Eh? I have for a friend installed mplayer completely in a single command-line: apt-get install mplayer. I usually emerge mplayer. xine I've fallen more out of favor (on average, mplayer gets support for new codecs faster, and pays more attention to the native movie resolution and aspect ration than xine, but every time I have toyed with it it has been straightforward.
In answer to your question, it depends. If the application is threaded, then yes. An OS will not divide instructions within a single process or thread. However, many modern apps are threaded, or at least the Toolkit they make use of can take advantage of threading, and there is almost never a situation where only a single, single-threaded application is the only thing running on a system in this day and age.
OpenLDAP with SSL works fine in my experience, never had a headache from setup to implementation. Active Directory does indeed work out of the box. But when third party tools need to interact with it or you need some schema changes, things can go to hell in a handbasket quickly.
Your argument about client policy management is referring to Windows client policy management. I will give that Windows is superior to Linux at distributing policies to clients, but we are talking about Linux across the board. You have a lot more power across the board when you don't have to rely on samba to accomplish things. Also, with NT4 clients (which is often unavoidable in Windows networks on a budget), Samba actually offers a bit more power and flexibilty when dealing with those 'legacy' clients.
Remote software installation better on Windows? You have got to be kidding. Some applications do work fine for doing convenient remote applications. Sometimes Terminal Services is required. I have seen apps that will only successfully install from the console (or, by extension VNC).
I'll admit the MMC is a decent remote administration tool, but I would not give it as much credit to say it is good at managing multiple systems at once. I haven't really seen anything under windows that is any better than anything under linux as far as managing groups of computers at once.
Remote administration under Windows is much more of a pain than any *nix. Almost anything can be done through ssh and the system doesn't care. For gui, all X11 windows are created equal, whether local or remote. X11 is a bit talky in terms of bandwidth, but it is rarely needed. Windows administration first off requires GUI to be forwarded. Second off, Remote Desktop frequently behaves differently from the console, making VNC a requisite practically for those apps that break in RDP world. Why the hell VNC would be needed for much in Linux is beyond me. I rarely have to use X11 even.
And to say Windows 2000 is kerberos with no dicking around is a travesty. Have you ever tried to use the built-in facilities for anything other than Windows clients, or try to get Windows clients to authenticate against an alternative LDAP/Kerberos implementation? They bastardized kerberos just enough to make it desirable to be an all-ms shop. That is their business, making non-ms interoperating with MS too clunky to try. For an all Windows network it is fine, but in that case it might as well be something proprietary, so kerberos is just a buzz word hinting at interoperability that just isn't there.
You seem to have been comparing built in facilities to third party applications when oit comes to Enterprise monitoring. I haven't really bothered to try many third party products when it comes to this area, and I'm not sure what *exactly* you mean by enterprise monitoring specifically, so I'll leave this alone.
And finally, with regards to automatic updating. No sane administrator trying to maintain a consistant environment blindly runs auto-update. One, you test out patches before giving the big ok to mass deployment. For another, Windows updates requires reboots 99% of the time for update package installation. That really makes reliabily sink. If you are really crazy enough to do auto-updates and trust parties outside your organization, you can easily use up2date automatically or apt as a massive cron job.
My final point is that clearly you are a relatively seasoned Windows administrator. I have been in that role too. Both times they let me go in favor of a cheaper administration who was 'good enough'. These replacements often have no idea how to fully exploit the features available in Windows. When talking with them, they never know that AD is an LDAP system, or even what Kerberos is. The only thing they ever do is vnc (yes *vnc in*) to the domain controller to modify user accounts not realizing the power of mmc to make it easier. That is the extent to which they interact with AD. These are the people who cannot by themselves efficiently manage larger networks.
And it is becoming increasingly hard for businesses to tell the good from the bad. The market is so saturated of people who were pretty decent and jumped at the 'get your MCSE with us' commercials, that finding good administration is hard. Linux scares these people by and large, so the market of Linux administrators is a lot more pure. If and when RHCE becomes 'hot' like mcse, you'll see a lot more junk Linux admins too...
I agree that people are brainwashed to think plain text is evil... XML is an improvement, since it is typically human readable, but if it is a smallish application with a reasonably small search space, then plain text, or even XML is much better than a binary DB.
However, I think something on the scale of a filesystem needs something more efficient, both in terms of storage space efficiency and performance. A plain text fs index would be large and slow to query compared with an optimized database format. That does not mean, of course, that the data has to have special tools to access it. Quite the contrary, when implemented at the OS level, textual representations can be generated rather quickly on the fly for editing if you so choose. I seem to recall reading up on some of the ideas in the next-gen Reiser filesystem. Their example was/etc/passwd as a directory, able to cd in and see the data listed in different ways simply by going to different directories and doing an ls. Individual entries could be opened in a text editor, or vi/etc/passwd and you get the view you are accustomed to. It's all the same to the filesystem, no matter how you do it. If the lowlevel filesystem drivers are aware of the workings of the database, they can and should implement transparent text editing facilities (and structures convenient to shells). The MS registry is a train wreck for many reasons. One of those is they never intended on heavy end-user direct modification, so it is ugly and not well integrated in terms of access. It *could* be as easy as editing a text file when implemented at a low enough level, it's just that people don't.
Of course I don't agree with the goal of this article either. Too obtrusive to the user. I would prefer something along the lines of 'locate' with more sophisticated and up-to-date indexing methods and more comprehensive attribute recording. Take the many many hints already available rather than asking a user to provide more. Would have to be implemented on a lower level to do it right, but a small price to pay to avoid using find when up-to-the-minute info is needed.
Nothing about the file systems prohibit your functionality, that is all application level stuff. Hard/Soft Links are there if the save dialog supported it. But for the common user, I think that dialog would be excessively confusing (I can't imagine how you would offer the selection graphically, and whatever delimiter you choose if by text could cause a great deal of confusion to a user unfamiliar with it who just happens to try to use the character as part of a single filename). For those that want it, there are links, aliases,.lnk files in the appropriate operating systems, all easily accessible through file managers. Too many people and systems focus and getting as far away from the file manager as possible, turning it into basically a launch platform and having most file management tasks done in the open/save dialogs. This is sloppy. I like some ideas as exemplified by rox (http://rox.sf.net/). Applications developed purely with rox in mind don't have open/save dialogs, they offer convenient methods of calling up a rox window in a reasonable location. Both loading and saving are done through drag and drop. Smoothly integrating the file manager with applications works very well, and means the user does not have to become accustomed to different, inconsistant methods of file management. Operations should be intuitive, consistant, and symmetric. If you can drag and drop to open a file (and all apps should open the file, not paste it in as some do), the same should be possible to save. Of course this is all unreasonable now, people are used to open/save dialogs and this just would be confusing, but it would have been nice if things had gone that way in the mainstream....
For two, I think application preferences regarding this matter work well. Imagine that you had a system that did choose the default directory based on a 'find -type f -exec file {};' sort of mechanism (and by some miracle worked without a performance hit). This means the default directory could change between two save dialogs opening without the user necessarily doing anything to change this. Say they do spring cleaning on a directory of spreadsheets, and categorize each and put them all in different directories. Now they make a new one, go to save it, probably not paying attention to the path since they are used to knowing where it goes, and it picks the biggest category. Incorrect behavior.
I have not known a single person who desperately needs or wants a revolution with regard to file organization. They almost always know where to look for things, and in the rare event they do not, file search is there. Boosting the speed and complexity of that functionality without impacting the general user experience is where the most good can be done. No reason to subject computer users in general to a file organization paradigm shift (not to mention legacy applications).
Care to back that up with anything at all? Too many flaws.
One, assuming your statement is correct (which I'm sure it isn't) nothing about Serial ATA demands platter based devices.
I cannot see how you can make that statement anyway. You could make a stronger argument for CD-ROM. Drives there have no control over the media they read in terms of durability. They know about standards, and generally must cater to the least common denominator in terms of spinning discs so fast they shatter. There the limit has been in the 50x-60x area with a single laser reading. But now you have multiple lasers to read different parts of the disc, reducing seek times and throughput beyond what was the 'physical limit' of a single laser reading a single disc spinning no faster than 50-60x read speed. So drive heads can change in design to exceed whatever logic you see.
Also, a hard drive is a highly controlled environment. The materials chosen for the platters is well known and RPM can be pumped up. When material fails, you can change the radius, thickness, and to an extent the material you use to get higher RPM. Beyond that, you can use multiple platters with independent drive heads, to acheive a highly controlled RAID-0 performance boost within the drive.
GTK+ is not the worst. In fact, with gtkmm or PyGtk, it is quite palleteable. I would say gtkmm and qt are on equal footing.
As far as UI design is concerned, it really reflects the different design goals of GTK+ versus Apple APIs. GTK+ strives to provide maximal flexibility. Let the developer chose however he wants to design the app and provide some way to implement it. Flys in the face of consistancy, but there is a reason. Apple kills flexibility to enforce consistancy. Though I respect both goals, and think GTK+OSX is a REALLY good step for providing ports of complex apps quickly versus XDarwin, I agree that developers should consider how badly GTK+ apps will stick out like a sore thumb in OSX. They can work to make some widgets more seamless (i.e. menubar, button appearance, etc.), but others will always stand out (detachable handleboxes, for example. They can be quite useful (on multi-head configurations), but I don't think they are anywhere to be found in OSX).
I know, not cross-platform, but the goodness of python in Cocoa...
As to the original story, I see the value as a porting mechanism of 'legacy' applications, but I fear that it will become a crutch for those who would otherwise provide a native Cocoa port. XDarwin already makes a lot of projects view a native Cocoa port as not absolutely necessary and not worth the effort. I would agree.... IF XDarwin was hardware accelerated, and worked perfectly consistantly in rootless mode, but neither is the case, even with OrborosX..
It's called IPSec. Dammit, just because it is used almost exclusively as a VPN solution, no one considers it as a solution for wireless security, and it indeed is a kick ass solution. I allow on my personal network access to FORWARD only if they are able to establish an IPSec connection. INPUT only on ESP and UDP port 500 (and DNS, ssh, for other reasons). I feel a lot more confident in this setup. The only problem I still foresee is that of systems on the same network being able to talk to each other unencrypted. The setup of mine (as well as most), however, is that servers are wired and wireless entities exchange no significant traffic between themselves...
They have parachutes in several smaller planes. In fact, I think in the past couple of months the first emergency-use of such a chute in a personal aircraft occured and was successful. I understood there were some obstacles in getting a parachute system that would be effective to work for a small craft, and that there is no way in hell you could support the weight of even the passenger compartment by itself.
Re:Don't forget killer whales...
on
Starcraft
·
· Score: 2
Not Killer whales, humpback whales.
Killer whales aren't even endangered or anything, they are doing great in the wild still...
It isn't in their best interest though. Everyone would jump on the do-not-call list. A lot of people say they never would buy from a telemarketer, but they sometimes cave, and that is what they would stand to lose. Don't like telemarketers, and I honestly don't listen to anyone, but I can see their logic.
A note to everyone out there doing this sort of thing. Use example.com, it can never be registered to a valid address and I've never seen an email address verification that rejects it.
I always use some string @example.com
If I have to see the email, I give them myemailaddress-randomstring@myrealdomain.com, and then put a.forward-randomstring in later to/dev/null. I have full control over my email server running postfix, so I can do whatever I want.
Filling out random email addresses that could point to valid addresses not owned by you is not necessary.
A small severance package. Yep, laid off yesterday, this close to Christmas. At the same time, invited to the company Christmas party. Yeah, I feel like going... I know, tough times and all, but seems like really crappy timing. Now if I get a better job before Christmas, well, I will consider this a good thing....
x86 is by far the most prominent platform for linux, and with good reason. It is the most widespread, accessible, and affordable platform that, hardware for hardware, is comparable or better than the other hardware out there. When people buy PPC, Sparc, PA-Risc or MIPS workstations, they are rarely ever buying for the sake of the hardware underneath, they are buying for the explicit purpose of running OSX, AIX, Solaris, HP-UX, or Irix, not to put linux on it. For that they could get high quality x86 hardware cheaper and do the same thing.
IT are never appreciated til they are gone...
on
The New IT Crisis
·
· Score: 2
I know three companies who in the last week have laid off their entire IT departments. They see that their servers are pretty much always up, and whatever IT is doing, the people managing do not see the impact. They canned everyone, and I'm just waiting to see the horrible mess that results.
As to the automation issue, it is pretty much already very easily possible, but is hard to get in place most places, because of managers making technical decisions they shouldn't be. They read papers and get enthralled with buzzwords and insist that the technology be used, despite how it will not fit in nicely with the existing configuration. Good equipment has great monitoring capabilities. On software, even Windows has decent capabilities for widespread automated and remote maintenance. I have had the experience of seeing a need, pointing out the problem and strongly recommend a solution, have management flat out go a different direction that I said would be a bad idea, and then be complained to that the solution was not working as well as it should be...
Semi-on topic, but flat out avoiding logging and squashing every flame that comes along is not necessarily in the best interest of the environment.
Protected the trees from fire and logging changes the environment. For example, some pine trees take a brute force approach of many many many seedlings that are individually weak and the expectation is that a fire will wipe most out, but the few that remain will be fine..... Others, take a long time to develop a few saplings while fireproofing. Without fires, these slow and steady species get starved out. I don't know too much about this stuff (my wife is the Bio/Botany/Genetics person), but moderating logging can serve as a good, efficient replacement for raging wildfires (which interfere too much with our lives to tolerate beyond small, controlled, often wasteful controlled burns).
A firewire drive in a fireproof safe, shouldn't they cancel each other out?
But seriously, putting tapes, drives, or even optical media in a fireproof safe I would not classify as being safe from fire, unless you have a really kick ass fireproof safe that *also* insures the heat level doesn't get high enough to cause problems. Just keeping the temperature below the combustion point of most materials isn't enough.....
It's newsworthy because they are not doing the standard MS thing. If they gave out any non-MS laptop, that is certainly news. Many Universities issue Windows laptops, and this goes against the grain. Probably more about economics (Apple giving price breaks) than philosophy, but still very intriguing.
Most office workers don't use their work email address for anything much, so no one knows about their email address aside from maybe family and business contacts, of course spam will be lower. Any sort of server-side mail filtering is a gamble, to say the least. If anything gets mis-catagorized, it could be a disaster.... My company has discussed it and decided it is better to let users deal with it than risk trashing legitimate mail.
The implication seems to be they just want a 30 day grace period. Does not seem to be unreasonable to me. After the 30 day grace period I guess it is open season. The only part that bothers me is that the terms seem to indicate that the 30 day period starts from their acknowledgement that you submitting a flaw, not from the time of acknowledgement. If they chose to not acknowledge responses in a timely manner, that could be a problematic loophole.
Flash player 6, and actually, none of the sites I typically am visiting have any flash content or ads... Strange but true. Trying a nightly now to see how it goes...
I'm sorry, but for all that Chimera is hailed, is is a piece of crap browser. I've been using it and it just crashes constantly and lacks a lot of features. Features that I sorely miss from mozilla/phoenix while using Chimera are:
Smart Bookmarks (searching from location bar very convenient, am using what I feel is a kludge of a javascript monstrosity set as my search page to search by selection or pop up a dialog if there is no selection, decent, but still not as cool).
Type-ahead find: very nifty feature.
Ability to have hrefs that request new windows open in tabs. I like tabs and don't like sites breaking my preferred usage paradigm.
Freaking close buttons on the tabs. I hate having to right click, control-click, or click and hold to close a tab that is not the active tab. Just annoying.
The first is to me the biggest issue. I just had to rant that Chimera is not 'all-that'. If it didn't crash so much and at *least* had smart bookmarks, then maybe. OmniWeb and IE are just too feature barren, Opera misrenders some important pages to me, and Mozilla is too slow. Phoenix has been decent, but middle-click doesn't work and sometimes it gets a bit confused in the MacOSX builds... Well, enough of my rant..
I'm looking at a new system for HTPC purposes. Since the mplayer/freevo/xmame/zsnes combo is such a killer ui-wise (been trying it on my desktop to evaluate), linux seems the way to go.
However, the problem becomes TV-out. Now it looks like nVidia, Matrox, and ATI have decent TV-out capabilities.
Matrox 3D is unacceptable, so I've discounted that (some pretty neat GL visualizations on xmms wouldn't look as good), so I'm left with ATI and nVidia. Researched nVidia and within two minutes saw that TV was fully supported in the binary-only drivers. On ATI's side of the fence.... they neither seem to have TV-out support in any binary form nor are willing to release specs (or else face the wrath of ??AA for providing a path for Macrovision to be disabled), so the only TV-out support is through using an unaccelerated framebuffer, completely unacceptable.
ATI's efforts seem to be less comprehensive than nVidia's, both in terms of scope (only 8500 and 9700?), and in terms of functionality. All of this stuff should not be exclusive with respect to each other. I would love to see this happen, as there are a couple of decent motherboard's with built in TV-Out that use RV100 chips...
Slashdot Mirror
Eh? I have for a friend installed mplayer completely in a single command-line: apt-get install mplayer. I usually emerge mplayer. xine I've fallen more out of favor (on average, mplayer gets support for new codecs faster, and pays more attention to the native movie resolution and aspect ration than xine, but every time I have toyed with it it has been straightforward.
In answer to your question, it depends. If the application is threaded, then yes. An OS will not divide instructions within a single process or thread. However, many modern apps are threaded, or at least the Toolkit they make use of can take advantage of threading, and there is almost never a situation where only a single, single-threaded application is the only thing running on a system in this day and age.
I know, flamebait, but I'll bite.
OpenLDAP with SSL works fine in my experience, never had a headache from setup to implementation. Active Directory does indeed work out of the box. But when third party tools need to interact with it or you need some schema changes, things can go to hell in a handbasket quickly.
Your argument about client policy management is referring to Windows client policy management. I will give that Windows is superior to Linux at distributing policies to clients, but we are talking about Linux across the board. You have a lot more power across the board when you don't have to rely on samba to accomplish things. Also, with NT4 clients (which is often unavoidable in Windows networks on a budget), Samba actually offers a bit more power and flexibilty when dealing with those 'legacy' clients.
Remote software installation better on Windows? You have got to be kidding. Some applications do work fine for doing convenient remote applications. Sometimes Terminal Services is required. I have seen apps that will only successfully install from the console (or, by extension VNC).
I'll admit the MMC is a decent remote administration tool, but I would not give it as much credit to say it is good at managing multiple systems at once. I haven't really seen anything under windows that is any better than anything under linux as far as managing groups of computers at once.
Remote administration under Windows is much more of a pain than any *nix. Almost anything can be done through ssh and the system doesn't care. For gui, all X11 windows are created equal, whether local or remote. X11 is a bit talky in terms of bandwidth, but it is rarely needed. Windows administration first off requires GUI to be forwarded. Second off, Remote Desktop frequently behaves differently from the console, making VNC a requisite practically for those apps that break in RDP world. Why the hell VNC would be needed for much in Linux is beyond me. I rarely have to use X11 even.
And to say Windows 2000 is kerberos with no dicking around is a travesty. Have you ever tried to use the built-in facilities for anything other than Windows clients, or try to get Windows clients to authenticate against an alternative LDAP/Kerberos implementation? They bastardized kerberos just enough to make it desirable to be an all-ms shop. That is their business, making non-ms interoperating with MS too clunky to try. For an all Windows network it is fine, but in that case it might as well be something proprietary, so kerberos is just a buzz word hinting at interoperability that just isn't there.
You seem to have been comparing built in facilities to third party applications when oit comes to Enterprise monitoring. I haven't really bothered to try many third party products when it comes to this area, and I'm not sure what *exactly* you mean by enterprise monitoring specifically, so I'll leave this alone.
And finally, with regards to automatic updating. No sane administrator trying to maintain a consistant environment blindly runs auto-update. One, you test out patches before giving the big ok to mass deployment. For another, Windows updates requires reboots 99% of the time for update package installation. That really makes reliabily sink. If you are really crazy enough to do auto-updates and trust parties outside your organization, you can easily use up2date automatically or apt as a massive cron job.
My final point is that clearly you are a relatively seasoned Windows administrator. I have been in that role too. Both times they let me go in favor of a cheaper administration who was 'good enough'. These replacements often have no idea how to fully exploit the features available in Windows. When talking with them, they never know that AD is an LDAP system, or even what Kerberos is. The only thing they ever do is vnc (yes *vnc in*) to the domain controller to modify user accounts not realizing the power of mmc to make it easier. That is the extent to which they interact with AD. These are the people who cannot by themselves efficiently manage larger networks.
And it is becoming increasingly hard for businesses to tell the good from the bad. The market is so saturated of people who were pretty decent and jumped at the 'get your MCSE with us' commercials, that finding good administration is hard. Linux scares these people by and large, so the market of Linux administrators is a lot more pure. If and when RHCE becomes 'hot' like mcse, you'll see a lot more junk Linux admins too...
I agree that people are brainwashed to think plain text is evil... XML is an improvement, since it is typically human readable, but if it is a smallish application with a reasonably small search space, then plain text, or even XML is much better than a binary DB.
/etc/passwd as a directory, able to cd in and see the data listed in different ways simply by going to different directories and doing an ls. Individual entries could be opened in a text editor, or vi /etc/passwd and you get the view you are accustomed to. It's all the same to the filesystem, no matter how you do it. If the lowlevel filesystem drivers are aware of the workings of the database, they can and should implement transparent text editing facilities (and structures convenient to shells). The MS registry is a train wreck for many reasons. One of those is they never intended on heavy end-user direct modification, so it is ugly and not well integrated in terms of access. It *could* be as easy as editing a text file when implemented at a low enough level, it's just that people don't.
However, I think something on the scale of a filesystem needs something more efficient, both in terms of storage space efficiency and performance. A plain text fs index would be large and slow to query compared with an optimized database format. That does not mean, of course, that the data has to have special tools to access it. Quite the contrary, when implemented at the OS level, textual representations can be generated rather quickly on the fly for editing if you so choose. I seem to recall reading up on some of the ideas in the next-gen Reiser filesystem. Their example was
Of course I don't agree with the goal of this article either. Too obtrusive to the user. I would prefer something along the lines of 'locate' with more sophisticated and up-to-date indexing methods and more comprehensive attribute recording. Take the many many hints already available rather than asking a user to provide more. Would have to be implemented on a lower level to do it right, but a small price to pay to avoid using find when up-to-the-minute info is needed.
Nothing about the file systems prohibit your functionality, that is all application level stuff. Hard/Soft Links are there if the save dialog supported it. But for the common user, I think that dialog would be excessively confusing (I can't imagine how you would offer the selection graphically, and whatever delimiter you choose if by text could cause a great deal of confusion to a user unfamiliar with it who just happens to try to use the character as part of a single filename). For those that want it, there are links, aliases, .lnk files in the appropriate operating systems, all easily accessible through file managers. Too many people and systems focus and getting as far away from the file manager as possible, turning it into basically a launch platform and having most file management tasks done in the open/save dialogs. This is sloppy. I like some ideas as exemplified by rox (http://rox.sf.net/). Applications developed purely with rox in mind don't have open/save dialogs, they offer convenient methods of calling up a rox window in a reasonable location. Both loading and saving are done through drag and drop. Smoothly integrating the file manager with applications works very well, and means the user does not have to become accustomed to different, inconsistant methods of file management. Operations should be intuitive, consistant, and symmetric. If you can drag and drop to open a file (and all apps should open the file, not paste it in as some do), the same should be possible to save. Of course this is all unreasonable now, people are used to open/save dialogs and this just would be confusing, but it would have been nice if things had gone that way in the mainstream....
For two, I think application preferences regarding this matter work well. Imagine that you had a system that did choose the default directory based on a 'find -type f -exec file {};' sort of mechanism (and by some miracle worked without a performance hit). This means the default directory could change between two save dialogs opening without the user necessarily doing anything to change this. Say they do spring cleaning on a directory of spreadsheets, and categorize each and put them all in different directories. Now they make a new one, go to save it, probably not paying attention to the path since they are used to knowing where it goes, and it picks the biggest category. Incorrect behavior.
I have not known a single person who desperately needs or wants a revolution with regard to file organization. They almost always know where to look for things, and in the rare event they do not, file search is there. Boosting the speed and complexity of that functionality without impacting the general user experience is where the most good can be done. No reason to subject computer users in general to a file organization paradigm shift (not to mention legacy applications).
Care to back that up with anything at all? Too many flaws.
One, assuming your statement is correct (which I'm sure it isn't) nothing about Serial ATA demands platter based devices.
I cannot see how you can make that statement anyway. You could make a stronger argument for CD-ROM. Drives there have no control over the media they read in terms of durability. They know about standards, and generally must cater to the least common denominator in terms of spinning discs so fast they shatter. There the limit has been in the 50x-60x area with a single laser reading. But now you have multiple lasers to read different parts of the disc, reducing seek times and throughput beyond what was the 'physical limit' of a single laser reading a single disc spinning no faster than 50-60x read speed. So drive heads can change in design to exceed whatever logic you see.
Also, a hard drive is a highly controlled environment. The materials chosen for the platters is well known and RPM can be pumped up. When material fails, you can change the radius, thickness, and to an extent the material you use to get higher RPM. Beyond that, you can use multiple platters with independent drive heads, to acheive a highly controlled RAID-0 performance boost within the drive.
GTK+ is not the worst. In fact, with gtkmm or PyGtk, it is quite palleteable. I would say gtkmm and qt are on equal footing.
As far as UI design is concerned, it really reflects the different design goals of GTK+ versus Apple APIs. GTK+ strives to provide maximal flexibility. Let the developer chose however he wants to design the app and provide some way to implement it. Flys in the face of consistancy, but there is a reason. Apple kills flexibility to enforce consistancy. Though I respect both goals, and think GTK+OSX is a REALLY good step for providing ports of complex apps quickly versus XDarwin, I agree that developers should consider how badly GTK+ apps will stick out like a sore thumb in OSX. They can work to make some widgets more seamless (i.e. menubar, button appearance, etc.), but others will always stand out (detachable handleboxes, for example. They can be quite useful (on multi-head configurations), but I don't think they are anywhere to be found in OSX).
What about this?.
I know, not cross-platform, but the goodness of python in Cocoa...
As to the original story, I see the value as a porting mechanism of 'legacy' applications, but I fear that it will become a crutch for those who would otherwise provide a native Cocoa port. XDarwin already makes a lot of projects view a native Cocoa port as not absolutely necessary and not worth the effort. I would agree.... IF XDarwin was hardware accelerated, and worked perfectly consistantly in rootless mode, but neither is the case, even with OrborosX..
It's called IPSec. Dammit, just because it is used almost exclusively as a VPN solution, no one considers it as a solution for wireless security, and it indeed is a kick ass solution. I allow on my personal network access to FORWARD only if they are able to establish an IPSec connection. INPUT only on ESP and UDP port 500 (and DNS, ssh, for other reasons). I feel a lot more confident in this setup. The only problem I still foresee is that of systems on the same network being able to talk to each other unencrypted. The setup of mine (as well as most), however, is that servers are wired and wireless entities exchange no significant traffic between themselves...
They have parachutes in several smaller planes. In fact, I think in the past couple of months the first emergency-use of such a chute in a personal aircraft occured and was successful. I understood there were some obstacles in getting a parachute system that would be effective to work for a small craft, and that there is no way in hell you could support the weight of even the passenger compartment by itself.
Not Killer whales, humpback whales.
Killer whales aren't even endangered or anything, they are doing great in the wild still...
And they also aren't technically whales....
It isn't in their best interest though. Everyone would jump on the do-not-call list. A lot of people say they never would buy from a telemarketer, but they sometimes cave, and that is what they would stand to lose. Don't like telemarketers, and I honestly don't listen to anyone, but I can see their logic.
A note to everyone out there doing this sort of thing. Use example.com, it can never be registered to a valid address and I've never seen an email address verification that rejects it.
.forward-randomstring in later to /dev/null. I have full control over my email server running postfix, so I can do whatever I want.
I always use some string @example.com
If I have to see the email, I give them myemailaddress-randomstring@myrealdomain.com, and then put a
Filling out random email addresses that could point to valid addresses not owned by you is not necessary.
A small severance package. Yep, laid off yesterday, this close to Christmas. At the same time, invited to the company Christmas party. Yeah, I feel like going... I know, tough times and all, but seems like really crappy timing. Now if I get a better job before Christmas, well, I will consider this a good thing....
x86 is by far the most prominent platform for linux, and with good reason. It is the most widespread, accessible, and affordable platform that, hardware for hardware, is comparable or better than the other hardware out there. When people buy PPC, Sparc, PA-Risc or MIPS workstations, they are rarely ever buying for the sake of the hardware underneath, they are buying for the explicit purpose of running OSX, AIX, Solaris, HP-UX, or Irix, not to put linux on it. For that they could get high quality x86 hardware cheaper and do the same thing.
I know three companies who in the last week have laid off their entire IT departments. They see that their servers are pretty much always up, and whatever IT is doing, the people managing do not see the impact. They canned everyone, and I'm just waiting to see the horrible mess that results.
As to the automation issue, it is pretty much already very easily possible, but is hard to get in place most places, because of managers making technical decisions they shouldn't be. They read papers and get enthralled with buzzwords and insist that the technology be used, despite how it will not fit in nicely with the existing configuration. Good equipment has great monitoring capabilities. On software, even Windows has decent capabilities for widespread automated and remote maintenance. I have had the experience of seeing a need, pointing out the problem and strongly recommend a solution, have management flat out go a different direction that I said would be a bad idea, and then be complained to that the solution was not working as well as it should be...
Semi-on topic, but flat out avoiding logging and squashing every flame that comes along is not necessarily in the best interest of the environment.
Protected the trees from fire and logging changes the environment. For example, some pine trees take a brute force approach of many many many seedlings that are individually weak and the expectation is that a fire will wipe most out, but the few that remain will be fine..... Others, take a long time to develop a few saplings while fireproofing. Without fires, these slow and steady species get starved out. I don't know too much about this stuff (my wife is the Bio/Botany/Genetics person), but moderating logging can serve as a good, efficient replacement for raging wildfires (which interfere too much with our lives to tolerate beyond small, controlled, often wasteful controlled burns).
A firewire drive in a fireproof safe, shouldn't they cancel each other out?
But seriously, putting tapes, drives, or even optical media in a fireproof safe I would not classify as being safe from fire, unless you have a really kick ass fireproof safe that *also* insures the heat level doesn't get high enough to cause problems. Just keeping the temperature below the combustion point of most materials isn't enough.....
They want money, and they want guaranteed steady money projects, not just a chance at money with RSA.
It's newsworthy because they are not doing the standard MS thing. If they gave out any non-MS laptop, that is certainly news. Many Universities issue Windows laptops, and this goes against the grain. Probably more about economics (Apple giving price breaks) than philosophy, but still very intriguing.
Most office workers don't use their work email address for anything much, so no one knows about their email address aside from maybe family and business contacts, of course spam will be lower. Any sort of server-side mail filtering is a gamble, to say the least. If anything gets mis-catagorized, it could be a disaster.... My company has discussed it and decided it is better to let users deal with it than risk trashing legitimate mail.
The implication seems to be they just want a 30 day grace period. Does not seem to be unreasonable to me. After the 30 day grace period I guess it is open season. The only part that bothers me is that the terms seem to indicate that the 30 day period starts from their acknowledgement that you submitting a flaw, not from the time of acknowledgement. If they chose to not acknowledge responses in a timely manner, that could be a problematic loophole.
Flash player 6, and actually, none of the sites I typically am visiting have any flash content or ads... Strange but true. Trying a nightly now to see how it goes...
I'm sorry, but for all that Chimera is hailed, is is a piece of crap browser. I've been using it and it just crashes constantly and lacks a lot of features. Features that I sorely miss from mozilla/phoenix while using Chimera are:
Smart Bookmarks (searching from location bar very convenient, am using what I feel is a kludge of a javascript monstrosity set as my search page to search by selection or pop up a dialog if there is no selection, decent, but still not as cool).
Type-ahead find: very nifty feature.
Ability to have hrefs that request new windows open in tabs. I like tabs and don't like sites breaking my preferred usage paradigm.
Freaking close buttons on the tabs. I hate having to right click, control-click, or click and hold to close a tab that is not the active tab. Just annoying.
The first is to me the biggest issue. I just had to rant that Chimera is not 'all-that'. If it didn't crash so much and at *least* had smart bookmarks, then maybe. OmniWeb and IE are just too feature barren, Opera misrenders some important pages to me, and Mozilla is too slow. Phoenix has been decent, but middle-click doesn't work and sometimes it gets a bit confused in the MacOSX builds... Well, enough of my rant..
I'm looking at a new system for HTPC purposes. Since the mplayer/freevo/xmame/zsnes combo is such a killer ui-wise (been trying it on my desktop to evaluate), linux seems the way to go.
However, the problem becomes TV-out. Now it looks like nVidia, Matrox, and ATI have decent TV-out capabilities.
Matrox 3D is unacceptable, so I've discounted that (some pretty neat GL visualizations on xmms wouldn't look as good), so I'm left with ATI and nVidia. Researched nVidia and within two minutes saw that TV was fully supported in the binary-only drivers. On ATI's side of the fence.... they neither seem to have TV-out support in any binary form nor are willing to release specs (or else face the wrath of ??AA for providing a path for Macrovision to be disabled), so the only TV-out support is through using an unaccelerated framebuffer, completely unacceptable.
ATI's efforts seem to be less comprehensive than nVidia's, both in terms of scope (only 8500 and 9700?), and in terms of functionality. All of this stuff should not be exclusive with respect to each other. I would love to see this happen, as there are a couple of decent motherboard's with built in TV-Out that use RV100 chips...