The point is that unless every PC in service before 2008 is decommissioned by 2008, comparing the prevalence of linux on total number of computers in service to the percentage of currently shipping computer is incorrect.
For example, suppose there were 10000 computers in service at the start of 2008, but only 250 (2.5%) running linux. Suppose also that in 2008, 1000 new computers ship, 10% of which run linux (ie 100). Suppose every computer that shipped with linux still runs linux, and that no windows pc's were converted. You now have 11000 in service PC's, of which 350 run linux. Only 3.2% of the in service PC's run linux, but we're shipping 10% linux. Oh, mercy, the windows piracy!
> 1) 10g was made for linux, all other versions are a "port"
FALSE. At least the database itself was "made first" for Solaris and Windows (primary platforms) and everything else after. I think the release dates verify this.
Releases of the database after 10g will probably (but not necessarily) switch from Solaris to Linux as a primary development platform. So you're half correct. This is largely irrelevant anyway since it is the same codebase (one of Oracle's strengths).
That might be true, except for the fact that Solaris, Linux, Tru64, HPUX, AIX5l were all available for download on OTN before the Windows port of 10g was....
1. counterproductive, since they're just refusing traffic. AFAIK linuxtoday wasn't publishing a copy (which I could see being argued as theft... but besides which this tactic wouldn't have worked against either), but posting a link to information week's article
2. pointless, since the people that are reading these types of articles might have a better idea of how the web works than the publisher apparently does, and realize that by simply copy and pasting the URL into the address bar (or by blocking their browser's reporting of HTTP referer) they can read the article without hitting the useless block.
The problem is that when you have one, eg, webserver, listening on a given IP it has no way of knowing that you want one site or another. That is why you need the host header. Think about it. Any dns implementation which fails to return an A record for a given name when it exists isn't just going to have problems with virtual hosting on single ip... Its busted!
To me what the problem boils down to is memorizing facts. Whether you're taking a cert for Windows or for some variety of Unix, etc, the problem is when memorizing facts is considered learning.
Being an MCSE or having RedHat or Sun cert shouldn't disqualify you from getting a job or getting respect, but a much more telling quality is how interested you are in the hows and whys of the way things work. Being able to deduce things for yourself. I think the problems arise in many cases when people are entering IT for the $$$ and aren't really interested in their work. They take their MCSE courses and don't bother to think about what they're "learning", or to go and try things themselves.
The challenge for a company offering certifications that they want to be worth something is to be able to select the people that can figure things out for themselves and solve real problems.
Someone outside the U.S. could implement a plugin that has the same API's as the binary iPlanet plugin using openssl library... and then we wouldn't need to wait until the RSA patent expires...
Certain countries (not many mind you) ban the sale of region restricted dvd players. If more countries banned their sale, the MPAA, etc would lose that stupid battle.
I think the same argument applies here -- UCITA or no, if many other countries prohibit the sale of backdoored software a la UCITA, software manufacturers won't avail themselves of their new-found U.S. rights. Hit them where they pay attention -- in the wallet.
I don't know that it would be effective since: (1) NSI would be in breach of contract (IANAL tho) (2) There would be huge publicity advertising MAPS and RBL to all kinds of people that have never heard of them before (3) Even if they do delete the vix.com domain, you'll note that this still wouldn't stop the use of RBL by any domain which slaves the rbl.maps domain, since you configure bind with the IP's of the master's for the zone.
Slashdot Mirror
The point is that unless every PC in service before 2008 is decommissioned by 2008, comparing the prevalence of linux on total number of computers in service to the percentage of currently shipping computer is incorrect.
For example, suppose there were 10000 computers in service at the start of 2008, but only 250 (2.5%) running linux. Suppose also that in 2008, 1000 new computers ship, 10% of which run linux (ie 100). Suppose every computer that shipped with linux still runs linux, and that no windows pc's were converted. You now have 11000 in service PC's, of which 350 run linux. Only 3.2% of the in service PC's run linux, but we're shipping 10% linux. Oh, mercy, the windows piracy!
Canada uses dollars, not pounds.
That might be true, except for the fact that Solaris, Linux, Tru64, HPUX, AIX5l were all available for download on OTN before the Windows port of 10g was....
Yeah, but it is
... but besides which this tactic wouldn't have worked against either), but posting a link to information week's article
1. counterproductive, since they're just refusing traffic. AFAIK linuxtoday wasn't publishing a copy (which I could see being argued as theft
2. pointless, since the people that are reading these types of articles might have a better idea of how the web works than the publisher apparently does, and realize that by simply copy and pasting the URL into the address bar (or by blocking their browser's reporting of HTTP referer) they can read the article without hitting the useless block.
No.
... Its busted!
The problem is that when you have one, eg, webserver, listening on a given IP it has no way of knowing that you want one site or another. That is why you need the host header. Think about it. Any dns implementation which fails to return an A record for a given name when it exists isn't just going to have problems with virtual hosting on single ip
To me what the problem boils down to is memorizing facts. Whether you're taking a cert for Windows or for some variety of Unix, etc, the problem is when memorizing facts is considered learning.
Being an MCSE or having RedHat or Sun cert shouldn't disqualify you from getting a job or getting respect, but a much more telling quality is how interested you are in the hows and whys of the way things work. Being able to deduce things for yourself. I think the problems arise in many cases when people are entering IT for the $$$ and aren't really interested in their work. They take their MCSE courses and don't bother to think about what they're "learning", or to go and try things themselves.
The challenge for a company offering certifications that they want to be worth something is to be able to select the people that can figure things out for themselves and solve real problems.
Someone outside the U.S. could implement a plugin that has the same API's as the binary iPlanet plugin using openssl library ... and then we wouldn't need to wait until the RSA patent expires...
Certain countries (not many mind you) ban the sale of region restricted dvd players. If more countries banned their sale, the MPAA, etc would lose that stupid battle.
I think the same argument applies here -- UCITA or no, if many other countries prohibit the sale of backdoored software a la UCITA, software manufacturers won't avail themselves of their new-found U.S. rights. Hit them where they pay attention -- in the wallet.
I don't know that it would be effective since:
(1) NSI would be in breach of contract (IANAL tho)
(2) There would be huge publicity advertising MAPS
and RBL to all kinds of people that have never
heard of them before
(3) Even if they do delete the vix.com domain, you'll
note that this still wouldn't stop the use of RBL
by any domain which slaves the rbl.maps domain,
since you configure bind with the IP's of the
master's for the zone.