Sawmill. It ain't free (Speech nor Beer), but neither is Google Analytics. If you put a price on your users' privacy, anyways.
Also, it's a bit more techie-oriented than GA, a bit more faster and quite a bit more powerful. In the end, it's a matter of taste - I like it.
You're obviously right about the difficulty of fixing problems, where the openness of Open Source really comes to play. It certainly sucks to know that only a single bit needed to be flipped to make something work that doesn't; and not getting any kind of support from the manufacturer sucks even worse.
Though, in defense of nVidia your problem does seem rather unique. Until very recently it was my understanding that most any screen made in the past decade ought to provide an EDID -- the standard's fifteen bloody years old. While I understand your aggravation against nVidia, I'd also point a finger the other way, to whomever would build a high-end screen without a working EDID implementation. Single-Link DVI is sufficient for 1080p60, so we're probably looking at screens over $2k a pop.
Anyways, while Open Source (with an equal feature set)) is always better than closed solutions, I even more strongly prefer a working over an incomplete or non-working solution. I'd love to switch to a free (speech) driver, but in my experience, none of them works even remotely as well as the kernel-tainting blob.
Nice, didn't know that. I'm planning to wait till about Q3/2009 for a new performance rig, but if ATi manages to catch up to nVidia's performance 'till then, I may just opt for the really open option. Thanks.
Three things:
- None of the futures you mentioned contradicts any of the others. Quite obviously Blue Bonnet won't predict the future of the storage market and Minute Maid won't be the first companyto know about new processes in CPU manufacturing.
- What's the future according to Minute Maid anyways? Really, I'm intrigued!
- Did you notice the interesting parallel between the future according to ATT and what the american government seems to be steering to? More bars in more places (and as many people behind them as possible(?))? What a strange coincidence...
nVidia will probably continue their controversial blob model (i.e. you get a binary object plus the source to a kernel module that, with the help of said object, works as a driver). Purists rage against it because it's against freedom and-so-on, pragmatists tend to like the full 3D acceleration that comes with it.
Intel is going the Open Source road, trying to be as open as possible. Unfortunately, from a performance PoV their hardware sucks. Their products are intended as consumer-level, chipset integrated solutions and, considering that, work nicely. Don't try any 3D games, though.
ATi opened a lot of specs, so community-developed and completely open drivers are on the horizon. Unfortunately the horizon is quite far away and the movement towards it is similar to a kid on a tricycle. The situation is prone to improve though. Performance-wise, ATi may be a good choice if you'd like to play the occasional game, but they don't really compare to nVidia (which is unlikely to change soon).
In the end, I'm going to stick to nVidia in the near future, using intel wherever low energy consumption is strongly desired (i.e. notebooks and similar). ATi just ain't my cup of tea, I wouldn't be putting a red card in a Windows box either, but my preference of nVintel is just such -- a preference. Go with whatever suits you best.
Businesses start migrating some time after SP1, which still is pretty recent. Expect those 15% of Vista to go up in the next few quarters, going on for years (hey, some businesses are migrating from Win2k right now).
Linux? Not so much. Many businesses are slowly getting to know Linux on the server side of their business, gradually switching some boxes first, before migrating any important parts. Some already are in a later stage of this transition and may be sampling Linux on the desktop. Very few are as far as actually switching on a large scale would necessitate.
OS X? Not so much, usually way less. Linux has the great advantage of getting a foot in the door thanks to being free (beer), OS X is frickin' expensive. Free run-anywhere LiveCDs could've made a big difference, especially during the start phases of Vista, but Apple decided to pass on that chance. They probably get to keep the small, "hip" and creative businesses, but those tend to get by with less than ten desktops overall while MSFT gets to keep the more than ten thousand seat enterprises.
Also, responding to both of your initial examples (Linux ISOs, MMO (I'm guessing WoW, right?) patches): Both publishers you speak of are just that, publishers. They pay a lot for any kind of useful transit bandwidth to their clients and want/need to minimize that. Every byte of data handled in P2P fashion is a byte that doesn't use up their expensive bandwidth.
ISPs have to support the whole network from the server on which the content lies down to the customers. Spreading downloads to a P2P model doesn't lessen the amount of data traveling through that, it might only take some load off their download servers (at the expense of traffic predictability), which doesn't make that much of a difference as an $50/mo leased box will be able to satisfy two GbE links, which is equal to about 650 terabytes a month at perfect or some 200 TB (200k movies at 1 GB/movie) with realistic values. Stack a couple of these with a midsize storage backend and you've got an almost unbeatable solution.
Okay, trying again. Three possibilities and their effect on the telco central and more remote (i.e. from core to last mile and client) networks and (the expensive part) transit bandwidth:
A: Streaming servers near the central network, exclusively server-to-client traffic. Some network strain at the edge of the centre, little processing power required in both servers and routing equipment due to single, predictable streams. Little last-mile strain, single, easy-to-shape stream. Little to no upstream usage from client (this is expensive), no transit (this is *really* expensive).
B: P2P within telcos network. Same (bandwidth-wise) strain within network centre -- the P2P data needs to be transferred from one client to another. More processing power because of lots of different streams from one client to another. More processing power required for the distribution servers when parts need to be seeded to clients. Also, some additional strain on the outer parts because of client upstream. Still no transit.
C: B, spread out over the internet. Same intranet strain in core and outer network as in B, extra internet traffic which means having to buy transit which is expensive.
Look, don't get me wrong -- I like the P2P model from a user's perspective. It means there's lots of content, available for free, which is great. I like it even more from the perspective of an internet content publisher. A few uploads using my expensive bandwidth but hundreds of people can get my work.
But, as I should've explained enough by now, there's nothing to like about P2P from a telco's perspective. Especially not for streaming and when compared to a solution as simple as a few boxes sendfile()ing that content out for nearly no expense. P2P is unpredictable, tends to hog bandwidth and confuses the hell out of most traffic shaping methods.
The gov't bureaucracy takes it's time, speeding that up is usually impossible. Usually, there are interesting alternative routes, though. (I, for one, got a passport in less than 48 hours from asking about for it to holding it in my hand)
The same goes for our telcos, though insisting on a delivery date and generally being a pain in their lower back will help things (as oppossed to the B permit).
Well, anyways: Welcome to Switzerland, enjoy your stay;)
Strangely enough, $(cent)i(euro)nt(currency)£(currency)g¥ is the only word I could think of containing all of c, e, l, o, s and y. An interesting connection, really.
Also, neither cent, euro or generic currency symbols are supported with or without JS in the new discussion system, making this post way less funnier:/
How often has a BitTorrent download that did not mostly originate from dedicated seeds on server-grade connections max out most of your bandwidth for 95% of the time it took to be downloaded?
Linear protocols like http or ftp are perfectly suited for streaming because they obtain the content the way you want to consume it: Starting at point x, progressing towards the end.
P2P protocols are, by their very nature, based in small blocks so lots of slow peers can replicate sets of files quickly amongst each other, but that also means large parts can be done while you're waiting for a slow peer to finish those first few kilobytes. Good priorization can set off some of that, but not all of it. Servers acting as seeds can set off all of it, but participating in a BitTorrent swarm, serving up "random" pieces requires a bit more work than zero-copy-sendfile()ing a file on a system that's optimized for just that. From some experience in high-bandwidth (several nodes with 100+mbps links plus others hooked up via DSL, quite life-like) BitTorrent situations, the former will only be profitable if 60-80% of the distribution is done by "other" peers.
You're absolutely right. Unfortunately the few grand that can actually be raised aren't nearly enough to seriously participate in the game.
If they're used, however, to start or perfect an entirely new game in another league, that league may just outperform the old one and suddenly become interesting.
If Ogg Vorbis was to suddenly provide 192 kbps of AAC fidelity in an 48 kbps stream, do you think manufacturers would just pass the "60 000 songs instead of 20 000 the competition offers" argument up? They can and will play the marketing game and may just be convinced by technical merits that let them compete better.
I'm not advocating to donate less to KDE, but apart from the page you mentioned, KDE gets corporate support (as already mentioned) and has several patrons.
Also, IIRC, KDE is very closely related to SuSE which then again belongs to Novell now. With that kind of background support, they don't seem to be struggling to pay their bandwidth bills, so I tend to stick to smaller projects without corporate backing. I find my money to make more of a difference than it'd make to KDE, Gnome, OpenSuSE or Fedora (and so on).
I'd rather donate to people who actually create useful stuff than pointless and annoying campaigns like DefectiveByDesign, BadVista or PlayOgg. The FSF has some great projects (fighting software patent and drm legislation, the RIAA Expert Witness fund), but Microsoft badmouthing or format advancement are total wastes of money (to me).
Software and formats should be adopted based on their merits (i.e. be better than the rest and rely mostly on word-of-mouth propaganda) and not throw valuable resources into the bottomless pit that is modern marketing.
Oookay, you sure that's your definitive answer? Just asking, 'cause it's a pretty dumb one. Their service will obviously running on their servers which will obviously be located in their network. They won't send themselves a bandwidth bill for that since, well, suing themselves for not paying would be pretty dumb and you don't send invoices if you're not going to collect, eh?
Now there may be some costs for the actual servers, but we're talking the easiest kind of data to distribute here -- predictable, linear flows of data. Their actual cost will probably be around 3-5 cents per gig of movie. Plus, of course, portal/application maintenance, plus network maintenance. If they're expecting to charge $3 per movie and each is about one gig, switching to p2p distribution can save them 1-2% of the distribution cost at the expense of massively worse service quality and really turning them to hypocrites. Seems unlikely.
Depends. Try driving from Berne to Zurich around 1700. Train (station to station, both quite centralized) takes 58 minutes while driving (Highway on- to offramp, rather off centre) at an ideal 124 km/h (120 km/h legal limit plus 4 km/h safety margin for laser measurements) those 121.5 kilometers will take 0.8 minutes more. Assuming a very optimistic 110 km/h, driving will actually add about 15% to your travel time.
Berne - Zurich may be a somewhat unfair example as it's one of the most important sections -- but large parts of the swiss train network are similarly fast and extremely on time. Going by train will get you to most places equally fast as driving and let you do work, eat relaxedly, enjoy a movie or enjoy some quality time with your significant other joining the rail high club. Or so I heard. Heh.
Also, regarding your sig: Wine 1.0 is scheduled (and apparently moving ahead on time) for release in june, 1.0-rc1 is out.
Also, discussion2 has a bug: Text entered after clicking the "Preview" button, before the textbox disappears to make room for the actual preview isn't displayed there but submitted anyway. PoC: the trailing d in parent.
The ad clearly discourages people from publicly disclosing their SSN. It's a bit like Jackass -- interesting things are done by professionals who proceed to discourage viewers from repeating said things.
I'm pretty sure normal customers could get an extended contract for a few grand more -- risky behaviour deserves jacked up rates. Auto insurance for a racecar driver or medical for a stunt man should be more expensive because they are way more probable to create lots of expenses.d
Nope. Calls diverted to your answering machine* are not subject to roaming charges as they don't leave the original destination network. I'm not quite sure about the "minutes" situation over in the U.S., but in the civilized world (speaking from a GSM PoV), incoming and discarded calls will never cost you money, minutes or anything else. I suspect the very same is true over in the US of A as the argument for charging minutes for incoming calls is that they can be discarded (and sent to voice mail).
* "Answering machine" as in "hosted voice mail". This includes ATT's Visual Voicemail as well as all solutions provided by your carrier. There are (few, mostly for the Symbian platform) applications that'll take the call and store messages on your phone, but you'd know if you'd installed one of these.
What the..? Killing her would be a terrible waste of valuable resources. Group her and similarly convicted people together in teams of about five, add a guard and an assault rifle. Who need their house painted, a shed built or their fields ploughed get to hire such teams for $25/hr of which $20 goes to the guard, $2 towards foods to keep the prisoners alive and $3 towards a giant pit into which the prisoners are locked at night.
Lots of problems (overflowing prisons, too much criminality, illegal immigration) solved in one absolutely constitutional plan that totally respects human rights. Heh.
I think it's age 13 and up; children below that age appear to be legally protected against (m)any sorts of non-governmental privacy invation. I don't know the exact legal basis, but along with many other services the Windows version of Adobe Flash gathers a lot more information for people 13 and over.
Slashdot Mirror
Sawmill. It ain't free (Speech nor Beer), but neither is Google Analytics. If you put a price on your users' privacy, anyways.
Also, it's a bit more techie-oriented than GA, a bit more faster and quite a bit more powerful. In the end, it's a matter of taste - I like it.
You're obviously right about the difficulty of fixing problems, where the openness of Open Source really comes to play. It certainly sucks to know that only a single bit needed to be flipped to make something work that doesn't; and not getting any kind of support from the manufacturer sucks even worse.
Though, in defense of nVidia your problem does seem rather unique. Until very recently it was my understanding that most any screen made in the past decade ought to provide an EDID -- the standard's fifteen bloody years old. While I understand your aggravation against nVidia, I'd also point a finger the other way, to whomever would build a high-end screen without a working EDID implementation. Single-Link DVI is sufficient for 1080p60, so we're probably looking at screens over $2k a pop.
Anyways, while Open Source (with an equal feature set)) is always better than closed solutions, I even more strongly prefer a working over an incomplete or non-working solution. I'd love to switch to a free (speech) driver, but in my experience, none of them works even remotely as well as the kernel-tainting blob.
Nice, didn't know that. I'm planning to wait till about Q3/2009 for a new performance rig, but if ATi manages to catch up to nVidia's performance 'till then, I may just opt for the really open option. Thanks.
Well played, sir.
Three things: - None of the futures you mentioned contradicts any of the others. Quite obviously Blue Bonnet won't predict the future of the storage market and Minute Maid won't be the first companyto know about new processes in CPU manufacturing.
- What's the future according to Minute Maid anyways? Really, I'm intrigued!
- Did you notice the interesting parallel between the future according to ATT and what the american government seems to be steering to? More bars in more places (and as many people behind them as possible(?))? What a strange coincidence...
nVidia will probably continue their controversial blob model (i.e. you get a binary object plus the source to a kernel module that, with the help of said object, works as a driver). Purists rage against it because it's against freedom and-so-on, pragmatists tend to like the full 3D acceleration that comes with it.
Intel is going the Open Source road, trying to be as open as possible. Unfortunately, from a performance PoV their hardware sucks. Their products are intended as consumer-level, chipset integrated solutions and, considering that, work nicely. Don't try any 3D games, though.
ATi opened a lot of specs, so community-developed and completely open drivers are on the horizon. Unfortunately the horizon is quite far away and the movement towards it is similar to a kid on a tricycle. The situation is prone to improve though. Performance-wise, ATi may be a good choice if you'd like to play the occasional game, but they don't really compare to nVidia (which is unlikely to change soon).
In the end, I'm going to stick to nVidia in the near future, using intel wherever low energy consumption is strongly desired (i.e. notebooks and similar). ATi just ain't my cup of tea, I wouldn't be putting a red card in a Windows box either, but my preference of nVintel is just such -- a preference. Go with whatever suits you best.
Get a T61 with Linux, available right from Lenovo and listed on their site. :)
Businesses start migrating some time after SP1, which still is pretty recent. Expect those 15% of Vista to go up in the next few quarters, going on for years (hey, some businesses are migrating from Win2k right now).
Linux? Not so much. Many businesses are slowly getting to know Linux on the server side of their business, gradually switching some boxes first, before migrating any important parts. Some already are in a later stage of this transition and may be sampling Linux on the desktop. Very few are as far as actually switching on a large scale would necessitate.
OS X? Not so much, usually way less. Linux has the great advantage of getting a foot in the door thanks to being free (beer), OS X is frickin' expensive. Free run-anywhere LiveCDs could've made a big difference, especially during the start phases of Vista, but Apple decided to pass on that chance. They probably get to keep the small, "hip" and creative businesses, but those tend to get by with less than ten desktops overall while MSFT gets to keep the more than ten thousand seat enterprises.
Also, responding to both of your initial examples (Linux ISOs, MMO (I'm guessing WoW, right?) patches): Both publishers you speak of are just that, publishers. They pay a lot for any kind of useful transit bandwidth to their clients and want/need to minimize that. Every byte of data handled in P2P fashion is a byte that doesn't use up their expensive bandwidth.
ISPs have to support the whole network from the server on which the content lies down to the customers. Spreading downloads to a P2P model doesn't lessen the amount of data traveling through that, it might only take some load off their download servers (at the expense of traffic predictability), which doesn't make that much of a difference as an $50/mo leased box will be able to satisfy two GbE links, which is equal to about 650 terabytes a month at perfect or some 200 TB (200k movies at 1 GB/movie) with realistic values. Stack a couple of these with a midsize storage backend and you've got an almost unbeatable solution.
Okay, trying again. Three possibilities and their effect on the telco central and more remote (i.e. from core to last mile and client) networks and (the expensive part) transit bandwidth:
A: Streaming servers near the central network, exclusively server-to-client traffic. Some network strain at the edge of the centre, little processing power required in both servers and routing equipment due to single, predictable streams. Little last-mile strain, single, easy-to-shape stream. Little to no upstream usage from client (this is expensive), no transit (this is *really* expensive).
B: P2P within telcos network. Same (bandwidth-wise) strain within network centre -- the P2P data needs to be transferred from one client to another. More processing power because of lots of different streams from one client to another. More processing power required for the distribution servers when parts need to be seeded to clients. Also, some additional strain on the outer parts because of client upstream. Still no transit.
C: B, spread out over the internet. Same intranet strain in core and outer network as in B, extra internet traffic which means having to buy transit which is expensive.
Look, don't get me wrong -- I like the P2P model from a user's perspective. It means there's lots of content, available for free, which is great. I like it even more from the perspective of an internet content publisher. A few uploads using my expensive bandwidth but hundreds of people can get my work.
But, as I should've explained enough by now, there's nothing to like about P2P from a telco's perspective. Especially not for streaming and when compared to a solution as simple as a few boxes sendfile()ing that content out for nearly no expense. P2P is unpredictable, tends to hog bandwidth and confuses the hell out of most traffic shaping methods.
The gov't bureaucracy takes it's time, speeding that up is usually impossible. Usually, there are interesting alternative routes, though. (I, for one, got a passport in less than 48 hours from asking about for it to holding it in my hand)
;)
The same goes for our telcos, though insisting on a delivery date and generally being a pain in their lower back will help things (as oppossed to the B permit).
Well, anyways: Welcome to Switzerland, enjoy your stay
Strangely enough, $(cent)i(euro)nt(currency)£(currency)g¥ is the only word I could think of containing all of c, e, l, o, s and y. An interesting connection, really. :/
Also, neither cent, euro or generic currency symbols are supported with or without JS in the new discussion system, making this post way less funnier
How often has a BitTorrent download that did not mostly originate from dedicated seeds on server-grade connections max out most of your bandwidth for 95% of the time it took to be downloaded?
Linear protocols like http or ftp are perfectly suited for streaming because they obtain the content the way you want to consume it: Starting at point x, progressing towards the end.
P2P protocols are, by their very nature, based in small blocks so lots of slow peers can replicate sets of files quickly amongst each other, but that also means large parts can be done while you're waiting for a slow peer to finish those first few kilobytes. Good priorization can set off some of that, but not all of it. Servers acting as seeds can set off all of it, but participating in a BitTorrent swarm, serving up "random" pieces requires a bit more work than zero-copy-sendfile()ing a file on a system that's optimized for just that. From some experience in high-bandwidth (several nodes with 100+mbps links plus others hooked up via DSL, quite life-like) BitTorrent situations, the former will only be profitable if 60-80% of the distribution is done by "other" peers.
You're absolutely right. Unfortunately the few grand that can actually be raised aren't nearly enough to seriously participate in the game.
If they're used, however, to start or perfect an entirely new game in another league, that league may just outperform the old one and suddenly become interesting.
If Ogg Vorbis was to suddenly provide 192 kbps of AAC fidelity in an 48 kbps stream, do you think manufacturers would just pass the "60 000 songs instead of 20 000 the competition offers" argument up? They can and will play the marketing game and may just be convinced by technical merits that let them compete better.
I'm not advocating to donate less to KDE, but apart from the page you mentioned, KDE gets corporate support (as already mentioned) and has several patrons.
Also, IIRC, KDE is very closely related to SuSE which then again belongs to Novell now. With that kind of background support, they don't seem to be struggling to pay their bandwidth bills, so I tend to stick to smaller projects without corporate backing. I find my money to make more of a difference than it'd make to KDE, Gnome, OpenSuSE or Fedora (and so on).
I'd rather donate to people who actually create useful stuff than pointless and annoying campaigns like DefectiveByDesign, BadVista or PlayOgg. The FSF has some great projects (fighting software patent and drm legislation, the RIAA Expert Witness fund), but Microsoft badmouthing or format advancement are total wastes of money (to me).
Software and formats should be adopted based on their merits (i.e. be better than the rest and rely mostly on word-of-mouth propaganda) and not throw valuable resources into the bottomless pit that is modern marketing.
Oookay, you sure that's your definitive answer? Just asking, 'cause it's a pretty dumb one.
Their service will obviously running on their servers which will obviously be located in their network. They won't send themselves a bandwidth bill for that since, well, suing themselves for not paying would be pretty dumb and you don't send invoices if you're not going to collect, eh?
Now there may be some costs for the actual servers, but we're talking the easiest kind of data to distribute here -- predictable, linear flows of data. Their actual cost will probably be around 3-5 cents per gig of movie. Plus, of course, portal/application maintenance, plus network maintenance. If they're expecting to charge $3 per movie and each is about one gig, switching to p2p distribution can save them 1-2% of the distribution cost at the expense of massively worse service quality and really turning them to hypocrites. Seems unlikely.
Depends. Try driving from Berne to Zurich around 1700. Train (station to station, both quite centralized) takes 58 minutes while driving (Highway on- to offramp, rather off centre) at an ideal 124 km/h (120 km/h legal limit plus 4 km/h safety margin for laser measurements) those 121.5 kilometers will take 0.8 minutes more. Assuming a very optimistic 110 km/h, driving will actually add about 15% to your travel time.
Berne - Zurich may be a somewhat unfair example as it's one of the most important sections -- but large parts of the swiss train network are similarly fast and extremely on time. Going by train will get you to most places equally fast as driving and let you do work, eat relaxedly, enjoy a movie or enjoy some quality time with your significant other joining the rail high club. Or so I heard. Heh.
Also, regarding your sig: Wine 1.0 is scheduled (and apparently moving ahead on time) for release in june, 1.0-rc1 is out.
Also, discussion2 has a bug: Text entered after clicking the "Preview" button, before the textbox disappears to make room for the actual preview isn't displayed there but submitted anyway. PoC: the trailing d in parent.
The ad clearly discourages people from publicly disclosing their SSN. It's a bit like Jackass -- interesting things are done by professionals who proceed to discourage viewers from repeating said things.
I'm pretty sure normal customers could get an extended contract for a few grand more -- risky behaviour deserves jacked up rates. Auto insurance for a racecar driver or medical for a stunt man should be more expensive because they are way more probable to create lots of expenses.d
Nope. Calls diverted to your answering machine* are not subject to roaming charges as they don't leave the original destination network. I'm not quite sure about the "minutes" situation over in the U.S., but in the civilized world (speaking from a GSM PoV), incoming and discarded calls will never cost you money, minutes or anything else. I suspect the very same is true over in the US of A as the argument for charging minutes for incoming calls is that they can be discarded (and sent to voice mail).
* "Answering machine" as in "hosted voice mail". This includes ATT's Visual Voicemail as well as all solutions provided by your carrier. There are (few, mostly for the Symbian platform) applications that'll take the call and store messages on your phone, but you'd know if you'd installed one of these.
Hm you're right. I even had a thought about re-using her organs before writing that post... ;)
Ah well, just kill her quickly and cheaply...
What the..? Killing her would be a terrible waste of valuable resources. Group her and similarly convicted people together in teams of about five, add a guard and an assault rifle. Who need their house painted, a shed built or their fields ploughed get to hire such teams for $25/hr of which $20 goes to the guard, $2 towards foods to keep the prisoners alive and $3 towards a giant pit into which the prisoners are locked at night.
Lots of problems (overflowing prisons, too much criminality, illegal immigration) solved in one absolutely constitutional plan that totally respects human rights. Heh.
I think it's age 13 and up; children below that age appear to be legally protected against (m)any sorts of non-governmental privacy invation. I don't know the exact legal basis, but along with many other services the Windows version of Adobe Flash gathers a lot more information for people 13 and over.