Businesses fear Open Source like the plague because they're afraid of govenments "buying" software then declaring it "Open Source" they don't have to pay.
How the hell is that supposed to work? If you contract me to produce some software for you, and I use open source, you still have to pay me the agreed amount or see me in court.
Any company's lawyers will understand this. If they make such an argument, they are simply lying. Their real motive is that they don't want you to see some parts of the code. This could be because they're embarrassed by the shoddy quality. More often it's because there are things there in addition to what you think you paid for.
In the case of Diebold, they made this very clear before the 2004 election, when then-CEO Wally O'Dell said - in writing - to the Ohio Republicans that he would deliver their state to George Bush. He lived up to that promise, and there are good grounds to suspect that this wasn't at all accidental. They want their code secret so that we can't find out some of the things they've got hidden there.
In the case of elections, paranoia is simply rational. History tells us that the people running an election will cheat if given the slightest opportunity. Secret code makes cheating very easy, and the assumption should always be that secrecy like this is to hide what's going on.
The only practical way to get honest elections with computerized equipment is to require that all the code be open and visible to the public. Anything less is a guarantee of dishonest elections.
(Guaranteeing that the published code is actually what's running inside the machine is another issue. We need a way to do that, too.)
Postmodernism is what happens when the rate of intellectual change approaches C'.... C' is something I just made up,
I think you've illustrated the primary problem here.
There is a rather old distinction in logic, philosophy and linguistics: the symbol/referent distinction. This is the difference between a symbol and the thing that it represents. There are a lot of logical errors and social problems that are based on confusion of these.
Much of the problem is the tendency to think that if something has been named, then it exists. This leads to the belief in things as great as our various gods, societies, corporations, and such, down to things as fleeting and inconsequential as postmodernism and Web 2.0.
Marketers do tend to understand this, and use the logical confusion to their advantage. This is the primary origin of the term "Web 2.0". It was invented as a marketing tool, to trick people into thinking that they had invented something new. Much of the critisism is pointing out that there's really nothing behind the term.
Commenting on Web 2.0 and postmodernism by blithely inventing a "C'", with a definition that isn't amenable to any sort of testing, is a good way of illustrating what's really going on here.
[Bush] has encountered virtually no resistance or scrutiny from Congress, and has skillful deceptive tactictians who, in a very real, cynical, Machiavellian sense, have artfully deceived the entire world, America included, into turning over as much power as possible to them and their cronies.
I don't believe they have been deceptive at all. From well before Bush was elected governor of Texas, his behavior and policies have been quite thoroughly documented and the story has been available for anyone to read who has the slightest interest in the subject. And if you look at the public information before the invasion of Iraq, you'll find that all of the Bush gangs claims had been thoroughly and very publicly shown false. They were reduced to arguing that they had to invade because of what Saddam might do in the future.
The problem is that about half the American voting population approves of Bush's policies and actions. This isn't from ignorance. We've all been told repeatedly and in great detail of his gang's lies and corruption. Anyone voting for him should be assumed to know about it. Their support implies that they approve.
In particular, many people have explained in great detail why Bush's war in Iraq is highly illegal under both international and American law. He and most of his Cabinet could easily be impleached and thrown in jail for the rest of their lives. This isn't going to happen. The reason is that most of the Senate (including most of the Democrats) know and approve of it all.
... one is about lying about sexual relations with a woman and the other is about destroying a fundamental structure of our government.
Yeah. But the real problem is: Our current Republican leadership thinks that the former is a serious problem worth prosecuting. The latter is not just acceptable behavior; it's policy.
12% admitted that they often remain online longer than expected.
When you consider the effect of buggy browsers, combined with confusing, poorly-designed web sites, this is just the normal situation for all of us. It's not a symptom of any problem in the user at all.
Animal and plant breeders have a phrase "hybrid vigor". Google for it.
The theoretical explanation seems a bit weak so far. It's mostly based on the observation that dominant traits tend to be adaptive, while recessive traits tend to be maladaptive. This isn't always true, of course; it's just a statistical pattern. But it would explain why first-generation crosses between distantly-related strains tend to produce vigorous offspring.
So in the future the dimwitted underclass will end up running society and subjugating the better-looking upperclass?
Actually, this has already happened numerous times in the past couple centuries. Look into the family histories of the rulers of the Ottoman, Russian and Chinese empires for instructive examples. And they're just the best-documented of all the cases known.
The US is now on the familiar path to developing its small class of hereditary idiots who occupy the country's top offices. Stay tuned and see how it turns out.
The British do seem to have figured it out, though. They've walled off their group of idiot rulers so they're merely entertaining, but no longer a danger to society, and nobody trying to replace them. The citizenry is probably a long way from thinking that their rulers would make better sausage.
I looked through the review for an idea of what the phrase "first class" might mean to the writers, but I didn't spot it. This was a way of getting at the real question: What sort of web site is being pushed here?
My suspicion, based on lots of other similar examples of advice on building web sites the Right Way, is that "first class" means flashy and entertaining. Maybe I'm wrong, but experience here says to expect the worst.
Not that I'm objecting to that. There's a demand for entertainment on the web. But there's also a demand for informative sites, and the design criteria for information and entertainment are very different.
Most of the web sites I've been involved with are of the informative type. For such sites, the canonical "first class" site is google.com, not (for example) yahoo.com. For a more complex site, news.google.com works pretty well.
But I didn't get a feel from the review or the author's web site what their criteria for "first class" might be. Unless I get a clear idea what sort of web site is being pushed, I probably won't be spending my money on any book of advice.
(I might also obsever that most entertainment sites are really in need of good advice on how to do the job right. But I probably don't need to tell that to anyone who visits them often.;-)
..., it should require the "I agree". It's more of a CYA thing to do.
Well, maybe, but people do routinely point out that the GPL doesn't require that you agree with its provisions. If you don't agree, you simply don't follow the extra rights that it gives you. You just use the software like you would anything else, under the terms of your local copyright laws. You don't need anyone's permission to do that. And it doesn't matter whether or not you agree with the copyright laws; they are in effect whether you agree or not.
So how does any user clicking add any CYA aspect to the GPL?
The only motivation I can see here is satisfying the users' expectation that they'll have to click on an (unreadable legalese) agreement. But that doesn't mean that clicking has any legal import in this case. You're merely following a custom that many users now expect, even in a case where it has no legal import.
My firefox very, very rarely crashes (once every few *months* Java or Flash bring it down).
Well, sure, it crashes rarely. But I find that I have to kill it every other day, because its memory usage has grown to several times the mere gigabyte of memory that I have. Closing all its windows only releases a small amount of its claim on memory. So I kill it, all my other apps speed up, and a new FF runs quickly after I've reloaded the pages I was working on.
Saying that something "very rarely crashes" is a bit disingenuous, when what it does is drag the system (itself included) to a near halt. "But it didn't crash!" Yeah, right.
(This is on both a linux RH system and a Mac with OSX 10.4.whatever the current version is. YMMV, of course, depending on what you're using FF for.)
So when did Microsoft ever "compete" with any other company? That's not how they do business, y'know.
Consider that they got their start as a subcontractor for IBM, and used IBM's economic clout to enforce "agreements" with retail vendors that effectively locked out other startups. That was so successful that they've never had much of a motive to "compete" in any ordinary sense of the word. They don't compete; they make deals. When they have to, they engage in classical price wars, but that's a last resort. They used a price war to bankrupt Netscape, of course. But their usual approach is more like how they're now trying to lock out all those pesky vendors of Windows security software. They'll succeed at that, too, as they have with all sorts of other software.
That's how big business works in the Real World [TM], y'know. You don't profit by competing. You profit by ensuring that the customers don't know about and can't easily find the competitors' products.
The iPod is an abberation. MS screwed up, and failed to take the appropriate steps before the masses became aware of this new product. It's a bit late for them now, though they still have a chance of winning if they can make it sufficiently difficult for Windows users to use iPods. As a last resort, they can give away their own player "free", i.e., with the price included in a Vista system.
Well duh, of course it's caused by a programmer. Doesn't mean there's a good reason for it,...
I was replying to the original comment:
But there's no reason...
Note the lack of the word "good" in the original. I agree that there's no good reason; I was just pointing out that there's a reason. And yes, it's a fairly obvious reason, if perhaps not a good one.
Actually, if you look at the origin of the mozilla suite in the Mosaic browser, I'd claim that it was reasonable to implement it this way at first. Building Mosaic for the first time was a bit of a task, and it's no surprise that little issues like background threads eating up cpu time might have been ignored. Mosaic was, to a great extent, a "proof of concept", and it was spectacularly successful for its time.
But that was then. Now, the marketers are taking advantage of this to saturate our cpus for the purpose of distracting us with active images. This is growing into a serious problem. There would be a real advantage to a browser with the ability to block this waste of cpu.
Firefox has some tools for controling cpu use. But the tools are scattered, ad hoc, incomplete, and difficult to use. There's a gimmick to stop active images. There's a gimmick to turn javascript on and off. There's an extension to block flash. There's no way I know to block a tag. There's no consistency to any of this. Opera has some good ideas, but it's also incomplete. I wonder if there's a way to unify the problem into a single, consistent tool?
But there's no reason [active images] should be eating CPU when the window is minimized or when I'm looking at a different tab.
Actually, there's a very good reason: Some programmer implemented it that way.
That's the reason that most software does what it does. If there are several ways that something could be done, you can expect that all of them will be implemented in some software somewhere. It's basically at the programmer's whim.
In the case of active images, there's no sensible reason for keeping them running when they're not visible on the screen. But stopping them and restarting them takes explicit actions on the part of the software. This requires that the programmer(s) be aware of the issue and write the code to do it right. You'd expect that initial releases of image-displaying software wouldn't do this.
It's quite difficult for a user to verify that it's actually an active image that's eating the cpu. But it can be done. Use the context menu to load the image into a separate tab. Wait for cpu usage to stabilize, and note the browser's cpu usage. Close that tab, wait a bit, and note the browser's cpu usage. If it has decreased, you have some partial evidence. Re-create the tab with the active image, and see if the cpu usage goes back up. If so, you have better evidence. Close the tab, and see if the cpu usage declines. If so, it's really good evidence.
I've done this, and concluded that FF does indeed keep images "running" when they're not visible on the screen. This happens if I switch to another tab, and also if I minimize the window.
Even worse, it happens with all "active" elements. Javascript keeps running, as do any video plugins. The only way to stop them is to disable the feature globally (or per site with Opera).
There's lots of room for future improvements here.
Right now Firefox is using about 60% of my CPU, and it's minimized. (Before you ask, I also have Flashblock, so there shouldn't be any Flash apps running.)
Try turning off java and javascript. And set images to loop only once.
These steps won't get everything that uses cpu when "idle", but they all help. Advertisers like to use all these things to make their ads distractive. There are some other cpu-eating features that probably can't be turned off, such as the auto-refresh in a page's tags.
What would really help is a thorough study of all the ways that a page can continue to use cpu. We need a list of them, and an on/off flag for each one. If FF (or seamonkey) would give us something like this, it would be another good advertising point.
> Perhaps they can start talking about tienamen square maturely instead of pretending it didn't happen.
*snort* Good luck on that one. I still have yet to hear a mature discussion outside of university walls on the internment of Americans during WWII:...
I have, but I'd agree that it's not a common topic of conversation. I even grew up in the Seattle area, with a lot of friends of Asian background, and I remember a number of calm discussions of the shameful event. Most of the people that I knew, both of Japanese and European background, just viewed it as "Yeah, isn't it horrible how people treat each other sometimes?"
Finding non-academic discussions of the topic isn't difficult any more. You can, of course, find a good summary at wikipedia. I also found that googling for "Japanese American internment" gets about 1.4 million hits, and a quick check of a few shows that reasoned, factual articles and discussions seem common.
Of course, the rabid bigot types won't be part of such discussions. And I do remember a few such people from my childhood. We mostly ignored them, while being aware that they were the sort of people who would do it again if given a chance. I didn't have a lot of friends like that, though.
We're seeing this right now, of course, with the American treatment of Arab-Americans and others of Middle-Eastern origin. It's not to the point of mass-internment camps, but those of us following the story are aware of what's going on. This includes the government's mistreatment of people, using an attack by unrelated other people as an excuse. And the mistreatment includes arrest and indefinite incarceration without charges or trial. In some cases, it includes torture of people later shown to be innocent of any suspected crimes.
Lots of us are quite aware of our country's history in such matters. Attempts to suppress knowledge of this history are not very successful, and any American who wants to learn the facts can now find them quite easily.
The Chinese leaders may have realized that the economic value to them of access to sites like wikipedia is greater than any danger that people might read politically unacceptable information. After all, the US doesn't collapse because its people have access to information about the more sordid parts of their government's history. Most Chinese citizens will remain as safely ignorant as most Americans are.
: On a somewhat related note, are new features even what people want in a browser?
What I'd like, and what would make me a frequent user of a browser that did it, is an "anti-feature": A simple-to-use tool that lets me turn off (and occasionally on) all of the so-called features that soak up cpu.
The problem is that I often have a number of browsers open with lots of windows and tabs. After a while, I find that the browsers are using most of my cpu. If I want to do something that uses more than a small amount of cpu, I have to shut down all my browsers.
With various browsers, I've discovered how to disable most of these cpu-eating features. I have flashblock installed in the browsers that will accept it. I turn off java and javascript. I set images to run only once (in the browsers that let me do this). And so on. But it doesn't solve the problem. All the browsers, including firefox, eventually start gobbling cpu, and I have to kill them.
What I want is a window that lists all the "features" that can use cpu time. The window should let me enable and disable each such feature separately. Even better would be something like flashblock, that lets me enable one such feature just once in a single page. Next best is what Opera does: It lets me enable a few such features on a per-site basis, and blocks the rest. (But it doesn't do this for everything, and I have to kill Opera at times when it suddely goes to 80% of the cpu.)
This is a major problem if you do a lot of web testing, and need to run a lot of browsers. It's also a problem if you're involved in things that require following a number of blogs or other such online discussions (like slashdot;-). It's even a problem if you're following a dozen or so webcomics, since most of them have ads, and some of the ads suddenly take over your cpu.
I also find that I have to shut down most of my browsers before I burn a CD. A browser suddenly grabbing 90% of the cpu for a second or so can really kill a CD write operation.
If the FF folks would give us a good tool for this, it could be a very popular new feature. And it would teach the advertisers a much-needed lesson.
Heh. Here I was all prepared to to point out that the obvious answer is "both", but I noticed that every single reply so far has said the same thing. So rather than be rated redundant, I'll just point out that there is a standard name for this particular bit of illogic.
If authors had any sense, they'd be jumping onto the anti-copyright bandwagon. Such laws were, we're told repeatedly, created to give authors and artists control over their creations, and to guarantee them income from sales. More and more, the actual effect is to take away the creators' control, giving control and profits primarily to their corporate masters.
This story just illustrates that some authors have figured out that it helps to let their readers know what's available. And the copyright question basically asks whether a publisher has the right to block communication between an author and the audience.
Maybe we do need some sort of copyright laws. But authors don't need the current copyright laws. That's what's keeping most of them poor.
There is the conjecture that pi is a "normal" number, in which case Sagan's circular feature is there regardless of the base that you use. It's there an infinite number of times, in fact.
But to my knowledge, pi hasn't yet been proved "normal". It just appears so from statistics on the first trillion or so digits in base 10, which is hardly a proof.
To provide a book that passes muster with SW developers would be a blunder.
Depends. You're right, if the only goal is to maximise the sales of the book. If you want to improve the quality of software, you're wrong, though. This book just got a "don't bother" review in a tech forum. So we computer geeks won't read it, and we won't know what (if any) useful suggestions he might have for us.
Of course, you could take this as an example of the problem that he's describing, in the publishing industry. Like most commercial software, the book was "managed" by people whose main (and perhaps only) motive is to maximise sales. Such managers have little interest in what happens after the sale, except insofar as it might affect subsequent sales. In particular, neither software nor publishing managers are concerned with writing software or books that appeal to their fields' "geeks". They want the product out the door by a fixed date, and they want whatever gimmicks will increase sales to the masses.
The result isn't all that difficult to understand. Especially if you've been involved in software and/or publishing, and seen firsthand the pressures that management puts on the authors.
Of course, here and there you do find good managers, and sometimes they can help a quality product sneak through the process.
I don't think it's a bad thing that Microsoft has made it impractical to charge for a web browser. How is it a bad thing if they make it impractical to charge for anti-virus software?
Good point. And we might generalize it a bit. We often read here that old canard "You get what you pay for". With software, not only is this not generally true; what's more common is that with software, price and quality are typically inversely related.
Microsoft is merely doing its part to maintain this situation. They do it in a somewhat subtle way: They pretend that much of it is free, but you do in fact pay for IE and for MS's anti-virus software, as part of the price for their entire "system". You get crappy, poorly-functioning software, of course, in agreement with the price-quality rule. If you want quality, you have to download and install either shareware or free software.
Actually, there is somewhat of a parallel for this outside of computers. It's well known that, if you want quality audio or video equipment, you don't buy the all-in-one "systems". Those are simple purchases, and the components do work together (and are typically integrated into one box so that they appear to be a single product). But to get quality, you have to buy individual components, and interconnect them yourself. This takes time for study and wiring, but the end result will be much better quality.
Microsoft systems are like this. They sell as a "system", but the overall quality is low, especially since the components generally don't inter-operate nearly as well as advertised. Like A/V equipment, if you want quality, you'll just have to spend the time to install the quality components yourself.
The difference is that, with quality A/V equipment, the good stuff usually costs more than the crappy "integrated system" box. With software, the good stuff is usually a lot cheaper than the integrated junk. And when you look at all the hair-pulling and time-wasting futzing you've gotta do with MS software, the "component" software is often easier to get running right. So with both price and time, the quality stuff is cheaper than an all-in-one "system".
But with software, nobody much knows how to make things interoperate well.
I care if the OSS projects I use suddenly drop all US resident contributors.
Which reminds me of a question that I've never seen answered: Suppose I'm an American who has been given an account on a machine in country X. I ssh to that machine and develop some software. Did I develop it in the US or in X?
This isn't an entirely hypothetical question. Actually, I'm sometimes not aware of where in the world some of the machines are that I'm working on; I sometimes just have a hostname and/or IP address, and the location isn't relevant to my work, so I don't bother learning where it is.
Shame on us all, we let a bunch of money hungry subhumans rule the world.
This probably isn't very interesting to the majority of slashdot readers, who we'd expect to have the knowedge and sense to have long ago turned off javascript and all other scripting things in their browsers. Right? Right.....
Actually, men can not only get breast cancer, but they can also produce milk. Google for "witch's milk" for some descriptions of this in newborns. This happens in around 5% of newborns of both sexes, and normally stops within a week or two. But most of us no longer believe that it's caused by witches. It's now usually attributed to the mother's hormones that cross the placenta and affect the almost-born fetus.
Slashdot Mirror
Businesses fear Open Source like the plague because they're afraid of govenments "buying" software then declaring it "Open Source" they don't have to pay.
How the hell is that supposed to work? If you contract me to produce some software for you, and I use open source, you still have to pay me the agreed amount or see me in court.
Any company's lawyers will understand this. If they make such an argument, they are simply lying. Their real motive is that they don't want you to see some parts of the code. This could be because they're embarrassed by the shoddy quality. More often it's because there are things there in addition to what you think you paid for.
In the case of Diebold, they made this very clear before the 2004 election, when then-CEO Wally O'Dell said - in writing - to the Ohio Republicans that he would deliver their state to George Bush. He lived up to that promise, and there are good grounds to suspect that this wasn't at all accidental. They want their code secret so that we can't find out some of the things they've got hidden there.
In the case of elections, paranoia is simply rational. History tells us that the people running an election will cheat if given the slightest opportunity. Secret code makes cheating very easy, and the assumption should always be that secrecy like this is to hide what's going on.
The only practical way to get honest elections with computerized equipment is to require that all the code be open and visible to the public. Anything less is a guarantee of dishonest elections.
(Guaranteeing that the published code is actually what's running inside the machine is another issue. We need a way to do that, too.)
Postmodernism is what happens when the rate of intellectual change approaches C'. ... C' is something I just made up,
I think you've illustrated the primary problem here.
There is a rather old distinction in logic, philosophy and linguistics: the symbol/referent distinction. This is the difference between a symbol and the thing that it represents. There are a lot of logical errors and social problems that are based on confusion of these.
Much of the problem is the tendency to think that if something has been named, then it exists. This leads to the belief in things as great as our various gods, societies, corporations, and such, down to things as fleeting and inconsequential as postmodernism and Web 2.0.
Marketers do tend to understand this, and use the logical confusion to their advantage. This is the primary origin of the term "Web 2.0". It was invented as a marketing tool, to trick people into thinking that they had invented something new. Much of the critisism is pointing out that there's really nothing behind the term.
Commenting on Web 2.0 and postmodernism by blithely inventing a "C'", with a definition that isn't amenable to any sort of testing, is a good way of illustrating what's really going on here.
Sorta like fighting fire with fire, I guess.
[Bush] has encountered virtually no resistance or scrutiny from Congress, and has skillful deceptive tactictians who, in a very real, cynical, Machiavellian sense, have artfully deceived the entire world, America included, into turning over as much power as possible to them and their cronies.
I don't believe they have been deceptive at all. From well before Bush was elected governor of Texas, his behavior and policies have been quite thoroughly documented and the story has been available for anyone to read who has the slightest interest in the subject. And if you look at the public information before the invasion of Iraq, you'll find that all of the Bush gangs claims had been thoroughly and very publicly shown false. They were reduced to arguing that they had to invade because of what Saddam might do in the future.
The problem is that about half the American voting population approves of Bush's policies and actions. This isn't from ignorance. We've all been told repeatedly and in great detail of his gang's lies and corruption. Anyone voting for him should be assumed to know about it. Their support implies that they approve.
In particular, many people have explained in great detail why Bush's war in Iraq is highly illegal under both international and American law. He and most of his Cabinet could easily be impleached and thrown in jail for the rest of their lives. This isn't going to happen. The reason is that most of the Senate (including most of the Democrats) know and approve of it all.
... one is about lying about sexual relations with a woman and the other is about destroying a fundamental structure of our government.
Yeah. But the real problem is: Our current Republican leadership thinks that the former is a serious problem worth prosecuting. The latter is not just acceptable behavior; it's policy.
12% admitted that they often remain online longer than expected.
When you consider the effect of buggy browsers, combined with confusing, poorly-designed web sites, this is just the normal situation for all of us. It's not a symptom of any problem in the user at all.
Animal and plant breeders have a phrase "hybrid vigor". Google for it.
The theoretical explanation seems a bit weak so far. It's mostly based on the observation that dominant traits tend to be adaptive, while recessive traits tend to be maladaptive. This isn't always true, of course; it's just a statistical pattern. But it would explain why first-generation crosses between distantly-related strains tend to produce vigorous offspring.
"Further research is needed."
So in the future the dimwitted underclass will end up running society and subjugating the better-looking upperclass?
Actually, this has already happened numerous times in the past couple centuries. Look into the family histories of the rulers of the Ottoman, Russian and Chinese empires for instructive examples. And they're just the best-documented of all the cases known.
The US is now on the familiar path to developing its small class of hereditary idiots who occupy the country's top offices. Stay tuned and see how it turns out.
The British do seem to have figured it out, though. They've walled off their group of idiot rulers so they're merely entertaining, but no longer a danger to society, and nobody trying to replace them. The citizenry is probably a long way from thinking that their rulers would make better sausage.
I looked through the review for an idea of what the phrase "first class" might mean to the writers, but I didn't spot it. This was a way of getting at the real question: What sort of web site is being pushed here?
;-)
My suspicion, based on lots of other similar examples of advice on building web sites the Right Way, is that "first class" means flashy and entertaining. Maybe I'm wrong, but experience here says to expect the worst.
Not that I'm objecting to that. There's a demand for entertainment on the web. But there's also a demand for informative sites, and the design criteria for information and entertainment are very different.
Most of the web sites I've been involved with are of the informative type. For such sites, the canonical "first class" site is google.com, not (for example) yahoo.com. For a more complex site, news.google.com works pretty well.
But I didn't get a feel from the review or the author's web site what their criteria for "first class" might be. Unless I get a clear idea what sort of web site is being pushed, I probably won't be spending my money on any book of advice.
(I might also obsever that most entertainment sites are really in need of good advice on how to do the job right. But I probably don't need to tell that to anyone who visits them often.
..., it should require the "I agree". It's more of a CYA thing to do.
Well, maybe, but people do routinely point out that the GPL doesn't require that you agree with its provisions. If you don't agree, you simply don't follow the extra rights that it gives you. You just use the software like you would anything else, under the terms of your local copyright laws. You don't need anyone's permission to do that. And it doesn't matter whether or not you agree with the copyright laws; they are in effect whether you agree or not.
So how does any user clicking add any CYA aspect to the GPL?
The only motivation I can see here is satisfying the users' expectation that they'll have to click on an (unreadable legalese) agreement. But that doesn't mean that clicking has any legal import in this case. You're merely following a custom that many users now expect, even in a case where it has no legal import.
My firefox very, very rarely crashes (once every few *months* Java or Flash bring it down).
Well, sure, it crashes rarely. But I find that I have to kill it every other day, because its memory usage has grown to several times the mere gigabyte of memory that I have. Closing all its windows only releases a small amount of its claim on memory. So I kill it, all my other apps speed up, and a new FF runs quickly after I've reloaded the pages I was working on.
Saying that something "very rarely crashes" is a bit disingenuous, when what it does is drag the system (itself included) to a near halt. "But it didn't crash!" Yeah, right.
(This is on both a linux RH system and a Mac with OSX 10.4.whatever the current version is. YMMV, of course, depending on what you're using FF for.)
So when did Microsoft ever "compete" with any other company? That's not how they do business, y'know.
...
Consider that they got their start as a subcontractor for IBM, and used IBM's economic clout to enforce "agreements" with retail vendors that effectively locked out other startups. That was so successful that they've never had much of a motive to "compete" in any ordinary sense of the word. They don't compete; they make deals. When they have to, they engage in classical price wars, but that's a last resort. They used a price war to bankrupt Netscape, of course. But their usual approach is more like how they're now trying to lock out all those pesky vendors of Windows security software. They'll succeed at that, too, as they have with all sorts of other software.
That's how big business works in the Real World [TM], y'know. You don't profit by competing. You profit by ensuring that the customers don't know about and can't easily find the competitors' products.
The iPod is an abberation. MS screwed up, and failed to take the appropriate steps before the masses became aware of this new product. It's a bit late for them now, though they still have a chance of winning if they can make it sufficiently difficult for Windows users to use iPods. As a last resort, they can give away their own player "free", i.e., with the price included in a Vista system.
Let's see what they do
Well duh, of course it's caused by a programmer. Doesn't mean there's a good reason for it, ...
...
I was replying to the original comment:
But there's no reason
Note the lack of the word "good" in the original. I agree that there's no good reason; I was just pointing out that there's a reason. And yes, it's a fairly obvious reason, if perhaps not a good one.
Actually, if you look at the origin of the mozilla suite in the Mosaic browser, I'd claim that it was reasonable to implement it this way at first. Building Mosaic for the first time was a bit of a task, and it's no surprise that little issues like background threads eating up cpu time might have been ignored. Mosaic was, to a great extent, a "proof of concept", and it was spectacularly successful for its time.
But that was then. Now, the marketers are taking advantage of this to saturate our cpus for the purpose of distracting us with active images. This is growing into a serious problem. There would be a real advantage to a browser with the ability to block this waste of cpu.
Firefox has some tools for controling cpu use. But the tools are scattered, ad hoc, incomplete, and difficult to use. There's a gimmick to stop active images. There's a gimmick to turn javascript on and off. There's an extension to block flash. There's no way I know to block a tag. There's no consistency to any of this. Opera has some good ideas, but it's also incomplete. I wonder if there's a way to unify the problem into a single, consistent tool?
But there's no reason [active images] should be eating CPU when the window is minimized or when I'm looking at a different tab.
Actually, there's a very good reason: Some programmer implemented it that way.
That's the reason that most software does what it does. If there are several ways that something could be done, you can expect that all of them will be implemented in some software somewhere. It's basically at the programmer's whim.
In the case of active images, there's no sensible reason for keeping them running when they're not visible on the screen. But stopping them and restarting them takes explicit actions on the part of the software. This requires that the programmer(s) be aware of the issue and write the code to do it right. You'd expect that initial releases of image-displaying software wouldn't do this.
It's quite difficult for a user to verify that it's actually an active image that's eating the cpu. But it can be done. Use the context menu to load the image into a separate tab. Wait for cpu usage to stabilize, and note the browser's cpu usage. Close that tab, wait a bit, and note the browser's cpu usage. If it has decreased, you have some partial evidence. Re-create the tab with the active image, and see if the cpu usage goes back up. If so, you have better evidence. Close the tab, and see if the cpu usage declines. If so, it's really good evidence.
I've done this, and concluded that FF does indeed keep images "running" when they're not visible on the screen. This happens if I switch to another tab, and also if I minimize the window.
Even worse, it happens with all "active" elements. Javascript keeps running, as do any video plugins. The only way to stop them is to disable the feature globally (or per site with Opera).
There's lots of room for future improvements here.
Right now Firefox is using about 60% of my CPU, and it's minimized. (Before you ask, I also have Flashblock, so there shouldn't be any Flash apps running.)
Try turning off java and javascript. And set images to loop only once.
These steps won't get everything that uses cpu when "idle", but they all help. Advertisers like to use all these things to make their ads distractive. There are some other cpu-eating features that probably can't be turned off, such as the auto-refresh in a page's tags.
What would really help is a thorough study of all the ways that a page can continue to use cpu. We need a list of them, and an on/off flag for each one. If FF (or seamonkey) would give us something like this, it would be another good advertising point.
> Perhaps they can start talking about tienamen square maturely instead of pretending it didn't happen.
...
*snort* Good luck on that one. I still have yet to hear a mature discussion outside of university walls on the internment of Americans during WWII:
I have, but I'd agree that it's not a common topic of conversation. I even grew up in the Seattle area, with a lot of friends of Asian background, and I remember a number of calm discussions of the shameful event. Most of the people that I knew, both of Japanese and European background, just viewed it as "Yeah, isn't it horrible how people treat each other sometimes?"
Finding non-academic discussions of the topic isn't difficult any more. You can, of course, find a good summary at wikipedia. I also found that googling for "Japanese American internment" gets about 1.4 million hits, and a quick check of a few shows that reasoned, factual articles and discussions seem common.
Of course, the rabid bigot types won't be part of such discussions. And I do remember a few such people from my childhood. We mostly ignored them, while being aware that they were the sort of people who would do it again if given a chance. I didn't have a lot of friends like that, though.
We're seeing this right now, of course, with the American treatment of Arab-Americans and others of Middle-Eastern origin. It's not to the point of mass-internment camps, but those of us following the story are aware of what's going on. This includes the government's mistreatment of people, using an attack by unrelated other people as an excuse. And the mistreatment includes arrest and indefinite incarceration without charges or trial. In some cases, it includes torture of people later shown to be innocent of any suspected crimes.
Lots of us are quite aware of our country's history in such matters. Attempts to suppress knowledge of this history are not very successful, and any American who wants to learn the facts can now find them quite easily.
The Chinese leaders may have realized that the economic value to them of access to sites like wikipedia is greater than any danger that people might read politically unacceptable information. After all, the US doesn't collapse because its people have access to information about the more sordid parts of their government's history. Most Chinese citizens will remain as safely ignorant as most Americans are.
: On a somewhat related note, are new features even what people want in a browser?
;-). It's even a problem if you're following a dozen or so webcomics, since most of them have ads, and some of the ads suddenly take over your cpu.
What I'd like, and what would make me a frequent user of a browser that did it, is an "anti-feature": A simple-to-use tool that lets me turn off (and occasionally on) all of the so-called features that soak up cpu.
The problem is that I often have a number of browsers open with lots of windows and tabs. After a while, I find that the browsers are using most of my cpu. If I want to do something that uses more than a small amount of cpu, I have to shut down all my browsers.
With various browsers, I've discovered how to disable most of these cpu-eating features. I have flashblock installed in the browsers that will accept it. I turn off java and javascript. I set images to run only once (in the browsers that let me do this). And so on. But it doesn't solve the problem. All the browsers, including firefox, eventually start gobbling cpu, and I have to kill them.
What I want is a window that lists all the "features" that can use cpu time. The window should let me enable and disable each such feature separately. Even better would be something like flashblock, that lets me enable one such feature just once in a single page. Next best is what Opera does: It lets me enable a few such features on a per-site basis, and blocks the rest. (But it doesn't do this for everything, and I have to kill Opera at times when it suddely goes to 80% of the cpu.)
This is a major problem if you do a lot of web testing, and need to run a lot of browsers. It's also a problem if you're involved in things that require following a number of blogs or other such online discussions (like slashdot
I also find that I have to shut down most of my browsers before I burn a CD. A browser suddenly grabbing 90% of the cpu for a second or so can really kill a CD write operation.
If the FF folks would give us a good tool for this, it could be a very popular new feature. And it would teach the advertisers a much-needed lesson.
Heh. Here I was all prepared to to point out that the obvious answer is "both", but I noticed that every single reply so far has said the same thing. So rather than be rated redundant, I'll just point out that there is a standard name for this particular bit of illogic.
If authors had any sense, they'd be jumping onto the anti-copyright bandwagon. Such laws were, we're told repeatedly, created to give authors and artists control over their creations, and to guarantee them income from sales. More and more, the actual effect is to take away the creators' control, giving control and profits primarily to their corporate masters.
This story just illustrates that some authors have figured out that it helps to let their readers know what's available. And the copyright question basically asks whether a publisher has the right to block communication between an author and the audience.
Maybe we do need some sort of copyright laws. But authors don't need the current copyright laws. That's what's keeping most of them poor.
There is the conjecture that pi is a "normal" number, in which case Sagan's circular feature is there regardless of the base that you use. It's there an infinite number of times, in fact.
But to my knowledge, pi hasn't yet been proved "normal". It just appears so from statistics on the first trillion or so digits in base 10, which is hardly a proof.
To provide a book that passes muster with SW developers would be a blunder.
Depends. You're right, if the only goal is to maximise the sales of the book. If you want to improve the quality of software, you're wrong, though. This book just got a "don't bother" review in a tech forum. So we computer geeks won't read it, and we won't know what (if any) useful suggestions he might have for us.
Of course, you could take this as an example of the problem that he's describing, in the publishing industry. Like most commercial software, the book was "managed" by people whose main (and perhaps only) motive is to maximise sales. Such managers have little interest in what happens after the sale, except insofar as it might affect subsequent sales. In particular, neither software nor publishing managers are concerned with writing software or books that appeal to their fields' "geeks". They want the product out the door by a fixed date, and they want whatever gimmicks will increase sales to the masses.
The result isn't all that difficult to understand. Especially if you've been involved in software and/or publishing, and seen firsthand the pressures that management puts on the authors.
Of course, here and there you do find good managers, and sometimes they can help a quality product sneak through the process.
Look, increasing the costs of your bandwidth isn't stable proposition in a free market.
So where in the world is there a free market in bandwidth? I've never heard of such a thing.
Someone will come along and be cheaper.
Not if they can't get a license to operate, they won't.
I wish the article would mention what the dark cloud implies. Do scientists think it's a storm? Does it suggest water or life?
Well, if you look at the blow-up of the storm, you can see that it's maybe about 30 pixels. Not a whole lot of information to go on.
But we can expect a few other telescopes will be aimed at it. We might hear a few more bits of information in a few months.
I don't think it's a bad thing that Microsoft has made it impractical to charge for a web browser. How is it a bad thing if they make it impractical to charge for anti-virus software?
Good point. And we might generalize it a bit. We often read here that old canard "You get what you pay for". With software, not only is this not generally true; what's more common is that with software, price and quality are typically inversely related.
Microsoft is merely doing its part to maintain this situation. They do it in a somewhat subtle way: They pretend that much of it is free, but you do in fact pay for IE and for MS's anti-virus software, as part of the price for their entire "system". You get crappy, poorly-functioning software, of course, in agreement with the price-quality rule. If you want quality, you have to download and install either shareware or free software.
Actually, there is somewhat of a parallel for this outside of computers. It's well known that, if you want quality audio or video equipment, you don't buy the all-in-one "systems". Those are simple purchases, and the components do work together (and are typically integrated into one box so that they appear to be a single product). But to get quality, you have to buy individual components, and interconnect them yourself. This takes time for study and wiring, but the end result will be much better quality.
Microsoft systems are like this. They sell as a "system", but the overall quality is low, especially since the components generally don't inter-operate nearly as well as advertised. Like A/V equipment, if you want quality, you'll just have to spend the time to install the quality components yourself.
The difference is that, with quality A/V equipment, the good stuff usually costs more than the crappy "integrated system" box. With software, the good stuff is usually a lot cheaper than the integrated junk. And when you look at all the hair-pulling and time-wasting futzing you've gotta do with MS software, the "component" software is often easier to get running right. So with both price and time, the quality stuff is cheaper than an all-in-one "system".
But with software, nobody much knows how to make things interoperate well.
I care if the OSS projects I use suddenly drop all US resident contributors.
Which reminds me of a question that I've never seen answered: Suppose I'm an American who has been given an account on a machine in country X. I ssh to that machine and develop some software. Did I develop it in the US or in X?
This isn't an entirely hypothetical question. Actually, I'm sometimes not aware of where in the world some of the machines are that I'm working on; I sometimes just have a hostname and/or IP address, and the location isn't relevant to my work, so I don't bother learning where it is.
Shame on us all, we let a bunch of money hungry subhumans rule the world.
When was it any different?
This probably isn't very interesting to the majority of slashdot readers, who we'd expect to have the knowedge and sense to have long ago turned off javascript and all other scripting things in their browsers. Right? Right .....
Actually, men can not only get breast cancer, but they can also produce milk. Google for "witch's milk" for some descriptions of this in newborns. This happens in around 5% of newborns of both sexes, and normally stops within a week or two. But most of us no longer believe that it's caused by witches. It's now usually attributed to the mother's hormones that cross the placenta and affect the almost-born fetus.