Which is fine when the qualities discussed are precisely freedom-related, but when people 180 their views of objective technical qualities depending on who the adversary is, that is just plain hypocritical.
Link to one single comment by someone who has done that.
In fact, link to one single comment by someone who claims mp3s are technically superior to wma. The only reason we like them better is that they're somewhat more open. WMA does sound a lot better.
Fortunately, you really don't have to choose -- you could just use ogg or flac, both of which are likely better quality than mp3, and both of which are unencumbered.
Given that they cost (new) around five times more than the same thing digitally, or than what I think it's worth, that's a small comfort.
I don't need to back them up, because the disks *are* the backup.
Please don't ever say that again. I know what you meant -- at least, I really hope I do -- but let's avoid this situation.
Why buy it as a download, when my very first act after downloading it would be to burn it onto a CD as a backup?
Because even if the download is flac, you can still probably fit about one and a half album per backup CD, and you absolutely can fit at least 10 or so to a single-layer DVD. If the download is mp3, that's more like ten albums per CD (and 150 or more to a DVD).
Also, because it's likely cheaper, and much easier to buy it directly from the band, or close enough -- Internet labels like Magnatune take only some 50%, which sounds like a lot until you consider that it's going to cost over 90% with the big CD labels.
1) Free, open source, no royalties. Costs you nothing to allow it, even if you already have mp3.
2) Because you can re-encode to anything else.
3) Because saying "digital music" is just as easy as saying "mp3 file", and is actually far more likely to be understood.
I'm not saying there shouldn't be mp3 files also. But I am saying that everywhere there's mp3, there should also be flac derived from the original source, if it's available. (Assuming the "original source" isn't mp3.)
That's the way I most often see it done. mp3 for people who want it to "just work" without having to think about it, and flac so I can re-encode it to a better format if my player supports it.
You have to pay your engineers to research and implement support. In fact, there may be a validated and certified MP3 implementation already available for your hardware, but not an OGG implementation.
There's already an open source reference implementation. I'm not sure, but it might be public domain -- in any case, it's been ported enough places that I imagine it's fairly portable C by now.
You have to pay your lawyers to verify that it's patent- and royalty-free.
Or you can just tell your bosses that it is, says so right there on the site, Slashdot agrees, etc, and they can choose to ignore the lawyers.
Seriously, that's about as dumb as having to pay your lawyers to find out if you need to pay for Linux licenses on your servers.
You may need to increase the amount of processing power or memory to handle the additional codec.
That's the legitimate one.
You'll need to perform additional testing to make sure OGG files actually work.
If you don't already have an automated testing framework, you really should get one. It'll save you money in the long run.
You'll need to account for additional support costs if the OGG support is broken but the MP3 support isn't.
Also legitimate -- but I thought that was the point of the above testing?
You have to have your marketing department do extra research to determine if the additional sales of your media player because of OGG support cover the additional costs.
Well, if all of the above become cheap enough, you could support pretty much every format out there. (Or you could, you know, ship your player with RockBox, but I guess that's too obvious.)
Cheap enough, and you can simply tell your marketing department "We're doing it this way" -- just like with the legal department.
I'll agree that it's a poor design to require so much memory to boot, but give it some swap and you're good. I've found that on very old hardware, once it's properly booted and a good chunk of things swapped out, it performs decently enough.
Credit where credit is due, the BSD license does not require Apple to release their changes at all, so doing so on a delayed timetable is still a lot better than most vendors.
I didn't say they were violating the license, just that it annoys me how they choose to contribute.
I'd rather have them contribute than not contribute, though.
I doubt it. Why do you suppose Apple publishes the Darwin source or any of the other low-level technologies they code (ZeroConf, LaunchD, etc.). They publish them because they actually see the business case for sharing the work with other companies and gaining wider adoption of said technologies.
And why do you suppose they started with a kernel they could get under a BSD license?
If Apple were to fork the latest WineHQ Wine, thus requiring them to give code back (LGPL'd and all), I'd argue that it's not because they're so willing to share the code, but because it would take them much more work to bring any of the other forks up to speed. I don't necessarily believe they'd avoid releasing at all, but rather, that they'd stall and delay and maybe give something back every year or two.
Is there and 'add new hardware' control panel. I recently bought a TV card that was supposed to 'work out of the box' on linux. But as usual I had to fuck around with config files telling kernel modules to load.
Well, to be entirely fair, "add new hardware" is fairly irrelevant most of the time.
Stuff that works out of the box tends to be autodetected, even if it does require those config file tweaks. Stuff that doesn't or can't work out of the box wouldn't fit very well with the standard Windows "add new hardware" dialog -- it's not as if there's going to be a driver CD -- although I believe there's a similar option for adding in support for various disabled-by-default binary kernel modules (nVidia, for example).
In KDE -- in ANY KDE app -- I can type KIO URLs. What this means is that Unix paths are a subset of what I can open and save to.
Other examples: zip:/home/sanity/foo.zip/bar.png -- throw it in Gwenview and, rather than unzipping the gallery somewhere, I just view as-is. Or kate zip:/home/sanity/foo.odt/content.xml -- apparently, it's read-only, but still pretty damned cool. Konqueror and Dolphin tend to discover these kinds of things on their own -- clicking a zip doesn't open file-roller or ark or anything similar in dolphin, it just opens it as a folder, transparently switching to the "zip" interface.
Or the ssh access. fish://user@host/ and it's as if I've mounted their filesystem. I mean, sure, I use ssh itself most of the time, but there's something almost beautifully blasphemous about using GUI editors and tools on a remote machine, especially when they don't have to know they're doing it. (Obviously Kate works, but so does KWord, Krita... like I said, any KDE app. Imagine that -- KWord fish://root@webserver/etc/apache2/httpd.conf)
And that's just what I've sort of picked up casually, with things like tip of the day.
I don't mean to disrespect Rox, nor am I completely defending the idea of software bloat, but we do actually use some of that "crap in the background" -- and once you get past the boot time, KDE 3, even, is pretty usable on really pathetic hardware. (Also, I may have to use Rox a bit to know if it can do what I want -- I tend to use the commandline for 90% of my file management, and Dolphin/Konq for 10% (including just showing off how freakin cool it is.)
You mentioned scripting? Looks like KDE 4 is planning that, if there isn't EcmaScript (JavaScript) all over the place already.
I forget whose sig it was, but it fits beautifully: "We do what we must because we can."
there is nothing wrong with content producers wanting to control how their content, in particular, the stuff they paid to generate, from being indexed.
True, nothing wrong with wanting. But as Jayne says, "If wishes were horses, we'd all be eat'n steak!"
(I know that's not the original quote, but I like Jayne's version.)
Gee, just a thought, but what about a way to display a summary and an ad chosen by the content producer along with the summary? Advertisers would spend lots for that kind of exposure.
Yes, they would. And if they got it, they'd find out very quickly how stupendously bad an idea it is.
You see, the reason Google's on top right now is that they were the simple, fast search engine. When it was between Google and Yahoo, Yahoo's homepage was almost as big as Slashdot's, with a tiny little "search" box somewhere, maybe, and Google was just better at getting results to you quickly anyway.
In other words... As much of a monopoly as they have, if Google ever does implement something like that, I'm switching to another search. Windows Live search, even. I imagine enough people would do that, or block them, that the "value" of those ads would plummet.
Ok, regardless of the legal fairness, I'd think removing those previews would actually reduce the likelihood of me visiting such a site.
Almost never do I see a Google result and say "Ok, I know all I needed to, not going to click." More often, I see one and say "Gee, looks like that site won't be very helpful, let's move on to the next one." I can only imagine my response would be like that, only more so, towards anyone who could allow Google to index them without allowing Google to summarize them.
It's just Adobe has a large, old codebase that has been transitioned across so many different architectures (68000 series, Power PC series, and now x86 32/64) that re-engineering it will be difficult.
I'm confused...
Were this Microsoft, I could understand, given that chunks of DOS were carried around for so long, and Win16 after that, but then, even modern kick-ass 64-bit dual-core processors can still run 16-bit code -- probably 8-bit code, if there is such a thing.
But I'm confused... Photoshop has run on Windows for a long time, which means x86. It also ran on 68k and PowerPC. And you're telling me it's an old, inflexible codebase? What they've done with it suggests that it would have to be extremely portable... guess not.
But then, I keep forgetting... Flash runs on everything, Windows, Linux, Mac, x86, PPC, ARM even, but not 64-bit. Either they don't care, or they've got a fairly huge engineering problem on their hands.
Disclaimer: I didn't even read grandparent, and your post seems reasonable enough.
What I find frustrating about Apple is their need to so tightly control every bit of code they borrow. Look how long it's taken for Webkit to go back into Konqueror... and don't even get me started on BSD/Darwin, whose policy seems to be "Open whenever we feel like it."
Thus, I suspect that, were Apple to include Wine, they'd fork it, improve it quite a lot (though largely in ways that can't easily be integrated back into Wine), assuming they didn't just fork Crossover, Cedega, or the newest version of Wine that's not LGPL'd. I don't know who to blame for this situation, actually -- it seems like Apple is not playing nice with others, yet with all the code there (well, most of the time), it seems like the projects which got forked could be re-integrating a lot of Apple improvements a lot faster.
Then you will be happy to know that for instance ODF still supports binary blobs in it's XML. Mayby OASIS isn't listing that as a main feature though...
Sometime I'll look it up, but I'm fairly sure I've never seen an ODF file with binary anywhere but as a separate file in the zip, and itself in a standard format.
So to extend ODF you would either have to add all of OMML to ODF or to change the w3c MathML standard.
Or say something like "It's MathML, plus this." This is done in the HD-DVD spec, in at least a few places where it includes standards like the DOM, but with fairly major changes/extensions.
And that is just one example. It would require more extending than is originally in the ODF spec to accomodate the needs of MS Office.
That is what I meant, actually. Maybe I shouldn't have used the word "extending"?
Regardless, I'd rather have a more limited format that's actually feasible to implement than "LineSpacingLikeWord95" bullshit.
Ok, suppose I release my work under a license which requires the following notice to be kept: "Permission is hereby granted, free of charge, to anyone who comes into possession of this software to modify, redistribute, and deal in the software without restriction provided that this notice remains intact along with the warranty disclaimer below."
Now I'm even more confused. How does this not cover derivative works?
It will take awhile for them to figure it out.
on
Futurama Returns!
·
· Score: 1
A quote from TFA:
It might be a very brief window when DVDs are so powerful. If the show had been on 10 years earlier, we'd be dead. A few years from now, when Internet speeds are better, maybe one person buys it and shares it with a hundred of their friends.
Or maybe right now, you can release it on the Internet along with a region-free DVD, and Fox becomes irrelevant? Because right now, Internet speeds are plenty good enough for BitTorrent to work, and wouldn't it be great if it was your medium, instead of only for piracy? Is the creator of Futurama really that fucking clueless?
Well, except that Fox probably owns at least some of the Futurama IP, so they won't let that happen, because they don't want to be irrelevant.
First, why is it that the number of connections has anything to do with it? I thought it was the amount of bandwidth used -- that is, number of packets sent?
Second, if this is really the case, then why won't Comcast simply come clean, and put in their FAQ "You get x simultaneous open TCP connections. Go over that, and we will close them for you."
/etc/init.d/ was designed to take the system from power-up to login, and from logout to power-down.
Networking is not unique in being something that falls entirely outside of this.
On a server, networking is required for anything it's serving.
It's more like a dependency system. (Like, but not exactly the same.)
Networking is needed for Apache. Yet, it helps no one to have networking enabled when the cable is physically disconnected. It may be needed to login on some clients, just as the root filesystem must be mounted -- yet in other cases, various mountpoints and network connections are ephemeral.
It makes more sense to me to have a standard system based around events. That is, you define "Cable plugged in event -> network up", but maybe also "User changed wifi networks" events. And so on.
The fact is, an event system is a superset of init.d, and init.d isn't really even sufficient for servers anymore. Or, well, ever -- that's why we have things like cron.
1: If you download a copy of the GCC under the GPL v3, are you licensing your patents which the GCC infringes on to all third parties?
I fail to see how it takes a lot of discussion to arrive at a "no". The GPL is a license to redistribute. If you're only downloading a copy of the GCC, you're not bound by any license, period. If you're redistributing it, that's when it affects you.
Other lawyers in the SFLC have suggested that one is *not* allowed to change licenses on such files unless they are modified.
I'm not sure I follow. Why not?
Even if it required a modification, how large of one? Couldn't you just change a comment?
DJBs solution is to impose a non-native method of managing daemons, which though compatible across different systems, is not compatible with all of the other services present on the system. This is a right, royal, pain in the backside.
That is not the point.
The point is to develop a better one, in the hope that it will replace all the various native ones.
I'm not defending just how much of a failure svc is at that, or the whole licensing bullshit, but I've got to admire the guy for having the balls to try -- to write daemontools in the first place..
I disagree that its the spirit so much as an interpretation of the spirit. Sure, it's the interpretation of the original author, but that might not be the spirit developers picked up on when they read the GPL so theres certainly room to complain.
Not really. No one's forcing you to use GPL, nor are they forcing you to use any particular variant of GPL.
Specifically, the code signing/hardware clause I take issue with as I see hardware and software as two separate things, with the software's license having no place mucking with hardware.
It doesn't. However, it does specify which hardware you're allowed to distribute it with / run it on.
That's hardly new -- for instance, Apple software can only be run on Apple hardware, and most versions of Windows Vista may not be run anywhere but inside a virtual machine. I imagine you'll find most software comes with similar restrictions.
But the really common misconception here is that the software does nothing to the hardware, it only specifies which hardware you're allowed to run it on. And I think that's entirely fair. For example, the Java license used to prevent it being used in nuclear facilities...
Theres legit reasons to not want arbitrary code running on a device. Look at how much crap Rockstar got in over people going out of their way to modify their software(GTA3) to get to a sex scene that is otherwise not at all accessible.
That's fine. They just can't use GPL'd code, then.
Now imagine what happens when, say, TiVo(let's face it, the reason people care about this clause) has a fork that allows any user to easily share their shows and create private mesh nets of tv shows, including a few PC clients as archive dumps so that people can have access to all tv shows they want.
When that happens, most people will go "meh" and go back to BitTorrent.
Let's face it, DRM does. Not. Work.
While we're at it, let's not get confused -- that doesn't mean that DRM isn't a problem, and that it couldn't be deadly to F/OSS. What it means is that only one person has to crack the DRM on a piece of media, and then that media is free forever (on the Internet). If we're talking about signed binaries, or similar restrictions, that's entirely different -- everyone has to crack their own hardware now, to get it to run anything but the stock software. (Just look at iPhones for an example.)
Yeah, that would be awesome, and the end user would be better off for it, but you can't tell me TiVo wouldn't be in for a world of even more ill will from Big Media(tm), if not outright lawsuits as they're profitting directly from these forks.
What would they possibly be sued for?
More relevantly, this is, again, entirely TiVo's choice: Either use entirely GPLv2 stuff, or write your own damned software (or buy/license it), or you have to play by our rules.
I don't understand why it suddenly becomes so complicated, as if we're forcing TiVo to do anything. All we're forcing them to do is not freeload off our current and future work -- in fact, they can freeload all they like off our past work.
But what about more distant forks? Lets say you had a web based virus scanner where people could upload a file and it would run a bunch of your custom checks and also a GPL virus scanner's scan on it and give you the results. Should all of that code be forced to be released?
All the code that's required. I'm not sure I see how your custom checks are "infected", unless you link them directly into the GPL'd code.
I like the GPL, but its just too messy and situational. I think Public Domain's best feature is that it has none of this gray area and just is what it is, making things like SQLite so easy to embed.
I like that too, as a consumer. But as a developer, I doubt I'd
On my system networking is done by deleting the mess of scripts that came with the distro and writing a simple script with a few ifconfig and iwconfig commands.
Why?
I mean, seriously, do you also delete the "mess of scripts that came with the distro" for, say, starting a graphical login, and just run 'startx' yourself? Or maybe real men don't even use login/getty, and you wrote your own login prompt?
Out of curiosity, what do you do on foreign wireless networks? Just manually iwlist and iwconfig every one?
What's wrong with "ifconfig eth1 down"? That works on every distro.
That you would even ask that tells me you haven't thought particularly hard about this.
What's wrong with "killall -9 apache"? It works on every distro (well, except the ones where apache is called apache2, or httpd, or my-custom-apache-build). It guarantees that the service will go down. And it's flatly retarded, in every case but yours, and of questionable intelligence in your case.
Go look in (on Gentoo)/etc/conf.d/net, or (on Ubuntu)/etc/networking. Specifically, look at the documentation. There's a lot more going on, even for the stupidly common case. Quick question: Where does your DHCP client daemon go if you just down the interface directly? Or maybe it's a more complex one -- what happens to your virtual machine configs? Or your static routes?
It's not the end of the world, probably, but neither is killall -9 apache, probably.
And while manually maintaining a bunch of simple ifconfig (read: static) commands, and just dealing with it when something breaks -- that's fine for you, but it's fairly naive to suggest that everyone else will tolerate that. There's a reason those distro-specific scripts were created in the first place.
(One more hint: When I plug my Ubuntu laptop in, it notices it has a link and tries to DHCP. When I unplug it, it starts looking for wireless networks. That's a ton of scripts I don't have to write, or even run.)
The biggest problem you may have here is that there are two competing definitions -- the real one, and the ones the ISPs made up.
The real one goes: ISPs shall be neutral with respect to network traffic. This is really, really, ridiculously, ludicrously simple: you put a router between your customer and the Internet. You do not put any firewall or packet shaping rules there.
There's a lot of ways to be more specific and less possible to poke legal holes in it. But that's the part of it that's as simple as, for instance, "Don't steal other people's shit." There's all kinds of ways to steal other people's shit, and there are separate laws for most of them, but the definition of stealing really is pretty simple, most of the time -- taking something from someone else without their consent.
Now, the other definition is just the opposite: The ISPs, of course, don't like net neutrality legislation -- they would rather have the governments not regulate them. So they twist it into how net neutrality is supposed to be about the government remaining neutral about the ISPs (net providers). And so you get morons like this, except most of them think they're FOR net neutrality.
c) Such traffic consists of unsolicited commercial email, and the customer has requested that the ISP filter inbound email to remove spam.
Don't make it spam-specific. Make it possible for the consumers to opt-in to very specific and clearly defined filters -- that is, if it claims to filter spam, it will not also filter bittorrent. And make sure that's opt-in, not opt-out, so that unless people are specifically requesting some sort of filter or shaping, they don't get it.
Slashdot Mirror
Link to one single comment by someone who has done that.
In fact, link to one single comment by someone who claims mp3s are technically superior to wma. The only reason we like them better is that they're somewhat more open. WMA does sound a lot better.
Fortunately, you really don't have to choose -- you could just use ogg or flac, both of which are likely better quality than mp3, and both of which are unencumbered.
Given that they cost (new) around five times more than the same thing digitally, or than what I think it's worth, that's a small comfort.
Please don't ever say that again. I know what you meant -- at least, I really hope I do -- but let's avoid this situation.
Because even if the download is flac, you can still probably fit about one and a half album per backup CD, and you absolutely can fit at least 10 or so to a single-layer DVD. If the download is mp3, that's more like ten albums per CD (and 150 or more to a DVD).
Also, because it's likely cheaper, and much easier to buy it directly from the band, or close enough -- Internet labels like Magnatune take only some 50%, which sounds like a lot until you consider that it's going to cost over 90% with the big CD labels.
1) Free, open source, no royalties. Costs you nothing to allow it, even if you already have mp3.
2) Because you can re-encode to anything else.
3) Because saying "digital music" is just as easy as saying "mp3 file", and is actually far more likely to be understood.
I'm not saying there shouldn't be mp3 files also. But I am saying that everywhere there's mp3, there should also be flac derived from the original source, if it's available. (Assuming the "original source" isn't mp3.)
That's the way I most often see it done. mp3 for people who want it to "just work" without having to think about it, and flac so I can re-encode it to a better format if my player supports it.
There's already an open source reference implementation. I'm not sure, but it might be public domain -- in any case, it's been ported enough places that I imagine it's fairly portable C by now.
Or you can just tell your bosses that it is, says so right there on the site, Slashdot agrees, etc, and they can choose to ignore the lawyers.
Seriously, that's about as dumb as having to pay your lawyers to find out if you need to pay for Linux licenses on your servers.
You may need to increase the amount of processing power or memory to handle the additional codec.That's the legitimate one.
If you don't already have an automated testing framework, you really should get one. It'll save you money in the long run.
Also legitimate -- but I thought that was the point of the above testing?
Well, if all of the above become cheap enough, you could support pretty much every format out there. (Or you could, you know, ship your player with RockBox, but I guess that's too obvious.)
Cheap enough, and you can simply tell your marketing department "We're doing it this way" -- just like with the legal department.
I'll agree that it's a poor design to require so much memory to boot, but give it some swap and you're good. I've found that on very old hardware, once it's properly booted and a good chunk of things swapped out, it performs decently enough.
I didn't say they were violating the license, just that it annoys me how they choose to contribute.
I'd rather have them contribute than not contribute, though.
And why do you suppose they started with a kernel they could get under a BSD license?
If Apple were to fork the latest WineHQ Wine, thus requiring them to give code back (LGPL'd and all), I'd argue that it's not because they're so willing to share the code, but because it would take them much more work to bring any of the other forks up to speed. I don't necessarily believe they'd avoid releasing at all, but rather, that they'd stall and delay and maybe give something back every year or two.
Well, to be entirely fair, "add new hardware" is fairly irrelevant most of the time.
Stuff that works out of the box tends to be autodetected, even if it does require those config file tweaks. Stuff that doesn't or can't work out of the box wouldn't fit very well with the standard Windows "add new hardware" dialog -- it's not as if there's going to be a driver CD -- although I believe there's a similar option for adding in support for various disabled-by-default binary kernel modules (nVidia, for example).
In KDE -- in ANY KDE app -- I can type KIO URLs. What this means is that Unix paths are a subset of what I can open and save to.
Other examples: zip:/home/sanity/foo.zip/bar.png -- throw it in Gwenview and, rather than unzipping the gallery somewhere, I just view as-is. Or kate zip:/home/sanity/foo.odt/content.xml -- apparently, it's read-only, but still pretty damned cool. Konqueror and Dolphin tend to discover these kinds of things on their own -- clicking a zip doesn't open file-roller or ark or anything similar in dolphin, it just opens it as a folder, transparently switching to the "zip" interface.
Or the ssh access. fish://user@host/ and it's as if I've mounted their filesystem. I mean, sure, I use ssh itself most of the time, but there's something almost beautifully blasphemous about using GUI editors and tools on a remote machine, especially when they don't have to know they're doing it. (Obviously Kate works, but so does KWord, Krita... like I said, any KDE app. Imagine that -- KWord fish://root@webserver/etc/apache2/httpd.conf)
And that's just what I've sort of picked up casually, with things like tip of the day.
I don't mean to disrespect Rox, nor am I completely defending the idea of software bloat, but we do actually use some of that "crap in the background" -- and once you get past the boot time, KDE 3, even, is pretty usable on really pathetic hardware. (Also, I may have to use Rox a bit to know if it can do what I want -- I tend to use the commandline for 90% of my file management, and Dolphin/Konq for 10% (including just showing off how freakin cool it is.)
You mentioned scripting? Looks like KDE 4 is planning that, if there isn't EcmaScript (JavaScript) all over the place already.
I forget whose sig it was, but it fits beautifully: "We do what we must because we can."
Let's just not pay attention to them, even that much.
True, nothing wrong with wanting. But as Jayne says, "If wishes were horses, we'd all be eat'n steak!"
(I know that's not the original quote, but I like Jayne's version.)
Yes, they would. And if they got it, they'd find out very quickly how stupendously bad an idea it is.
You see, the reason Google's on top right now is that they were the simple, fast search engine. When it was between Google and Yahoo, Yahoo's homepage was almost as big as Slashdot's, with a tiny little "search" box somewhere, maybe, and Google was just better at getting results to you quickly anyway.
In other words... As much of a monopoly as they have, if Google ever does implement something like that, I'm switching to another search. Windows Live search, even. I imagine enough people would do that, or block them, that the "value" of those ads would plummet.
Ok, regardless of the legal fairness, I'd think removing those previews would actually reduce the likelihood of me visiting such a site.
Almost never do I see a Google result and say "Ok, I know all I needed to, not going to click." More often, I see one and say "Gee, looks like that site won't be very helpful, let's move on to the next one." I can only imagine my response would be like that, only more so, towards anyone who could allow Google to index them without allowing Google to summarize them.
I'm confused...
Were this Microsoft, I could understand, given that chunks of DOS were carried around for so long, and Win16 after that, but then, even modern kick-ass 64-bit dual-core processors can still run 16-bit code -- probably 8-bit code, if there is such a thing.
But I'm confused... Photoshop has run on Windows for a long time, which means x86. It also ran on 68k and PowerPC. And you're telling me it's an old, inflexible codebase? What they've done with it suggests that it would have to be extremely portable... guess not.
But then, I keep forgetting... Flash runs on everything, Windows, Linux, Mac, x86, PPC, ARM even, but not 64-bit. Either they don't care, or they've got a fairly huge engineering problem on their hands.
Disclaimer: I didn't even read grandparent, and your post seems reasonable enough.
What I find frustrating about Apple is their need to so tightly control every bit of code they borrow. Look how long it's taken for Webkit to go back into Konqueror... and don't even get me started on BSD/Darwin, whose policy seems to be "Open whenever we feel like it."
Thus, I suspect that, were Apple to include Wine, they'd fork it, improve it quite a lot (though largely in ways that can't easily be integrated back into Wine), assuming they didn't just fork Crossover, Cedega, or the newest version of Wine that's not LGPL'd. I don't know who to blame for this situation, actually -- it seems like Apple is not playing nice with others, yet with all the code there (well, most of the time), it seems like the projects which got forked could be re-integrating a lot of Apple improvements a lot faster.
Sometime I'll look it up, but I'm fairly sure I've never seen an ODF file with binary anywhere but as a separate file in the zip, and itself in a standard format.
Or say something like "It's MathML, plus this." This is done in the HD-DVD spec, in at least a few places where it includes standards like the DOM, but with fairly major changes/extensions.
That is what I meant, actually. Maybe I shouldn't have used the word "extending"?
Regardless, I'd rather have a more limited format that's actually feasible to implement than "LineSpacingLikeWord95" bullshit.
Now I'm even more confused. How does this not cover derivative works?
A quote from TFA:
Or maybe right now, you can release it on the Internet along with a region-free DVD, and Fox becomes irrelevant? Because right now, Internet speeds are plenty good enough for BitTorrent to work, and wouldn't it be great if it was your medium, instead of only for piracy? Is the creator of Futurama really that fucking clueless?
Well, except that Fox probably owns at least some of the Futurama IP, so they won't let that happen, because they don't want to be irrelevant.
First, why is it that the number of connections has anything to do with it? I thought it was the amount of bandwidth used -- that is, number of packets sent?
Second, if this is really the case, then why won't Comcast simply come clean, and put in their FAQ "You get x simultaneous open TCP connections. Go over that, and we will close them for you."
Networking is not unique in being something that falls entirely outside of this.
It's more like a dependency system. (Like, but not exactly the same.)
Networking is needed for Apache. Yet, it helps no one to have networking enabled when the cable is physically disconnected. It may be needed to login on some clients, just as the root filesystem must be mounted -- yet in other cases, various mountpoints and network connections are ephemeral.
It makes more sense to me to have a standard system based around events. That is, you define "Cable plugged in event -> network up", but maybe also "User changed wifi networks" events. And so on.
The fact is, an event system is a superset of init.d, and init.d isn't really even sufficient for servers anymore. Or, well, ever -- that's why we have things like cron.
I fail to see how it takes a lot of discussion to arrive at a "no". The GPL is a license to redistribute. If you're only downloading a copy of the GCC, you're not bound by any license, period. If you're redistributing it, that's when it affects you.
I'm not sure I follow. Why not?
Even if it required a modification, how large of one? Couldn't you just change a comment?
That much is probably true.
The point is to develop a better one, in the hope that it will replace all the various native ones.
I'm not defending just how much of a failure svc is at that, or the whole licensing bullshit, but I've got to admire the guy for having the balls to try -- to write daemontools in the first place..
Not really. No one's forcing you to use GPL, nor are they forcing you to use any particular variant of GPL.
It doesn't. However, it does specify which hardware you're allowed to distribute it with / run it on.
That's hardly new -- for instance, Apple software can only be run on Apple hardware, and most versions of Windows Vista may not be run anywhere but inside a virtual machine. I imagine you'll find most software comes with similar restrictions.
But the really common misconception here is that the software does nothing to the hardware, it only specifies which hardware you're allowed to run it on. And I think that's entirely fair. For example, the Java license used to prevent it being used in nuclear facilities...
That's fine. They just can't use GPL'd code, then.
When that happens, most people will go "meh" and go back to BitTorrent.
Let's face it, DRM does. Not. Work.
While we're at it, let's not get confused -- that doesn't mean that DRM isn't a problem, and that it couldn't be deadly to F/OSS. What it means is that only one person has to crack the DRM on a piece of media, and then that media is free forever (on the Internet). If we're talking about signed binaries, or similar restrictions, that's entirely different -- everyone has to crack their own hardware now, to get it to run anything but the stock software. (Just look at iPhones for an example.)
What would they possibly be sued for?
More relevantly, this is, again, entirely TiVo's choice: Either use entirely GPLv2 stuff, or write your own damned software (or buy/license it), or you have to play by our rules.
I don't understand why it suddenly becomes so complicated, as if we're forcing TiVo to do anything. All we're forcing them to do is not freeload off our current and future work -- in fact, they can freeload all they like off our past work.
All the code that's required. I'm not sure I see how your custom checks are "infected", unless you link them directly into the GPL'd code.
I like that too, as a consumer. But as a developer, I doubt I'd
Why?
I mean, seriously, do you also delete the "mess of scripts that came with the distro" for, say, starting a graphical login, and just run 'startx' yourself? Or maybe real men don't even use login/getty, and you wrote your own login prompt?
Out of curiosity, what do you do on foreign wireless networks? Just manually iwlist and iwconfig every one?
That you would even ask that tells me you haven't thought particularly hard about this.
What's wrong with "killall -9 apache"? It works on every distro (well, except the ones where apache is called apache2, or httpd, or my-custom-apache-build). It guarantees that the service will go down. And it's flatly retarded, in every case but yours, and of questionable intelligence in your case.
Go look in (on Gentoo) /etc/conf.d/net, or (on Ubuntu) /etc/networking. Specifically, look at the documentation. There's a lot more going on, even for the stupidly common case. Quick question: Where does your DHCP client daemon go if you just down the interface directly? Or maybe it's a more complex one -- what happens to your virtual machine configs? Or your static routes?
It's not the end of the world, probably, but neither is killall -9 apache, probably.
And while manually maintaining a bunch of simple ifconfig (read: static) commands, and just dealing with it when something breaks -- that's fine for you, but it's fairly naive to suggest that everyone else will tolerate that. There's a reason those distro-specific scripts were created in the first place.
(One more hint: When I plug my Ubuntu laptop in, it notices it has a link and tries to DHCP. When I unplug it, it starts looking for wireless networks. That's a ton of scripts I don't have to write, or even run.)
The biggest problem you may have here is that there are two competing definitions -- the real one, and the ones the ISPs made up.
The real one goes: ISPs shall be neutral with respect to network traffic. This is really, really, ridiculously, ludicrously simple: you put a router between your customer and the Internet. You do not put any firewall or packet shaping rules there.
There's a lot of ways to be more specific and less possible to poke legal holes in it. But that's the part of it that's as simple as, for instance, "Don't steal other people's shit." There's all kinds of ways to steal other people's shit, and there are separate laws for most of them, but the definition of stealing really is pretty simple, most of the time -- taking something from someone else without their consent.
Now, the other definition is just the opposite: The ISPs, of course, don't like net neutrality legislation -- they would rather have the governments not regulate them. So they twist it into how net neutrality is supposed to be about the government remaining neutral about the ISPs (net providers). And so you get morons like this, except most of them think they're FOR net neutrality.
Hope that clears some things up.
Don't make it spam-specific. Make it possible for the consumers to opt-in to very specific and clearly defined filters -- that is, if it claims to filter spam, it will not also filter bittorrent. And make sure that's opt-in, not opt-out, so that unless people are specifically requesting some sort of filter or shaping, they don't get it.
But yes, it is pretty easy to define.