If your Scrum Master looks anything like a traditional Project Manager then you're doing it (very, very) wrong.
It's a very different role. There is no directly mapping role of Project Manager in the Scrum method as the responsibilities of a traditional PM are split across the Scrum Master (facilitator and process enforcer/advocate), the Project Owner (project direction, task priority, release planning), and the Team (accountability, communication).
What's amusing here is that I do own an iPhone and love it; It's one of the best consumer inventions in a generation.
The OP's question was regarding the spirit of the GPL, the morals involved in using it in such a manor. I'm not arguing for or against the GPL itself, simply examining this particular case by the standards requested (the spirit, goals, morals the GPL enshrines). Based on those standards this use is a clear betrayal of what the GPL is for. Legally it'll hold up in court, sure, but that's neither the question nor the point.
That said: I'm a huge proponent of actual free, actually open source software. The GPL however, is anything but. It's the manifesto of a twisted movement with incredibly misguided values and overtly sinister methods. Those are the standards however, which the OP asked the world to hold their work up to.
I strongly support the OP's work, choices, and reasoning. That's probably because it is in such direct opposition to the values of the GPL, a product of a bunch of loons I am similarly in strong opposition to.
This is very much against the spirit of the GPL from a couple of directions.
1) From one side is the real $$ cost to the end user if they wish to modify and improve the software ("Freedom for the user"). Sure, the source is free...but the dev kit is not ($99 I think?) nor is the Mac you are required to own to run the dev kit at all. Even if you gave the app away, this point would still stand.
2) Another is the app store: Apple controls it, so there's little reason to believe someone else improving it a bit and publishing it again would be allowed to ("Freedom for the developer").
3) But mostly it is the real spirit of the GPL: To force all software to be "free" (however you choose to mis-define "free"), if it is yours or not. This application of the GPL does not force Apple to make the dev kit "free", the iPhone OS "free", or development on the iPhone "free". In fact if it's actually any good it promotes the closed nature of the iPhone ("Freedom for software itself").
"The creative commons license is a copyright, it's just a different (more permissive) kind of copyright."
Technically, yes. Functionally, no. A copyright intrinsically grants the holder an array of powerful controls over the work, controls which are the entire reason to bother having a copyright system at all. Creative Commons attempts to license away all but the most trivial and vain of those powers, effectively neutering the entire core of what a copyright is and what it is for. What is left is a copyright in name only. For all practical purposes it's giving the work into the public domain.
----------
Offering one's work effectively to the public domain may well be a noble cause. But the disingenuous rhetoric about it still being a copyright, etc is more then a little bit dishonest and frankly, evil.
Especially Swing dancing (Lindy Hop, Balboa, a bunch of others). It's a running joke in the swing dancing community that all the leads are engineers and all the follows are school teachers because 9 times out of 10 it's true. There's something about swing dancing in particular that works very well with the engineer brain, probably all the physics involved.
Opportunity: Woman outnumber men typically
Advantage: Woman love men who want to learn to dance almost as much as they love men who can dance.
Classes: Are very social and "solo" friendly. They all rotate everyone constantly, so even if you came with a partner you'd rotate away from them almost immediately. This lets you meet a lot of people in a way that automatically breaks the ice; Zero effort or courage involved, it just happens as part of the normal class.
Dances: As long as you follow the basic rules you'll never get turned down for a dance. They are: Rule 0: Don't be stinky Rule 1: Do no harm (don't be the guy that yanks and twists arms, for any reason) Rule 2: Have fun (even the most "basic" of dances will make someone's night if you're enjoying yourself) Rule 3: Dance on beat (You don't need fantastic rhythm, just don't dance faster or slower then the music or it'll make her brain hurt)
Confidence: Knowing how to dance and physically interacting with the opposite sex on a near constant basis goes a long way to building social confidence, composure, and getting ride of that "needy vibe" that freaks woman out.
Some starting points: www.SoCalSwing.com / www.LALindyHop.com (Los Angeles, Orange County) www.SwingTalk.com / www.lindylist.com (San Francisco area) www.Yehoodi.com (New York, National)
Avoid "ballroom" dance studios like the plague...they'll just milk you for money and isolate you from the larger dance community.
The analogy isn't really correct. We're talking about "free" software, intended to be reused w/o monetary compensation. For the music analogy we'd have to be talking about closed source, commercial software being copied into another product.
Music performances and computer source code are also fundamentally different things. Again we'd need to change the analogy from recorded music performance to written sheet music to even begin to be equatable (the exception are "remix artists", but frankly they are more makers of sonic collage then musician).
That all can be reduced to ones and zeros does not make them all the same thing.
ShineTheLight muses that people will take another person's work, without credit, and pass it off as their own. I would propose the GPL encourages this type of behavior rather then discourages it.
By creating a license model that is so legally toxic, the GPL pushes code reuse underground. Even if the copier wanted to give credit or contribute code back to the community, they could not because to do so would be to effectively sign a legal confession that they are using the original code at all. The effect of the GPL is that it says, "Give us every last byte of your code, hide all of it underground, or completely reinvent the wheel from scratch (which may cost you your job if you don't have enough time)". It's only reasonable that someone put in such a position would simply choose to hide the infraction.
In sharp contrast the BSD license encourage compliance, credit, and open contributions by throwing away the GPL's ridiculous assumption that "whatever is mine is mine and whatever is yours is mine too". Developers want to give proper credit; Adopting code from respected sources is a display of intelligence in software, and showing respect by giving credit gains mutual respect. Developers want to contribute code; Show and tell to the world again, is a time honored way of displaying skill and gaining respect. The GPL however, puts developers into an entirely unreasonable position with no good answer (no, drinking the koolaid is not a reasonable answer most of the time) and so forces developers to take the least unreasonable of the available (practical) options.
Without sync you've still got nothing. rename() affects the directory (the hard links) not the file itself. rename() absolute does not ensure your data is on disk.
Your example ensures your code has performed all it's write calls before moving the file into its final location (a check against the code crashing more then the FS), which is good. But again, without sync it's only about your own code crashing...it has nothing to do with the FS.
If you care, sync. Yes, it has a performance penalty; Reliability is costly.
"Okay, maybe this is a good thing, because maybe it just means that a default Ubuntu does a very good job of including enough apps that the average user can do everything they need to do. Or maybe it just means that most people, unlike me, don't enjoy playing with software."
There is, of course, another option more obvious, simple, and likely possibility: Ubuntu's software acquisition system (find, review, fetch, install) is not so great.
Case in point: The iPhone with iTunes. Users are all searching for, reviewing, downloading, and installing within minutes of opening the box. All on their own motivation, without frustration, without much if any documentation. iPhone users are thrilled to tell others (grandma and super geek alike) about their phone, the applications they've installed, all it can do and do really, really well. And everyone they talk to wants to hear about it because, face it...the iPhone does not suck.
Linux on the desktop however? Good luck with that.
As mentioned a bit below, Cygwin has largely eliminated most any need to run an actual Unix system on a desktop. Windows + Cygwin makes a hell of a better desktop workstation then any Linux distro to date (and likely ever). Face it...Windows does a hell of a better job running Unix software then Unix does running Windows software...and Windows software by and large is better desktop software.
-------
There's a reason I'm a BSD fan: I applaud pro-Unix development rather then the Linux mentality of being anti-Microsoft. Pro-positions are almost always more productive then anti-positions. I'll never recommend Windows for a server platform just as I'll similarly never recommend any Unix as a desktop OS. Anyone that "must" have a Unix OS on their desktop should be running Mac OS.
They had a lot of ambitious ideas, but they pushed it out way too early and it showed. It's just another example of bean counters pushing to see a fast ROI and so they pressure the devs into writing too much too fast. That always results in kludgy, fragile code...guaranteed death for an MMORPG. Shadowbane was dead before it left closed beta.
The business plan of MMORPGs is long game; They need a subscriber base that at least remains constant but ideally grows over time. But that requires a critical mass and when a game falls on its face early in its days...players abandon it never to return and it will never reach that critical mass. Players are fickle and combined with the fact that MMORPGs are more about the other players playing then the game itself, losing players means losing game value.
The bean counters however, are only concerned about the short game; What's the revenue this quarter. As a result they care more about the initial on sale then they do about subscribers.
When I played Jumpgate in the beta (oh, so long ago now...) while there were Jumpgate between sectors, you still had to cross each sector to get to the next gate in the line: There were no short cut gates.
Many sectors took 45+ mins at max speed of a very upgraded ship to cross. 45 mins of going in a single straight line pointed at the next jumpgate. And often you'd have to cross a dozen of such sectors (no joke) to get where you needed to go. That's double digit hours of real world "game" time, all while listening to hypnotic techo music that's incredibly sleep inducing.
And you couldn't AFK that 45 mins: If you hit a rock in the middle and blew up, it could cost you *days* of real world time to fully recover. And even in sectors where you knew you had a clear path, if you were off target on the next jumpgate you could run into it and blow up (and some were 90 degrees from where you were coming from, forcing you to make an S turn at the end anyway).
If you didn't hit a rock, and didn't hit the next jumpgate, you could still get attacked by other players or some of the hostile NPC "space creatures" around. And crashing in Jumpgate was *HUGELY* expensive to recover from.
---------
Thank god for worm hole theory and non-realistic gameplay: It saved the space sim.
---------
BTW, shameless plug:
Allegiance, the best game you've never played. And yes, it has "alephs" to jump though (thank the gods!)
Re:It'll suck as bad, or worse, then then 2nd did.
on
Deus Ex 3 Announced
·
· Score: 1
Deus Ex's problem was lack of marketing mostly, and maybe a bit to ahead of its time. Ion Storm was in a huge internal mess at the time and didn't really try to do much of anything with Deus Ex. The game got great press because it was a truly, undeniably fantastic game (a rarity...in a game press climate where the review of a game hasn't anything to do with the quality).
It's also much more of a thinking game and quite a bit of it was lost on people that weren't well educated. The story and story items (like the fantastic newspaper articles in the game) relied heavily on real events, conspiracy theory, etc. Most of the story just went over people's heads, most especially gamer teens (the "only" market anyone was looking at at the time).
Now...the industry is well aware of the older, better educated gamers. A game like the original Deus Ex, with the convincing, deep, thought provoking story line and detail oriented game play (the inventory and skills management, the tactical choices to make at every step) would do quite well. Of course...they'd have to market it...and they'd have to stay the hell away from the craptastic FPS shooter experience that is any console.
Which is about as likely as Bush pulling out of Iraq for xmas.
It'll suck as bad, or worse, then then 2nd did...
on
Deus Ex 3 Announced
·
· Score: 4, Interesting
I remember reading all the articles and developer interviews as the 2nd game was being designed and built. What was clearly apparent more then anything else was how completely blind they were to what made the first game such a huge hit. They gave themselves credit for a long list of aspects of the first game that barely had anything to do with its success and completely ignored everything that made the game great. The file result was no surprise to anyone that read those interviews and dev blogs.
And then...in the aftermath of the sequel...their interviews again showed they had no idea why their game was a complete and total flop.
They'll screw it up; There's really no chance in hell of them not completely screwing the pooch again. They haven't a clue what they did right or what they did wrong. Go replay the first game; It was great, it's still great, but it was a fluke. The industry isn't setup to create great games like that anymore.
In my view it's really hard to say that SVN is CVS done right. While it has improved on a great many aspects of CVS, for others it has been an amazingly huge step backwards. Practical reporting of any kind is a huge feature hole of SVN. It takes an extreme amount of non-trivial wrapper code to even start pulling useful information out of SVN (there are a ton of edge cases). The history of a particular path is incredibly difficult to reconstruct (there's no clean way to list PEG versions). -The developers of SVN don't believe path history matters at all.
CVS's reporting abilities aren't great either, but at least CVS can let you log from one tag to another...SVN, not at all. -You can...but my god does it take a huge amount of non-trivial code to do so. Scripting reports with CVS is far, far easier to do and do reliably then SVN (and far faster to generate), even allowing for SVN's --xml option (which is great, but doesn't substitute for real reporting functionality).
SVN can't tell you copied-to points (for example, to tags) only copied-from and so it becomes a significant research effort to figure out what real revision/branch of a file is in which tag. Per file...
There's a ton more that's flat out bad in SVN. Would I recommend it over CVS? Yes...but just barely. SVN is much more of a step to the side then a step forward.
That all said...Linus working from his email inbox of patch attachments instead of using CVS, SVN, or anything was just insanely dumb. At least he finally got away from that a few years ago. Linus, like most smart people in the world, is not smart at everything. In fact he excels in only a very, very tiny subset of the software process and is flat out horrible at most of the rest of it. His fan boys however, believe him to be their God and most are simply too ignorant to know any better.
Funny...FreeBSD has been able to adapt to these "competing" priorities very smoothly for a dog's life.
How many times have the "smart" people developing Linux completely swapped out the scheduler wholesale? The Linux developers would do themselves, and their users, a huge service by adopting FreeBSD's scheduler design if not the implementation outright.
It's my biggest pet peeve with Linux: Under any load whatsoever, interactive response goes down the drain. And really, it always has. This is a historic issue with Linux that a great many accept simply because they don't know any better. They don't know what a quality Unix system actually is like. "It's better then Windows" is their benchmark...
The Linux kernel devs should be smart enough and humble enough to know when they are out of their league and clearly, writing a scheduler that's even just better then Windows 95 used is beyond their abilities. It's high time they leveraged the open source model and actually borrowed the code that works instead of thinking they have the brain cells to fix it. But they won't, ever, because as smart as they are...they aren't smart enough to know when they are dumb.
Agreed. And since I check my LiveJournal friends page more often then I check email, throwing NetFlix's RSS feed into my friends page makes it trivial to keep up on what new stuff has been released.
I forgot to add: Cygwin - Trivially blowing away 99% of duel-system needs.
So we're down to 1% of "needs". Lets review the follow ups:
Brandybuck - "I've had eight PC systems in my life" Then STFU. You are the poster child for my argument.
SanityInAnarchy - "I play Quake3 on Linux, becaues it's much faster and better that way." *Yawn* Another clueless uber-geek. If you were a Linux user and booted Windows to play games, you might have had a hint of a valid argument. But the other way around? *bizz*, you lose, thanks for playing.
Pharmboy At least Pharmboy had some valid cases, albeit still only talking about 0.0001% of cases. And that's really my point: 99.999% of cases made for dual-booting are simply bogus, the actual tiny fraction of valid cases are themselves mostly done as short-term hacks. So why spend so much time on super-perfecting dual-boot? And why the hell should it get anywhere near front-page slashdot coverage?
bperkins boot loader vs multi-boot loader. -Strawman, give me your email and I'll paypal you a $1 to buy a clue. WRT BIOS updates you're just wrong. -Most bios updates are done via floopy, via DOS or no OS.
heatdeath Please quote where I mentioned the word "laptop". WTF are you talking about? Although it is funny to see you thinking you can post and mod at the same time. It's a real shame...the collective IQ of/. is dropping like a rock...
Once upon a time computer hardware was rather expensive. Spending as much as a decent used car would cost on a 2nd or 3rd box to run another OS wasn't a practical reality for a great many.
But today? A rather high powered machine can be had new for a couple hundred. Add to this fact that 99.9999% of users who want/need multiple OSes have gone through a few upgrade cycles already and thus have hardware otherwise doing nothing...and it quickly becomes apparent that optimizing a system to handle booting multiple OSes is nothing but a complete and total waste of time.
If you want to multi-boot for some geek fun, more power to you, but so far as practical use is concerned there is absolutely, possitively, zip, zero, zlitch use for it. This story is front page news worthy...how exactly?
I know Make. Make is a friend of mine. Ant is no Make.
Ant can automate Java builds slightly better then a.bat or shell script...but that's sadly about where it ends.
The major problem is that Ant has no built in dependancy system. Dependant tasks don't count as they have no targets and thus nothing to check for dependancies. All real dependancy checking in Ant is embedded in the Ant task code...which varies task to task and worse yet has little to no documentation (not even documented to be taking place or not!). Examples are javac, zip, copy tasks.
Ant + Javamake gets closer to a Make replacement, but not completely.
But really, to be accurate Ant is a portable.bat/shellscript replacement...and that's it. Claiming it is a legit replacement for Make is to completely misunderstand what Make actually is and does. That doesn't mean Ant isn't useful, it is, but it is no Make.
Or even as low as $5/month. Switch your line to "metered service" for local calls. You'll be charged by the minute for all local calls even as close as next door (after a $3.00/month buffer), but you're not using the line anyway right?
I have a phone attached to my line for emergencies (like calling my cell so I can find where I left the bugger), with the ringer turned off.
So "Naked DSL" saves you $5/month...and maybe some $20 setup charge or such. *yawn* This is news worthy?
All open source business models boil down to about the same thing, selling "support" of one kind or another.
But this model has some major drawbacks, the worst of which is that it actually encourages the authors to write bad software, ignore documentation, etc. The better the software becomes, the more clear the documentation, the less support is actually needed. The open source "business model" actually penalizes good work...and the better the work, the higher the penalties.
Very large consulting forms (IBM, etc) can still make this model work as most of their "support" comes not from supporting the original software but rather from "integration" work; writing kludge software to get app A to talk to B and C and move data between D, E, and F. IOW they are supporting the infrastructure at large...not just their "own software". This business model however does not scale down; it only really works at the high end.
There is one other open source business model that can sometimes work, but typically only for a single developer. This is the model of "honor" or "donation" software such as BitTorrent, where the author is the high tech version of a street musician playing for tips. But I'd hardly call such a "business model".
Don't get me wrong, I'm a huge advocate of the open source movement...I just believe the people who really think you can create a practical business theory around it are confused. Almost without exception every "successful" open source business is either writing trash (eg RedHat), doing integration work (eg IBM), or begging for spare change (eg BitTorrent).
I thought the "Free" in "Free Software" was meant as "Free Speech"? Now there's a boycott effort in the free software community to try and limit MS's freedom of speech?
Ok, so I'm no big fan of MS either, but I must conclude that anyone who takes part in such an effort has lost any moral argument about "free software" being at all about "free speech".
Slashdot Mirror
If your Scrum Master looks anything like a traditional Project Manager then you're doing it (very, very) wrong.
It's a very different role. There is no directly mapping role of Project Manager in the Scrum method as the responsibilities of a traditional PM are split across the Scrum Master (facilitator and process enforcer/advocate), the Project Owner (project direction, task priority, release planning), and the Team (accountability, communication).
1) That's entirely not the point of the GPL. If an author truly wished for such an outcome they would use a BSD style license, not the GPL.
What's amusing here is that I do own an iPhone and love it; It's one of the best consumer inventions in a generation.
The OP's question was regarding the spirit of the GPL, the morals involved in using it in such a manor. I'm not arguing for or against the GPL itself, simply examining this particular case by the standards requested (the spirit, goals, morals the GPL enshrines). Based on those standards this use is a clear betrayal of what the GPL is for. Legally it'll hold up in court, sure, but that's neither the question nor the point.
That said: I'm a huge proponent of actual free, actually open source software. The GPL however, is anything but. It's the manifesto of a twisted movement with incredibly misguided values and overtly sinister methods. Those are the standards however, which the OP asked the world to hold their work up to.
I strongly support the OP's work, choices, and reasoning. That's probably because it is in such direct opposition to the values of the GPL, a product of a bunch of loons I am similarly in strong opposition to.
This is very much against the spirit of the GPL from a couple of directions.
1) From one side is the real $$ cost to the end user if they wish to modify and improve the software ("Freedom for the user"). Sure, the source is free...but the dev kit is not ($99 I think?) nor is the Mac you are required to own to run the dev kit at all. Even if you gave the app away, this point would still stand.
2) Another is the app store: Apple controls it, so there's little reason to believe someone else improving it a bit and publishing it again would be allowed to ("Freedom for the developer").
3) But mostly it is the real spirit of the GPL: To force all software to be "free" (however you choose to mis-define "free"), if it is yours or not. This application of the GPL does not force Apple to make the dev kit "free", the iPhone OS "free", or development on the iPhone "free". In fact if it's actually any good it promotes the closed nature of the iPhone ("Freedom for software itself").
"The creative commons license is a copyright, it's just a different (more permissive) kind of copyright."
Technically, yes. Functionally, no.
A copyright intrinsically grants the holder an array of powerful controls over the work, controls which are the entire reason to bother having a copyright system at all.
Creative Commons attempts to license away all but the most trivial and vain of those powers, effectively neutering the entire core of what a copyright is and what it is for.
What is left is a copyright in name only. For all practical purposes it's giving the work into the public domain.
----------
Offering one's work effectively to the public domain may well be a noble cause. But the disingenuous rhetoric about it still being a copyright, etc is more then a little bit dishonest and frankly, evil.
I have Moderator points: How do I use them to mod down the original story for being invalid, kneejerk, flamebait?
This.
Especially Swing dancing (Lindy Hop, Balboa, a bunch of others). It's a running joke in the swing dancing community that all the leads are engineers and all the follows are school teachers because 9 times out of 10 it's true. There's something about swing dancing in particular that works very well with the engineer brain, probably all the physics involved.
Opportunity: Woman outnumber men typically
Advantage: Woman love men who want to learn to dance almost as much as they love men who can dance.
Classes: Are very social and "solo" friendly. They all rotate everyone constantly, so even if you came with a partner you'd rotate away from them almost immediately. This lets you meet a lot of people in a way that automatically breaks the ice; Zero effort or courage involved, it just happens as part of the normal class.
Dances: As long as you follow the basic rules you'll never get turned down for a dance. They are:
Rule 0: Don't be stinky
Rule 1: Do no harm (don't be the guy that yanks and twists arms, for any reason)
Rule 2: Have fun (even the most "basic" of dances will make someone's night if you're enjoying yourself)
Rule 3: Dance on beat (You don't need fantastic rhythm, just don't dance faster or slower then the music or it'll make her brain hurt)
Confidence: Knowing how to dance and physically interacting with the opposite sex on a near constant basis goes a long way to building social confidence, composure, and getting ride of that "needy vibe" that freaks woman out.
Some starting points:
www.SoCalSwing.com / www.LALindyHop.com (Los Angeles, Orange County)
www.SwingTalk.com / www.lindylist.com (San Francisco area)
www.Yehoodi.com (New York, National)
Avoid "ballroom" dance studios like the plague...they'll just milk you for money and isolate you from the larger dance community.
The analogy isn't really correct. We're talking about "free" software, intended to be reused w/o monetary compensation. For the music analogy we'd have to be talking about closed source, commercial software being copied into another product.
Music performances and computer source code are also fundamentally different things. Again we'd need to change the analogy from recorded music performance to written sheet music to even begin to be equatable (the exception are "remix artists", but frankly they are more makers of sonic collage then musician).
That all can be reduced to ones and zeros does not make them all the same thing.
ShineTheLight muses that people will take another person's work, without credit, and pass it off as their own. I would propose the GPL encourages this type of behavior rather then discourages it.
By creating a license model that is so legally toxic, the GPL pushes code reuse underground. Even if the copier wanted to give credit or contribute code back to the community, they could not because to do so would be to effectively sign a legal confession that they are using the original code at all. The effect of the GPL is that it says, "Give us every last byte of your code, hide all of it underground, or completely reinvent the wheel from scratch (which may cost you your job if you don't have enough time)". It's only reasonable that someone put in such a position would simply choose to hide the infraction.
In sharp contrast the BSD license encourage compliance, credit, and open contributions by throwing away the GPL's ridiculous assumption that "whatever is mine is mine and whatever is yours is mine too". Developers want to give proper credit; Adopting code from respected sources is a display of intelligence in software, and showing respect by giving credit gains mutual respect. Developers want to contribute code; Show and tell to the world again, is a time honored way of displaying skill and gaining respect. The GPL however, puts developers into an entirely unreasonable position with no good answer (no, drinking the koolaid is not a reasonable answer most of the time) and so forces developers to take the least unreasonable of the available (practical) options.
Without sync you've still got nothing. rename() affects the directory (the hard links) not the file itself. rename() absolute does not ensure your data is on disk.
Your example ensures your code has performed all it's write calls before moving the file into its final location (a check against the code crashing more then the FS), which is good. But again, without sync it's only about your own code crashing...it has nothing to do with the FS.
If you care, sync. Yes, it has a performance penalty; Reliability is costly.
"Okay, maybe this is a good thing, because maybe it just means that a default Ubuntu does a very good job of including enough apps that the average user can do everything they need to do. Or maybe it just means that most people, unlike me, don't enjoy playing with software."
There is, of course, another option more obvious, simple, and likely possibility: Ubuntu's software acquisition system (find, review, fetch, install) is not so great.
Case in point: The iPhone with iTunes. Users are all searching for, reviewing, downloading, and installing within minutes of opening the box. All on their own motivation, without frustration, without much if any documentation. iPhone users are thrilled to tell others (grandma and super geek alike) about their phone, the applications they've installed, all it can do and do really, really well. And everyone they talk to wants to hear about it because, face it...the iPhone does not suck.
Linux on the desktop however? Good luck with that.
As mentioned a bit below, Cygwin has largely eliminated most any need to run an actual Unix system on a desktop. Windows + Cygwin makes a hell of a better desktop workstation then any Linux distro to date (and likely ever). Face it...Windows does a hell of a better job running Unix software then Unix does running Windows software...and Windows software by and large is better desktop software.
-------
There's a reason I'm a BSD fan: I applaud pro-Unix development rather then the Linux mentality of being anti-Microsoft. Pro-positions are almost always more productive then anti-positions. I'll never recommend Windows for a server platform just as I'll similarly never recommend any Unix as a desktop OS. Anyone that "must" have a Unix OS on their desktop should be running Mac OS.
They had a lot of ambitious ideas, but they pushed it out way too early and it showed. It's just another example of bean counters pushing to see a fast ROI and so they pressure the devs into writing too much too fast. That always results in kludgy, fragile code...guaranteed death for an MMORPG. Shadowbane was dead before it left closed beta.
The business plan of MMORPGs is long game; They need a subscriber base that at least remains constant but ideally grows over time. But that requires a critical mass and when a game falls on its face early in its days...players abandon it never to return and it will never reach that critical mass. Players are fickle and combined with the fact that MMORPGs are more about the other players playing then the game itself, losing players means losing game value.
The bean counters however, are only concerned about the short game; What's the revenue this quarter. As a result they care more about the initial on sale then they do about subscribers.
When I played Jumpgate in the beta (oh, so long ago now...) while there were Jumpgate between sectors, you still had to cross each sector to get to the next gate in the line: There were no short cut gates.
Many sectors took 45+ mins at max speed of a very upgraded ship to cross. 45 mins of going in a single straight line pointed at the next jumpgate. And often you'd have to cross a dozen of such sectors (no joke) to get where you needed to go. That's double digit hours of real world "game" time, all while listening to hypnotic techo music that's incredibly sleep inducing.
And you couldn't AFK that 45 mins: If you hit a rock in the middle and blew up, it could cost you *days* of real world time to fully recover. And even in sectors where you knew you had a clear path, if you were off target on the next jumpgate you could run into it and blow up (and some were 90 degrees from where you were coming from, forcing you to make an S turn at the end anyway).
If you didn't hit a rock, and didn't hit the next jumpgate, you could still get attacked by other players or some of the hostile NPC "space creatures" around. And crashing in Jumpgate was *HUGELY* expensive to recover from.
---------
Thank god for worm hole theory and non-realistic gameplay: It saved the space sim.
---------
BTW, shameless plug:
Allegiance, the best game you've never played. And yes, it has "alephs" to jump though (thank the gods!)
Deus Ex's problem was lack of marketing mostly, and maybe a bit to ahead of its time. Ion Storm was in a huge internal mess at the time and didn't really try to do much of anything with Deus Ex. The game got great press because it was a truly, undeniably fantastic game (a rarity...in a game press climate where the review of a game hasn't anything to do with the quality).
It's also much more of a thinking game and quite a bit of it was lost on people that weren't well educated. The story and story items (like the fantastic newspaper articles in the game) relied heavily on real events, conspiracy theory, etc. Most of the story just went over people's heads, most especially gamer teens (the "only" market anyone was looking at at the time).
Now...the industry is well aware of the older, better educated gamers. A game like the original Deus Ex, with the convincing, deep, thought provoking story line and detail oriented game play (the inventory and skills management, the tactical choices to make at every step) would do quite well. Of course...they'd have to market it...and they'd have to stay the hell away from the craptastic FPS shooter experience that is any console.
Which is about as likely as Bush pulling out of Iraq for xmas.
I remember reading all the articles and developer interviews as the 2nd game was being designed and built. What was clearly apparent more then anything else was how completely blind they were to what made the first game such a huge hit. They gave themselves credit for a long list of aspects of the first game that barely had anything to do with its success and completely ignored everything that made the game great. The file result was no surprise to anyone that read those interviews and dev blogs.
And then...in the aftermath of the sequel...their interviews again showed they had no idea why their game was a complete and total flop.
They'll screw it up; There's really no chance in hell of them not completely screwing the pooch again. They haven't a clue what they did right or what they did wrong. Go replay the first game; It was great, it's still great, but it was a fluke. The industry isn't setup to create great games like that anymore.
In my view it's really hard to say that SVN is CVS done right. While it has improved on a great many aspects of CVS, for others it has been an amazingly huge step backwards. Practical reporting of any kind is a huge feature hole of SVN. It takes an extreme amount of non-trivial wrapper code to even start pulling useful information out of SVN (there are a ton of edge cases). The history of a particular path is incredibly difficult to reconstruct (there's no clean way to list PEG versions). -The developers of SVN don't believe path history matters at all.
CVS's reporting abilities aren't great either, but at least CVS can let you log from one tag to another...SVN, not at all. -You can...but my god does it take a huge amount of non-trivial code to do so. Scripting reports with CVS is far, far easier to do and do reliably then SVN (and far faster to generate), even allowing for SVN's --xml option (which is great, but doesn't substitute for real reporting functionality).
SVN can't tell you copied-to points (for example, to tags) only copied-from and so it becomes a significant research effort to figure out what real revision/branch of a file is in which tag. Per file...
There's a ton more that's flat out bad in SVN. Would I recommend it over CVS? Yes...but just barely. SVN is much more of a step to the side then a step forward.
That all said...Linus working from his email inbox of patch attachments instead of using CVS, SVN, or anything was just insanely dumb. At least he finally got away from that a few years ago. Linus, like most smart people in the world, is not smart at everything. In fact he excels in only a very, very tiny subset of the software process and is flat out horrible at most of the rest of it. His fan boys however, believe him to be their God and most are simply too ignorant to know any better.
Linux has been swapping schedulers on a regular basis for over a decade now.
They need to give up.
They need to borrow a schedule design that actually works.
It'll never happen of course. Too many egos in the way.
Funny...FreeBSD has been able to adapt to these "competing" priorities very smoothly for a dog's life.
How many times have the "smart" people developing Linux completely swapped out the scheduler wholesale? The Linux developers would do themselves, and their users, a huge service by adopting FreeBSD's scheduler design if not the implementation outright.
It's my biggest pet peeve with Linux: Under any load whatsoever, interactive response goes down the drain. And really, it always has. This is a historic issue with Linux that a great many accept simply because they don't know any better. They don't know what a quality Unix system actually is like. "It's better then Windows" is their benchmark...
The Linux kernel devs should be smart enough and humble enough to know when they are out of their league and clearly, writing a scheduler that's even just better then Windows 95 used is beyond their abilities. It's high time they leveraged the open source model and actually borrowed the code that works instead of thinking they have the brain cells to fix it. But they won't, ever, because as smart as they are...they aren't smart enough to know when they are dumb.
Agreed. And since I check my LiveJournal friends page more often then I check email, throwing NetFlix's RSS feed into my friends page makes it trivial to keep up on what new stuff has been released.
I forgot to add: Cygwin - Trivially blowing away 99% of duel-system needs.
/. is dropping like a rock...
So we're down to 1% of "needs". Lets review the follow ups:
Brandybuck - "I've had eight PC systems in my life"
Then STFU. You are the poster child for my argument.
SanityInAnarchy - "I play Quake3 on Linux, becaues it's much faster and better that way."
*Yawn* Another clueless uber-geek. If you were a Linux user and booted Windows to play games, you might have had a hint of a valid argument. But the other way around? *bizz*, you lose, thanks for playing.
Pharmboy
At least Pharmboy had some valid cases, albeit still only talking about 0.0001% of cases. And that's really my point: 99.999% of cases made for dual-booting are simply bogus, the actual tiny fraction of valid cases are themselves mostly done as short-term hacks. So why spend so much time on super-perfecting dual-boot? And why the hell should it get anywhere near front-page slashdot coverage?
bperkins
boot loader vs multi-boot loader. -Strawman, give me your email and I'll paypal you a $1 to buy a clue.
WRT BIOS updates you're just wrong. -Most bios updates are done via floopy, via DOS or no OS.
heatdeath
Please quote where I mentioned the word "laptop". WTF are you talking about? Although it is funny to see you thinking you can post and mod at the same time. It's a real shame...the collective IQ of
Once upon a time computer hardware was rather expensive. Spending as much as a decent used car would cost on a 2nd or 3rd box to run another OS wasn't a practical reality for a great many.
But today? A rather high powered machine can be had new for a couple hundred. Add to this fact that 99.9999% of users who want/need multiple OSes have gone through a few upgrade cycles already and thus have hardware otherwise doing nothing...and it quickly becomes apparent that optimizing a system to handle booting multiple OSes is nothing but a complete and total waste of time.
If you want to multi-boot for some geek fun, more power to you, but so far as practical use is concerned there is absolutely, possitively, zip, zero, zlitch use for it. This story is front page news worthy...how exactly?
I know Make. Make is a friend of mine. Ant is no Make.
.bat or shell script...but that's sadly about where it ends.
.bat/shellscript replacement...and that's it. Claiming it is a legit replacement for Make is to completely misunderstand what Make actually is and does. That doesn't mean Ant isn't useful, it is, but it is no Make.
Ant can automate Java builds slightly better then a
The major problem is that Ant has no built in dependancy system. Dependant tasks don't count as they have no targets and thus nothing to check for dependancies. All real dependancy checking in Ant is embedded in the Ant task code...which varies task to task and worse yet has little to no documentation (not even documented to be taking place or not!). Examples are javac, zip, copy tasks.
Ant + Javamake gets closer to a Make replacement, but not completely.
But really, to be accurate Ant is a portable
Or even as low as $5/month. Switch your line to "metered service" for local calls. You'll be charged by the minute for all local calls even as close as next door (after a $3.00/month buffer), but you're not using the line anyway right?
I have a phone attached to my line for emergencies (like calling my cell so I can find where I left the bugger), with the ringer turned off.
So "Naked DSL" saves you $5/month...and maybe some $20 setup charge or such. *yawn* This is news worthy?
All open source business models boil down to about the same thing, selling "support" of one kind or another.
But this model has some major drawbacks, the worst of which is that it actually encourages the authors to write bad software, ignore documentation, etc. The better the software becomes, the more clear the documentation, the less support is actually needed. The open source "business model" actually penalizes good work...and the better the work, the higher the penalties.
Very large consulting forms (IBM, etc) can still make this model work as most of their "support" comes not from supporting the original software but rather from "integration" work; writing kludge software to get app A to talk to B and C and move data between D, E, and F. IOW they are supporting the infrastructure at large...not just their "own software". This business model however does not scale down; it only really works at the high end.
There is one other open source business model that can sometimes work, but typically only for a single developer. This is the model of "honor" or "donation" software such as BitTorrent, where the author is the high tech version of a street musician playing for tips. But I'd hardly call such a "business model".
Don't get me wrong, I'm a huge advocate of the open source movement...I just believe the people who really think you can create a practical business theory around it are confused. Almost without exception every "successful" open source business is either writing trash (eg RedHat), doing integration work (eg IBM), or begging for spare change (eg BitTorrent).
I thought the "Free" in "Free Software" was meant as "Free Speech"? Now there's a boycott effort in the free software community to try and limit MS's freedom of speech?
Ok, so I'm no big fan of MS either, but I must conclude that anyone who takes part in such an effort has lost any moral argument about "free software" being at all about "free speech".