I agree, the crass humour and inside jokes meant there was no-one I would have been comfortable suggesting this to.
It's also unfortunate that the last show will be a LugRadio Live broadcast. The regular shows generally have enough interesting content to compensate for the often tiresome gags. But the live shows are a different story. It seems they take special delight in insulting not only each other, but their guests and sponsors as well.
And the segment bashing unions for stifling open source was just pathetic. I know they were annoyed by the inconvenience, but if I'd travelled any distance to see them live, I'd have been pissed that they chose to focus on this instead of something that actually relates to Linux. I just turned it off.
"Just becasue you are an unethical bastard, doesn't mean everyone else is."
You're right, it was all my fault.
Still, I think it's worth pointing out that in situations that rewarded proper documentation, I was more than happy to do so. If you want people to act a certain way, you should encourage them to do so, rather than expecting them to suffer for it.
I've never been told by a TA to find the answer and work backwards
It doesn't have to be that blatant. In my undergrad chem labs, we were marked based on how well we ran the experiments, with the relative success based on the yield we got. So accurately reporting that we got 50% yield was enough to pass the lesson, barely. We quickly learned that doubling the reagent volumes, without reporting that we had done so, would bring our yields up to the 80-90% range we needed to get a good mark. So, without ever being explicitly told to cheat, we learned that cheating was valued over truthful reporting and acted accordingly.
In a way, this is very much in keeping with the way science is practiced. Failed experiments, well run and accurately written up, do not get published. We are rewarded for our results (true or false), not our ability as experimenters.
In my botany labs, if an experiment failed (i.e., the plant died) I reported that truthfully, and was marked based on the quality of the report, not the actual outcome of the experiment. This was reflective of the quality of teaching in our botany department as a whole, and one reason why I'm a botanist today.
I'm a software developer. I write thousands of lines of code every year. And you know what? I was paid for writing each one of those, with no expectation of continuing profits until I die. Why should some high-school dropout with no useful skills other than wailing (off-pitch, no less) into a microphone expect to be any different? Similarly, why should any corporation expect that? A key difference is you were paid for your code. Most musicians are not being paid up front - they need some way to charge for their service after the fact. I agree, there's no reason to grant anyone a perpetual monopoly. But to suggest that musicians are paid to record an album, and the money from CD sales is all 'extra' is simply not the reality they are working in.
Some people might be embarrassed to tell musicians and movie makers that they shouldn't be compensated for the work they do musicians and movie makers have never been compensated for the work they do. musicians have been given few cents over $20 worth single album sales by the distributors, and told to go on unending world tours in order to earn money for themselves. What's worse, the industry siphoning off 99% of their profit, or you stealing 100% of what's left? It seems to me that no-one is valuing the work of the musician here.
... what draws me to his works is the extreme depth and development... Tolkien had the mythical history of Middle-Earth more or less figured out by the time LOTR was published to the extent that some of the languages are fully functional (Quenya and Sindarin especially).
There is nothing second rate about Tolkien, except perhaps to a world of small attention spans and desire for quick (and cheap) thrills...
When I was fourteen I would have agreed with you. I was totally captivated by the dwarvish runes and all that 'extreme depth'. But rereading the books twenty years later, it comes across as more than a little self-indulgent. Detail is a good thing when it adds texture and context to your story. But when you feel the need to inform your audience of the translations of every minor character's name in a half dozen fictional languages, that's taking it a little too far. Do we really need to know that Gandalf's half-brother's cousin's second wife is known as Wilma to the Dwarves of Khazaa Dum, and Betty to the Elves of Mirkwood?
Is my short attention span to blame for me not being endlessly fascinated by meticulous, excessive, and ultimately pointless details?
huh. I don't use MS anymore, but I thought I read that multiple desktops was one of the new features of Vista. I can't find anything about it on microsoft.com, and google only showed me a bunch of third-party projects, so I guess I was mistaken. My bad.
Now that I'm used to my multi-desktop arrangement it would be hard to go back...
I understand where you're coming from. But I think the UI design does actually make a lot of sense in a Linux context. Some of the problems you describe may be a consequence of a flawed port to windows. But when you're working with virtual desktops those problems completely disappear.
In my case, I'm reading slashdot in my 'firefox' desktop. When I want to get back to editing a picture I hotkey over to the 'gimp' desktop where all the windows are laid out how I want them. When I need to check mail, I hop over to my mail desktop and so on. Nothing is ever minimized, and I never find myself alt-tabbing through a dozen unrelated windows, as would happen when all my apps are on the same desktop.
Given that 99% of my computer use is confined to 3 apps (browser, editor, mail), a six desktop layout is more than enough room for all, without becoming too complex to navigate. Now that MS supports multiple desktops as well, at least in Vista, people may start discovering that there are saner ways to arrange multiple windows than squeezing them all onto the toolbar of a single desktop.
However, there is still some fixed cost to be absorbed somewhere in the chain to support the administration and management of legitimate peer review. Presently, publishers absorb this cost.
Not quite. Peer reviewers are not paid for their efforts, and the associate editors that manage them are not paid for their work. The only people that get paid in any of the journals in my discipline are the technical people responsible for actually assembling the articles, and possibly the top editor who oversees the associate editors. The actual cost of production is tiny compared to the price charged for a subscription.
A colleague of mine is involved in a small non-profit journal, and he figures he needs to charge less than half of what the mainstream journals do in order to cover his costs. Considering that the big journals will benefit from a substantially larger subscription/content ratio, they really are making out like bandits.
We have the tools within the academic and library communities to take control of our own publications, what we need is a shift in thinking, and some way to reward running a journal that is on par with the professional prestige associated with actually publishing in it.
Agreed. All I'm saying is that ESR's version of the story seems to suggest that the only reason he's not as highly regarded as RMS is because he chose not to be, and not because he isn't really bringing anything to the table.
RMS continues to present a considered, consistent and insightful perspective on issues related to Free Software. You may or may not agree with what he says, but he says it well and is a good spokesperson for the FSF. On the other hand, while ESR occasionally has an interesting observation, he seems to be primarily interested in singing his own praises. I think that's why he's not taken as seriously as RMS -- not because his insightful analysis of political movements in the 1800s informed his decision to deliberately removed himself from the public spotlight. That's a self-serving crock.
That's not irony. I think you missed his point. RMS remains a charismatic figurehead of an important organisation, and ESR is not. Lest we think this is due to any failing on the part of ESR, he informs us that he chose to remain out of the spotlight, in order to serve the greater good. Opinions will differ as to whether this is more self-serving revisionism, or if he really did exercise that kind of foresight.
I'm glad that ESR sees at least one of his goals as being to reduce the amount of political crap flying around in the community
After he'd already equated Free Software with moralizing and conducting ideological warfare, you could be forgiven for thinking he was more interested in spreading bullshit that containing it.
yp.
Re:The best tools stay out of the way...
on
Goodbye Cruel Word
·
· Score: 3, Informative
Fundamentally the issue with interfaces is not providing features piled on features, but figuring out how to craft a tool that people can use to get work done rather than having to learn how to use the tool.
That's fine, if you just want to write letters to your friends and family, or update a personal blog, or whatever. But if writing is something you do professionally, what is wrong with investing an afternoon or a weekend in learning how to use a truly powerful editor? My work involves a combination of technical writing, popular writing, and coding. I could do all of these using Microsoft Word, or Word in combination with Notepad for coding, with very minimal time required to get going.
But investing a week (over a period of several months) in learning to use Emacs to serve my needs has paid off dividends. When you consider that most of us spend 40+ hours a week, 48+ weeks a year, editing text of one kind or another, I think the expectation that a good tool is one that take no effort to *start* using is misguided. If you are going to be spending a large chunk of your life doing a particular task, a little short term pain to gain access to a tool that will grow with your needs over the rest of your career is really not such a burden.
Emacs is not the answer to everyone's needs, of course. But I think anyone who is at all technically savvy should at least consider learning to use a proper editor.
Traditionally, copyright was for the life of the author + some reasonably large number.
The only tradition associated with copyright terms is the practice of extending them beyond the previous limit. Copyright started out at 14 years[1] with an optional 7 year extension. Actually, I guess there is also a second tradition, that of abusing the monopoly granted to copyright holders regardless of the term.
The only objection to teaching intelligent design in schools is when it is taught in science class, as it is clearly not science.
So ignorant teachers are free to spout nonsense to children, as long as it's done out-of-context? By that logic, it would be ok to teach kids that 2 + 2 = 5, as long as you don't do it in math class.
True enough, as far as individuals are concerned. But institutions like schools, libraries, businesses etc. are a lot more risk-averse than individuals. Responsibly-run organizations will be following the law regardless of what the RCMP decide to do.
yp.
science is too big to avoid black boxes
on
Open Source Math
·
· Score: 1
>> Black box science does not work.
That's unfairly harsh. Nearly all science done today is black box science to some extent. I work on molecular systematics. I understand a great deal about the biology of the system I work on, the mathematical underpinnings of the multivariate analyses I do, and some of the biochemistry involved in the reactions I run. However, I do not know how a PCR machine works, beyond the trivial observation that it heats stuff up and cools it down. I could not tell you how the gel I load onto a sequencer converts the samples I load into the picture I analyze three hours later, beyond an equally naive explanation.
This does not invalidate my work, but it does highlight the importance of using open source protocols. Outside of math and physics research, science is far too complex for any individual to understand every aspect of their research apparatus. We rely on each other to fill in the gaps. I have the skills and training to validate a small component of my research. But by using well-documented procedures and relatively widely available equipment, the bits I can't validate myself can be checked by others with appropriate background. But if I use proprietary protocols, reagents, or software, then I can no longer rely on the community to validate my work.
I wish I could say all my work used Free Software and open protocols. It's something I strive towards, and advocate for in areas where I have any say. But too many researchers are too focussed on producing high quality results in their specific area of specialty, to the extent that they rely on proprietary shortcuts in areas that they see as peripheral.
Better yet, stop by your local used book store and get last year's edition. Math doesn't change that fast. You can spend $150 for this year's edition, or get the same material for 90% off used. The only reason to buy a new math textbook is if you have to work the specific exercises assigned by a professor. I taught myself basic-to-advanced calculus using three different textbooks purchased for less than $10 together.
One thing that is worth investing in as a self-learner is a Schaum's book with the worked examples. For $40 or less, especially if you find one used, you can have 1000's of practice problems with detailed answers. It's about the closest thing to having a real live teacher walk you through the tough stuff.
that's a different suit than "they are using BusyBox but they aren't making all the source to everything on their hardware available".
I think the more tricky points here are about what it will do to the related works (aggregates, etc.)
This is very clearly spelled out in the GPL. From version 3:
Inclusion of a covered {i.e., GPL} work in an aggregate does not cause this License to apply to the other parts of the aggregate.
As far as the difference between a png decoder and a Stephen King novel, there is no difference in terms of copyright. Because you consider a novel to be inherently more valuable than an unmodified bit of GPL code is immaterial. What you think about it doesn't matter, both works are covered by copyright, and you are bound by law to respect that copyright. And it doesn't matter if you modify the decoder. The key event is distribution. You can modify it all you like and keep your modifications to yourself, *until you distribute it*. Then you are required to release the source to anyone you give the binary to. And that's the case for unmodified code too - if you distribute it, you have to make the source available.
You're very confused about what the GPL does. If Monsoon is selling a piece of hardware that is running GPL code, that is very clearly distribution. If they haven't modified the code in doing so, they are still required to make the source available to their customers should they request it. And if they have modified the source at all, their modifications must be made available. Whether or not there is other code included in the product that is unrelated to busybox is immaterial. They can put all the proprietary code they like in their product, but anything derived from GPL code must be made available to customers who request it.
Some of your particular points:
Now, the question is: are the things that use BusyBox derivative of BusyBox or not?
They don't even need to be 'derivative'. They just need to use BusyBox, and they are bound to provide the sources on request.
Does use of BusyBox by Monsoon mean that every bit of Monsoon's other code has to be released as well?
Absolutely not. The GPL doesn't spread to other code. Programmers spread GPL code to other projects. Any derivative of GPL code has to be released when it's distributed, but other unrelated code in the same product is not effected. See almost any Linux distribution for examples of GPL and proprietary code being distributed side by side.
I would bet that if Monsoon can demonstrate that they are just using BusyBox out of convenience and not because it is crucial to their software, the worst that will happen is that they will have to replace the BusyBox with some other code.
This is just plain ridiculous. Can I distribute photocopies of a Stephen King novel, just because it's more convenient than writing my own book?
I don't think they can be forced to release their source code.
If it is derived from GPL code they most certainly can.
I guess the question then is, if I write "hello world" for Linux, and with my hello world distribute Linux as well, does that mean I have to release the source to "hello world"? That's the question I'm fairly certain people are going to see here.
I'm fairly certain you don't have the foggiest idea how the GPL works. You should take a stroll over to gnu.org before you continue with your uninformed speculation.
No, no-one has ever suggested that GPLv3 somehow permits the unauthorized distribution of proprietary software. This is your own kooky reading of the debate.
What has been claimed is that by participating in the distribution of GPLv3 programs via the SUSE certificates, MS will be forced to comply with GPLv3 with respect to the software *in the SUSE distribution*. If true, this means that they forfeit their right to sue anyone, whether or not they are using SUSE, over any patents they claim are violated by the GPLv3 software that MS distributes.
I can't comment on whether or not this interpretation will hold up in court, and of course SUSE doesn't include any GPLv3 code *yet*. But your suggestion that RMS is somehow hoping to use GPLv3 to gain access to MS software is just plain wrong.
Emacs comes with fine tutorial, available from the help menu, or via C-h t. It comes with a complete online reference manual. A tutorial introduction to elisp, aimed at getting non-programmers up to speed wrt customizing their Emacs. And there's a two-volume set for anyone interested in serious programming in elisp. And O'Reilly has a good manual as well, if you want to pay for it.
And you definitely don't have to learn elisp to use the advanced features of Emacs. You have access to a very rich suite of editing functions with standard emacs. If you want the same (comparable, if not identical) features available with Vim? you will be just fine without knowing anything about lisp.
However, with a little time invested you can increase your productivity by customizing functions. I suppose all editors worth their salt provide regexp search and replace. But if you want to automate complicated stuff having a full extension language on hand is a huge plus. For example, I'm writing a latex document, and I want to be able to pull out all the figures/tables/footnotes to a separate file. This requires a fairly sophisticated regexp, as it has to handle nested parentheses and various options for the different environment types. I don't think it can be done with a one-line regexp. It can be done in a dozen lines of elisp, and nothing too difficult to sort out since the real heavy lifting of the regexps is already done with standard functions. Of course, I'm a rank beginner at this stuff, but I'm hooked after seeing how easy it was to make a fairly complex and useful function.
I guess that dedicated TeX editors probably already have such features built-in. But another benefit of doing this with emacs is that I don't need to learn a different interface to do similar manipulations to code, mail, html...
I don't think it did, really. It only looks like that because it is so readily extended. If you compare Emacs to ed or vi or similar, it's going to look very complex and bloaty. But the better comparison is probably with bash.
Bash gives you a pretty basic programming language, and then you take a few Unix apps and pipe them together and you can do just about anything. Same thing with Emacs - you start with a relatively basic text editor and a solid extension language. A few decades later those basic ingredients have been sculpted by the user-community to provide a huge variety of features that rms never imagined way back when. It's exactly because elisp is simple and modular that Emacs has become so incredibly feature-rich.
If you really wanted to you could probably excise most of the extensions out of Emacs to trim it down. The fact that this hasn't happened suggests that most of us Emacs-ites appreciate having those features available, even if we only have time to learn to use a handful at a time.
Sure, you may see a lot of stuff when you look at what's included, but there's really no tractable way to see what's not included. I bet there's a lot that's not a part of the Debian repo that you or I have never even heard about
True, but that's really not important. What is important is that the stuff you *have* heard about is there. Browsers: check. Office Suites: check. Email clients: check. That covers most regular users right there. Of course, it's Debian so you also get all the development tools, sound, printing, multimedia etc. Granted, there are some issues with some of these things, but those aren't so much packaging problems as more general licensing issues.
I've run into plenty of academic packages that weren't Java (actually, I've never run into graphics-related research that wasn't implemented in C or C++).
This is really a problem with academia, rather than linux. I'd prefer to use tools that are written in portable standard C with a decent command-line interface, but Java is better than having to switch computers to use something once or twice. As far as the debian repos go, it will likely never make sense to include these sorts of packages, which by their nature will have a very small, and usually very technically proficient, user base.
One problem shows up because of how Debian's stable/testing/unstable segregation works, and how that's not effective for most people's desktops. I want the latest stable KDE, but I want bleeding-edge PidginIM
With all due respect, this is not a problem with Debian, it's a problem with what you expect a distro to do. Stable and bleeding-edge are mutually exclusive concepts for any sort of mass-production system like vanilla-Debian. If you are in a position to a) need to mix and match stable and unstable packages and b) you actually know what that entails, then you should know that you are going to have to do some custom tweaking of your own to satisfy your requirements. But this doesn't reflect a shortcoming of Debian with respect to regular users. It just means that you have very exacting requirements, and you're unlikely to find an answer in a prepackaged form. Most non-technical users won't have these sorts of requirements, so it doesn't matter that to achieve them requires skills they don't have.
>> This is one of the areas where Free Software is far, far ahead of what Windows currently has.
> Right up until the software you want isn't in the repo, or is broken. Then it falls way, way behind.
I disagree. First, if you're running Debian then there is very little that isn't in the repository. The only things I use that aren't available through aptitude are some very specialized niche programs developed by academics to solve very particular problems. Most of these are in Java, and so the installation process is identical on Linux and MS.
Second, you assume that even if something isn't in the repository for a Linux distro it would already have a point-and-click installer for MSWindows. If something is not in the Debian repository it is generally either very specialized or in an early stage of development. In many cases these are things that aren't yet available as anything but source code for any platform. So the real comparison is not between compiling source for Linux vs clicking on an.exe for windows. Your real options are more likely compile from source for Linux, which can be difficult, or compile from source for MSWindows, which is generally even harder, if it's even possible.
The real issue here is that with Free Software you often have access to programs much earlier in the development cycle than you get with proprietary projects. So it looks more difficult, but in reality it's just a consequence of the more transparent development process. You can certainly handle the standard browsing/email/word-processing tasks without ever compiling anything.
Slashdot Mirror
I agree, the crass humour and inside jokes meant there was no-one I would have been comfortable suggesting this to.
It's also unfortunate that the last show will be a LugRadio Live broadcast. The regular shows generally have enough interesting content to compensate for the often tiresome gags. But the live shows are a different story. It seems they take special delight in insulting not only each other, but their guests and sponsors as well.
And the segment bashing unions for stifling open source was just pathetic. I know they were annoyed by the inconvenience, but if I'd travelled any distance to see them live, I'd have been pissed that they chose to focus on this instead of something that actually relates to Linux. I just turned it off.
yp.
"Just becasue you are an unethical bastard, doesn't mean everyone else is."
You're right, it was all my fault.
Still, I think it's worth pointing out that in situations that rewarded proper documentation, I was more than happy to do so. If you want people to act a certain way, you should encourage them to do so, rather than expecting them to suffer for it.
yp.
I've never been told by a TA to find the answer and work backwards
It doesn't have to be that blatant. In my undergrad chem labs, we were marked based on how well we ran the experiments, with the relative success based on the yield we got. So accurately reporting that we got 50% yield was enough to pass the lesson, barely. We quickly learned that doubling the reagent volumes, without reporting that we had done so, would bring our yields up to the 80-90% range we needed to get a good mark. So, without ever being explicitly told to cheat, we learned that cheating was valued over truthful reporting and acted accordingly.
In a way, this is very much in keeping with the way science is practiced. Failed experiments, well run and accurately written up, do not get published. We are rewarded for our results (true or false), not our ability as experimenters.
In my botany labs, if an experiment failed (i.e., the plant died) I reported that truthfully, and was marked based on the quality of the report, not the actual outcome of the experiment. This was reflective of the quality of teaching in our botany department as a whole, and one reason why I'm a botanist today.
yp.
... what draws me to his works is the extreme depth and development ... Tolkien had the mythical history of Middle-Earth more or less figured out by the time LOTR was published to the extent that some of the languages are fully functional (Quenya and Sindarin especially).
There is nothing second rate about Tolkien, except perhaps to a world of small attention spans and desire for quick (and cheap) thrills...
When I was fourteen I would have agreed with you. I was totally captivated by the dwarvish runes and all that 'extreme depth'. But rereading the books twenty years later, it comes across as more than a little self-indulgent. Detail is a good thing when it adds texture and context to your story. But when you feel the need to inform your audience of the translations of every minor character's name in a half dozen fictional languages, that's taking it a little too far. Do we really need to know that Gandalf's half-brother's cousin's second wife is known as Wilma to the Dwarves of Khazaa Dum, and Betty to the Elves of Mirkwood?
Is my short attention span to blame for me not being endlessly fascinated by meticulous, excessive, and ultimately pointless details?
yp
huh. I don't use MS anymore, but I thought I read that multiple desktops was one of the new features of Vista. I can't find anything about it on microsoft.com, and google only showed me a bunch of third-party projects, so I guess I was mistaken. My bad.
Now that I'm used to my multi-desktop arrangement it would be hard to go back...
yp.
I understand where you're coming from. But I think the UI design does actually make a lot of sense in a Linux context. Some of the problems you describe may be a consequence of a flawed port to windows. But when you're working with virtual desktops those problems completely disappear.
In my case, I'm reading slashdot in my 'firefox' desktop. When I want to get back to editing a picture I hotkey over to the 'gimp' desktop where all the windows are laid out how I want them. When I need to check mail, I hop over to my mail desktop and so on. Nothing is ever minimized, and I never find myself alt-tabbing through a dozen unrelated windows, as would happen when all my apps are on the same desktop.
Given that 99% of my computer use is confined to 3 apps (browser, editor, mail), a six desktop layout is more than enough room for all, without becoming too complex to navigate. Now that MS supports multiple desktops as well, at least in Vista, people may start discovering that there are saner ways to arrange multiple windows than squeezing them all onto the toolbar of a single desktop.
yp.
However, there is still some fixed cost to be absorbed somewhere in the chain to support the administration and management of legitimate peer review. Presently, publishers absorb this cost.
Not quite. Peer reviewers are not paid for their efforts, and the associate editors that manage them are not paid for their work. The only people that get paid in any of the journals in my discipline are the technical people responsible for actually assembling the articles, and possibly the top editor who oversees the associate editors. The actual cost of production is tiny compared to the price charged for a subscription.
A colleague of mine is involved in a small non-profit journal, and he figures he needs to charge less than half of what the mainstream journals do in order to cover his costs. Considering that the big journals will benefit from a substantially larger subscription/content ratio, they really are making out like bandits.
We have the tools within the academic and library communities to take control of our own publications, what we need is a shift in thinking, and some way to reward running a journal that is on par with the professional prestige associated with actually publishing in it.
yp.
We'd be sad, but it wouldn't end FSF.
Agreed. All I'm saying is that ESR's version of the story seems to suggest that the only reason he's not as highly regarded as RMS is because he chose not to be, and not because he isn't really bringing anything to the table.
RMS continues to present a considered, consistent and insightful perspective on issues related to Free Software. You may or may not agree with what he says, but he says it well and is a good spokesperson for the FSF. On the other hand, while ESR occasionally has an interesting observation, he seems to be primarily interested in singing his own praises. I think that's why he's not taken as seriously as RMS -- not because his insightful analysis of political movements in the 1800s informed his decision to deliberately removed himself from the public spotlight. That's a self-serving crock.
yp.
That's not irony. I think you missed his point. RMS remains a charismatic figurehead of an important organisation, and ESR is not. Lest we think this is due to any failing on the part of ESR, he informs us that he chose to remain out of the spotlight, in order to serve the greater good. Opinions will differ as to whether this is more self-serving revisionism, or if he really did exercise that kind of foresight.
yp
I'm glad that ESR sees at least one of his goals as being to reduce the amount of political crap flying around in the community
After he'd already equated Free Software with moralizing and conducting ideological warfare, you could be forgiven for thinking he was more interested in spreading bullshit that containing it.
yp.
Fundamentally the issue with interfaces is not providing features piled on features, but figuring out how to craft a tool that people can use to get work done rather than having to learn how to use the tool.
That's fine, if you just want to write letters to your friends and family, or update a personal blog, or whatever. But if writing is something you do professionally, what is wrong with investing an afternoon or a weekend in learning how to use a truly powerful editor? My work involves a combination of technical writing, popular writing, and coding. I could do all of these using Microsoft Word, or Word in combination with Notepad for coding, with very minimal time required to get going.
But investing a week (over a period of several months) in learning to use Emacs to serve my needs has paid off dividends. When you consider that most of us spend 40+ hours a week, 48+ weeks a year, editing text of one kind or another, I think the expectation that a good tool is one that take no effort to *start* using is misguided. If you are going to be spending a large chunk of your life doing a particular task, a little short term pain to gain access to a tool that will grow with your needs over the rest of your career is really not such a burden.
Emacs is not the answer to everyone's needs, of course. But I think anyone who is at all technically savvy should at least consider learning to use a proper editor.
yp
Traditionally, copyright was for the life of the author + some reasonably large number.
The only tradition associated with copyright terms is the practice of extending them beyond the previous limit. Copyright started out at 14 years[1] with an optional 7 year extension. Actually, I guess there is also a second tradition, that of abusing the monopoly granted to copyright holders regardless of the term.
yp.
[1]http://download.nowis.com/index.cfm?phile=FreeCulture.html&tipe=text/html#2_2_1
The only objection to teaching intelligent design in schools is when it is taught in science class, as it is clearly not science.
So ignorant teachers are free to spout nonsense to children, as long as it's done out-of-context? By that logic, it would be ok to teach kids that 2 + 2 = 5, as long as you don't do it in math class.
yp.
True enough, as far as individuals are concerned. But institutions like schools, libraries, businesses etc. are a lot more risk-averse than individuals. Responsibly-run organizations will be following the law regardless of what the RCMP decide to do.
yp.
>> Black box science does not work.
That's unfairly harsh. Nearly all science done today is black box science to some extent. I work on molecular systematics. I understand a great deal about the biology of the system I work on, the mathematical underpinnings of the multivariate analyses I do, and some of the biochemistry involved in the reactions I run. However, I do not know how a PCR machine works, beyond the trivial observation that it heats stuff up and cools it down. I could not tell you how the gel I load onto a sequencer converts the samples I load into the picture I analyze three hours later, beyond an equally naive explanation.
This does not invalidate my work, but it does highlight the importance of using open source protocols. Outside of math and physics research, science is far too complex for any individual to understand every aspect of their research apparatus. We rely on each other to fill in the gaps. I have the skills and training to validate a small component of my research. But by using well-documented procedures and relatively widely available equipment, the bits I can't validate myself can be checked by others with appropriate background. But if I use proprietary protocols, reagents, or software, then I can no longer rely on the community to validate my work.
I wish I could say all my work used Free Software and open protocols. It's something I strive towards, and advocate for in areas where I have any say. But too many researchers are too focussed on producing high quality results in their specific area of specialty, to the extent that they rely on proprietary shortcuts in areas that they see as peripheral.
yp.
Better yet, stop by your local used book store and get last year's edition. Math doesn't change that fast. You can spend $150 for this year's edition, or get the same material for 90% off used. The only reason to buy a new math textbook is if you have to work the specific exercises assigned by a professor. I taught myself basic-to-advanced calculus using three different textbooks purchased for less than $10 together.
One thing that is worth investing in as a self-learner is a Schaum's book with the worked examples. For $40 or less, especially if you find one used, you can have 1000's of practice problems with detailed answers. It's about the closest thing to having a real live teacher walk you through the tough stuff.
yp.
that's a different suit than "they are using BusyBox but they aren't making all the source to everything on their hardware available".
I think the more tricky points here are about what it will do to the related works (aggregates, etc.)
This is very clearly spelled out in the GPL. From version 3:
As far as the difference between a png decoder and a Stephen King novel, there is no difference in terms of copyright. Because you consider a novel to be inherently more valuable than an unmodified bit of GPL code is immaterial. What you think about it doesn't matter, both works are covered by copyright, and you are bound by law to respect that copyright. And it doesn't matter if you modify the decoder. The key event is distribution. You can modify it all you like and keep your modifications to yourself, *until you distribute it*. Then you are required to release the source to anyone you give the binary to. And that's the case for unmodified code too - if you distribute it, you have to make the source available.
You're very confused about what the GPL does. If Monsoon is selling a piece of hardware that is running GPL code, that is very clearly distribution. If they haven't modified the code in doing so, they are still required to make the source available to their customers should they request it. And if they have modified the source at all, their modifications must be made available. Whether or not there is other code included in the product that is unrelated to busybox is immaterial. They can put all the proprietary code they like in their product, but anything derived from GPL code must be made available to customers who request it.
Some of your particular points:
Now, the question is: are the things that use BusyBox derivative of BusyBox or not?
They don't even need to be 'derivative'. They just need to use BusyBox, and they are bound to provide the sources on request.
Does use of BusyBox by Monsoon mean that every bit of Monsoon's other code has to be released as well?
Absolutely not. The GPL doesn't spread to other code. Programmers spread GPL code to other projects. Any derivative of GPL code has to be released when it's distributed, but other unrelated code in the same product is not effected. See almost any Linux distribution for examples of GPL and proprietary code being distributed side by side.
I would bet that if Monsoon can demonstrate that they are just using BusyBox out of convenience and not because it is crucial to their software, the worst that will happen is that they will have to replace the BusyBox with some other code.
This is just plain ridiculous. Can I distribute photocopies of a Stephen King novel, just because it's more convenient than writing my own book?
I don't think they can be forced to release their source code.
If it is derived from GPL code they most certainly can.
I guess the question then is, if I write "hello world" for Linux, and with my hello world distribute Linux as well, does that mean I have to release the source to "hello world"? That's the question I'm fairly certain people are going to see here.
I'm fairly certain you don't have the foggiest idea how the GPL works. You should take a stroll over to gnu.org before you continue with your uninformed speculation.
No, no-one has ever suggested that GPLv3 somehow permits the unauthorized distribution of proprietary software. This is your own kooky reading of the debate.
What has been claimed is that by participating in the distribution of GPLv3 programs via the SUSE certificates, MS will be forced to comply with GPLv3 with respect to the software *in the SUSE distribution*. If true, this means that they forfeit their right to sue anyone, whether or not they are using SUSE, over any patents they claim are violated by the GPLv3 software that MS distributes.
I can't comment on whether or not this interpretation will hold up in court, and of course SUSE doesn't include any GPLv3 code *yet*. But your suggestion that RMS is somehow hoping to use GPLv3 to gain access to MS software is just plain wrong.
yp.
Emacs comes with fine tutorial, available from the help menu, or via C-h t. It comes with a complete online reference manual. A tutorial introduction to elisp, aimed at getting non-programmers up to speed wrt customizing their Emacs. And there's a two-volume set for anyone interested in serious programming in elisp. And O'Reilly has a good manual as well, if you want to pay for it.
And you definitely don't have to learn elisp to use the advanced features of Emacs. You have access to a very rich suite of editing functions with standard emacs. If you want the same (comparable, if not identical) features available with Vim? you will be just fine without knowing anything about lisp.
However, with a little time invested you can increase your productivity by customizing functions. I suppose all editors worth their salt provide regexp search and replace. But if you want to automate complicated stuff having a full extension language on hand is a huge plus. For example, I'm writing a latex document, and I want to be able to pull out all the figures/tables/footnotes to a separate file. This requires a fairly sophisticated regexp, as it has to handle nested parentheses and various options for the different environment types. I don't think it can be done with a one-line regexp. It can be done in a dozen lines of elisp, and nothing too difficult to sort out since the real heavy lifting of the regexps is already done with standard functions. Of course, I'm a rank beginner at this stuff, but I'm hooked after seeing how easy it was to make a fairly complex and useful function.
I guess that dedicated TeX editors probably already have such features built-in. But another benefit of doing this with emacs is that I don't need to learn a different interface to do similar manipulations to code, mail, html...
yp.
I don't think it did, really. It only looks like that because it is so readily extended. If you compare Emacs to ed or vi or similar, it's going to look very complex and bloaty. But the better comparison is probably with bash.
Bash gives you a pretty basic programming language, and then you take a few Unix apps and pipe them together and you can do just about anything. Same thing with Emacs - you start with a relatively basic text editor and a solid extension language. A few decades later those basic ingredients have been sculpted by the user-community to provide a huge variety of features that rms never imagined way back when. It's exactly because elisp is simple and modular that Emacs has become so incredibly feature-rich.
If you really wanted to you could probably excise most of the extensions out of Emacs to trim it down. The fact that this hasn't happened suggests that most of us Emacs-ites appreciate having those features available, even if we only have time to learn to use a handful at a time.
yp.
Sure, you may see a lot of stuff when you look at what's included, but there's really no tractable way to see what's not included. I bet there's a lot that's not a part of the Debian repo that you or I have never even heard about
True, but that's really not important. What is important is that the stuff you *have* heard about is there. Browsers: check. Office Suites: check. Email clients: check. That covers most regular users right there. Of course, it's Debian so you also get all the development tools, sound, printing, multimedia etc. Granted, there are some issues with some of these things, but those aren't so much packaging problems as more general licensing issues.
I've run into plenty of academic packages that weren't Java (actually, I've never run into graphics-related research that wasn't implemented in C or C++).
This is really a problem with academia, rather than linux. I'd prefer to use tools that are written in portable standard C with a decent command-line interface, but Java is better than having to switch computers to use something once or twice. As far as the debian repos go, it will likely never make sense to include these sorts of packages, which by their nature will have a very small, and usually very technically proficient, user base.
One problem shows up because of how Debian's stable/testing/unstable segregation works, and how that's not effective for most people's desktops. I want the latest stable KDE, but I want bleeding-edge PidginIM
With all due respect, this is not a problem with Debian, it's a problem with what you expect a distro to do. Stable and bleeding-edge are mutually exclusive concepts for any sort of mass-production system like vanilla-Debian. If you are in a position to a) need to mix and match stable and unstable packages and b) you actually know what that entails, then you should know that you are going to have to do some custom tweaking of your own to satisfy your requirements. But this doesn't reflect a shortcoming of Debian with respect to regular users. It just means that you have very exacting requirements, and you're unlikely to find an answer in a prepackaged form. Most non-technical users won't have these sorts of requirements, so it doesn't matter that to achieve them requires skills they don't have.
yp
>> This is one of the areas where Free Software is far, far ahead of what Windows currently has.
> Right up until the software you want isn't in the repo, or is broken. Then it falls way, way behind.
I disagree. First, if you're running Debian then there is very little that isn't in the repository. The only things I use that aren't available through aptitude are some very specialized niche programs developed by academics to solve very particular problems. Most of these are in Java, and so the installation process is identical on Linux and MS.
Second, you assume that even if something isn't in the repository for a Linux distro it would already have a point-and-click installer for MSWindows. If something is not in the Debian repository it is generally either very specialized or in an early stage of development. In many cases these are things that aren't yet available as anything but source code for any platform. So the real comparison is not between compiling source for Linux vs clicking on an .exe for windows. Your real options are more likely compile from source for Linux, which can be difficult, or compile from source for MSWindows, which is generally even harder, if it's even possible.
The real issue here is that with Free Software you often have access to programs much earlier in the development cycle than you get with proprietary projects. So it looks more difficult, but in reality it's just a consequence of the more transparent development process. You can certainly handle the standard browsing/email/word-processing tasks without ever compiling anything.
yp