This seems staggerly out of touch; are you aware that the nuclear weapons in question were built from technology the Clinton administration donated in exchange for not developing nuclear weapons?
With all due respect, you don't know this. The DPRK has simply said they possess nuclear weapons; they have further stated that they've had them for a long time. There was a scare about them having nukes in 1993; that was only a year into Clinton's first term. Who knows when or how they got them; let's not forget that they are (at least on the surface of it) on peaceable terms with China, another major nuclear power. They were previously on good terms with the Soviets. There are lots of places they could have gotten the tech from.
Clinton may have been responsible, but he just as easily could have had nothing to do with it. Your statement makes it sound as though there is authoritative evidence damning him, when in fact there is none.
Try not to spread mindless bullshit just because it seems likely to you and fits in with your politics. It advances no one's knowledge.
It's a good question, on the surface, but the problem is that you are looking to end users as the source of a software company's primary income. This is perhaps true in the gaming industry, but I see it as being less true in other industries.
Discounting the Macintosh user for a moment, the average Windows user is presented with a large amount of warez and other pirated software and very few seem to have any problems whatsoever installing cracked versions of software they didn't buy. I know essentially no one who bought MS Office, for example. Even Adobe Photoshop is generally only purchased by professionals.
However, sales of software remain profitable -- because the big purchasers of software are companies. They are much more motivated (see the BSA) to make sure that they have legally purchased software, as the ramifications are far greater, and the cost of software, which might be prohibitive for the end user, represents a smaller investment for a company, especially when offset against the fees incurred by unlicensed software discovered in an audit.
Macintosh users are a little bit different. This is probably partly a cultural thing, partly because Macs have classically been more expensive and higher quality than their PC counterparts (suggesting that people that purchase them aren't afraid to pay for quality) and because, due to their minority status, there just isn't as much in the way of easily accessible pirated software (although this wasn't the case at all back in the days of System 7).
Getting around to your point, I think that while some of what you say is correct -- Linux users are spoiled by high quality, free (both as in beer and as in speech) software, and so the average user is reticent to purchase possibly inferior proprietary solutions, and is even willing to make do with inferior free solutions if available.
But this is not the case of a company, and we see here the same situation as with Windows: it is in the company's best interest to buy Oracle, or say, Lotus Notes for Linux (if it existed), Photoshop, or Office. A hobbiest may be willing to risk OpenOffice mucking up his word document, but a company would fork out the cash for MS Office if they had the choice to.
This brings us to your one unfounded allegation -- the rest I think were relatively insightful -- that Linux distributions are equivalent to different versions of Mac OS or Windows.
Actually, a Linux "distribution" is just that: a bunch of software written by other people and bundled by a distributor. In most cases, the distributor conforms to certain norms with respect to filesystem structure, configuration file syntax, and the like. The reason for this is not so much because they are motivated by goodwill -- although this is perhaps also the case -- but rather because the software they bundle was not written by them, and its authors wrote it expecting the system to behave in a particular way. Because the software is open source, it is possible for them to hack it to conform to their own, special setup -- but this takes time and effort and needs to be redone everytime they introduce a new version of their software. It's just not worth it.
So, Oracle and IBM standardize on RedHat, for example; some people standardize on SuSE. But the differences between these two systems are cosmetic, really. I can get binaries designed for these systems to run on my Debian box without issues.
When a vendor says, "This software is designed for RHEL 3" or whatever, what they mean is that it was built and linked against the particular versions of libraries in that distribution's version; it may also mean that they depend on certain versions of utilities present in that distribution. Because all of these are Free Software, I can quite easily go and reproduce that environment, if I want, on any other system. It's really not a big deal at all.
That's why, despite the fact that most proprietary Linux software is released with RedHat in mind, I can find it repa
Honestly, are you just totally daft? Your (relatively) low UID suggests that you've been around for a while, but it apparently isn't an indicator of resourcefulness. You do realize that we have the web and google, right?
Or was asking for a source just a half-assed way to indicate your belief that the OP's story was apocryphal?
I put "bill gates ten years two operating systems windows linux" into Google and got this link as the first hit.
That's fewer characters typed than it took you to write your completely moronic post.
Someone on here was saying a few days ago that people on the internet that ask for a cite couldn't give a rat's ass about the veracity of the information they're calling into question. At the time, I thought he was just being overly cynical, but now I'm really starting to wonder.
This has been ten minutes of my already too-short life wasted. I sure as hell hope you appreciate the link that you could have gotten in ten seconds if you hadn't been so god cursed lazy.
Sure, stable is a bit behind for desktops. But while you and I may want the latest and the greatest, in a production environment, forced updates suck because breakage is likely to occur. Of course, you must update from time to time, but it makes everyone's life easier if it doesn't happen often. In this context, Debian Stable is perfect.
When I first started with Debian (around Hamm), the devs generally recommended stable for personal use. Nowadays, they usually don't bother. Stable's focus has shifted, as a result of the slow update cycle that handling so many architectures forces on them. As it turns out, this has been a very good thing.
Debian testing and even unstable are much more like other distributions, and as we're free to track them, I don't really see what the problem is. It used to be that unstable was pretty much broken most of the time and really an experts-only sort of distribution, but nowadays most of the total breakage is in experimental and even unstable is pretty stable, despite it having the latest and the greatest.
The Mag-Lev that connects Shanghai's Long Yang Road subway station with Pudong International Airport has a maximum speed of 431kmph (there is a spedometer in each car to advertise its cruising speed). This is not its actual maximum speed, but given the length of the track (I think it only goes about 40km) it doesn't make sense to accelerate further.
While the TGV has gone 500kmph in speed tests, it rarely passes 350 when actually carrying passengers. Having said that, the TGV is a much smoother ride than the Mag-Lev, but then it is also much more mature technology.
I do believe the Mag-Lev in China is the fastest train in the world, at the moment, in terms of actual speed achieved in regular use. Of course, a mag-lev design removes track friction and so it makes perfect sense that it should be faster than any rail-based alternative.
The Mag-Lev was designed by Germans, though, IIRC, so I'm not sure it's an example of Chinese innovation.
I'm not so sure. I'm thinking long term, understand. Not tomorrow or the next day, but years down the road.
The problem is that a browser is an incredibly complex piece of code, and what with the W3C making recommendations left and right, it isn't getting any simpler.
When you're a company, producing code costs money. Browsers, unlike say, operating systems, are already considered a commodity by the public. Whereas most people think that paying for an OS makes perfect sense, no one is ready to pay for a browser -- they're all free. So it follows that you can't really make any money selling a browser, unless you have some special niche (like Opera does with their mobile phone stuff).
The money you can make (a la Opera) is ok for a small to medium sized business, but not at all interesting from the perspective of a big player like MS.
Now, normally a counter like "the end user doesn't care about these things" would be right on the money, but consider who makes the web that people look at: not Joe Sixpack. Not the unwashed masses. Web developers. Designers.
These people are aware of the W3C, even if the end-consumer is not. These people are aware the CSS3 recommendations, SVG, MathML, XML namespaces and an integrated DOM, etc -- and they are excited about these features. They want to use them to develop awesome websites, webapps. The possibilities are endless, and only one thing stands in the way: Internet Explorer.
This has generated a tremendous amount of animosity among web developers towards IE. They hate it. These people are not Free Software zealots; many of them use Windows exclusively. They don't like Firefox because it's Free -- they like it because it is delivering on the new technologies that promise to make the web an exciting, interesting place.
And they author the web. Don't underestimate their influence.
Firefox is in active development, and it is already much more capable than IE is -- by this I mean that it supports more of the standards, more technologies, and whatsmore, what isn't supported yet is in the process of being implemented by someone, somewhere. If there's a bug, it gets fixed.
The only thing that keeps IE in the game is marketshare.
When you consider that IE6 had it's last feature addition in 1998, it's actually quite an impressive browser: for all its quirks its support of standards isn't that bad. But its support reflects the cutting edge in 1998, not 2005. Time keeps marching on, too. 1998 is getting further and further away, the software older and older and older. Sure, corporate sites will consider IE6 the lowest common denominator for a long time, designing around its limitations -- but they aren't the ones that drive content on the web. Right now, there are already sites that are standards compliant and don't display as intended in IE -- and the authors make no bones about pointing it out. After a while, IE users will see that many such sites exist -- their numbers can only increase.
IE6 will die because MS doesn't care about it anymore. They aren't updating it. They don't support the standards that web devs want. Do you remember how long people supported NS4? IE6 will be the same. But that won't stop the web from moving forward.
Now, MS might change their minds and throw a bunch of money into IE7 or MSN or whatever their new browser is called, bring it up to desired standards compliance in order to keep control of the web. As long as they support open standards, I could care less. I don't mind hacking my CSS/XHTML to work in all the browsers around, as long as those are minor hacks.
But when SVG support starts showing up in official releases of Mozilla and Co, it's going to be hard to swallow IE. The likely work-around for this will be content negotiation, which will mean that when a Firefox browser visits my site they'll get lightweight SVG, and an IE user will be stuck with bulky bitmaps. They won't be able to resize
Actually, Sarge has been the testing release since Woody was released, which was 2 and half years ago I think.
Actually, "testing" and "unstable" always exist -- you can always track them. The name of testing changes each time there's a release -- so back when potato was the stable release, woody was the testing release (sarge was not the unstable release -- the unstable release is always called sid). When woody became stable, they started calling testing sarge. When sarge becomes stable, testing will get a new name, and so on ad infinitum.
So for example, I run testing on my laptop. That means that I run sarge, currently. But the day sarge becomes stable, I will actually be running whatever the new testing is called.
You don't consider locking someone up (possibly indefinitely), without accusing them and without trial, to be infringing on their right to life? Sure, it's not execution -- but who knows how long they'll be held? Who knows whether they're guilty or not, and if they are, of what?
I'll admit that I may have jumped on you prematurely -- I find the whole "us vs. the terrorists" mentality frustrating, and perhaps I wrongly assumed it of you -- but I think that stripping due process very much interferes with a person's right to live their lives as free individuals. From my perspective, locking me up forever isn't much different from killing me. In all likelyhood, the "terrorists" we're holding in Cuba aren't going to be there forever, but they certainly have been there a long time.
At any rate, at this point it's a semantic argument. It sounds as though we basically agree.
I realize that you're being deliberately obtuse, but the use of quotes around the term "terrorist" was clearly meant to give the term a certain uncertainty. Who and what constitutes a terrorist very much depends on your politics, and frankly, many people in the US these days seem to use that term a little bit too freely.
Your picking the extreme "suicide bomber" example says nothing about the OP's point; suicide bombers may be considred terrorists, but not all the people that are considered terrorists are suicide bombers, or even necessarily guilty of a crime anyone can prove in a court of law (what a hassle habeus corpus is -- let's just get rid of it), or even accused of a crime.
The fact that this is not just says nothing about what ought to be done with real criminals. It simply points out that just because the scare term "terrorist" is applied to a person does not mean that they should immediately have all their rights stripped and be sent off to an offshore prison facility where the pesky rights our great nation affords to people in its territory are null and void.
I don't speak for the OP, of course, but I believe that was his point. Whatsmore, I think you understood that full well.
While you may be right, I feel obligated to point out that your "vendor shipments" data is pretty much completely useless when one is discussing a Free OS like Linux or the BSDs, because, put simply, there is no simple indicator for the amount of instances deployed.
With Mac OS X, simply looking at how many units Apple shipped will tell you how many people are adopting Mac OS X. It's quite an accurate indicator and it can be trusted.
With Linux, there are many commercial distributions (Red Hat, SuSE, etc) as well a number of wildly popular free ones (Debian, Slackware). You can buy servers with these pre-installed, but I would be surprised if most Linux deployments were obtained this way. As there are no restrictions on redistribution of GPL'd code (unless changes have been made, obviously, in which case those changes must be made available), how many instances of say, Red Hat Linux sold will tell you nothing whatever about how many systems actually run the OS.
Even with Windows, it's difficult to say; Windows comes pre-installed on most systems whether you want it or not. While I wouldn't expect that a startlingly large percentage of Windows machines have their OS replaced immediately after being bought, it is undeniable that a number of them do (my laptop, for example, came with Windows XP, but now runs Debian). Again, I'm not saying we're a huge percentage -- but we are a percentage.
Really, Mac OS X is pretty much the only OS on the market today whose deployment can be accurately measured with unit shipments (although I don't doubt that there are some people out there that would buy PPC hardware and run Linux or *BSD -- I would, no offense to OS X).
So basically, Apple being the world's largest vendor of UNIX means only that Apple sells more UNIX than any other company -- which means nothing with respect to Linux, and IIRC, at least currently, there are more Linux boxes than there are Mac OS X boxes, by a long shot (although I wouldn't be surprised, given Apple's rising popularity, if this were to change in the near future).
Anyway, I'm not disparaging the Macintosh, or your facts, just pointing out that your argument is pretty much non sequitur in this context. If you were comparing Apple to say, Sun, you'd be much more on the mark.
Exactly. In almost exact analogy to the doc format, the GIMP is capable of opening PSD documents, despite the format being undocumented and tempermental -- although it doesn't always do a good job of it.
Photoshop, on the other hand, cannot open XCF documents -- despite the fact that much less work is required to do so, as the document format is open and documented.
Rather like Abiword/OOo, which can both open.doc files with varying degrees of success, while MS Office is incapable of opening the Abiword/OOo native formats, despite those being open and documented.
Really, I don't think any of this says anything about lack of interoperability on either side, even though on the face of it MS appears less interoperable with other formats. One, because it's hardly surprising that the dominant player wouldn't bother adding support for his competitor's formats, no matter how easy that support might be, and two, because Bill Gates wasn't talking about Linux/Windows interoperability (at least, in my reading) but rather Windows/Windows interoperability and Linux/Linux interoperability.
In that respect, it's entirely possible that he has a point -- but it's a point that hasn't been neglected by the Linux community, even if it persists. freedesktop.org has published a large number of interoperability standards aimed at getting things to behave in a standard way between widget sets, and current incarnations of GNOME/KDE apparently interoperate rather well (I wouldn't know, I don't use em).
With respect to interapplication interoperability (as in, Office suite interoperability), the recently discussed Open Document Format has already gained fairly widespread acceptance in the word processing group of free programs, and is expected to replace their native formats as default in subsequent releases, guaranteeing that OO and Abiword, for example, do interoperate.
I'm not particularly concerned about that stuff though, as I'm lucky enough to not have to deal with word processors too often.
Still, even if Bill's rant was intended (on his part) as FUD, we should always take a good look at ourselves and see whether his criticisms have merit. My take, in this case, is that he isn't entirely wrong, but he's pointed this issue out when the solution is already well underway. It seems we're ahead of him in that respect.
If the license holder can prevent you from doing something with the software, then it is technically proprietary software.
Don't you mean "copyright holder" and not "license holder"? The license holder, by definition, can't prevent you from doing anything -- the license only specifies under what terms the non-copyright holder may distribute the software (usually, none).
In defense of RMS, he has often stated that the copyleft legal hack would be unnecessary in a world without copyright. You understand that the GPL grants rights -- it does not limit them. Copyright law in the US (and in all nations that are Berne convention signatories) default to an "all rights reserved" status, where only the copyright holder has any distribution rights at all as long as the copyright is valid (it usually expires after some time, but how long depends on the material being copyrighted).
What this means, practically, is that you are by default "prevented" from doing anything (related to distribution, ie, copying) with software that you don't explicitly hold the copyright to.
You are correct that the GPL does not grant all rights; there are some that it reserves. Specifically, it does not grant the right to distribute a binary form of the software without the source code, and aditionally requires that any derivative works also be available under the GPL.
In this sense, the BSD license (and cognates) grant more rights, but even they do not grant all rights.
The only "license" that grants all rights is not a license at all: public domain. On the face of it, public domain code may seem like a good idea, but in fact, it isn't. Why not? Because of liability. If I author code and contribute it to the public domain, I am not making distribution of the code contigent on the receiver abandoning his or her right to take me to court should the code not behave as expected. This is why every license (including BSD) have a clause about "MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE."
Without this clause (a clause that you legally cannot enforce with PD, because you have voluntarily forfeited copyright, and thus there is no licensing or legal agreement) you are liable, implicity, if someone should use your code and have it not work as they expected it to.
Given all the legal explanations I've just given you, you should understand that the statement you've made makes no sense. It makes no sense because licenses, by their definition, cannot "prevent" you from doing anything, as no license by definition means "all rights reserved to the copyright holder and none for anyone else."
The exception, as I stated, is Public Domain, in which you renounce your copyright. Not a good idea from a legal liability standpoint -- use BSD or MIT, it's almost equivalent.
Given this, your understanding of what "proprietary software" is is most certainly somewhat muddled, as no license grants all rights, for sound legal reasons. That would mean that all software is proprietary, which would make the term rather useless, wouldn't it?
At any rate, "proprietary software" is a somewhat subjective term -- I believe it lacks a formal definition. Normally, the implication is that it's software that isn't free, but what free software is will be different depending on whom you ask -- RMS, Theo De Raadt, and ESR all have different views on this question. None is objectively "right", so you'll have to find your own middle ground.
Hopefully this information proves useful in helping you find it.
I'm sure RMS appreciates your nod to him, but he actually doesn't want you to just randomly put GNU/ in front of every instance of Linux. In this case, you should actually just say Linux (and RMS would agree with me), because you aren't talking about GNU/Linux the OS, but Linux the kernel (which Linus wrote, or at least began). Linus did not write GNU/Linux -- most of that OS was written by other people, with most of the userland tools and libraries that make it function coming from the GNU project (hence the GNU/ monicker).
The GNU/Linux / Linux distinction is actually useful, because the OS and the kernel are distinctly different things. I usually just say Linux when it's clear which one I mean, but even if you don't appreciate RMS's point of view, you should understand that a kernel and an operating system are not the same thing.
You're describing the bootstrapping problem, which if I recall correctly is actually described on the GNU website, somewhere.
To answer your question, the GNU devs started out with proprietary operating systems -- primarily SunOS, I think -- and took advantage of the modularity of UNIX to replace one utility at a time. This is why the kernel was the last piece -- because most of what makes a UNIX system run actually resides in user space.
Attempts to create free versions of other OS types -- ReactOS comes to mind -- have a harder time following this example, because most other operating systems are not designed in such a modular way. So they start with the kernel.
There's the question of legality, and then there's the question of ethics. The GPL is designed to only restrict those rights that could be used to restrict the rights of other would be users; in other words, it takes the Libertarian "my right to swing my fist ends where your nose begins" mantra and applies it to software licenses.
The original BSD "advertising clause" does what you refer to -- force credit-giving at the license level -- at it was such a flop that even BSDers shun it. The legal ramifications of such a clause were far too cumbersome to warrant inclusion.
All of this is well and good, but there are a few things that we would all do well to remember:
1) RMS, like him or not, authored the GPL.
2) The existance of the GPL has created a sharing-friendly environment for the creation of free software. BSD is a great license, but most people don't want to see their code swallowed into a proprietary product (the fact that BSD people see no issue with this is great, but most people don't share their view).
3) With respect to the Linux kernel: Linus wrote the Linux kernel to run GNU software. Think about that. I don't mean, "he designed the kernel to run with a GNU userland", although that is also the case. If you read his early usenet posts, he wanted to have kernel that was UNIX-like enough that he could run GNU-utilities. It was his goal, so to speak, at that time.
4) Today, much of what makes your GNU/Linux system run is GNU software -- around 20%, IIRC, on average more than code from any other project (including the kernel itself, by a long shot). Of the programs and libraries that are required to even get your computer to boot (I'm not talking ls here), the percentage is much, much higher.
5) The Linux kernel, itself, can be compiled as intended with only one compiler: gcc. Further, it requires the binutils chain and GNU make to complete the build. These tools (with the exception of GNU make) are so stable and widely used that the BSDs, too, depend on them -- making the GNU project a necessary prerequisite for even BSD development. I don't doubt that had GNU not existed, the BSDs would have gone ahead and developed their own dev tools -- they still dream about doing it, and there are projects every now and then that aim to replace them -- but the fact remains, gcc and the binutils chain form the backbone of the world's free software. There is no equivalent to them. In fact, they're so good, that many proprietary UNIX vendors ship them instead of their own chains these days.
6) A GNU-less Linux kernel (leaving the tools required to build it out of the picture for a second) may exist on some random embedded systems (and even then, not necessarily, as glibc is typically used to interface with the kernel), but there is no server or desktop system that is capable of running without GNU tools. There is no other non-GNU software (with the exception of the kernel itself) that can make this claim. Lots of popular software (like X, for example) is an integral part of GNU/Linux -- but only GNU software can boast such deep integration into the system that you cannot even successfully boot without it.
To me, this seems like a good reason to give GNU credit. A lot of credit. Because the only reason we have free operating systems today is because of this project. Because RMS was so freakishly devoted to freedom that he went ahead and set out to build an entirely free OS, from scratch. GNU is also the only one project that has ever succeeded. No one else ever has, think about this. People may not use the HURD much -- it isn't stable, that's for sure -- but you can run the HURD with only GNU software. Every single other Free Software OS, including the BSDs, requires GNU software somewhere (although admittedly, if you're not keen on compiling anything, you may be able to use the system without GNU software -- but you couldn't develop it).
As someone who majored in pure math, I have to disagree vehemently. Computers don't belong in math classes, period, and overhead projectors (ack) aren't a whole lot better.
No, a good math class contains an engaging teacher, students with pencils and paper, and one of a blackboard or white board -- both of which have advantages and disadvantages for the teacher, but are mostly the same to the students. Models may be useful as well, if you're doing something low-dimensional that benefits from visualisation.
Computers can't do math -- all they can do, really, is computation. Which is not to say that they can't be useful, but to introduce them into a math class is to underscore computation in mathematics, when what is important is logic. Math is much more a subset of philosophy than science, a fact commonly misunderstood by non-math people. These non-math people unfortunately include administrators, who in the interest of "bringing computers into the classroom" are wont to force math teachers to use these new-fangled gizmos.
There is no substitute for explanation, for discourse, or for student/teacher bond. Any subject can be exciting and interesting to any student as long as the teacher is able to communicate his or her love of the subject in a way that the student can identify with. Looking at a computer screen in a classroom is tremendously dull, and while it might be manipulated effectively by the teacher, it cannot be manipulated effectively by a student that might want to jump in with his or her insight. Math is not a spectator sport -- student involvement is essential to proper understanding of the subject matter, especially at more advanced levels.
This, incidentally, is one of the reasons that I hate overhead projectors -- they project an 8x11 sized scrap of transparency onto a light surface. This might be great for the teacher, but when you want to bring a student up, it's much harder to find a place for him or her to find a place to write on that tiny scrap.
Further, overhead projectors encourage what is without a doubt the most sinful teaching style -- note preparation. Especially in math, which requires a great deal of thought -- a lubricated brain, so to speak -- boring lectures cannot be tolerated, if you expect your students to come away with an even cursory understanding of the material you present. A teacher that writes out all his notes on transparency is able to simply put up the notes, one sheet at a time, and comment. You know the style! God, just thinking about it makes me want to puke. It encourages a rut, discourages tangential conversations, proofs, and forays -- in other words, it's no better than just reading the textbook to the student, which he or she is quite capable of doing him/herself, thank you very much.
The thought of powerpoint is even worse -- all the fun of the overhead projector with even less editability.
Furthermore, the use of calculators is just silly. Counting is for computers. Proof is for mathematicians. Why on earth would you ever need to cope with numbers more complex than a few rather small natural numbers as coefficients? You might have a case in a trigonometry class but you said you were doing multivariate analysis.
Computers' place in math: a computer lab somewhere with Mathematica/Maple for the undergrads that haven't yet learned to divorce the abstract from the tangible. Certainly not in the classroom, except for very specific purposes (visualisation of recurrences, differential equations, vector fields, that sort of thing, in low-d situations).
doubt -> doubt (n) doubt + less -> without doubt (adj) doubt + less + ly -> describes an action or adjective which acts or describes something in a manner lacking doubt (adverb of manner, constructed from an adjective, as in "hungrily").
In this case, we take the adjective "informative" and use an adverb to describe how it is infomative; in this sense, I indicate that I do not doubt that it is informative -- in fact, the construction is inherently passive as it deemphasizes the agent doubting, so the implication is that "one" would not doubt that it is informative. In other words, we are speaking of the response that I/no one doubts is quite informative -- but we can use just three words: the doubtlessly informative response.
Got it?
Yeah, I know, I love grammar. Linguist geek and all.
I know he based it on another program, but lets be fair here: what makes fetchmail far, far better than any other similar program is its SMTP integration, which AFAIK popclient did not support. Of course, it wasn't ESR's idea, either (he admits this much) -- but he did write the code to do it. Not that this is complex in the slightest or anything.
Further, given the name popclient, I would presume that it probably only supported POP. Even if it didn't, fetchmail supports everything under the sun. But then, by ESR's own admission, he attempted to mimic Linus Torvalds in his management of the project -- something about proving his Cathedral vs Bazaar theory correct, or the like. Anyway, this means that he accepted a lot of patches and such from a lot of different people, so you could (if you wanted) make the argument that most of the code in fetchmail is not his. But then, that's true of Linus and Linux, too.
One thing is certain -- fetchmail is useful, and has been managed by him, at least. I'll give him credit for it. I don't give him credit for much else.
I believe his point was that RMS is extremely, extremely consistant. From your perspective, this may be taken to mean "he is consistantly a nutjob". This, however, stems from your disagreement with his first principles, not from the insight he deduces from them.
RMS has an opinion -- a very strong opinion -- that you (and many others) disagree with. He could be called a fanatic (and this is what the OP was talking about) because his opinion is, from the perspective of most people, "way the fuck out there". I take it this is your position, and it's a valid one.
ESR, on the other hand, is "reasonable" about certain things -- in fact, some of what he says is very insightful -- but sometimes he's just kooky, and by this I don't mean that he espouses a value or set of values that the mainstream finds odd, but rather, that he is at times painfully irrational and, well, beefheaded.
I've read pretty much everything on his website. To me, it seems as though he wants to be RMS. By this I mean he wants the community to respect him the way that RMS is respected (don't laugh now, whatever your personal feelings about him, he has a large following). He says, in numerous places, that RMS's creation of the GNU project -- the project that is essentially responsible, either directly or indirectly, for 80% of all the free/open source software we use today -- is actually nothing particularly special. That anyone could have done it. That in fact, he could have done it.
He says, again and again, that RMS was just "in the right place at the right time". He belittles RMS. Attacks him publicly, personally, but at the same time addresses him by his first name and claims that the two are or have been friends (something I'm in no position to affirm or deny). RMS, on the other hand, never says anything about ESR. He talks about the Open Source movement in a less than flattering way sometimes, but he almost always has a well-thought out reason for it, even if no one really cares what his reasons are.
I disagree with RMS sometimes, and I disagree with ESR sometimes. I agree with them both sometimes. RMS is fanatical about Free Software; but ESR is just a fanatic.
With respect, Russ, there are other Slashdot readers that would like to hear your views on this. While jbn-o may have (in true hacker tradition, might I add) come across a bit harshly with his point, his point nonetheless stands and those of us following this thread would appreciate a non-sarcastic response from you.
For what it's worth, I too am, as you say, in error, because I have the same apparent misconception of the difference between the Open Source movement and the Free Software movement as jbn-o does, and I'd wager that I'm not alone.
As this is a public forum, with many interested would-be contributors, I think taking your doubtlessly informative response to jbn-o's erroneous assessment private is somewhat at odds with your new position, whose duties presumably include acting as ambassador for OSI.
I'm genuinely curious and open-minded, understand. But at the moment, jbn-o has my attention, and you do not. I had hoped to see something more from you.
I'm not a huge fan of ESR, but in the interest of giving credit where credit is due, he is responsible for fetchmail, which is admittedly an extremely useful (if not particularly exciting) piece of software.
He also wrote an INTERCAL compiler for UNIX-like systems, I think. I think he also wrote a build configuration system (like menuconfig) for the Linux kernel which was ultimately rejected -- but I'll hold off on disparaging him for it, as I've never had anything accepted either, and submitting something on LKML is more than most people do.
Of course, ESR is best known for his gift of gab, not his contributed code. When he's speaking about something he actually has a cursory understanding of, he's usually capable of being quite insightful. Unfortunately, he's one of those intelligent people who believes that because his opinions are respected in one arena, he is automatically an expert in all arenae -- and he attempts frequently to back his esoteric points of view -- especially regarding homosexuality, racism, and the like -- with rather flawed logic.
Of course, ESR is only human, and just because he's been a public figure of some note doesn't mean that we should hold him to an unrealistically high standard.
He's just not particularly well rounded. Writes well, though.
Slashdot Mirror
You, sir, are an idiot.
With all due respect, you don't know this. The DPRK has simply said they possess nuclear weapons; they have further stated that they've had them for a long time. There was a scare about them having nukes in 1993; that was only a year into Clinton's first term. Who knows when or how they got them; let's not forget that they are (at least on the surface of it) on peaceable terms with China, another major nuclear power. They were previously on good terms with the Soviets. There are lots of places they could have gotten the tech from.
Clinton may have been responsible, but he just as easily could have had nothing to do with it. Your statement makes it sound as though there is authoritative evidence damning him, when in fact there is none.
Try not to spread mindless bullshit just because it seems likely to you and fits in with your politics. It advances no one's knowledge.
It's a good question, on the surface, but the problem is that you are looking to end users as the source of a software company's primary income. This is perhaps true in the gaming industry, but I see it as being less true in other industries.
Discounting the Macintosh user for a moment, the average Windows user is presented with a large amount of warez and other pirated software and very few seem to have any problems whatsoever installing cracked versions of software they didn't buy. I know essentially no one who bought MS Office, for example. Even Adobe Photoshop is generally only purchased by professionals.
However, sales of software remain profitable -- because the big purchasers of software are companies. They are much more motivated (see the BSA) to make sure that they have legally purchased software, as the ramifications are far greater, and the cost of software, which might be prohibitive for the end user, represents a smaller investment for a company, especially when offset against the fees incurred by unlicensed software discovered in an audit.
Macintosh users are a little bit different. This is probably partly a cultural thing, partly because Macs have classically been more expensive and higher quality than their PC counterparts (suggesting that people that purchase them aren't afraid to pay for quality) and because, due to their minority status, there just isn't as much in the way of easily accessible pirated software (although this wasn't the case at all back in the days of System 7).
Getting around to your point, I think that while some of what you say is correct -- Linux users are spoiled by high quality, free (both as in beer and as in speech) software, and so the average user is reticent to purchase possibly inferior proprietary solutions, and is even willing to make do with inferior free solutions if available.
But this is not the case of a company, and we see here the same situation as with Windows: it is in the company's best interest to buy Oracle, or say, Lotus Notes for Linux (if it existed), Photoshop, or Office. A hobbiest may be willing to risk OpenOffice mucking up his word document, but a company would fork out the cash for MS Office if they had the choice to.
This brings us to your one unfounded allegation -- the rest I think were relatively insightful -- that Linux distributions are equivalent to different versions of Mac OS or Windows.
Actually, a Linux "distribution" is just that: a bunch of software written by other people and bundled by a distributor. In most cases, the distributor conforms to certain norms with respect to filesystem structure, configuration file syntax, and the like. The reason for this is not so much because they are motivated by goodwill -- although this is perhaps also the case -- but rather because the software they bundle was not written by them, and its authors wrote it expecting the system to behave in a particular way. Because the software is open source, it is possible for them to hack it to conform to their own, special setup -- but this takes time and effort and needs to be redone everytime they introduce a new version of their software. It's just not worth it.
So, Oracle and IBM standardize on RedHat, for example; some people standardize on SuSE. But the differences between these two systems are cosmetic, really. I can get binaries designed for these systems to run on my Debian box without issues.
When a vendor says, "This software is designed for RHEL 3" or whatever, what they mean is that it was built and linked against the particular versions of libraries in that distribution's version; it may also mean that they depend on certain versions of utilities present in that distribution. Because all of these are Free Software, I can quite easily go and reproduce that environment, if I want, on any other system. It's really not a big deal at all.
That's why, despite the fact that most proprietary Linux software is released with RedHat in mind, I can find it repa
Jesus Christ.
Honestly, are you just totally daft? Your (relatively) low UID suggests that you've been around for a while, but it apparently isn't an indicator of resourcefulness. You do realize that we have the web and google, right?
Or was asking for a source just a half-assed way to indicate your belief that the OP's story was apocryphal?
I put "bill gates ten years two operating systems windows linux" into Google and got this link as the first hit.
That's fewer characters typed than it took you to write your completely moronic post.
Someone on here was saying a few days ago that people on the internet that ask for a cite couldn't give a rat's ass about the veracity of the information they're calling into question. At the time, I thought he was just being overly cynical, but now I'm really starting to wonder.
This has been ten minutes of my already too-short life wasted. I sure as hell hope you appreciate the link that you could have gotten in ten seconds if you hadn't been so god cursed lazy.
Sure, stable is a bit behind for desktops. But while you and I may want the latest and the greatest, in a production environment, forced updates suck because breakage is likely to occur. Of course, you must update from time to time, but it makes everyone's life easier if it doesn't happen often. In this context, Debian Stable is perfect.
When I first started with Debian (around Hamm), the devs generally recommended stable for personal use. Nowadays, they usually don't bother. Stable's focus has shifted, as a result of the slow update cycle that handling so many architectures forces on them. As it turns out, this has been a very good thing.
Debian testing and even unstable are much more like other distributions, and as we're free to track them, I don't really see what the problem is. It used to be that unstable was pretty much broken most of the time and really an experts-only sort of distribution, but nowadays most of the total breakage is in experimental and even unstable is pretty stable, despite it having the latest and the greatest.
The Mag-Lev that connects Shanghai's Long Yang Road subway station with Pudong International Airport has a maximum speed of 431kmph (there is a spedometer in each car to advertise its cruising speed). This is not its actual maximum speed, but given the length of the track (I think it only goes about 40km) it doesn't make sense to accelerate further.
While the TGV has gone 500kmph in speed tests, it rarely passes 350 when actually carrying passengers. Having said that, the TGV is a much smoother ride than the Mag-Lev, but then it is also much more mature technology.
I do believe the Mag-Lev in China is the fastest train in the world, at the moment, in terms of actual speed achieved in regular use. Of course, a mag-lev design removes track friction and so it makes perfect sense that it should be faster than any rail-based alternative.
The Mag-Lev was designed by Germans, though, IIRC, so I'm not sure it's an example of Chinese innovation.
I'm not so sure. I'm thinking long term, understand. Not tomorrow or the next day, but years down the road.
The problem is that a browser is an incredibly complex piece of code, and what with the W3C making recommendations left and right, it isn't getting any simpler.
When you're a company, producing code costs money. Browsers, unlike say, operating systems, are already considered a commodity by the public. Whereas most people think that paying for an OS makes perfect sense, no one is ready to pay for a browser -- they're all free. So it follows that you can't really make any money selling a browser, unless you have some special niche (like Opera does with their mobile phone stuff).
The money you can make (a la Opera) is ok for a small to medium sized business, but not at all interesting from the perspective of a big player like MS.
Now, normally a counter like "the end user doesn't care about these things" would be right on the money, but consider who makes the web that people look at: not Joe Sixpack. Not the unwashed masses. Web developers. Designers.
These people are aware of the W3C, even if the end-consumer is not. These people are aware the CSS3 recommendations, SVG, MathML, XML namespaces and an integrated DOM, etc -- and they are excited about these features. They want to use them to develop awesome websites, webapps. The possibilities are endless, and only one thing stands in the way: Internet Explorer.
This has generated a tremendous amount of animosity among web developers towards IE. They hate it. These people are not Free Software zealots; many of them use Windows exclusively. They don't like Firefox because it's Free -- they like it because it is delivering on the new technologies that promise to make the web an exciting, interesting place.
And they author the web. Don't underestimate their influence.
Firefox is in active development, and it is already much more capable than IE is -- by this I mean that it supports more of the standards, more technologies, and whatsmore, what isn't supported yet is in the process of being implemented by someone, somewhere. If there's a bug, it gets fixed.
The only thing that keeps IE in the game is marketshare.
When you consider that IE6 had it's last feature addition in 1998, it's actually quite an impressive browser: for all its quirks its support of standards isn't that bad. But its support reflects the cutting edge in 1998, not 2005. Time keeps marching on, too. 1998 is getting further and further away, the software older and older and older. Sure, corporate sites will consider IE6 the lowest common denominator for a long time, designing around its limitations -- but they aren't the ones that drive content on the web. Right now, there are already sites that are standards compliant and don't display as intended in IE -- and the authors make no bones about pointing it out. After a while, IE users will see that many such sites exist -- their numbers can only increase.
IE6 will die because MS doesn't care about it anymore. They aren't updating it. They don't support the standards that web devs want. Do you remember how long people supported NS4? IE6 will be the same. But that won't stop the web from moving forward.
Now, MS might change their minds and throw a bunch of money into IE7 or MSN or whatever their new browser is called, bring it up to desired standards compliance in order to keep control of the web. As long as they support open standards, I could care less. I don't mind hacking my CSS/XHTML to work in all the browsers around, as long as those are minor hacks.
But when SVG support starts showing up in official releases of Mozilla and Co, it's going to be hard to swallow IE. The likely work-around for this will be content negotiation, which will mean that when a Firefox browser visits my site they'll get lightweight SVG, and an IE user will be stuck with bulky bitmaps. They won't be able to resize
Actually, Sarge has been the testing release since Woody was released, which was 2 and half years ago I think.
Actually, "testing" and "unstable" always exist -- you can always track them. The name of testing changes each time there's a release -- so back when potato was the stable release, woody was the testing release (sarge was not the unstable release -- the unstable release is always called sid). When woody became stable, they started calling testing sarge. When sarge becomes stable, testing will get a new name, and so on ad infinitum.
So for example, I run testing on my laptop. That means that I run sarge, currently. But the day sarge becomes stable, I will actually be running whatever the new testing is called.
Well, in the wise words of Descartes: "Posto, ergo sum."
Or was that CowboyNeal?
You don't consider locking someone up (possibly indefinitely), without accusing them and without trial, to be infringing on their right to life? Sure, it's not execution -- but who knows how long they'll be held? Who knows whether they're guilty or not, and if they are, of what?
I'll admit that I may have jumped on you prematurely -- I find the whole "us vs. the terrorists" mentality frustrating, and perhaps I wrongly assumed it of you -- but I think that stripping due process very much interferes with a person's right to live their lives as free individuals. From my perspective, locking me up forever isn't much different from killing me. In all likelyhood, the "terrorists" we're holding in Cuba aren't going to be there forever, but they certainly have been there a long time.
At any rate, at this point it's a semantic argument. It sounds as though we basically agree.
I realize that you're being deliberately obtuse, but the use of quotes around the term "terrorist" was clearly meant to give the term a certain uncertainty. Who and what constitutes a terrorist very much depends on your politics, and frankly, many people in the US these days seem to use that term a little bit too freely.
Your picking the extreme "suicide bomber" example says nothing about the OP's point; suicide bombers may be considred terrorists, but not all the people that are considered terrorists are suicide bombers, or even necessarily guilty of a crime anyone can prove in a court of law (what a hassle habeus corpus is -- let's just get rid of it), or even accused of a crime.
The fact that this is not just says nothing about what ought to be done with real criminals. It simply points out that just because the scare term "terrorist" is applied to a person does not mean that they should immediately have all their rights stripped and be sent off to an offshore prison facility where the pesky rights our great nation affords to people in its territory are null and void.
I don't speak for the OP, of course, but I believe that was his point. Whatsmore, I think you understood that full well.
While you may be right, I feel obligated to point out that your "vendor shipments" data is pretty much completely useless when one is discussing a Free OS like Linux or the BSDs, because, put simply, there is no simple indicator for the amount of instances deployed.
With Mac OS X, simply looking at how many units Apple shipped will tell you how many people are adopting Mac OS X. It's quite an accurate indicator and it can be trusted.
With Linux, there are many commercial distributions (Red Hat, SuSE, etc) as well a number of wildly popular free ones (Debian, Slackware). You can buy servers with these pre-installed, but I would be surprised if most Linux deployments were obtained this way. As there are no restrictions on redistribution of GPL'd code (unless changes have been made, obviously, in which case those changes must be made available), how many instances of say, Red Hat Linux sold will tell you nothing whatever about how many systems actually run the OS.
Even with Windows, it's difficult to say; Windows comes pre-installed on most systems whether you want it or not. While I wouldn't expect that a startlingly large percentage of Windows machines have their OS replaced immediately after being bought, it is undeniable that a number of them do (my laptop, for example, came with Windows XP, but now runs Debian). Again, I'm not saying we're a huge percentage -- but we are a percentage.
Really, Mac OS X is pretty much the only OS on the market today whose deployment can be accurately measured with unit shipments (although I don't doubt that there are some people out there that would buy PPC hardware and run Linux or *BSD -- I would, no offense to OS X).
So basically, Apple being the world's largest vendor of UNIX means only that Apple sells more UNIX than any other company -- which means nothing with respect to Linux, and IIRC, at least currently, there are more Linux boxes than there are Mac OS X boxes, by a long shot (although I wouldn't be surprised, given Apple's rising popularity, if this were to change in the near future).
Anyway, I'm not disparaging the Macintosh, or your facts, just pointing out that your argument is pretty much non sequitur in this context. If you were comparing Apple to say, Sun, you'd be much more on the mark.
Exactly. In almost exact analogy to the doc format, the GIMP is capable of opening PSD documents, despite the format being undocumented and tempermental -- although it doesn't always do a good job of it.
.doc files with varying degrees of success, while MS Office is incapable of opening the Abiword/OOo native formats, despite those being open and documented.
Photoshop, on the other hand, cannot open XCF documents -- despite the fact that much less work is required to do so, as the document format is open and documented.
Rather like Abiword/OOo, which can both open
Really, I don't think any of this says anything about lack of interoperability on either side, even though on the face of it MS appears less interoperable with other formats. One, because it's hardly surprising that the dominant player wouldn't bother adding support for his competitor's formats, no matter how easy that support might be, and two, because Bill Gates wasn't talking about Linux/Windows interoperability (at least, in my reading) but rather Windows/Windows interoperability and Linux/Linux interoperability.
In that respect, it's entirely possible that he has a point -- but it's a point that hasn't been neglected by the Linux community, even if it persists. freedesktop.org has published a large number of interoperability standards aimed at getting things to behave in a standard way between widget sets, and current incarnations of GNOME/KDE apparently interoperate rather well (I wouldn't know, I don't use em).
With respect to interapplication interoperability (as in, Office suite interoperability), the recently discussed Open Document Format has already gained fairly widespread acceptance in the word processing group of free programs, and is expected to replace their native formats as default in subsequent releases, guaranteeing that OO and Abiword, for example, do interoperate.
I'm not particularly concerned about that stuff though, as I'm lucky enough to not have to deal with word processors too often.
Still, even if Bill's rant was intended (on his part) as FUD, we should always take a good look at ourselves and see whether his criticisms have merit. My take, in this case, is that he isn't entirely wrong, but he's pointed this issue out when the solution is already well underway. It seems we're ahead of him in that respect.
Don't you mean "copyright holder" and not "license holder"? The license holder, by definition, can't prevent you from doing anything -- the license only specifies under what terms the non-copyright holder may distribute the software (usually, none).
In defense of RMS, he has often stated that the copyleft legal hack would be unnecessary in a world without copyright. You understand that the GPL grants rights -- it does not limit them. Copyright law in the US (and in all nations that are Berne convention signatories) default to an "all rights reserved" status, where only the copyright holder has any distribution rights at all as long as the copyright is valid (it usually expires after some time, but how long depends on the material being copyrighted).
What this means, practically, is that you are by default "prevented" from doing anything (related to distribution, ie, copying) with software that you don't explicitly hold the copyright to.
You are correct that the GPL does not grant all rights; there are some that it reserves. Specifically, it does not grant the right to distribute a binary form of the software without the source code, and aditionally requires that any derivative works also be available under the GPL.
In this sense, the BSD license (and cognates) grant more rights, but even they do not grant all rights.
The only "license" that grants all rights is not a license at all: public domain. On the face of it, public domain code may seem like a good idea, but in fact, it isn't. Why not? Because of liability. If I author code and contribute it to the public domain, I am not making distribution of the code contigent on the receiver abandoning his or her right to take me to court should the code not behave as expected. This is why every license (including BSD) have a clause about "MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE."
Without this clause (a clause that you legally cannot enforce with PD, because you have voluntarily forfeited copyright, and thus there is no licensing or legal agreement) you are liable, implicity, if someone should use your code and have it not work as they expected it to.
Given all the legal explanations I've just given you, you should understand that the statement you've made makes no sense. It makes no sense because licenses, by their definition, cannot "prevent" you from doing anything, as no license by definition means "all rights reserved to the copyright holder and none for anyone else."
The exception, as I stated, is Public Domain, in which you renounce your copyright. Not a good idea from a legal liability standpoint -- use BSD or MIT, it's almost equivalent.
Given this, your understanding of what "proprietary software" is is most certainly somewhat muddled, as no license grants all rights, for sound legal reasons. That would mean that all software is proprietary, which would make the term rather useless, wouldn't it?
At any rate, "proprietary software" is a somewhat subjective term -- I believe it lacks a formal definition. Normally, the implication is that it's software that isn't free, but what free software is will be different depending on whom you ask -- RMS, Theo De Raadt, and ESR all have different views on this question. None is objectively "right", so you'll have to find your own middle ground.
Hopefully this information proves useful in helping you find it.
Yes, apparently he was told just to go to the conference, but he considered that advice harmful.
I'm sure RMS appreciates your nod to him, but he actually doesn't want you to just randomly put GNU/ in front of every instance of Linux. In this case, you should actually just say Linux (and RMS would agree with me), because you aren't talking about GNU/Linux the OS, but Linux the kernel (which Linus wrote, or at least began). Linus did not write GNU/Linux -- most of that OS was written by other people, with most of the userland tools and libraries that make it function coming from the GNU project (hence the GNU/ monicker).
The GNU/Linux / Linux distinction is actually useful, because the OS and the kernel are distinctly different things. I usually just say Linux when it's clear which one I mean, but even if you don't appreciate RMS's point of view, you should understand that a kernel and an operating system are not the same thing.
You're describing the bootstrapping problem, which if I recall correctly is actually described on the GNU website, somewhere.
To answer your question, the GNU devs started out with proprietary operating systems -- primarily SunOS, I think -- and took advantage of the modularity of UNIX to replace one utility at a time. This is why the kernel was the last piece -- because most of what makes a UNIX system run actually resides in user space.
Attempts to create free versions of other OS types -- ReactOS comes to mind -- have a harder time following this example, because most other operating systems are not designed in such a modular way. So they start with the kernel.
There's the question of legality, and then there's the question of ethics. The GPL is designed to only restrict those rights that could be used to restrict the rights of other would be users; in other words, it takes the Libertarian "my right to swing my fist ends where your nose begins" mantra and applies it to software licenses.
The original BSD "advertising clause" does what you refer to -- force credit-giving at the license level -- at it was such a flop that even BSDers shun it. The legal ramifications of such a clause were far too cumbersome to warrant inclusion.
All of this is well and good, but there are a few things that we would all do well to remember:
1) RMS, like him or not, authored the GPL.
2) The existance of the GPL has created a sharing-friendly environment for the creation of free software. BSD is a great license, but most people don't want to see their code swallowed into a proprietary product (the fact that BSD people see no issue with this is great, but most people don't share their view).
3) With respect to the Linux kernel: Linus wrote the Linux kernel to run GNU software. Think about that. I don't mean, "he designed the kernel to run with a GNU userland", although that is also the case. If you read his early usenet posts, he wanted to have kernel that was UNIX-like enough that he could run GNU-utilities. It was his goal, so to speak, at that time.
4) Today, much of what makes your GNU/Linux system run is GNU software -- around 20%, IIRC, on average more than code from any other project (including the kernel itself, by a long shot). Of the programs and libraries that are required to even get your computer to boot (I'm not talking ls here), the percentage is much, much higher.
5) The Linux kernel, itself, can be compiled as intended with only one compiler: gcc. Further, it requires the binutils chain and GNU make to complete the build. These tools (with the exception of GNU make) are so stable and widely used that the BSDs, too, depend on them -- making the GNU project a necessary prerequisite for even BSD development. I don't doubt that had GNU not existed, the BSDs would have gone ahead and developed their own dev tools -- they still dream about doing it, and there are projects every now and then that aim to replace them -- but the fact remains, gcc and the binutils chain form the backbone of the world's free software. There is no equivalent to them. In fact, they're so good, that many proprietary UNIX vendors ship them instead of their own chains these days.
6) A GNU-less Linux kernel (leaving the tools required to build it out of the picture for a second) may exist on some random embedded systems (and even then, not necessarily, as glibc is typically used to interface with the kernel), but there is no server or desktop system that is capable of running without GNU tools. There is no other non-GNU software (with the exception of the kernel itself) that can make this claim. Lots of popular software (like X, for example) is an integral part of GNU/Linux -- but only GNU software can boast such deep integration into the system that you cannot even successfully boot without it.
To me, this seems like a good reason to give GNU credit. A lot of credit. Because the only reason we have free operating systems today is because of this project. Because RMS was so freakishly devoted to freedom that he went ahead and set out to build an entirely free OS, from scratch. GNU is also the only one project that has ever succeeded. No one else ever has, think about this. People may not use the HURD much -- it isn't stable, that's for sure -- but you can run the HURD with only GNU software. Every single other Free Software OS, including the BSDs, requires GNU software somewhere (although admittedly, if you're not keen on compiling anything, you may be able to use the system without GNU software -- but you couldn't develop it).
Now, I don't always
As someone who majored in pure math, I have to disagree vehemently. Computers don't belong in math classes, period, and overhead projectors (ack) aren't a whole lot better.
No, a good math class contains an engaging teacher, students with pencils and paper, and one of a blackboard or white board -- both of which have advantages and disadvantages for the teacher, but are mostly the same to the students. Models may be useful as well, if you're doing something low-dimensional that benefits from visualisation.
Computers can't do math -- all they can do, really, is computation. Which is not to say that they can't be useful, but to introduce them into a math class is to underscore computation in mathematics, when what is important is logic. Math is much more a subset of philosophy than science, a fact commonly misunderstood by non-math people. These non-math people unfortunately include administrators, who in the interest of "bringing computers into the classroom" are wont to force math teachers to use these new-fangled gizmos.
There is no substitute for explanation, for discourse, or for student/teacher bond. Any subject can be exciting and interesting to any student as long as the teacher is able to communicate his or her love of the subject in a way that the student can identify with. Looking at a computer screen in a classroom is tremendously dull, and while it might be manipulated effectively by the teacher, it cannot be manipulated effectively by a student that might want to jump in with his or her insight. Math is not a spectator sport -- student involvement is essential to proper understanding of the subject matter, especially at more advanced levels.
This, incidentally, is one of the reasons that I hate overhead projectors -- they project an 8x11 sized scrap of transparency onto a light surface. This might be great for the teacher, but when you want to bring a student up, it's much harder to find a place for him or her to find a place to write on that tiny scrap.
Further, overhead projectors encourage what is without a doubt the most sinful teaching style -- note preparation. Especially in math, which requires a great deal of thought -- a lubricated brain, so to speak -- boring lectures cannot be tolerated, if you expect your students to come away with an even cursory understanding of the material you present. A teacher that writes out all his notes on transparency is able to simply put up the notes, one sheet at a time, and comment. You know the style! God, just thinking about it makes me want to puke. It encourages a rut, discourages tangential conversations, proofs, and forays -- in other words, it's no better than just reading the textbook to the student, which he or she is quite capable of doing him/herself, thank you very much.
The thought of powerpoint is even worse -- all the fun of the overhead projector with even less editability.
Furthermore, the use of calculators is just silly. Counting is for computers. Proof is for mathematicians. Why on earth would you ever need to cope with numbers more complex than a few rather small natural numbers as coefficients? You might have a case in a trigonometry class but you said you were doing multivariate analysis.
Computers' place in math: a computer lab somewhere with Mathematica/Maple for the undergrads that haven't yet learned to divorce the abstract from the tangible. Certainly not in the classroom, except for very specific purposes (visualisation of recurrences, differential equations, vector fields, that sort of thing, in low-d situations).
"Do you care if it falls?"
"What?"
"The Roman Empire."
A pause. "Fuuuuuuuuuuuuuuuuuck it..."
doubt -> doubt (n)
doubt + less -> without doubt (adj)
doubt + less + ly -> describes an action or adjective which acts or describes something in a manner lacking doubt (adverb of manner, constructed from an adjective, as in "hungrily").
In this case, we take the adjective "informative" and use an adverb to describe how it is infomative; in this sense, I indicate that I do not doubt that it is informative -- in fact, the construction is inherently passive as it deemphasizes the agent doubting, so the implication is that "one" would not doubt that it is informative. In other words, we are speaking of the response that I/no one doubts is quite informative -- but we can use just three words: the doubtlessly informative response.
Got it?
Yeah, I know, I love grammar. Linguist geek and all.
I know he based it on another program, but lets be fair here: what makes fetchmail far, far better than any other similar program is its SMTP integration, which AFAIK popclient did not support. Of course, it wasn't ESR's idea, either (he admits this much) -- but he did write the code to do it. Not that this is complex in the slightest or anything.
Further, given the name popclient, I would presume that it probably only supported POP. Even if it didn't, fetchmail supports everything under the sun. But then, by ESR's own admission, he attempted to mimic Linus Torvalds in his management of the project -- something about proving his Cathedral vs Bazaar theory correct, or the like. Anyway, this means that he accepted a lot of patches and such from a lot of different people, so you could (if you wanted) make the argument that most of the code in fetchmail is not his. But then, that's true of Linus and Linux, too.
One thing is certain -- fetchmail is useful, and has been managed by him, at least. I'll give him credit for it. I don't give him credit for much else.
I believe his point was that RMS is extremely, extremely consistant. From your perspective, this may be taken to mean "he is consistantly a nutjob". This, however, stems from your disagreement with his first principles, not from the insight he deduces from them.
RMS has an opinion -- a very strong opinion -- that you (and many others) disagree with. He could be called a fanatic (and this is what the OP was talking about) because his opinion is, from the perspective of most people, "way the fuck out there". I take it this is your position, and it's a valid one.
ESR, on the other hand, is "reasonable" about certain things -- in fact, some of what he says is very insightful -- but sometimes he's just kooky, and by this I don't mean that he espouses a value or set of values that the mainstream finds odd, but rather, that he is at times painfully irrational and, well, beefheaded.
I've read pretty much everything on his website. To me, it seems as though he wants to be RMS. By this I mean he wants the community to respect him the way that RMS is respected (don't laugh now, whatever your personal feelings about him, he has a large following). He says, in numerous places, that RMS's creation of the GNU project -- the project that is essentially responsible, either directly or indirectly, for 80% of all the free/open source software we use today -- is actually nothing particularly special. That anyone could have done it. That in fact, he could have done it.
He says, again and again, that RMS was just "in the right place at the right time". He belittles RMS. Attacks him publicly, personally, but at the same time addresses him by his first name and claims that the two are or have been friends (something I'm in no position to affirm or deny). RMS, on the other hand, never says anything about ESR. He talks about the Open Source movement in a less than flattering way sometimes, but he almost always has a well-thought out reason for it, even if no one really cares what his reasons are.
I disagree with RMS sometimes, and I disagree with ESR sometimes. I agree with them both sometimes. RMS is fanatical about Free Software; but ESR is just a fanatic.
With respect, Russ, there are other Slashdot readers that would like to hear your views on this. While jbn-o may have (in true hacker tradition, might I add) come across a bit harshly with his point, his point nonetheless stands and those of us following this thread would appreciate a non-sarcastic response from you.
For what it's worth, I too am, as you say, in error, because I have the same apparent misconception of the difference between the Open Source movement and the Free Software movement as jbn-o does, and I'd wager that I'm not alone.
As this is a public forum, with many interested would-be contributors, I think taking your doubtlessly informative response to jbn-o's erroneous assessment private is somewhat at odds with your new position, whose duties presumably include acting as ambassador for OSI.
I'm genuinely curious and open-minded, understand. But at the moment, jbn-o has my attention, and you do not. I had hoped to see something more from you.
I'm not a huge fan of ESR, but in the interest of giving credit where credit is due, he is responsible for fetchmail, which is admittedly an extremely useful (if not particularly exciting) piece of software.
He also wrote an INTERCAL compiler for UNIX-like systems, I think. I think he also wrote a build configuration system (like menuconfig) for the Linux kernel which was ultimately rejected -- but I'll hold off on disparaging him for it, as I've never had anything accepted either, and submitting something on LKML is more than most people do.
Of course, ESR is best known for his gift of gab, not his contributed code. When he's speaking about something he actually has a cursory understanding of, he's usually capable of being quite insightful. Unfortunately, he's one of those intelligent people who believes that because his opinions are respected in one arena, he is automatically an expert in all arenae -- and he attempts frequently to back his esoteric points of view -- especially regarding homosexuality, racism, and the like -- with rather flawed logic.
Of course, ESR is only human, and just because he's been a public figure of some note doesn't mean that we should hold him to an unrealistically high standard.
He's just not particularly well rounded. Writes well, though.