Some people brought up that the employees end up paying taxes when they sell the stock they acquired by exercising the options.
Which would you propose:
1) The company pays tax and the employees get huge gains on stock options tax free.
2) The employee pays the tax on the huge gain and the company does not (the present situation)
3) Both the employee and the company pay tax on the options. (good for the government, bad for the company, no difference for the employee)
4) Stock options are outlawed because no one should be able to profit from their own hard work.
When the employees own a lot of company stock, the distinction between the employees and the company gets blurred a bit. But "employee owners" can be fired. Other stock holders can't.
My sister is an (traditional) animator and is in the Cartoonists Union.
I don't recall her ever complaining about it being expensive or hard entering the union.
The work itself was harder to get than the union membership.
In Animation you frequently get laid off at the end of the film. Then your off for several months.
Then you might be working for a different studio next time.
You get your benefits through the union.
Being in the union does not prevent you from taking non-union work.
So TCP/IP and the Internet were originally DARPA funded.
So by your logic, they are also in the DOD, FBI, and NSA's best interest.
The tcpdump utility uses the libpcap, which was developed at Laurence Berkeley NATIONAL LABORATORY!
I suppose you will just have to balance the risk of being cracked by non-government individuals versus the risk of using US government developed network monitoring tools to protect yourself.
When I was in graduate school. They took us on a tour of Stanford Linear Accelerator.
They had the same problem of wanting to store all the data for all the transducers in the detector.
But they could only store it if they triggered on
one in a million events (roughly).
So they used a lot of electronics to generate the trigger signal (think at least one 50ft trailer filled with relay racks).
Of course, this was around 1986 or so.
At that time, they used a VAX 11/780 to store the data onto digital tapes (the 9 track 6250bpi variety). They also used Fujitsu IBM mainframe clones things.
And they used a lot of FORTRAN.
I could see why Linux would be very popular in Physics today.
Once they capture the data, they have to plot it.
They could use GNUPlot now. I was in Physics
just before Unix and X-Windows moved in.
I believe the transition was driven by the RISC
processors gaining ground from the CISC (like the
VAX instruction set).
They had something like GNUPlot called TopDrawer.
Physicists have such big computing needs that they
will use anything they can get their hands on.
So Linux and FreeBSD have great potential.
They are also used to sharing their source code.
It isn't strictly a government thing. There are American scientists working on experiments at CERN in Switzerland, in Japan (I can't remember the name of the center), and at Fermilab in Illinois.
There are also plenty of people of other nationalities working side by side with them.
Big sciense is done internationally.
The interesting thing is that when the results are published, the paper will have two hundred or so names on it. This includes graduate students that helped build the detectors years ago and are no longer on the project.
But if the discovery is done at Fermilab, a different set of two hundred names will be on the paper.
1) The two parties are very close to each other
(at least in compared to other democratic countries).
2) The two parties politics are right in the middle of the political spectrum.
3) There are lots of people whos views are not right down the middle so they always prefer the same party they always vote for.
This doesn't mean that they don't think about their vote. It means that every time, the things that are important to them result in the same conclusion.
If you are pro-Union, which party usually has your choice in it?
If you are anti-Gun Control, which party usually has your choice in it?
You can check voting records all you want, but you will most likely reach the same conclusion.
I don't understand why anyone with programming skills would put up with 70 hour weeks.
I have bee in the business 20 years and the vast majority of that time was 40-50 hour weeks.
In a crunch, I occasionally worked 60 hour weeks.
But I never work for startups and I find that contracting (and getting paid for every hour)
tends to prevent 60 to 70 hour weeks.
Re:Irrelevant to most of us
on
Death March
·
· Score: 2
I don't think that NineNine implied that he wouldn't put in his share of work.
It is not a contractor's job to solve management problems. It is a contractor's job to
do the work assigned to them.
Saying you don't care is one way of protecting yourself emotionally when you find yourself
working for un-reasonable people.
If a project is behind schedule and your efforts can make a difference, then by all means work harder.
But what do you do when the project schedule was never realistic? After you give management your
opinion and they don't want to hear it, then what do you do? Do you burn yourself out?
"Death March" isn't talking about projects that are a little behind schedule.
Re:Irrelevant to most of us
on
Death March
·
· Score: 1
That presumes that you don't mind working with stressed out people (including managers).
I can't be happy working around a bunch of un-happy people, so I will usually look for another contract position.
There are plenty of them out there.
>> I have gone and done something about it. I wrote a letter and sent it to both my Senators.
>> You can as well. I've put the letter up for download here.
>> Sorry about it being a word doc, but I wrote it at work and our network admin is a M$ nut.
The ideal system from the ISPs perspective would be one which monitor TCP/IP traffic against each user.
The "high-traffic" users could be marked and you
could have the router start dropping packets for those users.
The one or two abusers would decide "this ISP sucks" and switch to another.
So you modify a Radius deamon to provide a map of IP numbers to users and mod the routing code on a Linux box doing your routing.
Where I once worked, there was a program written in C that was in production. Many of the programmers kept saying "that program needs to be re-written in C++".
This program processed records from a flat file from a non-Unix system, applying them to an SQL database. It also used setjmp/longjmp in such a way that if "anything really bad" happened, it did a longjmp to the main record processing loop. Because it kept going and processed all the good records after the bad ones, you did not have to restart the job and get the program to start in the middle of the file.
This may seem like a small thing, but in a production billing environment, it was a big deal. This program worked so well, there was really no need to re-write it.
If you were starting from scratch in C++, you ould do exactly the same thing with careful use of exceptions. This means logging an error at the point of detection of a failure and not putting catch blocks at every call level. Putting catch blocks in too many levels causes masking of error conditions. You only want to put the catch block at the point where you can recover.
Define fixed length arrays everywhere. Get the code working. Let the Crackers discover them later.
Have you noticed lately how many open source C projects have had buffer overrun fixes lately?
In C++, you can frequently use an in-memory stream instead of an array. This allows you to avoid
defining all those constant array sizes and trying
to guess what the maximum valid size of data will be.
Years ago, in my days of C programming, someone
was trying to fill me in on all the lore of
techniques to avoid allowing the user to crash
your program with invalid input.
I wonder if this body of lore is larger or smaller
than the that of the C++ world.
I have done both (5 years C, 6 1/2 years C++).
I prefer C++.
Slashdot Mirror
Someone at ESO should fix the headline to match the copy.
2 Billion km would by something like 20 times the length of our galaxy.
The internal switch from OpenMail to Exchange happened several months ago.
This was discussed on Slashdot a long time ago.
Some people brought up that the employees end up paying taxes when they sell the stock they acquired by exercising the options.
Which would you propose:
1) The company pays tax and the employees get huge gains on stock options tax free.
2) The employee pays the tax on the huge gain and the company does not (the present situation)
3) Both the employee and the company pay tax on the options. (good for the government, bad for the company, no difference for the employee)
4) Stock options are outlawed because no one should be able to profit from their own hard work.
When the employees own a lot of company stock, the distinction between the employees and the company gets blurred a bit. But "employee owners" can be fired. Other stock holders can't.
Logically, it makes perfect sense if MS Office is a better product than Corel Office.
What an underhanded technique, kill your competition by coming out with a better product.
So the poster could be arguing that Microsoft did not kill Corel, their better product did.
Of course, people can argue which is better. Just like they argue emacs vs. vi.
This reminds me of the advertizement that was a Robert Klein comedy routine.
"You get every album ever recorded."
Only now: "You get every message ever recorded."
This used to be the way to install drivers on SunOS (and later Solaris). This was before
.o) and use insmod, rmmod
dynamically loadable drivers.
You added your entries to cdevsw (or something like that). Recompiled the driver table and
and re-linked the kernel.
Later on, you could build a dynamically loadable module (still a
just like Linux today.
My sister is an (traditional) animator and is in the Cartoonists Union.
I don't recall her ever complaining about it being expensive or hard entering the union.
The work itself was harder to get than the union membership.
In Animation you frequently get laid off at the end of the film. Then your off for several months.
Then you might be working for a different studio next time.
You get your benefits through the union.
Being in the union does not prevent you from taking non-union work.
As far as I know, Sybase ASE is not free for "e-commerce".
So TCP/IP and the Internet were originally DARPA funded.
So by your logic, they are also in the DOD, FBI, and NSA's best interest.
The tcpdump utility uses the libpcap, which was developed at Laurence Berkeley NATIONAL LABORATORY!
I suppose you will just have to balance the risk of being cracked by non-government individuals versus the risk of using US government developed network monitoring tools to protect yourself.
Red Hat is not out of business yet. But they are not making money either. Also, their stock price is down around 7 bucks right now.
What company is actually making a profit from Linux service and technical support?
Subaru is Japanese for the Pleiades asterism.
They named the car after the asterism. The logo
is a simplified version of it.
The conic solid must stop somewhere before the moon nd planets, or these would be in the airspace of the country they were over at the time.
In Netscape 6, select Tasks, Composer.
If for some reason, this crashes (I haven't tried
it in 6 yet), you can always use Composer from
Netscape 4.7.
I use NS6 PRE3 on Windows NT.
It crashes once or twice a day.
It also has problems on my Linux box at home.
I keep using it and sending the URLS back through the automatic bug report software, hoping one day
they will get it more stable.
Where I work (as a web developer) they are slowly moving toward the attitude that there is only one
browser (IE). They even run IE under HPUX.
So I hope the Netscape team hurrys up.
When I was in graduate school. They took us on a tour of Stanford Linear Accelerator.
They had the same problem of wanting to store all the data for all the transducers in the detector.
But they could only store it if they triggered on
one in a million events (roughly).
So they used a lot of electronics to generate the trigger signal (think at least one 50ft trailer filled with relay racks).
Of course, this was around 1986 or so.
At that time, they used a VAX 11/780 to store the data onto digital tapes (the 9 track 6250bpi variety). They also used Fujitsu IBM mainframe clones things.
And they used a lot of FORTRAN.
I could see why Linux would be very popular in Physics today.
Once they capture the data, they have to plot it.
They could use GNUPlot now. I was in Physics
just before Unix and X-Windows moved in.
I believe the transition was driven by the RISC
processors gaining ground from the CISC (like the
VAX instruction set).
They had something like GNUPlot called TopDrawer.
Physicists have such big computing needs that they
will use anything they can get their hands on.
So Linux and FreeBSD have great potential.
They are also used to sharing their source code.
It isn't strictly a government thing. There are American scientists working on experiments at CERN in Switzerland, in Japan (I can't remember the name of the center), and at Fermilab in Illinois.
There are also plenty of people of other nationalities working side by side with them.
Big sciense is done internationally.
The interesting thing is that when the results are published, the paper will have two hundred or so names on it. This includes graduate students that helped build the detectors years ago and are no longer on the project.
But if the discovery is done at Fermilab, a different set of two hundred names will be on the paper.
You fail to take into account the following:
1) The two parties are very close to each other
(at least in compared to other democratic countries).
2) The two parties politics are right in the middle of the political spectrum.
3) There are lots of people whos views are not right down the middle so they always prefer the same party they always vote for.
This doesn't mean that they don't think about their vote. It means that every time, the things that are important to them result in the same conclusion.
If you are pro-Union, which party usually has your choice in it?
If you are anti-Gun Control, which party usually has your choice in it?
You can check voting records all you want, but you will most likely reach the same conclusion.
I have never worked 70 hour weeks.
I don't understand why anyone with programming skills would put up with 70 hour weeks.
I have bee in the business 20 years and the vast majority of that time was 40-50 hour weeks.
In a crunch, I occasionally worked 60 hour weeks.
But I never work for startups and I find that contracting (and getting paid for every hour)
tends to prevent 60 to 70 hour weeks.
I don't think that NineNine implied that he wouldn't put in his share of work.
It is not a contractor's job to solve management problems. It is a contractor's job to
do the work assigned to them.
Saying you don't care is one way of protecting yourself emotionally when you find yourself
working for un-reasonable people.
If a project is behind schedule and your efforts can make a difference, then by all means work harder.
But what do you do when the project schedule was never realistic? After you give management your
opinion and they don't want to hear it, then what do you do? Do you burn yourself out?
"Death March" isn't talking about projects that are a little behind schedule.
That presumes that you don't mind working with stressed out people (including managers).
I can't be happy working around a bunch of un-happy people, so I will usually look for another contract position.
There are plenty of them out there.
>> I have gone and done something about it. I wrote a letter and sent it to both my Senators.
>> You can as well. I've put the letter up for download here.
>> Sorry about it being a word doc, but I wrote it at work and our network admin is a M$ nut.
So do a File, Save As, HTML.
The ideal system from the ISPs perspective would be one which monitor TCP/IP traffic against each user.
The "high-traffic" users could be marked and you
could have the router start dropping packets for those users.
The one or two abusers would decide "this ISP sucks" and switch to another.
So you modify a Radius deamon to provide a map of IP numbers to users and mod the routing code on a Linux box doing your routing.
A piece of cake!
> resembles, almost too closely, the GOTO ...
Actually, it is more like setjmp, longjmp.
Where I once worked, there was a program written in C that was in production. Many of the programmers kept saying "that program needs to be re-written in C++".
This program processed records from a flat file from a non-Unix system, applying them to an SQL database. It also used setjmp/longjmp in such a way that if "anything really bad" happened, it did a longjmp to the main record processing loop. Because it kept going and processed all the good records after the bad ones, you did not have to restart the job and get the program to start in the middle of the file.
This may seem like a small thing, but in a production billing environment, it was a big deal. This program worked so well, there was really no need to re-write it.
If you were starting from scratch in C++, you ould do exactly the same thing with careful use of exceptions. This means logging an error at the point of detection of a failure and not putting catch blocks at every call level. Putting catch blocks in too many levels causes masking of error conditions. You only want to put the catch block at the point where you can recover.
Typical example of a C project.
Define fixed length arrays everywhere. Get the code working. Let the Crackers discover them later.
Have you noticed lately how many open source C projects have had buffer overrun fixes lately?
In C++, you can frequently use an in-memory stream instead of an array. This allows you to avoid
defining all those constant array sizes and trying
to guess what the maximum valid size of data will be.
Years ago, in my days of C programming, someone
was trying to fill me in on all the lore of
techniques to avoid allowing the user to crash
your program with invalid input.
I wonder if this body of lore is larger or smaller
than the that of the C++ world.
I have done both (5 years C, 6 1/2 years C++).
I prefer C++.
This crashes, but it crashes in the caller's code.
You never make it into the function (which may be in a library).
So if I have:
int* p = NULL;
int i = *p;
It will crash. This has nothing to do with reference arguments to functions but rather dereferencing null pointers.
I like references because, as one function
calls another, I don't have to keep doing
assert (p) or if p != NULL, over and over again.