Like the old margerine ad claiming "there is no difference" this piece reads like an ad. I would hesitate to put any stock in it for a number of reasons. Having installed and configured both I cannot see how Fedora 8 is anything but YARHR (yet another red hat release) i.e., bumping the version as "development theater" with ittle actual improvement. There's little difference bttn F8 and F7 much less any of the earlier releases.
Since installation is similar (though the Ubuntu "live" CD allows for better hardware driver validation), the first thing to compare is the gui package front-ends. Both are good but Synaptic is better than Yumex, far better in actual use. It is intuitive for those who are not Linux gurus where Yumex is not. Deb packages also tend to have fewer dependencies and there are more of them. Firefox3 for example, was available to Ubuntu users first.
Second most important item is the kernel, mainly the wireless drivers. Ubuntu wins here, particularly on laptops and older hardware. Example: adding a wep key. Click and paste in Ubuntu where it's easier to edit the poorly documented text files in Fedora.
One of my pet peeves is default security. Run 'netstat -anp' on a newly installed RH box and you'll be shocked to see how much is running and listening for network connections. Big difference from Ubuntu where you will likely see a much smaller process table and only ports 22 (ssh) and 68 (dhcp) open to the world.
Otherwise both have their high and low points. The big downer is the stuff that gets "deprecated" and made incompatible with previous release for no good reason. This is mostly GNU's fault to be sure. Sometimes I think they break stuff just to differentiate Linux from Unix. I really dislike Linux upgrades because so much breaks, far more than in a BSD, IBM, and Sun OS upgrades. Rewriting shell scripts to account for parameter differences that have no evident rational gets old after the 4th or 5th time (say "nslookup has been deprecated" three times fast, but wait, now it's been un-deprecated, ah but the output format has been changed, again...). But I digress, and am grateful to all FOSS coders, especially those who don't make work difficult for those of us who install, upgrade, and manage their systems.
Not really sure why RedHat is allowing its distribution to fall so far behind. I suppose they're fat and happy to get paid for RHEL support, RHEL bugfixes, and RHEL repos. Like SCO before them, IMO, it's a short-term business model that won't hold up to Debian's community process much longer.
MYTH: MSFT and Intel constitute the evil Wintel cartel Some posts are such obvious astroturfing that they make you wish for downmod points for the advertising. Note the flaimbait reference to "evil cartel". Follow the link to a "Wikipedia article", almost certainly penned by fellow astroturfers. See in particular the citation regarding Intel belonging to the OLPC project since summer. Of course there's no mention of what Intel did or gave to the project over the past "2 months" (hint: virtually nothing).
These are the same reasons DELL sells only Intel and until just recently only Microsoft: pure anti-trust tactics. Wholly illegal in the US, or at least they would be if it wasn't for a Regan/Bush-appointed judiciary. Sad to say, as an American, but the world's center of open and progressive government has shifted to the EU thanks to the sleazy business practices of Intel, Microsoft, Enron, Haliburton, and the right wing of the Republican party.
Maybe if a lot of people hadn't pushed so hard to "internationalize" DNS management and just left well enough alone, we'd not be having these problems. Are you forgetting the US Corporation that took over Internic registration and policy (from SRI)? It was worse than ICANN, charging $100 per domain per year, hanging on to expired domains and auctioning them for a profit, not letting domain holders move to another registrar weeks before a domains was to expire... That corporation was Network Solutions. Yes, Verisign did purchase and spin-off NS, but it retained their core policies and continues to leverage its authority over registrars, fee-wise and otherwise, in ways that are far from democratic.
As bad as ICANN is I cannot help but think how many thousands of dollars and pre-expired domains I would have lost had ICANN not stepped in. Perhaps the devil you know IS better, but they re still not protecting the public interest.
Under Cerf ICANN has degenerated into a politically paralyzed organization unable to pass ANY legislation. The biggest downside to this so far is the Storm worm. Storm only exists because "Domain Tasters" can register a domain for free and use it until the Tasting period expires. ICANN could fix this policy in a single meeting.
Domain Tasting works particularly well in combination with "fast flush" DNS. FF violates RFCs and is a particularly nefarious way of covering one's tracks, but registrars are helpless to do anything about it without an ICANN policy. Again, ICANN could fix this policy issue in a single meeting.
Without Tasting and FF there would be no Storm worm, and Storm only exists due to ICANN's dysfunctional board of directors. If Dengate-Thrush cannot fix ICANN is suspect Storm will force the US government to take over ICANN, probably by executive decree, and we all know how well the US government manages things (USPS, TSA, Amtrak,...)
Call me a troll but it seems Theo is trying to nitpick on grammar to force changes back into the OpenBSD kernel. Contradicting the "BSD feel". I don't know. It seems what Theo is doing is simply pointing out how GPL authors are taking BSD code and not giving back. No different really, than the proprietary takers (MS) they often criticize.
BSD is like a doting parent who loves their child unconditionally. GPL is like a parent who only loves their child as long as their child loves them back. The family stays together, not for the kids, but only because of the love of the BSD parent.
You can't quantify SA productivity.
I respectfully disagree. I must disagree as well, but with a twist. In my experience the only people who can evaluate SA productivity are other SA's. It is perfectly analogous to evaluating programmers. You can't do it by lines of code, you can't even do it by uptime (which can be out of the engineer's hands), but given a little time you can do it in comparison to others.
That's not to say there are no metrics, in fact there are quite a few that aren't being measured. Just that the metrics most worth measuring are relative. Sure you have to start with the SA/coder's value to the organization using all the standard tools: cost of downtime, benefit to employees, etc. But the bottom line is relative cost. We all know that a bad coder can write enough dead-end code to cost a company several times his or her salary in recoding. The same is true of SA's.
Spam is a good example. A good email SA knows how to maintain spamassassin, bayes, clam-av, etc and save everyone from losing mail to false-positives, losing functionality to spam volume, losing data and systems to viruses and trojans, losing disk space to worthless spam folders, and losing important communications. Most email admins, however, know only MS Exchange, which costs big bucks, and whose spam and virus filters also cost big bucks, far more than their admin's salaries in most cases, even though Exchange and family doesn't perform half as well as the open source alternatives.
The bottom line: there is a lot of ROI SAs and coders add to the organization but we've never done a good job of measuring it. I mean just try and find papers on this by searching Google or Yahoo, good luck...
No question Google's home page beats the pants off Yahoo's. Have you ever tried Yahoo without cookies? ("unsupported browser..." cruft)
It's difficult to see how Google and Yahoo compare other than by their search and email services. As far as search goes Google's multi-lingual results bug me to no end. You can't select a language without enabling cookies. Problem with cookies is they facilitate tracking. Problem with tracking is it biases search results by basing them on your previous searches. That and increasingly advertiser-slanted results keep me a fan of scroogle.org over either Yahoo or Google. Deja Vu Alta Vista.
On the email side Google's spam filters seem to be better, but their javascript is worse, sometimes much worse, particularly over slow connections. I doubt that many Google engineers even remember sub-broadband connectivity, much less test their applications under it. That's probably a good part of the reason why Yahoo is more popular outside of the US. Well, that and privacy laws.
In the long run I think Yahoo has the better chance of coming out on top. Google has some great ideas but nothing that cannot be copied. Yahoo, OTOH, isn't so much about ideas as execution. They can and probably will take whatever Google does and put a spit-shine on it. Well, whatever excepting perhaps search. We're still waiting to see what Yahoo can do with search.
ping -s www.yahoo.com Is there a reason you omitted the ping statistics line? My own ping didn't jive. Not surprising if you know what Akamai is, and why any site using it will never see slow ping times.
The 80-column limit comes from the size of an IBM punched card. Urban myth perhaps? There's no relation between screen widths and punch cards.
80 character lines are derived from centuries of ergonomic experiments to find the most easily read line. Books used it long before there were computers. Newspapers use even shorter lines to fit more on a page and facilitate page scanning.
Not sure how much these ergonomics apply to code though. I will say that I tend to use 2 space indentation and still find long lines unreadable.
My vote: study history and do the research i.e, do your homework before proposing such a change. The old adage "if it ain't broke don't fix it" applies to computers the same as anything else. Wish we could convince some of our Linux coders of this before they change the flags to application XXX yet again...
Given the text of the article, and section headers like "The Desperate Panic of the Apple Haters", you can't really take it seriously as an analysis. As with most advocacy pieces there's really nothing in-depth to warrant the read.
So I'm remaining skeptical until long term sales figures are available. From the spec to-date it doesn't appear the iPhone has any features not found on other phones for some time now. Will Apple implement in a way that makes up for the lack of new technology, as it did with the iPod?
One thing I do see as a drawback is the touchpad. Cell phones, Blackberrys, and keyboards have keys for a reason: speed and accuracy of typing. No tactile feedback on closely spaced keys is likely to make the iPhone less user-friendly than you'd expect, if cheaper to manufacture.
I've never met a single developer who even knew or cared about XSS I'm one who does care, but other than SSL there really is little that can be done about XSS. Sure you can put ever more layers of validation in the client, all of which can be worked-around with various levels of difficulty. That's the wrong tool for this job. The right tool is SSL and session tracking.
Second problem is you cannot build security on a foundation of shifting sand. Witness the security holes introduced in Firefox2 that didn't exist in Firefox1, or Microsoft's insistence on using activex after all these years of known vulnerabilities. The worst offender, however, is javascript. No Javascript implementation is without vulnerabilities.
ARS must have rushed the fact checking to get this article out. Truth is that ARIN does not, and has never, made a best effort at anything except to charge ISPs for address space and let them reap a 500 to 1000% profit reselling it. ARIN has done nothing substantive to promote IPv6, and ARIN looks the other way at hundreds of existing, unused, large IPv4 network allocations.
I've worked at Silicon Valley companies with multiple class B allocations that could have easily put them behind NAT gateways and firewalls. The University of California campuses have many class Bs and will tell you they "can't do NAT to the dormitories because it's too difficult to track". That's 65K address per class B and there are dozens of these, and several class As, that are just waiting to be reclaimed.
What these class A and B-owning organizations are doing is holding on to vacant land as long as they can, until it becomes valuable, at which point they hope to sell it at a big profit.
ARIN is doing the same thing by failing to reclaim these allocations. They're just waiting for demand to climb like California real-estate to begin cashing-in. This is exactly what Network Solutions/Verisign did with domain names when they had a government-protected monopoly. Have we forgotten so soon, one year domain registration was free (via SRI), and the mext year it was $100 per year per domain (via Verisign), despite actual costs of $7/year. This scenario should also be familiar to those who have had to change telephone area codes, sometimes more than once, until enough people complained (of course that was when the FCC was in Democratic hands. With Republicans the Telcos have once-again been cleaning up).
So believe the hype, but remember, if you fail to look a little deeper we will soon be paying the price, in increased ISP fees, for this wholly artificial IPv4 address shortage.
Is the JVM really less "heavy or overkill" than Flash? If you have a reasonably up-to-date system there is no difference in the end-user experience.
More importantly, Java does run on a 64-bit Linux desktop. Flash and a number of other proprietary so-called standard products (by Adobe, Real, etc) do not.
Security - Javascript is NOT designed to secure a web app
If this isn't the understatement of the year I'll eat my webapp.
security needs to happen on the server side, out of necessity
This implies that spam filtering also needs to happen on the server side... which will also defies logic.
I've programmed in dozens of languages, and few are as flexible and enjoyable as Javascript
Google astroturfing aside, all you can truthfully say in favor of javascript is that
it is better than the language used by sendmail.cf. By better I mean easier to read, debug, and test.
The only reason anyone in their right mind would use javascript today is because it is the only option.
Personally, I'm hoping JavaFX has something more like Python's syntax i.e, easily understood by someone
who does not owe their employment to the understanding of it.
Seems like a bit premature to label most of these "failed". We just need to separate concepts from implementation, that and keep in ExtremeTech's need to sell papers.
The BTX Form Factor
Home systems need to be quieter more than cooler. Colo servers are
better off with larger/fewer drives and more energy efficient cpus.
High-Definition Video on the PC
PCs are replacing TV in this respect. People want better content more
than prettier pictures. Then too there's the problem with bandwidth
and storage capacities. This is a technology whose time will come when
the supporting infrastructure catches up.
High Fidelity Digital Audio
Demand is there but cheap storage is not
Windows XP Professional x64 Edition
Microsoft FUD, just like IPnP.
High Definition Optical Drives
WTF? CDs are dead.
Copy Protection for Music CDs
RIAA/MPAA/Sony FUD.
iPod Competitors Emerge
As the owner of a Creative MP3 player it seems to me that iPod
competitors have always had better technology but need to match Apple's
marketing.
The Digital Home
Haven't you been eating your X10 spam? Seriously though, this is as
promising a field as it gets, especially regarding energy efficiency.
Only reason there's little development is the fear of Microsoft's
monopoly. Thanks due utterly clueless Republican economic policies.
Google's Gmail Service
Privacy issues, spam problems, nothing really new about this service.
Dual Graphics Cards
Preempted by workspace ergonomics, cubicle size, and multiple virtual
desktops.
I know too many kids who sit at their computers to avoid going outside and playing with other kids. If this is your reason for limiting their time at the keyboard then it is a good thing, though 30 minutes is unrealistic. I wouldn't be too concerned until the 3rd hour or so.
It fundamentally depends on what they would be doing without the PC. If they would be watching TV then by all means let 'em type. PCs make kids smarter whereas TVs only make people lazy and stupid. Both PCs and TVs, however, contribute to the skyrocketing rate of childhood obesity and that's a _very_ bad thing.
As long as they're walking or cycling to school, playing sports afterwards, and not overweight then I don't think you really want to limit their computer time.
Why a politically motivated person like Murdoch would want Skype has nothing to do with it's ability to turn a profit. It is one and the same reason Google created gmail, AOL, Yahoo and MS their email and IM services, in a word: data-mining (ok, two words).
When you sign-up for any of these servers you'll note there is no guarantee that anything you type or say will be kept private or secret. On the contrary everything must be sent through the provider's servers where they can parse it at will. From stock tips to source code, these services are a gold mine for the providers.
I though soft updates made journaling unneeded and everything slower?
Not even close. Soft Updates, background writes, and background fsck do not protect against data loss and corruption like journalling. The only thing you ever need to worry about on journalling filesystems is a hardware error, and that mainly on non-SCSI drives without RAID.
Also, journalling tends to speed up more types of writes than it slows. On busy filesystems it is almost universally faster.
A journalling filesystem should bring FreeBSD to about the level of performance and reliablity that Solaris reached 7 years ago. It's about time!
An opinion piece is something that lists the name of her mother (not PJ) and also gives a street address along with pictures of the outside of where she lives?
On top of that he says he found "nothing unethical" about the article. How could you ever trust a publication with an editor like that?
I will be checking my all of magazines for any reference to "Fuat Kircaali" or "Sys-Con Media" and not purchasing anything of the sort.
> Asterisk's primary appeal is that it integrates POTS
From what I've used of Asterisk, past 2 years, POTS is its main drawback. From buggy Digium cards to poor QA. 'cvs checkout -r HEAD' seems to be the standard tech support. Thanks but no thanks, especially when 'cvs checkout' broke it in the first place and the last 4 HEAD checkouts didn't fix it. Asterisk and POTS is NOT by any measure production quality.
Wrong. -STABLE and -RELEASE are two different things
I think you missed my point. Obviously -STABLE is not a release. But going by FreeBSD's terminology it's not a release candidate either. The category before RC is BETA. By that common definition -STABLE is no different from BETA whereas -CURRENT is ALPHA.
My main question was how do you have 3 RCs without the first 2 not being RCs at all?
Slashdot Mirror
Like the old margerine ad claiming "there is no difference" this piece reads like an ad. I would hesitate to put any stock in it for a number of reasons. Having installed and configured both I cannot see how Fedora 8 is anything but YARHR (yet another red hat release) i.e., bumping the version as "development theater" with ittle actual improvement. There's little difference bttn F8 and F7 much less any of the earlier releases.
Since installation is similar (though the Ubuntu "live" CD allows for better hardware driver validation), the first thing to compare is the gui package front-ends. Both are good but Synaptic is better than Yumex, far better in actual use. It is intuitive for those who are not Linux gurus where Yumex is not. Deb packages also tend to have fewer dependencies and there are more of them. Firefox3 for example, was available to Ubuntu users first.
Second most important item is the kernel, mainly the wireless drivers. Ubuntu wins here, particularly on laptops and older hardware. Example: adding a wep key. Click and paste in Ubuntu where it's easier to edit the poorly documented text files in Fedora.
One of my pet peeves is default security. Run 'netstat -anp' on a newly installed RH box and you'll be shocked to see how much is running and listening for network connections. Big difference from Ubuntu where you will likely see a much smaller process table and only ports 22 (ssh) and 68 (dhcp) open to the world.
Otherwise both have their high and low points. The big downer is the stuff that gets "deprecated" and made incompatible with previous release for no good reason. This is mostly GNU's fault to be sure. Sometimes I think they break stuff just to differentiate Linux from Unix. I really dislike Linux upgrades because so much breaks, far more than in a BSD, IBM, and Sun OS upgrades. Rewriting shell scripts to account for parameter differences that have no evident rational gets old after the 4th or 5th time (say "nslookup has been deprecated" three times fast, but wait, now it's been un-deprecated, ah but the output format has been changed, again...). But I digress, and am grateful to all FOSS coders, especially those who don't make work difficult for those of us who install, upgrade, and manage their systems.
Not really sure why RedHat is allowing its distribution to fall so far behind. I suppose they're fat and happy to get paid for RHEL support, RHEL bugfixes, and RHEL repos. Like SCO before them, IMO, it's a short-term business model that won't hold up to Debian's community process much longer.
These are the same reasons DELL sells only Intel and until just recently only Microsoft: pure anti-trust tactics. Wholly illegal in the US, or at least they would be if it wasn't for a Regan/Bush-appointed judiciary. Sad to say, as an American, but the world's center of open and progressive government has shifted to the EU thanks to the sleazy business practices of Intel, Microsoft, Enron, Haliburton, and the right wing of the Republican party.
As bad as ICANN is I cannot help but think how many thousands of dollars and pre-expired domains I would have lost had ICANN not stepped in. Perhaps the devil you know IS better, but they re still not protecting the public interest.
Under Cerf ICANN has degenerated into a politically paralyzed organization unable to pass ANY legislation. The biggest downside to this so far is the Storm worm. Storm only exists because "Domain Tasters" can register a domain for free and use it until the Tasting period expires. ICANN could fix this policy in a single meeting.
...)
Domain Tasting works particularly well in combination with "fast flush" DNS. FF violates RFCs and is a particularly nefarious way of covering one's tracks, but registrars are helpless to do anything about it without an ICANN policy. Again, ICANN could fix this policy issue in a single meeting.
Without Tasting and FF there would be no Storm worm, and Storm only exists due to ICANN's dysfunctional board of directors. If Dengate-Thrush cannot fix ICANN is suspect Storm will force the US government to take over ICANN, probably by executive decree, and we all know how well the US government manages things (USPS, TSA, Amtrak,
No different really, than the proprietary takers (MS) they often criticize.
Interesting that this story would get approved by Slashdot given that Opera is closed source and they don't release a distribution for 64bit Linux.
BSD is like a doting parent who loves their child unconditionally. GPL is like a parent who only loves their child as long as their child loves them back. The family stays together, not for the kids, but only because of the love of the BSD parent.
In a nutshell
That's not to say there are no metrics, in fact there are quite a few that aren't being measured. Just that the metrics most worth measuring are relative. Sure you have to start with the SA/coder's value to the organization using all the standard tools: cost of downtime, benefit to employees, etc. But the bottom line is relative cost. We all know that a bad coder can write enough dead-end code to cost a company several times his or her salary in recoding. The same is true of SA's.
Spam is a good example. A good email SA knows how to maintain spamassassin, bayes, clam-av, etc and save everyone from losing mail to false-positives, losing functionality to spam volume, losing data and systems to viruses and trojans, losing disk space to worthless spam folders, and losing important communications. Most email admins, however, know only MS Exchange, which costs big bucks, and whose spam and virus filters also cost big bucks, far more than their admin's salaries in most cases, even though Exchange and family doesn't perform half as well as the open source alternatives.
The bottom line: there is a lot of ROI SAs and coders add to the organization but we've never done a good job of measuring it. I mean just try and find papers on this by searching Google or Yahoo, good luck...
Metric man oh where can you be.
No question Google's home page beats the pants off Yahoo's. Have you ever tried Yahoo without cookies? ("unsupported browser..." cruft)
It's difficult to see how Google and Yahoo compare other than by their search and email services. As far as search goes Google's multi-lingual results bug me to no end. You can't select a language without enabling cookies. Problem with cookies is they facilitate tracking. Problem with tracking is it biases search results by basing them on your previous searches. That and increasingly advertiser-slanted results keep me a fan of scroogle.org over either Yahoo or Google. Deja Vu Alta Vista.
On the email side Google's spam filters seem to be better, but their javascript is worse, sometimes much worse, particularly over slow connections. I doubt that many Google engineers even remember sub-broadband connectivity, much less test their applications under it. That's probably a good part of the reason why Yahoo is more popular outside of the US. Well, that and privacy laws.
In the long run I think Yahoo has the better chance of coming out on top. Google has some great ideas but nothing that cannot be copied. Yahoo, OTOH, isn't so much about ideas as execution. They can and probably will take whatever Google does and put a spit-shine on it. Well, whatever excepting perhaps search. We're still waiting to see what Yahoo can do with search.
ping www.yahoo.com
--- www.yahoo-ht3.akadns.net ping statistics ---
11 packets transmitted, 11 packets received, 0% packet loss
round-trip min/avg/max/stddev = 11.643/12.185/12.660/0.278 ms
ping www.google.com
--- www.l.google.com ping statistics ---
11 packets transmitted, 8 packets received, 0% packet loss
round-trip min/avg/max/stddev = 14.059/14.581/15.155/0.360 ms
80 character lines are derived from centuries of ergonomic experiments to find
the most easily read line. Books used it long before there were computers.
Newspapers use even shorter lines to fit more on a page and facilitate page
scanning.
Not sure how much these ergonomics apply to code though. I will say that I
tend to use 2 space indentation and still find long lines unreadable.
My vote: study history and do the research i.e, do your homework before
proposing such a change. The old adage "if it ain't broke don't fix it" applies
to computers the same as anything else. Wish we could convince some of our
Linux coders of this before they change the flags to application XXX yet again...
Given the text of the article, and section headers like "The Desperate Panic of the Apple Haters", you can't really take it seriously as an analysis. As with most advocacy pieces there's really nothing in-depth to warrant the read.
So I'm remaining skeptical until long term sales figures are available. From the spec to-date it doesn't appear the iPhone has any features not found on other phones for some time now. Will Apple implement in a way that makes up for the lack of new technology, as it did with the iPod?
One thing I do see as a drawback is the touchpad. Cell phones, Blackberrys, and keyboards have keys for a reason: speed and accuracy of typing. No tactile feedback on closely spaced keys is likely to make the iPhone less user-friendly than you'd expect, if cheaper to manufacture.
Second problem is you cannot build security on a foundation of shifting sand. Witness the security holes introduced in Firefox2 that didn't exist in Firefox1, or Microsoft's insistence on using activex after all these years of known vulnerabilities. The worst offender, however, is javascript. No Javascript implementation is without vulnerabilities.
ARS must have rushed the fact checking to get this article out. Truth is that ARIN does not, and has never, made a best effort at anything except to charge ISPs for address space and let them reap a 500 to 1000% profit reselling it. ARIN has done nothing substantive to promote IPv6, and ARIN looks the other way at hundreds of existing, unused, large IPv4 network allocations.
I've worked at Silicon Valley companies with multiple class B allocations that could have easily put them behind NAT gateways and firewalls. The University of California campuses have many class Bs and will tell you they "can't do NAT to the dormitories because it's too difficult to track". That's 65K address per class B and there are dozens of these, and several class As, that are just waiting to be reclaimed.
What these class A and B-owning organizations are doing is holding on to vacant land as long as they can, until it becomes valuable, at which point they hope to sell it at a big profit.
ARIN is doing the same thing by failing to reclaim these allocations. They're just waiting for demand to climb like California real-estate to begin cashing-in. This is exactly what Network Solutions/Verisign did with domain names when they had a government-protected monopoly. Have we forgotten so soon, one year domain registration was free (via SRI), and the mext year it was $100 per year per domain (via Verisign), despite actual costs of $7/year. This scenario should also be familiar to those who have had to change telephone area codes, sometimes more than once, until enough people complained (of course that was when the FCC was in Democratic hands. With Republicans the Telcos have once-again been cleaning up).
So believe the hype, but remember, if you fail to look a little deeper we will soon be paying the price, in increased ISP fees, for this wholly artificial IPv4 address shortage.
More importantly, Java does run on a 64-bit Linux desktop. Flash and a number of
other proprietary so-called standard products (by Adobe, Real, etc) do not.
If this isn't the understatement of the year I'll eat my webapp.
security needs to happen on the server side, out of necessity
This implies that spam filtering also needs to happen on the server side... which will also defies logic.
I've programmed in dozens of languages, and few are as flexible and enjoyable as Javascript
Google astroturfing aside, all you can truthfully say in favor of javascript is that it is better than the language used by sendmail.cf. By better I mean easier to read, debug, and test.
The only reason anyone in their right mind would use javascript today is because it is the only option. Personally, I'm hoping JavaFX has something more like Python's syntax i.e, easily understood by someone who does not owe their employment to the understanding of it.
to separate concepts from implementation, that and keep in ExtremeTech's
need to sell papers.
Home systems need to be quieter more than cooler. Colo servers are
better off with larger/fewer drives and more energy efficient cpus.
PCs are replacing TV in this respect. People want better content more
than prettier pictures. Then too there's the problem with bandwidth
and storage capacities. This is a technology whose time will come when
the supporting infrastructure catches up.
Demand is there but cheap storage is not
Microsoft FUD, just like IPnP.
WTF? CDs are dead.
RIAA/MPAA/Sony FUD.
As the owner of a Creative MP3 player it seems to me that iPod
competitors have always had better technology but need to match Apple's
marketing.
Haven't you been eating your X10 spam? Seriously though, this is as
promising a field as it gets, especially regarding energy efficiency.
Only reason there's little development is the fear of Microsoft's
monopoly. Thanks due utterly clueless Republican economic policies.
Privacy issues, spam problems, nothing really new about this service.
Preempted by workspace ergonomics, cubicle size, and multiple virtual
desktops.
I know too many kids who sit at their computers to avoid going outside and playing with other kids. If this is your reason for limiting their time at the keyboard then it is a good thing, though 30 minutes is unrealistic. I wouldn't be too concerned until the 3rd hour or so.
It fundamentally depends on what they would be doing without the PC. If they would be watching TV then by all means let 'em type. PCs make kids smarter whereas TVs only make people lazy and stupid. Both PCs and TVs, however, contribute to the skyrocketing rate of childhood obesity and that's a _very_ bad thing.
As long as they're walking or cycling to school, playing sports afterwards, and not overweight then I don't think you really want to limit their computer time.
Why a politically motivated person like Murdoch would want Skype has nothing to do with it's ability to turn a profit. It is one and the same reason Google created gmail, AOL, Yahoo and MS their email and IM services, in a word: data-mining (ok, two words).
When you sign-up for any of these servers you'll note there is no guarantee that anything you type or say will be kept private or secret. On the contrary everything must be sent through the provider's servers where they can parse it at will. From stock tips to source code, these services are a gold mine for the providers.
r7
I though soft updates made journaling unneeded and everything slower?
Not even close. Soft Updates, background writes, and background fsck do not protect against data loss and corruption like journalling. The only thing you ever need to worry about on journalling filesystems is a hardware error, and that mainly on non-SCSI drives without RAID.
Also, journalling tends to speed up more types of writes than it slows. On busy filesystems it is almost universally faster.
A journalling filesystem should bring FreeBSD to about the level of performance and reliablity that Solaris reached 7 years ago. It's about time!
r7
An opinion piece is something that lists the name of her mother (not PJ) and also gives a street address along with pictures of the outside of where she lives?
On top of that he says he found "nothing unethical" about the article. How could you ever trust a publication with an editor like that?
I will be checking my all of magazines for any reference to "Fuat Kircaali" or "Sys-Con Media" and not purchasing anything of the sort.
Wozniak spoke extensively about this period at Apple in a great interview at Gnomedex. It's available in high quality streaming audio on ITConversations: http://www.itconversations.com/shows/detail214.htm l
> Asterisk's primary appeal is that it integrates POTS
From what I've used of Asterisk, past 2 years, POTS is its main drawback. From buggy Digium cards to poor QA. 'cvs checkout -r HEAD' seems to be the standard tech support. Thanks but no thanks, especially when 'cvs checkout' broke it in the first place and the last 4 HEAD checkouts didn't fix it. Asterisk and POTS is NOT by any measure production quality.
r7
Wrong. -STABLE and -RELEASE are two different things
I think you missed my point. Obviously -STABLE is not a release. But going by FreeBSD's terminology it's not a release candidate either. The category before RC is BETA. By that common definition -STABLE is no different from BETA whereas -CURRENT is ALPHA.
My main question was how do you have 3 RCs without the first 2 not being RCs at all?
Really looks like a bad case of NIH to me.
R7
Can someone clarify FreeBSD's terminology? I thought a release candidate was different from a beta (known in FreeBSD-speak as a -STABLE).