Then there's the proposed $100 per flight user fee.
And insurance. At $500k insured hull, you're looking at quite a substantial figure. My guess from my own insurance premium would be $13k.
And we have capital cost, which, at a low 4.5% interest rate, amounts to $22,500 a year for half a mill.
There's nothing cheap about owning an airplane. (I own a glider, and even there, the insurance and some minimal maintenance cost add up to a rather high per-flight cost...) You don't own an aircraft because you're trying to save money!
What are they actually using in terms of special apps?
I suspect most of it are web-based eGovernment applications, perhaps accounting (SAP?), on top of OpenOffice. The GNU/Linux applications involved are all very stable by now, so this seems like a reasonable decision.
The press release actually mentions an increase in workstations from 1,500 to 9,500, and a reduction in system malfunctions. I don't think it is plausible to have either 70 or 46 actual support tickets, as suggested by the description here. That doesn't make sense given the number of machines involved, whether they're running Windows or GNU/Linux or whatever.
Besides, the PR compares the modern-day GNU/Linux installation to Windows NT. Seriously?
(PS: Was it the German foreign affairs office that changed back to Windows recently, due to general user unhappiness?)
This is tyranny of a set of minorities, who all expect to have their irrational beliefs respected and tolerated.
Is this another instance of what the ecologist Hardin called the Tragedy of the Commons? By making use of simple, seemingly reasonable demands, a large number of individuals exploits a shared resource (culture, education) up to the point where the total of the actions disables the system. Religious and political freedom may now destroy the education system and with it the environment that makes these freedoms viable in the first place.
We were looking into CalDAV just a few days ago, and we couldn't find a decent web client that supports it. (We tried Horde Kronolith, but the client UI turned out to be poor, and support for CalDAV or even just WebDAV nonexistent.)
The problem with this is that you would need to take one of the standard Dells (closely matching a MacBook Pro) and then configure the MacBook to match the Dell. I suspect that the result would be much closer, or maybe the Dell coming out top.
You don't need Chinese people to get heavily accented English. In fact, English varies a lot. If you're in Yorkshire (England) or in Ayrshire (Scotland), in Singapore, Brisbane or Nashville, you'll find extremely different accents. A good speech corpus will contain large samples of as many accents as possible, including meta-data that allows people to filter this to produce an acoustic model that is tailored to intended target users.
But the same applies to recording modalities. Depending on whether you're building dialogue systems that run over the phone, or whether you want to recognize dictated input, you will have to have different recordings. The acoustic qualities of the channel differ (phone vs. high-quality microphone), and while that can be simulated and compensated for to some extent, the fact that people speak differently in spontaneous conversation vs. dictation will pretty much break most recognizer - unless you have purpose-built models.
These are some of the reasons why it's hard to find an all-purpose corpus, and why the existing corpora are so expensive to license. (Not only does each have a limited market - they are very expensive to collect, too.) So it's great that the effort is being made - I hope it'll eventually lead to improvements in the freely available recognition engines, too.
There's nothing wrong with "making a nice little incoming on the side", is there?
It's a contractual matter, and as such it'll probably depend on the particular copyright arrangements in place. They will likely differ for affiliate professors (who are academic freelancers and just contracted for particular classes) and tenured ones, who are basically full-time & permanent staff.
Not quite. Doesn't TCAS, the collision avoidance system in use, rely on transponder signals (radar) and / or communication with the TCAS in the other aircraft to give the right instructions?
Buildings neither have transponders nor a TCAS...
There are other plans in place, including computer-simulated "walls", which would prevent airliners from entering exclusion zones.
Well, Ozone depletion seems to actually lower temperatures, thereby masking global warming.
Both Ozone hole and global warming have similar causes though (emissions, in general). Don't forget that Ozone forming low on the ground (in smog) is another another problem.
The fact that phenomena are related and causality is not as easy as you think or thought it was isn't the scientist's fault. What matters is what we know now, based on current evidence. And that is that global warming is much greater than a normal, cyclical effect. It is clear that it is man-made, and there are absolutely plausible known mechanisms for this.
Sure, an administration can repress scientists and support the mineral oil business (owned by the President's family) and a globally harmful lifestyle. But that is not going to change the realities, and it's not going to save our habitat.
While LaTeX is a technically great typesetting system for all of these cases, human factors play a crucial role here. What type of people study arts, history, classics, or also law or philosophy? Even in some more scientific subjects like psychology, you tend to find people who will very much prefer a WYSIWYG system. LaTeX looks like programming to them - the level of abstraction is too high.
That is not to say that a history student can't learn LaTeX. Hell, they would probably be happy with it and consistently use it. But getting there is too card, in the general case.
If you're in science, LaTeX is the way to go. On the Mac, you can use Aquamacs Emacs which is easy to use and comes with the AUCTeX environment to edit LaTeX comfortably. TeXShop is a good alternative.
However, if you're in the humanities or so, your students will likely be unable to learn LaTeX in reasonable time. In that case, I'd recommend Papyrus from ROM Logicware, which is a very fast text processor that can deal with large documents. (Their web site is crap, but try the Papyrus demo!). There are alternative editors, but Papyrus seems like the fastest word processor to me. Also, it's available for different systems, not just for Macs, and they have cheap licenses, I think.
Boy, what do you have to do to get a $1500 fine? I heard backing up from a highway toll station (to get into another lane) in Italy is pretty darn expensive... What were your $600 (was $1500) for?
So i've been working on a free and open source project for about two years, with releases coming out for about a year. The financial side is desolate. I don't despair (I don't live from it), but there is little monetary incentive to continue.
Income from donations is fairly small. Maybe $100 a month on average, more after releases. Overall, it doesn't justify the development effort.
Making money on support contracts is not likely to proove effective. My project is an advanced distribution of a text editor that's mainly used by geeks, and there in either a non-commercial (FOSS) environment, or by people who have been using the editor for many years. They're happy to solve their problems themselves, or simply ask a mailing list, where there is plenty of free support. Companies tend to use IDEs for their development efforts, and dedicated tools such as Dreamweaver to produce HTML. The third major application, LaTeX, is common only in academic circle (where people help themselves rather than paying for external support).
I could try to make $1 on each download. I'm allowed to do that under the GPL. That'll significantly decrease my user base. And it'll make Paypal rich, rather than me. Micropayment solutions aren't workable these days - there is no standard (except Paypal, maybe), and the fees for small sums are substantial.
Charging for the binary might mean copyrighting the layout of the distribution. That's against the spirit of free software, to some extent. People might fork off new versions. There are folks out there who simply care about this ideology, but don't see that developers need to pay their bills, too.
Ad revenue? Miniscule. And I've just seen that companies like Macupdate, which account for a certain amount of traffic, deliberately cut off any revenue stream by deep-linking so I can't show ads or ask people to donate when they download (I have no control over the download servers).
Bad market conditions with the wrong users for a workable business model. I don't know if it's the philosophy behind it. As a developer, I love the idea of free software. But since I have to pay my rent, I wonder if I can ever get some compensation for what I'm doing.
The reason why I continue is that users are supportive in a social way. Some donate and show their appreciation. Some blog about the project. Hardly anyone contributes code (Mac users...), but a lot of them send in good bug reports and support it that way. RMS and the FSF say, well, money is not the ultimate goal for all of us. Yes that's right. But as an answer to the problem, it's not enough.
A standard is a technical scheme which is wide-spread. Some people call it "quasi-standard", but I think that's prescriptivism. On the other hand, the W3C publishes a lot of technical recommendations, which never become a standard.
Creativity in the user interface isn't neccessarily good. In the most cases, it's very very bad for the user.
iPods (maybe not the Shuffle) are a standard in the market of personal stereos. But the point was more about a USB memory stick that behaves like a drive. Works on my Mac, works on Windows. Automatically. If there's an automount that works with the USB sticks out there, then I'd say a Linux box can deal with such standard periphery. Maybe it can do so, I don't know. It's just that it didn't work for me, last time I've tried.
I also remember PDFs inside a pdflatex'ed document being rendered wrongly (black instead of transparent). Something that worked just fine on Macs and Windows machines, but the Linux boxes never rendered it right (I think acroread did, but some free software didn't).
Let's not go over single bugs that we've discovered here and there. The problem is the general picture, and I think GNU/Linux is catching up there. But it's not quite there yet.
Seems that UI standards is a total internal thing for any given app
On my Mac, almost all apps stick to guidelines. While written by a single vendor (Apple), they are adhered to across most applications for the benefit of users. On my Linux boxes, applications differ greatly - compare xview to acroread. Compare copy&paste. In the shell, same story. Sometimes it's '--verbose' and '-v', sometimes it's '-verbose'. Sometimes input goes in stdin, sometimes a file needs to be specified, or '-' for stdin.
As for new hardware standards: if I plugged in my iPod Shuffle (behaves like a standard USB memory stick), would I automatically get a new mount in/mnt? Not here, where we have Fedora Core. My officemate recently installed Windows over his Linux and he was so happy to have sound output. The machine's a recent Dell. You could call "Dell" a standard in today's world.
Standards, #1: When installing things on a Linux (or Unix, for that matter) machine, (i.e../configure, make, install), I bump into dependency chains from hell. Different versions of various libraries. This is a problem with the distribution technique. Libaries do not establish consistent and future-proof APIs. KDE and GNOME use different APIs to some extent. Existing standards (e.g. X11) lack features. On my Mac, I have nice.app bundles that usually contain everything that's needed, so you get around this issue. In the times of 1$=1GB, the cost is neglegible. Applications adhere to UI standards (which they don't on GNU/Linux, and which they often do on Windows).
Standards, #2: Various hardware standards and emerging standards aren't well supported on Linux. As someone said before, the software needs to work with a lot of different file formats and allow plugging in a lot of different hardware. Standards are there to help. They don't impede on my freedom - in the contrary, they buy me valuable time when I don't have to hunt down drivers. On my Mac, everything just works. And that's what I would expect from an OS and its applications were I supposed to install it instead on my laptop.
Linux machines loaded with Firefox, at Circuit City or Best Buy, for $200 less than the Windows counterparts.
Except that it would be around $20 less, or so. OEM licenses are cheap.
I guess the advantage of a free operating system is that it's libre rather than gratis.
(That's a different story for server-level features, of course.)
The one-buffer-one-frame thing can be turned off with one mouse click (Options menu). Same for the frame appearance themes, which many people find useful to distinguish different types of buffers shown in the frames..emacs is loaded, but the customizations (from custom and the Options menu) that you're not supposed to mess with anyways are stored inside the Library/Preferences hierarchy where they belong on a Mac.
About the customizations you're right to some extend - Aquamacs has different defaults. It's not meant to be for the long-time Emacs geek who doesn't actually want an Emacs that behaves differently. In your case I'd really recommend the Carbon Emacs Package. But for most other people, we think that Aquamacs defaults make a lot of sense.
I use in-ear headphones which are placed inside the ear canal. They shut off noise better than activate noise cancelling headphones can (better than the Bose Quiet Comfort I used to have) and enable me to listen to music at much lower volume.
I use the Etymotic 6i phones - they're already pretty good in terms of sound quality. For a couple hundred bucks more, you can get something from their professional range - these are supposed to be superb.
Slashdot Mirror
Then there's the proposed $100 per flight user fee. And insurance. At $500k insured hull, you're looking at quite a substantial figure. My guess from my own insurance premium would be $13k. And we have capital cost, which, at a low 4.5% interest rate, amounts to $22,500 a year for half a mill. There's nothing cheap about owning an airplane. (I own a glider, and even there, the insurance and some minimal maintenance cost add up to a rather high per-flight cost...) You don't own an aircraft because you're trying to save money!
What are they actually using in terms of special apps? I suspect most of it are web-based eGovernment applications, perhaps accounting (SAP?), on top of OpenOffice. The GNU/Linux applications involved are all very stable by now, so this seems like a reasonable decision. The press release actually mentions an increase in workstations from 1,500 to 9,500, and a reduction in system malfunctions. I don't think it is plausible to have either 70 or 46 actual support tickets, as suggested by the description here. That doesn't make sense given the number of machines involved, whether they're running Windows or GNU/Linux or whatever. Besides, the PR compares the modern-day GNU/Linux installation to Windows NT. Seriously? (PS: Was it the German foreign affairs office that changed back to Windows recently, due to general user unhappiness?)
This is tyranny of a set of minorities, who all expect to have their irrational beliefs respected and tolerated. Is this another instance of what the ecologist Hardin called the Tragedy of the Commons? By making use of simple, seemingly reasonable demands, a large number of individuals exploits a shared resource (culture, education) up to the point where the total of the actions disables the system. Religious and political freedom may now destroy the education system and with it the environment that makes these freedoms viable in the first place.
Wow, that's big news. I heard him speak last year when he got the Rumelhart prize - the story about how he worked out Bayesian networks is humble. I think, some paper napkins were involved. The whole lecture is archived here: http://thesciencenetwork.org/programs/cogsci-2011/rumelhart-lecture-judea-pearl
We were looking into CalDAV just a few days ago, and we couldn't find a decent web client that supports it. (We tried Horde Kronolith, but the client UI turned out to be poor, and support for CalDAV or even just WebDAV nonexistent.)
What a sorry state of affairs...
Sorry, that's a contradiction.
because it wouldn't have to worry about pilots or passengers getting tired.
What? That's certainly not the reason why jets cruise at high altitudes, and I don't know what passengers getting tired has to do with it anyways.
Not insightful. Mod parent down.
The problem with this is that you would need to take one of the standard Dells (closely matching a MacBook Pro) and then configure the MacBook to match the Dell.
I suspect that the result would be much closer, or maybe the Dell coming out top.
/ from a mac fanboy
hey better than nothing. or as the statistical NLP people tend to say: there's only one thing that's better than data. more data!
You don't need Chinese people to get heavily accented English. In fact, English varies a lot. If you're in Yorkshire (England) or in Ayrshire (Scotland), in Singapore, Brisbane or Nashville, you'll find extremely different accents. A good speech corpus will contain large samples of as many accents as possible, including meta-data that allows people to filter this to produce an acoustic model that is tailored to intended target users.
But the same applies to recording modalities. Depending on whether you're building dialogue systems that run over the phone, or whether you want to recognize dictated input, you will have to have different recordings. The acoustic qualities of the channel differ (phone vs. high-quality microphone), and while that can be simulated and compensated for to some extent, the fact that people speak differently in spontaneous conversation vs. dictation will pretty much break most recognizer - unless you have purpose-built models.
These are some of the reasons why it's hard to find an all-purpose corpus, and why the existing corpora are so expensive to license. (Not only does each have a limited market - they are very expensive to collect, too.) So it's great that the effort is being made - I hope it'll eventually lead to improvements in the freely available recognition engines, too.
There's nothing wrong with "making a nice little incoming on the side", is there?
It's a contractual matter, and as such it'll probably depend on the particular copyright arrangements in place. They will likely differ for affiliate professors (who are academic freelancers and just contracted for particular classes) and tenured ones, who are basically full-time & permanent staff.
You cannot produce "megawatts per day". That's non-sensical. You can produce N megawatts at any time, or M megawatt hours per day.
Not quite. Doesn't TCAS, the collision avoidance system in use, rely on transponder signals (radar) and / or communication with the TCAS in the other aircraft to give the right instructions? Buildings neither have transponders nor a TCAS... There are other plans in place, including computer-simulated "walls", which would prevent airliners from entering exclusion zones.
Both Ozone hole and global warming have similar causes though (emissions, in general). Don't forget that Ozone forming low on the ground (in smog) is another another problem.
The fact that phenomena are related and causality is not as easy as you think or thought it was isn't the scientist's fault. What matters is what we know now, based on current evidence. And that is that global warming is much greater than a normal, cyclical effect. It is clear that it is man-made, and there are absolutely plausible known mechanisms for this.
Sure, an administration can repress scientists and support the mineral oil business (owned by the President's family) and a globally harmful lifestyle. But that is not going to change the realities, and it's not going to save our habitat.
While LaTeX is a technically great typesetting system for all of these cases, human factors play a crucial role here. What type of people study arts, history, classics, or also law or philosophy? Even in some more scientific subjects like psychology, you tend to find people who will very much prefer a WYSIWYG system. LaTeX looks like programming to them - the level of abstraction is too high.
That is not to say that a history student can't learn LaTeX. Hell, they would probably be happy with it and consistently use it. But getting there is too card, in the general case.
However, if you're in the humanities or so, your students will likely be unable to learn LaTeX in reasonable time. In that case, I'd recommend Papyrus from ROM Logicware, which is a very fast text processor that can deal with large documents. (Their web site is crap, but try the Papyrus demo!). There are alternative editors, but Papyrus seems like the fastest word processor to me. Also, it's available for different systems, not just for Macs, and they have cheap licenses, I think.
Hi there,
o ryid=127
have you seen this one?
The world seems to be laughing about your city administration...
http://www.centos.org/modules/news/article.php?st
Cheers
Dave
Boy, what do you have to do to get a $1500 fine? I heard backing up from a highway toll station (to get into another lane) in Italy is pretty darn expensive... What were your $600 (was $1500) for?
So i've been working on a free and open source project for about two years, with releases coming out for about a year. The financial side is desolate. I don't despair (I don't live from it), but there is little monetary incentive to continue.
Income from donations is fairly small. Maybe $100 a month on average, more after releases. Overall, it doesn't justify the development effort.
Making money on support contracts is not likely to proove effective. My project is an advanced distribution of a text editor that's mainly used by geeks, and there in either a non-commercial (FOSS) environment, or by people who have been using the editor for many years. They're happy to solve their problems themselves, or simply ask a mailing list, where there is plenty of free support. Companies tend to use IDEs for their development efforts, and dedicated tools such as Dreamweaver to produce HTML. The third major application, LaTeX, is common only in academic circle (where people help themselves rather than paying for external support).
I could try to make $1 on each download. I'm allowed to do that under the GPL. That'll significantly decrease my user base. And it'll make Paypal rich, rather than me. Micropayment solutions aren't workable these days - there is no standard (except Paypal, maybe), and the fees for small sums are substantial.
Charging for the binary might mean copyrighting the layout of the distribution. That's against the spirit of free software, to some extent. People might fork off new versions. There are folks out there who simply care about this ideology, but don't see that developers need to pay their bills, too.
Ad revenue? Miniscule. And I've just seen that companies like Macupdate, which account for a certain amount of traffic, deliberately cut off any revenue stream by deep-linking so I can't show ads or ask people to donate when they download (I have no control over the download servers).
Bad market conditions with the wrong users for a workable business model. I don't know if it's the philosophy behind it. As a developer, I love the idea of free software. But since I have to pay my rent, I wonder if I can ever get some compensation for what I'm doing.
The reason why I continue is that users are supportive in a social way. Some donate and show their appreciation. Some blog about the project. Hardly anyone contributes code (Mac users...), but a lot of them send in good bug reports and support it that way. RMS and the FSF say, well, money is not the ultimate goal for all of us. Yes that's right. But as an answer to the problem, it's not enough.
A standard is a technical scheme which is wide-spread. Some people call it "quasi-standard", but I think that's prescriptivism. On the other hand, the W3C publishes a lot of technical recommendations, which never become a standard.
Creativity in the user interface isn't neccessarily good. In the most cases, it's very very bad for the user.
iPods (maybe not the Shuffle) are a standard in the market of personal stereos. But the point was more about a USB memory stick that behaves like a drive. Works on my Mac, works on Windows. Automatically. If there's an automount that works with the USB sticks out there, then I'd say a Linux box can deal with such standard periphery. Maybe it can do so, I don't know. It's just that it didn't work for me, last time I've tried.
I also remember PDFs inside a pdflatex'ed document being rendered wrongly (black instead of transparent). Something that worked just fine on Macs and Windows machines, but the Linux boxes never rendered it right (I think acroread did, but some free software didn't).
Let's not go over single bugs that we've discovered here and there. The problem is the general picture, and I think GNU/Linux is catching up there. But it's not quite there yet.
On my Mac, almost all apps stick to guidelines. While written by a single vendor (Apple), they are adhered to across most applications for the benefit of users. On my Linux boxes, applications differ greatly - compare xview to acroread. Compare copy&paste. In the shell, same story. Sometimes it's '--verbose' and '-v', sometimes it's '-verbose'. Sometimes input goes in stdin, sometimes a file needs to be specified, or '-' for stdin.
As for new hardware standards: if I plugged in my iPod Shuffle (behaves like a standard USB memory stick), would I automatically get a new mount in /mnt? Not here, where we have Fedora Core. My officemate recently installed Windows over his Linux and he was so happy to have sound output. The machine's a recent Dell. You could call "Dell" a standard in today's world.
Standards, #1: ./configure, make, install), I bump into dependency chains from hell. Different versions of various libraries. This is a problem with the distribution technique. Libaries do not establish consistent and future-proof APIs. KDE and GNOME use different APIs to some extent. Existing standards (e.g. X11) lack features. On my Mac, I have nice .app bundles that usually contain everything that's needed, so you get around this issue. In the times of 1$=1GB, the cost is neglegible. Applications adhere to UI standards (which they don't on GNU/Linux, and which they often do on Windows).
When installing things on a Linux (or Unix, for that matter) machine, (i.e.
Standards, #2:
Various hardware standards and emerging standards aren't well supported on Linux. As someone said before, the software needs to work with a lot of different file formats and allow plugging in a lot of different hardware. Standards are there to help. They don't impede on my freedom - in the contrary, they buy me valuable time when I don't have to hunt down drivers. On my Mac, everything just works. And that's what I would expect from an OS and its applications were I supposed to install it instead on my laptop.
Except that it would be around $20 less, or so. OEM licenses are cheap.
I guess the advantage of a free operating system is that it's libre rather than gratis. (That's a different story for server-level features, of course.)
Can someone explain to me what exactly the harm is in having children occasionally trip over a picture with a naked breast (or worse, oh my God!)?
Parents getting questions about what these naked people do on that picture?
I agree that porn doesn't reflect real sex life, but are James Bond or 24 an accurate reflection of work in intelligence agencies?
Gun fights in the afternoon program is OK. Naked people making love is not. I don't get it.
The one-buffer-one-frame thing can be turned off with one mouse click (Options menu). Same for the frame appearance themes, which many people find useful to distinguish different types of buffers shown in the frames. .emacs is loaded, but the customizations (from custom and the Options menu) that you're not supposed to mess with anyways are stored inside the Library/Preferences hierarchy where they belong on a Mac.
About the customizations you're right to some extend - Aquamacs has different defaults. It's not meant to be for the long-time Emacs geek who doesn't actually want an Emacs that behaves differently. In your case I'd really recommend the Carbon Emacs Package. But for most other people, we think that Aquamacs defaults make a lot of sense.
I use in-ear headphones which are placed inside the ear canal. They shut off noise better than activate noise cancelling headphones can (better than the Bose Quiet Comfort I used to have) and enable me to listen to music at much lower volume.
I use the Etymotic 6i phones - they're already pretty good in terms of sound quality. For a couple hundred bucks more, you can get something from their professional range - these are supposed to be superb.