I remember working for a certain networking hardware company that was not cisco and was fiercely competing with Cisco..... However all the site's switches and routing equipment that wasn't used explicitly for testing or development purposes? Cisco......
A *lot* of companies won't eat their own dog food, and that is really funny to me...
In most any circumstance, anyone with an association in good standing with an employer has an almost infinite advantage over anyone else, even if on paper they are much much much less apparently qualified. I know, this just agrees with your point, and I too have been pissed off by this phenomenon, but I have to admit, there is sound reasoning behind it.
The major problem with hiring people is that it is typically a long term investment with a good deal of commitment on the part of the employer. For obvious reasons, there is an extreme benefit in determining the qualifications of a candidate before hiring. Despite all best efforts in interviews and resumes, it really isn't possible to gauge a candidate very accurately except to weed out the most incompetent. Even if someone isn't up to par, they can fake it in an interview long enough to look better than another candidate that just can't deliver a good sales pitch at interview time. Social networking provides the applicant credibility, and provides the employer better accountability ('Jim recommended him, so Jim has ultimate responsibility').
Of course, further benefits include lower startup time for a new employee and, over time, increased morale and teamwork, the foundation of which was already partially built off of company time before even the first day. Someone coming into work with a few familiar faces will tend to hit the ground running better, being less timid about getting starting advice and knowing a comfortable person to ask about things as they start running. Even without asking questions, the closer contact allows the existing employ to detect problems long before they would have been seen with a stranger.
So yes, to those without the networks, it is really unfair, and it is unfair when you build a really nice social network only to have it shattered by a site shutdown (both have been my situation), but it does increase candidate review reliability overall and increase net productivity.
All that said, I think this should go out the window in the sales/marketing world. The best evaluation is the pitch they throw for themselves in a short term. If a guy can sell you on taking him for a job over other candidates, I can't imagine a better qualification for that line of work...
If they are too obvious and blatent, clearly demonstrating a huge amount of resource to discredit a competitor backfires. If it is worth that much resource and *requires* that much resource to make a competitor look bad, MS would be more obviously fearful of linux and if it takes a lot of money to offset the good press adequately, that also makes a statement about the quality of linux vs. MS.
Kinda like if someone makes a huge deal of how they don't worry about something, and they repeatedly evoke that fact without solicitation, you know they are lying.
Me too. Like I will take any opportunity to reinvent how it happened, maybe it would turn out like one of those exciting, last-minute hollywood escapes. It takes a concious effort to supress such optimism and remember all that happened is set in stone and had an unhappy ending.
The suggestion is rather unhelpful. I am quite happily employed with a good job, after several months of searching early last year. If you can't even sell yourself to employers, you certainly are going to have an extremely difficult time selling your skills to customers. In other words, good business means someone with a head for marketing (i.e. lying and exagerrating). It also means just someone who is simply good at business in general over technical skills. It also means a lot of startup resource, which is as rare as employment.
Personally, the path I took to employment was finding a contracting company, who is more willing to talk to candidates, figure out the right lies and exagerrations they can pull off on behalf of the applicant, and pass them off to the potential buyer in a nice, favorable light. I wasn't told about their lies and exagerrations until the buyer was interviewing me, and I was mad I was caught off guard, but it seemed to be par for the course. I served my time as a contractor and the company was so satisfied with my work, I got hired on as a full time employee. Since then I've been getting friends who have had difficulty to use contracting companies, with near 100% success. Representing your skillset and professional experience is a lot harder than you may realize, and it takes the marketing 'guts' of the companies to add the lies it thinks you can pull off in the interview, or at least to get you to the interview so even though the company discovers you have exagerrations, they know your skill is good and the lies aren't your fault...
Actually, it depends on the wiring layout. If the initial wiring job is meticulously done and the servers have decent back layouts, it isn't so bad. I've delt with setups where only network and power are plugged in, and those are particularly sparse in 1U systems.
Blade servers are of course really easy, a 7U chassis to hold 14 servers, with only a few connections to share amongst them.
SuSE was the only significant contributor anyway..
on
United Linux Dead
·
· Score: 1
Novell knows SuSE can fly on its own, any excuse to get them out of helping other somewhat unsucessful linux companies is good news to them. I know, Turbo and Conectiva have enjoyed moderate success in Asia and South America, but in the global scheme of things, they didn't have much of a market share.
Well that is all and good, but when you are on the road far removed from such luxury as ethernet or even a computer, nothing beats picking up a phone, dialing a dial-in service, and making modem noises with your mouth for some quick web browsing.
RedHat has in the recent past tried to make it crystal clear that if you want a long period of support, long time of provided updates, and a long product lifecycle in general, that you shell out for the 'enterprise' editions. 2.1 is a 7.x era product and is still well supported and remained the 'latest product' (as defined relative to the RedHat enterprise offering) for a long time. The release of RHEL3 has done nothing to slow that support down, and it looks like these enterprise editions will be similar to MS product lifecycles, which is reasonable. So this move is consistant with their strategy. Their take is that the 'freeloaders' will buy into the Enterprise product line, and if they don't, they weren't worth the effort to appease in the first place. Perhaps a tad short sighted in the scheme of things (bad public image is apparent), but they have failed to really break out of their state as a fledgling company with their old strategy, and, from the business perspective, had little choice and not much to risk. They hope to make RHEL a corporate standard, and therefore being short on new features relative to the community will not be so obvious, and then the companies can feel good about long lifecycles and their 'latest and greatest' Red Hat.
Of course, the bad thing is that these *extremely* short lifecycles will be held up high by the likes of MS as examples of how RedHat will leave you out in the cold long before MS will. Even if not completely true, it has enough truth in it for MS to put a strong, believable, verifiable spin on the situation. That is the consequence of this strategic change that they will have to face. And don't try to make it sound like 7.x is *ancient*, it feels that way to the Linux community because that is the pace it is used to moving at, but in a company, it is still a 'new' product.
I personally use Gentoo, but in professional work I deal primarily with SuSE and RedHat, and for both technical and business reasons, I think SuSE has managed to get things right. With SuSE, they have a much more complete, coherent feeling solution. Things just work. Their strategy to all sorts of things is far more flexible once you appreciate it. And with the Enterprise edition, they have enough partnerships in place to truly offer a comprehensive solution. In dealing with RH Enterprise offerings, it is essentially RH9 with some spit and polish. No extras, nothing you couldn't really get from any free distribution, with only RH support to differntiate it. SLES, however, includes a few niceties, such as an included, well behaved, supported JVM. Sure, you can download those for free, but it is important in such a product to have a complete solution out of the box.
Couple this to their pricing model (RH WS costs at least $179, SuSE Professional costs $79), and it seems like a much more reasonable product when compared to the likes of RH and MS.
For North America and Europe, SuSE and RedHat are virtually the only 'professional' Linux platform solutions. Others have some fantastic technical merits, but are not real professional-grade businesses for the enterprise to deal with. I love Gentoo, I like Debian, and on technical merit alone I would place both above RedHat and SuSE (as long as the user is a highly competent linux enthusiast), but the support infrastructures are simply not there in a meaningful way as far as businesses are concerned.
I agree that film is superior and will likely always have a charm all its own.... Exactly like analog vinyl records will always have their charm, but like analog vinyl records, the technology is catching up extremely fast to soon be 'good enough' for even the most discerning of viewers (note I said 'soon', implying not today, but the future). Your estimate on high resolution cameras will likely prove to be extremely pessimistic judging from the rate of everything else electronic.
As to price, that argument made absolutely zero sense. You confuse the technology with the final medium. You can get analog film developed onto really good, or really crappy photo paper, and even the best will fade with time. Sure, the negative will typically hold out *better*, but will still over time degrade. You can print out images from digital cameras onto the exact same range of photo paper pretty much, with the same results. Sure you use an inket on a sheet of normal paper and it will be crap, but you can go high end and even send images to be developed by professional labs just like film. And unlike the negatives of film, the digital picture can be preserved with guaranteed perfection, and can be duplicated at will flawlessly. Even if the media it is stored on degrades to unreadable, it was probably copied to new media before that. This scenario is simply not possible with analog strategies without infinite precision. Duplication will degrade the result without question in film.
Different formats is a repeat of number 1. There is a more obvious tradeoff between amount of view captured on camera versus quality in digital, but provided you used incredibly high resolution digital format, you can produce whatever size you want. As the fidelity of cameras increase, this issue goes away completely.
Now I agree film's nature gives it great advantage on the count of quality, but durability, that is simply impossibly tilted toward digital.
Assuming you store the media properly and choose good media, it may very well last 40 years. And if you can read the data in its entirety, then it is pristine.
Now forget about the media. CD-R media properly stored will outlive CDR drive technology, meaning people will compile CDR collections to DVD media in the near future, and repeat the process as new media technology/capacity comes out. The data is perfectly preserved across new media. Even if CDR was the end-all media that no one would go beyond, you can duplicate old CDRs to new CDRs as desired. Though the digital nature of the pictures means less detail, it also means perfectly (and verifiably) transferable. If a negative goes, any means to copy will be analog methods, and thus without infinite precision, loss of detail is inevitable. For each duplication with even the utmost detail, the picture will degrade some.
Yes, that is true, but consider that the amount of resources invested overall increases, as investments are less frequently total losses with this forewarning. Bad economically for the geographic locale predicted to quake, good in general for investment. Fewer resources lost and lower risk all comes out to healthier investments. This is all assuming that false positives are *extremely* rare and that it is also capable of predicting >90% of disasters, change either variable and the picture changes.... Of course some investments would go up (construction companies and the like would clean up on 'quake-proofing' non-movable structures).
Now, back to the geographic locale's state. Sucks to be them economically, but let's say you had the choice of having equal chance at having investments near your house, or knowing that in ~3 months, a catastrophic quake that could kill you is extremely likely. The economic problems are both temporary and offset by the value of increasing awareness to save lives. 4 months later after the quake happens, no further risk is seen and companies are already lined up to do reconstruction of whatever was destroyed. 3 months of warning allows a community to do a lot to protect investments from harm and prepare a rapid recovery plan for high-risk, high-value structures that may be destroyed. So while in the short term economic conditions are potentially bad, having 3 months warning provides better long term economic circumstances.
Seeing as how I can't find an x40, either it is a mistake, or you are using a model that is not even released.... Seems hardly fair if the latter is the case to expect FreeBSD to support the hardware reliably...
As to the notion of never running out of address space 'never, never' as he puts it, I wouldn't be so sure. The 32-bit address space provides 4.2 billion addresses. With that in mind, we are much nearer to exhaustion than current usage would dictate. It is all about the allocation, and if sloppy allocation occurs, the 128-bit address space of IPv6 could be exhausted too. For example, the architecture of current implementations make it so that the smallest subnet anyone will likely allocate are 64-bit networks, and use MAC addresses (or something else, but still 64-bit, because it's easy), so immediately you take the address space down tremendously. Still should be well more than enough for everyone on earth to have a/64 network, but it has yet to be seen whether certain organizations might, for the hell of it, get allocated/8 networks because they can. As near as I can tell, the high 16 bits seem to be somewhat protected, but you never know what will happen. If there is a grab for/8 networks among big players, you have the same problems that IPv4 has today.
As to security implications, it is true that implementations will be for the short term future less tested and therefore likely to contain critical flaws, but still IPv6 code is receiving a fair amount of testing, and critical flaws will not be quite so devastating as you may think, no more than an Apache, Linux Kernel, or MS security exposure, which we have seen all of in fairly recent history without the sky falling.... Of course the wrinkle in this is a lot of the 'home router' concepts that happen to protect common home systems will cease to provide that protection. They provide NAT features, therefore masking to an extent the system behind the device. Despite what the author says about NAT being bad because it doesn't protect against things like browser exploits and physical intruders, NAT is on the level of firewalling in terms of protection. Any reasonable network security person will realize that browser exploits, email worms, and physical intrusion must always be kept in mind, and it has nothing to do with NAT or firewalling. NAT remains effective at, for example, fending off web server and rpc attacks from unsuspecting or experimenting workstations. If NAT goes away (hopefully), people need to be mindful of good old firewalling strategies. Implementations are maturing (experimental ip6tables implementation, for example, is approaching closely the ipv4 iptables featureset). If cable/dsl 'routers' revert to hubs in a wealth of addressing, I expect either cable/dsl 'firewall' devices or increased ISP vigilance to deal with the more widespread system exposure.
All that said, I like IPv6 (my desktop, gateway, and laptops are using IPv6 and each have public IPv6 addresses, keep NAT on IPv4 on some systems), but I (and everyone else) has been waiting and watching a long long time and no encouraging migrations are yet to be seen, and I doubt the near future will bring any incentive to push such a change.
Actually, in my experience, tg3 is *extremely* broken compared to bcm5700 on many applicable cards. Negotiation of link state being a serious problem on some cards.
All check, what's your point? On my A31p the only button not working is the 'ThinkPad' button, the rest present keysyms (except the volume, which works regardless of OS support). DVD playback works about as well as anywhere else (XvMC is the only thing looming that will give additional booost), wireless ethernet is a prism chipset, power management is both apm and acpi. Sound is intel_8x0.
Not a single hardware feature on this laptop fails under linux, and working on other Thinkpads, it holds true across the board in my experience.
I suppose it is your choice, but why bother getting a Mac and skipping out on MacOSX? Their hardware is crappy, Apple cares more about looking nice than lasting more than the warranty period. Once upon a time, they made stuff that worked and stayed durable, but now they make overpriced cheap flimsy plastic and offer hostile customer support when things break. The only thing I like about Apple is that OSX is truly a fantastic blend of 'Grandma usability' and BSD power. Linux can run just as well for 99.9% of users out there on x86 hardware (if you find/make a decent x86 with the same money you shell out for a Mac).
If you couldn't tell, I bought into the hype and bought an iBook a while back and have regretted it since. Spent months trying to convince support the motherboard needed replacement due to video failures, (now well documented), and they eventuall did. The power supply started shocking my lap after 3 months, it was replaced and the replacement died just after warranty was up, and I had to shell out 80 bucks to keep using the damn thing. On top of all that, the lid latch broke and the ethernet port stopped working, so now I can only use the wireless. And each time I called about stuff under Warranty, they were more interested in trying to sell me extended warranties or insisting that Apple didn't do direct hardware warranty service , or admitting they did, but my problems were obviously because I abused/dropped the laptop or something, which never occured.
Meanwhile my Thinkpad has served me well, and it was cheaper. Even a crappy Presario laptop has outlasted that crap iBook. I buy a new system usually at least annually, and Apple has me so off to their products. It is a shame because I think the Apple software developers/testers have *really* got something right in the midsts of everything else going to crap.
You're thinking on hobbyist scale. It as free as your time for hobbyists, and for the computer saavy, that is cheap. Now for those who barely figured out Windows in the first place, not viable. For companies pricing out expensive infrastructures (trying to plan farms of servers and/or provide a common supported corporate desktop), using those resources for any sort of experimentation is completely out of the question. Even delaying deployment schedules for the sake of experimentation is highly unlikely. The managers of the people implementing will do them the 'favor' of doing the research and telling them the direction. This advertising is less about convinincing undecided parties, but reassuring people with existing Microsoft setups that the potential savings in TCO is at least uncertain and that an investment in migrating their infrastructure will not see a sufficient return on investment.
Of course common sense could make this campaign backfire horribly. For example, holding up the shining example of low-end MS run WinTel systems beating out high-end mainframe systems has to set some alarm off in even the most uneducated mind. To get the figures they wanted, they had to change the whole hardware situation to the most unbalanced comparison imaginable. Of course that would hold true, and it has absolutely nothing to do with the software running on the systems, mainframes are intended to give ultimate performance, not affordability. Why didn't they do the comparison with equal hardware? Why did they feel the need to give Linux a handicap? Was MS unable to back up their point without such a drastic measure?
perhaps more clearly written as: $40.25 per "megabit of throughput per second". $1.79 per "peak request per second".
So, by there reckoning, to go from a system capable of handling x requests per second to x+1 requests per second in wintel world, it is 1.79. Just to explain the terminology.
You missed his point. Let's say, hypothetically, KISS said 'kiss off, we are violationg the GPL and don't give a rat's ass.' Ok, so know they are known to be in willful violation of the GPL, and therefore lose license to redistirubte within the bounds of copyright law. The GPL is not a contract, and therefore not adhering to the GPL only means redistribution is prohibited by *copyright* law. Now take them to court because they are redistributing the content without a license. Easy case to prove, but the penalty is based on damages. How much does a free software project lose by having another company copy their stuff? Lost monetary gain is not an issue, as they already give it away for free. So yeah, the company is easily proved to violate copyright, but in doing so, the monetary payout in terms of penalty is likely to be low.
So within the bounds of the legal system, I think the GPL holds little sway. However, what companies are intimidated by is the possibility of a boycott. The market of poeple who care about the GPL and stay aware of these cases maybe relatively small, but still sizable. For an embedded device company in particular, where the main draw is the hardware and the software merely an enabler for selling the hardware, the cost of disclosing source is infinitely smaller than even 300 lost customers due to boycott. Now for software packages that offer little benefit over a lesser known open source package they based themselves on, things change, as disclosing source under GPL means that a) competing almost as good free package gets attention and they lose sales and b) people can redistribute their product under terms of GPL and they face a new competing product they helped to develop and receive no benefit. However, for these companies they would likely want to keep the whole thing quiet (don't want potential customers finding out there is an open source project out there that is almost as good as their commercial product), and make a donation in exchange for the project not complaining loudly about the violation.
What, will it now support them new-fangled horseless carriages?
I know, the joke is that X seems so out of date feature wise in terms of eye-candy cruft, but the average technology level of the 20th century isn't too impressive.
Slashdot Mirror
I remember working for a certain networking hardware company that was not cisco and was fiercely competing with Cisco..... However all the site's switches and routing equipment that wasn't used explicitly for testing or development purposes? Cisco......
A *lot* of companies won't eat their own dog food, and that is really funny to me...
In most any circumstance, anyone with an association in good standing with an employer has an almost infinite advantage over anyone else, even if on paper they are much much much less apparently qualified. I know, this just agrees with your point, and I too have been pissed off by this phenomenon, but I have to admit, there is sound reasoning behind it.
The major problem with hiring people is that it is typically a long term investment with a good deal of commitment on the part of the employer. For obvious reasons, there is an extreme benefit in determining the qualifications of a candidate before hiring. Despite all best efforts in interviews and resumes, it really isn't possible to gauge a candidate very accurately except to weed out the most incompetent. Even if someone isn't up to par, they can fake it in an interview long enough to look better than another candidate that just can't deliver a good sales pitch at interview time. Social networking provides the applicant credibility, and provides the employer better accountability ('Jim recommended him, so Jim has ultimate responsibility').
Of course, further benefits include lower startup time for a new employee and, over time, increased morale and teamwork, the foundation of which was already partially built off of company time before even the first day. Someone coming into work with a few familiar faces will tend to hit the ground running better, being less timid about getting starting advice and knowing a comfortable person to ask about things as they start running. Even without asking questions, the closer contact allows the existing employ to detect problems long before they would have been seen with a stranger.
So yes, to those without the networks, it is really unfair, and it is unfair when you build a really nice social network only to have it shattered by a site shutdown (both have been my situation), but it does increase candidate review reliability overall and increase net productivity.
All that said, I think this should go out the window in the sales/marketing world. The best evaluation is the pitch they throw for themselves in a short term. If a guy can sell you on taking him for a job over other candidates, I can't imagine a better qualification for that line of work...
If they are too obvious and blatent, clearly demonstrating a huge amount of resource to discredit a competitor backfires. If it is worth that much resource and *requires* that much resource to make a competitor look bad, MS would be more obviously fearful of linux and if it takes a lot of money to offset the good press adequately, that also makes a statement about the quality of linux vs. MS.
Kinda like if someone makes a huge deal of how they don't worry about something, and they repeatedly evoke that fact without solicitation, you know they are lying.
Me too. Like I will take any opportunity to reinvent how it happened, maybe it would turn out like one of those exciting, last-minute hollywood escapes. It takes a concious effort to supress such optimism and remember all that happened is set in stone and had an unhappy ending.
The suggestion is rather unhelpful. I am quite happily employed with a good job, after several months of searching early last year. If you can't even sell yourself to employers, you certainly are going to have an extremely difficult time selling your skills to customers. In other words, good business means someone with a head for marketing (i.e. lying and exagerrating). It also means just someone who is simply good at business in general over technical skills. It also means a lot of startup resource, which is as rare as employment.
Personally, the path I took to employment was finding a contracting company, who is more willing to talk to candidates, figure out the right lies and exagerrations they can pull off on behalf of the applicant, and pass them off to the potential buyer in a nice, favorable light. I wasn't told about their lies and exagerrations until the buyer was interviewing me, and I was mad I was caught off guard, but it seemed to be par for the course. I served my time as a contractor and the company was so satisfied with my work, I got hired on as a full time employee. Since then I've been getting friends who have had difficulty to use contracting companies, with near 100% success. Representing your skillset and professional experience is a lot harder than you may realize, and it takes the marketing 'guts' of the companies to add the lies it thinks you can pull off in the interview, or at least to get you to the interview so even though the company discovers you have exagerrations, they know your skill is good and the lies aren't your fault...
Actually, it depends on the wiring layout. If the initial wiring job is meticulously done and the servers have decent back layouts, it isn't so bad. I've delt with setups where only network and power are plugged in, and those are particularly sparse in 1U systems.
Blade servers are of course really easy, a 7U chassis to hold 14 servers, with only a few connections to share amongst them.
Novell knows SuSE can fly on its own, any excuse to get them out of helping other somewhat unsucessful linux companies is good news to them. I know, Turbo and Conectiva have enjoyed moderate success in Asia and South America, but in the global scheme of things, they didn't have much of a market share.
I see you took my advice, awesome ;)
Well that is all and good, but when you are on the road far removed from such luxury as ethernet or even a computer, nothing beats picking up a phone, dialing a dial-in service, and making modem noises with your mouth for some quick web browsing.
I suggest an addendum to your sig:
Gentoo: But I don't mind waiting for tomorrow..
(Not meant to be a flame, I use Gentoo too and love it and the tradeoff is worth it, but it just had to be said).
RedHat has in the recent past tried to make it crystal clear that if you want a long period of support, long time of provided updates, and a long product lifecycle in general, that you shell out for the 'enterprise' editions. 2.1 is a 7.x era product and is still well supported and remained the 'latest product' (as defined relative to the RedHat enterprise offering) for a long time. The release of RHEL3 has done nothing to slow that support down, and it looks like these enterprise editions will be similar to MS product lifecycles, which is reasonable. So this move is consistant with their strategy. Their take is that the 'freeloaders' will buy into the Enterprise product line, and if they don't, they weren't worth the effort to appease in the first place. Perhaps a tad short sighted in the scheme of things (bad public image is apparent), but they have failed to really break out of their state as a fledgling company with their old strategy, and, from the business perspective, had little choice and not much to risk. They hope to make RHEL a corporate standard, and therefore being short on new features relative to the community will not be so obvious, and then the companies can feel good about long lifecycles and their 'latest and greatest' Red Hat.
Of course, the bad thing is that these *extremely* short lifecycles will be held up high by the likes of MS as examples of how RedHat will leave you out in the cold long before MS will. Even if not completely true, it has enough truth in it for MS to put a strong, believable, verifiable spin on the situation. That is the consequence of this strategic change that they will have to face. And don't try to make it sound like 7.x is *ancient*, it feels that way to the Linux community because that is the pace it is used to moving at, but in a company, it is still a 'new' product.
I personally use Gentoo, but in professional work I deal primarily with SuSE and RedHat, and for both technical and business reasons, I think SuSE has managed to get things right. With SuSE, they have a much more complete, coherent feeling solution. Things just work. Their strategy to all sorts of things is far more flexible once you appreciate it. And with the Enterprise edition, they have enough partnerships in place to truly offer a comprehensive solution. In dealing with RH Enterprise offerings, it is essentially RH9 with some spit and polish. No extras, nothing you couldn't really get from any free distribution, with only RH support to differntiate it. SLES, however, includes a few niceties, such as an included, well behaved, supported JVM. Sure, you can download those for free, but it is important in such a product to have a complete solution out of the box.
Couple this to their pricing model (RH WS costs at least $179, SuSE Professional costs $79), and it seems like a much more reasonable product when compared to the likes of RH and MS.
For North America and Europe, SuSE and RedHat are virtually the only 'professional' Linux platform solutions. Others have some fantastic technical merits, but are not real professional-grade businesses for the enterprise to deal with. I love Gentoo, I like Debian, and on technical merit alone I would place both above RedHat and SuSE (as long as the user is a highly competent linux enthusiast), but the support infrastructures are simply not there in a meaningful way as far as businesses are concerned.
I agree that film is superior and will likely always have a charm all its own.... Exactly like analog vinyl records will always have their charm, but like analog vinyl records, the technology is catching up extremely fast to soon be 'good enough' for even the most discerning of viewers (note I said 'soon', implying not today, but the future). Your estimate on high resolution cameras will likely prove to be extremely pessimistic judging from the rate of everything else electronic.
As to price, that argument made absolutely zero sense. You confuse the technology with the final medium. You can get analog film developed onto really good, or really crappy photo paper, and even the best will fade with time. Sure, the negative will typically hold out *better*, but will still over time degrade. You can print out images from digital cameras onto the exact same range of photo paper pretty much, with the same results. Sure you use an inket on a sheet of normal paper and it will be crap, but you can go high end and even send images to be developed by professional labs just like film. And unlike the negatives of film, the digital picture can be preserved with guaranteed perfection, and can be duplicated at will flawlessly. Even if the media it is stored on degrades to unreadable, it was probably copied to new media before that. This scenario is simply not possible with analog strategies without infinite precision. Duplication will degrade the result without question in film.
Different formats is a repeat of number 1. There is a more obvious tradeoff between amount of view captured on camera versus quality in digital, but provided you used incredibly high resolution digital format, you can produce whatever size you want. As the fidelity of cameras increase, this issue goes away completely.
Now I agree film's nature gives it great advantage on the count of quality, but durability, that is simply impossibly tilted toward digital.
Assuming you store the media properly and choose good media, it may very well last 40 years. And if you can read the data in its entirety, then it is pristine.
Now forget about the media. CD-R media properly stored will outlive CDR drive technology, meaning people will compile CDR collections to DVD media in the near future, and repeat the process as new media technology/capacity comes out. The data is perfectly preserved across new media. Even if CDR was the end-all media that no one would go beyond, you can duplicate old CDRs to new CDRs as desired. Though the digital nature of the pictures means less detail, it also means perfectly (and verifiably) transferable. If a negative goes, any means to copy will be analog methods, and thus without infinite precision, loss of detail is inevitable. For each duplication with even the utmost detail, the picture will degrade some.
Yes, that is true, but consider that the amount of resources invested overall increases, as investments are less frequently total losses with this forewarning. Bad economically for the geographic locale predicted to quake, good in general for investment. Fewer resources lost and lower risk all comes out to healthier investments. This is all assuming that false positives are *extremely* rare and that it is also capable of predicting >90% of disasters, change either variable and the picture changes.... Of course some investments would go up (construction companies and the like would clean up on 'quake-proofing' non-movable structures).
Now, back to the geographic locale's state. Sucks to be them economically, but let's say you had the choice of having equal chance at having investments near your house, or knowing that in ~3 months, a catastrophic quake that could kill you is extremely likely. The economic problems are both temporary and offset by the value of increasing awareness to save lives. 4 months later after the quake happens, no further risk is seen and companies are already lined up to do reconstruction of whatever was destroyed. 3 months of warning allows a community to do a lot to protect investments from harm and prepare a rapid recovery plan for high-risk, high-value structures that may be destroyed. So while in the short term economic conditions are potentially bad, having 3 months warning provides better long term economic circumstances.
Seeing as how I can't find an x40, either it is a mistake, or you are using a model that is not even released.... Seems hardly fair if the latter is the case to expect FreeBSD to support the hardware reliably...
But still a bit harsh on IPv6....
/64 network, but it has yet to be seen whether certain organizations might, for the hell of it, get allocated /8 networks because they can. As near as I can tell, the high 16 bits seem to be somewhat protected, but you never know what will happen. If there is a grab for /8 networks among big players, you have the same problems that IPv4 has today.
As to the notion of never running out of address space 'never, never' as he puts it, I wouldn't be so sure. The 32-bit address space provides 4.2 billion addresses. With that in mind, we are much nearer to exhaustion than current usage would dictate. It is all about the allocation, and if sloppy allocation occurs, the 128-bit address space of IPv6 could be exhausted too. For example, the architecture of current implementations make it so that the smallest subnet anyone will likely allocate are 64-bit networks, and use MAC addresses (or something else, but still 64-bit, because it's easy), so immediately you take the address space down tremendously. Still should be well more than enough for everyone on earth to have a
As to security implications, it is true that implementations will be for the short term future less tested and therefore likely to contain critical flaws, but still IPv6 code is receiving a fair amount of testing, and critical flaws will not be quite so devastating as you may think, no more than an Apache, Linux Kernel, or MS security exposure, which we have seen all of in fairly recent history without the sky falling.... Of course the wrinkle in this is a lot of the 'home router' concepts that happen to protect common home systems will cease to provide that protection. They provide NAT features, therefore masking to an extent the system behind the device. Despite what the author says about NAT being bad because it doesn't protect against things like browser exploits and physical intruders, NAT is on the level of firewalling in terms of protection. Any reasonable network security person will realize that browser exploits, email worms, and physical intrusion must always be kept in mind, and it has nothing to do with NAT or firewalling. NAT remains effective at, for example, fending off web server and rpc attacks from unsuspecting or experimenting workstations. If NAT goes away (hopefully), people need to be mindful of good old firewalling strategies. Implementations are maturing (experimental ip6tables implementation, for example, is approaching closely the ipv4 iptables featureset). If cable/dsl 'routers' revert to hubs in a wealth of addressing, I expect either cable/dsl 'firewall' devices or increased ISP vigilance to deal with the more widespread system exposure.
All that said, I like IPv6 (my desktop, gateway, and laptops are using IPv6 and each have public IPv6 addresses, keep NAT on IPv4 on some systems), but I (and everyone else) has been waiting and watching a long long time and no encouraging migrations are yet to be seen, and I doubt the near future will bring any incentive to push such a change.
Actually, in my experience, tg3 is *extremely* broken compared to bcm5700 on many applicable cards. Negotiation of link state being a serious problem on some cards.
All check, what's your point? On my A31p the only button not working is the 'ThinkPad' button, the rest present keysyms (except the volume, which works regardless of OS support). DVD playback works about as well as anywhere else (XvMC is the only thing looming that will give additional booost), wireless ethernet is a prism chipset, power management is both apm and acpi. Sound is intel_8x0.
Not a single hardware feature on this laptop fails under linux, and working on other Thinkpads, it holds true across the board in my experience.
I suppose it is your choice, but why bother getting a Mac and skipping out on MacOSX? Their hardware is crappy, Apple cares more about looking nice than lasting more than the warranty period. Once upon a time, they made stuff that worked and stayed durable, but now they make overpriced cheap flimsy plastic and offer hostile customer support when things break. The only thing I like about Apple is that OSX is truly a fantastic blend of 'Grandma usability' and BSD power. Linux can run just as well for 99.9% of users out there on x86 hardware (if you find/make a decent x86 with the same money you shell out for a Mac).
If you couldn't tell, I bought into the hype and bought an iBook a while back and have regretted it since. Spent months trying to convince support the motherboard needed replacement due to video failures, (now well documented), and they eventuall did. The power supply started shocking my lap after 3 months, it was replaced and the replacement died just after warranty was up, and I had to shell out 80 bucks to keep using the damn thing. On top of all that, the lid latch broke and the ethernet port stopped working, so now I can only use the wireless. And each time I called about stuff under Warranty, they were more interested in trying to sell me extended warranties or insisting that Apple didn't do direct hardware warranty service , or admitting they did, but my problems were obviously because I abused/dropped the laptop or something, which never occured.
Meanwhile my Thinkpad has served me well, and it was cheaper. Even a crappy Presario laptop has outlasted that crap iBook. I buy a new system usually at least annually, and Apple has me so off to their products. It is a shame because I think the Apple software developers/testers have *really* got something right in the midsts of everything else going to crap.
You're thinking on hobbyist scale. It as free as your time for hobbyists, and for the computer saavy, that is cheap. Now for those who barely figured out Windows in the first place, not viable. For companies pricing out expensive infrastructures (trying to plan farms of servers and/or provide a common supported corporate desktop), using those resources for any sort of experimentation is completely out of the question. Even delaying deployment schedules for the sake of experimentation is highly unlikely. The managers of the people implementing will do them the 'favor' of doing the research and telling them the direction. This advertising is less about convinincing undecided parties, but reassuring people with existing Microsoft setups that the potential savings in TCO is at least uncertain and that an investment in migrating their infrastructure will not see a sufficient return on investment.
Of course common sense could make this campaign backfire horribly. For example, holding up the shining example of low-end MS run WinTel systems beating out high-end mainframe systems has to set some alarm off in even the most uneducated mind. To get the figures they wanted, they had to change the whole hardware situation to the most unbalanced comparison imaginable. Of course that would hold true, and it has absolutely nothing to do with the software running on the systems, mainframes are intended to give ultimate performance, not affordability. Why didn't they do the comparison with equal hardware? Why did they feel the need to give Linux a handicap? Was MS unable to back up their point without such a drastic measure?
perhaps more clearly written as:
$40.25 per "megabit of throughput per second".
$1.79 per "peak request per second".
So, by there reckoning, to go from a system capable of handling x requests per second to x+1 requests per second in wintel world, it is 1.79. Just to explain the terminology.
You missed his point. Let's say, hypothetically, KISS said 'kiss off, we are violationg the GPL and don't give a rat's ass.' Ok, so know they are known to be in willful violation of the GPL, and therefore lose license to redistirubte within the bounds of copyright law. The GPL is not a contract, and therefore not adhering to the GPL only means redistribution is prohibited by *copyright* law. Now take them to court because they are redistributing the content without a license. Easy case to prove, but the penalty is based on damages. How much does a free software project lose by having another company copy their stuff? Lost monetary gain is not an issue, as they already give it away for free. So yeah, the company is easily proved to violate copyright, but in doing so, the monetary payout in terms of penalty is likely to be low.
So within the bounds of the legal system, I think the GPL holds little sway. However, what companies are intimidated by is the possibility of a boycott. The market of poeple who care about the GPL and stay aware of these cases maybe relatively small, but still sizable. For an embedded device company in particular, where the main draw is the hardware and the software merely an enabler for selling the hardware, the cost of disclosing source is infinitely smaller than even 300 lost customers due to boycott. Now for software packages that offer little benefit over a lesser known open source package they based themselves on, things change, as disclosing source under GPL means that a) competing almost as good free package gets attention and they lose sales and b) people can redistribute their product under terms of GPL and they face a new competing product they helped to develop and receive no benefit. However, for these companies they would likely want to keep the whole thing quiet (don't want potential customers finding out there is an open source project out there that is almost as good as their commercial product), and make a donation in exchange for the project not complaining loudly about the violation.
Easy, they could complain that their market was so devastated by the new offering that it would no longer be profitable to compete.
What, will it now support them new-fangled horseless carriages?
I know, the joke is that X seems so out of date feature wise in terms of eye-candy cruft, but the average technology level of the 20th century isn't too impressive.
I've found that the Pallas MPI benchmark suite is a lot more interesting and informative than Linpack, just FYI.