I thought you could already use DTrace on Linux, and if they GPL their stuff, it will all be ported to Linux. The article says that it would be hard, but you know it would happen. Linux does not have DTrace. You're right that it will probably happen. Eventually, after much work. And ZFS isn't looking like it'll be an easy addition either. There doesn't look to be an equivalent of Zones either -- Linux has some nice security module hook in the kernel thanks to work by the NSA, but right now it is largely unused (even distros that enable SELinux have very lax policies, and fairly basic management). Again, that might arrive, at some indeterminate time in the future. Considering that your original post was proclaiming:
...and all the functionality of Linux has jumped ahead of Solaris... arguing that Linux may eventually catch up with these powerful Solaris features is a little disingenuous don't you think? Linux and Solaris are both worth having, depending on what you need. I look forward to what this project, and the OpenSolaris project, can put together.
"but the fact that the built-in make, vi, grep, etc. are still basically unmodified"
Who cares? Do they work? That depends on your measure of "work". They do the raw bare minimum one would expect from such things, but the GNU versions tend to come with a lot of comforts that you start taking for granted after not very long. Its nothing you can't technically live without, but it does start to feel awfully spartan. A good comparison might be Solaris grep and GNU grep, or perhaps Solaris diff and GNU diff. Nothing wrong with the Solaris versions, but the GNU versions have some useful extra options, and more flexible regexps.
Zatoichi? Nah, Lucas should go back to ripping off Kurosawa. How about Seven samurai with six ex-jedi and one farmers son? Or perhaps a nice Star Wars version of Yojimbo and Sanjuro?
The network desktop has been tried many times in the past, by Microsoft (badly) with "ActiveDesktop" and in theory with XAML and.NET, and by Sun in various forms. All the efforts I've seen so far just don't cut it. That doesn't mean it isn't a good idea -- I think there's real promise in a distributed approach to the desktop -- just that it is hard to execute well. Stumbling blocks in the past have included: a lack of real network transparency (the "online" aspect was a thin veneer rather than being truly transparent); lack of sufficient bandwidth (all the "online" stuff was pitifully slow, and ignored); and security, security, security.
To succeed you need a system that doesn't view the network as a bolted on thing, but integrates it at the core; Plan9 comes to mind on that front. At least X11 has network transparency, but it needs to be more efficient (think NX), and have far better security built in to really work for this. Bandwidth will slowly but surely fix itself. That leaves security -- and there's a lot required to make that happen. It is an ambitious and worthy goal, but in this case it is possibly a case of biting off more than you can chew: if it isn't transparent, efficient and secure, it isn't going anywhere, and fulfilling those requirements would require vast architectural changes.
Re:A good start, but still some holes to fill.
on
Linux as A Musician's OS?
·
· Score: 2, Informative
I second that, actually what I want is an application that provides a singing tutor. I have a pretty good voice, but I flub quite a bit of notes and my sense of pitch could be better. I suspect that Solfege may be what you're after. It's a nice little program that can test you on recognising and singing various intervals etc. Definitely worth checking out if you want to improve your ear.
GNOME, sure, but KDE? Granted, I'm lazy and haven't done any research, but what's gone KDE's way while Novell pumps GNOME? Novell bought Ximian, and that has resulted in significant contributions to GNOME from Novell via the Ximian staff. Novell also bought SUSE, and that has resulted in significant contributions to KDE from Novell via the SUSE staff (since SUSE was one of the major contributors of KDE code). Think of it this way: Novell, unlike Redhat, has been quite seriously pursuing the corporate desktop (Redhat has been halfheartedly pursuing it at best, and instead focusing on servers); that has involved quite a bit of user testing and usability studies, the results of which can be applied to both desktops.
Yes, but with only one set of mouse pointer and keyboard focus. Uh, no. Watch the demo. You can have independent focus for each keyboard/mouse set that you care to hotplug in.
give them credit where credit is due. The same research facility came up with another similar technology where you can attach multiple (unlimited) mice to a single PC and all operate independently. Remarkable. I bet no one has ever thought of that before. This is why Microsoft is so successful. They come up with these ideas first, and then implement basic research prototypes that will likely never appear in any shipping product. By comparison open source and X11 has only managed a complete system that lets you hotplug as many mice and keyboards as like that you can install and have working right now.
What's wrong with chalk and slates? Nothing, they work just fine. We don't use them anymore though, becaue pencil and paper are now cheap enough, and better. With a good quality display this would be better than paper texts. Not revolutionarily better, just a little better, but potentially enough better.
Can you search your handwritten notes for keywords and phrases? Can you even do that with your text beyond the limite range of search terms in the index? Can you attach your notes and commentary directly to the relevant passages of the text (staple handwritten pages into the text I guess - hardly ideal or practical)? Can you do worked exercises directly into your textbook -- if so, how many such empty pages do they offer you? Can you have all your course material, notes, exercises, text etc. all kept well organised together with no effort? Or do you have a stack of notes, a text you often don't read, and piles of scrawled exercises that you have to look up in the text to even find what the question was?
You don't need to make it utterly impervious, merely sufficiently difficult, and sufficiently unrewarding, that is isn't worthwhile. Sufficiently difficult can be achieved by having everything bar the textbooks in firmware, and don't provide easy access. Sufficiently unrewarding can be achieved by just making the device itself extremely limited, and largely single purpose. If all you can really achieve by hacking it is a slightly different eBook reader because the hardware is extremely limited then most people won't bother. Sure, you'll never prevent everyone from hacking it, but you don't have to. You just have to make it hard enough that the vast majority of kids won't bother.
A further point based on another comment I saw here: in theory with an eBook based system you could allow schools to basically compile their own textbook from a selection of chapters. Most courses don't use all of a book, and many textbooks are as unweildy as they are because the authors want to cover all the possible topics that a teacher might want to cover. If you could create, and load onto the reader, a book custom built from only the chapters you'll be covering in the course that would be of benefit. Again, this really falls onto textbook publishers to actually offer such options, which I'm not sure they would be inclined to do.
Where is the wisdom we have lost in knowledge? Where is the knowledge we have lost in information?
-- T.S. Eliot, Choruses from the Rock
Information is not the same as knowledge -- you need to piece the information together in a structured way for it to be meaningful. Knowledge is not the same as wisdom, or insight -- you need an understanding of how the knowledge connects together, and how it relates. Right now we have a glut of information. The job of a teacher, in this day and age, is to help teach the students how to put all that information together to build knowledge; how to learn. Often that is going to be done by example, by bringing structure and sense to information as it is provided. Wisdom is, of course, much harder to impart -- only the best teachers can do that; thankfully there are some.
That would be a convenience, but wouldn't solve the problems of pornography, broken equipment, network costs, hacking, etc. I think the solution to that is to not provide a hackable device, but just a very simple reader with a basic tablet interface -- getting stuff on and off can be a non-trivial task because ultimately it will be done once a year when the next years worth of textbooks gets loaded on by the school. It's not a general purpose computer, it's a slightly advanced eBook reader with a non-standard interface for loading new material. That drops out the porn (sure, some kids will figure out a way to get it on there, but its no worse than the kids bringing in Playboy magazines -- you're never going to stop it, you just have to make it decidedly non-trvial), and the network costs. Hopefully such a special purpose device, being as simple as it is, should be much cheaper to manufacture.
Nor would switching tablets for laptops necessarily do anything to improve achievement. A special purpose reader that has all your textbooks with good search facilities and the ability to annotate (via the simple tablet interface), bookmark, etc. would be an improvement over ordinary textbooks -- presuming the reader itself is of good enough quality. Being able to take notes directly onto the textbook, work on problems directly into the text, draw digrams, add bookmarks search tags, and generally have the text more firmly integrated into the course by making it central to all work, is going to be a good thing. It's not a revolution, but it would be an improvement. Of course this requires two things before it is feasible: (1) eBook readers have to be of good enough quality to be an acceptable replacement for paper. (2) Text sellers have to actually sell their eBook versions for significantly less than their printed paper copies. Part (1) is all about the quality of the resolution, and general display. Right now it sucks. ePaper, or eInk, or whatever they call it, shows real promise in this area, but it's still very new. Part (2) is actually the harder one. There's not too much point in this if a printed dead tree copy is as cheap as an eBook -- students can fork over the cash for the heavy version and scrawl in the textbookm themselves; it wouldn't be quite as good as the eBook option, but it would narrow the gap sufficiently. If, on the other hand, eBooks are signficantly cheaper (as we would reasonably expect them to be) then there's enough good economic sense behind moving to eBook reader devices to properly motivate it.
Technology could be very effective in schools. Hell, technology is very effective in schools now. Mass produced paper and ball point pens are a significant improvement over the chalk slates of yesteryear. They in turn, were better then the wet clay tablets of further back. The key here is that you need to introduce technology that can be targetted toward making specific things that we expect children to do at school easier. Just giving them a laptop is "introducing technology to schools", but it isn't introducing anything terribly helpful to current methods of teaching and education. It's a cheap political solution that sounds forward looking because it is "technology", but it does nothing. What would work? How about eBook readers with all the kids textbooks pre-loaded onto it? Better yet, an eBook reader with basic tablet functionality to let the student annotate the PDFs, and write notes -- that's as much as it needs to do. Much more and it is just a distraction. Simple targetted devices are the key. The OLPC project is onto something -- they are quite targetted in the software and OS they are putting on there -- maybe not ideal, but it is something. Why are we not seeing eBook readers taken up? First because politicians tend to be stupid, but second, it is because (let's face it) current eBook readers suck: they just aren't that good. When we have ePaper or similar with much better DPI (and they are starting to appear) things might change. Until then the technology simply isn't good enough to replace pen and paper.
Why do you assume multi-platform programs would use the same UI? Does AOL instant messenger on Mac and Windows use the same UI? Then think of Adium as the OS X UI. Is it that bad that it has a different name? Ultimately they're using the same core libraries, so at heart they're the same, they just have different UIs.
I'm really tired of the chronic mantra that "Linux/FreeBSD/whatever doesn't need memory/CPU speed/whatever" -- it's a classic piece of misdirection. Yes, Linux itself can run on a stripped-down system -- but GNU/Linux is a memory hog, particularly when GUI interfaces are involved. When the wrong GUI interfaces, that are designed for more powerful desktop machines, are used then yes, the memory requirements increase. Then again this looks like a GUI to me, running just fine on a fairly stripped down device -- it has only as much RAM as the original spec for the XO laptops.
Do you honestly that Joe is going to opt for the $800 "vista ready" computer when it looks as though the $500 "ubuntu loaded" one is right next to it on the virtual shelf?
Um, care to quote a real price on the unit? Seeings as where Dell sells PCs with no OS for about the same prices as a Windows machine my guess it that a Linux machine won't be any cheaper. I think the GPP was referring to the fact that machines that can be advertised as "Vista Ready" will require rather higher specs (and hence be noticeable more expensive) than a machine that will run Ubuntu. Sure, on identical hardware there will likely be little or no price difference, but the hardware requirements of the two OSs are distinctly unequal, and if all you need is a low spec box, why spend all the extra money just because Vista requires fancier hardware?
It's called XP. If you think for one second that users who migrated from XP to vista and hated vista are more willing to go to an "unknown" OS versus going back to XP you are out of your mind. Sure, but there has been enough demand for XP that Dell has started offering it as an option on new machines. Obviously there are a lot of people who would prefer not to have Vista. In a sense making this deal with Canonical can be seen as future proofing. Right now Dell can offer XP as an alternative to Vista, and has enough demand that they felt the need to offer it as an option. That option is not going to last forever, however, since come the end of the year MS is going to stop providing XP licenses. If Dell wants to continue to offer alternatives to Vista they'll have to be offering something other than XP -- drawing up a deal with Canonical and getting Ubuntu on machines prior to the end of the year gives them that option. Who knows, maybe when faced with Ubuntu as the alternative to Vista people will take Vista and the demand for something other than Vista will dry up. It makes sense, from Dell's perspective, to at least have something on the table when MS pulls the XP rug out from under them though.
First, maybe that was obvious to you from the initial post but it wasn't to me. Because most run of the mill software conferences you attend have audiences filled with people who are apparently religious about the distinction between Free and free? And they have Stallman in attendance as a speaker? Come now, it wasn't that hard to read between the lines and figure out the context. I think it was more a case of your own prejudices resulting in a kneejerk reaction without bothering to actually read and consider what was said.
Um... in spite of Richard Stallman's rather pathetic attempt to redefine the English language, that is what the term "free software" actually means. You cannot legitimately criticize the Oracle representative for using the English language correctly. Given what Bruce was saying, I think it was implicit that this was a Free Software conference (Stallman was there, giving a speech, and the audience knew, and cared deeply, about the distinction between Free and free). Under those circumstances I think you can very legitmately criticise the Oracle guy. If I go to an Oracle conference and spend my time talking about Delphi (that's where the oracle was after all) and Pythia, and the latest archaeological findings, I think I can reasonably expect to get criticised, despite the that I am using the English language meaning of oracle correctly. Have a little awareness of your audience, and the context in which you are speaking...
The audience tore the Oracle guy to shreds and insisted that he say "cost-less" instead of "Free" for the rest of the talk.
Stallman hasn't been made dictator yet, you know, not even in Cuba. We're still allowed to use "free" in its normal meaning. Sure, but it seemed pretty clear that the conference in question was a Free software conference, and while you are allowed to use free in its normal sense there, you can expect to get heckled for it. If I go to a math conference promising a talk on "Group Theory" and then start talking about the behaviour of mobs of people, well, I can expect some flak for that. That doesn't mean mathematicians control the meaning of "group", but it does reflect the fact that you should really have a clue about your audience and what they mean by key words -- just as the Oracle guy should have done.
My project has been migrating our 300+ machines from Solaris to Red Hat over the past couple of years. We were all excited at first, now we all miss Solaris...with Linux, there's one thing we can't do - we can't keep the machines from locking up. The beauty of the Linux/Solaris situation compared to, say Windows/Solaris, is that a switch in either direction is relatively painless. There are differences, and enough minor hurdles to keep a migration team busy, but for an end user the migration is practically transparent. You're still using the same Desktop Environment, with pretty much all the same applications. That makes it a lot easier to choose whichever one is going to suit your purpose, because you only need to change the underlying layer.
For those of us who just want to generate some simple graphs for papers and such, what do people use? I've messed with Excel, gnuplot, R, and now I'm using ploticus. Anyone have better solutions? To be honest I would suggest you try messing with Gnuplot some more -- it is actually a lot better, and produces much nicer plots, than it seems at first. The trick is to use a different terminal type than "x11", which is pretty crappy; the output looks remarkably different if you use "png", "svg" or "postscript". Here are some examples of plots I've done with Gnuplot: [1], [2], [3][4] (for the last link, note that vertical text alignment renders fine in inkscape, just not on Wikipedia -- download the svg file to see).
If that's still not tickling your fancy then I would suggest matplotlib which is actually pretty versatile, and produces good looking plots. There's also PyX if you're looking for slightly more raw graphical interaction with nice output. Truth be told, however, after messing around with many of the same options you have, I've found that Gnuplot, once you get over the initial learning hurdle and figure out how to turn out nice looking plots, is the fastest and easiest way to turn out plots and charts.
Slashdot Mirror
...and all the functionality of Linux has jumped ahead of Solaris... arguing that Linux may eventually catch up with these powerful Solaris features is a little disingenuous don't you think? Linux and Solaris are both worth having, depending on what you need. I look forward to what this project, and the OpenSolaris project, can put together.Who cares? Do they work? That depends on your measure of "work". They do the raw bare minimum one would expect from such things, but the GNU versions tend to come with a lot of comforts that you start taking for granted after not very long. Its nothing you can't technically live without, but it does start to feel awfully spartan. A good comparison might be Solaris grep and GNU grep, or perhaps Solaris diff and GNU diff. Nothing wrong with the Solaris versions, but the GNU versions have some useful extra options, and more flexible regexps.
Zatoichi? Nah, Lucas should go back to ripping off Kurosawa. How about Seven samurai with six ex-jedi and one farmers son? Or perhaps a nice Star Wars version of Yojimbo and Sanjuro?
The network desktop has been tried many times in the past, by Microsoft (badly) with "ActiveDesktop" and in theory with XAML and .NET, and by Sun in various forms. All the efforts I've seen so far just don't cut it. That doesn't mean it isn't a good idea -- I think there's real promise in a distributed approach to the desktop -- just that it is hard to execute well. Stumbling blocks in the past have included: a lack of real network transparency (the "online" aspect was a thin veneer rather than being truly transparent); lack of sufficient bandwidth (all the "online" stuff was pitifully slow, and ignored); and security, security, security.
To succeed you need a system that doesn't view the network as a bolted on thing, but integrates it at the core; Plan9 comes to mind on that front. At least X11 has network transparency, but it needs to be more efficient (think NX), and have far better security built in to really work for this. Bandwidth will slowly but surely fix itself. That leaves security -- and there's a lot required to make that happen. It is an ambitious and worthy goal, but in this case it is possibly a case of biting off more than you can chew: if it isn't transparent, efficient and secure, it isn't going anywhere, and fulfilling those requirements would require vast architectural changes.
What's wrong with chalk and slates? Nothing, they work just fine. We don't use them anymore though, becaue pencil and paper are now cheap enough, and better. With a good quality display this would be better than paper texts. Not revolutionarily better, just a little better, but potentially enough better.
Can you search your handwritten notes for keywords and phrases? Can you even do that with your text beyond the limite range of search terms in the index? Can you attach your notes and commentary directly to the relevant passages of the text (staple handwritten pages into the text I guess - hardly ideal or practical)? Can you do worked exercises directly into your textbook -- if so, how many such empty pages do they offer you? Can you have all your course material, notes, exercises, text etc. all kept well organised together with no effort? Or do you have a stack of notes, a text you often don't read, and piles of scrawled exercises that you have to look up in the text to even find what the question was?
You don't need to make it utterly impervious, merely sufficiently difficult, and sufficiently unrewarding, that is isn't worthwhile. Sufficiently difficult can be achieved by having everything bar the textbooks in firmware, and don't provide easy access. Sufficiently unrewarding can be achieved by just making the device itself extremely limited, and largely single purpose. If all you can really achieve by hacking it is a slightly different eBook reader because the hardware is extremely limited then most people won't bother. Sure, you'll never prevent everyone from hacking it, but you don't have to. You just have to make it hard enough that the vast majority of kids won't bother.
A further point based on another comment I saw here: in theory with an eBook based system you could allow schools to basically compile their own textbook from a selection of chapters. Most courses don't use all of a book, and many textbooks are as unweildy as they are because the authors want to cover all the possible topics that a teacher might want to cover. If you could create, and load onto the reader, a book custom built from only the chapters you'll be covering in the course that would be of benefit. Again, this really falls onto textbook publishers to actually offer such options, which I'm not sure they would be inclined to do.
Where is the wisdom we have lost in knowledge?
Where is the knowledge we have lost in information?
-- T.S. Eliot, Choruses from the Rock
Information is not the same as knowledge -- you need to piece the information together in a structured way for it to be meaningful. Knowledge is not the same as wisdom, or insight -- you need an understanding of how the knowledge connects together, and how it relates. Right now we have a glut of information. The job of a teacher, in this day and age, is to help teach the students how to put all that information together to build knowledge; how to learn. Often that is going to be done by example, by bringing structure and sense to information as it is provided. Wisdom is, of course, much harder to impart -- only the best teachers can do that; thankfully there are some.
(1) eBook readers have to be of good enough quality to be an acceptable replacement for paper.
(2) Text sellers have to actually sell their eBook versions for significantly less than their printed paper copies.
Part (1) is all about the quality of the resolution, and general display. Right now it sucks. ePaper, or eInk, or whatever they call it, shows real promise in this area, but it's still very new. Part (2) is actually the harder one. There's not too much point in this if a printed dead tree copy is as cheap as an eBook -- students can fork over the cash for the heavy version and scrawl in the textbookm themselves; it wouldn't be quite as good as the eBook option, but it would narrow the gap sufficiently. If, on the other hand, eBooks are signficantly cheaper (as we would reasonably expect them to be) then there's enough good economic sense behind moving to eBook reader devices to properly motivate it.
Technology could be very effective in schools. Hell, technology is very effective in schools now. Mass produced paper and ball point pens are a significant improvement over the chalk slates of yesteryear. They in turn, were better then the wet clay tablets of further back. The key here is that you need to introduce technology that can be targetted toward making specific things that we expect children to do at school easier. Just giving them a laptop is "introducing technology to schools", but it isn't introducing anything terribly helpful to current methods of teaching and education. It's a cheap political solution that sounds forward looking because it is "technology", but it does nothing. What would work? How about eBook readers with all the kids textbooks pre-loaded onto it? Better yet, an eBook reader with basic tablet functionality to let the student annotate the PDFs, and write notes -- that's as much as it needs to do. Much more and it is just a distraction. Simple targetted devices are the key. The OLPC project is onto something -- they are quite targetted in the software and OS they are putting on there -- maybe not ideal, but it is something. Why are we not seeing eBook readers taken up? First because politicians tend to be stupid, but second, it is because (let's face it) current eBook readers suck: they just aren't that good. When we have ePaper or similar with much better DPI (and they are starting to appear) things might change. Until then the technology simply isn't good enough to replace pen and paper.
Um, care to quote a real price on the unit? Seeings as where Dell sells PCs with no OS for about the same prices as a Windows machine my guess it that a Linux machine won't be any cheaper. I think the GPP was referring to the fact that machines that can be advertised as "Vista Ready" will require rather higher specs (and hence be noticeable more expensive) than a machine that will run Ubuntu. Sure, on identical hardware there will likely be little or no price difference, but the hardware requirements of the two OSs are distinctly unequal, and if all you need is a low spec box, why spend all the extra money just because Vista requires fancier hardware?
Out of curiousity -- have you looked into Teradata at all? It certainly offers scalability to ridicolous data volumes.
Stallman hasn't been made dictator yet, you know, not even in Cuba. We're still allowed to use "free" in its normal meaning. Sure, but it seemed pretty clear that the conference in question was a Free software conference, and while you are allowed to use free in its normal sense there, you can expect to get heckled for it. If I go to a math conference promising a talk on "Group Theory" and then start talking about the behaviour of mobs of people, well, I can expect some flak for that. That doesn't mean mathematicians control the meaning of "group", but it does reflect the fact that you should really have a clue about your audience and what they mean by key words -- just as the Oracle guy should have done.
Bar charts work just fine. For fancier plotting options your best bet is R.
If that's still not tickling your fancy then I would suggest matplotlib which is actually pretty versatile, and produces good looking plots. There's also PyX if you're looking for slightly more raw graphical interaction with nice output. Truth be told, however, after messing around with many of the same options you have, I've found that Gnuplot, once you get over the initial learning hurdle and figure out how to turn out nice looking plots, is the fastest and easiest way to turn out plots and charts.