What software run in a school/business environment needs to be run as an administrator? Stop spreading FUD.
It's not FUD.
I work in a small OCR shop. We scan alot of legal documents and convert them to PDF using Adobe Capture (not my choice, I prefer OCRShopXTR).
Capture, both the OCR and scanning components, will either refuse to run or keep crashing if not run as an administrator. Same goes for Kodak's scanning software (which is, incidentally, some of the worst and most user-unfriednly software I have ever seen). Adobe Acrobat will not run properly as a user without r/w to Program Files.
There would probably be an even bigger list if I dodn't have to run nearly everyone as a power user anyway (there's Winamp too, but we don't use that at work).
And please note I don't blame MS for this. Everything since Win2K has a had a great system of ACL's and user privs, but the devs have been lazy and not bothered to follow the MS's recommendations and are still stuck in the 9x days (although some of MS'ssoftware suffers from the same problems), so because half of the software out there doesn't run in an unpriveliged environment, MS's are half-forced into making everyone an administrator.
Stupid I know, but to call the GP "FUD" is disingenuous.
Horses for courses I guess; I'm a long-time Opera user, and I hate the little X's on the tabs! I just use a mouse gesture or a shift-click on the tab or a Ctrl-W to close tabs depending on whether I'm using the mouse or keyboard.
Thankfully for me, you can turn them off (tools > preferences > general > show close button on each tab).
But yeah, I don't like the way Firefox handles alot of it's UI myself, and the close tab button way up the top is a bit of a PITA for user who haven't yet grasped the power of mouse gestures...
Most Linux advocates don't just hand their grandma a copy of Linux From Scratch and leave them to it, you know...
My first ever experience with Linux was Mandrake 8.2 Beta. I found the installation of the OS much easier than (my first) windows install two weeks previously; It Just Worked. Once I'd gotten my head around how it handled installing things, it was easy.
These days it's brain-sputteringly simple. Every desktop distro worth its salt has a graphical package installation utility that explains exactly which packages are available, and what they are used for - usualy sorted into relevant categories. Heck, even installing things as complicated (in the Linux world) as kernel modules/drivers is usually simple.
Granted, some pre-alpha drivers require some confugling, but once they go stable and are added to the kernel, it becomes practically impossible not to set things up properly, thanks to the marvell of things like hotplug. Even relatively complicated pieces of software (that no Joe Schmoe would install anyway) such as Apache and MySQL have GUI configurators available. In the realm of pure desktop productivity/leisure software, I've not encountered a single package that didn't just install and give me a nice clickable button in my K/foot/whatever menu.
If there's anything wrong with the way modern desktop distros handle packaging, it's educating the users away from the "download arbitrary.exe from random website, double click to install" mindset. Whilst I agree that a unified packaing system would be great, different distros are tuned for different usage models, so (to me at least) it makes sense that they'll handle things differently. Forcing Linux to adopt the windows method of software installation will create more problems than it'll solve.
FWIW, my girlfriend is entirely happy with the Gentoo/KDE box I've donated to her.
Thanks for the post, an awful lot of articles I've read don't often go into to performance issues like that.
AFAIK, all modern x86 chips alredy use register renaming, but would the additional overhead of SMT make that slower?
Re: the cache issues, it seems to be something that Intel have (maybe) tried to solve by dumping ever-increasing amount of cache on their chips (although of course they need generous cache alread because of their relatively high memory latency). Obviously I'm thinking that the AMD64's considerably lower memory access latency will help with cache-thrashing in this regard.
As regards to the manafacturing issues, I'm led to believe that Intel's SMT logic takes up about 5% of the die space, although it's a relatively simple implementation and it's because of this that it can lead to performance degradation in some applications (apparently). I don't have any figures on how much silicon SMT consumes on a POWER5, but I assume it's more than that.
So whilst it might not be an economical option for the largely single-process workloads on desktop machines, might a good SMT implementation not be of great benefit to the Opteron line of server/workstation chips? They generally have to deal with alot more threads/processes than the Athlon64's, and as I see it they are already starting to tread on the toes of the big iron chips in some of the 4-8 way server stakes.
The article mentions the POWER5 chip and it's implementation of SMT and how it behaves with multi-core chips (i.e. how it can devote all threads on one core to a single task, with the other core(s) sharing the workload via SMT) and how it's rather more impressive that the HyperThreading[TM] on Intel P4's, although I'm not a microprocesor guru.
Whilst I can understand AMD's decision not to put SMT in their current processors, with the recent focus on multi-core and multi-threading I think they'd be foolish not to think about it soon, and (as someone not very up on non-x86 chips) it seems IBM's POWER5 is a good base to emulate. Does anyone have any information on SMT implementations in POWER other chips like Sparc and Itanium?
Making a browser fully 100% standards compatible is one thing. Basic CSS support is another. IE currently is miles away from the former and still not close enough to the latter to warrant as anything other than nearly-a-decade-old browser technology in my book.
Browsers other than IE at least try to follow standards.
We'll support the features people want = we'll support the features that are most used = we'll bugfix the features used by the World's Most Popular Browser [TM] = we're actually going gto do jack shit for interoperability.
I'd love to believe that MS really are going to fix IE's utterly appalling CSS support that other browsers sorted out years ago, but judging by IE's track record I don't hold out alot of hope. Hack, am I the only one who failed to notice any significant extra features in the much vaunted IE6?
Whilst I have absolutely no problems getting multimedia to play back on my system, I still have one thing to say:
IMHO media players for Linux suck.
There, I said it. As far s features go, Linux media players seem almost unsurpassed. There's lots of high quality code that manages to play back my movies using less CPU than my windows partition, and give me higher quality video along with it, as well an inbuilt support for subtitles, remote controls, DVD's... the works. Brilliant.
But the media frontend/GUI's are appalling. I may have been spoilt by having used Media Player Classic on windows for the past three years, but is there any technical reason we can't have a GUI like that in Linux? A nice, simple interface with hordes of functionality hidden beneath the surface for anyone who needs it? Instead we get "skinnable" Hi-Fi-esque (why?) GUI's with tiny, fiddly buttons that seem to require a zillion different mouse clicks, key presses and command line options before you can cajole them into doing what you want.
At the moment, the only media player I've found that comes remotely close is Totem which seems to be a sort of clone of MPC, but it's still nowhere near as usable (it has the drag'n'drop playlist and the double-click-to-fullscreen I want, but no hover-up control panel in fullscreen) and if you're not a GNOME user it still requires a boatload of other apps and services to be installed as dependencies (but then GNOME bloat is another rant in itself). Xine-ui is probably the second best IMHO, and that's not saying much; silly, ugly, built-in libs and no in-movie playback controls. mplayer's GTK (not even GTK2) frontend gmplayer is abysmally hard to use and has virtually no options. Whilst I understand that these apps are entirely volunteer-driven, I find it astonishing that the volunteers are happy with them.
And before you criticise, yes I am actually learning C++ so that I can help make a media player GUI I can live with. Althgouh I don't expect that'll stop me being ritually ignored/modden into oblivion for daring to question Linux's usability.
Pity they didn't extend the same courtesy to 2K - a version that, IMHO, is functionally identical to XP but with a bit of new chrome, and hence why I object to being blackmailed into paying £250 solely for the privilege of a number jumping from 5.0 to 5.1.
As soon as DirectX and modern GFX drivers aren't available on 2K, I'm jumping entirely to WineX.
Stricly speaking, I think it's a meta-meta-dupe. It's a/. article about a CNet article about previous/. articles - many of which, in turn, have been dupes themselves or dupes about articles.
If/. dupes this story, we might be lucky enough to see an infinite recursion loop whic... urgh, my head hurts.
Don't know if they're available in the states, but XFX make a passively cooled nVidia 6600 (non-GT) with dual DVI ports and TV-out. Retails here for about £80 (~$140), so prolly availble in the US for about $100. Whether this is "reasonably priced" for you I don't know, but it's a good little card for 2D work and the odd bit of OpenGL eye-candy, and will hold it's own in games.
Effing crikey, what is wrong with/.'ers?! Is it really so difficult to understand how the BBC works?
Every single time the BBC is mentioned we have hordes of people denigrating this "government agency" what with their jackbooted thugs collecting the TV tax from poor British peasants.
The Beeb does not work like that. The TV license is a voluntary contribution. Don't want to pay it? Then don't own a TV capable of receiving broadcast signals.
Yes, the license fee is collected by the government and passed on to the BBC. But other than this the BBC is entirely independent, much to the chagrin of the government alot of the time. It's the BBC's relative impartiality that has contributed to it becoming one of the most respected media insitutions the world over.
All these people perputaully insisting that the BBC is some Stalinistic socialist plot set to destroy all lesser media companies is either a mass troll or wilful ignorance on a nationwide scale.
Unless I'm very much mistaken, the Pentium M was designed from the very beginning to use very little power, which I think is more important to it's overall power usage than it's fab process.
I wasn't aware Intel were using strained silicon, and on further reading it seems they're also developing their own brand of SOI without IBM's patents, but rumour has it that a) it won't be deployed until 2007 with 65/45nm chips and b) it isn't as effective. Guess we'll have to wait and see.
Obviously the Pentium M arch is higly promising, I just wish that Intel would get on with promoting it as a CPU instead of flogging the advanced-stages-of-decomposition horse that is Netburst.
Is this the same IBM that co-develop fab tech with AMD, who currently have the lowest power consumption 90nm core available?
IIRC, SOI and strained silicon are both IBM patents that Intel has been itching to get their hands on, but IBM will only license them in exchange for IP that Intel isn't willing to part with. As such, Intel's current deeply piplelined, high clock arch is kinda sunk on their current 90nm process.
Damn straight - even a horribly hampered PS3 would make a quite powerful frontend machine. And if one or two of those vector units could be coaxed into decoding MPEG2/HDTV as they showed in the tech demo, you would have one very cool little media box.
Is there anyone here who used the PS2 Linux kit? Was all the software for it (especially the compiler) Sony-authorised, or could you pretty much install anything you want on it?
At least with the PS3, the primary CPU is PPC-based, for which GCC already has pretty good support, and IBM and Sony have already pledged that Cell will be fairly open - heck, you could prolly have a boot-from-DVD MythTV distro. Now all we need to do is hope the media bods at Sony don't try and cripple it in such a way that it can't run "untrusted" software without a mod chip or other such DMCA-baiting measures.
(P.S. Hard drive aside, does anyone know if it's possible to shoehorn a PXE boot routine into kit like this? As easy as a hard drive would be, availability and price will no doubt be unfriendly)
Milkdrop is the reason I've never needed to take any hallucinogenic drugs.
Same here. Now imagine when you have a few friends over for dinner, and a few of them happen to make a "special" mushroom omelette. Just for fun, I set my two Linux machines up running xscreensaver's and my windows machine running Milkdrop. There were some fairly ecstatic grins going around...:D
On a more serious note, I cannot frickin' wait until a MilkDrop variant gets ported to XMMS (or, even better, Beep Media Player), even if it's only plugin-compatible and doesn't use the same code. Milkdrop is hands-down the thing that has kept me clinging to Winamp these past few years, and I find myself booting into windows just to play music with it. IIRC it's already been added to CVS builds of the Xbox Media Centre, and hopefully it'll also amke its way into MythMusic too.
I never thought I'd say this, btu thank you AOL for helping make my life more... colourful... swirly... urgh, I feel sick...!
Here is an article where the chief financial officer of nVidia confirms that the supposedly "in-game" footage from the new PS3 is a load of cobblers, cos the RSX chip isn't finished yet and doesn't exist in a workable form.
Sigh... it's the emotion engine/missile guidance systems all over again.
I'm not sure it's related to a TTL issue, but the (appallingly slow and unreliable) DNS from my ISP (BT OpenWoe - no, not my choice!) b0rked on pretty much every highly Akamai-sed site I know about.
Google.co.uk wouldn't resolve (although.com would). Image servers for yahoo, IMDB and a boatload of others were nowhere to be seen. Miscellaneous other sites would slow to a crawl as my router struggled to resolve fifteen different ad servers.
I set up Bind a week ago just for a laugh, and haven't had a single problem ever since. Can someone more fluent in the wherefores of DNS explain if this has anything to do with the TTL issue? I thought it might be that the IP addresses for the servers in question changed frequently and that British Telecom's notoriously bad DNS wasn't catching up fast enough, but if anyone has a better explanantion I'd love to hear it:)
That's interesting, considering that a browser-only installation of Mozilla is leaner than FireFox. What went wrong?
Probably the problems discussed in the article:) I'm not a developer and know sod-all about coding so I'm just pulling things out of my proverbial donkey, but the speed of FF development has come along at a frightening pace. IIRC the objective was to create a replacement for IE using Gecko ASAP; there's probably been some hurried code that's injected a bit of bloat. There's also a few UI quirks that need addressing (like *not* re-rendering the page when it can be pulled out of memory cache).
The excessive memory footprint of FireFox is what stops me from using it, so that would be welcome.
I think once the embedded browser gains a bit more momentum, expect to see lots of Gecko and GUI crosstalk between the projects. You can't really add much more functionality to Gecko at the mo since it renders pretty much anything exceptionally well indeed, so I'm hoping for alot of work to be done slimming it down, along with better modularity of FF and its extensions, as well XUL improvements. Etc etc etc, yadda yadda yadda wishlist...
Until then, Opera will remain my browser of choice as it does what it says on the tin with a minimum of fuss (even though there are some quirks under Linux that I'm no happy with). FF is catching up very quickly though, and I think it's a testament to the devs talent that such a solid project could be knocked up so quickly.
Whilst I agree wholeheartedly with your comment (I'm an Opera user myself, under Linux and windows) and I was quite surprised you weren't modded into obliviion by some of the seemingly anti-Opera/. members, remember that Opera makes most of their money from the embedded market, especially mobile phones and PDA's. Hence there's an awful lot of incentive to get the browser to be as lightweight as possible. Similarly, alot of Opera's innovations have been related to making a browser useful with limited screen size (tabs, small screen rendering) and navigation tools (gestures, incredibly extensive shortcuts).
Conversely, FF's main aim was to develop a leaner browser than the Moz suite whilst still maintaining a Moz/Opera-like level of functionality. Now that the browser is more-or-less set in stone, expect to see alot of work being done in the smaller/faster areas. Especially with the up-and-coming Gecko-powered embedded browser that's being worked on.
Make sure you're running X at 100dpi (try checking with xdpyinfo | grep resolution). Running at slightly off-kilter resolutions can muck up some fonts; personally, on my boxes fonts are gorgeous.
The two servers I admin at work have both been runnning testing for well over 100 days since I upgraded them. It might not be recommended for production use, but it works fine in this instance - just as well as stable in fact (my home server runs stable), it just takes a bit more effort to maintain.
If you're worried about security updates but still want modern stuff on top, there's always backports.org
Slashdot Mirror
What software run in a school/business environment needs to be run as an administrator? Stop spreading FUD.
It's not FUD.
I work in a small OCR shop. We scan alot of legal documents and convert them to PDF using Adobe Capture (not my choice, I prefer OCRShopXTR).
Capture, both the OCR and scanning components, will either refuse to run or keep crashing if not run as an administrator. Same goes for Kodak's scanning software (which is, incidentally, some of the worst and most user-unfriednly software I have ever seen). Adobe Acrobat will not run properly as a user without r/w to Program Files.
There would probably be an even bigger list if I dodn't have to run nearly everyone as a power user anyway (there's Winamp too, but we don't use that at work).
And please note I don't blame MS for this. Everything since Win2K has a had a great system of ACL's and user privs, but the devs have been lazy and not bothered to follow the MS's recommendations and are still stuck in the 9x days (although some of MS'ssoftware suffers from the same problems), so because half of the software out there doesn't run in an unpriveliged environment, MS's are half-forced into making everyone an administrator.
Stupid I know, but to call the GP "FUD" is disingenuous.
Horses for courses I guess; I'm a long-time Opera user, and I hate the little X's on the tabs! I just use a mouse gesture or a shift-click on the tab or a Ctrl-W to close tabs depending on whether I'm using the mouse or keyboard.
Thankfully for me, you can turn them off (tools > preferences > general > show close button on each tab).
But yeah, I don't like the way Firefox handles alot of it's UI myself, and the close tab button way up the top is a bit of a PITA for user who haven't yet grasped the power of mouse gestures...
Most Linux advocates don't just hand their grandma a copy of Linux From Scratch and leave them to it, you know...
.exe from random website, double click to install" mindset. Whilst I agree that a unified packaing system would be great, different distros are tuned for different usage models, so (to me at least) it makes sense that they'll handle things differently. Forcing Linux to adopt the windows method of software installation will create more problems than it'll solve.
My first ever experience with Linux was Mandrake 8.2 Beta. I found the installation of the OS much easier than (my first) windows install two weeks previously; It Just Worked. Once I'd gotten my head around how it handled installing things, it was easy.
These days it's brain-sputteringly simple. Every desktop distro worth its salt has a graphical package installation utility that explains exactly which packages are available, and what they are used for - usualy sorted into relevant categories. Heck, even installing things as complicated (in the Linux world) as kernel modules/drivers is usually simple.
Granted, some pre-alpha drivers require some confugling, but once they go stable and are added to the kernel, it becomes practically impossible not to set things up properly, thanks to the marvell of things like hotplug. Even relatively complicated pieces of software (that no Joe Schmoe would install anyway) such as Apache and MySQL have GUI configurators available. In the realm of pure desktop productivity/leisure software, I've not encountered a single package that didn't just install and give me a nice clickable button in my K/foot/whatever menu.
If there's anything wrong with the way modern desktop distros handle packaging, it's educating the users away from the "download arbitrary
FWIW, my girlfriend is entirely happy with the Gentoo/KDE box I've donated to her.
Thanks for the post, an awful lot of articles I've read don't often go into to performance issues like that.
AFAIK, all modern x86 chips alredy use register renaming, but would the additional overhead of SMT make that slower?
Re: the cache issues, it seems to be something that Intel have (maybe) tried to solve by dumping ever-increasing amount of cache on their chips (although of course they need generous cache alread because of their relatively high memory latency). Obviously I'm thinking that the AMD64's considerably lower memory access latency will help with cache-thrashing in this regard.
As regards to the manafacturing issues, I'm led to believe that Intel's SMT logic takes up about 5% of the die space, although it's a relatively simple implementation and it's because of this that it can lead to performance degradation in some applications (apparently). I don't have any figures on how much silicon SMT consumes on a POWER5, but I assume it's more than that.
So whilst it might not be an economical option for the largely single-process workloads on desktop machines, might a good SMT implementation not be of great benefit to the Opteron line of server/workstation chips? They generally have to deal with alot more threads/processes than the Athlon64's, and as I see it they are already starting to tread on the toes of the big iron chips in some of the 4-8 way server stakes.
One thing that has been interesting me lately, after reading a series of Anadtech articles on current and near-future processor tech is the possible inclusion of SMT (oft marketed as Hyperthreading by Intel) on AMD cores.
The article mentions the POWER5 chip and it's implementation of SMT and how it behaves with multi-core chips (i.e. how it can devote all threads on one core to a single task, with the other core(s) sharing the workload via SMT) and how it's rather more impressive that the HyperThreading[TM] on Intel P4's, although I'm not a microprocesor guru.
Whilst I can understand AMD's decision not to put SMT in their current processors, with the recent focus on multi-core and multi-threading I think they'd be foolish not to think about it soon, and (as someone not very up on non-x86 chips) it seems IBM's POWER5 is a good base to emulate. Does anyone have any information on SMT implementations in POWER other chips like Sparc and Itanium?
Making a browser fully 100% standards compatible is one thing. Basic CSS support is another. IE currently is miles away from the former and still not close enough to the latter to warrant as anything other than nearly-a-decade-old browser technology in my book.
Browsers other than IE at least try to follow standards.
What stupid, stupid reasoning.
We'll support the features people want = we'll support the features that are most used = we'll bugfix the features used by the World's Most Popular Browser [TM] = we're actually going gto do jack shit for interoperability.
I'd love to believe that MS really are going to fix IE's utterly appalling CSS support that other browsers sorted out years ago, but judging by IE's track record I don't hold out alot of hope. Hack, am I the only one who failed to notice any significant extra features in the much vaunted IE6?
Whilst I have absolutely no problems getting multimedia to play back on my system, I still have one thing to say:
IMHO media players for Linux suck.
There, I said it. As far s features go, Linux media players seem almost unsurpassed. There's lots of high quality code that manages to play back my movies using less CPU than my windows partition, and give me higher quality video along with it, as well an inbuilt support for subtitles, remote controls, DVD's... the works. Brilliant.
But the media frontend/GUI's are appalling. I may have been spoilt by having used Media Player Classic on windows for the past three years, but is there any technical reason we can't have a GUI like that in Linux? A nice, simple interface with hordes of functionality hidden beneath the surface for anyone who needs it? Instead we get "skinnable" Hi-Fi-esque (why?) GUI's with tiny, fiddly buttons that seem to require a zillion different mouse clicks, key presses and command line options before you can cajole them into doing what you want.
At the moment, the only media player I've found that comes remotely close is Totem which seems to be a sort of clone of MPC, but it's still nowhere near as usable (it has the drag'n'drop playlist and the double-click-to-fullscreen I want, but no hover-up control panel in fullscreen) and if you're not a GNOME user it still requires a boatload of other apps and services to be installed as dependencies (but then GNOME bloat is another rant in itself). Xine-ui is probably the second best IMHO, and that's not saying much; silly, ugly, built-in libs and no in-movie playback controls. mplayer's GTK (not even GTK2) frontend gmplayer is abysmally hard to use and has virtually no options. Whilst I understand that these apps are entirely volunteer-driven, I find it astonishing that the volunteers are happy with them.
And before you criticise, yes I am actually learning C++ so that I can help make a media player GUI I can live with. Althgouh I don't expect that'll stop me being ritually ignored/modden into oblivion for daring to question Linux's usability.
Pity they didn't extend the same courtesy to 2K - a version that, IMHO, is functionally identical to XP but with a bit of new chrome, and hence why I object to being blackmailed into paying £250 solely for the privilege of a number jumping from 5.0 to 5.1.
As soon as DirectX and modern GFX drivers aren't available on 2K, I'm jumping entirely to WineX.
For those who don't get the joke from a non-native english speaker;
s/extatic/eggstatic/
Stricly speaking, I think it's a meta-meta-dupe. It's a /. article about a CNet article about previous /. articles - many of which, in turn, have been dupes themselves or dupes about articles.
/. dupes this story, we might be lucky enough to see an infinite recursion loop whic... urgh, my head hurts.
If
Don't know if they're available in the states, but XFX make a passively cooled nVidia 6600 (non-GT) with dual DVI ports and TV-out. Retails here for about £80 (~$140), so prolly availble in the US for about $100. Whether this is "reasonably priced" for you I don't know, but it's a good little card for 2D work and the odd bit of OpenGL eye-candy, and will hold it's own in games.
Effing crikey, what is wrong with /.'ers?! Is it really so difficult to understand how the BBC works?
Every single time the BBC is mentioned we have hordes of people denigrating this "government agency" what with their jackbooted thugs collecting the TV tax from poor British peasants.
The Beeb does not work like that. The TV license is a voluntary contribution. Don't want to pay it? Then don't own a TV capable of receiving broadcast signals.
Yes, the license fee is collected by the government and passed on to the BBC. But other than this the BBC is entirely independent, much to the chagrin of the government alot of the time. It's the BBC's relative impartiality that has contributed to it becoming one of the most respected media insitutions the world over.
All these people perputaully insisting that the BBC is some Stalinistic socialist plot set to destroy all lesser media companies is either a mass troll or wilful ignorance on a nationwide scale.
Unless I'm very much mistaken, the Pentium M was designed from the very beginning to use very little power, which I think is more important to it's overall power usage than it's fab process.
I wasn't aware Intel were using strained silicon, and on further reading it seems they're also developing their own brand of SOI without IBM's patents, but rumour has it that a) it won't be deployed until 2007 with 65/45nm chips and b) it isn't as effective. Guess we'll have to wait and see.
Obviously the Pentium M arch is higly promising, I just wish that Intel would get on with promoting it as a CPU instead of flogging the advanced-stages-of-decomposition horse that is Netburst.
Is this the same IBM that co-develop fab tech with AMD, who currently have the lowest power consumption 90nm core available?
IIRC, SOI and strained silicon are both IBM patents that Intel has been itching to get their hands on, but IBM will only license them in exchange for IP that Intel isn't willing to part with. As such, Intel's current deeply piplelined, high clock arch is kinda sunk on their current 90nm process.
Damn straight - even a horribly hampered PS3 would make a quite powerful frontend machine. And if one or two of those vector units could be coaxed into decoding MPEG2/HDTV as they showed in the tech demo, you would have one very cool little media box.
Is there anyone here who used the PS2 Linux kit? Was all the software for it (especially the compiler) Sony-authorised, or could you pretty much install anything you want on it?
At least with the PS3, the primary CPU is PPC-based, for which GCC already has pretty good support, and IBM and Sony have already pledged that Cell will be fairly open - heck, you could prolly have a boot-from-DVD MythTV distro. Now all we need to do is hope the media bods at Sony don't try and cripple it in such a way that it can't run "untrusted" software without a mod chip or other such DMCA-baiting measures.
(P.S. Hard drive aside, does anyone know if it's possible to shoehorn a PXE boot routine into kit like this? As easy as a hard drive would be, availability and price will no doubt be unfriendly)
Milkdrop is the reason I've never needed to take any hallucinogenic drugs.
:D
Same here. Now imagine when you have a few friends over for dinner, and a few of them happen to make a "special" mushroom omelette. Just for fun, I set my two Linux machines up running xscreensaver's and my windows machine running Milkdrop. There were some fairly ecstatic grins going around...
On a more serious note, I cannot frickin' wait until a MilkDrop variant gets ported to XMMS (or, even better, Beep Media Player), even if it's only plugin-compatible and doesn't use the same code. Milkdrop is hands-down the thing that has kept me clinging to Winamp these past few years, and I find myself booting into windows just to play music with it. IIRC it's already been added to CVS builds of the Xbox Media Centre, and hopefully it'll also amke its way into MythMusic too.
I never thought I'd say this, btu thank you AOL for helping make my life more... colourful... swirly... urgh, I feel sick...!
...of all those who haven't seen it yet;
Here is an article where the chief financial officer of nVidia confirms that the supposedly "in-game" footage from the new PS3 is a load of cobblers, cos the RSX chip isn't finished yet and doesn't exist in a workable form.
Sigh... it's the emotion engine/missile guidance systems all over again.
Yoiks! Someone else in the world that's heard of Laika!
:D
Congrats on the most appropriate bastardisation of Bad Times ever
I'm not sure it's related to a TTL issue, but the (appallingly slow and unreliable) DNS from my ISP (BT OpenWoe - no, not my choice!) b0rked on pretty much every highly Akamai-sed site I know about.
.com would). Image servers for yahoo, IMDB and a boatload of others were nowhere to be seen. Miscellaneous other sites would slow to a crawl as my router struggled to resolve fifteen different ad servers.
:)
Google.co.uk wouldn't resolve (although
I set up Bind a week ago just for a laugh, and haven't had a single problem ever since. Can someone more fluent in the wherefores of DNS explain if this has anything to do with the TTL issue? I thought it might be that the IP addresses for the servers in question changed frequently and that British Telecom's notoriously bad DNS wasn't catching up fast enough, but if anyone has a better explanantion I'd love to hear it
Probably the problems discussed in the article
The excessive memory footprint of FireFox is what stops me from using it, so that would be welcome.
I think once the embedded browser gains a bit more momentum, expect to see lots of Gecko and GUI crosstalk between the projects. You can't really add much more functionality to Gecko at the mo since it renders pretty much anything exceptionally well indeed, so I'm hoping for alot of work to be done slimming it down, along with better modularity of FF and its extensions, as well XUL improvements. Etc etc etc, yadda yadda yadda wishlist...
Until then, Opera will remain my browser of choice as it does what it says on the tin with a minimum of fuss (even though there are some quirks under Linux that I'm no happy with). FF is catching up very quickly though, and I think it's a testament to the devs talent that such a solid project could be knocked up so quickly.
Whilst I agree wholeheartedly with your comment (I'm an Opera user myself, under Linux and windows) and I was quite surprised you weren't modded into obliviion by some of the seemingly anti-Opera /. members, remember that Opera makes most of their money from the embedded market, especially mobile phones and PDA's. Hence there's an awful lot of incentive to get the browser to be as lightweight as possible. Similarly, alot of Opera's innovations have been related to making a browser useful with limited screen size (tabs, small screen rendering) and navigation tools (gestures, incredibly extensive shortcuts).
Conversely, FF's main aim was to develop a leaner browser than the Moz suite whilst still maintaining a Moz/Opera-like level of functionality. Now that the browser is more-or-less set in stone, expect to see alot of work being done in the smaller/faster areas. Especially with the up-and-coming Gecko-powered embedded browser that's being worked on.
Make sure you're running X at 100dpi (try checking with xdpyinfo | grep resolution). Running at slightly off-kilter resolutions can muck up some fonts; personally, on my boxes fonts are gorgeous.
The two servers I admin at work have both been runnning testing for well over 100 days since I upgraded them. It might not be recommended for production use, but it works fine in this instance - just as well as stable in fact (my home server runs stable), it just takes a bit more effort to maintain.
If you're worried about security updates but still want modern stuff on top, there's always backports.org