Well, no, it amounted to a crime because basically they effectively sold the casinos a product saying "Don't worry, this won't get hacked, it's solid", and then used their inside knowledge to hack it themselves. This amounted to defrauding the casinos - little better than an online retailer saying "don't worry, we'll keep your credit card information safe" and then promptly maxing out the credit cards of anyone who shops there. I imagine you would want to see law enforcement involved with any online retailer that did that. Just because the victim here happens to be a massively rich, morally questionable company doesn't mean they are not a victim, and don't deserve to be protected by the same laws as everyone else.
You don't win big when you've got a good scam like that. It's tempting, but really, you just shouldn't do it, it's a dead give away.
Worst case I ever heard of: A guy who had worked on PNRGs for casinos (yes, way back when such things were deemed good enough) decided to cash in, so he got together with a friend and wrote a quick program to sync in the the PRNG given a reasonable number of inputs. The PRNGs were mostly (and still are sometimes!) used for the keno games. He had his friend up in the hotel room with a laptop, and phoned up the numbers from a few rounds of keno. They got what seemed to be a reasonable sync, so he put a massive amount of cash predicting the next 10 numbers in order (which has stupendous returns (naturally)). Bang, up come all 10 numbers, in order. The police arrested his accomplice in the hotel room about 10 minutes later...
The downside would be that not 'everyone' can use a PC, the way they can today, since MS Windows is by far the most newbie-friendly operating system availible for PC.
Did you get around to playing the BeOS? Now that was a user friendly operating system, and it was available for PC. Easily comparable in ease of use to Macintosh, and realistically better than Windows. Of course, it never had quite the same level of hardware support and applications available to compare to Windows - but then if there wasn't any Windows, all of a sudden that wouldn't be an issue.
Unfortunately (mostly due to aggressive leveraging of monopoly upon OEMs forcing manufacturers to never provide PCs with BeOS preinstalled) BeOS is now effectively dead. There are some efforts to resurrect it, but if you actually want to give the real thing a go, some interesting licensing games ended up with YellowTab managing to sell what amounts to BeOS 6 as Zeta. Check it out, it is actually very impressive.
A Microsoft spokeswoman said that many of these newly disclosed documents were not relevant to the trial, which focuses on Microsoft pricing actions.
Oh, of course, sorry. Yes, these documents aren't relevant for the current trial, so we should just ignore them completely and pretend they don't exist.
"These are not the documents you are looking for..."
There are rules for typesetting documents. TeX (and by extension, LaTeX) uses those rules.
So the standard, boring "letter" and "article" styles that just scream out: "THIS DOCUMENT WAS TYPESET IN LATEX!" are the rule and we shouldn't deviate from them?
Nope, by all means write your own - it's not that hard. I wrote up some company standard document classes at one company I worked at, providing a standardised (but very different from the standard TeX look) for all company documents produced in TeX. The bonus was that I wrote 2 document classes - 1 for documents, 1 for presentations, in such a way that providing you added \summary{summary of paragraph here} at the beginning of sections and paragraphs the same latex source could produce a document or a presentation depending on which document class you used - very useful.
Orginally I did this because I was in the research department and needed to typeset a lot of mathematics. Typsetting math in Word is bad enough, typesettig it in Powerpoint is utterly diabolical. In the end, however, I got known for producing the best looking documents in the company - my document class closely mirrored the Word template (and was close enough for the marketing department to okay it:-), but the better fonts, and better layout engine of TeX just made the whole thing look nicer. The differences were subtle, but the end result was a much better looking, more professional document.
It doesn't take much effort at all to write a new document class for LaTeX - it sounds hard, but it is easy to inherit from one of the base classes and then just rejig things to suit your preferred style. It is definitely worth the time and effort!
What about other critical features being able to place figures and text-frames exactly where you want them (and not where LaTeX wants to misplace them) or tracking changes/version control?
Minipages, parboxes, and styles like floatflt all make complex figure placement quite painless (certainly no harder than complex figure placement in MS Word). As for version control and change tracking - given that latex is pure text it is pretty damn easy to keep latex files in CVS which provides far better version control than MS Word. If you really want, you can keep latex documents in Visual SourceSafe, as I once did at a Windows based company.
Do I sound annoyed? Well, I am annoyed. You would be too if your every PhD student would initially insist on using LaTeX for his manuscripts ("i'm not gonna touch M$ word with 10ft pole!") and then expects me to make notes on a print-out.
Really? They must be quite slow then - I just use pdflatex and get people to use PDF annotation facilities to make notes - works brilliantly.
I wonder what this major result is going to be? KPanel? Metaciwin? Konqilus?
How about throwing some weight behind the development of this project? It wouldn't take too much effort to have this remarkably robust and efficient (it already works very well for all basic Qt themes, but struggles a bit with more complicated ones - Mosfet's liquid themes etc.), and having that done you could quite easily claim a unified look for all apps without any hassle - plus a single point at which to change themes - change the KDE style and it automatically effects all GTK apps.
There are plenty of good easy goals to aim for here.
Other than going for OpenBSD and lacking some functionality, what else do you propose?
How about making SELinux with a good default security policy the standard setup for all distributions using the 2.6 kernel?
The quality and power of SELinux in terms of security is literally light years ahead of any other commonly available Operating system (except, perhaps an obscure BSD fork which I believe was implementing a similar security structure).
Honestly, SELinux really is that good, and has been fully folded into the 2.6 kernel. People just need to start using it.
Argh. Too many stupid jokes are the only things getting moderated up at the moment.
I'm so pleased to have heard about this - I've been after this sort of thing for a long time now: case mods etc. that have elegant or classical styling instead of the usual "how many lights can I stick on it" crap.
I'm quite sick of beige boxes, and ugly designs - why can't more companies go for something like this? How about some nice brushed steel keyboards and mice? How about a nice (fake) tortoiseshell keyboard and mouse combo?
Apple turns out a ine array of beautiful elegant designs, but all the PC accessories just look like they escaped from the set of a cheap sci fi film set.
Yeah, except it looked more like the interviwer asking him what he'd do in that contingency rather than him bringing it up as a joke (in which case, I think he gave the correct response). It does make one wonder about the quality of journalists these days though...
But if Microsoft open-sourced Windows, which Messman said he did not think would happen as this was a huge cash cow for the Redmond, Wash., software company, Novell would help its customers use open-source Windows if this happened and they wanted it, he added.
If DeBeers starts giving away diamonds for free, we'll be sure to make sure our clients get some. In the event that a magic fairy inserts $50 billion into our bank account, we'll share that with our clients.
If a user does not know how to run a windows machine (keeping up to date on patches, running antivirus software, etc) then please explain to me how they'll be able to admin a linux machine.
No idea. An unfortunately MacOS X is also well known for it's extreme complexity and difficulty to use.
The reason most (or all) viruses are written for Windows is because that's where they'll do the most damage, since most people use Windows.
All fine and well, but it will help you if you switch, because then you'll be joining the happy minority that don't worry about such things.
Of course if everyone switches it will be a problem, but really, what are the odds of that actually happening?
It;s all fine and well to say "If everyone switched we'd still have the same problems with viruses", but realistically, everyone isn't going to switch. A lot of people are heavily locked into their current platform - so, if you can, switch...
Jedidiah.
Everyone has their own research division...
on
AT&T Labs' Brain Drain
·
· Score: 5, Insightful
It's a nice idea - every company has their own pure research division to solve all those interesting problems, and the IP stays within the company... except, very few companies can afford to do this.
Then again, look at what's come out of these sorts of pure research labs: C, UNIX, WIMP interfaces, etc., even Java, to some extent, could well be considered the output of such a process.
These aren't technologies you can bottle and sell. The value of these sorts of things is the productivity gains they provide. That's not to say the bottle and sell it approach hasn't been tried, but in the end the real meat is often in the abstract ideas, and even with the current patent system you can't patent purely abstract concepts. That is, all these ideas have been cloned, reinvented, or otherwise copied in one form or another.
Which brings me to my point - if you can't bottle it and sell it, if your competitors are just going to end up making a near duplicate anyway, why are you trying to fund this research lab all by yourself? No one doubts the quality of the work that can come out of these places, so why aren't there more cases of a group of various companuies banding together to fund a research group*? I'm not even talking about joining up with your direct competition - surely it wouldn't be that hard to have a group of companies that are not directly competing yet are all interested in managing to bring about a new, better, computer interface etc.
This "go it alone" attitude is sinking a lot of potentially incredibly valuable research simply because companies don't seem to be able to cooperate.
Jedidiah
* Note, for instance, that OSDL is exactly this sort of thing. A research group funded by a wide range of backers all interested in pushing forward computing. And it seems to be a model that's working well!
So with YaST going open source and having a much larger developer base willing to scratch odd itches, I wonder if we'll get a GNOME/GTK port of YaST that will get included in Ximian Desktop?
So is this a sign of the "We are really taking open source on board" that Novell has been trying to sell us, or is this just an internal SuSE decision? To be honest, I'm quietly hoping this was a Novel call, and that it's a sign that we have a big player really taking open source and GPL seriously. That, and hopefully it would be a sign that Novell might eventually start open sourcing some of their own applications, which would be a tremendous boost for FOSS.
Of course we need another office suite - as long as it supports compatible formats, who cares how many we have? Choice is good, and, more importantly a bit of competition is good. Right now everything is largely locked into the MS Office paradigm of how to do things, but there are other ways of doing these sorts of applications. The GoBe Productive suite, for instance, while not a direct MS Office offers a different and very nice style of doing some of these things. The more innovative and new thinking we can bring to the party the better we will be.
I really do fail to believe that the basic MS Office style word processor and spreadsheet are the pinnacle of design for such applications.
That's the price of being on the cutting edge of course. More patient individuals can wait for the clue patch to be merged into the kernel, and their distribution to release a nice binary of that kernel version. My understanding is that both RedHat and SuSE are working on this, but Lindows and Xandros will not be including the clue module in their future kernel packages, claiming their users can't deal with such complicated features. There is still much debate as to whether Mandrake should include it or not. On the other hand Gentoo and Slackware have similarly already decided not to include the clue patched kernel in their standard release, under the presumption that all their users already have it installed. Finally, the clue patch is anticipated to be available in Debian Stable some time in the next 4 years. It is already available in Debian Unstable but has been known to crash some (read "all") systems that have installed it.
I think he's talking about floating toolbars, which usually aren't included in the dock/taskbar on Windows or OS X.
This is probably an issue with ICCCM or whatever the WM protocol is called.
Thereshould be a simple "windowlist skip" hint that can be applied to windows - in fact, there already is - you notice the panel, and various other objects (which are windows - try a different window manager that doesn't take the hints and manage them like anything else) never show up on the task bar, so obviously such things are readily available.
But I want to have GIMP in one button, but separate xterms in separate buttons, so how do I do that?
Use a window manager that can handle such things properly would be a good start. Enlightenment has "Window Groups" where you can configure a set of windows to all respond to actions (so, for instance, iconifying one window in the group iconifies them all). You can also configure which actions are handled by the whole group, and which are individual (so maximzing a window is done by a single window, not everything in the group.
In my experience enlightenment's handling of this was a little clunky (it took too much tme to "configure"), but it was very effective once you had it going.
The real question is: why are the standard window managers for GNOME and KDE (metacity and kwin) not implementing something like this - it seems obvious, and if it can be made to hook with the taskbar, then all your problems are solved.
The key here is that, as other people have pointed out, having and MDI means the program with the MDI has to reimplement it's own window handling (I always hated that about MDI apps in windows - their window hanling inside the MDI sucked), which is stupid when you have window managers to handle that sort of thing - what ought to be happening (instead of forcing applications to write their own MDI) is for window managers to be taking on enough functionality to make MDI like behaviour trivial for a user to arrange.
Unfortunately, I wouldn't hold your breath on that one - not from mainstream window managers anyway.
Hmmm, comparing KDE with GTK I would have to say mostly it follows the usual difference between GNOME and KDE that has been apparent in the last year or two: GNOME has focussed on a slimmed down, simplified model with emphasis on clean and simple, while KDE has focussed on providing options.
To be honest, however, from what I've gathered the GNOME people have been far more influenced by Apple than KDE.
And finally - when you come down to it, it's a file selector, there;s not a whole lot innovative you can do with it. The KDE file selector doesn't look overly different from the Windows one, so really, is it any surprise that GNOME follows a vaguely similar line?
Slashdot Mirror
Well, no, it amounted to a crime because basically they effectively sold the casinos a product saying "Don't worry, this won't get hacked, it's solid", and then used their inside knowledge to hack it themselves. This amounted to defrauding the casinos - little better than an online retailer saying "don't worry, we'll keep your credit card information safe" and then promptly maxing out the credit cards of anyone who shops there. I imagine you would want to see law enforcement involved with any online retailer that did that. Just because the victim here happens to be a massively rich, morally questionable company doesn't mean they are not a victim, and don't deserve to be protected by the same laws as everyone else.
Jedidiah.
They broke the contract involving confidentiality they signed when working for the company that wrote the PRNG fir starters,
I believe some sort of fraud charges also resulted.
Jedidiah.
You don't win big when you've got a good scam like that. It's tempting, but really, you just shouldn't do it, it's a dead give away.
Worst case I ever heard of: A guy who had worked on PNRGs for casinos (yes, way back when such things were deemed good enough) decided to cash in, so he got together with a friend and wrote a quick program to sync in the the PRNG given a reasonable number of inputs. The PRNGs were mostly (and still are sometimes!) used for the keno games. He had his friend up in the hotel room with a laptop, and phoned up the numbers from a few rounds of keno. They got what seemed to be a reasonable sync, so he put a massive amount of cash predicting the next 10 numbers in order (which has stupendous returns (naturally)). Bang, up come all 10 numbers, in order. The police arrested his accomplice in the hotel room about 10 minutes later...
Jedidiah.
The downside would be that not 'everyone' can use a PC, the way they can today, since MS Windows is by far the most newbie-friendly operating system availible for PC.
Did you get around to playing the BeOS? Now that was a user friendly operating system, and it was available for PC. Easily comparable in ease of use to Macintosh, and realistically better than Windows. Of course, it never had quite the same level of hardware support and applications available to compare to Windows - but then if there wasn't any Windows, all of a sudden that wouldn't be an issue.
Unfortunately (mostly due to aggressive leveraging of monopoly upon OEMs forcing manufacturers to never provide PCs with BeOS preinstalled) BeOS is now effectively dead. There are some efforts to resurrect it, but if you actually want to give the real thing a go, some interesting licensing games ended up with YellowTab managing to sell what amounts to BeOS 6 as Zeta. Check it out, it is actually very impressive.
Jedidiah.
They could always license Acrobat from Adobe and integrate that in...
Jedidiah
From the article:
A Microsoft spokeswoman said that many of these newly disclosed documents were not relevant to the trial, which focuses on Microsoft pricing actions.
Oh, of course, sorry. Yes, these documents aren't relevant for the current trial, so we should just ignore them completely and pretend they don't exist.
"These are not the documents you are looking for..."
Jedidiah.
Nope, by all means write your own - it's not that hard. I wrote up some company standard document classes at one company I worked at, providing a standardised (but very different from the standard TeX look) for all company documents produced in TeX. The bonus was that I wrote 2 document classes - 1 for documents, 1 for presentations, in such a way that providing you added \summary{summary of paragraph here} at the beginning of sections and paragraphs the same latex source could produce a document or a presentation depending on which document class you used - very useful.
Orginally I did this because I was in the research department and needed to typeset a lot of mathematics. Typsetting math in Word is bad enough, typesettig it in Powerpoint is utterly diabolical. In the end, however, I got known for producing the best looking documents in the company - my document class closely mirrored the Word template (and was close enough for the marketing department to okay it
It doesn't take much effort at all to write a new document class for LaTeX - it sounds hard, but it is easy to inherit from one of the base classes and then just rejig things to suit your preferred style. It is definitely worth the time and effort!
Jedidiah.
What about other critical features being able to place figures and text-frames exactly where you want them (and not where LaTeX wants to misplace them) or tracking changes/version control?
Minipages, parboxes, and styles like floatflt all make complex figure placement quite painless (certainly no harder than complex figure placement in MS Word). As for version control and change tracking - given that latex is pure text it is pretty damn easy to keep latex files in CVS which provides far better version control than MS Word. If you really want, you can keep latex documents in Visual SourceSafe, as I once did at a Windows based company.
Do I sound annoyed? Well, I am annoyed. You would be too if your every PhD student would initially insist on using LaTeX for his manuscripts ("i'm not gonna touch M$ word with 10ft pole!") and then expects me to make notes on a print-out.
Really? They must be quite slow then - I just use pdflatex and get people to use PDF annotation facilities to make notes - works brilliantly.
Jedidiah
I wonder what this major result is going to be? KPanel? Metaciwin? Konqilus?
How about throwing some weight behind the development of this project? It wouldn't take too much effort to have this remarkably robust and efficient (it already works very well for all basic Qt themes, but struggles a bit with more complicated ones - Mosfet's liquid themes etc.), and having that done you could quite easily claim a unified look for all apps without any hassle - plus a single point at which to change themes - change the KDE style and it automatically effects all GTK apps.
There are plenty of good easy goals to aim for here.
Jedidiah.
You just contradicted your sig.
Jedidiah.
Other than going for OpenBSD and lacking some functionality, what else do you propose?
How about making SELinux with a good default security policy the standard setup for all distributions using the 2.6 kernel?
The quality and power of SELinux in terms of security is literally light years ahead of any other commonly available Operating system (except, perhaps an obscure BSD fork which I believe was implementing a similar security structure).
Honestly, SELinux really is that good, and has been fully folded into the 2.6 kernel. People just need to start using it.
Jedidiah
15" TFT 458 Euros
17" TFT 604 Euros
19" TFT 1090 Euros
Which is not all that bad considering how nice they look - here's a german shop selling them for those that are finding the site slashdotted.
Jedidiah
Argh. Too many stupid jokes are the only things getting moderated up at the moment.
I'm so pleased to have heard about this - I've been after this sort of thing for a long time now: case mods etc. that have elegant or classical styling instead of the usual "how many lights can I stick on it" crap.
I'm quite sick of beige boxes, and ugly designs - why can't more companies go for something like this? How about some nice brushed steel keyboards and mice? How about a nice (fake) tortoiseshell keyboard and mouse combo?
Apple turns out a ine array of beautiful elegant designs, but all the PC accessories just look like they escaped from the set of a cheap sci fi film set.
About bloody time, that's all I have to say.
Jedidiah.
Yeah, except it looked more like the interviwer asking him what he'd do in that contingency rather than him bringing it up as a joke (in which case, I think he gave the correct response). It does make one wonder about the quality of journalists these days though...
Jedidiah.
From the article:
But if Microsoft open-sourced Windows, which Messman said he did not think would happen as this was a huge cash cow for the Redmond, Wash., software company, Novell would help its customers use open-source Windows if this happened and they wanted it, he added.
If DeBeers starts giving away diamonds for free, we'll be sure to make sure our clients get some. In the event that a magic fairy inserts $50 billion into our bank account, we'll share that with our clients.
Sure. Whatever.
Jedidiah.
If a user does not know how to run a windows machine (keeping up to date on patches, running antivirus software, etc) then please explain to me how they'll be able to admin a linux machine.
No idea. An unfortunately MacOS X is also well known for it's extreme complexity and difficulty to use.
Jedidiah.
Switching won't really help.
The reason most (or all) viruses are written for Windows is because that's where they'll do the most damage, since most people use Windows.
All fine and well, but it will help you if you switch, because then you'll be joining the happy minority that don't worry about such things.
Of course if everyone switches it will be a problem, but really, what are the odds of that actually happening?
It;s all fine and well to say "If everyone switched we'd still have the same problems with viruses", but realistically, everyone isn't going to switch. A lot of people are heavily locked into their current platform - so, if you can, switch...
Jedidiah.
It's a nice idea - every company has their own pure research division to solve all those interesting problems, and the IP stays within the company... except, very few companies can afford to do this.
Then again, look at what's come out of these sorts of pure research labs: C, UNIX, WIMP interfaces, etc., even Java, to some extent, could well be considered the output of such a process.
These aren't technologies you can bottle and sell. The value of these sorts of things is the productivity gains they provide. That's not to say the bottle and sell it approach hasn't been tried, but in the end the real meat is often in the abstract ideas, and even with the current patent system you can't patent purely abstract concepts. That is, all these ideas have been cloned, reinvented, or otherwise copied in one form or another.
Which brings me to my point - if you can't bottle it and sell it, if your competitors are just going to end up making a near duplicate anyway, why are you trying to fund this research lab all by yourself? No one doubts the quality of the work that can come out of these places, so why aren't there more cases of a group of various companuies banding together to fund a research group*? I'm not even talking about joining up with your direct competition - surely it wouldn't be that hard to have a group of companies that are not directly competing yet are all interested in managing to bring about a new, better, computer interface etc.
This "go it alone" attitude is sinking a lot of potentially incredibly valuable research simply because companies don't seem to be able to cooperate.
Jedidiah
* Note, for instance, that OSDL is exactly this sort of thing. A research group funded by a wide range of backers all interested in pushing forward computing. And it seems to be a model that's working well!
So with YaST going open source and having a much larger developer base willing to scratch odd itches, I wonder if we'll get a GNOME/GTK port of YaST that will get included in Ximian Desktop?
Anyone want to give some odds?
Jedidiah.
So is this a sign of the "We are really taking open source on board" that Novell has been trying to sell us, or is this just an internal SuSE decision? To be honest, I'm quietly hoping this was a Novel call, and that it's a sign that we have a big player really taking open source and GPL seriously. That, and hopefully it would be a sign that Novell might eventually start open sourcing some of their own applications, which would be a tremendous boost for FOSS.
Jedidiah.
Of course we need another office suite - as long as it supports compatible formats, who cares how many we have? Choice is good, and, more importantly a bit of competition is good. Right now everything is largely locked into the MS Office paradigm of how to do things, but there are other ways of doing these sorts of applications. The GoBe Productive suite, for instance, while not a direct MS Office offers a different and very nice style of doing some of these things. The more innovative and new thinking we can bring to the party the better we will be.
I really do fail to believe that the basic MS Office style word processor and spreadsheet are the pinnacle of design for such applications.
Jedidiah.
That's the price of being on the cutting edge of course. More patient individuals can wait for the clue patch to be merged into the kernel, and their distribution to release a nice binary of that kernel version. My understanding is that both RedHat and SuSE are working on this, but Lindows and Xandros will not be including the clue module in their future kernel packages, claiming their users can't deal with such complicated features. There is still much debate as to whether Mandrake should include it or not. On the other hand Gentoo and Slackware have similarly already decided not to include the clue patched kernel in their standard release, under the presumption that all their users already have it installed. Finally, the clue patch is anticipated to be available in Debian Stable some time in the next 4 years. It is already available in Debian Unstable but has been known to crash some (read "all") systems that have installed it.
Hope this helps.
Jedidiah.
I think he's talking about floating toolbars, which usually aren't included in the dock/taskbar on Windows or OS X.
This is probably an issue with ICCCM or whatever the WM protocol is called.
Thereshould be a simple "windowlist skip" hint that can be applied to windows - in fact, there already is - you notice the panel, and various other objects (which are windows - try a different window manager that doesn't take the hints and manage them like anything else) never show up on the task bar, so obviously such things are readily available.
Jedidiah.
But I want to have GIMP in one button, but separate xterms in separate buttons, so how do I do that?
Use a window manager that can handle such things properly would be a good start. Enlightenment has "Window Groups" where you can configure a set of windows to all respond to actions (so, for instance, iconifying one window in the group iconifies them all). You can also configure which actions are handled by the whole group, and which are individual (so maximzing a window is done by a single window, not everything in the group.
In my experience enlightenment's handling of this was a little clunky (it took too much tme to "configure"), but it was very effective once you had it going.
The real question is: why are the standard window managers for GNOME and KDE (metacity and kwin) not implementing something like this - it seems obvious, and if it can be made to hook with the taskbar, then all your problems are solved.
The key here is that, as other people have pointed out, having and MDI means the program with the MDI has to reimplement it's own window handling (I always hated that about MDI apps in windows - their window hanling inside the MDI sucked), which is stupid when you have window managers to handle that sort of thing - what ought to be happening (instead of forcing applications to write their own MDI) is for window managers to be taking on enough functionality to make MDI like behaviour trivial for a user to arrange.
Unfortunately, I wouldn't hold your breath on that one - not from mainstream window managers anyway.
Jedidiah.
To be honest, however, from what I've gathered the GNOME people have been far more influenced by Apple than KDE.
And finally - when you come down to it, it's a file selector, there;s not a whole lot innovative you can do with it. The KDE file selector doesn't look overly different from the Windows one, so really, is it any surprise that GNOME follows a vaguely similar line?
Jedidiah.