I personally think it should be self-evident to any educated person that such expensive and painful policies are not adopted uselessly and painlessly. (I can't argue about "invasively".) The only questions are "what is the use?", "what is the point?".
Here's a reasonable hypothesis: We are under de facto martial law, in direct contravention of the Posse Commitatus Act, and the purpose is to condition the populace to unquestioning obedience to the surveillance state.
If you think my hypothesis unreasonable, I invite you to suggest a superior alternative.
You forget, this is Canada. You have no rights. The only reason it's not a playground for fascist butchers is that they're all acting like Doug and Dave MacKenzie.
Now in the U.S., you'd get the twice the brutality, but you would have the comfort of knowing that it was illegal, although of course no court in the land would give a flying wahoo about that.
Code is not apolitical by any stretch of the imagination. It is often *purely* a form of political expression, with no other intention on the part of the author. The functions of the code in such cases are side-effects which are necessary to appeal to the audience, like L. Frank Baum writing Oz books to promote the free silver cause, or Hitler's swagger and punch.
Problem? No, that's the *appeal*. Permutation fascination. James Joyce on acid. Racter | travesty | eliza. What's amazing is that they speak so well for such a large community.
> BTW, I wouldn't consider him a professional > nay-sayer, but rather skeptical, analytical (both > good qualities in a scientist) and out spoken > (which can be good or bad)
Skepticism can easily exceed the bounds of intellectual honesty. When such an excess becomes ingrained and habitual, self-justifying delusion sets in.
Analysis of the unknown is folly. That's why the scientific method consists of the creative generation of hypothesis, which is then confirmed or disconfirmed by experimentation.
The bottom line in science is not analysis, or orthodox dogma, or arguments from authority, but the cold, hard facts of experimental evidence, and the delusive skepticism of ideologues such as Park pollute the public mind, as witness the ignorant comments in this slashdot article, or worse yet create in credulous factions of the public a reactionary embrace of the entire range of heterodox opinion, rather than just those elements contrary to orthodoxy which are well-attested by observation.
You know, this is a *very* interesting point. A StrongARM PDA is quite attractive as a web server. I'd eschew the microdrive, just keep static content in flash, and use the CF slot for a 100bTX interface instead.
As server blades, these things would be a *dream*: Ultra low power, trivial heat dissipation, built-in LCD console. And the best part is, they have consumer-level economies of scale, so that price/performance is way ahead of anything you can drop in a 2U backplane.
Someone should make a rack mount for these things.
Agreed, HTTP is not designed for RPC, so writing an RPC system on top of it is sometimes painful, but with software, one man's pain is all that it takes for everyone to benefit from that implementation, so why should anyone care?
Oh, but I forgot, this was a Microsoft guy talking, and he assumes that everyone and his brother has to reimplement the same code because nobody would *dream* of sharing their "intellectual property". Pfft. I'll ignore that guy.
NeWS, Motif, CDE, now Gnome. I think the CDE experience blinded Sun to the KDE advantage, because KDE incorporates too much CDE icing. It's really too bad, because KDE provides a superior component architecture, and it much more advanced in it's functional development than is Gnome.
Doesn't the constitution specify that no treaty is effective or binding unless ratified by Congress?
Now I know that much of the Constitution is an irrelevant theoretic excercise, since Roosevelt established an autocratic presidency by threatening to pack the supreme court in order to get the grotesquely unconstitutional ruling of washburn in 1942, but surely this core element of the document is still in force!
Well it's not FUD if it's true. I would prefer to address the substance of the arguments rather than relying on demagoguery.
Peronsally, I think the decline, if not death, of Solaris is inevitable -- indeed, manifest already -- so that Sun will be doing the same think IBM is doing now, very soon, if they can get their egos under control long enough to earn their shareholders some money.
Your confidence is misplaced. The arguments would have to be very dissimilar. Microsoft doesn't make robust hardware for enterprise scale application services.
Sun makes great hardware, for these kinds of apps, where robustness and scalability are crucial. It's a sucky horrible place to work, and most of their software is useless (Java excepted) but the hardware is way cool.
Now if Sun would drop Solaris, those mainframe sales would be going to Linux on Sun hardware instead. But some clueless suit bought SVR5 10 years ago and still can't admit that it was a big, expensive mistake.
Well then, you should buy a starfire, because you can keep it running, slap in new processors, new memory, and then suck them into a running partition.
It seems that most of the criticisms of Shahin Kahn's article are based on ignorance. It's a fair assessment of the liabilities of using mainframe hardware for typical modern web service applications. IBM tried to save the mainframe from declining market share in a very ingenious way, and Linux and IBM have benefited from it, but that doesn't mean that it is competetive with Sun's hardware offerings for the same application environments.
Not all of Kahn's objections to VMs are valid, however. The robustness arguments are good, but the performance ones are short-sighted. While s/390 Linux may not be tuned today, you can be confidently assured that it will be soon -- even if IBM has to fork the kernel to do it.
>For large organizations who (like mine) made a >$50mil investment in moving to J2EE applications as >a corporate standard, the Sun stamp of approval is >absolutely necessary.
Think about *why* certification is a checklist item: You want to know that your application code will run, to the degree to which you have adhered to the certified interface contracts. The same goal can often be met by using a single revision of an open-source solution -- just don't upgrade. The certified COTS solution and the open-source solution have real cost tradeoffs, and I can't comment intelligently on how they play in your applications, but I do hope for the sake of your organization that you will actually analyze and weigh those trade-offs, rather than discounting one alternative because it doesn't satisfy a derived requirement without business legitimacy.
For good cross-platform consolidation of access management, I recommend using Novell Directory Services (a nicely compatible LDAP implementation is included, and more), in conjunction with Kerberos v5. Keys can be kept in NDS LDAP, and auth tickets (which make powerusers of multiple machines and services on the network very happy because they don't have to enter passwords every 15 seconds) granted by krb5. I haven't tried integrating standard krb5 and microsoft krb5-alike systems in one network before, and can imagine that there may be some issues that need to be finessed here, but if you just avoid the MS implementation altogether, you can end-run those issues.
Writev has nothing to do with altivec really. It is an ancient unix optimization in which a number of write(2) operations are combined in order to save on syscalls. It's like an old vector machine's scatter/gather, but for memory-disk operations instead of register-memory operations. One could conceivably use some altivec insns in the implementation of writev, but one could do that for any syscall implementation, really, so it's not directly bearing.
While it is conceivable that writev might be implemented using some altivec insns on G>=4, writev really doesn't have anything to do with altivec. writev is an optimization introduced in the pre-BSD days, to allow one to make several writes with just one syscall. The 'v' in writev is the vector of regions to write in the file.
I stopped using windows/outlook/pgp, and
switched to KDE/KMail/gpg, and I find it easier
to use, as well as eliminating most of my security
problems. Encrypting on the wire is great, but
it does no good if some Outloook VBScript virus
has installed a backdoor on your machine.
If you don't need to buy a current model -- and in
this case, you don't -- then you could buy a couple
of refurb older generation laptops for the price of
one zaurus. Mount one under the front seat of your
car, and tote the other to parties.
Slashdot Mirror
> uselessly, pointlessly, invasively.
I personally think it should be self-evident to any
educated person that such expensive and painful
policies are not adopted uselessly and painlessly.
(I can't argue about "invasively".) The only
questions are "what is the use?", "what is the
point?".
Here's a reasonable hypothesis: We are under
de facto martial law, in direct contravention of
the Posse Commitatus Act, and the purpose is to
condition the populace to unquestioning obedience
to the surveillance state.
If you think my hypothesis unreasonable, I invite
you to suggest a superior alternative.
You forget, this is Canada. You have no rights.
The only reason it's not a playground for fascist
butchers is that they're all acting like Doug and
Dave MacKenzie.
Now in the U.S., you'd get the twice the brutality,
but you would have the comfort of knowing that it
was illegal, although of course no court in the land
would give a flying wahoo about that.
Knitting needles?!?! Why, she could have been
knitting an.... AFGHAN!
Because people get killed all the time, every day,
for their beliefs, or for their plans to stop the
killing.
Code is not apolitical by any stretch of the
imagination. It is often *purely* a form of
political expression, with no other intention
on the part of the author. The functions of the
code in such cases are side-effects which are
necessary to appeal to the audience, like L.
Frank Baum writing Oz books to promote the
free silver cause, or Hitler's swagger and punch.
Problem? No, that's the *appeal*. Permutation
fascination. James Joyce on acid. Racter |
travesty | eliza. What's amazing is that they
speak so well for such a large community.
58,000 megatons seems bigger than 58,000,000 tons
because it is 1000 times bigger than 58,000,000 tons.
No, FC is not an advanced form of SCSI. You can
run SCSI over FC-AL, however. You can also run
SCSI over OC384 or a 300 baud modem now (iSCSI).
> BTW, I wouldn't consider him a professional
> nay-sayer, but rather skeptical, analytical (both
> good qualities in a scientist) and out spoken
> (which can be good or bad)
Skepticism can easily exceed the bounds of
intellectual honesty. When such an excess
becomes ingrained and habitual, self-justifying
delusion sets in.
Analysis of the unknown is folly. That's why
the scientific method consists of the creative
generation of hypothesis, which is then confirmed
or disconfirmed by experimentation.
The bottom line in science is not analysis,
or orthodox dogma, or arguments from authority,
but the cold, hard facts of experimental evidence,
and the delusive skepticism of ideologues such as
Park pollute the public mind, as witness the
ignorant comments in this slashdot article, or
worse yet create in credulous factions of the
public a reactionary embrace of the entire range
of heterodox opinion, rather than just those
elements contrary to orthodoxy which are
well-attested by observation.
You know, this is a *very* interesting point.
A StrongARM PDA is quite attractive as a web
server. I'd eschew the microdrive, just keep
static content in flash, and use the CF slot
for a 100bTX interface instead.
As server blades,
these things would be a *dream*: Ultra low power,
trivial heat dissipation, built-in LCD console.
And the best part is, they have consumer-level
economies of scale, so that price/performance
is way ahead of anything you can drop in a 2U
backplane.
Someone should make a rack mount for these things.
Agreed, HTTP is not designed for RPC, so writing
an RPC system on top of it is sometimes painful,
but with software, one man's pain is all that it
takes for everyone to benefit from that
implementation, so why should anyone care?
Oh, but I forgot, this was a Microsoft guy talking,
and he assumes that everyone and his brother has to
reimplement the same code because nobody would
*dream* of sharing their "intellectual property".
Pfft. I'll ignore that guy.
NeWS, Motif, CDE, now Gnome. I think the CDE
experience blinded Sun to the KDE advantage,
because KDE incorporates too much CDE icing.
It's really too bad, because KDE provides a
superior component architecture, and it much
more advanced in it's functional development
than is Gnome.
Doesn't the constitution specify that no treaty
is effective or binding unless ratified by Congress?
Now I know that much of the Constitution is
an irrelevant theoretic excercise, since Roosevelt
established an autocratic presidency by threatening
to pack the supreme court in order to get the
grotesquely unconstitutional ruling of washburn
in 1942, but surely this core element of the
document is still in force!
Well it's not FUD if it's true. I would prefer to
address the substance of the arguments rather than
relying on demagoguery.
Peronsally, I think the decline, if not death,
of Solaris is inevitable -- indeed, manifest
already -- so that Sun will be doing the same
think IBM is doing now, very soon, if they can
get their egos under control long enough to
earn their shareholders some money.
Your confidence is misplaced. The arguments
would have to be very dissimilar. Microsoft
doesn't make robust hardware for enterprise scale
application services.
Sun makes great hardware, for these kinds of apps,
where robustness and scalability are crucial.
It's a sucky horrible place to work, and most of
their software is useless (Java excepted) but
the hardware is way cool.
Now if Sun would drop Solaris, those mainframe
sales would be going to Linux on Sun hardware
instead. But some clueless suit bought SVR5
10 years ago and still can't admit that it was
a big, expensive mistake.
Well then, you should buy a starfire, because you
can keep it running, slap in new processors, new
memory, and then suck them into a running
partition.
It seems that most of the criticisms of Shahin
Kahn's article are based on ignorance. It's a
fair assessment of the liabilities of using
mainframe hardware for typical modern web service
applications. IBM tried to save the mainframe
from declining market share in a very ingenious
way, and Linux and IBM have benefited from it,
but that doesn't mean that it is competetive
with Sun's hardware offerings for the same
application environments.
Not all of Kahn's objections to VMs are valid,
however. The robustness arguments are good, but
the performance ones are short-sighted. While
s/390 Linux may not be tuned today, you can be
confidently assured that it will be soon -- even
if IBM has to fork the kernel to do it.
>For large organizations who (like mine) made a
>$50mil investment in moving to J2EE applications as
>a corporate standard, the Sun stamp of approval is
>absolutely necessary.
Think about *why* certification is
a checklist item: You want to know
that your application code will run, to the
degree to which you have adhered to the
certified interface contracts. The same goal
can often be met by using a single revision
of an open-source solution -- just
don't upgrade. The certified COTS solution
and the open-source solution have real cost
tradeoffs, and I can't comment intelligently
on how they play in your applications, but
I do hope for the sake of your organization
that you will actually analyze and weigh those
trade-offs, rather than discounting one
alternative because it doesn't satisfy a
derived requirement without business
legitimacy.
personally, i think the only morally respectable
response to this state of affairs is terrorism.
For good cross-platform consolidation of access
management, I recommend using Novell Directory
Services (a nicely compatible LDAP implementation
is included, and more), in conjunction with Kerberos
v5. Keys can be kept in NDS LDAP, and auth tickets
(which make powerusers of multiple machines and
services on the network very happy because they
don't have to enter passwords every 15 seconds)
granted by krb5. I haven't tried integrating
standard krb5 and microsoft krb5-alike systems
in one network before, and can imagine that there
may be some issues that need to be finessed here,
but if you just avoid the MS implementation
altogether, you can end-run those issues.
Writev has nothing to do with altivec really.
It is an ancient unix optimization in which a
number of write(2) operations are combined in
order to save on syscalls. It's like an old
vector machine's scatter/gather, but for memory-disk
operations instead of register-memory operations.
One could conceivably use some altivec insns in the
implementation of writev, but one could do that for
any syscall implementation, really, so it's not
directly bearing.
While it is conceivable that writev might be
implemented using some altivec insns on G>=4,
writev really doesn't have anything to do with
altivec. writev is an optimization introduced
in the pre-BSD days, to allow one to make several
writes with just one syscall. The 'v' in writev
is the vector of regions to write in the file.
That's a problem with C++, not a problem with
operator overloading.
I stopped using windows/outlook/pgp, and
switched to KDE/KMail/gpg, and I find it easier
to use, as well as eliminating most of my security
problems. Encrypting on the wire is great, but
it does no good if some Outloook VBScript virus
has installed a backdoor on your machine.
If you don't need to buy a current model -- and in
this case, you don't -- then you could buy a couple
of refurb older generation laptops for the price of
one zaurus. Mount one under the front seat of your
car, and tote the other to parties.
Nah, just use serial console, and connect it to
a terminal server.
Also, I would turn on swap. You don't need to
mount any FS partitions to mount swap.
.