After finding it in a Slashdot comment, I really want to play with the Colors! painting program. What's the quickest way to get that up and running, assuming that I know absolutely nothing about DS homebrew?
Those have significant direct and indirect benefits (lowered cost and less environmental impact primarily). I guess if you're a security zealot you wouldn't even consider the tradeoff, but for those of us in the real world it's probably not a tough sell.
I think his point was to make people aware of the tradeoff. He's telling them that those reasons are perfectly valid, but you should know what new risks you're exposing your systems to so you can make that decision with full knowledge. I do know people who equate VMWare with security, and that's just not true. He wants everyone to understand that.
Firstly, Vista isn't painful. I've tried it, I use it, it's fine.
Oh yeah? Whew, I guess that settles it then.
On the other hand, we bought a new laptop for my wife. I thought we'd see what the pre-installed Vista was like before upgrading it to XP, and I admit that it wasn't awful, except that it insists on disconnecting from our AP at least once every five minutes. I wish it had a setting for "stay connected to the damn AP until I tell you otherwise, OK?".
Again, this is the pre-installed version, not one that I self-installed and potentially messed up on my own. It has at least one repeatable and verifiable issue right out of the box. It's not a major one, granted, but doesn't exactly inspire confidence.
Although "computers" is on the list, "computer software" is not and other subcategories are sufficiently narrow that I think it would be if it were significant (eg "motor vehicle parts" and "automobiles" are counted seperately). Boeing is in there via "aircraft", but Hollywood doesn't seem to be listed unless you count "organic chemicals" as meaning Britney's drug-fueled butt.
Even if MS's entire $44 billion revenue in 2006 was from outside sales, they'd still only make up about 4% of total exports. I think that MS and Hollywood are pretty high on the list of visibility, but in raw numbers they don't seem to be all that important to the economy, at least not to the point that we should bend the rest of it to their will.
IBM big iron is designed for those people who cannot use anything else. They have no competition. It doesn't matter if you prefer mysql or whatever - it cannot handle those kinds of loads, because it can't scale up to clusters of hundreds of thousands of CPUs.
Very true. I remember when Google used to use MySQL on Linux to power their searches but gave up in despair as the whole thing collapsed around them.
Also, if you want to break you stuff up into a lot of different classes, with each class in a different file, then it quickly becomes a large mess of stuff to include in each page. I find that this is one of the major advantages of having a compiled language.
Hey, now, as much as I dislike PHP I don't see that "include_once 'foo';" is that much more complicated than "#include ". Also, the "compiled language" concept is a wholly different animal. Having to run a compiler on a piece of code doesn't automatically make it more manageable. Also, I don't know of many purely interpreted languages commonly used anymore. Python compiles to bytecode upon execution (and caches that bytecode for quick startup the next time), as does PHP if you're using any of the free or commercial accelerators.
I detest PHP, but not for any of the reasons you mentioned.
Okay, that's all well and good, but let's look at the one metric that well determine if your way is the correct way. Does your way result in things breaking?
When it comes down to it, MS agreed with you and used that as their metric.
And here we are discussing how disastrous that metric has been for them long-term, now that they're locked into its inevitable conclusions.
Talk about a false dichotomy! Do you really think the two are at all related?
Definitely. The old OS model allowed certain shortcuts such as hacks that directly patched the code segments of other programs that were running to change their behavior. The new protected memory model flat-out makes that hackery impossible, so it was up to programs to add explicit support for message passing and other external control systems. There isn't a message passing system in the world that's as fast as just overwriting a destination application's buffers with new data.
That's just one example of why some things are inherently slower if done right. Sometimes it's just not avoidable. That doesn't mean that the new way is inefficient or bad, just different.
I was never into Macs back in the day so I can't comment on old vs. new Finder or spring loaded folders, etc., but I find it telling that the only people who seem to seriously dislike the new Finder are the ones who seriously loved the old one. To everyone else it's pretty spiffy and a reasonably good model of how such things are supposed to work. That is, I'm not at all convinced that the old Finder was actually superior; it's just that people liked it that way, darnit, and anything different is inferior by definition.
None of that has anything to do with multitasking or event loop handling and you know it.
You're right: it doesn't. I'm not sure why you even brought it up.
I know in my work as a software engineer that I would literally by thrown out on my head if I went through our code and started modifying existing methods so that they had "better output" without fixing everything single method that referenced it to deal with this new "better output."
Conversely, I know in my work as software engineer that I would literally by thrown out on my head if I expected a given method to always behave in a certain ways even though that way was never documented anywhere and purely a side effect of that specific implementation. They would then proceed to kick me in the butt if I dared complain that someone reimplemented it in a better way that was 100% backward compatible with the published specs for that method.
Seriously, if newFoo() works exactly like foo()'s documented behavior from previous releases, then it is foo() and should be treated as such. I can't blame them for not wanting fooLikeDos5(), fooLikeWin30(), fooLikeWin311(), fooLikeWin98(), and fooLikeWinXP() all cluttering up the API.
Now, changing foo() so that it no longer behaved like the old documentation would be unforgivable, but I haven't heard anyone say that's what they did. I'm far from a MS apologist but I can't really hold this against them.
I can completely recognize that 10.0 was pretty rough when moving from OS 9.
Old Macs had a flaw (yes, I said it) where holding down the mouse button would freeze the rest of the computer.
Including the network stack.
We noticed this because when the rest of the office would play MP3s from our graphics guy's Mac's shared folder, everyone's audio would randomly and simultaneously drop out. We eventually realized that it happened when he was holding Photoshop's menus open for a long time while he pondered which filter to apply to some image.
People who found 10.0 to be rough were the ones who convinced themselves that pre-10 was the paragon of computing and The Way Things Were Meant To Be. To everyone else, OS 9 and earlier were just horrible.
I think the answer about what they should do is create a new method called newFoo() and deprecate the old method so that everything that uses the old method doesn't break.
Why should they? If foo() and newFoo() have the same defined API but newFoo offers additional, completely optional functionality, I don't see a compelling reason for maintaining both of them. I suppose you could define foo() like
int foo() { /* Legacy support for incompetent programmers */ return newFoo() ? 1 : 0; }
but I don't think it's incumbent upon MS to atone for everyone's sins. They have enough of their own to worry about. The problem is that they adopted the policy of supporting all that brokenness long ago and now people demand it. What I suggested for.NET is about the only way I can think of to break that cycle.
I agree, you can't blame Adobe for not wanting to fix a 5 year old application which is 3 versions out of date. You can blame Microsoft though for breaking it. If it worked it should continue to work.
I find myself doing the unusual position of supporting MS here. A lot of applications did things The Wrong Way or used undefined, undocumented behaviors that they should never have relied upon.
An example for coders: imagine a system function named "foo" that returns 0 on success or nonzero on failure. The XP implementation happened to return 1 as its specific, unchanging value of nonzero, although that was never documented anywhere. It just did. In Vista, foo is modified so that it still returns 0 on success, or one of many defined constant values specifying exactly which error occurred. Finally, imagine that lazy programmers who should've been writing
if !foo() { handle error }
were instead writing
if foo() == 1 { handle error }
because those two have been functionally identical for a few years.
In that all-too-common scenario, what is MS supposed to do? Their main options are:
Revert to the old behavior, preventing all new software from benefitting from that functionality.
Keep the new, better behavior but endure endless whining from people using fundamentally broken software.
Change the program loader to wrap each executable and library with checks like
if progname == 'adobeBrokenv1.3' { use old api }
I don't envy them the hole they dug for themselves. They would have been far better off if long ago they'd made it clear that their published API was a contract. If you follow it to the letter then your programs would continue to work. If you break with it, all bets are off. Everyone else does this. If you link against GNU libc and your software suddenly crashing, its maintainers would look at what you're doing and either fix libc or tell you that you'd screwed up and to fix your software to follow the published docs. Instead, MS once again used a greedy algorithm to optimize in the short term for developers! developers! developers! no matter how badly that screwed up their underpinnings.
Honestly, I think.NET is perhaps their last chance to get this right. I think they should take a hardline and change constant values and randomize undefined return values and otherwise deliberate tweak things so that API-compliant software would still work fine but everything else would crash horribly every other month or so. It'd be a painful transition for people used to the idea of MS doing the work of fixing vendor software for the rest of eternity, but maybe they could finally get rid of the backward bug compatibility albatross around their neck.
But alas, yeah, it's the truth. Here I sit at work pecking away next to a boring black PC case hooked to a boring black monitor with boring black speakers. At least I have my beige Model M keyboard to break up the monotony (which is something you probably don't hear often (although you definitely hear the keyboard itself)).
I'd take a picture of me holding a "HI SQUIGGLESLASH" Sharpie sign, but I haven't gotten Bluetooth working yet. Actually, that fact alone should give me some Linux cred.
I'm just wondering if you would be coming to the defense of Microsoft had Greenpeace leveled the same allegations against them
Yep. There are plenty of reasons to dislike any given large company, but Greenpeace hasn't raised any legitimate ones.
So the problem apparently is that they're goring your sacred cow
I'm typing this from a Dell running Gutsy Gibbon. I don't even own an iPod. Sorry, but it's probably a bit harder to write me off as a Mac fanboy than you seem to wish.
Baiting aside, I do think that the criticism may be somewhat more relevant when leveled at Apple than at competitors, because Apple has always cultivated that valley, consumer-friendly persona.
Relevant, perhaps, but still a lie. In this one specific case, Apple's already ahead of almost all of their competitors and they're still improving. It's like Greenpeace decided to pick on Prius drivers for less-than-perfect fuel economy while ignoring SUVs.
Put yourself in MSFT's shoes and imagine what you would do. A security issue crops up. One team comes back with a solution that does not break all the competitors products. The other team comes up with a solution that incidentally breaks competitors products. Which one will you pick as "critical security update"?
I'd pick the one that doesn't further drive IT employees worldwide to despise my product because every "critical security update" breaks things that were working yesterday. MS seems to have missed the distinction between optimization and greedy algorithms.
Please provide a method on how to backup my TV, my couch, my oven, etc, so when I'm robbed I'm good to go.
Good point. I am now ripping the tape drive out of my server because if I can't make backups of my kitchen table then I shouldn't be allowed to make backups of my home directory.
Does your logic truly make sense to you, or are you just being an ass for the sake of it?
But apple actually tries to make their products Green after the first busting from greenpeace
Correction: Apple was "green" before the first tantrum. Their sole reaction to Greenpeace's childish outburst was to publicize exactly how green they were even before GP got involved.
This kind of thing can ruin a quiet little family company like Apple Computer.
Sarcasm aside, it can indeed hurt their stock price, causing problems for its millions of owners (through mutual funds in 401Ks, etc). And just because they're not a small ma-and-pop store, why do they deserve to be libeled by those cretins? Apple isn't being punished for wrongdoing. Apple's being punished for being well-liked. That's not fair or justifiable in any society I'd care to be a part of.
Slashdot Mirror
After finding it in a Slashdot comment, I really want to play with the Colors! painting program. What's the quickest way to get that up and running, assuming that I know absolutely nothing about DS homebrew?
I think his point was to make people aware of the tradeoff. He's telling them that those reasons are perfectly valid, but you should know what new risks you're exposing your systems to so you can make that decision with full knowledge. I do know people who equate VMWare with security, and that's just not true. He wants everyone to understand that.
Your 8GB hard drive will miss you.
See also: KMFDM's "Witness". It isn't as lyrically strong, but the singer sounds eerily like they really mean it.
Oh yeah? Whew, I guess that settles it then.
On the other hand, we bought a new laptop for my wife. I thought we'd see what the pre-installed Vista was like before upgrading it to XP, and I admit that it wasn't awful, except that it insists on disconnecting from our AP at least once every five minutes. I wish it had a setting for "stay connected to the damn AP until I tell you otherwise, OK?".
Again, this is the pre-installed version, not one that I self-installed and potentially messed up on my own. It has at least one repeatable and verifiable issue right out of the box. It's not a major one, granted, but doesn't exactly inspire confidence.
Solid proof that nothing makes a joke funnier than explaining it so that others can understand the punchline.
According to the CIA World Factbook, the biggest exports are:
Although "computers" is on the list, "computer software" is not and other subcategories are sufficiently narrow that I think it would be if it were significant (eg "motor vehicle parts" and "automobiles" are counted seperately). Boeing is in there via "aircraft", but Hollywood doesn't seem to be listed unless you count "organic chemicals" as meaning Britney's drug-fueled butt.
Even if MS's entire $44 billion revenue in 2006 was from outside sales, they'd still only make up about 4% of total exports. I think that MS and Hollywood are pretty high on the list of visibility, but in raw numbers they don't seem to be all that important to the economy, at least not to the point that we should bend the rest of it to their will.
I suspect you're right. The biggest question is whether it will be as an example or as a warning.
Very true. I remember when Google used to use MySQL on Linux to power their searches but gave up in despair as the whole thing collapsed around them.
Pretty much all Mac programs have that "problem" since its window manager uses click-to-focus. It's not a bug; it's a feature.
Hey, now, as much as I dislike PHP I don't see that "include_once 'foo';" is that much more complicated than "#include ". Also, the "compiled language" concept is a wholly different animal. Having to run a compiler on a piece of code doesn't automatically make it more manageable. Also, I don't know of many purely interpreted languages commonly used anymore. Python compiles to bytecode upon execution (and caches that bytecode for quick startup the next time), as does PHP if you're using any of the free or commercial accelerators.
I detest PHP, but not for any of the reasons you mentioned.
When it comes down to it, MS agreed with you and used that as their metric.
And here we are discussing how disastrous that metric has been for them long-term, now that they're locked into its inevitable conclusions.
Definitely. The old OS model allowed certain shortcuts such as hacks that directly patched the code segments of other programs that were running to change their behavior. The new protected memory model flat-out makes that hackery impossible, so it was up to programs to add explicit support for message passing and other external control systems. There isn't a message passing system in the world that's as fast as just overwriting a destination application's buffers with new data.
That's just one example of why some things are inherently slower if done right. Sometimes it's just not avoidable. That doesn't mean that the new way is inefficient or bad, just different.
I was never into Macs back in the day so I can't comment on old vs. new Finder or spring loaded folders, etc., but I find it telling that the only people who seem to seriously dislike the new Finder are the ones who seriously loved the old one. To everyone else it's pretty spiffy and a reasonably good model of how such things are supposed to work. That is, I'm not at all convinced that the old Finder was actually superior; it's just that people liked it that way, darnit, and anything different is inferior by definition.
None of that has anything to do with multitasking or event loop handling and you know it.You're right: it doesn't. I'm not sure why you even brought it up.
Conversely, I know in my work as software engineer that I would literally by thrown out on my head if I expected a given method to always behave in a certain ways even though that way was never documented anywhere and purely a side effect of that specific implementation. They would then proceed to kick me in the butt if I dared complain that someone reimplemented it in a better way that was 100% backward compatible with the published specs for that method.
Seriously, if newFoo() works exactly like foo()'s documented behavior from previous releases, then it is foo() and should be treated as such. I can't blame them for not wanting fooLikeDos5(), fooLikeWin30(), fooLikeWin311(), fooLikeWin98(), and fooLikeWinXP() all cluttering up the API.
Now, changing foo() so that it no longer behaved like the old documentation would be unforgivable, but I haven't heard anyone say that's what they did. I'm far from a MS apologist but I can't really hold this against them.
Old Macs had a flaw (yes, I said it) where holding down the mouse button would freeze the rest of the computer.
Including the network stack.
We noticed this because when the rest of the office would play MP3s from our graphics guy's Mac's shared folder, everyone's audio would randomly and simultaneously drop out. We eventually realized that it happened when he was holding Photoshop's menus open for a long time while he pondered which filter to apply to some image.
People who found 10.0 to be rough were the ones who convinced themselves that pre-10 was the paragon of computing and The Way Things Were Meant To Be. To everyone else, OS 9 and earlier were just horrible.
Why should they? If foo() and newFoo() have the same defined API but newFoo offers additional, completely optional functionality, I don't see a compelling reason for maintaining both of them. I suppose you could define foo() like
but I don't think it's incumbent upon MS to atone for everyone's sins. They have enough of their own to worry about. The problem is that they adopted the policy of supporting all that brokenness long ago and now people demand it. What I suggested for .NET is about the only way I can think of to break that cycle.
I find myself doing the unusual position of supporting MS here. A lot of applications did things The Wrong Way or used undefined, undocumented behaviors that they should never have relied upon.
An example for coders: imagine a system function named "foo" that returns 0 on success or nonzero on failure. The XP implementation happened to return 1 as its specific, unchanging value of nonzero, although that was never documented anywhere. It just did. In Vista, foo is modified so that it still returns 0 on success, or one of many defined constant values specifying exactly which error occurred. Finally, imagine that lazy programmers who should've been writing
were instead writing
because those two have been functionally identical for a few years.
In that all-too-common scenario, what is MS supposed to do? Their main options are:
I don't envy them the hole they dug for themselves. They would have been far better off if long ago they'd made it clear that their published API was a contract. If you follow it to the letter then your programs would continue to work. If you break with it, all bets are off. Everyone else does this. If you link against GNU libc and your software suddenly crashing, its maintainers would look at what you're doing and either fix libc or tell you that you'd screwed up and to fix your software to follow the published docs. Instead, MS once again used a greedy algorithm to optimize in the short term for developers! developers! developers! no matter how badly that screwed up their underpinnings.
Honestly, I think .NET is perhaps their last chance to get this right. I think they should take a hardline and change constant values and randomize undefined return values and otherwise deliberate tweak things so that API-compliant software would still work fine but everything else would crash horribly every other month or so. It'd be a painful transition for people used to the idea of MS doing the work of fixing vendor software for the rest of eternity, but maybe they could finally get rid of the backward bug compatibility albatross around their neck.
Thanks for making my morning. :-)
But alas, yeah, it's the truth. Here I sit at work pecking away next to a boring black PC case hooked to a boring black monitor with boring black speakers. At least I have my beige Model M keyboard to break up the monotony (which is something you probably don't hear often (although you definitely hear the keyboard itself)).
I'd take a picture of me holding a "HI SQUIGGLESLASH" Sharpie sign, but I haven't gotten Bluetooth working yet. Actually, that fact alone should give me some Linux cred.
"So, I paid for [a single working copy of] XP, and I [got a single working copy of] XP. I'm happy."
Fixed your clarification for you.
I doubt there's any court in the world that would convict you of using a product after paying for it.
Yep. There are plenty of reasons to dislike any given large company, but Greenpeace hasn't raised any legitimate ones.
So the problem apparently is that they're goring your sacred cowI'm typing this from a Dell running Gutsy Gibbon. I don't even own an iPod. Sorry, but it's probably a bit harder to write me off as a Mac fanboy than you seem to wish.
Baiting aside, I do think that the criticism may be somewhat more relevant when leveled at Apple than at competitors, because Apple has always cultivated that valley, consumer-friendly persona.Relevant, perhaps, but still a lie. In this one specific case, Apple's already ahead of almost all of their competitors and they're still improving. It's like Greenpeace decided to pick on Prius drivers for less-than-perfect fuel economy while ignoring SUVs.
I'd pick the one that doesn't further drive IT employees worldwide to despise my product because every "critical security update" breaks things that were working yesterday. MS seems to have missed the distinction between optimization and greedy algorithms.
Good point. I am now ripping the tape drive out of my server because if I can't make backups of my kitchen table then I shouldn't be allowed to make backups of my home directory.
Does your logic truly make sense to you, or are you just being an ass for the sake of it?
Said groups can encourage this by doing so themselves.
Correction: Apple was "green" before the first tantrum. Their sole reaction to Greenpeace's childish outburst was to publicize exactly how green they were even before GP got involved.
Sarcasm aside, it can indeed hurt their stock price, causing problems for its millions of owners (through mutual funds in 401Ks, etc). And just because they're not a small ma-and-pop store, why do they deserve to be libeled by those cretins? Apple isn't being punished for wrongdoing. Apple's being punished for being well-liked. That's not fair or justifiable in any society I'd care to be a part of.