventually able to read the sector.
Ok, there is more to performance than the scripting, especially when you are dealing with vector and animation drawing.
You said "Performance features", and then you mentioned an example that LOADS RASTERISED IMAGES. Why would I even think of the speed of the vector and animation drawing engine when you were talking about a Socket/URLLoader and Bitmap implementation that has zero need for vectors and virtually zero need for animation? Now that we've clarified it's the animation library you're talking about, can you provide an example of an animation that works really slowly in Flash but is smooth in Silverlight?
The part you also omit for the viewers is that ActionScript 3 is 'faster' which MOST people and sites do not use yet.
As opposed to Silverlight, which is used by so many more sites right now, right? Oh, wait...
The older Flash 8 and earlier standards are significantly slower than JavaScript, and this is ALSO not a comparison of the native code speeds of Silverlight that use.NET.
Really? The way you wrote it originally certainly made it sound like a comparison. I guess I was confused.
The Flash 8 will remain the industry standard until Flash Lite is brought up to Flash 9 standards, and by that time SilverLight will have just as much of an opportunity to entice developers, especially Windows developers that can take existing code sets and existing WPF/.NET applicaitons and build for Silverlight.
What does market share have anything to do with performance features? Why are you even mentioning this when we were supposedly discussing the improved performance of Silverlight over Flash?
You are also dismissing the importance of Silverlight using 'standard' languages. ActionScript is is a hack on hack of a animation scripting language and in no way compares to even JScript let alone running native C# or even VB code.
Wait, wait. ActionScript, a language designed from the beginning around animation (since that's all that Flash did back when it was first introduced) is a "hack on hack of a animation scripting language [sic]", as opposed to JavaScript, which is designed for providing basic scripting functionality and Web page interactivity?? Excuse me for being incredulous, but.. this is a ridiculous statement.
Why yes it does, but it still DOES NOT SUPPORT multi-cast Streaming. There is more to HD support than just providing the VC1 codec and rendering it properly, there is the bandwidth and server side issues as well.
Yes, I agree that Flash doesn't support multicast. However, before you can multicast video over the Internet, you also need to get ISPs to support multicast, and... a whole heaping lot of them don't, which makes support for this particular feature rather irrelevant at this stage of the game. For what it's worth, Flash doesn't support UDP either, which is far more important for real-time video streaming.
And again, this is with Player 9 only content as this is the first version of Flash to try to offer accelerated Video playback, something Silverlight already handles easily as it was designed around HD VC1.
Only the 2007 version of Flash offers features in the... 2007 version of Silverlight? Wha?
You really need to go look up more about the Flex - Flash confusion people like to add to this argument. The flash format is not as open as you seem to think it is, nor is Flex fully interoperable as well.
Bloody hell, mate. I need to "look up more about the Flex - Flash confusion"? What? I certainly didn't say that the SWF specification was open without caveats, but you've yet to say what exactly is the "secret format" that Flash has that Silverlight does
You've written a nice list, but I just don't see anything on it that's really, well, valid.
1) As far as performance in general is concerned, ActionScript 3 is extremely fast. Though I definitely wouldn't say the same about ActionScript 2, it's not fair to compare an old version of Flash against a recent version of Silverlight.
2) Flash Player 9 Update 3, which was released in December of last year, supports H.264 and HE-AAC.
3) Flex uses a similar XML-based format called MXML for describing applications. Of course, "easier" is relative -- I'm sure if you've been working as a Windows programmer forever it's easier, but maybe not for someone that isn't used to how Microsoft does things. Also, what's a "secret format" that Flash has? The entire SWF specification is open (well, except to use to build a Flash player, which is pretty stupid), and ActionScript is based on the ECMAScript specification.
4a) Flash has a "rich vector/bitmap based environment" (whatever that means -- it can draw on bitmaps and do transformations and effects, and it can draw vector shapes), and has since forever. How is this any worse than what Silverlight has (speaking as someone that has not used Silverlight)?
4b) No, you can't use any language you want, but I don't necessarily see this as a huge advantage, since it adds an amount of additional complexity that could easily be problematic. You can't ask for "a Silverlight" programmer, now you have to ask for "a Silverlight programmer that also knows Python/C#/whatever" -- this will really narrow your potential hiring pool.
5) Flash has ExternalInterface which provides 100% seamless interaction between Flash and JavaScript, and is hardly "heavy".
6) Have you even looked at what Flash provides lately? ActionScript 3 is an extremely capable language. Without giving any specific examples of features that don't/can't exist in Flash, but that do in Silverlight, it's hard to respond to this. Provide an example and we'll talk.
7) I've not personally experienced performance issues with Flash applications on OS X, but YMMV. Since I don't use Windows, it's hard for me to say if something runs more slowly than it would on a Windows box, but I never ran anything that seemed slow or that pegged my CPU. I've heard that it's slower on PPC architectures, but Windows never ran on PPC to begin with, so who knows how Flash would run on Windows if there were a PPC version. I've never ever run a Silverlight application, so I can't confirm your allegation that it works better, either.
8) Can you provide a specific example of how the security model of Silverlight is more any more secure? Flash code runs in a sandboxed virtual machine ("managed code" for non-Microsofties out there) too, and has since the beginning of time. Saying "see recent Flash updates" just says to me that Adobe has addressed potential security issues that may have existed, and hardly damns the platform as being somehow tragically insecure. (And, in fact, the recent security updates to Flash are nothing more than hardening against some potential XSS attacks.)
9) Sounds like MXML, again. Don't repeat yourself, you already mentioned XAML once.;) Talking about "Display PDF" as if it were some markup language makes no sense, too, since Display is an application for viewing PDF files -- nothing more.
Now, I'm certainly no Flash apologist -- up until about a month ago I refused to touch it, and ActionScript 2 is unbelievably shitty -- and certainly if we were comparing against Flash 8 or earlier running ActionScript 2 you'd have some valid points, but nothing on your list actually seems to me to be a valid reason why Silverlight is better than Flash here and now. And again, despite your protests that Microsoft is developing an OS X version of Silverlight, and is working with Mono to develop a Linux version, they have not been above releasing software for platforms and then dropping it without cause in the past, and I haven't seen them changing their colours.
Safari 3/WebKit's CSS support is excellent, probably the best now, or at the very least on par with Gecko. I don't know how many of the fixes that Apple has made to their rendering engine have worked their way back into KHTML, though. Konqueror 3.5.2 still had some issues when I last tried it, but most of them seemed related to a weak JavaScript implementation and not so much CSS, though there were a couple CSS bugs I saw (relative units were sized incorrectly and text-indent did not work correctly). According to bug reports filed in KDE's bug tracker, though, it has been improved muchly since then (it's up to 3.5.7 now), so chances are any issues you had before don't exist in the current version.
You can set SQL modes, such as STRICT_ALL_TABLES, that will cause MySQL to reject invalid data instead of truncating it. There is documentation about the various SQL modes.
More inconsistent function naming (verb location):
file_get_contents() vs. get_html_translation_table()
Even within the same extension:
imagesetstyle() vs. imagecolorset()
Flipped haystack and needle:
strpos(haystack, needle) vs. in_array(needle, haystack)
Speed:
Scutigena Computer Language Performance Comparison (see graphs) There used to be another site that you could compare one language's speed relative to another that also showed PHP as one of the slowest. I can't seem to find it now, though. Also PHP5 might compare a bit more favourably, but this is all I could find after a quick Google search. Perhaps more importantly, PHP drags the speed of other things down (like Apache), since even though the core is supposedly thread-safe, nobody seems to know which extensions are and aren't, so eg. Apache needs to be run in prefork mpm instead of using a threaded mpm.
I think PHP is overall a fairly decent language; I've used it for many years with great success. But it does have major problems, and it would be nice for them to get fixed instead of pushed aside. (I read some minutes from a PHP 6 meeting a while ago where they touched on the issue of consistency, and the PHP Group decided that it wasn't important enough to fix. It's really annoying to me to need a PHP-aware IDE or a manual always handy to program in a language because the arguments and function names are so non-uniform.)
It sounds like, since you never let it go there, it keeps trying to go to the IE7 "First Start" page, which, contrary to all logic, is not a local page but is instead hosted on an MSN server. This page allows you to..er..change default IE settings. Seems rather stupid and insecure to me, but hey.
How is it elitist snobbery? You're asking the browser to handle invalid data for which the only true method for handling it is to bail out, since it's by definition impossible to parse it for proper meaning (it can only be guessed at, and just like dealing with vagueness in human languages, different software can have a different opinion about how the vagueness should be handled). XML parsers do exactly that, which is why I wish more people would start writing sites using XHTML (with the application/xhtml+xml MIME-type), so that this generation of ugly tag soup Web sites will eventually close.
If I started leaving out punctuation in my writing (missing close tags), switched what I was talking about mid-sentence without making a note of it (mismatched tags), added a bunch of obscure terms that only my close group of friends use (proprietary tags), put words where they made no sense (extra tags), and added a bunch of gibberish at random in the middle of my sentences (invalid tags), would you be elitist for asking me to clarify my writing when you were unable to read it?
Honestly, it's so easy to write "valid" markup that I'm convinced that only gross incompetence and disturbing amounts of laziness are the reasons that any invalid markup exists today. The rules for HTML and XML are very, very simple; The W3C provides an online validator that's VERY good at explaining exactly what's wrong with a document, and trying to give anyone a pass for generating invalid data while blaming Safari for interpreting differently (not incorrectly, since there is no 'correct' way to handle invalid data other than bailing out) the very same invalid data is doing the Web a disservice.
Slashdot Mirror
Maybe you should try using AdBlock to only block those advertisers that engage in such practises then? It's not an all-or-nothing affair.
If using a different Web browser to access a server causes it to crash, you have more serious things to worry about, like finding another vendor that doesn't write software that takes down your server when it's accessed in a perfectly reasonable manner.
Using 32-bit unsigned integers gives a maximum of 4GB of addressable space. It had been 2GB until their recent firmware update. Also, there are no disks larger than 1TB currently on the market, so 4TB is also a practical limit.
Um, no, try again. RAID-6 is n+2 redundancy, not n+3. RAID-10 is n+2 on a good day but you are really only guaranteed n+1, since if both mirrored disks fail then you are screwed.
What do you think a "Release Candidate" is, exactly? If there are no bugs, it becomes the final released version. Hence the name.
Actually, many of them probably would if they could, but the Americans with Disabilities Act makes it illegal to do so.
The issue you describe in IE has probably nothing to do with the nesting and everything to do with hasLayout. Also, if you've got more than 2 or 3 <div> inside each-other, you should re-evaluate what you're doing and probably use a more appropriate element (ul, ol, dl, p, h1-h6, etc).
I strongly disagree. Unless you want to end up having to do like this article suggests and make a bunch of changes every time a new browser comes out (and maybe you do -- I bet it's good for business), it makes much more sense to test against something that you know adheres to defined standards and then doing minor fixups for IE at the end.
If you're writing and testing against IE, and you write a line of CSS that doesn't do what you expect and change it to make it work, but the reason it wasn't working isn't because you wrote it wrong but because IE calculates some dimensions incorrectly (read: hasLayout), then when you get around to testing it in everything else (and by "everything else" I mean "Firefox", since this is what the IE-first crowd seems to think means "everything else") it's broken. Now compound this issue 20 times, because there are 20 distinct things in the CSS that cause IE to fuck up. Maybe there are also some combinations of things that trigger a bug. So now, instead of writing hacks to work around IE's brokenness, you are writing hacks and sending different code conditionally to "work around" the browsers that are rendering it properly. Suddenly, when IE8 comes around and fixes the bugs you're relying on in IE, you've got a broken Web site again. It's just a bad idea, and getting things working across all the other browsers, frankly, takes a mere fraction of the time it does to get things touched up properly for IE.
Uh. min-width is in IE7. I'm surprised you don't know that if you're testing "primary in Internet Explorer".
Not to be high and mighty, but you really really really should develop against a standards-compliant browser *first*, which means any one of Konqueror, Safari, Opera, or Firefox, and then hack IE once you're all done using conditional comments. Since all the browser vendors other than Microsoft do a good job of adhering to standards at this point, by testing against one of those browsers you can pretty well guarantee you will be functioning in the rest of them. It makes much more sense than to test on the outlier (IE) and then try to fiddle with it until it works in everything else.
I'm quite confident that none of my sites will need to be updated for IE8 as long as Microsoft are doing their jobs, because the sites are written to conform to standards and only use conditional comments with special CSS for browsers <= IE7. That means that when IE8 rolls around, it will get served the same standards-compliant code as everything else and (for once) will not break on it.
"Deal with"? I personally couldn't live without having a global hotkey to control background apps (like Amarok) and a compose key. Can you explain why having fewer keys is better in your opinion? I'm very curious to understand, since I use mine constantly, and the only thing I can figure is that people are so blinded with anti-Microsoft rage that there's a little Windows logo on their keyboard that they'd rather go without the added functionality than have to see a silly logo.
Did you try unchecking the option in the QuickTime control panel?
All modern disks ship with some unused spare sectors that are used to remap onto failed sectors. This occurs all inside the drive's firmware, so even though the computer thinks it's addressing the same sector, in actuality the drive is pulling data from the remapped spare. The firmware is smart enough to only remap sectors when you try to write to a bad one, though, because if it decided to remap a bad sector that had data on it that you needed, you'd not be able to get back that data even if the disk was eventually able to read the sector.
You said "Performance features", and then you mentioned an example that LOADS RASTERISED IMAGES. Why would I even think of the speed of the vector and animation drawing engine when you were talking about a Socket/URLLoader and Bitmap implementation that has zero need for vectors and virtually zero need for animation? Now that we've clarified it's the animation library you're talking about, can you provide an example of an animation that works really slowly in Flash but is smooth in Silverlight?
As opposed to Silverlight, which is used by so many more sites right now, right? Oh, wait...
Really? The way you wrote it originally certainly made it sound like a comparison. I guess I was confused.
What does market share have anything to do with performance features? Why are you even mentioning this when we were supposedly discussing the improved performance of Silverlight over Flash?
Wait, wait. ActionScript, a language designed from the beginning around animation (since that's all that Flash did back when it was first introduced) is a "hack on hack of a animation scripting language [sic]", as opposed to JavaScript, which is designed for providing basic scripting functionality and Web page interactivity?? Excuse me for being incredulous, but.. this is a ridiculous statement.
Yes, I agree that Flash doesn't support multicast. However, before you can multicast video over the Internet, you also need to get ISPs to support multicast, and... a whole heaping lot of them don't, which makes support for this particular feature rather irrelevant at this stage of the game. For what it's worth, Flash doesn't support UDP either, which is far more important for real-time video streaming.
Only the 2007 version of Flash offers features in the... 2007 version of Silverlight? Wha?
Bloody hell, mate. I need to "look up more about the Flex - Flash confusion"? What? I certainly didn't say that the SWF specification was open without caveats, but you've yet to say what exactly is the "secret format" that Flash has that Silverlight does
You've written a nice list, but I just don't see anything on it that's really, well, valid.
;) Talking about "Display PDF" as if it were some markup language makes no sense, too, since Display is an application for viewing PDF files -- nothing more.
1) As far as performance in general is concerned, ActionScript 3 is extremely fast. Though I definitely wouldn't say the same about ActionScript 2, it's not fair to compare an old version of Flash against a recent version of Silverlight.
2) Flash Player 9 Update 3, which was released in December of last year, supports H.264 and HE-AAC.
3) Flex uses a similar XML-based format called MXML for describing applications. Of course, "easier" is relative -- I'm sure if you've been working as a Windows programmer forever it's easier, but maybe not for someone that isn't used to how Microsoft does things. Also, what's a "secret format" that Flash has? The entire SWF specification is open (well, except to use to build a Flash player, which is pretty stupid), and ActionScript is based on the ECMAScript specification.
4a) Flash has a "rich vector/bitmap based environment" (whatever that means -- it can draw on bitmaps and do transformations and effects, and it can draw vector shapes), and has since forever. How is this any worse than what Silverlight has (speaking as someone that has not used Silverlight)?
4b) No, you can't use any language you want, but I don't necessarily see this as a huge advantage, since it adds an amount of additional complexity that could easily be problematic. You can't ask for "a Silverlight" programmer, now you have to ask for "a Silverlight programmer that also knows Python/C#/whatever" -- this will really narrow your potential hiring pool.
5) Flash has ExternalInterface which provides 100% seamless interaction between Flash and JavaScript, and is hardly "heavy".
6) Have you even looked at what Flash provides lately? ActionScript 3 is an extremely capable language. Without giving any specific examples of features that don't/can't exist in Flash, but that do in Silverlight, it's hard to respond to this. Provide an example and we'll talk.
7) I've not personally experienced performance issues with Flash applications on OS X, but YMMV. Since I don't use Windows, it's hard for me to say if something runs more slowly than it would on a Windows box, but I never ran anything that seemed slow or that pegged my CPU. I've heard that it's slower on PPC architectures, but Windows never ran on PPC to begin with, so who knows how Flash would run on Windows if there were a PPC version. I've never ever run a Silverlight application, so I can't confirm your allegation that it works better, either.
8) Can you provide a specific example of how the security model of Silverlight is more any more secure? Flash code runs in a sandboxed virtual machine ("managed code" for non-Microsofties out there) too, and has since the beginning of time. Saying "see recent Flash updates" just says to me that Adobe has addressed potential security issues that may have existed, and hardly damns the platform as being somehow tragically insecure. (And, in fact, the recent security updates to Flash are nothing more than hardening against some potential XSS attacks.)
9) Sounds like MXML, again. Don't repeat yourself, you already mentioned XAML once.
Now, I'm certainly no Flash apologist -- up until about a month ago I refused to touch it, and ActionScript 2 is unbelievably shitty -- and certainly if we were comparing against Flash 8 or earlier running ActionScript 2 you'd have some valid points, but nothing on your list actually seems to me to be a valid reason why Silverlight is better than Flash here and now. And again, despite your protests that Microsoft is developing an OS X version of Silverlight, and is working with Mono to develop a Linux version, they have not been above releasing software for platforms and then dropping it without cause in the past, and I haven't seen them changing their colours.
Regards,
MySQL 5.0 has triggers, too, since 5.0.2.
Safari 3/WebKit's CSS support is excellent, probably the best now, or at the very least on par with Gecko. I don't know how many of the fixes that Apple has made to their rendering engine have worked their way back into KHTML, though. Konqueror 3.5.2 still had some issues when I last tried it, but most of them seemed related to a weak JavaScript implementation and not so much CSS, though there were a couple CSS bugs I saw (relative units were sized incorrectly and text-indent did not work correctly). According to bug reports filed in KDE's bug tracker, though, it has been improved muchly since then (it's up to 3.5.7 now), so chances are any issues you had before don't exist in the current version.
You are confusing patents with copyright.
You can set SQL modes, such as STRICT_ALL_TABLES, that will cause MySQL to reject invalid data instead of truncating it. There is documentation about the various SQL modes.
Sure, I'll give you some.
Inconsistent function naming (underscores):
substr_compare() vs.
strcmp()
More inconsistent function naming (verb location):
file_get_contents() vs.
get_html_translation_table()
Even within the same extension:
imagesetstyle() vs.
imagecolorset()
Flipped haystack and needle:
strpos(haystack, needle) vs.
in_array(needle, haystack)
Speed:
Scutigena Computer Language Performance Comparison (see graphs)
There used to be another site that you could compare one language's speed relative to another that also showed PHP as one of the slowest. I can't seem to find it now, though. Also PHP5 might compare a bit more favourably, but this is all I could find after a quick Google search. Perhaps more importantly, PHP drags the speed of other things down (like Apache), since even though the core is supposedly thread-safe, nobody seems to know which extensions are and aren't, so eg. Apache needs to be run in prefork mpm instead of using a threaded mpm.
I think PHP is overall a fairly decent language; I've used it for many years with great success. But it does have major problems, and it would be nice for them to get fixed instead of pushed aside. (I read some minutes from a PHP 6 meeting a while ago where they touched on the issue of consistency, and the PHP Group decided that it wasn't important enough to fix. It's really annoying to me to need a PHP-aware IDE or a manual always handy to program in a language because the arguments and function names are so non-uniform.)
Macs haven't used CR as the line delimiter since OS 9. OS X uses LF, like every other BSD and *nix.
No. 11500000 * .01% = 11500000 * .0001 = 1150 ;)
Do you work for Verizon?
Konqueror is like Windows Explorer on steroids. It is awesome. You should try it.
It sounds like, since you never let it go there, it keeps trying to go to the IE7 "First Start" page, which, contrary to all logic, is not a local page but is instead hosted on an MSN server. This page allows you to..er..change default IE settings. Seems rather stupid and insecure to me, but hey.
How is it elitist snobbery? You're asking the browser to handle invalid data for which the only true method for handling it is to bail out, since it's by definition impossible to parse it for proper meaning (it can only be guessed at, and just like dealing with vagueness in human languages, different software can have a different opinion about how the vagueness should be handled). XML parsers do exactly that, which is why I wish more people would start writing sites using XHTML (with the application/xhtml+xml MIME-type), so that this generation of ugly tag soup Web sites will eventually close.
If I started leaving out punctuation in my writing (missing close tags), switched what I was talking about mid-sentence without making a note of it (mismatched tags), added a bunch of obscure terms that only my close group of friends use (proprietary tags), put words where they made no sense (extra tags), and added a bunch of gibberish at random in the middle of my sentences (invalid tags), would you be elitist for asking me to clarify my writing when you were unable to read it?
Honestly, it's so easy to write "valid" markup that I'm convinced that only gross incompetence and disturbing amounts of laziness are the reasons that any invalid markup exists today. The rules for HTML and XML are very, very simple; The W3C provides an online validator that's VERY good at explaining exactly what's wrong with a document, and trying to give anyone a pass for generating invalid data while blaming Safari for interpreting differently (not incorrectly, since there is no 'correct' way to handle invalid data other than bailing out) the very same invalid data is doing the Web a disservice.
Just my $0.02, adjusted for inflation.
That's a really great idea! Good thing nobody's thought of it before, maybe you should get a patent... ;)