If Opera and Chrome can manage it, I don't see why IE and FF shouldn't be able to, and they don't feel the need to tack spoilers and rims on to the browsers and tell me how extreeeeeeeeeeeeeeme they are.
"Not being designed" for "more than X tabs" is a get out - what people really mean when they say that is "we didn't design our tabbing schema to be scalable, so if you're one of the users that runs into one of our unspecified design limitations we'll just say you don't fit in with 90% of the market and ignore you. SEP".
Yes, I am a riot to eavesdrop when on the phone to tech support centres:)
Latest version of 3.5, and I listed my extensions in another post.
Thanks for the link though - I remember applying a few of those when I first started running into problems a few years back (2.x was abominable for memory usage), but I found they were only marginally effective - they key issue is page fragmentation, so even if you place hard limits on the amount of memory to be used as cache you still end up with that memory being "exhausted", eventually leaving little to no space for cache... and you're back to crappy performance again.
Keyconfig Mouse gestures redox Oldbar Tab Mix Plus User Agent Switcher Adblock plus Download Statusbar Flashblock
No plugins other than Java enabled either.
Before you tell me "extension XYZ is buggy, don't use it!" (everyone always says this whenever tabmix is mentioned), please don't. These are what I consider my essential set that make firefox usable to me; remove any one of those and FF becomes seriously annoying to use. Not a large set of plugins by any means either. If you know of other extensions that do the same thing then please let me know and I'll have a look at them. Until then please don't assume I'm one of those morons that install everything shiny and then wonders why the purple monkey won't go away and keeps telling them they have a limp penis.
FYI, I've actually got three different systems at my desk, including a laptop and a quad core workstation with 16GB of RAM (firefox gets unusable once it eats up about 2.5GB of it though). As the AC points out, this doesn't help you when an app is crashing due to eating all available memory. The trick is to fix the app, not throw more and more hardware at it until it appears to work fine. Each machine tends to be used for a different task and I don't have a single machine I just use just for web-browsing;
Vista had a similar problem - as Russinovich's blogs have pointed out, there were serious bugs with vista, most noticeably WDDM and DirectX causing excessive memory usage (duplication of objects in RAM and VRAM) which was a heavy factor in it's unpopularity; throwing more hardware at it wasn't an option with the rise of the netbook. So MS was forced to fix these problems with Win7 and we now have a version of windows that'll run semi-comfortably within 1GB - I can't stand the default UI, but it's underbelly is a massive improvement on vista. The FF and IE team would do well to learn this lesson, being the number 1 and 2 applications that 90% of people spend 90% of their computer using.
Not offtopic; both IE and FF do have serious, serious problems with memory after any amount of time spent browsing. I typically keep 30-50 tabs open in FF and it'll hit the 1.8GB barrier within two to three days, and crash. Time and time again I'm told I'm "using it wrong". I don't use IE nearly as much, but the job demands it occasionally - and it too will tend to glom up with cruft if you leave 10+ tabs running for more than a few hours.
Incessantly annoying, and one of the prime reasons I'm still an Opera loyalist, despite the absence of a fire'n'forget ad blocker - it's fast out of the box, has sane defaults and doesn't slow down over time (I spend about 45mins on every new user account to get FF to a state where I find it as usable as Opera, which takes me 10mins of configuring), I just can't use it at work since it doesn't get along with our proxy.
The problem with IE and FF is that because I'm not a common user profile (I'm always told people can't cope with having more than five tabs open; suck it, I have excellent spatial awareness) and the memory problems are damned hard to fix (heck, submitting a decent bug report about it is hard) - so for most people it's easier to just add a whizz-bang feature with a meaningless name that sounds impressive to a non-techie and SEP the memory problems away.
I know for a fact that you won't be having any s*x when I get you fired from your j*b for blasphemy, Mrs. Hiram T. Washbucket from Nashville, Alaska. It's just a shame that saying something that the self-appointed religious/non-religious/power-hungry-maniacs majority don't agree with isn't punishable by death any more, as the world would be such a better place with all the objectively objectionable people removed.
Whether the guy was crapflooding the site isn't the issue here, folks. Crapflooding is just like being an arsehole - annoying, but not illegal. Ban his IP from posting if he's crapflooding. Reporting using the word "pussy" to his boss in the hope that he gets fired as a result is an plainly obvious vengeance burn. Yeah, the guy was a prick for posting "pussy" all the time, but at least he had a sense (albeit limited) of humour.
Agreed. Opera Mini is streets ahead of any of the blackberry browsers, whether it be speed (it's faster downloading and rendering pages), UI (it's alot easier to use on a phone and much faster to scroll) and ease of reading (the BB browser has one of the worst placements of elements I've ever seen).
The S60 browser (I have both this and Opera on my Nokia) is pretty good, although the UI is *slightly* clunkier than opera and navigating on a non-touchscreen phone is a little bit more cumbersome.
Some companies do place restrictions on third party apps on their blackberries though - thankfully, alot of our staffers work from the road via blackberry and 3G, and so Opera and Google Maps were allowed onto our BES to rapturous applause. If you're not as lucky as me, you might need to kidnap your BES admins and threaten them that you'll make them be guinea pigs for using the BB encryption software;)
What I can't understand about the awesome resolution independence of KDE4 (been using 4.3 for a month or so now) is their stubborn refusal to actually get widgets to scale sensibly. The fact that every desktop element is scalable is well cool, but it's not polished overall - on my two main workstations (2x 20" 1680x1050, 1x 24" 1900x1200) I have shedloads of desktop real estate and I'd like a big fat taskbar... but if I raise it to double or triple height, the clock scales in height *and* width, so that you end up with a colossal clock taking up 20% of your horizontal taskbar space. At least the clock in XP was sane enough to add extra information (day, date). If anyone knows of any workarounds, I'd kill to hear them.
Secondly, my laptops - one has a very nice 11" 1360x768 LED screen; vertical real estate is at a premium. However, the KDE4 taskbar wastes a shedload of vertical space by insitsting on having curvy corners for all the app placeholders, and if you have a double-height taskbar you still have space wasted with gaps to seperate the text from the curved edges. I've looked high and low for options to turn the curves off, as well as for themes without an obsession for putting curves on everything (I swear, within five years "usability" experts will have us all using perfectly circular buttons for everything).
TLDR: KDE did a great job with a scalable UI (and most apps follow the guidelines well), but it needs alot of polish before I'd call it completely ready for prime time or crotchety power users like me that dislike young whippersnappers on their lawn.
Seems clear to me - Hollywood has been successfully infiltrated by the Slashdot Illuminati. For years they've been conditioning the technocrats into accepting continual recycling of the same information, the powerful few who once could have stood against them, with an endless series of dupes. Disguising this ruse as mere incompetence and laziness, they've now bribed, blackmailed and assassinated their way up into the highest echelons of the media industry so they can try and assert their control over the proles as well.
Think that the coming of talentless no-hopers like $any_fucking_pop_artist appearing all over the top of the charts is a coincidence? Think again. 1997, slashdot opens. 1997, Britney is given her first record contract... paid for by a bank account traceable back to the now defunct (and suspiciously absent from any business records - just try looking for them on google, they're not there) MaldaBates Holdings registered in the Cayman Islands... who were also the 51% stakeholders in Jive Records. Coincidence?
Mark my words, behind the friendly green glow of slashdot lies a shadowy consortium of power brokers, vulture capitalists, politicians, marketing consultants and soda manufacturers. They're out for control, and they're winning.
Posting anonymously in case CowboyNeal is watching.
Thanks for that, I figured it couldn't be as simple as solely relying on telco-added firmware mods to flag $app as a non-mobile program. Seems like the whole industry is geared towards making some bits more expensive than others... an attitude I do find surprising (especially seeing as my nokia came bundled with a boatload of VoIP apps) - the geek in me would just expect the phone to just incorporate a bog-standard modem/packet interface and then make it available to whatever apps requested data, but I guess that's why I'll never be a businessman...!
Not Verizon I'm afraid, I'm in the UK using a Nokia E52. I hear about all the Verizon horror stories though... but my advice has always to never accept free gifts from strangers;) I had a crippled phone once but thankfully it's really easy to get them unlocked. But no-one would dare do things like disabling bluetooth over here... touch wood;)
Editing ROMs? My phone has it built in natively - plug into a machine and it'll offer itself as a 3G card (or USB mass storage, or data mode, or any number of USB thingies). I don't use phones that have carrier-specific firmware on them.
Nothing is incredibly strong stuff though - witness that it's almost impossible to tear toilet paper or a cheque book along the perforated lines, clearly indicating that less matter means a stronger material. I hypothesize that if we could find a way to remove 99.999999% of the matter from, say, common or garden steel we'd have something as tough as neutronium whilst weighing the same as a dried Mexican Staring Frog.
However, I'm convinced someone has stolen my idea and already incorporated it into modern blister packs.
Forgive me if I'm asking a silly question, but how does your telco know whether what you're looking up is part of the "mobile" web or is hooked up to your laptop...? Surely they can't tell what you're actually doing with your phone without actually doing a traffic analysis? They both just use plain ol' protocols on top of TCP/IP, right?
I ask because I'm considering buying a data bolt-on for my phone and they make a big distinction between the "mobile internet" and "normal internet" (although they don't understand the different between the internet and the WWW so getting an answer out of them is impossible), as one for of data inexplicably costs more than the other, so it's obvious to me that they're gouging. But do they really have any way of detecting what you're doing with your data connection?
As someone else points out, it's reasonably trivial to re-enable your root account in ubuntu. However, I mostly just use `sudo su -` for quick'n'dirty shell access.
Parent is right about Kubuntu putting an ugly face on KDE; whilst I'm prepared to admit that KDE4 is approaching feature parity with 3.5, kubuntu has been plagued with bugs since its release that made life impossible. I was also bitten by the non-functioning wireless (across three different WLAN cards, one intel, two atheros) and on a current freshly installed Karmic system, plasma-desktop routinely chomps CPU and RAM until crashing (currently spiking at 70% CPU usage and 1.4GB RAM) - problems I've never run into on other KDE distros. So far, anyway;) Maybe I'll see what Debian's KDE is like...
Not disagreeing with you per se (I concede that the possibility of a hash collision is infinitesimal with SHA256) but even so wouldn't a collision be the worst kind of failure - namely silent data corruption?
Does anyone know if the ZFS code incorporates a mode where you can enforce checking of the blocks bit-for-bit in the event of the hashes being the same? More IO intensive but it's a checkbox for all those "data integrity is paramount" applications.
Don't just try Sumatra, use it. If, like me, 99.9% of your time spent with PDF's is reading and searching them (yay ubiquitous tech docs and white papers), your computer will love you for using Sumatra. Almost instantaneous startup, tiny memory footprint, bitchin' fast rendering and, as far as I'm aware, no endless litany of security holes. Another bonus is not having to disable browser-hanging plugins.
As soon as someone comes up with a decent replacement for flash, we'll have achieved desktop nirvana;)
Best thing our new management has done was to open the budgets. We all got an email telling us the expenditure for the previous year and expected expenditure for this year - the bosses were going to ask the managers to get the techie types, like myself, to go over each item and see if they were neccesary... but before they could organise a meeting they were already inundated with "this contract expired two years ago!" and "we can totally negotiate a better price on this!" messages from the front lines.
A management that understand and engages with underlings will result in a mutually beneficial relationship - I get to do the most techie work I can do and get away with minimal paperwork, simply because everyone from the top to the bottom understands the process and none of us want to waste time with redundancy. Management that don't understand their serfs and refuse to engage will continually create counter-intuitive, expensive and demoralising practices that will eventually cause all the gives-a-shit workers to leave for a management that cares... leaving the pencil pushing jobsworths to help run the company into the ground. This is true of all professions, not just computing - it's just more acutely observed in computing due to techies generally not being trustful of a (frequently closed) hierarchy.
Last time I looked at the Intel X25 specs, you could write to them at their maximum rate 24/7 for something like a decade before the cells would start to fail writing (but you'd still be able to read them). I doubt you'd see such longevity from a spinning rust hard drive.
Don't know if it's just the shitty exchange rate here in the UK, but price-per-GB on SSD's has skyrocketed recently; about four months ago I paid £230 for my 120GB OCZ Agility; the same drive is now on sale here for £300. Other brands of SSD's show the same huge jumps (which is a shame as I'm waiting on 6Gbps SATA before I jump onto a new workstation) whereas everything else seems more or less static. Either the price of NAND has risen considerably, or demand is such that supply is low, or manufacturers have decided they can shaft us.
Still though, I'm not buying another spinning-rust hard drive unless I really need the space. Affordable SSD's are the most important thing to happen to storage this decade.
We use a cluster of HP LeftHand appliances at work which does all those things without FC; the discs are 7k SATA and some 10k SAS and everything goes over 10Gb iSCSI, which is shared between two HP blade centres, each holding ten nodes of dual Xeon 5500 and 96GB RAM, connected via a pair of Cisco Nexus switches. Performance is limited somewhat by the fact that some mope overruled me speccing the LeftHand's with a primary 10Gb interface and secondary bonded 2x1Gb interfaces but the cluster will still easily do 600MB/s and ~2000 IOPS which is more than enough for our 300 or so VM's (and we've still got kit dedicated to dev and testing at the moment). The LeftHand's mirror the data to our other data centre synchronously, and VMware lets us do quick migration between sites to maintain loads (and we can run everything from a single site without a hitch).
I had my doubts over iSCSI as well, and it did take alot more hand-holding than my old beloved 2Gb fibre from our EMC Celerra to get it to decent performance (mostly due to Broadcom's crappy NIC not supporting 9k jumbo frames) but iSCSI is a good way to do high performance storage for considerably less than FC. We're mulling testing FCoE on the same hardware to avoid iSCSI/network misconfiguration pitfalls but we're perfectly happy with it for now.
Not disagreeing with you saying that FC isn't dead (I still think it's great where you can afford it, and EMC quoted us a fantastic deal on a CX4 & Brocade 4 and 8Gb switches - just the LeftHand solution gave us 150% as much storage for the same price, and all our VMware licensing thrown in. Recessions are great for getting a good deal.
Meanwhile in real life, governments the world over are in the pockets of the media industry and their slavish public can't take it in the arse fast enough. Sarkozy is just a politician who's more openly "available" for influence than others, but there's plenty more worms in the EU woodwork. The number of politicians I've seen parroting, word for word, the latest anti-customer campaign about how piracy eats up 92% of the global GDP or some such bullshit makes you lose all faith in humani... sorry, in sentient life the world over.
"I don't know which species is worse. You don't see them fucking each other over for a percentage."
Except the google study didn't display any evidence of this happening - there was no correlation between higher temperatures and higher failure rates on mechanical drives.
Even if it were, it'd be easy to rememdy - boot all your servers off SSD and keep them in a "hot" room. Keep your SANs-full-o'-spinning-rust in a "cold" room. You've just saved a fortune in air con despite being unable to convince your CTO that heat isn't as big a killer as many people claim it to be.
We had a power failure at one of our data centres, due to a combination of a stupid JCB driver and IBM's ineptitude (not keeping the diesel tanks full). Power for the servers was restored about six hours before the air-con was back up and running, and most of our equipment got cooked (ambient temp ranging from 35 to 40 degrees depending which part of the data centre you stood in) - we demanded IBM guarantee us a 3hr turnaround on any parts that died for the next 6 months due to heat failure. 18 months later and our hard drive failure graph is the same as it ever was.
Shoddy components on hardware is another matter I guess, but we've never had any hardware die due to a single faulty component apart from the occasional RAID card. Expect, as "hot" DC's become more common, that the heat thresholds on lots of enterprise equipment will increase... for a price, of course;)
Slashdot Mirror
If Opera and Chrome can manage it, I don't see why IE and FF shouldn't be able to, and they don't feel the need to tack spoilers and rims on to the browsers and tell me how extreeeeeeeeeeeeeeme they are.
"Not being designed" for "more than X tabs" is a get out - what people really mean when they say that is "we didn't design our tabbing schema to be scalable, so if you're one of the users that runs into one of our unspecified design limitations we'll just say you don't fit in with 90% of the market and ignore you. SEP".
Yes, I am a riot to eavesdrop when on the phone to tech support centres :)
Latest version of 3.5, and I listed my extensions in another post.
Thanks for the link though - I remember applying a few of those when I first started running into problems a few years back (2.x was abominable for memory usage), but I found they were only marginally effective - they key issue is page fragmentation, so even if you place hard limits on the amount of memory to be used as cache you still end up with that memory being "exhausted", eventually leaving little to no space for cache... and you're back to crappy performance again.
Don't know if you ever saw Pavlov's blog about the fragmentation issue, but it makes an excellent read: http://blog.pavlov.net/2007/11/10/memory-fragmentation/
Keyconfig
Mouse gestures redox
Oldbar
Tab Mix Plus
User Agent Switcher
Adblock plus
Download Statusbar
Flashblock
No plugins other than Java enabled either.
Before you tell me "extension XYZ is buggy, don't use it!" (everyone always says this whenever tabmix is mentioned), please don't. These are what I consider my essential set that make firefox usable to me; remove any one of those and FF becomes seriously annoying to use. Not a large set of plugins by any means either. If you know of other extensions that do the same thing then please let me know and I'll have a look at them. Until then please don't assume I'm one of those morons that install everything shiny and then wonders why the purple monkey won't go away and keeps telling them they have a limp penis.
FYI, I've actually got three different systems at my desk, including a laptop and a quad core workstation with 16GB of RAM (firefox gets unusable once it eats up about 2.5GB of it though). As the AC points out, this doesn't help you when an app is crashing due to eating all available memory. The trick is to fix the app, not throw more and more hardware at it until it appears to work fine. Each machine tends to be used for a different task and I don't have a single machine I just use just for web-browsing;
Vista had a similar problem - as Russinovich's blogs have pointed out, there were serious bugs with vista, most noticeably WDDM and DirectX causing excessive memory usage (duplication of objects in RAM and VRAM) which was a heavy factor in it's unpopularity; throwing more hardware at it wasn't an option with the rise of the netbook. So MS was forced to fix these problems with Win7 and we now have a version of windows that'll run semi-comfortably within 1GB - I can't stand the default UI, but it's underbelly is a massive improvement on vista. The FF and IE team would do well to learn this lesson, being the number 1 and 2 applications that 90% of people spend 90% of their computer using.
Not offtopic; both IE and FF do have serious, serious problems with memory after any amount of time spent browsing. I typically keep 30-50 tabs open in FF and it'll hit the 1.8GB barrier within two to three days, and crash. Time and time again I'm told I'm "using it wrong". I don't use IE nearly as much, but the job demands it occasionally - and it too will tend to glom up with cruft if you leave 10+ tabs running for more than a few hours.
Incessantly annoying, and one of the prime reasons I'm still an Opera loyalist, despite the absence of a fire'n'forget ad blocker - it's fast out of the box, has sane defaults and doesn't slow down over time (I spend about 45mins on every new user account to get FF to a state where I find it as usable as Opera, which takes me 10mins of configuring), I just can't use it at work since it doesn't get along with our proxy.
The problem with IE and FF is that because I'm not a common user profile (I'm always told people can't cope with having more than five tabs open; suck it, I have excellent spatial awareness) and the memory problems are damned hard to fix (heck, submitting a decent bug report about it is hard) - so for most people it's easier to just add a whizz-bang feature with a meaningless name that sounds impressive to a non-techie and SEP the memory problems away.
Sad but true.
</minirant>
Guess what? People have sex.
I know for a fact that you won't be having any s*x when I get you fired from your j*b for blasphemy, Mrs. Hiram T. Washbucket from Nashville, Alaska. It's just a shame that saying something that the self-appointed religious/non-religious/power-hungry-maniacs majority don't agree with isn't punishable by death any more, as the world would be such a better place with all the objectively objectionable people removed.
Whether the guy was crapflooding the site isn't the issue here, folks. Crapflooding is just like being an arsehole - annoying, but not illegal. Ban his IP from posting if he's crapflooding. Reporting using the word "pussy" to his boss in the hope that he gets fired as a result is an plainly obvious vengeance burn. Yeah, the guy was a prick for posting "pussy" all the time, but at least he had a sense (albeit limited) of humour.
Agreed. Opera Mini is streets ahead of any of the blackberry browsers, whether it be speed (it's faster downloading and rendering pages), UI (it's alot easier to use on a phone and much faster to scroll) and ease of reading (the BB browser has one of the worst placements of elements I've ever seen).
The S60 browser (I have both this and Opera on my Nokia) is pretty good, although the UI is *slightly* clunkier than opera and navigating on a non-touchscreen phone is a little bit more cumbersome.
Some companies do place restrictions on third party apps on their blackberries though - thankfully, alot of our staffers work from the road via blackberry and 3G, and so Opera and Google Maps were allowed onto our BES to rapturous applause. If you're not as lucky as me, you might need to kidnap your BES admins and threaten them that you'll make them be guinea pigs for using the BB encryption software ;)
What I can't understand about the awesome resolution independence of KDE4 (been using 4.3 for a month or so now) is their stubborn refusal to actually get widgets to scale sensibly. The fact that every desktop element is scalable is well cool, but it's not polished overall - on my two main workstations (2x 20" 1680x1050, 1x 24" 1900x1200) I have shedloads of desktop real estate and I'd like a big fat taskbar... but if I raise it to double or triple height, the clock scales in height *and* width, so that you end up with a colossal clock taking up 20% of your horizontal taskbar space. At least the clock in XP was sane enough to add extra information (day, date). If anyone knows of any workarounds, I'd kill to hear them.
Secondly, my laptops - one has a very nice 11" 1360x768 LED screen; vertical real estate is at a premium. However, the KDE4 taskbar wastes a shedload of vertical space by insitsting on having curvy corners for all the app placeholders, and if you have a double-height taskbar you still have space wasted with gaps to seperate the text from the curved edges. I've looked high and low for options to turn the curves off, as well as for themes without an obsession for putting curves on everything (I swear, within five years "usability" experts will have us all using perfectly circular buttons for everything).
TLDR: KDE did a great job with a scalable UI (and most apps follow the guidelines well), but it needs alot of polish before I'd call it completely ready for prime time or crotchety power users like me that dislike young whippersnappers on their lawn.
Seems clear to me - Hollywood has been successfully infiltrated by the Slashdot Illuminati. For years they've been conditioning the technocrats into accepting continual recycling of the same information, the powerful few who once could have stood against them, with an endless series of dupes. Disguising this ruse as mere incompetence and laziness, they've now bribed, blackmailed and assassinated their way up into the highest echelons of the media industry so they can try and assert their control over the proles as well.
Think that the coming of talentless no-hopers like $any_fucking_pop_artist appearing all over the top of the charts is a coincidence? Think again. 1997, slashdot opens. 1997, Britney is given her first record contract... paid for by a bank account traceable back to the now defunct (and suspiciously absent from any business records - just try looking for them on google, they're not there) MaldaBates Holdings registered in the Cayman Islands... who were also the 51% stakeholders in Jive Records. Coincidence?
Mark my words, behind the friendly green glow of slashdot lies a shadowy consortium of power brokers, vulture capitalists, politicians, marketing consultants and soda manufacturers. They're out for control, and they're winning.
Posting anonymously in case CowboyNeal is watching.
> sudo make me a sandwich
> I'm sorry Dave, but under USPTO 7617530 I can't allow you to do that
Thanks for that, I figured it couldn't be as simple as solely relying on telco-added firmware mods to flag $app as a non-mobile program. Seems like the whole industry is geared towards making some bits more expensive than others... an attitude I do find surprising (especially seeing as my nokia came bundled with a boatload of VoIP apps) - the geek in me would just expect the phone to just incorporate a bog-standard modem/packet interface and then make it available to whatever apps requested data, but I guess that's why I'll never be a businessman...!
Not Verizon I'm afraid, I'm in the UK using a Nokia E52. I hear about all the Verizon horror stories though... but my advice has always to never accept free gifts from strangers ;) I had a crippled phone once but thankfully it's really easy to get them unlocked. But no-one would dare do things like disabling bluetooth over here... touch wood ;)
Editing ROMs? My phone has it built in natively - plug into a machine and it'll offer itself as a 3G card (or USB mass storage, or data mode, or any number of USB thingies). I don't use phones that have carrier-specific firmware on them.
Nothing is incredibly strong stuff though - witness that it's almost impossible to tear toilet paper or a cheque book along the perforated lines, clearly indicating that less matter means a stronger material. I hypothesize that if we could find a way to remove 99.999999% of the matter from, say, common or garden steel we'd have something as tough as neutronium whilst weighing the same as a dried Mexican Staring Frog.
However, I'm convinced someone has stolen my idea and already incorporated it into modern blister packs.
Forgive me if I'm asking a silly question, but how does your telco know whether what you're looking up is part of the "mobile" web or is hooked up to your laptop...? Surely they can't tell what you're actually doing with your phone without actually doing a traffic analysis? They both just use plain ol' protocols on top of TCP/IP, right?
I ask because I'm considering buying a data bolt-on for my phone and they make a big distinction between the "mobile internet" and "normal internet" (although they don't understand the different between the internet and the WWW so getting an answer out of them is impossible), as one for of data inexplicably costs more than the other, so it's obvious to me that they're gouging. But do they really have any way of detecting what you're doing with your data connection?
As someone else points out, it's reasonably trivial to re-enable your root account in ubuntu. However, I mostly just use `sudo su -` for quick'n'dirty shell access.
Parent is right about Kubuntu putting an ugly face on KDE; whilst I'm prepared to admit that KDE4 is approaching feature parity with 3.5, kubuntu has been plagued with bugs since its release that made life impossible. I was also bitten by the non-functioning wireless (across three different WLAN cards, one intel, two atheros) and on a current freshly installed Karmic system, plasma-desktop routinely chomps CPU and RAM until crashing (currently spiking at 70% CPU usage and 1.4GB RAM) - problems I've never run into on other KDE distros. So far, anyway ;) Maybe I'll see what Debian's KDE is like...
Not disagreeing with you per se (I concede that the possibility of a hash collision is infinitesimal with SHA256) but even so wouldn't a collision be the worst kind of failure - namely silent data corruption?
Does anyone know if the ZFS code incorporates a mode where you can enforce checking of the blocks bit-for-bit in the event of the hashes being the same? More IO intensive but it's a checkbox for all those "data integrity is paramount" applications.
Don't just try Sumatra, use it. If, like me, 99.9% of your time spent with PDF's is reading and searching them (yay ubiquitous tech docs and white papers), your computer will love you for using Sumatra. Almost instantaneous startup, tiny memory footprint, bitchin' fast rendering and, as far as I'm aware, no endless litany of security holes. Another bonus is not having to disable browser-hanging plugins.
As soon as someone comes up with a decent replacement for flash, we'll have achieved desktop nirvana ;)
Best thing our new management has done was to open the budgets. We all got an email telling us the expenditure for the previous year and expected expenditure for this year - the bosses were going to ask the managers to get the techie types, like myself, to go over each item and see if they were neccesary... but before they could organise a meeting they were already inundated with "this contract expired two years ago!" and "we can totally negotiate a better price on this!" messages from the front lines.
A management that understand and engages with underlings will result in a mutually beneficial relationship - I get to do the most techie work I can do and get away with minimal paperwork, simply because everyone from the top to the bottom understands the process and none of us want to waste time with redundancy. Management that don't understand their serfs and refuse to engage will continually create counter-intuitive, expensive and demoralising practices that will eventually cause all the gives-a-shit workers to leave for a management that cares... leaving the pencil pushing jobsworths to help run the company into the ground. This is true of all professions, not just computing - it's just more acutely observed in computing due to techies generally not being trustful of a (frequently closed) hierarchy.
Last time I looked at the Intel X25 specs, you could write to them at their maximum rate 24/7 for something like a decade before the cells would start to fail writing (but you'd still be able to read them). I doubt you'd see such longevity from a spinning rust hard drive.
Don't know if it's just the shitty exchange rate here in the UK, but price-per-GB on SSD's has skyrocketed recently; about four months ago I paid £230 for my 120GB OCZ Agility; the same drive is now on sale here for £300. Other brands of SSD's show the same huge jumps (which is a shame as I'm waiting on 6Gbps SATA before I jump onto a new workstation) whereas everything else seems more or less static. Either the price of NAND has risen considerably, or demand is such that supply is low, or manufacturers have decided they can shaft us.
Still though, I'm not buying another spinning-rust hard drive unless I really need the space. Affordable SSD's are the most important thing to happen to storage this decade.
We use a cluster of HP LeftHand appliances at work which does all those things without FC; the discs are 7k SATA and some 10k SAS and everything goes over 10Gb iSCSI, which is shared between two HP blade centres, each holding ten nodes of dual Xeon 5500 and 96GB RAM, connected via a pair of Cisco Nexus switches. Performance is limited somewhat by the fact that some mope overruled me speccing the LeftHand's with a primary 10Gb interface and secondary bonded 2x1Gb interfaces but the cluster will still easily do 600MB/s and ~2000 IOPS which is more than enough for our 300 or so VM's (and we've still got kit dedicated to dev and testing at the moment). The LeftHand's mirror the data to our other data centre synchronously, and VMware lets us do quick migration between sites to maintain loads (and we can run everything from a single site without a hitch).
I had my doubts over iSCSI as well, and it did take alot more hand-holding than my old beloved 2Gb fibre from our EMC Celerra to get it to decent performance (mostly due to Broadcom's crappy NIC not supporting 9k jumbo frames) but iSCSI is a good way to do high performance storage for considerably less than FC. We're mulling testing FCoE on the same hardware to avoid iSCSI/network misconfiguration pitfalls but we're perfectly happy with it for now.
Not disagreeing with you saying that FC isn't dead (I still think it's great where you can afford it, and EMC quoted us a fantastic deal on a CX4 & Brocade 4 and 8Gb switches - just the LeftHand solution gave us 150% as much storage for the same price, and all our VMware licensing thrown in. Recessions are great for getting a good deal.
Sarkozy's just had an unhappy love affair and he doesn't see why anyone else should have a good time.
Meanwhile in real life, governments the world over are in the pockets of the media industry and their slavish public can't take it in the arse fast enough. Sarkozy is just a politician who's more openly "available" for influence than others, but there's plenty more worms in the EU woodwork. The number of politicians I've seen parroting, word for word, the latest anti-customer campaign about how piracy eats up 92% of the global GDP or some such bullshit makes you lose all faith in humani... sorry, in sentient life the world over.
"I don't know which species is worse. You don't see them fucking each other over for a percentage."
Except the google study didn't display any evidence of this happening - there was no correlation between higher temperatures and higher failure rates on mechanical drives.
http://hardware.slashdot.org/article.pl?sid=07/02/18/0420247
http://www.engadget.com/2007/02/18/massive-google-hard-drive-survey-turns-up-very-interesting-thing/
http://labs.google.com/papers/disk_failures.pdf
Even if it were, it'd be easy to rememdy - boot all your servers off SSD and keep them in a "hot" room. Keep your SANs-full-o'-spinning-rust in a "cold" room. You've just saved a fortune in air con despite being unable to convince your CTO that heat isn't as big a killer as many people claim it to be.
We had a power failure at one of our data centres, due to a combination of a stupid JCB driver and IBM's ineptitude (not keeping the diesel tanks full). Power for the servers was restored about six hours before the air-con was back up and running, and most of our equipment got cooked (ambient temp ranging from 35 to 40 degrees depending which part of the data centre you stood in) - we demanded IBM guarantee us a 3hr turnaround on any parts that died for the next 6 months due to heat failure. 18 months later and our hard drive failure graph is the same as it ever was.
Shoddy components on hardware is another matter I guess, but we've never had any hardware die due to a single faulty component apart from the occasional RAID card. Expect, as "hot" DC's become more common, that the heat thresholds on lots of enterprise equipment will increase... for a price, of course ;)