Maybe they've developed a new inventive approach. If so, fair enough. If not, then the patent will not be enforceable and they've just burned cash on it.
Validity of a patent is something that is very difficult to attack. The courts assume that a patent is valid and it's expensive and time consuming to counter it, no matter how ridiculous. The USPTO doesn't proactively attempt to invalidate patents (note that the patents for swinging and playing with cats are both still valid), you must be willing to pay (possibly very large sums of) money and spend time in court to counteract a patent, and (usually) you have to wait until you get sued to do it. Frivolous patents harm all of us, whether or not they're ever used.
Umm, you do know that it's a necessary condition of patentability that the invention is sufficiently disclosed so that it can be worked by one skilled in the art. If not, then the patent is declared invalid. Again, they will have justed wasted their money.
The first part of this is true, the second is not. The PTO is _supposed_ to enforce this before the granting of the patent. They're also really shitty at this and totally fall down, review pretty much any of the patents posted in Slashdot stories for examples. The Eolas patent is one of the best examples.
Re:What in God's name...?
on
Core CSS (2nd ed.)
·
· Score: 2, Interesting
For what it's worth, most of the IE extensions are usefull, and should be absorbed into the spec/implemented by other browsers. The ones I miss most often are overflow-x and overflow-y, but the layout options for non-Western character sets are neat too. According to MS, most (althought not all) of the extentions have been submitted to the W3C, so they should get off thier asses and add them to the spec so Mozilla devs can support them without feeling naughty.
CSS positioning is fundamentally broken. It's simply impossible to acurately recreate some of the stuff tables do, and even where it's not it's a (great deal) more work to recreate many simple layouts. Even something as simple as "center this element on it's parent" is impossible without absolute positioning.
For styling CSS is da bomb. For layout, template based pages are at least as (if not more) maintainable, have the advantage of being cross-browser compliant (including text-based browsers), and are, in most cases, easier to understand and maintain. If you're going for pixel-perfect positioning, you'll want CSS though.
The whole seperation of content thing is crap, though. The structure of the HTML still affects both the layout and the formatting, so there's no way you can get a totally clean break. Accept that fact that alterations to the sites HTML will require changes to the CSS and vice versa.
The cross-browser issues are a real pain with CSS - it's worse than the hacks we all used back in 93 to make stuff render correctly. IE may be broken and non-compilant but it's still got the 80-90% market share and if you're doing any large-scale site it MUST render correctly on IE. That means giving up on a whole ton of CSS functionality, and especially if you're using it for positioning. One more reason to stick with tables for another couple years.
Maybe CSS 3 will adress some of the issues. For the time being, I ain't real impressed. It's certainly a step up on font tags, but it's got a long way to go before it's usefull for positioning.
I should rather say that the implementation of GC in the most popular languages that uses it precludes them. D (digitalmars.com/d) has automatic objects and GC and if it had the backing of Sun or Microsoft I think it'd be tremendously popular.
Just to be clear, I think that the argument that sharing with a million people is "limited" is ridiculous. However, defining indefinite copyright duration as "limited" is equally ridiculous and I was pointing out the hypocrisy, not trying to justify piracy.
Klowners work (at least the bubble-tux one) doesn't have any license information. This is a pretty severe oversight on his part imo, and it would explain how this could be an accident rather than an intentional/uncaring thieving of art. It's a site specifically intented for people to download and use the images, after all.
That said, people should have enough copyright know-how (certainly the commercial artist who did the flash movie) to know better than to just randomly grab pics.
Well, I dunno. Copyright law says that fair use is "limited" sharing, right? And the Constitution says that copyright is for a "limited" time, right? And Congress and the Supreme Court say that any amount of time less than forever is limited, so therefore sharing with any amount of people less than everyone is equally "limited".
It's not a problem with GC so much but that the implementation of GC in the language precludes the constructs that let you efficently manage other resources. I'm only aware of one language that has both GC and automatic objects. Java is probably the most poplar GCed language/environment, and it certainly brought GC into the mainstream, so it's particular resource management flaws are often seen to be part of GC. (Functional languages that use closures to manage resources are a different issue, and I have no idea how GC interacts with that if at all).
Sun is being misleading. You can do that, but it's a lousy idea. Read up on when finalization occurs (the relevant docs were posted in part elsewhere in this thread). Basically, it's run when the object is GCed, not when it goes out of scope (which makes it useless for RAII), and it's not guaranted to be run at all. Finalization is not the same as destruction, and can't be used for the same purposes. Simple as that.
I'll try to clear it up a little more with a better example. Say you're locking a file. In C++, you'd create an object to control the lock.
class locker { locker() {lockfile();} ~locker() {unlockfile();} }; function WriteFile() { locker l; .... }
The file is locked when the function is entered, and is unlocked when the function is left, no mater how it's left - early return, exception, anything. There is no way to leave that function scope with the file locked. You can't do anything like that with Java, because it has no automatic objects and non-deterministic finalization.
This is one reason why I like C++ so much - because you aren't locked into a refcounting model. Especially with role-based templates, you can use any number of different styles of smart pointer (weak refs, copy on write, threadsafe, simple auto_pointers...) with very little trouble. A language that handles ref counting for you must do it in a more generic fashion and will lack either functionality or performance (or both) at one time or another. JIT and very smart compilers can help with some of this (eliminate resource locking if you can tell that a resource is never accessed from multiple threads) but they aren't a panacea and I don't think any current JIT implemetations can do this anyway.
D (see digitalmars.com/d, or the recent Slashdot article) is the holy grail here as far as I'm concerned - it has GC but also retains automatic objects you can use for RAII. On top of that, the GC is much more amenable to bypassing, unlike Java or C#s.
Thats pretty cool, but I suspect that the code gets really tangled if you've got a need to lock more than one resource - you'd need chains of closures. It's functionally (hehe!) equivilent to the C++ RAII model, but I think the C++ one is more concise and clear.
On the other hand,it can probably deal with transaction-style locks easier than RAII can - although I've seen a system to handle that that uses RAII objects combined with functors (instead of closures). It works almost identically to the Ruby model.
Finalizers in Java are equvilent, in C++ terms, to not doing any cleanup and relying on the OS to do it. Except it's even worse, because finalizers are not guaranted to be called. They do not provide the functionality of destructors, which are guaranteed to be called when the object goes out of scope. Using finalizers the same way a C++ programmer uses destructors is a HUGE programming error and will result in resource leaks.
It's the same right I've got to say "Hey, you can buy Pepsi at my store", if in fact I sell Pepsi. It's not infringment to refer to a specific product or company using it's mark - thats what the mark is for, after all. It'd be a violation if I didn't actually sell Pepsi, just my store brand. That said, I think thats the most marginal case except, as you say, if axa.co.uk isn't actually run by a company with a legit claim to the AXA mark. Since the site is down I'm going to give the benefit of the doubt, but note that just as I have no way to tell whether or not it's valid, neither does Google.
I have no way of seeing the ads that prompted the suite (presumably the ones on google.fr), but the ones available on the other Google sites do not, in fact, do this. They all use the AXA mark in relation to the AXA product itself (IE, competitive quotes comparing AXA to other insurance providers, business information about the stock symbol AXA, and reselling of AXA insurance). The only exception is an ad for axa.co.uk, and it's possible thats a competitor using the AXA mark illegaly, but I have no way of verifying that and neither does Google. There's no easy real world equvilent to computerized searching - it's certainly NOT the same as portraying one brand as another. The closest I can come is placing competing products on the same shelf (store brand w/ name brand, in particular), or offering coupons for a competing product when you buy a name brand product. And none of those things are illegal.
None of those companies sell insurance. Vault.com is a business site (the adword is probably because of the stock symbol, not the company name). Discountedorfree.com is a clearinghouse of low-cost crap, they've got the keyword because companies who advertise there resell AXA insurance. BuyXYZ.co.uk and 360clicks.co.uk are comparison services that provide competetive quotes. I'm not familiar with French law, but under US law I'm pretty sure that all of those (except possibly discountedorfree.com) would be reasonable "fair use" of trademarks. Especially vault.com - can you even claim a trademark on your stock symbol?
Oops, forgot about the adword on top. Axa.co.uk is down for me, but I'm going to presume that it is either the UK site for the AXA mentioned in the article or that it's a different company that owns the AXA trademark in the UK. In either case it's not a trademark violation.
Utter BS. Any company publishing a phone book will be perfectly happy to sell you an ad under any name you want. If you don't own the rights to that name and are infringing a trademark, it's your problem, and you'll be sued to pull the ad, not whoever publishes the phone book. The due dilligence required to ensure that no advertiser is breaking trademark is completely unreasonable, and on top of that it places the burden of defending the trademark on an uninterested party - namely Google. Assume for the moment that the AXA in the suite is actually violating some other companys trademark - now Google is being drawn into a case in which it has no stake.
The correct action here would be to sue the companies doing the advertising. And all this is even assuming that the case is legitimate, because the ads I see on the generic Google site (they've been pulled from google.fr) don't appear to infringe the trademark in any way.
Actually, this isn't (yet) true. As a private company, Google has no obligation whatsoever to be a souless capital seeking machine, and is perfectly free to decide that it's primarly an information service, with the advertising and cash flow being a useful side effect. And, judging from both Googles origins and it's actions I suspect thats more or less how the company views itself.
In the US, you're allowed to use trademarks when you're referring to a specific company or product, so it's perfectly legal to use a competitors brand in your ads if you're directly comparing yourself to them. I can't speak to what ads they're talking about (on plain old Google.com, the ads I come up with are for meta-services offering market information and such - I don't know exactly what AXA does but if they're an insurance company I don't think any of these guys compete), but as long as the ad itself doesn't present itself as being AXA, I don't see the problem. It's certainly not Googles problem - this is a pretty clear case where public interest is preserved by tightening the scope of trademarks, rather than broadening it.
Update: there's no ads at all for AXA on google.fr. Anyone know what actual ads these guys are bitching about?
The Bible contains tons of historical fact, and in fact it's well-researched by anthropologist and archeaologists. On the other hand, there's also tons of stuff in the Bible thats totally contradicted by the archaeological evidence, too. It's usefull to science in the same way that any very old text is usefull. One particular find isn't going to change anything.
I'm also a web developer. Firefox renders up to twice as fast as IE on any given page. The only times its slower is when loading plugins (acrobat in particular). Ctr-T to open a new tab is _far_ faster than a new IE window for me, since IE has to spawn a new process, then start loading the page involved. The web development tools available in Firefox and Mozilla dwarf IE - script debugger that doesn't suck, the ability to outline elements, view the dom model in real time, edit CSS styles in real time (this is _huge_), real time header information, usefull document property pages (for example, a list of all the links on a page) - all are great. This is to say nothing of IEs crappy CSS support, which really is a marginal issue for me since it has to work in IE so I'm stuck dealing with those issues no matter what my personal browser choice is.
Oh, I just did a test on my machine (in it's normal state for me, which is full of all sorts of windows and applications). Opening a new IE window took about 1/3rd of second from hitting Ctrl-N to page load, with a blank page. Opening a new Firefox tab took less time than it took my fingers to get back on home keys after hitting ctrl-T. Opening a new Firefox window took about the same 1/3rd of a second. This is on a P3-900 with 512 megs of ram, hardly high end.
In summary: There's plenty to bash IE about. I suspect that if you used FireFox for a week you'd get over your IE habits (thats about how long it took me to get used to ctrl-t instead of ctrl-n, f6 to highlight address bar instead of ctrl-tab, and to become irreversably addicted to find-as-you-type), you wouldn't want to switch back.
He doesn't have the options dialog on Office 97, and there's no dismisal option. You must navigate the Office menus to disable him. I should point out that he hangs office for a good 30 seconds the first time he loads, too. Stupid old hardware.
I interoperate with everyone else at work (who all use Office 97) without difficulty using openoffice. OOo doesn't export properly to PowerPoint 97 (although it imports fine, and it exports to PowerPoint 2000/XP fine), but thats a minor issue for me at best.
Granted, it's not like we're doing typesetting or anything with Word - but then, you shouldn't be. Your basic "words with styling and some pictures and maybe a table or 3" documents work without a hitch.
The best part of the lightbulb (as opposed to clippy) is that it stays out of your way and is trivially turned off. In fact, it doesn't bother me enough to turn it off, unlike Clippy, who makes me gnash my teeth in rage and almost crumple my mouse as I try to find a way to disable it. Note that clippy has no "Shut the hell up and leave me alone" option in his stupid little dialoge - you have to sit there looking at him while you dig through menus trying to disable him.
Slashdot Mirror
Validity of a patent is something that is very difficult to attack. The courts assume that a patent is valid and it's expensive and time consuming to counter it, no matter how ridiculous. The USPTO doesn't proactively attempt to invalidate patents (note that the patents for swinging and playing with cats are both still valid), you must be willing to pay (possibly very large sums of) money and spend time in court to counteract a patent, and (usually) you have to wait until you get sued to do it. Frivolous patents harm all of us, whether or not they're ever used.
Umm, you do know that it's a necessary condition of patentability that the invention is sufficiently disclosed so that it can be worked by one skilled in the art. If not, then the patent is declared invalid. Again, they will have justed wasted their money.
The first part of this is true, the second is not. The PTO is _supposed_ to enforce this before the granting of the patent. They're also really shitty at this and totally fall down, review pretty much any of the patents posted in Slashdot stories for examples. The Eolas patent is one of the best examples.
For what it's worth, most of the IE extensions are usefull, and should be absorbed into the spec/implemented by other browsers. The ones I miss most often are overflow-x and overflow-y, but the layout options for non-Western character sets are neat too. According to MS, most (althought not all) of the extentions have been submitted to the W3C, so they should get off thier asses and add them to the spec so Mozilla devs can support them without feeling naughty.
For styling CSS is da bomb. For layout, template based pages are at least as (if not more) maintainable, have the advantage of being cross-browser compliant (including text-based browsers), and are, in most cases, easier to understand and maintain. If you're going for pixel-perfect positioning, you'll want CSS though.
The whole seperation of content thing is crap, though. The structure of the HTML still affects both the layout and the formatting, so there's no way you can get a totally clean break. Accept that fact that alterations to the sites HTML will require changes to the CSS and vice versa.
The cross-browser issues are a real pain with CSS - it's worse than the hacks we all used back in 93 to make stuff render correctly. IE may be broken and non-compilant but it's still got the 80-90% market share and if you're doing any large-scale site it MUST render correctly on IE. That means giving up on a whole ton of CSS functionality, and especially if you're using it for positioning. One more reason to stick with tables for another couple years.
Maybe CSS 3 will adress some of the issues. For the time being, I ain't real impressed. It's certainly a step up on font tags, but it's got a long way to go before it's usefull for positioning.
SlickEdit (you know, the thing in TFA) is based on Eclipse.
I should rather say that the implementation of GC in the most popular languages that uses it precludes them. D (digitalmars.com/d) has automatic objects and GC and if it had the backing of Sun or Microsoft I think it'd be tremendously popular.
Just to be clear, I think that the argument that sharing with a million people is "limited" is ridiculous. However, defining indefinite copyright duration as "limited" is equally ridiculous and I was pointing out the hypocrisy, not trying to justify piracy.
That said, people should have enough copyright know-how (certainly the commercial artist who did the flash movie) to know better than to just randomly grab pics.
Well, I dunno. Copyright law says that fair use is "limited" sharing, right? And the Constitution says that copyright is for a "limited" time, right? And Congress and the Supreme Court say that any amount of time less than forever is limited, so therefore sharing with any amount of people less than everyone is equally "limited".
It's not a problem with GC so much but that the implementation of GC in the language precludes the constructs that let you efficently manage other resources. I'm only aware of one language that has both GC and automatic objects. Java is probably the most poplar GCed language/environment, and it certainly brought GC into the mainstream, so it's particular resource management flaws are often seen to be part of GC. (Functional languages that use closures to manage resources are a different issue, and I have no idea how GC interacts with that if at all).
I'll try to clear it up a little more with a better example. Say you're locking a file. In C++, you'd create an object to control the lock.
The file is locked when the function is entered, and is unlocked when the function is left, no mater how it's left - early return, exception, anything. There is no way to leave that function scope with the file locked. You can't do anything like that with Java, because it has no automatic objects and non-deterministic finalization.This is one reason why I like C++ so much - because you aren't locked into a refcounting model. Especially with role-based templates, you can use any number of different styles of smart pointer (weak refs, copy on write, threadsafe, simple auto_pointers...) with very little trouble. A language that handles ref counting for you must do it in a more generic fashion and will lack either functionality or performance (or both) at one time or another. JIT and very smart compilers can help with some of this (eliminate resource locking if you can tell that a resource is never accessed from multiple threads) but they aren't a panacea and I don't think any current JIT implemetations can do this anyway.
D (see digitalmars.com/d, or the recent Slashdot article) is the holy grail here as far as I'm concerned - it has GC but also retains automatic objects you can use for RAII. On top of that, the GC is much more amenable to bypassing, unlike Java or C#s.
On the other hand,it can probably deal with transaction-style locks easier than RAII can - although I've seen a system to handle that that uses RAII objects combined with functors (instead of closures). It works almost identically to the Ruby model.
Finalizers in Java are equvilent, in C++ terms, to not doing any cleanup and relying on the OS to do it. Except it's even worse, because finalizers are not guaranted to be called. They do not provide the functionality of destructors, which are guaranteed to be called when the object goes out of scope. Using finalizers the same way a C++ programmer uses destructors is a HUGE programming error and will result in resource leaks.
It's the same right I've got to say "Hey, you can buy Pepsi at my store", if in fact I sell Pepsi. It's not infringment to refer to a specific product or company using it's mark - thats what the mark is for, after all. It'd be a violation if I didn't actually sell Pepsi, just my store brand. That said, I think thats the most marginal case except, as you say, if axa.co.uk isn't actually run by a company with a legit claim to the AXA mark. Since the site is down I'm going to give the benefit of the doubt, but note that just as I have no way to tell whether or not it's valid, neither does Google.
I have no way of seeing the ads that prompted the suite (presumably the ones on google.fr), but the ones available on the other Google sites do not, in fact, do this. They all use the AXA mark in relation to the AXA product itself (IE, competitive quotes comparing AXA to other insurance providers, business information about the stock symbol AXA, and reselling of AXA insurance). The only exception is an ad for axa.co.uk, and it's possible thats a competitor using the AXA mark illegaly, but I have no way of verifying that and neither does Google. There's no easy real world equvilent to computerized searching - it's certainly NOT the same as portraying one brand as another. The closest I can come is placing competing products on the same shelf (store brand w/ name brand, in particular), or offering coupons for a competing product when you buy a name brand product. And none of those things are illegal.
Oops, forgot about the adword on top. Axa.co.uk is down for me, but I'm going to presume that it is either the UK site for the AXA mentioned in the article or that it's a different company that owns the AXA trademark in the UK. In either case it's not a trademark violation.
The correct action here would be to sue the companies doing the advertising. And all this is even assuming that the case is legitimate, because the ads I see on the generic Google site (they've been pulled from google.fr) don't appear to infringe the trademark in any way.
Actually, this isn't (yet) true. As a private company, Google has no obligation whatsoever to be a souless capital seeking machine, and is perfectly free to decide that it's primarly an information service, with the advertising and cash flow being a useful side effect. And, judging from both Googles origins and it's actions I suspect thats more or less how the company views itself.
Update: there's no ads at all for AXA on google.fr. Anyone know what actual ads these guys are bitching about?
The Bible contains tons of historical fact, and in fact it's well-researched by anthropologist and archeaologists. On the other hand, there's also tons of stuff in the Bible thats totally contradicted by the archaeological evidence, too. It's usefull to science in the same way that any very old text is usefull. One particular find isn't going to change anything.
Oh, I just did a test on my machine (in it's normal state for me, which is full of all sorts of windows and applications). Opening a new IE window took about 1/3rd of second from hitting Ctrl-N to page load, with a blank page. Opening a new Firefox tab took less time than it took my fingers to get back on home keys after hitting ctrl-T. Opening a new Firefox window took about the same 1/3rd of a second. This is on a P3-900 with 512 megs of ram, hardly high end.
In summary: There's plenty to bash IE about. I suspect that if you used FireFox for a week you'd get over your IE habits (thats about how long it took me to get used to ctrl-t instead of ctrl-n, f6 to highlight address bar instead of ctrl-tab, and to become irreversably addicted to find-as-you-type), you wouldn't want to switch back.
He doesn't have the options dialog on Office 97, and there's no dismisal option. You must navigate the Office menus to disable him. I should point out that he hangs office for a good 30 seconds the first time he loads, too. Stupid old hardware.
Granted, it's not like we're doing typesetting or anything with Word - but then, you shouldn't be. Your basic "words with styling and some pictures and maybe a table or 3" documents work without a hitch.
The best part of the lightbulb (as opposed to clippy) is that it stays out of your way and is trivially turned off. In fact, it doesn't bother me enough to turn it off, unlike Clippy, who makes me gnash my teeth in rage and almost crumple my mouse as I try to find a way to disable it. Note that clippy has no "Shut the hell up and leave me alone" option in his stupid little dialoge - you have to sit there looking at him while you dig through menus trying to disable him.