Let's hear it for duplicated work! While the rest of the world marches on, it is inspiring to see that at least a few are still toiling away to make software look like it did 5 years ago.
Nobody forces anyone to use RoadRunner. Just get another broadband service or go back to 56k if you absolutely *must* share your 0-d4y l337 m0v13z with the rest of the world.
Excuse me? Have you ever been to China? You're stupid-ass whining would be put into perspective if you're ever arrested for "subversion" or "spying". Do you get a jury of peers? Not fucking likely. Do you have any freedom of expression? No, you cannot even speak against government policies (whereas in America, I'm sure you're aware, you can scream as loud as you want to about any government policy, no matter how flawed your opinion).
So please, shut the fuck up until you have the slightest idea what life is like outside of middle school.
It can only lead to improvement. Remember the whole affair of dropping in a completely new VM right in the middle of the supposedly "stable" 2.4 kernel branch? BSD has never had problems like these, and it pains me to see Linux doing silly stuff like that. I await the day when I can safely assume that something located in the "stable" branch really is stable.
Why bother? If you have a UPS, all you need to do is let it alert your servers to the loss of external power and the servers can begin a clean shutdown sequence, certainly well within your 70 minute range. Most APC UPSes that I know of have a serial cable hookup. If you have more than one server hooked up to one UPS, I'm sure you could devise someway of one server recieving the power-down signal and broadcasting it to all your other machines over the network.
Of course, if you're worried about data integrity, you will have a mirror across multiple striped drives using multiple controllers. And then use a Journaled Filesystem to improve boot time.
This is a misinformed opinion, at best. Your RAID setup will only save data in the case of hardware failure (i.e. one of your disks fails). It will do nothing about incomplete writes. The whole purpose of journaled filesystems is to ensure that writes completed, to minimize filesystem corruption. It just so happens that the way it does this allows for a faster boot, which is an added bonus.
That said, releasing unreadable code, while not as useful as releasing highly documented, readable code, is still better than not releasing anything.
Unreadable code actually wastes more time than simply starting a project from scratch. As pointed out by the parent poster, people have to spend time going through a codebase before they give up on an OSS project and write their own software from scratch. Besides, releasing unreadable code is a violation of the spirit of the GPL; you might as well release binaries.
Sure, it can be problematic when one cannot understand or sometimes even read their software license, but there are greater problems afoot in the open source movement.
The primary problem is the absolute unreadability of most of the code. Now, I don't mean to flame or troll or anything like that, but there are a lot of parts of the Linux kernel (especially those written in assembler) that are difficult to read and understand. Why can't the Linux kernel maintainers write these illegible functions in a clearer, more concise language like Perl?
Speaking of Perl, there is a lot of awfully illegible code written with this language as well. Take Slashcode, for instance. Have you ever tried to read Slashcode, let alone make improvements to it? I have, and let me tell you, it ain't pretty. This kind of illegibility led me to dump a Perl-based site altogether in favor of something better-documented and more legible, such as (regrettably) ASP.NET.
The point of my little tirade here is simply thus: you defeat the purpose of opening your source when no one else can make heads or tails of it. Furthermoree, you are only shooting yourself in the foot when you try to audit the code, since you won't be able to use the "many eyes make every bug shallow" model to fix things (a concept first posited in Eric Raymond's masterpiece The Cathedral and the Bizarre). So please, comment your code, people, and code legibly.
How come this post got modded down to -1 territory? Methinks those with unlimited mod points have a conflict of interest.
But seriously, I was considering subscribing to the site to support my habit and help keep a great site like this up. I changed my mind when I saw the huge ads that the VA decided to put up to coerce people into subscriptions. Sheesh, why don't you just put them right in the middle of people's comments? They couldn't be much uglier.
I can't really see why they have the need to come out with ever faster graphics boards. I can play games like Deus Ex and CounterStrike fine already with my GeForce3 (in fact, it is already a little overkill). You really have to start asking the question "Are they ever going to make a game detailed enough to take advantage of this capacity?"
It seems that when Microsoft uses planned obsolescense, you all call it a monopoly and scream about it. But, for some reason, when a hardware company (which, might I add, doesn't have opensource drivers, thus stealing from the community) does it, you applaud them for "progress", a progress that is essentially pushing everyone who has a machine a year or more old to upgrade if they want to even play the latest games, much less enjoy them.
Yeah, and God forbid the major media companies are prevented from plugging their favorite companies while masquerading as unbiased news. Really, it's a double standard.
And for the record, the business plans you describe closely resemble those of almost all of the Linux companies, including VA Linux (you know, the people who own Slashdot?). Most companies, like Microsoft, make money the honest way, by selling something, rather than posing as an unbiased source whilst plugging their advertisers' products.
One is led to wonder just how much this guy has invested in "KnowNow". He sure couldn't get enough of plugging it in his column. How unprofessional, it's almost as bad as Slashdot plugging ThinkGeek crap here.
Slashdot Mirror
Let's hear it for duplicated work! While the rest of the world marches on, it is inspiring to see that at least a few are still toiling away to make software look like it did 5 years ago.
Nobody forces anyone to use RoadRunner. Just get another broadband service or go back to 56k if you absolutely *must* share your 0-d4y l337 m0v13z with the rest of the world.
So please, shut the fuck up until you have the slightest idea what life is like outside of middle school.
Correct me if I'm wrong, but didn't their culture spend several thousands of years as the most advanced on Earth?
It can only lead to improvement. Remember the whole affair of dropping in a completely new VM right in the middle of the supposedly "stable" 2.4 kernel branch? BSD has never had problems like these, and it pains me to see Linux doing silly stuff like that. I await the day when I can safely assume that something located in the "stable" branch really is stable.
Why bother? If you have a UPS, all you need to do is let it alert your servers to the loss of external power and the servers can begin a clean shutdown sequence, certainly well within your 70 minute range. Most APC UPSes that I know of have a serial cable hookup. If you have more than one server hooked up to one UPS, I'm sure you could devise someway of one server recieving the power-down signal and broadcasting it to all your other machines over the network.
This is a misinformed opinion, at best. Your RAID setup will only save data in the case of hardware failure (i.e. one of your disks fails). It will do nothing about incomplete writes. The whole purpose of journaled filesystems is to ensure that writes completed, to minimize filesystem corruption. It just so happens that the way it does this allows for a faster boot, which is an added bonus.
Now the whole internet can get BSODed when the routers running the network crash.
Yeah right. Hey, guess what? Just because you have a low uid doesn't mean that you're not a troll. Leave this guy alone, he never did anything to you.
He linked to a source of information from a US Government site. WTF are you talking about, troll?
Unreadable code actually wastes more time than simply starting a project from scratch. As pointed out by the parent poster, people have to spend time going through a codebase before they give up on an OSS project and write their own software from scratch. Besides, releasing unreadable code is a violation of the spirit of the GPL; you might as well release binaries.
The primary problem is the absolute unreadability of most of the code. Now, I don't mean to flame or troll or anything like that, but there are a lot of parts of the Linux kernel (especially those written in assembler) that are difficult to read and understand. Why can't the Linux kernel maintainers write these illegible functions in a clearer, more concise language like Perl?
Speaking of Perl, there is a lot of awfully illegible code written with this language as well. Take Slashcode, for instance. Have you ever tried to read Slashcode, let alone make improvements to it? I have, and let me tell you, it ain't pretty. This kind of illegibility led me to dump a Perl-based site altogether in favor of something better-documented and more legible, such as (regrettably) ASP.NET.
The point of my little tirade here is simply thus: you defeat the purpose of opening your source when no one else can make heads or tails of it. Furthermoree, you are only shooting yourself in the foot when you try to audit the code, since you won't be able to use the "many eyes make every bug shallow" model to fix things (a concept first posited in Eric Raymond's masterpiece The Cathedral and the Bizarre). So please, comment your code, people, and code legibly.
This certainly explains Linux's stellar security record of late.
Oh yeah, get in Enya, Pr0n_K1ng!
But seriously, I was considering subscribing to the site to support my habit and help keep a great site like this up. I changed my mind when I saw the huge ads that the VA decided to put up to coerce people into subscriptions. Sheesh, why don't you just put them right in the middle of people's comments? They couldn't be much uglier.
You are not!
Developer: Hey, my game is so inefficient, it makes the last generation of high end graphics boards choke! Yeah!
No one will pay for that kind of bullshit.
It seems that when Microsoft uses planned obsolescense, you all call it a monopoly and scream about it. But, for some reason, when a hardware company (which, might I add, doesn't have opensource drivers, thus stealing from the community) does it, you applaud them for "progress", a progress that is essentially pushing everyone who has a machine a year or more old to upgrade if they want to even play the latest games, much less enjoy them.
Earl! How the fuck have you been, you spastic hell?
Just put JonKatz and Fidel Castro in the same room. See who can stop talking first.
And for the record, the business plans you describe closely resemble those of almost all of the Linux companies, including VA Linux (you know, the people who own Slashdot?). Most companies, like Microsoft, make money the honest way, by selling something, rather than posing as an unbiased source whilst plugging their advertisers' products.
One is led to wonder just how much this guy has invested in "KnowNow". He sure couldn't get enough of plugging it in his column. How unprofessional, it's almost as bad as Slashdot plugging ThinkGeek crap here.
Hey, yeah! Maybe they can call it the "Gray Death"!
...in my pants!
Just wondering...