Cockup on my part in the previous post: where I put IPv4 I meant IPv6: I was meaning that perhaps we should concentrate on those who are looking favourably on IPv6 and let the rest either join the bandwagon later or be left in the dust.
Unfortunately as much benefit as there is even internally for the larger address space, many large companies and ISPs see the upfront setup costs (reconfiguring a large infrastructure isn't cheap, and cutting corners is far too risky in the modern 24/7 business world, and some equipment will need replacing too) and ignore the longer term savings in inconvenience, man time, and other hassle. They misunderstand the scale of the problem, not I, and it is only getting worse as they now have "yeah, but you said we'd have those problems years ago, and it hasn't happened yet" in their bag of ill-conceived reasons not to upgrade.
Our own hacked versions of NAT and so forth were necessary. Without them we would have been the ones left in the dust as our code would not operate on the rest of the world's IPv4 only infrastructure well enough so other OSs and software stacks would have taken that market share instead.
8 was released in 2009. IE9 last year. I'm not really sure it matters for google, but if you do custom web applications 3 years isn't really a long time to have to keep it alive.
IE8, while a significant improvement of 7 and 6 in a number of respects, was already seriously out of date on the day of release. The difference in testing and fixing effort associated with supporting IE8 compared to supporting IE9 is much larger than the two year difference in release dates would suggest.
Unfortunately for some of us, refusing to supporting IE8 is not a luxury we can currently afford. Hopefully the likes of Google taking this step will help push out corporate clients into the right decade (some of the largest banks in the UK have only upgraded to IE8 (from IE6) in their offices in the last year, in fact one that we deal with hasn't yet completed the transition).
The big thing with IE8 is that it's the last IE for windows XP. Which is why it has a larger markeshare than IE9 still.
If you are supporting corporates who are unworried about being embarrassingly behind the times, and other organizations like educational establishments (who also have legacy apps that won't work on decent browsers, and who (unlike the banks who are the bane of my work life) genuinely can't afford the time/investment required to fix that situation), the significant factor here is not the relationship between IE8 and XP, but the fact that IE8 will be officially supported from the point of view of security updates and fixes for other show-stopping bugs for as long as Windows 7 is (which will be supported in that way until something like 2022 IIRC). As our clients migrate away from XP in order to not be left vulnerable when XP falls out of support (or left having to pay MS large chunks of money if they don't upgrade in time and need fixes to problems found outside the final support window) I fully expect them to standardise on Windows 7 with IE8 for most of the next decade even if other big players follow Google's example.
No they don't. Any ISP/corp that doesn't want to use IPv6 is free to sit back and watch parts of the Internet become unavailable to them and their users. Of course by choosing this path they chose to eventually die, but it is their choice to make.
Doing something other than IPv6 simply because people won't make the effort is like sticking to horse-drawn vehicles because people don't want the hassle of having to visit petrol pumps and towns/cities don't want the hassle of constructing the required infrastructure.
While Google can refuse to specifically support them as they may have done in the past, what is to stop them simply paying to license those parts? If Google refuse to sell to them then they are big enough to create quite a stink that Google will not want: anti-trust allegations in both the US and EU much like MS has faced, and simple public statements of "this is what Google mean by free and open" which may harm their general reputation even without legal complaints/action.
They do at least have a profanity filter which is sometimes a little prudish, which would constrain some of the unflattering suggestions automatically. It Basically if a word would stop the auto-search-as-you-type operating, it shouldn't appear in suggested search terms unless one of your explicitly entered terms is already on the list.
Another think that keeps other terms near the top of the list is astro-turfing. Gaming Google is a common tactic in political circles. They won't do it to try bury an opponent in crap terms (as that may find them blocked by Google meaning they can't game in favour of themselves) but they will do it to skew their own results in a particular direction. Just get many people at home to search for "Mitt Romney tax cuts" and similar and it'll push "Mitt Romney " further down the list.
Almost certainly due to the number of 1024 bit certs that are out there.
Most CA's won't sign anything smaller than 2048 bit now, and that has been the case for a year or few, but what about companies that paid an absolute fortune for five year "enhanced validation" certificates or have their own CA for internal use and signed many many keys smaller than 1024 some years ago.
From a security standpoint 2048 should really be the cut-off, as it is elsewhere, but from a practicality view that simply wouldn't wash with a chunk of their userbase who would be very vocal in making sure MS (rather than any lax certificate review/renewal policies at their end) get the blame for any inconvenience caused so they'll not make that jump for a while (i.e. until 1024 bit certs are practically extinct, as those smaller than 1024 are now).
There are a few reason why some people object though, including but not limited to:
* They just don't like being followed. In some circumstances in real life (walking through an iffy part of a town you don't know well, or making your way through lion country) being followed is not a good feeling, and our brains which aren't as much evolved as we sometimes like to think don't make the distinction between being tracked physically and being tracked virtually.
* Some don't want certain browsing habits (porn being the main one) to be accidentally revealed to some of their circles (should they forget to engage incognito mode).
* Having heard the stories over the years and dismissed them as "yeah, you mean porn" but I do now know someone who nearly had a surprise soiled by gmail filling his screen with engagement ring adverts while his now-fiencée was around (again, incognito mode would have helped here).
* Back on the "being followed" feeling: this is happening in your own home, which makes it doubly odd if you are going to feel odd about it anyway.
* Some wonder, "if Google/bing/facebook are profiling me, who else might be?" - someone who collates enough information about you might be able to use it for various less useful things and that risk might not be worth the utility of relevant adverts.
* And my main objection: someone somewhere is making money off all this information in my profiles, and it isn't me nor am I getting any sort of cut!
I agree with DNT not being set by default. Make it an option on the default browser home page, then people can set it whenever they like, or just ignore it. Done.
But to Apache: "we do not support breaking open standards" hold no water what so ever when your way to express your love for standards is to patch your product such that it can completely ignore a generally accepted standard by default. That to my mind is a text-book example of hypocrisy.
And to the ad servers saying "if X then we'll just ignore DNT" I say fine: if you won't honour DNT I feel no guilt at all in completely blocking all your content. Thanks for playing. I only block ad networks that get on my nerves (auto playing sounds, overly irritating animations, malware riddled shite, and so forth), but this is on my list of things that get on my nerves.
For what it is worth I don't think DNT will make any difference at all, as it relies on everyone to play ball server-side and I barely trust anyone with a commercial or other interest in tracking people to play ball in anything other than hollow words, but that is no reason to not be irritated when you hear people say "we know and understand your preferences, but fuck you".
I don't for sure, but statements like that usually have a "my country is better than yours, ner ner neh ner ner" tone rather than a "every country is shite" one.
It's not bureucracy gone wild, just common citizen doing things
So you're quite happy to live in a world where every time you want to "do things" you have to go scouring through law books and beg the government for permission?
And I suppose you are happy living in a world where kids can't have decent chemistry sets because TERRORISM!!1!, and where it is difficult to get through an airport with a laptop because TERRORISTS!!!, in fact where you have to be intimately rubbed down by the TSA in said airport because TERROR!!!!!!!, and so on and so forth.
America: the land of the brainwashed-into-thinking-they-are-more-free-than-others.
Quake was well ahead of Duke 3D in technology terms, but as a game it never felt "complete" to me. Lots of ideas in there, but it didn't hang together in a way that suggested a thought-through narrative. Duke was hardly high art plot-wise, but at least it had one. Of course complex plot (or much plot at all) is not always needed but as a lot of the discussion above is about the relative coherence of HL's direction I fell it is relevant to subsequent discussion of Duke/Quake.
What the team behind Duke3D did was hack current technology a fair bit (using multiple maps in clever arrangements to get around the not-really-3D limitation with regard to in level design) and implement thir game using that. What iD did was truly push the limits of game engines (and the hardware they ran on) at the time, essentially becoming the first success of the next generation, but lost it a bit in terms of actuall game design. I think it comes down to the prevailing impression of iD: they make great game engines for other people to make great games with.
Don't get me wrong: I liked what I played of Quake (the first episode IIRC - up to the point of killing be big lava fellow with the lightning machine he helpfully never tried to get out of). But at times I think I was as much impressed by the technology as I was having fun.
Back closer to topic: HL did something more akin to Duke3D: take the best of current engine tech, with a few hacks/improvements, and use it to implement a relatively interesting game. HL2 pushed the boat out quite a bit further in terms of game engine technology, but it was still not a "next generation" revolution but again pushing the current tech a bit and using that to implement a good game.
Flip it around and make it necessary for content owners to provide their content for sale in order to make an infringement claim.
What about specially commissioned private works? If I have a poster design, video, or what-ever, made for myself or a company, would you have the right to copy it? What about my photos from a family holiday that are on my website? Can GreedyShister Ltd. copy them for use in their promotional materials simply because I'm not offering them for sale?
Works in progress would have similar problems: if I hand out a part finished work for people to look at and give feedback (or just because they are interested even if they won't have feedback they feel worth giving), can they copy that work in progress as I don't offer it for sale? The same for content that otherwise "leaks" out early. I suppose an "I intend to sell the finished work" clause might help here, but you would have to specify that every time you talk about the item as otherwise you open up to "well you didn't say..." arguments.
And of course it could easily be worked around the other way: all content producers have to do is say "I'll sell you this for {exorbitant value}" - no sales due to the price but it is offered for sale. You could add a "reasonableness" clause to your rule, but any such clause has no choice but to be vague and wide open to interpretation/argument.
PCs are not dying. They are just finally (nearly) completely commodity items.
Unfortunately for the likes of HP the thing that the market doesn't need in that situation is either large behemoth producers or an innovator. At this point all that happens (except for the high end which I cover in a second) is that the small players bring the price down to near cost while adding small improvements. There simply isn't anywhere massively different for PC form (box on/under desk, big screen) to go for now - people generally don't need more power or more features (hardware wise) than we already have there - hence all the R&D and consumer interest is going into phones and tablets. Of course there are the niche markets but the only one in the desktop arena is hard-core gamers and that isn't a very large market with the current state of gaming - the other power niches (high-end CAD users, other number crunchers, people with large databases to process, and so forth) are all moving (if they have not already moved) towards online processing or at very least their own little server farm (the likes of HP can still make a business there I'm sure, though that is not relevant to discussion about the "PC" market)
From a personal computing point of view we already have a reboot: it is the phones and tablets (and TVs & related set-top boxes, but in terms of both OS, other software, and hardware, these overlap the tablet market so much you can't really consider them much separate - the R&D for each feeds into all). If the likes of HP can't jump on that bandwagon they'll have to wait for (or have the luck to find or create) the next big thing or and/just hope their other markets (the server and large-scale "solution" markets) can make up for the drop-off in their personal computer market share.
There isn't going to be a revolution in the PC market. There will be many small improvements, maybe something large enough over time to be called evolution. There is no need for any such revolution: they do many jobs well, and to do the other jobs people want other forms are more suitable. They are not dying out though - they are just dropping into the relatively stagnant area of solved problems so you can't sell a new one to the user every year or two. The revolution is the developments in over areas (phones and tablets, for the time being), and even they are getting to the "we have all the features, the just need refining" stage already so in the next year or few we'll be saying the same things about tablets/phones/settops/TVs/etc needing a reboot.
Windows 8 is to Windows 7 what XP was to 2000:
* From the users PoV most of the changes are superficial (the new "skin" generally, the ribbon everywhere, explorer enhancements like the fancy long operation progress boxes)
* There is a lot more going on under the hood to make some of those changes possible, or to make other improvements like general efficiency and hardware support
* Most users won't know about, care about, or need to care about the above internal changes, so unless they want to UI update there is no reason to upgrade in the near future.
Many stuck with Windows 2000 until soon before it dropped out of security patch support because for many it worked so didn't need fixing, particularly because people didn't want to volunteer for a new UI when they were used to the old one (even though a lot of the newness could be turned off). I strongly suspect the transition from 7 (or Vista, or XP for those still using it) to 8 will be similarly drawn out, if not worse because the UI change is more significant.
The "registry" is still there?
Yes, and will be for many years to come. Far too much depends on that core feature for it to be removed in any way any time soon. Some code might be using alternate config storage methods/stores/APIs but if the registry went away much would stop working and translating everything it does to a new method (so said method can replace it transparently) would probably be somewhat impractical.
then the proper course of action would have been to remove the hosts file feature totally
IIRC you still need posix compliance (or the ability to claim it such that your claims can not be rubbished too easily) for your OS to be used in many US agencies, and the hosts file is one of the many minor points mentioned in that specification. Presumably that spec says something about having the feature, but does not say anything about effectively disabling it in this way.
The Windows 7 firewall can definitely do the job described by the poster a few comments up, which is what the poster you replied to was referring to.
There are things (a great many, depending on how far down the "it is rare people would need to" scale) iptables can do that the built-in Windows firewall can't do, but the matter currently being discussed certainly isn't one of them.
You are right to by cynical, but there is definitely more going on here than just the assault/rape case even if big bad Uncle Sam isn't the driving force.
If it were just because of that case it would have been handled quite differently - there are procedures in place for interviewing the accused here and all those options have been offered but refused: so they want him back there for some reason and the rape does not seem to warrant that determination on its own because at this point (IIRC he hasn't even officially been charged for that crime yet: they are officially just asking him to "help with their enquiries") those procedures should be more than sufficient.
To work that would need to be a fairly targeted attack: picking a few marks and working on them. A scatter-gun attack as usually used by scammers will simply alert the world to the problem and make all marks take the "no one will believe you" line, and the more targeted approach would take a lot more time and effort (and ability): while the payout could be more than worth it long term, I doubt any scammer will take the risk of waiting for as long as the scheme could take to "pay out".
The only case where this sort of attack is a concern IMO is when the attacker is someone that you know, rather than a more remote scammer, who sees reason to be out to get you.
2. Knowing proper spelling and grammar in your native language.
There have been suggestions that some of the scammers use this as a mark filter: people put off by the spelling/grammar would be unlilkely to follow through to the end anyway so put them off early so you can concentrate on the others. People who fall for the scam despite the presentation are better quality marks and more more likely to pay out (either because they have done something wrong and are feeling guilty, or because they don't speak the language well enough to spot the telltail problems, or simply because they are just plain thick).
Though I think it more likely that the simpler explanation (most of the scammers simply fail to create a good presentation in the target language) is more likely at least in most cases.
The implication is that they are not listening to people in general about issues that affect people in general when using their products, i.e. implying that there is a groundswell of opinion on some of the matters (so there are general problems rather than it just being a few fussy/opinionated/entitled individuals moaning) which they are ignoring.
Of course just moaning on random unrelated forums is not going to help, slashdot is not the right place for instance, people won't be heard if they don't speak in the right direction. Of course finding the right place to comment could be difficult for a large organisation like Google. TheOatmeal wasn't trying to be heard really (well, not specifically heard by Google), he was just making a passing reference to the sate of affairs for comic value, though he has managed what it would seem many have not and got the attention of someone who is in a relevant team at the company.
The 80Mbyte seems to be a network limit, not a bulk disc throughput limit. A quick test reading ~10Gb (in two large files not many small ones) locally on the server gets just under 120Mbyte/sec overall. I might be able to get better from those drives but I wasn't overly careful about stripe alignment and such (they are 4K sector 2Tb disks using Linux's software RAID). I'm not sure if those particular files are closer to the slow end of the spinning metal or closer to the fast end.
Pulling the same files to the Windows box from a Samba share is where I mentioned 80Mbyte (in fact a quick test with the same two files shows it top out at just over 65Mbyte). The server doesn't have anything SSD but the desktop box uses them for internal storage, I've not benchmarked copying from them to the external SSD though (I have copied large things that way, but just not paid attention while it happened, where I have watched the transfer rates while copying from the server. The server has no USB3, so plugging the drive in directly would not get any benefit from the local connection due to USB2 throughput limits.
I've not investigated where that drop from ~120Mbyte/s to ~65 comes from. It could be a mix of Samba not being terribly efficient (it does sit eating >65% of a CPU core while the operation proceeds), me having cheap NICs and a cheap switch, and me not having jumbo frames enabled anywhere (the network is shared with devices that I think don't support larger frames), and probably some other factors I've not considered.
For 24Tb of data you want more than 12x 2Tb drives. At least 18x 2Tb so you can apply RAID5 (or some compound RAID based on arrives of RAID5) or 24x2Tb for RAID1, so that you can survive at least some drive failure situations without data loss. Be careful of array rebuild times with drives that big though. ZFS is something to consider with an array this size (with smaller drives preferably as re-image/re-checksum times will not be short with this over large drives either).
You are right that anything like this would be much less hassle in the long run than many separate external drives, and their PSUs (something I'd not considered).
USB 2.0 provides 480Mbps of (theoretical) bandwidth. So unless you go Gigabit all over your network (not unreasonable), you won't beat it with a NAS. Even then, it's only 1-and-a-bit times as fast as USB working flat-out (and the difference being if you have multiple USB busses, you can get multiple drives working at once).
The 480Mbps is nowhere near what you will see in practise, unlike network speeds which are far closer to the rated maximum. Most USB drives I've seen top out at somewhere between 25 and 30MByte/sec, and if there are no other bottlenecks it isn't unusual to see 100Mbyte/sec from a gbit switched network. My main desktop pulls things from the fileserver at around 80Mbyte/sec, which is as fast as local reads tend to be on that array. So you are right about 100mbit networks: that'll be the bottleneck not USB, but gbit networking should outdo USB2 by at least a factor of 2, possibly 3, maybe even more if you have better drives in you main storage array than I do.
Before trying to run several USB drives to max out your network bandwidth, consider that you will taking the source disks too. Unless they are SSDs having 2, 3, or more concurrent bulk reads going on may not be any faster than one concurrent read as all the extra head movements will wipe out the bulk speed potential. If the OP's 24Tb is spread over numerous physical drives this need not ban an issue though (with planning careful enough to ensure there aren't two bulk processes reading from the same physical devices.
And USB 3.0 would beat it again.
That it would. I have an SSD in a USB3 enclosure, and it can happily consume 80Mbyte/sec read over my little network. It might even be able to do better than that: I've not measured a bulk write read from the internal SSD yet.
And 10Gb between the client and a server is an expensive network to deploy still. Granted, eSATA would probably be faster but there's nothing wrong with USB for such tasks if you *don't* want to provide Gigabit connections everywhere and (presumably) greater-than-gigabit backbones.
If I wanted more speed than USB3+gbit can provide (due to the size of data being backed up on each run) I'd be plugging the backup device(s) in locally to the source (vie eSATA, USB3, or such) rather than using the network (though again taking note to be careful how things are done if trying to use more than one backup device at once).
For the size of data being described, I'd not want a set of USB drives to be my primary backup solution though.
Slashdot Mirror
Actually, cockup in my comprehension completely. I replied to your post thinking you were replying to my post and completely confusing myself.
Note to self: don't post on a Sunday after a late drunken Saturday...
Cockup on my part in the previous post: where I put IPv4 I meant IPv6: I was meaning that perhaps we should concentrate on those who are looking favourably on IPv6 and let the rest either join the bandwagon later or be left in the dust.
Unfortunately as much benefit as there is even internally for the larger address space, many large companies and ISPs see the upfront setup costs (reconfiguring a large infrastructure isn't cheap, and cutting corners is far too risky in the modern 24/7 business world, and some equipment will need replacing too) and ignore the longer term savings in inconvenience, man time, and other hassle. They misunderstand the scale of the problem, not I, and it is only getting worse as they now have "yeah, but you said we'd have those problems years ago, and it hasn't happened yet" in their bag of ill-conceived reasons not to upgrade.
Our own hacked versions of NAT and so forth were necessary. Without them we would have been the ones left in the dust as our code would not operate on the rest of the world's IPv4 only infrastructure well enough so other OSs and software stacks would have taken that market share instead.
Only support current browsers
8 was released in 2009. IE9 last year. I'm not really sure it matters for google, but if you do custom web applications 3 years isn't really a long time to have to keep it alive.
IE8, while a significant improvement of 7 and 6 in a number of respects, was already seriously out of date on the day of release. The difference in testing and fixing effort associated with supporting IE8 compared to supporting IE9 is much larger than the two year difference in release dates would suggest.
Unfortunately for some of us, refusing to supporting IE8 is not a luxury we can currently afford. Hopefully the likes of Google taking this step will help push out corporate clients into the right decade (some of the largest banks in the UK have only upgraded to IE8 (from IE6) in their offices in the last year, in fact one that we deal with hasn't yet completed the transition).
The big thing with IE8 is that it's the last IE for windows XP. Which is why it has a larger markeshare than IE9 still.
If you are supporting corporates who are unworried about being embarrassingly behind the times, and other organizations like educational establishments (who also have legacy apps that won't work on decent browsers, and who (unlike the banks who are the bane of my work life) genuinely can't afford the time/investment required to fix that situation), the significant factor here is not the relationship between IE8 and XP, but the fact that IE8 will be officially supported from the point of view of security updates and fixes for other show-stopping bugs for as long as Windows 7 is (which will be supported in that way until something like 2022 IIRC). As our clients migrate away from XP in order to not be left vulnerable when XP falls out of support (or left having to pay MS large chunks of money if they don't upgrade in time and need fixes to problems found outside the final support window) I fully expect them to standardise on Windows 7 with IE8 for most of the next decade even if other big players follow Google's example.
ISP's/corporations have to commit to that
No they don't. Any ISP/corp that doesn't want to use IPv6 is free to sit back and watch parts of the Internet become unavailable to them and their users. Of course by choosing this path they chose to eventually die, but it is their choice to make.
Doing something other than IPv6 simply because people won't make the effort is like sticking to horse-drawn vehicles because people don't want the hassle of having to visit petrol pumps and towns/cities don't want the hassle of constructing the required infrastructure.
While Google can refuse to specifically support them as they may have done in the past, what is to stop them simply paying to license those parts? If Google refuse to sell to them then they are big enough to create quite a stink that Google will not want: anti-trust allegations in both the US and EU much like MS has faced, and simple public statements of "this is what Google mean by free and open" which may harm their general reputation even without legal complaints/action.
They do at least have a profanity filter which is sometimes a little prudish, which would constrain some of the unflattering suggestions automatically. It Basically if a word would stop the auto-search-as-you-type operating, it shouldn't appear in suggested search terms unless one of your explicitly entered terms is already on the list.
Another think that keeps other terms near the top of the list is astro-turfing. Gaming Google is a common tactic in political circles. They won't do it to try bury an opponent in crap terms (as that may find them blocked by Google meaning they can't game in favour of themselves) but they will do it to skew their own results in a particular direction. Just get many people at home to search for "Mitt Romney tax cuts" and similar and it'll push "Mitt Romney " further down the list.
Does anyone know why 1024 was selected?
Almost certainly due to the number of 1024 bit certs that are out there.
Most CA's won't sign anything smaller than 2048 bit now, and that has been the case for a year or few, but what about companies that paid an absolute fortune for five year "enhanced validation" certificates or have their own CA for internal use and signed many many keys smaller than 1024 some years ago.
From a security standpoint 2048 should really be the cut-off, as it is elsewhere, but from a practicality view that simply wouldn't wash with a chunk of their userbase who would be very vocal in making sure MS (rather than any lax certificate review/renewal policies at their end) get the blame for any inconvenience caused so they'll not make that jump for a while (i.e. until 1024 bit certs are practically extinct, as those smaller than 1024 are now).
There are a few reason why some people object though, including but not limited to:
I agree with DNT not being set by default. Make it an option on the default browser home page, then people can set it whenever they like, or just ignore it. Done.
But to Apache: "we do not support breaking open standards" hold no water what so ever when your way to express your love for standards is to patch your product such that it can completely ignore a generally accepted standard by default. That to my mind is a text-book example of hypocrisy.
And to the ad servers saying "if X then we'll just ignore DNT" I say fine: if you won't honour DNT I feel no guilt at all in completely blocking all your content. Thanks for playing. I only block ad networks that get on my nerves (auto playing sounds, overly irritating animations, malware riddled shite, and so forth), but this is on my list of things that get on my nerves.
For what it is worth I don't think DNT will make any difference at all, as it relies on everyone to play ball server-side and I barely trust anyone with a commercial or other interest in tracking people to play ball in anything other than hollow words, but that is no reason to not be irritated when you hear people say "we know and understand your preferences, but fuck you".
I don't for sure, but statements like that usually have a "my country is better than yours, ner ner neh ner ner" tone rather than a "every country is shite" one.
It's not bureucracy gone wild, just common citizen doing things
So you're quite happy to live in a world where every time you want to "do things" you have to go scouring through law books and beg the government for permission?
And I suppose you are happy living in a world where kids can't have decent chemistry sets because TERRORISM!!1!, and where it is difficult to get through an airport with a laptop because TERRORISTS!!!, in fact where you have to be intimately rubbed down by the TSA in said airport because TERROR!!!!!!!, and so on and so forth.
America: the land of the brainwashed-into-thinking-they-are-more-free-than-others.
Duke was dated when it came out (see Quake).
Quake was well ahead of Duke 3D in technology terms, but as a game it never felt "complete" to me. Lots of ideas in there, but it didn't hang together in a way that suggested a thought-through narrative. Duke was hardly high art plot-wise, but at least it had one. Of course complex plot (or much plot at all) is not always needed but as a lot of the discussion above is about the relative coherence of HL's direction I fell it is relevant to subsequent discussion of Duke/Quake.
What the team behind Duke3D did was hack current technology a fair bit (using multiple maps in clever arrangements to get around the not-really-3D limitation with regard to in level design) and implement thir game using that. What iD did was truly push the limits of game engines (and the hardware they ran on) at the time, essentially becoming the first success of the next generation, but lost it a bit in terms of actuall game design. I think it comes down to the prevailing impression of iD: they make great game engines for other people to make great games with.
Don't get me wrong: I liked what I played of Quake (the first episode IIRC - up to the point of killing be big lava fellow with the lightning machine he helpfully never tried to get out of). But at times I think I was as much impressed by the technology as I was having fun.
Back closer to topic: HL did something more akin to Duke3D: take the best of current engine tech, with a few hacks/improvements, and use it to implement a relatively interesting game. HL2 pushed the boat out quite a bit further in terms of game engine technology, but it was still not a "next generation" revolution but again pushing the current tech a bit and using that to implement a good game.
Flip it around and make it necessary for content owners to provide their content for sale in order to make an infringement claim.
What about specially commissioned private works? If I have a poster design, video, or what-ever, made for myself or a company, would you have the right to copy it? What about my photos from a family holiday that are on my website? Can GreedyShister Ltd. copy them for use in their promotional materials simply because I'm not offering them for sale?
Works in progress would have similar problems: if I hand out a part finished work for people to look at and give feedback (or just because they are interested even if they won't have feedback they feel worth giving), can they copy that work in progress as I don't offer it for sale? The same for content that otherwise "leaks" out early. I suppose an "I intend to sell the finished work" clause might help here, but you would have to specify that every time you talk about the item as otherwise you open up to "well you didn't say..." arguments.
And of course it could easily be worked around the other way: all content producers have to do is say "I'll sell you this for {exorbitant value}" - no sales due to the price but it is offered for sale. You could add a "reasonableness" clause to your rule, but any such clause has no choice but to be vague and wide open to interpretation/argument.
PCs are not dying. They are just finally (nearly) completely commodity items.
Unfortunately for the likes of HP the thing that the market doesn't need in that situation is either large behemoth producers or an innovator. At this point all that happens (except for the high end which I cover in a second) is that the small players bring the price down to near cost while adding small improvements. There simply isn't anywhere massively different for PC form (box on/under desk, big screen) to go for now - people generally don't need more power or more features (hardware wise) than we already have there - hence all the R&D and consumer interest is going into phones and tablets. Of course there are the niche markets but the only one in the desktop arena is hard-core gamers and that isn't a very large market with the current state of gaming - the other power niches (high-end CAD users, other number crunchers, people with large databases to process, and so forth) are all moving (if they have not already moved) towards online processing or at very least their own little server farm (the likes of HP can still make a business there I'm sure, though that is not relevant to discussion about the "PC" market)
From a personal computing point of view we already have a reboot: it is the phones and tablets (and TVs & related set-top boxes, but in terms of both OS, other software, and hardware, these overlap the tablet market so much you can't really consider them much separate - the R&D for each feeds into all). If the likes of HP can't jump on that bandwagon they'll have to wait for (or have the luck to find or create) the next big thing or and/just hope their other markets (the server and large-scale "solution" markets) can make up for the drop-off in their personal computer market share.
There isn't going to be a revolution in the PC market. There will be many small improvements, maybe something large enough over time to be called evolution. There is no need for any such revolution: they do many jobs well, and to do the other jobs people want other forms are more suitable. They are not dying out though - they are just dropping into the relatively stagnant area of solved problems so you can't sell a new one to the user every year or two. The revolution is the developments in over areas (phones and tablets, for the time being), and even they are getting to the "we have all the features, the just need refining" stage already so in the next year or few we'll be saying the same things about tablets/phones/settops/TVs/etc needing a reboot.
OT: Metro is effectively a "skin" or layer on 7?
Windows 8 is to Windows 7 what XP was to 2000:
* From the users PoV most of the changes are superficial (the new "skin" generally, the ribbon everywhere, explorer enhancements like the fancy long operation progress boxes)
* There is a lot more going on under the hood to make some of those changes possible, or to make other improvements like general efficiency and hardware support
* Most users won't know about, care about, or need to care about the above internal changes, so unless they want to UI update there is no reason to upgrade in the near future.
Many stuck with Windows 2000 until soon before it dropped out of security patch support because for many it worked so didn't need fixing, particularly because people didn't want to volunteer for a new UI when they were used to the old one (even though a lot of the newness could be turned off). I strongly suspect the transition from 7 (or Vista, or XP for those still using it) to 8 will be similarly drawn out, if not worse because the UI change is more significant.
The "registry" is still there?
Yes, and will be for many years to come. Far too much depends on that core feature for it to be removed in any way any time soon. Some code might be using alternate config storage methods/stores/APIs but if the registry went away much would stop working and translating everything it does to a new method (so said method can replace it transparently) would probably be somewhat impractical.
then the proper course of action would have been to remove the hosts file feature totally
IIRC you still need posix compliance (or the ability to claim it such that your claims can not be rubbished too easily) for your OS to be used in many US agencies, and the hosts file is one of the many minor points mentioned in that specification. Presumably that spec says something about having the feature, but does not say anything about effectively disabling it in this way.
The Windows 7 firewall can definitely do the job described by the poster a few comments up, which is what the poster you replied to was referring to.
There are things (a great many, depending on how far down the "it is rare people would need to" scale) iptables can do that the built-in Windows firewall can't do, but the matter currently being discussed certainly isn't one of them.
You are right to by cynical, but there is definitely more going on here than just the assault/rape case even if big bad Uncle Sam isn't the driving force.
If it were just because of that case it would have been handled quite differently - there are procedures in place for interviewing the accused here and all those options have been offered but refused: so they want him back there for some reason and the rape does not seem to warrant that determination on its own because at this point (IIRC he hasn't even officially been charged for that crime yet: they are officially just asking him to "help with their enquiries") those procedures should be more than sufficient.
can't wait to see for the BSOD on that.
It would bring a whole new meaning to the part that reads "beginning dump of memory".
To work that would need to be a fairly targeted attack: picking a few marks and working on them. A scatter-gun attack as usually used by scammers will simply alert the world to the problem and make all marks take the "no one will believe you" line, and the more targeted approach would take a lot more time and effort (and ability): while the payout could be more than worth it long term, I doubt any scammer will take the risk of waiting for as long as the scheme could take to "pay out".
The only case where this sort of attack is a concern IMO is when the attacker is someone that you know, rather than a more remote scammer, who sees reason to be out to get you.
There have been suggestions that some of the scammers use this as a mark filter: people put off by the spelling/grammar would be unlilkely to follow through to the end anyway so put them off early so you can concentrate on the others. People who fall for the scam despite the presentation are better quality marks and more more likely to pay out (either because they have done something wrong and are feeling guilty, or because they don't speak the language well enough to spot the telltail problems, or simply because they are just plain thick).
Though I think it more likely that the simpler explanation (most of the scammers simply fail to create a good presentation in the target language) is more likely at least in most cases.
The implication is that they are not listening to people in general about issues that affect people in general when using their products, i.e. implying that there is a groundswell of opinion on some of the matters (so there are general problems rather than it just being a few fussy/opinionated/entitled individuals moaning) which they are ignoring.
Of course just moaning on random unrelated forums is not going to help, slashdot is not the right place for instance, people won't be heard if they don't speak in the right direction. Of course finding the right place to comment could be difficult for a large organisation like Google. TheOatmeal wasn't trying to be heard really (well, not specifically heard by Google), he was just making a passing reference to the sate of affairs for comic value, though he has managed what it would seem many have not and got the attention of someone who is in a relevant team at the company.
The 80Mbyte seems to be a network limit, not a bulk disc throughput limit. A quick test reading ~10Gb (in two large files not many small ones) locally on the server gets just under 120Mbyte/sec overall. I might be able to get better from those drives but I wasn't overly careful about stripe alignment and such (they are 4K sector 2Tb disks using Linux's software RAID). I'm not sure if those particular files are closer to the slow end of the spinning metal or closer to the fast end. Pulling the same files to the Windows box from a Samba share is where I mentioned 80Mbyte (in fact a quick test with the same two files shows it top out at just over 65Mbyte). The server doesn't have anything SSD but the desktop box uses them for internal storage, I've not benchmarked copying from them to the external SSD though (I have copied large things that way, but just not paid attention while it happened, where I have watched the transfer rates while copying from the server. The server has no USB3, so plugging the drive in directly would not get any benefit from the local connection due to USB2 throughput limits. I've not investigated where that drop from ~120Mbyte/s to ~65 comes from. It could be a mix of Samba not being terribly efficient (it does sit eating >65% of a CPU core while the operation proceeds), me having cheap NICs and a cheap switch, and me not having jumbo frames enabled anywhere (the network is shared with devices that I think don't support larger frames), and probably some other factors I've not considered.
For 24Tb of data you want more than 12x 2Tb drives. At least 18x 2Tb so you can apply RAID5 (or some compound RAID based on arrives of RAID5) or 24x2Tb for RAID1, so that you can survive at least some drive failure situations without data loss. Be careful of array rebuild times with drives that big though. ZFS is something to consider with an array this size (with smaller drives preferably as re-image/re-checksum times will not be short with this over large drives either).
You are right that anything like this would be much less hassle in the long run than many separate external drives, and their PSUs (something I'd not considered).
USB 2.0 provides 480Mbps of (theoretical) bandwidth. So unless you go Gigabit all over your network (not unreasonable), you won't beat it with a NAS. Even then, it's only 1-and-a-bit times as fast as USB working flat-out (and the difference being if you have multiple USB busses, you can get multiple drives working at once).
The 480Mbps is nowhere near what you will see in practise, unlike network speeds which are far closer to the rated maximum. Most USB drives I've seen top out at somewhere between 25 and 30MByte/sec, and if there are no other bottlenecks it isn't unusual to see 100Mbyte/sec from a gbit switched network. My main desktop pulls things from the fileserver at around 80Mbyte/sec, which is as fast as local reads tend to be on that array. So you are right about 100mbit networks: that'll be the bottleneck not USB, but gbit networking should outdo USB2 by at least a factor of 2, possibly 3, maybe even more if you have better drives in you main storage array than I do.
Before trying to run several USB drives to max out your network bandwidth, consider that you will taking the source disks too. Unless they are SSDs having 2, 3, or more concurrent bulk reads going on may not be any faster than one concurrent read as all the extra head movements will wipe out the bulk speed potential. If the OP's 24Tb is spread over numerous physical drives this need not ban an issue though (with planning careful enough to ensure there aren't two bulk processes reading from the same physical devices.
And USB 3.0 would beat it again.
That it would. I have an SSD in a USB3 enclosure, and it can happily consume 80Mbyte/sec read over my little network. It might even be able to do better than that: I've not measured a bulk write read from the internal SSD yet.
And 10Gb between the client and a server is an expensive network to deploy still.
Granted, eSATA would probably be faster but there's nothing wrong with USB for such tasks if you *don't* want to provide Gigabit connections everywhere and (presumably) greater-than-gigabit backbones.
If I wanted more speed than USB3+gbit can provide (due to the size of data being backed up on each run) I'd be plugging the backup device(s) in locally to the source (vie eSATA, USB3, or such) rather than using the network (though again taking note to be careful how things are done if trying to use more than one backup device at once).
For the size of data being described, I'd not want a set of USB drives to be my primary backup solution though.