Then the FOSS people come along and say you should get a car based solely on the brand of transmission (kernel) inside.
More like the engine than the transmission--but even then it's a horrible analogy.
[...] Imagine if Apple constantly went on and on about OpenDarwin / XNU in their mass-market advertising. Or if Mozilla waged a "Get Gecko" campaign to end-users... They would be in the 1-2% penetration bracket nowadays with a nonsensical message like that.
Nope. The term "Linux" has come to encompass the common set of GNU and other userland tools and applications and desktop environments generally distributed with a Linux kernel--the entire "operating system", in the common meaning of the term. When IBM, Novell, the Linux Foundation, and others say "get Linux", they don't mean just the Linux kernel; they're clearly talking about the whole package. And nobody seems to have much trouble understanding this message.
Then there is the 'Get Ubuntu' crowd, which I admire (and I followed their advice). BUT Ubuntu is not a PC platform: It doesn't have a set UI to make life manageable for users and tech support alike, nor an SDK for app developers, nor a program for certifying hardware for the OS, nor a way to independently distribute application packages that will still work 6 months (nevermind 2-6 years) down the road.
Of course Ubuntu is a PC platform. The Ubuntu project itself is all about providing that set UI you desire. And clearly it has an SDK, it's just that the SDK is most properly referred to as "GNOME" or whatnot rather than "Ubuntu". Standardization is there, too; again, it's just that the standards aren't called "Ubuntu", but rather POSIX, LSB, or freedesktop.org, for instance, depending on what aspect of standardization you're interested in.
Applications cannot be independently distributed and still work down the road? My copies of VMware Workstation and Google Earth might have a thing or two to say about that. In fact Ubuntu provides dpkg, which means that it is much better off in terms of actual software installation and uninstallation than Windows is. (On the other hand, Windows remains the undisputed king of backwards compatibility, over OS X and Linux. However, time has shown that this can have significant long-term consequences for the overall quality of the operating system and its APIs.)
Many of your concerns here boil down to the fact that the actual standardization and unification efforts in the Linux world, to which developers can easily turn for instruction and assurance, don't all have the label "Ubuntu" slapped on them. My point is that, sure, you can consider Ubuntu as a platform unto itself if you'd like--but most aspects of GNU/Linux are now well-standardized enough that there's little need to narrow down to a single distribution for most purposes.
In fact, Firefox looks more like a PC platform in some important areas than any Linux distro.
Not only that, it completely ignores the probable rise of inexpensive and energy-efficient ARM-based netbooks. Windows 7 won't be running on those *at all*.
Yes, I do work in IT. And no, it isn't too much to ask of any competent admin, to have some basic awareness of the popular alternatives to expensive Microsoft solutions. Especially when those alternatives have generated as much press coverage as Likewise has.
I'm a busy admin too. Fortunately it doesn't take long at all to install Ubuntu Server, apt-get install likewise-open, and then type "domainjoin-cli join my.domain my-username" in the command line.
When you use being "busy" as an excuse for being ignorant of your options, you do your employer a disservice. That page you linked to hasn't had a major edit in two years or so, and it does not reflect the current best practices for setting up a simple Linux/Samba file server with AD integration. And no, no extra $$ is required for Ubuntu Server.
As a minor nitpick, TeX input (in conjunction with all images and other source files) does specify the rendered output exactly. Donald Knuth went to great lengths to ensure that a given TeX input will render identically on every machine on which TeX runs, even going so far as to use fixed-decimal numeric representations rather than whatever floating-point formats a given architecture may natively support.
Not that I'd send a TeX/LaTeX input file to someone when a PDF or PostScript file would suffice, of course.
You have to draw the line somewhere; I'm not sure it's correctly drawn right now.
Do you have to draw the line somewhere? Does the government actually have to step in and say, it's all right to put these substances in your body, but not those?
I disagree. I think it is not necessary. More importantly, I think the government has no right to tell us what we are and are not allowed to take into our own bodies.
I agree with you to a point, but I also blame Microsoft for implementing AutoRun in the first place, and then for making it so counter-intuitive to fully disable.
Octave is a great program (I switched from MATLAB too, after my bought-and-paid-for copy of MATLAB was broken by a simple OS X upgrade). But Octave is not a symbolic computer algebra system like Mathematica, Maple, Maxima, etc., so it cannot properly be called a Mathematica alternative.
Shh. Remember apple runs "fast" and is "glorious for multimedia!" somehow we skipped Linux and AMD, but hey, want to pay 2x as much for half the performance?
As for OS X versus Linux, the majority of multimedia encoding and decoding operations occur in userspace, so the Linux kernel has no speed advantage to offer there. Where the two operating systems do differentiate themselves is with their multimedia APIs and codecs; as a user of both Linux and OS X, it saddens me that OS X is so far ahead in this regard.
The server at http://www.genunix.org/, where this OpenSolaris 2008.11 ISO is hosted, is responding rather slowly right now (indirect Slashdotting?). So I want to point out that if you'd like to download this build and try it for yourself, you can get it as a torrent here.
First off, we really shouldn't count Macs as part of the equation. I haven't checked recently, but for a long time, OOo's support for MacOS X lagged way, way behind. It was essentially unusable.
No, we have to count Macs. One of the big bullet points on the OpenOffice 3 release notes was its new native Aqua support on OS X.
If I'm keeping track of a bunch of branches in my head, then that limits the amount of other stuff I can keep track of.
But the idea is that if your project demands you keep track of these things anyway, branching can help ease that burden.
The time I and the people on my team spend talking about the different branches is time we can't talk about actual features. If my ops guy or my product manager has to figure out what's where and what's elsewhere, then that's time, attention, and energy that they can't put into what they're really there for.
That's not at all applicable to the kind of local branching used in Git and Bzr, for the reasons I outlined in the previous post. You can make a gazillion branches in your local copy of the Git repository, and your ops guy and product manager won't even see them unless you explicitly push them back to the central repo.
As it turns out, the awesomeness of Git (and certain other distributed SCMs) can reduce that human cost to zero.
As I've mentioned elsewhere in this thread, I agree that some version control systems are better than others at handling branches. But the awesomeness of your tools can never reduce the cost to zero, because much of the cost isn't in the software.
Then what is the nature of this cost you refer to?
That's not to say git isn't awesome for certain situations, mind you. But branching and merging adds a fair bit of overhead, and anything that increases a project's overhead should be your last resort, not your first.
Well OK, those principles are appropriate for traditional tools like CVS and SVN. Branching and merging adds a lot of overhead if your SCM isn't built to handle it. But once you've used a modern SCM like Git or Bzr, you'll know that with Git you literally cannot branch too often. Don't think of Git branches like you think of Subversion branches, it's a whole different playing field.
I have a rule of thumb with Git: every time I set out to implement a new feature, if I think it's going to take more than a single commit to bring into the mainline, then I make a new branch for it. Working on three different features at the same time? Then use three different branches, and switch between them at will.
So what's the big difference between SVN and Git that allows such carefree use of branching? There are a few points to understand:
At one point, the SVN developers liked to make a big deal out of the fact that branching in SVN is a lot cheaper than it is in CVS. That's all well and good, but it doesn't mean much when SVN's handling of merges is so primitive. Git handles merges extremely well, in no small part due to the fact that, like Bzr, Git tracks content rather than changes. (If you browse through the recent history of the Linux kernel, you can witness 12- and even 20-way merges; with Git, merging really is no big deal.)
In Git (and in Bzr), branches are not immutable, permanent parts of your repository. You can delete (or simply not push) branches when you're done with them. This means you're free to make as many experimental branches as you like in your own local repository; you can take advantage of the full productivity power of Git branches without immortalizing your failed experiments in the central repo.
In Git (and in Bzr, but not in Mercurial), there doesn't have to be a 1-1 mapping of named branches between each copy of your distributed repository. This has the important result that you can name your experimental branch "experiment" in your local repo, without being concerned about polluting some global branch namespace. This does a way with a lot of the mental overhead associated with branches in SVN.
In conclusion, 90% of the time I see people advocating the use of branches only as a last resort, it's because they're using the wrong SCM =)
Something tells me that an email system that only depends on your hard drive with no net connection wouldn't be very useful.
What? You've got to be kidding, right? It is incredibly useful.
Except for a mail server and spam filtering software running on a VPS, my entire mail system resides on my laptop. I download new messages with fetchmail, sort them into mailboxes with procmail, read them with mutt, write them with vim, send them with msmtp. The upshot of this is that when my home Internet connection goes down, or when I'm out and about but don't have an access point, or should my server somehow catch fire -- I can still access all of my mailboxes locally. I can still edit and save drafts. I'm not stuck wondering, "Crap, when did she say that meeting will be?"; it's all right there.
This GMail outage exemplifies Reason #1 for using POP3 for email access.
Your 3-rd dimension is discrete. The push is for a continuous dimension.
All right, I have one question for you: Why?
Why would that be a worthy goal? Why should we assume that a three-dimensional browsing experience would be inherently better than a two-dimensional one? Note: Astute observations that 3>2 need not apply.
Don't be a fool. The fact that it is the default behavior is problematic enough, especially when people carelessly suggesting "Just use OpenDNS!" on Slashdot and elsewhere never seem to finish that breath with "...but be sure to sign up for an account with them, and log in to disable these features, and then install a dynamic DNS client on your computer and configure it to send updates to OpenDNS whenever your public IP address changes, otherwise they'll start hijacking your traffic again whenever you get a new IP address from your ISP."
So you tell me, why does it make any sense to recommend OpenDNS to anyone, when Level 3 and others have publicly-accessible servers that are faster and that respect users' privacy without gratuitous configuration and software installation?
It's troubling how many people will blindly recommend OpenDNS without understanding the huge problems with that service. Stay far, far away from OpenDNS - that is, unless you just don't care that they redirect all your Google queries through their own servers:
www.google.com is an alias for google.navigation.opendns.com. google.navigation.opendns.com has address 208.69.32.231 google.navigation.opendns.com has address 208.69.32.230
Or that they break with acceptable DNS behavior by sending you to their own advertising web server, rather than return a NXDOMAIN response, when a name cannot be resolved. (Good luck filtering spam with a DNSRBL if you're using OpenDNS.)
www.ajvelkajslkjalkvjeasl.com has address 208.69.32.132
Use Level3's anycast DNS servers instead: 4.2.2.1, 4.2.2.2,..., 4.2.2.6. They're faster than OpenDNS and they don't pull any of that nonsense on their users.
Slashdot Mirror
Then the FOSS people come along and say you should get a car based solely on the brand of transmission (kernel) inside.
More like the engine than the transmission--but even then it's a horrible analogy.
[...] Imagine if Apple constantly went on and on about OpenDarwin / XNU in their mass-market advertising. Or if Mozilla waged a "Get Gecko" campaign to end-users... They would be in the 1-2% penetration bracket nowadays with a nonsensical message like that.
Nope. The term "Linux" has come to encompass the common set of GNU and other userland tools and applications and desktop environments generally distributed with a Linux kernel--the entire "operating system", in the common meaning of the term. When IBM, Novell, the Linux Foundation, and others say "get Linux", they don't mean just the Linux kernel; they're clearly talking about the whole package. And nobody seems to have much trouble understanding this message.
Then there is the 'Get Ubuntu' crowd, which I admire (and I followed their advice). BUT Ubuntu is not a PC platform: It doesn't have a set UI to make life manageable for users and tech support alike, nor an SDK for app developers, nor a program for certifying hardware for the OS, nor a way to independently distribute application packages that will still work 6 months (nevermind 2-6 years) down the road.
Of course Ubuntu is a PC platform. The Ubuntu project itself is all about providing that set UI you desire. And clearly it has an SDK, it's just that the SDK is most properly referred to as "GNOME" or whatnot rather than "Ubuntu". Standardization is there, too; again, it's just that the standards aren't called "Ubuntu", but rather POSIX, LSB, or freedesktop.org, for instance, depending on what aspect of standardization you're interested in.
Applications cannot be independently distributed and still work down the road? My copies of VMware Workstation and Google Earth might have a thing or two to say about that. In fact Ubuntu provides dpkg, which means that it is much better off in terms of actual software installation and uninstallation than Windows is. (On the other hand, Windows remains the undisputed king of backwards compatibility, over OS X and Linux. However, time has shown that this can have significant long-term consequences for the overall quality of the operating system and its APIs.)
Many of your concerns here boil down to the fact that the actual standardization and unification efforts in the Linux world, to which developers can easily turn for instruction and assurance, don't all have the label "Ubuntu" slapped on them. My point is that, sure, you can consider Ubuntu as a platform unto itself if you'd like--but most aspects of GNU/Linux are now well-standardized enough that there's little need to narrow down to a single distribution for most purposes.
In fact, Firefox looks more like a PC platform in some important areas than any Linux distro.
Now that's just plain silly.
Not only that, it completely ignores the probable rise of inexpensive and energy-efficient ARM-based netbooks. Windows 7 won't be running on those *at all*.
Yes, I do work in IT. And no, it isn't too much to ask of any competent admin, to have some basic awareness of the popular alternatives to expensive Microsoft solutions. Especially when those alternatives have generated as much press coverage as Likewise has.
I'm a busy admin too. Fortunately it doesn't take long at all to install Ubuntu Server, apt-get install likewise-open, and then type "domainjoin-cli join my.domain my-username" in the command line.
When you use being "busy" as an excuse for being ignorant of your options, you do your employer a disservice. That page you linked to hasn't had a major edit in two years or so, and it does not reflect the current best practices for setting up a simple Linux/Samba file server with AD integration. And no, no extra $$ is required for Ubuntu Server.
As a minor nitpick, TeX input (in conjunction with all images and other source files) does specify the rendered output exactly. Donald Knuth went to great lengths to ensure that a given TeX input will render identically on every machine on which TeX runs, even going so far as to use fixed-decimal numeric representations rather than whatever floating-point formats a given architecture may natively support.
Not that I'd send a TeX/LaTeX input file to someone when a PDF or PostScript file would suffice, of course.
You have to draw the line somewhere; I'm not sure it's correctly drawn right now.
Do you have to draw the line somewhere? Does the government actually have to step in and say, it's all right to put these substances in your body, but not those?
I disagree. I think it is not necessary. More importantly, I think the government has no right to tell us what we are and are not allowed to take into our own bodies.
DoD needs a security nazi ( soup nazi style ).
That is to say, "No security for you!"?
I'm not sure I'm entirely sold on this idea...
I agree with you to a point, but I also blame Microsoft for implementing AutoRun in the first place, and then for making it so counter-intuitive to fully disable.
Octave is a great program (I switched from MATLAB too, after my bought-and-paid-for copy of MATLAB was broken by a simple OS X upgrade). But Octave is not a symbolic computer algebra system like Mathematica, Maple, Maxima, etc., so it cannot properly be called a Mathematica alternative.
Shh. Remember apple runs "fast" and is "glorious for multimedia!" somehow we skipped Linux and AMD, but hey, want to pay 2x as much for half the performance?
What? Are you actually suggesting that AMD's processors perform better than Intel's offerings? As a company AMD may have its strengths, but they cannot even remotely compete against Intel's Core 2, let alone Core i7.
As for OS X versus Linux, the majority of multimedia encoding and decoding operations occur in userspace, so the Linux kernel has no speed advantage to offer there. Where the two operating systems do differentiate themselves is with their multimedia APIs and codecs; as a user of both Linux and OS X, it saddens me that OS X is so far ahead in this regard.
The server at http://www.genunix.org/, where this OpenSolaris 2008.11 ISO is hosted, is responding rather slowly right now (indirect Slashdotting?). So I want to point out that if you'd like to download this build and try it for yourself, you can get it as a torrent here.
Pretty much anything in here: http://www.openbsd.org/cgi-bin/man.cgi
[...] Helped immensely by the spread of Christianity [...]
Unsubstantiated non-sequitur alert!
First off, we really shouldn't count Macs as part of the equation. I haven't checked recently, but for a long time, OOo's support for MacOS X lagged way, way behind. It was essentially unusable.
No, we have to count Macs. One of the big bullet points on the OpenOffice 3 release notes was its new native Aqua support on OS X.
You can even get DSL if you have a good enough falsetto.
If I'm keeping track of a bunch of branches in my head, then that limits the amount of other stuff I can keep track of.
But the idea is that if your project demands you keep track of these things anyway, branching can help ease that burden.
The time I and the people on my team spend talking about the different branches is time we can't talk about actual features. If my ops guy or my product manager has to figure out what's where and what's elsewhere, then that's time, attention, and energy that they can't put into what they're really there for.
That's not at all applicable to the kind of local branching used in Git and Bzr, for the reasons I outlined in the previous post. You can make a gazillion branches in your local copy of the Git repository, and your ops guy and product manager won't even see them unless you explicitly push them back to the central repo.
As it turns out, the awesomeness of Git (and certain other distributed SCMs) can reduce that human cost to zero.
As I've mentioned elsewhere in this thread, I agree that some version control systems are better than others at handling branches. But the awesomeness of your tools can never reduce the cost to zero, because much of the cost isn't in the software.
Then what is the nature of this cost you refer to?
That's not to say git isn't awesome for certain situations, mind you. But branching and merging adds a fair bit of overhead, and anything that increases a project's overhead should be your last resort, not your first.
Well OK, those principles are appropriate for traditional tools like CVS and SVN. Branching and merging adds a lot of overhead if your SCM isn't built to handle it. But once you've used a modern SCM like Git or Bzr, you'll know that with Git you literally cannot branch too often. Don't think of Git branches like you think of Subversion branches, it's a whole different playing field.
I have a rule of thumb with Git: every time I set out to implement a new feature, if I think it's going to take more than a single commit to bring into the mainline, then I make a new branch for it. Working on three different features at the same time? Then use three different branches, and switch between them at will.
So what's the big difference between SVN and Git that allows such carefree use of branching? There are a few points to understand:
In conclusion, 90% of the time I see people advocating the use of branches only as a last resort, it's because they're using the wrong SCM =)
Something tells me that an email system that only depends on your hard drive with no net connection wouldn't be very useful.
What? You've got to be kidding, right? It is incredibly useful.
Except for a mail server and spam filtering software running on a VPS, my entire mail system resides on my laptop. I download new messages with fetchmail, sort them into mailboxes with procmail, read them with mutt, write them with vim, send them with msmtp. The upshot of this is that when my home Internet connection goes down, or when I'm out and about but don't have an access point, or should my server somehow catch fire -- I can still access all of my mailboxes locally. I can still edit and save drafts. I'm not stuck wondering, "Crap, when did she say that meeting will be?"; it's all right there.
This GMail outage exemplifies Reason #1 for using POP3 for email access.
Try nspluginwrapper. (Which distro are you using? Nspluginwrapper is enabled by default in Ubuntu x86-64...)
I had a feeling someone would use this as a JavaScript-bashing opportunity. After all, JavaScript is the world's most misunderstood programming language.
Your 3-rd dimension is discrete. The push is for a continuous dimension.
All right, I have one question for you: Why?
Why would that be a worthy goal? Why should we assume that a three-dimensional browsing experience would be inherently better than a two-dimensional one? Note: Astute observations that 3>2 need not apply.
Isn't that kind of stupid to have a brain evolve a feature just to counteract another arbitrary feature?
Not necessarily. Check out Daniel Dennett's book Breaking the Spell for some interesting hypotheses as to why the placebo effect might be adaptive.
Don't be a fool. The fact that it is the default behavior is problematic enough, especially when people carelessly suggesting "Just use OpenDNS!" on Slashdot and elsewhere never seem to finish that breath with "...but be sure to sign up for an account with them, and log in to disable these features, and then install a dynamic DNS client on your computer and configure it to send updates to OpenDNS whenever your public IP address changes, otherwise they'll start hijacking your traffic again whenever you get a new IP address from your ISP."
So you tell me, why does it make any sense to recommend OpenDNS to anyone, when Level 3 and others have publicly-accessible servers that are faster and that respect users' privacy without gratuitous configuration and software installation?
It's troubling how many people will blindly recommend OpenDNS without understanding the huge problems with that service. Stay far, far away from OpenDNS - that is, unless you just don't care that they redirect all your Google queries through their own servers:
Or that they break with acceptable DNS behavior by sending you to their own advertising web server, rather than return a NXDOMAIN response, when a name cannot be resolved. (Good luck filtering spam with a DNSRBL if you're using OpenDNS.)
Use Level3's anycast DNS servers instead: 4.2.2.1, 4.2.2.2, ..., 4.2.2.6. They're faster than OpenDNS and they don't pull any of that nonsense on their users.