Thank goodness for John Levin and Meredith Baker. Voices of sanity above the din. I don't know what is going on at ICANN but it clearly has been disabled by special interests, groupthink, and dispersion of responsibility. Their "any tld is fine with us" plan (originally proposed by France's Internic) shows such a profound lack of concern for the consequences that it's clear the bulk of their membership is simply not technically qualified.
The downside this all illustrates, beyond any doubt, is that ICANN does not and can not work in its present format. It needs to be reconstituted to insure that all members have no conflicts of interest and sufficient experience and expertise with technical and security issues. I hope it can retain the non-profit status and multi-country membership, without being so inclusive (of small countries) that it cannot avoid being corrupted as ISO was when Microsoft bought the ISO's endorsement for OOXML, or ICANN itself was when Verisign did the same to win the exclusive contract for.com.
It appears within the system administration community that Ruby is well on its way to replacing Perl
Interesting observation but not what I'm seeing here in Silicon Valley. Perl is far less popular than in years past, but Ruby is not that popular among systems admins either. Instead Python is all the rage. I believe this is because Ruby is better suited to those who spend a majority of their time writing code whereas Python can be learned and used more easily without such frequent use. So sysadmins here, who don't spend the majority of their time writing code, are using mainly Python (and shell).
Not a statistic, just an observation across a dozen or so large and small Linux and Unix using organizations.
Of course Google "believes" personalized search is the future. It's in their advertising interest to learn as much about you as they can. That's where their revenue comes from. You can't really expect them to say anything that would negatively impact revenue.
I have not found that personalized search yields good quality results. That may be because most of my searches are about stuff I don't know rather than stuff I already know. More important is the choice of personalized or non-personalized search. What Google isn't talking about, and hasn't given consumers the option of to-date, is any choice in this matter (bean-counter mandate to be sure, "do no evil" marketing aside).
One thing the authors of this study seem to have forgotten is that there could still be other causes of Alzheimer's, and the HSV infection is simply opportunistic, as it is to other types of neural injury.
Nice try but pure fabrication. NAT, aka private address space, is not going away. Telcos/ILECs blocked NAT when IPv6 was being developed and have since then spent a lot on marketing IPv6 without NAT/rfc1918 as a solution too all our problems. In so doing they have delayed the adoption of IPv6 by many years. How much longer will their transparent opposition to IPv6 NAT delay the inevitable? That is the question. No, we are not going to assign public IP addresses to every network-enabled computer and other device. And no, we are not going to implement IPv6 until NAT is fully supported. This is the reality that those who claim, falsely, that NAT is not a solution, are trying to ignore.
Sadly, due to telco/ILEC influence there is not likely to be a single IPv6 NAT implement for several years. When it does happen, and it will, there is likely to have already been multiple IPv6 NAT implementations which network programmers will have a hard time reconciling. The problem is vendor lock-in, which astroturfing ILECs cannot achieve without blocking NAT, and in the process 'owning' all of your IP-enabled devices.
We would all be better off if A) youtube and other Flash sites made their content available in MP4 and other ISO standardized formats
They don't do that because they'd have to pay licensing fees.
What are you talking about. There are no licensing fees for MP4 or any of the ISO standardized codecs.
Youtube continues to offer Flash and no other formats because Adobe made them a deal they couldn't refuse. Other streaming video sites offer standardized, open formats, without licensing fees because Adobe hasn't (yet) made them similar offers. No different from MS buying AV vendors and canceling their Mac and Linux products, or IBM buying Lotus and canceling their SunOS products.
Linux users asked, and adobe listened. Great stuff.
Would be, except that we've been asking for several _years_ now. Isn't this kind of crap (releasing software for some OS and some architectures while ignoring others) why we need open source in the first place?
Looking forward, what assurance do we have that security issues will be addressed any better, or upgrades, or new features? We have none of course.
We would all be better off if A) youtube and other Flash sites made their content available in MP4 and other ISO standardized formats, and B) if Adobe published the Flash spec so others could develop better writers and viewers.
Yes we know that Adobe has to make a profit on their product to fund further development, but they don't have to do so using the same tie-in and lock-in that got MS sued for anti-trust.
when i was in school, warm-ups were there to avoid injuries
I think the big misconception here is that stretching is some kind of warm-up. It is not. Only exercise that elevates your heart-rate can have a warm-up effect.
So, warming-up _before_ stretching will improve performance and reduce injury but stretching while cold will only reduce performance and increase the risk of injury, all else being equal.
This is to ensure that the code that you freed... remains free.
This is not technically correct, the original code remains free regardless. What the GPL does is force you to release the code that you wrote on top of the original code. It's patently not about "remaining free". It is about "making free" i.e., imposing restrictions on what you can do with your modified code. This is a point that GPL proponents try to obfuscate, until a lawsuit is filed.
Bottom line is that the GPL has been the most successful license. I particularly appreciate the fact that Microsoft cannot simply take it for use in it's Windows illegal, anti-trust, locked-in code-base. What I cannot appreciate are the GPL proponents who freely take from BSD sources and give nothing back. Remember, 50% of Linux came from BSD. 0% of BSD came from Linux. How does that really differ from Microsoft's tactics?
The reason it was blocked was that it came from an IP that was current blacklisted for spamming and was clearly a dynamic IP, not that spamassassin recognized the message. Any mail from that IP would have been blocked. Spamassassin actually fell down pretty badly on the content analysis.
Partially correct, but you're forgetting that headers _are_ content as much as the body, and any properly configured Spamassassin takes full advantage of RBLs, RHSBLs, and CBLs to identify spam (as much as any other signature). On this (well configured) server anything above 6.0 is discarded, yielding no false positives and rare false negatives (~2 per week per account). Sure it would have scored higher if it had better analyzed the hrefs, but the point is that it recognized the messages as spam.
Make that "Companies doing business on the web without basic spam filters in place". Our mailservers all run Spamassassin which easily recognized and tagged these as spam: score=8.449 tests=[BAYES_50=0.001, DNS_FROM_OPENWHOIS=1.13, HELO_DYNAMIC_DHCP=1.398, HTML_MESSAGE=0.001, RCVD_IN_BL_SPAMCOP_NET=1.2, RCVD_IN_PBL=0.905, RCVD_IN_XBL=3.033, RDNS_NONE=0.1, SARE_MONEYTERMS=0.681]. Companies that can't even manage to implement basic spam filters are at a competitve disadvantage. Those that curtail their email correspondence are, well, a good oppotunity for short sellers.
Perhaps more revealing is how many Slashdot posters are missing the real point of this story. It's not phishing, which is old news, but the security flaws of a proprietary and closed source application. There's no way Adobe can secure Flash without taking it to open source and getting the resulting peer review. That's information security 101. Heck, Adobe hasn't even figured out how to release a version of Flash for 64bit Linux, which is a heck of a lot easier than security code audits. Appears that, despite claims to the contrary, Linux users care more about free as in wallets than free as in source.
Using SSL for trust based on the word of companies like Verisign is pointless - you have to do manual authentication
I wouldn't say it is pointless but it is not a panacea. Let's not forget that the sometimes dysfunctional methods browsers use to handle SSL certificates are an artifact not of security but of monetization. We can thank Verisign and Netscape for this, as they designed it that way back when Netscape granted Verisign an exclusive on browser certificates. Verislime milked this monopoly for all it was worth (as they did with domain registration about the same time). This is is one of the features IE improved on, and one of the reasons it eventually put Netscape out of business.
The sad part of all this is that Verislime still charges big bucks for worthless "premium" certs (at a 99% markup) and people are still foolish enough to purchase them. Worse is when open source browsers support this kind of "security theater" even though it only benefits a small number of providers and their unethical business practices.
Both Google and Yahoo (not sure about Hotmail) use global spam filters that everyone trains
This is a recipe for spam. You can't trust end-users with this. You also can't trust most email administrators to take care when marking things as spam/not spam. About all you can do is manually check (some of) the messages (some) users flag for inclusion to global Bayesian / pattern recognition filtering.
ISPs and email appliances that give end-users any control over their own filtering are shooting themselves in the foot. Spam filtering is a full time job and if you want to do it right you have to spend at least a couple of hours a day on it. 999 out of 1000 untrained users/admins get it wrong, accounting for the spam "statistics" published by Gartner and others. In reality, well managed email services have virtually no false positives and at most a few spam a week, while blocking or discarding the 95-99% of email messages that are spam (at the statistical average business or ISP).
The issue at large ISPs like gmail/yahoo is ROI. As long as users are willing to put up with spam there's no point in spending much money on effective tools and methods to identify and filter it.
Can't believe people take PC Magazine seriously any more. I mean really, when you have to resort to calling Usenet dead to sell papers...
This guy probably thinks usenet's dead because Google archiving of it has gone from bad to worse. Either that or it's dead because his ISP stopped maintaining a news server. But usenet is not dead, in fact it's working just fine as the tens of thousands of groups that just came up in tin from on my local ISP (via supernews) prove.
But then usenet never was about being popular, or archiving, or graphics, or forums, and it doesn't lend itself to making money. The many follow-on attempts to replicate usenet fall short of what still works great and is still the easiest way to get _content_ without the fluff or marketing.
No, usenet is not dead, not by a long shot, though there will always be those who will say it is to make a buck. There will probably also continue to be those who want us to think usenet is dead in order to sell their own profitable version of usenet, but follow the money and you'll find only wanna-be television content, as in most profit driven media, and lots of ads. Bottom line: Usenet will not advertise in PC magazine, and usenet will not die.
no one will get the patch because distros never package patches
That's one of the problems with binary distributions. FreeBSD sysadmins, OTOH, have only to cvsup or, if you don't even want to wait for that, edit the bind94/Makefile, change ISCVERSION from "9.4.2" to "9.4.3b2", run make makesum && make install clean, wait 3 minutes, and restart named.
But this bug also reflects poorly on ICS's Bind 9 development process. They could have reused bind v8 code which had already fixed these sequence randomization issues. Sadly, in their hast to reissue a "chock-full-o-features" bind v9 security took a back seat. Believe me this is not the only potential vulnerability in bind 9.
Actually the GPL was about building and making available better code. At one time that's what the GPL was about perhaps. Unfortunately, it now seems to be also about taking financial incentives out of the equation. The article makes the intent of the GPL clear in that respect.
This Slashdot thread, OTOH, should say a lot more to developers than the original article, or at least it should to those of us without the time or capital to open-source everything by putting a big red flag around GPL anything.
Whether that will, in the long term, increase or decrease the use of the GPL is too early to say. It does, however, indicate a bright future for software developed with a BSD license.
It means you can't even afford to run your own mail server or have someone do it for you. Which is probably not a big deal for non-business use, unless you are a systems administrator.
Gmail has other baggage though, as does Yahoo and most free email providers:
* messages are delivered and read in clear text, without any encryption (tls, https)
* Google, Yahoo et al searches and indexes all your email and attachments and keeps them even after you've deleted them,
providing unlimited opportunity for economic espionage (whether they take advantage of such access or not is another matter)
* does not provide the quality of spam filtering that some well run non-free providers do
But at least they are reliable, and their spam filters are not prone to FPs, which is more than you can say for Yahoo, Earthlink, etc.
Clearly this is a case of extremely low standards. Either that or designed-by an artist or structural engineer with a degree, but no experience, in architecture or interior design.
IMO what makes a workspace great is light, air, and natural materials. It has to have one or more windows, which _open_. No tinted glass either. Shades or blinds are also critical, and far greener than AC. Lighting should be warm. Not the sterile, cool panels covering cheap florescent bars that we are typically stuck with.
The floor has to be _wood_ (or cork, bamboo, etc), preferably covered by one or more wool carpets. No "modern" designs please.
The desk should also be wood.
And the walls have to be _thick_ for optimal acoustics. Quiet is important for "brain work".
Don't forget location. Walking distance to rail is ideal. Nothing like a morning's drive in traffic (or a bus ride) to start the day off wrong.
You never see this kind of quality in new construction, only in renovations. That's because profits are maximized when material is cheap (as in polyester) labor is cheap (as in unskilled) and locations are cheap (as in noplace you'd ever go otherwise). Architecture is also a lost art having been replaced by structural engineering without even a trimester of ergonomics or anything to do with craftsmanship.
Cities that can combine these elements will get the best employers and employees. Paving farmland for yet another concrete and glass business park will get, well, what you pay for...
RHEL and SUSE, and their free equivalents already do absolutely everything you need a server OS to do, are stable as hell(and way ahead of canonical when it comes to security Nothing could be further from the truth. A default install of all three will illustrate: * Best gui install and package tools:
1) Ubuntu (synaptic)
2) RHEL (yumex)
3) SuSE (yast) * Fewest unnecessary applications running and listening to open network ports (portmap, nfs, xfs,...):
1) Ubuntu
2) SuSE
3) RHEL * Do pkg deinstalls also remove dependencies:
YES) Ubuntu
NO) SuSE, RHEL * Best hardware compatibility (wifi drivers, etc):
1) Ubuntu
2) SuSE
3) RHEL
As to support, no Linux support is particularly good from my perspective (as a multi-decade sysadmin) and none compare to the Sun or IBM of old. That's the fault of poorly documented and sloppily designed GPL software for the most part, but also of proprietary x86 hardware manufacturers.
So there's a really big opportunity here, for the first company to do Linux support well. Ubuntu is currently the most promising candidate in this field, by a large margin (from the perspective of someone who works on all these OS and several others every day).
the best thing the ISO can do is formalize each "standard" I'm sure that would be a satisfactory outcome for Microsoft. It is, after all, true to their well established "embrace and extend" business method. But OOXML is not capable of being an actual standard for document interchange. Think about it for a second, simply adding xml tags to the beginning and end of a proprietary file/object is like calling a sow's ear "silk, i.e., nothing more than marketing-driven rhetoric. Using terminology that means one thing to market your completely different thing, and doing so with the blessing of an "international organization", is corrupt.
Thank MSISO, but no thanks.
Future relevance of ISO given their OOXML debacle
on
ISO Takes Control Of OOXML
·
· Score: 4, Insightful
The first question is the worthiness of OOXML to be an international standard That would be my second question. The first would be regarding ISO itself. Clearly this brings down ISO's stature as a standards setting organization by several notches. I mean how seriously can you take a standard that was adopted not on its technical merit, not because it was better than competing proposals, but because the voting members could be bought?
The offer is not what you think it is. Look at who is driving it and compare him with a guy named Phil whose company did exactly the same thing buy purchasing Ashton-Tate way back when. Amazing the similarities eh?
Board and executive egos aside there are cultures to consider. A friend who works at Yahoo tells me that up to 20% of his department would resign rather than work for Microsoft, and +20% of the rest would take advantage of it as long as they could. The way I see it is both corporate cultures are in decline. MS has not been a destination for the best and brightest, except perhaps right out of college, for at least 10 years. Yahoo still has clout but has problems with retention due to management. Unlike Google neither of these cultures has kept up with the rest of Silicon Valley for quite a few years.
But then it's most likely the money that will drive this decision, assuming it isn't derailed for the non-trivial anti-trust hurdles it won't likely survive. Microsoft has the money but then so did Borland, and so did Sun when they bought Cobalt... Like Kennedy in the Bay of Pigs, nobody is pointing out the elephant in the corner, that MS cannot handle a Yahoo acquisition (profitably). But then this isn't really about profitability is it? It's all about the next quarter, and the financial analysts will, as with the mortgage industry, not to look at it from any other angle.
For many of the same reasons there is no semblance of a secure electronic voting platform on the horizon. The reason is not that such a platform would be difficult to design. The reason is that it would not be profitable.
To be secure it would have to be open. In the case of voting platforms that means every line of code, every encryption algorithm, and all the hardware has to be open, published, and known. Nobody has yet figured out how to make enough money from such a system to outspend Diebold's lobbyists and earn considered from election officials.
Where have we seen this before, oh yea, SCO. Is there a difference? I don't see one and will predict exactly the same outcome. Trend will lose their suit and stockholders will foot the bill.
Sad that another board of directors would put their support behind litigation in place innovation, but then I guess someone must have been selling short to see any benefit in such a strategy.
The rest of us will need to be pro-active and not wait to replace Trend software, wherever it may be, with software from a more open and sustainable organization.
Slashdot Mirror
Thank goodness for John Levin and Meredith Baker. Voices of sanity above the din. I don't know what is going on at ICANN but it clearly has been disabled by special interests, groupthink, and dispersion of responsibility. Their "any tld is fine with us" plan (originally proposed by France's Internic) shows such a profound lack of concern for the consequences that it's clear the bulk of their membership is simply not technically qualified.
The downside this all illustrates, beyond any doubt, is that ICANN does not and can not work in its present format. It needs to be reconstituted to insure that all members have no conflicts of interest and sufficient experience and expertise with technical and security issues. I hope it can retain the non-profit status and multi-country membership, without being so inclusive (of small countries) that it cannot avoid being corrupted as ISO was when Microsoft bought the ISO's endorsement for OOXML, or ICANN itself was when Verisign did the same to win the exclusive contract for .com.
It appears within the system administration community that Ruby is well on its way to replacing Perl
Interesting observation but not what I'm seeing here in Silicon Valley. Perl is far less popular than in years past, but Ruby is not that popular among systems admins either. Instead Python is all the rage. I believe this is because Ruby is better suited to those who spend a majority of their time writing code whereas Python can be learned and used more easily without such frequent use. So sysadmins here, who don't spend the majority of their time writing code, are using mainly Python (and shell).
Not a statistic, just an observation across a dozen or so large and small Linux and Unix using organizations.
Of course Google "believes" personalized search is the future. It's in their advertising interest to learn as much about you as they can. That's where their revenue comes from. You can't really expect them to say anything that would negatively impact revenue.
I have not found that personalized search yields good quality results. That may be because most of my searches are about stuff I don't know rather than stuff I already know. More important is the choice of personalized or non-personalized search. What Google isn't talking about, and hasn't given consumers the option of to-date, is any choice in this matter (bean-counter mandate to be sure, "do no evil" marketing aside).
That's why I use Scroogle https://ssl.scroogle.org/scrapen8.html
One thing the authors of this study seem to have forgotten is that there could still be other causes of Alzheimer's, and the HSV infection is simply opportunistic, as it is to other types of neural injury.
NAT is not a permanent solution
Nice try but pure fabrication. NAT, aka private address space, is not going away. Telcos/ILECs blocked NAT when IPv6 was being developed and have since then spent a lot on marketing IPv6 without NAT/rfc1918 as a solution too all our problems. In so doing they have delayed the adoption of IPv6 by many years. How much longer will their transparent opposition to IPv6 NAT delay the inevitable? That is the question. No, we are not going to assign public IP addresses to every network-enabled computer and other device. And no, we are not going to implement IPv6 until NAT is fully supported. This is the reality that those who claim, falsely, that NAT is not a solution, are trying to ignore.
Sadly, due to telco/ILEC influence there is not likely to be a single IPv6 NAT implement for several years. When it does happen, and it will, there is likely to have already been multiple IPv6 NAT implementations which network programmers will have a hard time reconciling. The problem is vendor lock-in, which astroturfing ILECs cannot achieve without blocking NAT, and in the process 'owning' all of your IP-enabled devices.
See also
http://www.networkworld.com/news/2008/072109-nat-housley-qna.html
http://www.techworld.com/networking/features/index.cfm?featureid=4167
http://archives.devshed.com/forums/networking-100/security-gain-from-nat-top-5t-2323463.html
We would all be better off if A) youtube and other Flash sites made their content available in MP4 and other ISO standardized formats
They don't do that because they'd have to pay licensing fees.
What are you talking about. There are no licensing fees for MP4 or any of the ISO standardized codecs.
Youtube continues to offer Flash and no other formats because Adobe made them a deal they couldn't refuse. Other streaming video sites offer standardized, open formats, without licensing fees because Adobe hasn't (yet) made them similar offers. No different from MS buying AV vendors and canceling their Mac and Linux products, or IBM buying Lotus and canceling their SunOS products.
This most surely is why we need open source.
Linux users asked, and adobe listened. Great stuff.
Would be, except that we've been asking for several _years_ now. Isn't this kind of crap (releasing software for some OS and some architectures while ignoring others) why we need open source in the first place?
Looking forward, what assurance do we have that security issues will be addressed any better, or upgrades, or new features? We have none of course.
We would all be better off if A) youtube and other Flash sites made their content available in MP4 and other ISO standardized formats, and B) if Adobe published the Flash spec so others could develop better writers and viewers.
Yes we know that Adobe has to make a profit on their product to fund further development, but they don't have to do so using the same tie-in and lock-in that got MS sued for anti-trust.
when i was in school, warm-ups were there to avoid injuries
I think the big misconception here is that stretching is some kind of warm-up. It is not. Only exercise that elevates your heart-rate can have a warm-up effect.
So, warming-up _before_ stretching will improve performance and reduce injury but stretching while cold will only reduce performance and increase the risk of injury, all else being equal.
This is to ensure that the code that you freed... remains free.
This is not technically correct, the original code remains free regardless. What the GPL does is force you to release the code that you wrote on top of the original code. It's patently not about "remaining free". It is about "making free" i.e., imposing restrictions on what you can do with your modified code. This is a point that GPL proponents try to obfuscate, until a lawsuit is filed.
Bottom line is that the GPL has been the most successful license. I particularly appreciate the fact that Microsoft cannot simply take it for use in it's Windows illegal, anti-trust, locked-in code-base. What I cannot appreciate are the GPL proponents who freely take from BSD sources and give nothing back. Remember, 50% of Linux came from BSD. 0% of BSD came from Linux. How does that really differ from Microsoft's tactics?
The reason it was blocked was that it came from an IP that was current blacklisted for spamming and was clearly a dynamic IP, not that spamassassin recognized the message. Any mail from that IP would have been blocked. Spamassassin actually fell down pretty badly on the content analysis.
Partially correct, but you're forgetting that headers _are_ content as much as the body, and any properly configured Spamassassin takes full advantage of RBLs, RHSBLs, and CBLs to identify spam (as much as any other signature). On this (well configured) server anything above 6.0 is discarded, yielding no false positives and rare false negatives (~2 per week per account). Sure it would have scored higher if it had better analyzed the hrefs, but the point is that it recognized the messages as spam.
Make that "Companies doing business on the web without basic spam filters in place". Our mailservers all run Spamassassin which easily recognized and tagged these as spam: score=8.449 tests=[BAYES_50=0.001, DNS_FROM_OPENWHOIS=1.13, HELO_DYNAMIC_DHCP=1.398, HTML_MESSAGE=0.001, RCVD_IN_BL_SPAMCOP_NET=1.2, RCVD_IN_PBL=0.905, RCVD_IN_XBL=3.033, RDNS_NONE=0.1, SARE_MONEYTERMS=0.681]. Companies that can't even manage to implement basic spam filters are at a competitve disadvantage. Those that curtail their email correspondence are, well, a good oppotunity for short sellers.
Perhaps more revealing is how many Slashdot posters are missing the real point of this story. It's not phishing, which is old news, but the security flaws of a proprietary and closed source application. There's no way Adobe can secure Flash without taking it to open source and getting the resulting peer review. That's information security 101. Heck, Adobe hasn't even figured out how to release a version of Flash for 64bit Linux, which is a heck of a lot easier than security code audits. Appears that, despite claims to the contrary, Linux users care more about free as in wallets than free as in source.
Using SSL for trust based on the word of companies like Verisign is pointless - you have to do manual authentication
I wouldn't say it is pointless but it is not a panacea. Let's not forget that the sometimes dysfunctional methods browsers use to handle SSL certificates are an artifact not of security but of monetization. We can thank Verisign and Netscape for this, as they designed it that way back when Netscape granted Verisign an exclusive on browser certificates. Verislime milked this monopoly for all it was worth (as they did with domain registration about the same time). This is is one of the features IE improved on, and one of the reasons it eventually put Netscape out of business.
The sad part of all this is that Verislime still charges big bucks for worthless "premium" certs (at a 99% markup) and people are still foolish enough to purchase them. Worse is when open source browsers support this kind of "security theater" even though it only benefits a small number of providers and their unethical business practices.
Both Google and Yahoo (not sure about Hotmail) use global spam filters that everyone trains
This is a recipe for spam. You can't trust end-users with this. You also can't trust most email administrators to take care when marking things as spam/not spam. About all you can do is manually check (some of) the messages (some) users flag for inclusion to global Bayesian / pattern recognition filtering.
ISPs and email appliances that give end-users any control over their own filtering are shooting themselves in the foot. Spam filtering is a full time job and if you want to do it right you have to spend at least a couple of hours a day on it. 999 out of 1000 untrained users/admins get it wrong, accounting for the spam "statistics" published by Gartner and others. In reality, well managed email services have virtually no false positives and at most a few spam a week, while blocking or discarding the 95-99% of email messages that are spam (at the statistical average business or ISP).
The issue at large ISPs like gmail/yahoo is ROI. As long as users are willing to put up with spam there's no point in spending much money on effective tools and methods to identify and filter it.
Can't believe people take PC Magazine seriously any more. I mean really, when you have to resort to calling Usenet dead to sell papers...
This guy probably thinks usenet's dead because Google archiving of it has gone from bad to worse. Either that or it's dead because his ISP stopped maintaining a news server. But usenet is not dead, in fact it's working just fine as the tens of thousands of groups that just came up in tin from on my local ISP (via supernews) prove.
But then usenet never was about being popular, or archiving, or graphics, or forums, and it doesn't lend itself to making money.
The many follow-on attempts to replicate usenet fall short of what still works great and is still the easiest way to get _content_ without the fluff or marketing.
No, usenet is not dead, not by a long shot, though there will always be those who will say it is to make a buck. There will probably also continue to be those who want us to think usenet is dead in order to sell their own profitable version of usenet, but follow the money and you'll find only wanna-be television content, as in most profit driven media, and lots of ads. Bottom line: Usenet will not advertise in PC magazine, and usenet will not die.
no one will get the patch because distros never package patches
That's one of the problems with binary distributions. FreeBSD sysadmins, OTOH, have only to cvsup or, if you don't even want to wait for that, edit the bind94/Makefile, change ISCVERSION from "9.4.2" to "9.4.3b2", run make makesum && make install clean, wait 3 minutes, and restart named.
But this bug also reflects poorly on ICS's Bind 9 development process. They could have reused bind v8 code which had already fixed these sequence randomization issues. Sadly, in their hast to reissue a "chock-full-o-features" bind v9 security took a back seat. Believe me this is not the only potential vulnerability in bind 9.
This Slashdot thread, OTOH, should say a lot more to developers than the original article, or at least it should to those of us without the time or capital to open-source everything by putting a big red flag around GPL anything.
Whether that will, in the long term, increase or decrease the use of the GPL is too early to say. It does, however, indicate a bright future for software developed with a BSD license.
Gmail has other baggage though, as does Yahoo and most free email providers:
* messages are delivered and read in clear text, without any encryption (tls, https)
* Google, Yahoo et al searches and indexes all your email and attachments and keeps them even after you've deleted them,
providing unlimited opportunity for economic espionage (whether they take advantage of such access or not is another matter)
* does not provide the quality of spam filtering that some well run non-free providers do
But at least they are reliable, and their spam filters are not prone to FPs, which is more than you can say for Yahoo, Earthlink, etc.
Because wood is biodegradable and renewable, vs polyester which will accumulate in landfills basically forever.
Clearly this is a case of extremely low standards. Either that or designed-by an artist or structural engineer with a degree, but no experience, in architecture or interior design.
IMO what makes a workspace great is light, air, and natural materials. It has to have one or more windows, which _open_. No tinted glass either. Shades or blinds are also critical, and far greener than AC. Lighting should be warm. Not the sterile, cool panels covering cheap florescent bars that we are typically stuck with.
The floor has to be _wood_ (or cork, bamboo, etc), preferably covered by one or more wool carpets. No "modern" designs please.
The desk should also be wood.
And the walls have to be _thick_ for optimal acoustics. Quiet is important for "brain work".
Don't forget location. Walking distance to rail is ideal. Nothing like a morning's drive in traffic (or a bus ride) to start the day off wrong.
You never see this kind of quality in new construction, only in renovations. That's because profits are maximized when material is cheap (as in polyester) labor is cheap (as in unskilled) and locations are cheap (as in noplace you'd ever go otherwise). Architecture is also a lost art having been replaced by structural engineering without even a trimester of ergonomics or anything to do with craftsmanship.
Cities that can combine these elements will get the best employers and employees. Paving farmland for yet another concrete and glass business park will get, well, what you pay for...
* Best gui install and package tools:
1) Ubuntu (synaptic)
2) RHEL (yumex)
3) SuSE (yast)
* Fewest unnecessary applications running and listening to open network ports (portmap, nfs, xfs,
1) Ubuntu
2) SuSE
3) RHEL
* Do pkg deinstalls also remove dependencies:
YES) Ubuntu
NO) SuSE, RHEL
* Best hardware compatibility (wifi drivers, etc):
1) Ubuntu
2) SuSE
3) RHEL
As to support, no Linux support is particularly good from my perspective (as a multi-decade sysadmin) and none compare to the Sun or IBM of old. That's the fault of poorly documented and sloppily designed GPL software for the most part, but also of proprietary x86 hardware manufacturers.
So there's a really big opportunity here, for the first company to do Linux support well. Ubuntu is currently the most promising candidate in this field, by a large margin (from the perspective of someone who works on all these OS and several others every day).
Think about it for a second, simply adding xml tags to the beginning and end of a proprietary file/object is like calling a sow's ear "silk, i.e., nothing more than marketing-driven rhetoric. Using terminology that means one thing to market your completely different thing, and doing so with the blessing of an "international organization", is corrupt.
Thank MSISO, but no thanks.
The offer is not what you think it is. Look at who is driving it and compare him with a guy named Phil whose company did exactly the same thing buy purchasing Ashton-Tate way back when. Amazing the similarities eh?
Board and executive egos aside there are cultures to consider. A friend who works at Yahoo tells me that up to 20% of his department would resign rather than work for Microsoft, and +20% of the rest would take advantage of it as long as they could. The way I see it is both corporate cultures are in decline. MS has not been a destination for the best and brightest, except perhaps right out of college, for at least 10 years. Yahoo still has clout but has problems with retention due to management. Unlike Google neither of these cultures has kept up with the rest of Silicon Valley for quite a few years.
But then it's most likely the money that will drive this decision, assuming it isn't derailed for the non-trivial anti-trust hurdles it won't likely survive. Microsoft has the money but then so did Borland, and so did Sun when they bought Cobalt... Like Kennedy in the Bay of Pigs, nobody is pointing out the elephant in the corner, that MS cannot handle a Yahoo acquisition (profitably). But then this isn't really about profitability is it? It's all about the next quarter, and the financial analysts will, as with the mortgage industry, not to look at it from any other angle.
For many of the same reasons there is no semblance of a secure electronic voting platform on the horizon. The reason is not that such a platform would be difficult to design. The reason is that it would not be profitable.
To be secure it would have to be open. In the case of voting platforms that means every line of code, every encryption algorithm, and all the hardware has to be open, published, and known. Nobody has yet figured out how to make enough money from such a system to outspend Diebold's lobbyists and earn considered from election officials.
Where have we seen this before, oh yea, SCO. Is there a difference? I don't see one and will predict exactly the same outcome. Trend will lose their suit and stockholders will foot the bill.
Sad that another board of directors would put their support behind litigation in place innovation, but then I guess someone must have been selling short to see any benefit in such a strategy.
The rest of us will need to be pro-active and not wait to replace Trend software, wherever it may be, with software from a more open and sustainable organization.