Toshiba is right: the physical disc is dead. No one is going to buy Blu-Ray players like no one bought HD-DVD players. Everyone is going to download their HD movies onto servers.
What's the compelling HD download service? I mean this sounds good and all but it also sounds like what the HD-DVD people are just saying because they got beat. Is there anything that really suggests more contents is being downloaded than people will buy yet? Or is that just the sort of thing you say as you tuck your tail and walk away?
Stringer-san, if you're listening. It never is set up like this, never do you get a softball like this. Slash blu-ray player prices NOW! Attack when there is blood in the water!
I'm sorry but the auther hasn't presented any compelling reason why this is a 'good idea'(tm) and I can think of several reasons this is a 'bad idea'(tm). Do have I have to mention active X, proprietary languages, and 'broken' sites because of it? Then the need for Web.Devs. job skills increase significantly and become much more cumbersome.
There are some clear reasons to dump javascript. The obvious ones are its terrible performance and the complete lack of safty. Did you know there are port scanners built in javascript? You go to a web site and it quietly attempts to "navigate" to your internal network and it can report back what it finds. There are also
demonstration attacks where they'll do the same thing and post the common factory passwords to your router, I'm not sure if this has been done in the wild but I can
't think of a good reason for it not to happen.
Something like Java's notion of signed byte coded jars would be nice. As long as we're running any code from a remote system, a completely secure VM is needed and some way to grant trust with TLS/SSL could be lovely. While we're at it, giving just a little bit of attention to the performance would help also. It doesn't need to be compiled C like performance but Javascript is bloody slow and we just rely on more and more and more of it. My desktop has 4GB of ram, a core2 duo and firefox starts to really bog down with 10-15 tabs open, it's only frustrating because the older machine was replaced precisely because of that slowness.
If we were to come up with a more intelligent way of packaging javascript-ng, or whatever it is, and do some performance work with it. Another thing I'd like to see would be reuse and storing of common javascript packages. How many websites use scriptaculous? Any reason we can't pull a verison of it from a specific place?
Well they really shouldn't put that kind of stuff in an email to begin with if that's their policy.
Looking at the grand scheme, I can't imagine too many benefits to running your own exchange or notes, or whatever email system.
There are some security benefits and a whole lot of security risks. Even at businesses with full IT staffs, it's a pain, there are issues with storage and email retention, there are issues with their damn filters as they attempt to fight spam and viruses, there are issues with portability, issues with server maintenance. I'll mention spam again which seems to result in expensive, yet still crappy software products cobbelled on to exchange. On top of that all that, it's such a broken protocol to begin with, until some sort of robust replacement emerges (im2000 appears to be stalled) the best thing might be to outsource it to google, yahoo and hotmail. Hosting email just seems like a rat hole for most companies.
Now if Google is really smart about this, they'll cater it to companies, provide vanity domains and some extra security type features and control for management. Likewise, if I was starting another company tomorrow, I'd probably totally scuttle outlook and exchange and use gmail and google calendar and maybe the whole google beta suite of products as the foundation of my IT until I got profitable.
What's sicker still, if google, hotmail and yahoo could agree to a couple things, they could probably rewrite smtp themselves, who knows what percentage of actual email they transfer, they could add seamless auth and encryption and start to rid the world of spam.
Highschool is a bit old. I first encountered Logo on the Apple II around 2nd or 3rd grade.
You could do a lot more with it than turtle graphics but that was the draw. I look back fondly, it is probably one of the reasons I'm a software engineer now, it wasn't hard to make the computer do something that was satisfying to a you kid. I fear now that there aren't enough analogs, Squeak is very cool, Alice might be it, it's just that expectations are so different. SImply drawing a picture on the screen and writing little subroutines to make interesting designs was very satisfying the the 40columns of green text world.
Skype is good. It works, it's easy, it supports your platform. Seems like there will be ways to make money from it, that doesn't seem like a really difficult problem to solve.
Now what I don't get is how/why ebay is in the mix, this doesn't seem to have anything to do with what ebay does.
Actually, they explicitly forbid running services, which bittorrent tends to act like.
If you want to just buy access, you should upgrade and get a tier 1 peer and pay ARIN for some IP space and everything else... Comcast sells home users access and they can just about do whatever they want from their AUP and as a user you're completely free to terminate your contract with them and buy a different service. There are no bandwidth promises or anything else from $40/month vendors. You're really buying a lot less than you might think.
Has anyone actually established that this is at all true? "Widely reported" isn't the same thing as "true" I have yet to see any substantiation of this story, nor have I noticed any degradation in torrent use. Demonstrating that they are forging resets isn't that easy, some device might be resetting but pinning it to comcast is a little bit more challenging. It's definitely possible, with a handful of traces from different peers you could reasonably establish it. When this rumor first cropped up I had a handful of torrents running, pulling down legitimate Linux DVDs and I can't say I've noticed any change in anything via Comcast. Judging by the general network savvy of most of the torrent crowd, I'd place my wager on some jackass thinking it's taking too long to get something and trying to pin it on Comcast and the rest of the geek community jumping on the bandwagon. Just looks like a lot of speculation to me and then the collective hatred of large companies sort of picked it up and ran with it.
Whether or not they buy a vote seems almost immaterial to me. You know vendors tend to have seats on a lot of these types of standards committees and there is always a lot of personal protection going on. You'll never see Cisco at an IETF meeting backing something Cisco hasn't built already or backing something that is alternative to what Cisco is doing unless they've already decided to do it and by then they've usually built one.. IBM usually has an agenda before they get to ANSI or ISO or whatever the standards group is. Every company is that way. The smart ones go out of their ways to document the ever loving hell out of their technologies too so that it stands a better chance in these standards meetings and the really smart ones are open to accepting new ideas for their technology so that it can be standardized and have some consensus. (Do you really want to fight some obscure issue, of just ammend your standard to include it and make another company vote for it as a friend and supporter of the standard?)
What's more alarming to me is that there is simply no way that OOXML is a rational standard, the voters clearly are not expert at it, nobody is backing it with an alternative implementation. I don't even believe an alternative implementation is really possible at this point, it's just not clear to me. Can you imagine how the internet wouldn't even exist if IETF standards were approached this way? It is very clear to me that the folks voting on this standard have not read it, it's 7000 pages, there simply isn't a way that they did. I don't want to out right just bash MS but they came late to the game and they simply have no track record of pushing for open standards, it's almost against their very nature. To ramrod this though will ultimately just undermine what it means for something to be "standard" and standards committee members should be aware of that, this won't make OOXML the standard so much as it will undermine the very concept of a standard for this technology. The fact that nobody on the committee is putting the brakes on to me indicates just how broken this comittee is and that the standard should be either dropped or restarted. If they aren't taking is seriously, then let's just kill the standard, I'd rather have none than a bullshit one.
Open document formats is something that is fairly important. I bet you'd have trouble dealing with a lot of common document formats from just 15 years ago. Anyone process Wordperfect 4.2 and 5 files? How about Wordstar? Multimate anyone? Sure you can probably find a way to important them and make them usable but what about in another 5 years? As we digitize more documents, right now, we're almost making sure that in 100 years this will be a dark spot in history because they won't be able to process what records may exist, if they can get them off of the media (if the media is even good) It's good for mankind to produce some well defined, open and sane standards, it's also pretty good for business, how many formats does Office currently try to support? How much does that cost? Imagine if Office 2015 only supported like 3. I don't know what kinds of numbers MS spends on it, I'm guessing millions of dollars a year just on supporting Office file formats though and I couldn't imagine it really impacting the use of Office, it's a fine piece of software. I really don't even care if it's properly documented OOXML instead of the OASIS/OO.org XML format, it just needs to be properly documented and that documentation needs to be vetted before a vote happens. Maybe that's what MS really wants but these committee members are representing corporate interests as well as national ones in some cases and I can't possibly see how they can justify the job
they are doing. No standard is better than a really fucked up one.
I'm a little dismissive of the mystique around the required "super hackers" that never need to look for work but there is a ton of great advice on just hiring people.
I'm an engineer. Been there and done a lot of that. I'm not going to say I'm one of the greatest but not too shabby, I've built some stuff and made some good money, you, know left a few marks.. As a more senior guy I've been taking on more and more leadership and to be completely honest, I like to think there are things I know to look for and catch, but I suck at hiring and team building. At the end of the day it's about building products, selling them and making money and the balance between people you think are a good personality match vs. the people that are technically good enough vs. people that are actually motivated and want to work and be successful is hard. We've hired folks we though were good personality matches for the team and turned out to be terrible technically and completely unmotivated, as much as you might want to like them, you simply won't when they are trying to play big business "CYA" games and not actually contributing to the team.
I'm kind of dealing with a situation now, we're a small team, 4 or 5 developers and 2 testers. We hired the 5th developer based largely upon a recommendation from one of the testers. He's a marginal tester to be honest, good guy, just not super motivated, why we hired his recommendation is looking more and more stupid by the day, we value recommendations. After reading Joel's book I've found like 6 or 7 indicators that probably would have flagged this guy that we simply didn't think about. We were in a hurry, we thought the req would go away, etc.. Honestly, I'd rather have one fewer people and better morale that this guy, seriously in 6-8 months of having him, I cannot point to a single substantial contribution. Now we get to go through the process of firing him which sucks for every one involved also.
Basically, you always want smart people, you want motivated people, people that do a good job, people that have some passion, good communicators, strong team people that know what it is to be on a team, you want all of that stuff, all of the time and it's hard to find. We pretend that parts don't matter or don't matter as much. Having shitty people on your team just flat out sucks, doesn't matter how good everything else is.
I'm somewhat startled by the number of outright disbelievers.. It's clearly not as big or as bad (relatively speaking) as the.com collapse. Of course if you don't see it coming or don't think it will and it hits you, then that pretty much blows.
During.com anyone who could get Word to spit out HTML or fire up dreamweaver was capable of getting a fairly high paying job and "becoming a web developer." Not surprisingly, simply building web sites isn't such a great business. VCs and investors ate it up though for a while.
This time around there are 2 differences, more places are trying to get foreign tech laborers to do the actual "real work" and they build a web site here and the websites are fancier. There are some businesses that I just don't see lasting, I don't see how building an AJAX website is really that different than building just a website, still at the end of the day you need to make money and do something.
Maybe I'm getting old school here, but when I look at digg, and I see the top rated "news" stories of sorts, it really makes me think that the masses are really stupid. It's all very formulaic. New Harry Potter book on Piratebay before it comes out just isn't "technical" news to me, it's not really even news to me. Maybe I just don't get it, there is pirated shit on piratebay and that's somehow now news? Or someone figured out how to photograph or photocopy a book and pirate it that way, that's the news? (I mean, people did that in the 1980's, they just photocopy books..) Or maybe it's just "oh snap! MPAA/RIAA/whoever does books-AA you got served!" and while it's not news, someone wanted to say that and the masses "dugg" it. If someone somehow managed to sort of do the whole social networking thing and link up like minded folks (maybe some sort of passive IQ test) then maybe it'd all be more interesting. Personally, I can only take so many Lohan and Hilton stories before I just look elsewhere for "news."
Now I don't know if there is a bubble that is bursting exactly, seems like the money involved is a tiny fraction of what it used to be but if you think this kind of shit is the way forward, I certainly hope that it's not. We can do so much better.
What if the CPU does the decryption in realtime? Then you can use encrypted binaries to prevent certain types of attacks because the attacker would have to inject encrypted instructions in to an overflow...
I think a patent was just filed for this kind of technology.
Totally guy, I know, I'm still only using like 384 of the 640K that I'll ever need...
The big advantage is interconnect can be speed up dramatically. On clusters the interconnect fabric is the slow spot. Even on Pentium-D and typical SMP systems, the interconnect kind of sucks. The architectures are simple enough that it doesn't matter a ton, hypertransport is pretty good, relatively, but we're nowhere near the bandwidth of what the actual chip is capable of.
I think there has been enough history to demonstrate that we crave more cycles and can always find ways to use them for something or other. The question you should be asking is how will we use 80 core. SMP clustering has been proposed, we've pretty much figured out how to do 4-way SMP in a pretty efficient manner, so what if we make a cluster of 4-way machines out out of those 80 core? Then use some kind of grid technology on top of it?
Several vendors already have 80+ way multiprocessing system so I assume that it's not too hard to deal, it's not really that new, just getting it on a single waffer is. The bigger problem is the amount of single threadedness we have in out application code. In the UNIX world it has almost been taboo to spin off threads unless they are absolutely needed. The writing is on the wall though, I'd bone up on MPI/MPICH and other grid technologies and work/queue models and get better and spinning off threads... If Intel is going to ship 80way in the next 5 years, you'll probably have 8 to 16way on typical desktops in that time frame. All the video game systems are already multiprocessing, 3 to 6 wayish for the xbox360, and like 7 to 9ish for the PS3 and I'm not sure of the exact specs of Wii but it'll have at least 2 way with SMT. That amount of torque also screams virtualization, put a single app on a single "machine" and you can run at least 80 of them, if we simplify things enough we might even be able to make things secure.
When has it been done prior to any actions to create a monopoly?
I think that's a big difference here, people have latched on to a kind of worst case scenario that really hasn't happened anywhere and they're pushing that as the reason to regulate. I think if there is a real problem the regulate to fix it but so far there isn't a problem, just a potential threat. Subsequently, if ISPs were to start blocking some traffic and giving preferential treatment to other traffic, it creates a ripe market where an alternative ISP could really add value. Microsoft, Google, those are fricking huge companies with a lot of money, they are completely with the means to build a new large ISP. I think IBM has created multinational networks and sold them off at least twice now.
What are the down sides of net neutrality? What's the potential worst case there? We all get dumbed down to the same speed because that's fair? Something like that? ISPs refuse to innovate and increase speed? Or maybe they cannot increase speed until they can do it for all of their customers at the same time, which in effect will end innovation or any more speed. I simply don't see a lot of good coming from regulation before there is a problem to fix.. especially when we're talking about doing it to telcos that are already clustered fucked up.
I've been under the impression that jboss was looking for a suitor for a while now, since they've had a little bad blood with IBM. Either Oracle or HP, they want to cash in and there is nothing wrong with that.
Featurewise, they are the best opensource app platform going. Now does Larry integrate Jboss, harmoniously, with Oracle? That I very much doubt. I've been wrong before, look at all the stuff Sun is doing, I still don't trust them but they actually did it and they are slowly earning my respect and Oracle to act similarly.
From IBM's actions, it looks like they want to have a 2 prong strategy, you sell high dollar, industrial grade webspheres where you make a lot of jack and then you push geronimo and free stuff to the rest of the world so that they are building those types of applications in the first place. Oracle could be doing the same thing, Jboss is a damn fine entry grade product too, it can easily hang against many of the big boys.
By their testing methodology there were clear winners, it just wasn't what they had wanted so they said "no clear winner."
This is simple. CPU performance scales sublinearly. CPU cost scales nearly exponentially. It's very simple math, by their measurements and calculations the best price/performance will always be one of the cheapest CPUs. You get almost 10x the performance per dollar with current CPU prices.
There also happens to be a "sweet spot" where the cost increments start jumping more rapidly, that's where most "geeks" like to buy and somehow justify the performance is worth the $20-$50. The performance isn't that much though. They still want to justify the p4EE or the AthlonFX chips that cost over a grand.
Of course the rules change if you're actually processing the whole time and can save some hours
"I know this, an iPod is a device that plays music" -- Bruce Springsteen
So the palm had a great balance, calendar, notes, addresses, maybe a little more but that's 85%.
PocketPC comes along, word, IE, blah blah blah.
Who's going to buy a palm pilot without all that shit that the pocketPC has now? Just do the geek math on that one. At the same time, pocketpcs are a terrible laptop replacement. Why not just carry the thinkpad or powerbook around? I see lot's of people with palmPCs and very few that really use them. They get bought for features that aren't used. Meanwhile, palm hit this on the years back.
There is no discussion, Voltron is about team work and being stronger together than solo. The characters are rich and textured and have different levels of pride, greed and desire for personal glory. Blazing sword isn't formed immediately because usually they want to try to be the hero by them selves before the realize the value of the team and forming the mighty Voltron. At which point, Voltron is a fair and honorable hero and prefers to feel his opponents for weakness before dispatching with them. Plus several of the minor robeasts have been dispatched with other weapons from Voltron's arsenal, he's not one dimensional. He's sporting. Nothing to discuss, the characters are just too complex to allow for Voltron to immeditely form blazing sword. It's kind of like the "nuclear option" and not used until it has to be.
Slashdot Mirror
What's the compelling HD download service? I mean this sounds good and all but it also sounds like what the HD-DVD people are just saying because they got beat. Is there anything that really suggests more contents is being downloaded than people will buy yet? Or is that just the sort of thing you say as you tuck your tail and walk away?
It's not BSD either. It's the UofI License which is BSD like.
Stringer-san, if you're listening. It never is set up like this, never do you get a softball like this. Slash blu-ray player prices NOW! Attack when there is blood in the water!
There are some clear reasons to dump javascript. The obvious ones are its terrible performance and the complete lack of safty. Did you know there are port scanners built in javascript? You go to a web site and it quietly attempts to "navigate" to your internal network and it can report back what it finds. There are also demonstration attacks where they'll do the same thing and post the common factory passwords to your router, I'm not sure if this has been done in the wild but I can 't think of a good reason for it not to happen.
Something like Java's notion of signed byte coded jars would be nice. As long as we're running any code from a remote system, a completely secure VM is needed and some way to grant trust with TLS/SSL could be lovely. While we're at it, giving just a little bit of attention to the performance would help also. It doesn't need to be compiled C like performance but Javascript is bloody slow and we just rely on more and more and more of it. My desktop has 4GB of ram, a core2 duo and firefox starts to really bog down with 10-15 tabs open, it's only frustrating because the older machine was replaced precisely because of that slowness.
If we were to come up with a more intelligent way of packaging javascript-ng, or whatever it is, and do some performance work with it. Another thing I'd like to see would be reuse and storing of common javascript packages. How many websites use scriptaculous? Any reason we can't pull a verison of it from a specific place?
Looking at the grand scheme, I can't imagine too many benefits to running your own exchange or notes, or whatever email system. There are some security benefits and a whole lot of security risks. Even at businesses with full IT staffs, it's a pain, there are issues with storage and email retention, there are issues with their damn filters as they attempt to fight spam and viruses, there are issues with portability, issues with server maintenance. I'll mention spam again which seems to result in expensive, yet still crappy software products cobbelled on to exchange. On top of that all that, it's such a broken protocol to begin with, until some sort of robust replacement emerges (im2000 appears to be stalled) the best thing might be to outsource it to google, yahoo and hotmail. Hosting email just seems like a rat hole for most companies.
Now if Google is really smart about this, they'll cater it to companies, provide vanity domains and some extra security type features and control for management. Likewise, if I was starting another company tomorrow, I'd probably totally scuttle outlook and exchange and use gmail and google calendar and maybe the whole google beta suite of products as the foundation of my IT until I got profitable.
What's sicker still, if google, hotmail and yahoo could agree to a couple things, they could probably rewrite smtp themselves, who knows what percentage of actual email they transfer, they could add seamless auth and encryption and start to rid the world of spam.
You could do a lot more with it than turtle graphics but that was the draw. I look back fondly, it is probably one of the reasons I'm a software engineer now, it wasn't hard to make the computer do something that was satisfying to a you kid. I fear now that there aren't enough analogs, Squeak is very cool, Alice might be it, it's just that expectations are so different. SImply drawing a picture on the screen and writing little subroutines to make interesting designs was very satisfying the the 40columns of green text world.
Now what I don't get is how/why ebay is in the mix, this doesn't seem to have anything to do with what ebay does.
If you want to just buy access, you should upgrade and get a tier 1 peer and pay ARIN for some IP space and everything else... Comcast sells home users access and they can just about do whatever they want from their AUP and as a user you're completely free to terminate your contract with them and buy a different service. There are no bandwidth promises or anything else from $40/month vendors. You're really buying a lot less than you might think.
Has anyone actually established that this is at all true? "Widely reported" isn't the same thing as "true" I have yet to see any substantiation of this story, nor have I noticed any degradation in torrent use. Demonstrating that they are forging resets isn't that easy, some device might be resetting but pinning it to comcast is a little bit more challenging. It's definitely possible, with a handful of traces from different peers you could reasonably establish it. When this rumor first cropped up I had a handful of torrents running, pulling down legitimate Linux DVDs and I can't say I've noticed any change in anything via Comcast. Judging by the general network savvy of most of the torrent crowd, I'd place my wager on some jackass thinking it's taking too long to get something and trying to pin it on Comcast and the rest of the geek community jumping on the bandwagon. Just looks like a lot of speculation to me and then the collective hatred of large companies sort of picked it up and ran with it.
What good are the docs, for $50 or for free, if the they aren't correct and can't be used to produce and accurately parse ooxml documents?
What's more alarming to me is that there is simply no way that OOXML is a rational standard, the voters clearly are not expert at it, nobody is backing it with an alternative implementation. I don't even believe an alternative implementation is really possible at this point, it's just not clear to me. Can you imagine how the internet wouldn't even exist if IETF standards were approached this way? It is very clear to me that the folks voting on this standard have not read it, it's 7000 pages, there simply isn't a way that they did. I don't want to out right just bash MS but they came late to the game and they simply have no track record of pushing for open standards, it's almost against their very nature. To ramrod this though will ultimately just undermine what it means for something to be "standard" and standards committee members should be aware of that, this won't make OOXML the standard so much as it will undermine the very concept of a standard for this technology. The fact that nobody on the committee is putting the brakes on to me indicates just how broken this comittee is and that the standard should be either dropped or restarted. If they aren't taking is seriously, then let's just kill the standard, I'd rather have none than a bullshit one.
Open document formats is something that is fairly important. I bet you'd have trouble dealing with a lot of common document formats from just 15 years ago. Anyone process Wordperfect 4.2 and 5 files? How about Wordstar? Multimate anyone? Sure you can probably find a way to important them and make them usable but what about in another 5 years? As we digitize more documents, right now, we're almost making sure that in 100 years this will be a dark spot in history because they won't be able to process what records may exist, if they can get them off of the media (if the media is even good) It's good for mankind to produce some well defined, open and sane standards, it's also pretty good for business, how many formats does Office currently try to support? How much does that cost? Imagine if Office 2015 only supported like 3. I don't know what kinds of numbers MS spends on it, I'm guessing millions of dollars a year just on supporting Office file formats though and I couldn't imagine it really impacting the use of Office, it's a fine piece of software. I really don't even care if it's properly documented OOXML instead of the OASIS/OO.org XML format, it just needs to be properly documented and that documentation needs to be vetted before a vote happens. Maybe that's what MS really wants but these committee members are representing corporate interests as well as national ones in some cases and I can't possibly see how they can justify the job they are doing. No standard is better than a really fucked up one.
I'm a little dismissive of the mystique around the required "super hackers" that never need to look for work but there is a ton of great advice on just hiring people.
I'm an engineer. Been there and done a lot of that. I'm not going to say I'm one of the greatest but not too shabby, I've built some stuff and made some good money, you, know left a few marks.. As a more senior guy I've been taking on more and more leadership and to be completely honest, I like to think there are things I know to look for and catch, but I suck at hiring and team building. At the end of the day it's about building products, selling them and making money and the balance between people you think are a good personality match vs. the people that are technically good enough vs. people that are actually motivated and want to work and be successful is hard. We've hired folks we though were good personality matches for the team and turned out to be terrible technically and completely unmotivated, as much as you might want to like them, you simply won't when they are trying to play big business "CYA" games and not actually contributing to the team.
I'm kind of dealing with a situation now, we're a small team, 4 or 5 developers and 2 testers. We hired the 5th developer based largely upon a recommendation from one of the testers. He's a marginal tester to be honest, good guy, just not super motivated, why we hired his recommendation is looking more and more stupid by the day, we value recommendations. After reading Joel's book I've found like 6 or 7 indicators that probably would have flagged this guy that we simply didn't think about. We were in a hurry, we thought the req would go away, etc.. Honestly, I'd rather have one fewer people and better morale that this guy, seriously in 6-8 months of having him, I cannot point to a single substantial contribution. Now we get to go through the process of firing him which sucks for every one involved also.
Basically, you always want smart people, you want motivated people, people that do a good job, people that have some passion, good communicators, strong team people that know what it is to be on a team, you want all of that stuff, all of the time and it's hard to find. We pretend that parts don't matter or don't matter as much. Having shitty people on your team just flat out sucks, doesn't matter how good everything else is.
During .com anyone who could get Word to spit out HTML or fire up dreamweaver was capable of getting a fairly high paying job and "becoming a web developer." Not surprisingly, simply building web sites isn't such a great business. VCs and investors ate it up though for a while.
This time around there are 2 differences, more places are trying to get foreign tech laborers to do the actual "real work" and they build a web site here and the websites are fancier. There are some businesses that I just don't see lasting, I don't see how building an AJAX website is really that different than building just a website, still at the end of the day you need to make money and do something.
Maybe I'm getting old school here, but when I look at digg, and I see the top rated "news" stories of sorts, it really makes me think that the masses are really stupid. It's all very formulaic. New Harry Potter book on Piratebay before it comes out just isn't "technical" news to me, it's not really even news to me. Maybe I just don't get it, there is pirated shit on piratebay and that's somehow now news? Or someone figured out how to photograph or photocopy a book and pirate it that way, that's the news? (I mean, people did that in the 1980's, they just photocopy books..) Or maybe it's just "oh snap! MPAA/RIAA/whoever does books-AA you got served!" and while it's not news, someone wanted to say that and the masses "dugg" it. If someone somehow managed to sort of do the whole social networking thing and link up like minded folks (maybe some sort of passive IQ test) then maybe it'd all be more interesting. Personally, I can only take so many Lohan and Hilton stories before I just look elsewhere for "news."
Now I don't know if there is a bubble that is bursting exactly, seems like the money involved is a tiny fraction of what it used to be but if you think this kind of shit is the way forward, I certainly hope that it's not. We can do so much better.
It is a much more approachable and friendly group during the unstable tree periods.
I think a patent was just filed for this kind of technology.
The big advantage is interconnect can be speed up dramatically. On clusters the interconnect fabric is the slow spot. Even on Pentium-D and typical SMP systems, the interconnect kind of sucks. The architectures are simple enough that it doesn't matter a ton, hypertransport is pretty good, relatively, but we're nowhere near the bandwidth of what the actual chip is capable of.
I think there has been enough history to demonstrate that we crave more cycles and can always find ways to use them for something or other. The question you should be asking is how will we use 80 core. SMP clustering has been proposed, we've pretty much figured out how to do 4-way SMP in a pretty efficient manner, so what if we make a cluster of 4-way machines out out of those 80 core? Then use some kind of grid technology on top of it?
Several vendors already have 80+ way multiprocessing system so I assume that it's not too hard to deal, it's not really that new, just getting it on a single waffer is. The bigger problem is the amount of single threadedness we have in out application code. In the UNIX world it has almost been taboo to spin off threads unless they are absolutely needed. The writing is on the wall though, I'd bone up on MPI/MPICH and other grid technologies and work/queue models and get better and spinning off threads... If Intel is going to ship 80way in the next 5 years, you'll probably have 8 to 16way on typical desktops in that time frame. All the video game systems are already multiprocessing, 3 to 6 wayish for the xbox360, and like 7 to 9ish for the PS3 and I'm not sure of the exact specs of Wii but it'll have at least 2 way with SMT. That amount of torque also screams virtualization, put a single app on a single "machine" and you can run at least 80 of them, if we simplify things enough we might even be able to make things secure.
I think that's a big difference here, people have latched on to a kind of worst case scenario that really hasn't happened anywhere and they're pushing that as the reason to regulate. I think if there is a real problem the regulate to fix it but so far there isn't a problem, just a potential threat. Subsequently, if ISPs were to start blocking some traffic and giving preferential treatment to other traffic, it creates a ripe market where an alternative ISP could really add value. Microsoft, Google, those are fricking huge companies with a lot of money, they are completely with the means to build a new large ISP. I think IBM has created multinational networks and sold them off at least twice now.
What are the down sides of net neutrality? What's the potential worst case there? We all get dumbed down to the same speed because that's fair? Something like that? ISPs refuse to innovate and increase speed? Or maybe they cannot increase speed until they can do it for all of their customers at the same time, which in effect will end innovation or any more speed. I simply don't see a lot of good coming from regulation before there is a problem to fix.. especially when we're talking about doing it to telcos that are already clustered fucked up.
Sun stock has been in the toilet for so long, they have to do shit like this to try and raise it enough to get out.
Fuckers are trying to get away two pistons cheaper! Screw 'em. If it was a v-12 I'm there, but for only a v-10? Got to hell SCO!
Featurewise, they are the best opensource app platform going. Now does Larry integrate Jboss, harmoniously, with Oracle? That I very much doubt. I've been wrong before, look at all the stuff Sun is doing, I still don't trust them but they actually did it and they are slowly earning my respect and Oracle to act similarly.
From IBM's actions, it looks like they want to have a 2 prong strategy, you sell high dollar, industrial grade webspheres where you make a lot of jack and then you push geronimo and free stuff to the rest of the world so that they are building those types of applications in the first place. Oracle could be doing the same thing, Jboss is a damn fine entry grade product too, it can easily hang against many of the big boys.
This is simple. CPU performance scales sublinearly. CPU cost scales nearly exponentially. It's very simple math, by their measurements and calculations the best price/performance will always be one of the cheapest CPUs. You get almost 10x the performance per dollar with current CPU prices.
There also happens to be a "sweet spot" where the cost increments start jumping more rapidly, that's where most "geeks" like to buy and somehow justify the performance is worth the $20-$50. The performance isn't that much though. They still want to justify the p4EE or the AthlonFX chips that cost over a grand.
Of course the rules change if you're actually processing the whole time and can save some hours
So the palm had a great balance, calendar, notes, addresses, maybe a little more but that's 85%.
PocketPC comes along, word, IE, blah blah blah.
Who's going to buy a palm pilot without all that shit that the pocketPC has now? Just do the geek math on that one. At the same time, pocketpcs are a terrible laptop replacement. Why not just carry the thinkpad or powerbook around? I see lot's of people with palmPCs and very few that really use them. They get bought for features that aren't used. Meanwhile, palm hit this on the years back.
iPods play music.
Totally, Hiroshima was bombed intentionally... There was nothing accidental about it.
It's also blazing sword, not flaming sword.
There is no discussion, Voltron is about team work and being stronger together than solo. The characters are rich and textured and have different levels of pride, greed and desire for personal glory. Blazing sword isn't formed immediately because usually they want to try to be the hero by them selves before the realize the value of the team and forming the mighty Voltron. At which point, Voltron is a fair and honorable hero and prefers to feel his opponents for weakness before dispatching with them. Plus several of the minor robeasts have been dispatched with other weapons from Voltron's arsenal, he's not one dimensional. He's sporting. Nothing to discuss, the characters are just too complex to allow for Voltron to immeditely form blazing sword. It's kind of like the "nuclear option" and not used until it has to be.
What Cray processor are you thinking of? I think you're confused.
Well technically, Steve promised his customers a 3Ghz G5 and didn't deliver. We don't know what went on between Apple and IBM.