DigitalDame2 wrote: Watching how things like this play out is interesting to me because I want to believe that the internet will require everyone to be more responsible or lose. But the real question for me is at what point does total marketplace dominance trump that.
To answer the question somewhat literaly, at 83%.
If you have 30% of a market, you're a mover. If you have
more than 80%, you pretty much own it. The 83% number
came from U.S. court cases in the days of the
robber barons and the trust-busters.
As StCredZero said, a VM is a platform
and a target of development environments.
However, it doesn't need to have a single
implementation, any more than the i86
platform does, so I encourage people to
create multiple implementations of a given
VM ABI and target different languages
toward it.
This, like building on different hardware,
exposes bugs with great enthusiasm, which
fairly rapidly yield code quality and stability,
visible as a falling bug rate.
In effect, it's a way to trade more
bugs now for fewer later (;-)).
Note that this is just one reason for having a single VM for all languages. Security, optimization, etc. are others.
Actually, I'd like to see multiple implementations of a
class of (J)VMs, in addition to kripkenstein's
point about mutiple languages targeting it/them.
This would lead to a rapid shake-out of bugs and disfeatures.
A sucessful group I was in used to have a rule that you
could reccommend someone for a job if they were
better than you at something.
This resulted in a a steady increase in
quality people. Including at least two
superstars who eventually were bored with
easy stuff like kernel and driver escalations
and left for something harder.
Alas, the parent company
eventually laid the whole group off to save
jobs at headquarters.
Now I just rent my former colleagues
back from the companies they went to.
... applies whether or not I'm running software I own or somene else's, so merely owing the software they hack with the virus dosn't necessarily protect Microsoft against a charge that they've hacked my machine.
[Left as a comment on the blog] I've had much the same experience with electronic distribution, except in a much smaller scale. I was the co-author of the first edition of O'Reilly's "Using Samba", which was published under a free documntation license, and a copy was included in every download of the Samba program.
Using Samba was O'Reilly's best seller of the period, and jumped by all the other Samba books of the day.
It seems that people were printing small sections, making notes in the margin, and then buying the professionally printed book to have it in a portable format, but not to have to carry around huge inconvenient lumps of paper.
Assuming, of course, that you use the same
technique to write a kernel module to
extract protected content, as opposed to
this one for editing your own property (;-))
It might be nice to assume AfterEffects users have
professional knowlege from computer science, but
I wouldn't risk the sucess of a business on it.
A customer of mine has dozens of people who
they consider programmers, but in fact who are
graphics artists who happen to use flash authoring
tools. Becasue they know these peopel aren't professional
sysadmins, the company IT group is quite careful to make sure
updates that they send to the users are safe.
However, IT has little control over something sent directly
from the OS supplier to the end-user.
And when something mandatory, like viewing a
training film, triggers the "you gotta press OK"
message,
all the warnings in the world from IT aren't going to
overide the "watch this film" order from the
user's manager.
Thus the vendor has a duty, and it's a legal duty,
albeit one which is enforced by class action suits (;-))
My accountant woks for a video production firm (;-))
Joking aside, if your vendor sends you an update and
expects you to apply it, they have a duty to ensure that
thay've made a good-faith effort to ensure that
it isn't a root-kit or a brick-kit.
If they're not, they deserve public approbation
and a sharp smack to the wallet (suppliers often
don't have wrists).
That assumes that everyone is a sysadmin. I am, so
the suggestion is usable, but what if I was an
accountant? I get a mandatory training film on Sarbanes-Oxley
that says "upgrade your quicktime", I click the icon, and
my computer turns into a brick.
I'd claim the onus is on the distributor of
quicktime, that they test their updates and certify
that they have done due diligence to ensure
that they are not shipping, for example, a rootkit.
And if they haven't, then let litigatious customers
sue them into oblivion.
Use the recent Dtrace-fix kernel module to get tracing working,
and trace the offending program until you find the error.
Then write a kenel module to fix that.
lintux said:
IIRC this is done in mainframes for *ages* already...
Indeed, because CPUs for mainframes were sereriously
expensive, it would make sense to do "capacity on demand"
or "dynamic domain reconfiguration" (The current marketing
names for the "Limited Up/Downgrade model".
In fact, however, adding cores is seriously cheap: Azul
sells 48-core Java offload engines, Sun has 8-core chips,
and everybody credible has two.
That means buying excess capacity is easy and inexpensive,
so one can trivially size for your highest spikes
Really cool, expensive displays over the urinals, with ads looping continuously. Unfortunately, washrooms aren't well policed: one local display was used as a new storage location for the fire axe.
It is your duty to accept what the Judge says about the laws to be applied to the case, whether you agree or disagree with the law."
In at least Canada, this is a famous lie of omission: a
jury can find a defendant not guilty despite the law (see
"jury nullification" in both Canada and the U.S.).
This is why we have juries, you understand, not
logic programs, or even judges alone. Justice
trumps law, famously in the Morgentaler cases,
where he was repeatedly charged, and repeatedly
found innocent, whatever the Judge believed.
So we put a return instruction at the end of the title string, zeroed the registers and called the first byte of the string. If the resulting register contents weren't right, we executed a halt instruction.
I've always claimed that whenever Sun wrote a strange license, it was because their lawyers told them to.
You may recollect a small war between Sun and MS over the MS effort to "embrace and extend" Java.
I suspect we'll see more GPL3 and LGPS3 as it is shown in practice to provide the same patent potection as CDDL.
--dave
DigitalDame2 wrote: Watching how things like this play out is interesting to me because I want to believe that the internet will require everyone to be more responsible or lose. But the real question for me is at what point does total marketplace dominance trump that.
To answer the question somewhat literaly, at 83%.
If you have 30% of a market, you're a mover. If you have more than 80%, you pretty much own it. The 83% number came from U.S. court cases in the days of the robber barons and the trust-busters.
--dave
As StCredZero said, a VM is a platform and a target of development environments.
However, it doesn't need to have a single implementation, any more than the i86 platform does, so I encourage people to create multiple implementations of a given VM ABI and target different languages toward it.
This, like building on different hardware, exposes bugs with great enthusiasm, which fairly rapidly yield code quality and stability, visible as a falling bug rate.
In effect, it's a way to trade more bugs now for fewer later (;-)).
--dave
Actually, I'd like to see multiple implementations of a class of (J)VMs, in addition to kripkenstein's point about mutiple languages targeting it/them. This would lead to a rapid shake-out of bugs and disfeatures.
--dave
Can they develop some more of Monty?
I haven't seen anything new from him
for a long time.
--dave
A sucessful group I was in used to have a rule that you could reccommend someone for a job if they were better than you at something.
This resulted in a a steady increase in quality people. Including at least two superstars who eventually were bored with easy stuff like kernel and driver escalations and left for something harder.
Alas, the parent company eventually laid the whole group off to save jobs at headquarters.
Now I just rent my former colleagues back from the companies they went to.
--dave
First, build a language or system that runs existing programs.
Then change the compilers so they use MS-only, intel-only features by default
Then add attractive features at the source level.
Pretty soon, you can port *to* the new platform, but can't port away from it.
--dave
[PS: If you're already in that situation and want to port, send me private email]
... applies whether or not I'm running software
I own or somene else's, so merely owing the
software they hack with the virus dosn't
necessarily protect Microsoft against a
charge that they've hacked my machine.
-dave
Possibly he's keeping the duress a secret because he's under duress?
A credible threat, legal or illegal, usually contains a sub-threat about telling anyone, especially the cops.
--dave
[Left as a comment on the blog]
I've had much the same experience with electronic distribution,
except in a much smaller scale. I was the co-author of the first
edition of O'Reilly's "Using Samba", which was published under a
free documntation license, and a copy was included in every
download of the Samba program.
Using Samba was O'Reilly's best seller of the period, and jumped
by all the other Samba books of the day.
It seems that people were printing small sections, making
notes in the margin, and then buying the professionally
printed book to have it in a portable format,
but not to have to carry around huge inconvenient lumps of paper.
--dave
And the more things running on it and the more popular they are, the more hardware they sell.
After that comes the much smaller income you get from the commercial uses of MySQL.
--dave
Assuming, of course, that you use the same technique to write a kernel module to extract protected content, as opposed to this one for editing your own property (;-))
--dave
It might be nice to assume AfterEffects users have professional knowlege from computer science, but I wouldn't risk the sucess of a business on it.
A customer of mine has dozens of people who they consider programmers, but in fact who are graphics artists who happen to use flash authoring tools. Becasue they know these peopel aren't professional sysadmins, the company IT group is quite careful to make sure updates that they send to the users are safe. However, IT has little control over something sent directly from the OS supplier to the end-user.
And when something mandatory, like viewing a training film, triggers the "you gotta press OK" message, all the warnings in the world from IT aren't going to overide the "watch this film" order from the user's manager.
Thus the vendor has a duty, and it's a legal duty, albeit one which is enforced by class action suits (;-))
--dave (DRBrown.TSDC@HI-Multics.ARPA) c-b
My accountant woks for a video production firm (;-))
Joking aside, if your vendor sends you an update and expects you to apply it, they have a duty to ensure that thay've made a good-faith effort to ensure that it isn't a root-kit or a brick-kit.
If they're not, they deserve public approbation and a sharp smack to the wallet (suppliers often don't have wrists).
--dave
That assumes that everyone is a sysadmin. I am, so the suggestion is usable, but what if I was an accountant? I get a mandatory training film on Sarbanes-Oxley that says "upgrade your quicktime", I click the icon, and my computer turns into a brick.
I'd claim the onus is on the distributor of quicktime, that they test their updates and certify that they have done due diligence to ensure that they are not shipping, for example, a rootkit.
And if they haven't, then let litigatious customers sue them into oblivion.
--dave
Use the recent Dtrace-fix kernel module to get tracing working, and trace the offending program until you find the error. Then write a kenel module to fix that.
--dave
.. and see what it's doing, then load a kernel module
to undo it.
--dave
Indeed, because CPUs for mainframes were sereriously expensive, it would make sense to do "capacity on demand" or "dynamic domain reconfiguration" (The current marketing names for the "Limited Up/Downgrade model".
In fact, however, adding cores is seriously cheap: Azul sells 48-core Java offload engines, Sun has 8-core chips, and everybody credible has two.
That means buying excess capacity is easy and inexpensive, so one can trivially size for your highest spikes
--dave
Really cool, expensive displays over the
urinals, with ads looping continuously.
Unfortunately, washrooms aren't well
policed: one local display was used as
a new storage location for the fire axe.
--dave
Hmmn, it says here your license was issued in
Georgia. What is your business here in Moscow?
--dave
Thank you, kind sir!
In at least Canada, this is a famous lie of omission: a jury can find a defendant not guilty despite the law (see "jury nullification" in both Canada and the U.S.).
This is why we have juries, you understand, not logic programs, or even judges alone. Justice trumps law, famously in the Morgentaler cases, where he was repeatedly charged, and repeatedly found innocent, whatever the Judge believed.
--dave
So we put a return instruction at the end of the
title string, zeroed the registers and called the
first byte of the string. If the resulting
register contents weren't right, we executed
a halt instruction.
--dave
I suspect the more stringent licenses are for things that can be embraced and extended, possibly by a particular well-known competitor to Java (;-))
--dave
If you were btten by Bob's dog, you could either sue or charge the dog, and in the latter case have it whipped to death for assault.
Oddly enough, this was less effective than suing Bob, as he could just keep buying new dogs.
It's no longer legal in the British system, but it sounds like it's back in the U.S., but only for non-living pets (;-))
--dave