Some distros make it more or less easy to install rights-restricted software, like the stuff you need to play mp3s or DVDs. Neither Fedora/RHEL nor Debian allow nonfree software in their repos, but it's generally a fairly painless process to add a repo that does.
Ubuntu will, IIRC, ask you during the install process if you want to install such things, and Linux Mint comes with the media codecs by default. For other distributions you should research this issue.
Fedora and Ubuntu are the "big" distros, more or less, although Mageia seems to be climbing up DistroWatch lately. I had written off that project as dead when its Corporate Overlord bit the dust, but it's probably worth checking out. I hope I may say with enough accuracy that it is of similar quality to OpenSuse.
Fedora and Ubuntu have the biggest corporate backing and are likely to represent the most polished experiences. Ubuntu has its own way of doing things, most notably they have implemented at least two desktop environments (Unity and UNR) and their own startup process. Startup tends to be one of those big differences between distributions, but it's something you can safely ignore as a n00b user.
Fedora and Ubuntu use incompatible packaging systems, which tends to be irrelevant for a couple of reasons that aren't worth going over here. Generally you should figure that [a] any distro that is described as being derived from any other distro is package-compatible, and [b] it's very uncommon to need to install a package outside of your distribution's package management tools. We don't download software off websites, pretty much everything that you would ever want to install comes in the box.
It's hard to come up with too many more big important differences between these things, really. Desktop environments make a pretty big difference. Distros, not so much, especially among the big players.
Oh, and I forgot to mention. If you ever want to give yourself a real education in Linux, try Linux From Scratch. You'll probably even survive the experience. By contrast, slackware will be a friendly and trivial introduction, and Gentoo......sorry, my Gentoo joke is still compiling:(
It's much easier for you to specify your needs as there are hundreds of distros and packages that can be combined. To a first approximation pretty much all linux packages are available for all distributions.
Beyond that, most linux distributions are based off some other distribution. The description of Kubuntu as "Ubuntu, but with the KDE desktop environment" is perfectly descriptive.
So what distinguishes one distro from another? Besides what comes installed by default, the most significant difference is how those packages got there.
Debian is probably the distribution that the greatest number of other distributions are based on. It has a very very long testing cycle; it takes packages years to get into Debian's stable branch. Ubuntu is based on Debian unstable, and a shit-ton of things are based on Ubuntu, including Linux Mint.
Red Hat produces the next biggest family of linuxes. Fedora and Red Hat Enterprise Linux are more or less analogous to Debian unstable and stable, respectively, but I don't think very many people are dumb enough to try and base a distro on Fedora. CentOS is RHEL with the logo removed, and Scientific Linux is also based on RHEL.
Next up we have Gentoo, Arch, Slackware, and Suse.
I was going to put a joke about Gentoo here, but it's taking a while to compile. Gentoo is a rolling-release distro where most of the packages that you use are compiled on and for your machine. You mention gcc, this is related, but you will probably not ever use it directly. Compiling packages yourself can make them run faster, but it can take a long time.
Arch is a well-documented, rolling release distribution. I'm not sure what else to say about them honestly, but "well-documented" is one of the highest compliments I'm aware of.
Slackware is the oldest and most "unixy" of the distributions. It uses an old bootloader, old unix-style boot scripts, and by default boots to a text terminal. You should use Slackware if you want to retreat into a cave for five years, to emerge with a profound knowledge of unix, a full beard, and a solid opinion on whether emacs or vi(m) is the best text editor. I'm pretty sure these things are highly marketable. No, really.
Suse hasn't failed yet. The last time I checked, they had a wonderful, polished experience, and great admin/configuration tools. I have no idea why they don't have more users, except that there's already a shit-ton of options.
It's probably fair to say that Debian stable, RHEL, and any derivatives will have the longest testing cycle, and fewest updates in any given span of time. There are many more distributions for more specialized purposes, such as BackTrack for pen testing, Puppy for small installations, Bodhi for those seeking Enlightenment. You may have to figure out what you need on your own there.
Whew! Let's take a break for a minute.
All right. So with all that in mind, you can install, as previously mentioned, pretty much all the same stuff on any and every distro.
Here is a guide on desktop environments. If you're a n00b, you're probably going to want one of those.
We also have another guide for more experienced users, or those on resource-constrained systems, that covers some of the more popular window managers. Because sometimes 2GB of gnome libraries gets a bit old. For the truly adventurous, this post covers 30 Window Managers in 30 Days.
Honestly, there's really a pretty limited amount of advice that one can give about using any particular distro. They're all substantially similar. Without any specific information about what you want to use, a recommendation becomes, well, exactly what you were complaining about. "Use XYZ bec
I like Debian. Linux Mint Debian Edition is a good option, although I am using Crunchbang on my netbook. The latter is based on Debian stable.
Stability is the kind of virtue that you appreciate most in its absence. After an enthusiastic period of Fedora and Ubuntu use, I from time to time experienced issues with packages and drivers breaking on updates. These were usually resolvable, and forced a certain amount of CLI-foo on me, but there's only so many times one wants to wrangle with things that worked just fine yesterday.
Stability means having outdated versions of packages; you miss out on the new features as well as the new bugs. However, it's also pretty trivial to install packages from unstable if you really need them, and if all else fails you can compile from source (which is usually a painless process).
Ubuntu was certainly far less buggy than Fedora, and I certainly don't mind all you guys being Debian beta testers;) but my choice of OS is going to be heavily informed by whichever one has the longest testing cycle.
He should have stuck to bringing home pork. He didn't get to be the longest-serving Republican senator because he was bad at his job. Sarah Palin on the other hand......yeah okay maybe there's a pattern here.
Indulge me in a little hyperbole: for a friend of mine, hacking AT&T was a death sentence.
Lance Moore was involved with LulzSec, foolishly no doubt. As an AT&T technician of some sort, he acquired and subsequently distributed some internal corporate documents. The Justice department is liable to be a more accurate source of the specific complaints. He was caught. The FBI seized its opportunity to bring the hammer down. I've seen various figures given for the amount of jail time he was facing; somewhere between five and thirty. He was found dead by his own hand on February 24 of last year. His crime has by now likely been forgotten by all that were involved with it.
Sixteen other people were arrested the same day that he was arrested. I don't know their stories. The reader may judge whether justice was served.
$6k a year is doable. $16k a year would be quite pleasant. I would avoid the capital or other large cities. Actually getting a work permit or visa to either country is difficult to impossible, but I know people in both countries who have been there for decades on a tourist visa. Do note, this tends to limit your options for local employment; it's far better to work online.
There's essentially no native culture (or cuisine) in either place, "post-colonial" about sums it up. The police are nice enough but underpaid, the laws are enforced relatively arbitrarily and generally not in favor of extranjeros. If you're running a business, [a] congratulations for getting through the bureaucracy to accomplish this, and [b] you may from time to time expect to have laws about licenses and restrictions enforced against you that your (Tico) competition does not. I'm not sure whether I can really say that corruption was common, but it's probably fair to say that people were understanding about dealing with the laws and regulations -- or avoiding that, if necessary. I don't really consider this a bad thing, but if you have the expectation that the rule of law is going to be universally or rigidly applied, you may be disappointed.
The weather is beautiful, it's not terribly expensive to get to and from either country (at least, from the US), English is spoken by a good percentage of the population, utilities are cheap and reliable, health care is extremely affordable (medical tourism is common), internet is not that fast but widely available, and of course, knowledgeable tech workers are in high demand. In Costa Rica the beer is not good and relatively expensive, in Panama you can get two beers for $1. Computers are available, but expensive. It's probably going to be a good idea to buy in the US and work out a way to get it. I've heard both good and bad things about the mail system; I'd call it generally reliable, but the paranoid might want to find other means of receiving packages. If you end up going back and forth to the states a lot, you can make good money on the side bringing electronics back with you.
Panama is by far the cheaper of the two countries, you would probably be able to get by on less than $6k annually. I didn't like it quite as much because, at least in the places I frequented, cocaine was both common and extremely cheap there. That's fine for those who like that sort of thing, but generally I don't think it does much good for the community. Drug laws in both countries are sparingly enforced.
Roads are generally better in Panama; the country has a lot more money due to that whole canal thing. I can't recommend driving in Panama City, or anywhere in Costa Rica. Cars are absurdly expensive, and paradoxically people don't care about the lines on the road, the blinky things above them, the relative speed and velocity of other vehicles, or pedestrians.
Religion can be rational. If you take the axiom, "God exists," and develop it, you can have a perfectly sound (to the limits of Godel) logical system.
You can also do this with the axiom "God does not exist," or "Lizard Men rule the world," or any other concept.
The problem with rationality is that, absent empirical claims/testing, you can prove pretty much anything. The problem with empiricism is that "truth" becomes limited by your ability to measure things, i.e. everything you know is a little bit wrong.
Let me know whether rationality or empiricism has achieved the greatest result.
The only authoritative reference to how a system functions is the source code. If you don't have the source, at some point you will encounter the problem that is unsolvable without it. Unless of course the original programmer wrote about everything that could ever be done with his software.
With source code, all things are possible. Not necessarily pleasant or easy, but possible.
As long as we're characterizing each other, let's be accurate: I have never worked in an enterprise environment. Lately I've been working freelance. It tends to involve wearing a lot of hats: developer, sysadmin, marketing, legal. I can rely on my own skills at this stuff until a certain point, after which it's safer to turn over to a professional. Why? Because I don't have the hubris to think that I am expert enough to make decisions in areas that I have not studied. Clearly there we differ.
As a developer, I don't think "what would I like to use?" very often. Instead I ask myself "What is the best practice?" or "What is the most professional solution?" or "What is best in the long run?" or "What is the next guy that comes across this going to want to work with?"
You obviously are asking none of these questions. It's an amazing ignorance you're devoting yourself to. "If he was so clever, he'd be a developer" Truly an incredible statement, and the end of all meaningful discussion on the matter.
The point was not enterprise, but professionalism. Let's not get into semantic arguments.
There is in fact a couple issues with developing professionally on OSX. For one, you clearly have never used OSX Server, or you would not recommend it. Second, as I mentioned, development (as opposed to staging or production) is fine to do on OSX but this should not have anything to do with colocation. Lastly, packages are developed for Mac OSX as an afterthought. This is compounded by the lack of a built-in package manager. It's not necessarily the case that you will have more issues doing dev work in a non-standard environment, but you'll have different issues, which is just as bad in terms of finding good information about them.
All that aside, you're acting like this is a simple matter of preference. The big bad sysadmin could just choose to use OSX instead, and that's just as good, and you are just as qualified as he is to make that choice. Again, this is more like the choice between (e.g.) Photoshop and GIMP. One is a professional, industry-standard tool. The other is not. There may be some circumstances where it's okay not to use the industry standard tool, but you do not know enough to make that decision. It's not a costless decision, and it's not arbitrary.
I'm sure that you know all this, and have a really good business case that you've somehow just forgotten to explain. You would never make capricious decisions about critical infrastructure.
No. If you want your own personal copy of wordpress running on OSX, do that. Install XAMPP, wordpress, source control. It takes all of ten minutes -- less if you script it.
You do not need a publicly-accessible server to do development on. You sure as shit don't need colo.
The argument that people should use what they're comfortable with is nice, but not part of an enterprise server solution. Let's take an analogy here: I am a pretty decent cook. Given sufficient time and a minimum of kitchen gear, I can produce a tasty dish (or two or three). I am completely unqualified to be in a restaurant kitchen. I don't even know what I don't know about that profession.
You say you have a tasty recipe that you've whipped up with your Mac. You tell me that you want me and (24 x 5U) of my friends to make this recipe for all your internet customers. This will work, to a degree, but is unlikely to scale very well. Also, even though your home-cooked meal is tasty and does the same thing as that restaurant meal, you and that restaurant chef are not actually in the same business. In order to become a part of that business, you need to [a] study more, and [b] use the same professional tools that everyone else uses. Sometimes the only reason things are widely used is because they're widely used, but more often there's a reason that impacts the quality of the product.
It's fine to not be in the professional cooking business. But saying "I don't need a professional kitchen/chef" based on this experience is foolhardy. You don't know what you don't know.
The most popular hosting is Godaddy, with their $10 / month hosting account. (35% of sites are Godaddy sites.) The sites with hosting budgets of around $10-$50 month make up 95% of all sites.
As a web developer I may say categorically, fuck them. If you put a site on the web, it is your responsibility to make sure that it is secure. If you are not able to do that to a professional standard, you should not do it. In point of fact, there is a need for a licensing organization to prevent amateurs from practicing web development. The problem isn't that this website is exposing poor security practices, it's that it's not promoting professionalism.
We have passed the point where it's okay for the layman to host a site. Even if you're not collecting information about your users, you can still be attacked or used as an attack vector. The era of democratization of the web is over.
Passing the contract to Legal is the appropriate answer. Hiring someone else because of that is not. The idea of insisting that people sign an unmodified contract is reprehensible. If you can defend that, you simply haven't seen very many contracts. People put all sorts of stupid things in them, ranging from the unenforceable to the illegal, and just as often leave out crucial details (e.g. severability).
Turning down an otherwise qualified candidate because they happen to have a clue about contract law is a poor business practice -- the same goes for writing your own contracts. The appropriate thing to do when presented with a contract you don't like, is to give it to your lawyer so that he can make the appropriate changes. In point of fact, it's probably a good idea to have a lawyer check out any employment contract whether or not you like it.
Lawyers draft contracts to protect their clients. Essentially that means screwing over the other guy, to whatever degree they can get away with. Lumpy is perfectly correct: only a fool signs a contract as written, because it is guaranteed to have been written with someone else's interests in mind.
If all of the above is unconvincing, working freelance will quickly cure you of any inhibitions regarding modifying contracts.
I don't get the impression that you comprehend a word of what I wrote. "Sometimes it's okay to kill people" is not compatible with moral absolutism. Contrast the statement, "Killing is always wrong," or even better, "Abortion is always wrong."
The only way that you can make the statement "Murder is always wrong" is to use a circular definition, e.g. "Murder is wrongful killing."
You're welcome to be condescending but you should work on your reading comprehension. The definition of right and wrong are not being debated. You happened to pick a terrible example -- something that you do not in fact have an absolutist opinion on. In point of fact, most people don't have an absolutist opinion on it, even if they think they do. Your statements ("Depends," "...we're not sure...," "...have to make a judgement...") are about as far from absolute as you can get. So, thank you for proving my point. I recommend the wikipedia articles on moral absolutism, moral relativism, and moral universalism, so that you don't end up looking foolish when talking about what you believe in.
Moral absolutes aren't. You can't just say "Murder is wrong!" as an absolute statement and then quibble about what the definition of murder is. I mean, you can, if you want to tell yourself that you have some sort of superior moral character, but you're still engaged in moral relativism. Logically "Some killing is not murder" is equivalent to "Some killing is not wrong". You can feel free to take refuge in the tautological definition of murder ("wrongful killing") but otherwise you're pretty much hoist on your own petard here.
That is, sometimes it's okay to kill people, sometimes it isn't, and there's no reason to exclude broad categories of human knowledge (e.g. science, statistics, legal theory, ethics) when making decisions on the matter.
The Moblin alpha and beta releases booted in about 5 seconds, on a single-core Atom netbook, to a full linux desktop.
The alpha was mostly Fedora packages, the beta had a surprisingly sensible UI for the class of devices it was targeting, and would have done well for tablets, I think.
I wish that I knew how to make my computers do that again. You may talk about instant booting from ROM, but show me any combination of modern hardware and software that approaches the speeds of Moblin.
Not fixing a bug for a long time is not a "stable target". It's a failure to fix a bug -- which is exactly what TFA is talking about. You are not holding these things to the same standard.
Relying on buggy behavior is compounding the problem.
A "different" box model. Yeesh. No, I'm sure you're right about this. IE is the best browser. When it fails to fix bugs it's "stable". When it doesn't comply with the specs that's okay because those aren't the "important" features. If other browsers don't fix bugs, they're buggy and bad, and if they do they're a "moving target". When you have an inconsistent argument, your opponent must be reading it wrong.
Keep fighting the good fight, I suppose. One day we'll surely all see the light.
You blame Firefox and Chrome for being buggy. You admit that IE has bugs and target it exclusively. This is somehow not hypocritical. What does it matter that the bugs are "well known" ? Isn't that the same thing as FF and Chrome not fixing bugs for years?
Beyond that you're bitching that "I shouldn't have to support Chrome for free!" So don't, and quit bitching about it. If one of the people employing you wants Firefox support they can pay for it. Or if they feel that's unreasonable they can find someone else. Since when does being a professional not mean charging for your time?
I don't know how long you've been in the business, but if your big beef is font rendering, you don't remember very far back. IE6 had a broken box model, and you had to choose between using IE-only DX-provided effects or CSS.
Your bitching is hyperbolic, hypocritical, and counterproductive. Quit while you're ahead.
What world do you live in that non-OSS software is bug-free? Do you get a free pass if you don't comply to the specifications, but don't have bugs in how you do that?
What, besides your own prejudice, justifies supporting a browser that you admit has some serious bugs, and also does not properly implement the web standards?
I don't see any facts or evidence in your post -- that would presumably detract from your trolling. However, there's a simple fix for this and all other issues with browser compatibility: charge for it. Charge for compatibility beyond IE, and charge for any time spent submitting bug reports. No one is expecting you to work for free -- most of those OSS developers aren't working for free, either. Of course, if you need a reason to bitch about something more than you need money, you can keep doing what you're doing...
Governments are not businesses. They print money, they don't trade for it. Citizens might be analogous to a "shareholder" but in point of fact it's an entirely different legal arrangement. You might consider it a natural monopoly on the use of force, but citizens can't exactly cancel their subscription to that.
Governments are bad enough without trying to drag in inapplicable concepts to mismanage them with. They are not businesses, they should not be run as businesses, and electing someone to the Presidency because of his business acumen is like appointing them to head General Motors because they're a good auto mechanic.
There can be no connection between open source and the end of programming as a profession. The open-source hobbyist programmer is substantially in the minority. As far as can be measured, most open-source development is paid. For large projects this is invariably the case.
The necessity for the existence of the software vendor, or the necessity to preserve any rights for such, has not been established. You can talk all you want about how good it is for the customer, but you're not optimizing with their profit in mind, so let's be honest about that. Also, distributing the source and expecting that people will not copy it is wildly optimistic.
It must be a bizarre world you live in. Selling copies of software is almost quaint.
The business model of selling copies of software is becoming scarce. You're essentially arguing that this is solely due to licensing. I see it more as a market adjustment where we're collectively deciding to treat non-scarce commodities realistically.
SaaS is the same pig (selling copies) in a different blanket, so you should adjust your rant accordingly. The three pillars of the open source model are selling services, donations, and advertising. You might add a fourth in monetizing big data, but that pretty much tends to go hand in hand with advertising anyway.
The idea that people will stop writing software if they can't sell it is ludicrous. Taken to a logical extreme, you're implying that if you could not sell MS Office, people would not need word processors. If there is one concept that needs to be forever laid to rest as an argument, it is that people will stop doing something intrinsic to human nature because of some external phenomenon. People will stop writing code, and needing to have code written, just after we create the last piece of art, and sell it as the last business transaction. Approximately never, in other words.
Now, as far as your proposed licensing goes, since you haven't stipulated terms under which the binaries can change hands, they can't. Simply providing software to someone does not grant them the right to redistribute it. You must explicitly grant someone the right to modify and redistribute copyrighted works, or abrogate these rights by making the work public domain. Whether or not they have source code is pointless if they have no legal right to do anything but read it. So, as stated, your scheme is not compatible with the GPL and doesn't fall into any meaningful category of "free software".
Further, unless you've made something public domain, there's no way for downstream vendors to change the terms under which the code is distributed. If you have made it public domain, your ability to enforce control is null and void.
In summary, your licensing as written is a great way for customers to get sued for copyright infringement. Fixing that flaw either makes it public domain or mostly identical to some sort of copyleft license. To summarize the summary, you may want to read more about copyrights and basic economics. To summarize the summary of the summary, people are a problem.*
Seen it before. There is evidence to suggest that Einstein's theories are incomplete. He was not wrong in the sense that rocks will suddenly fly upwards. No source of "free" energy exists. Go troll elsewhere.
Slashdot Mirror
To expand on a couple points:
Some distros make it more or less easy to install rights-restricted software, like the stuff you need to play mp3s or DVDs. Neither Fedora/RHEL nor Debian allow nonfree software in their repos, but it's generally a fairly painless process to add a repo that does.
Ubuntu will, IIRC, ask you during the install process if you want to install such things, and Linux Mint comes with the media codecs by default. For other distributions you should research this issue.
Fedora and Ubuntu are the "big" distros, more or less, although Mageia seems to be climbing up DistroWatch lately. I had written off that project as dead when its Corporate Overlord bit the dust, but it's probably worth checking out. I hope I may say with enough accuracy that it is of similar quality to OpenSuse.
Fedora and Ubuntu have the biggest corporate backing and are likely to represent the most polished experiences. Ubuntu has its own way of doing things, most notably they have implemented at least two desktop environments (Unity and UNR) and their own startup process. Startup tends to be one of those big differences between distributions, but it's something you can safely ignore as a n00b user.
Fedora and Ubuntu use incompatible packaging systems, which tends to be irrelevant for a couple of reasons that aren't worth going over here. Generally you should figure that [a] any distro that is described as being derived from any other distro is package-compatible, and [b] it's very uncommon to need to install a package outside of your distribution's package management tools. We don't download software off websites, pretty much everything that you would ever want to install comes in the box.
It's hard to come up with too many more big important differences between these things, really. Desktop environments make a pretty big difference. Distros, not so much, especially among the big players.
Oh, and I forgot to mention. If you ever want to give yourself a real education in Linux, try Linux From Scratch. You'll probably even survive the experience. By contrast, slackware will be a friendly and trivial introduction, and Gentoo... ...sorry, my Gentoo joke is still compiling :(
It's much easier for you to specify your needs as there are hundreds of distros and packages that can be combined. To a first approximation pretty much all linux packages are available for all distributions.
Beyond that, most linux distributions are based off some other distribution. The description of Kubuntu as "Ubuntu, but with the KDE desktop environment" is perfectly descriptive.
So what distinguishes one distro from another? Besides what comes installed by default, the most significant difference is how those packages got there.
Debian is probably the distribution that the greatest number of other distributions are based on. It has a very very long testing cycle; it takes packages years to get into Debian's stable branch. Ubuntu is based on Debian unstable, and a shit-ton of things are based on Ubuntu, including Linux Mint.
Red Hat produces the next biggest family of linuxes. Fedora and Red Hat Enterprise Linux are more or less analogous to Debian unstable and stable, respectively, but I don't think very many people are dumb enough to try and base a distro on Fedora. CentOS is RHEL with the logo removed, and Scientific Linux is also based on RHEL.
Next up we have Gentoo, Arch, Slackware, and Suse.
I was going to put a joke about Gentoo here, but it's taking a while to compile. Gentoo is a rolling-release distro where most of the packages that you use are compiled on and for your machine. You mention gcc, this is related, but you will probably not ever use it directly. Compiling packages yourself can make them run faster, but it can take a long time.
Arch is a well-documented, rolling release distribution. I'm not sure what else to say about them honestly, but "well-documented" is one of the highest compliments I'm aware of.
Slackware is the oldest and most "unixy" of the distributions. It uses an old bootloader, old unix-style boot scripts, and by default boots to a text terminal. You should use Slackware if you want to retreat into a cave for five years, to emerge with a profound knowledge of unix, a full beard, and a solid opinion on whether emacs or vi(m) is the best text editor. I'm pretty sure these things are highly marketable. No, really.
Suse hasn't failed yet. The last time I checked, they had a wonderful, polished experience, and great admin/configuration tools. I have no idea why they don't have more users, except that there's already a shit-ton of options.
It's probably fair to say that Debian stable, RHEL, and any derivatives will have the longest testing cycle, and fewest updates in any given span of time. There are many more distributions for more specialized purposes, such as BackTrack for pen testing, Puppy for small installations, Bodhi for those seeking Enlightenment. You may have to figure out what you need on your own there.
Whew! Let's take a break for a minute.
All right. So with all that in mind, you can install, as previously mentioned, pretty much all the same stuff on any and every distro.
Here is a guide on desktop environments. If you're a n00b, you're probably going to want one of those.
We also have another guide for more experienced users, or those on resource-constrained systems, that covers some of the more popular window managers. Because sometimes 2GB of gnome libraries gets a bit old. For the truly adventurous, this post covers 30 Window Managers in 30 Days.
Honestly, there's really a pretty limited amount of advice that one can give about using any particular distro. They're all substantially similar. Without any specific information about what you want to use, a recommendation becomes, well, exactly what you were complaining about. "Use XYZ bec
I like Debian. Linux Mint Debian Edition is a good option, although I am using Crunchbang on my netbook. The latter is based on Debian stable.
Stability is the kind of virtue that you appreciate most in its absence. After an enthusiastic period of Fedora and Ubuntu use, I from time to time experienced issues with packages and drivers breaking on updates. These were usually resolvable, and forced a certain amount of CLI-foo on me, but there's only so many times one wants to wrangle with things that worked just fine yesterday.
Stability means having outdated versions of packages; you miss out on the new features as well as the new bugs. However, it's also pretty trivial to install packages from unstable if you really need them, and if all else fails you can compile from source (which is usually a painless process).
Ubuntu was certainly far less buggy than Fedora, and I certainly don't mind all you guys being Debian beta testers ;) but my choice of OS is going to be heavily informed by whichever one has the longest testing cycle.
Ted Stevens eviscerated Ted Stevens.
He should have stuck to bringing home pork. He didn't get to be the longest-serving Republican senator because he was bad at his job. Sarah Palin on the other hand... ...yeah okay maybe there's a pattern here.
Indulge me in a little hyperbole: for a friend of mine, hacking AT&T was a death sentence.
Lance Moore was involved with LulzSec, foolishly no doubt. As an AT&T technician of some sort, he acquired and subsequently distributed some internal corporate documents. The Justice department is liable to be a more accurate source of the specific complaints. He was caught. The FBI seized its opportunity to bring the hammer down. I've seen various figures given for the amount of jail time he was facing; somewhere between five and thirty. He was found dead by his own hand on February 24 of last year. His crime has by now likely been forgotten by all that were involved with it.
Sixteen other people were arrested the same day that he was arrested. I don't know their stories. The reader may judge whether justice was served.
$6k a year is doable. $16k a year would be quite pleasant. I would avoid the capital or other large cities. Actually getting a work permit or visa to either country is difficult to impossible, but I know people in both countries who have been there for decades on a tourist visa. Do note, this tends to limit your options for local employment; it's far better to work online.
There's essentially no native culture (or cuisine) in either place, "post-colonial" about sums it up. The police are nice enough but underpaid, the laws are enforced relatively arbitrarily and generally not in favor of extranjeros. If you're running a business, [a] congratulations for getting through the bureaucracy to accomplish this, and [b] you may from time to time expect to have laws about licenses and restrictions enforced against you that your (Tico) competition does not. I'm not sure whether I can really say that corruption was common, but it's probably fair to say that people were understanding about dealing with the laws and regulations -- or avoiding that, if necessary. I don't really consider this a bad thing, but if you have the expectation that the rule of law is going to be universally or rigidly applied, you may be disappointed.
The weather is beautiful, it's not terribly expensive to get to and from either country (at least, from the US), English is spoken by a good percentage of the population, utilities are cheap and reliable, health care is extremely affordable (medical tourism is common), internet is not that fast but widely available, and of course, knowledgeable tech workers are in high demand. In Costa Rica the beer is not good and relatively expensive, in Panama you can get two beers for $1. Computers are available, but expensive. It's probably going to be a good idea to buy in the US and work out a way to get it. I've heard both good and bad things about the mail system; I'd call it generally reliable, but the paranoid might want to find other means of receiving packages. If you end up going back and forth to the states a lot, you can make good money on the side bringing electronics back with you.
Panama is by far the cheaper of the two countries, you would probably be able to get by on less than $6k annually. I didn't like it quite as much because, at least in the places I frequented, cocaine was both common and extremely cheap there. That's fine for those who like that sort of thing, but generally I don't think it does much good for the community. Drug laws in both countries are sparingly enforced.
Roads are generally better in Panama; the country has a lot more money due to that whole canal thing. I can't recommend driving in Panama City, or anywhere in Costa Rica. Cars are absurdly expensive, and paradoxically people don't care about the lines on the road, the blinky things above them, the relative speed and velocity of other vehicles, or pedestrians.
Fun Facts: there are no addresses in Costa Rica. There are no roads connecting Central America with South America.
[a] I don't give two shits about Windows Phone.
[b] Add me to your list.
Thank you for posting this information. The article, as you have ascertained, is worthless and false. The same may be said of Microsoft.
Religion can be rational. If you take the axiom, "God exists," and develop it, you can have a perfectly sound (to the limits of Godel) logical system.
You can also do this with the axiom "God does not exist," or "Lizard Men rule the world," or any other concept.
The problem with rationality is that, absent empirical claims/testing, you can prove pretty much anything. The problem with empiricism is that "truth" becomes limited by your ability to measure things, i.e. everything you know is a little bit wrong.
Let me know whether rationality or empiricism has achieved the greatest result.
The only authoritative reference to how a system functions is the source code. If you don't have the source, at some point you will encounter the problem that is unsolvable without it. Unless of course the original programmer wrote about everything that could ever be done with his software.
With source code, all things are possible. Not necessarily pleasant or easy, but possible.
As long as we're characterizing each other, let's be accurate: I have never worked in an enterprise environment. Lately I've been working freelance. It tends to involve wearing a lot of hats: developer, sysadmin, marketing, legal. I can rely on my own skills at this stuff until a certain point, after which it's safer to turn over to a professional. Why? Because I don't have the hubris to think that I am expert enough to make decisions in areas that I have not studied. Clearly there we differ.
As a developer, I don't think "what would I like to use?" very often. Instead I ask myself "What is the best practice?" or "What is the most professional solution?" or "What is best in the long run?" or "What is the next guy that comes across this going to want to work with?"
You obviously are asking none of these questions. It's an amazing ignorance you're devoting yourself to. "If he was so clever, he'd be a developer" Truly an incredible statement, and the end of all meaningful discussion on the matter.
The point was not enterprise, but professionalism. Let's not get into semantic arguments.
There is in fact a couple issues with developing professionally on OSX. For one, you clearly have never used OSX Server, or you would not recommend it. Second, as I mentioned, development (as opposed to staging or production) is fine to do on OSX but this should not have anything to do with colocation. Lastly, packages are developed for Mac OSX as an afterthought. This is compounded by the lack of a built-in package manager. It's not necessarily the case that you will have more issues doing dev work in a non-standard environment, but you'll have different issues, which is just as bad in terms of finding good information about them.
All that aside, you're acting like this is a simple matter of preference. The big bad sysadmin could just choose to use OSX instead, and that's just as good, and you are just as qualified as he is to make that choice. Again, this is more like the choice between (e.g.) Photoshop and GIMP. One is a professional, industry-standard tool. The other is not. There may be some circumstances where it's okay not to use the industry standard tool, but you do not know enough to make that decision. It's not a costless decision, and it's not arbitrary.
I'm sure that you know all this, and have a really good business case that you've somehow just forgotten to explain. You would never make capricious decisions about critical infrastructure.
No. If you want your own personal copy of wordpress running on OSX, do that. Install XAMPP, wordpress, source control. It takes all of ten minutes -- less if you script it.
You do not need a publicly-accessible server to do development on. You sure as shit don't need colo.
The argument that people should use what they're comfortable with is nice, but not part of an enterprise server solution. Let's take an analogy here: I am a pretty decent cook. Given sufficient time and a minimum of kitchen gear, I can produce a tasty dish (or two or three). I am completely unqualified to be in a restaurant kitchen. I don't even know what I don't know about that profession.
You say you have a tasty recipe that you've whipped up with your Mac. You tell me that you want me and (24 x 5U) of my friends to make this recipe for all your internet customers. This will work, to a degree, but is unlikely to scale very well. Also, even though your home-cooked meal is tasty and does the same thing as that restaurant meal, you and that restaurant chef are not actually in the same business. In order to become a part of that business, you need to [a] study more, and [b] use the same professional tools that everyone else uses. Sometimes the only reason things are widely used is because they're widely used, but more often there's a reason that impacts the quality of the product.
It's fine to not be in the professional cooking business. But saying "I don't need a professional kitchen/chef" based on this experience is foolhardy. You don't know what you don't know.
The most popular hosting is Godaddy, with their $10 / month hosting account. (35% of sites are Godaddy sites.) The sites with hosting budgets of around $10-$50 month make up 95% of all sites.
As a web developer I may say categorically, fuck them. If you put a site on the web, it is your responsibility to make sure that it is secure. If you are not able to do that to a professional standard, you should not do it. In point of fact, there is a need for a licensing organization to prevent amateurs from practicing web development. The problem isn't that this website is exposing poor security practices, it's that it's not promoting professionalism.
We have passed the point where it's okay for the layman to host a site. Even if you're not collecting information about your users, you can still be attacked or used as an attack vector. The era of democratization of the web is over.
Passing the contract to Legal is the appropriate answer. Hiring someone else because of that is not. The idea of insisting that people sign an unmodified contract is reprehensible. If you can defend that, you simply haven't seen very many contracts. People put all sorts of stupid things in them, ranging from the unenforceable to the illegal, and just as often leave out crucial details (e.g. severability).
Turning down an otherwise qualified candidate because they happen to have a clue about contract law is a poor business practice -- the same goes for writing your own contracts. The appropriate thing to do when presented with a contract you don't like, is to give it to your lawyer so that he can make the appropriate changes. In point of fact, it's probably a good idea to have a lawyer check out any employment contract whether or not you like it.
Lawyers draft contracts to protect their clients. Essentially that means screwing over the other guy, to whatever degree they can get away with. Lumpy is perfectly correct: only a fool signs a contract as written, because it is guaranteed to have been written with someone else's interests in mind.
If all of the above is unconvincing, working freelance will quickly cure you of any inhibitions regarding modifying contracts.
I don't get the impression that you comprehend a word of what I wrote. "Sometimes it's okay to kill people" is not compatible with moral absolutism. Contrast the statement, "Killing is always wrong," or even better, "Abortion is always wrong."
The only way that you can make the statement "Murder is always wrong" is to use a circular definition, e.g. "Murder is wrongful killing."
You're welcome to be condescending but you should work on your reading comprehension. The definition of right and wrong are not being debated. You happened to pick a terrible example -- something that you do not in fact have an absolutist opinion on. In point of fact, most people don't have an absolutist opinion on it, even if they think they do. Your statements ("Depends," "...we're not sure...," "...have to make a judgement...") are about as far from absolute as you can get. So, thank you for proving my point. I recommend the wikipedia articles on moral absolutism, moral relativism, and moral universalism, so that you don't end up looking foolish when talking about what you believe in.
Moral absolutes aren't.
You can't just say "Murder is wrong!" as an absolute statement and then quibble about what the definition of murder is. I mean, you can, if you want to tell yourself that you have some sort of superior moral character, but you're still engaged in moral relativism. Logically "Some killing is not murder" is equivalent to "Some killing is not wrong". You can feel free to take refuge in the tautological definition of murder ("wrongful killing") but otherwise you're pretty much hoist on your own petard here.
That is, sometimes it's okay to kill people, sometimes it isn't, and there's no reason to exclude broad categories of human knowledge (e.g. science, statistics, legal theory, ethics) when making decisions on the matter.
The Moblin alpha and beta releases booted in about 5 seconds, on a single-core Atom netbook, to a full linux desktop.
The alpha was mostly Fedora packages, the beta had a surprisingly sensible UI for the class of devices it was targeting, and would have done well for tablets, I think.
I wish that I knew how to make my computers do that again. You may talk about instant booting from ROM, but show me any combination of modern hardware and software that approaches the speeds of Moblin.
"It's not about booting faster, it's about booting in five seconds." Sigh...
Not fixing a bug for a long time is not a "stable target". It's a failure to fix a bug -- which is exactly what TFA is talking about. You are not holding these things to the same standard.
Relying on buggy behavior is compounding the problem.
A "different" box model. Yeesh. No, I'm sure you're right about this. IE is the best browser. When it fails to fix bugs it's "stable". When it doesn't comply with the specs that's okay because those aren't the "important" features. If other browsers don't fix bugs, they're buggy and bad, and if they do they're a "moving target". When you have an inconsistent argument, your opponent must be reading it wrong.
Keep fighting the good fight, I suppose. One day we'll surely all see the light.
You blame Firefox and Chrome for being buggy. You admit that IE has bugs and target it exclusively. This is somehow not hypocritical. What does it matter that the bugs are "well known" ? Isn't that the same thing as FF and Chrome not fixing bugs for years?
Beyond that you're bitching that "I shouldn't have to support Chrome for free!" So don't, and quit bitching about it. If one of the people employing you wants Firefox support they can pay for it. Or if they feel that's unreasonable they can find someone else. Since when does being a professional not mean charging for your time?
I don't know how long you've been in the business, but if your big beef is font rendering, you don't remember very far back. IE6 had a broken box model, and you had to choose between using IE-only DX-provided effects or CSS.
Your bitching is hyperbolic, hypocritical, and counterproductive. Quit while you're ahead.
What world do you live in that non-OSS software is bug-free? Do you get a free pass if you don't comply to the specifications, but don't have bugs in how you do that?
What, besides your own prejudice, justifies supporting a browser that you admit has some serious bugs, and also does not properly implement the web standards?
I don't see any facts or evidence in your post -- that would presumably detract from your trolling. However, there's a simple fix for this and all other issues with browser compatibility: charge for it. Charge for compatibility beyond IE, and charge for any time spent submitting bug reports. No one is expecting you to work for free -- most of those OSS developers aren't working for free, either. Of course, if you need a reason to bitch about something more than you need money, you can keep doing what you're doing...
Governments are not businesses. They print money, they don't trade for it. Citizens might be analogous to a "shareholder" but in point of fact it's an entirely different legal arrangement. You might consider it a natural monopoly on the use of force, but citizens can't exactly cancel their subscription to that.
Governments are bad enough without trying to drag in inapplicable concepts to mismanage them with. They are not businesses, they should not be run as businesses, and electing someone to the Presidency because of his business acumen is like appointing them to head General Motors because they're a good auto mechanic.
There can be no connection between open source and the end of programming as a profession. The open-source hobbyist programmer is substantially in the minority. As far as can be measured, most open-source development is paid. For large projects this is invariably the case.
The necessity for the existence of the software vendor, or the necessity to preserve any rights for such, has not been established. You can talk all you want about how good it is for the customer, but you're not optimizing with their profit in mind, so let's be honest about that. Also, distributing the source and expecting that people will not copy it is wildly optimistic.
It must be a bizarre world you live in. Selling copies of software is almost quaint.
Ad hominem. Next argument, please.
Seriously, that Monty Python argument was more entertaining than this.
The business model of selling copies of software is becoming scarce. You're essentially arguing that this is solely due to licensing. I see it more as a market adjustment where we're collectively deciding to treat non-scarce commodities realistically.
SaaS is the same pig (selling copies) in a different blanket, so you should adjust your rant accordingly. The three pillars of the open source model are selling services, donations, and advertising. You might add a fourth in monetizing big data, but that pretty much tends to go hand in hand with advertising anyway.
The idea that people will stop writing software if they can't sell it is ludicrous. Taken to a logical extreme, you're implying that if you could not sell MS Office, people would not need word processors. If there is one concept that needs to be forever laid to rest as an argument, it is that people will stop doing something intrinsic to human nature because of some external phenomenon. People will stop writing code, and needing to have code written, just after we create the last piece of art, and sell it as the last business transaction. Approximately never, in other words.
Now, as far as your proposed licensing goes, since you haven't stipulated terms under which the binaries can change hands, they can't. Simply providing software to someone does not grant them the right to redistribute it. You must explicitly grant someone the right to modify and redistribute copyrighted works, or abrogate these rights by making the work public domain. Whether or not they have source code is pointless if they have no legal right to do anything but read it. So, as stated, your scheme is not compatible with the GPL and doesn't fall into any meaningful category of "free software".
Further, unless you've made something public domain, there's no way for downstream vendors to change the terms under which the code is distributed. If you have made it public domain, your ability to enforce control is null and void.
In summary, your licensing as written is a great way for customers to get sued for copyright infringement. Fixing that flaw either makes it public domain or mostly identical to some sort of copyleft license. To summarize the summary, you may want to read more about copyrights and basic economics. To summarize the summary of the summary, people are a problem.*
*with apologies to D.A.
Seen it before. There is evidence to suggest that Einstein's theories are incomplete. He was not wrong in the sense that rocks will suddenly fly upwards. No source of "free" energy exists. Go troll elsewhere.