If you're capturing packets from multiple machines and want to line up the captures, then you need to have accuracy.
If you're using SNMP to log equipment on the network, it helps to have everything as lined up as you can. Now, if you're a company doing this, typically you have your own time server and don't rely on this pool. But there are benefits to some to have more exacting time across all devices.
A lot of things depend on that if A happens before B, it gets timestamped as <= B.
Compiling things across a networked file system is enough to make you
want highly accurate time.
Operator overloading is an immensely useful paradigm, and this was not lost on Sun, but how it has been used in C++ in the past has set a clear precedent that most developers, left to their own devices with operator overloading, will tend to write code that is concise at the expense of being intuitive, regularly using operators in ways that bear little to no resemblance to their natural meaning, often resulting in code that is difficult for other people to understand and change later.
I don't know what typical C++ code looked like in the early 1990s,
but I'm sure it sucked.
But I can tell you I've seen a lot of bad C++ code written by people who did whatever pleased them,
and none of it abused operator overloading.
If this was what worried Sun, they made the wrong choice.
I think what really happened was that someone was on a Language Purist trip.
"We only need one single syntax for method calls", and so on.
Those text files are byzantine and subject to total failure, should one character be out of place. Have you ever tried to walk someone through typing in commands over the phone?
Have you ever tried to walk someone through a bloody GUI over the phone?
Or even by email/IM?
... and I've heard from many in the embedded world who state that Eclipse CDT has become the standard IDE in that arena.
I haven't used it, but it should be pointed out that Eclipse is now so big
that it's also used for a lot of stupid reasons -- that it's used for $foo doesn't
prove that it's a good fit for $foo, or for anything else.
I used to be in a project where we used Visual Studio 6.
In reality, it sucked, so I wrote all my code in Emacs.
I'm sorry, but that was as non-erotic as you can possibly get. If anyone considered it "scoring" that their sim gave a backrub and had a friendly kiss, or worse yet as some pornographic material to choke the chicken to, I'd seriously worry about their mental health.
You underestimate the power of human imagination. I've known people who have masturbated while interacting with a succubus in Nethack.
Uh, I think that is covered by the phrase "seriously worry about their mental health".
Well, according to http://it.slashdot.org/comments.pl?sid=281505&cid= 20383407 , none of the companies that voted YES in the Swedish vote "came in late" either. All were already members of the organization.
That doesn't match the
Computer Sweden article
which lists 24 out of 35 organizations in the working group as
newcomers,
18 of these 24 being "Microsoft Gold Certified Partner",
and cites a SIS representative saying there were eleven
members in late June
-- and that the big influx started on Thursday (two work days
before the vote?!).
Or most likely, getting the user's home directory so it knows where to find $HOME/.Skype to get the user's configuration settings. Virtually any program will do this, via the getpwnam function, section 3 of the Linux man page.
It's probably better to simply getenv("HOME") to find $HOME.
It's already in your process; why bother asking a file (or possibly a remote database)?
But yes, there are other mundane and valid reasons for reading/etc/passwd.
So now I'm sending mail to , for $foo in some of the domains I use frequently. I hope they'll write me back...
And all four responded quickly with proper, readable NDRs.
One large Microsoft-dominated tech company,
one mid-sized Microsoft-dominated tech company,
my sucky ex-ISP,
and the server in my closet.
Proper mail servers bounce during the SMTP session. Even AOL has LDAP integration so they can bounce during the SMTP session. If they can do it, anyone can.
Except a backup MX without access to an up-to-date copy of the full user database and
rewriting rules of the real MX that does local delivery.
Remember that a backup MX can (probably should) be far from the primary:
it is supposed to kick in when the primary
has lost network, suffered a disk crash, or is down for any kind of maintenance.
Am I the only one here that likes NDRs? If some important email is not delivered, I would much prefer that a sender is notified about that failure. Wouldn't you?
Yes.
I have mailed people and become seriously offended when the don't reply.
That's because I can assume that when I get no feedback, it's the recipient who blew it -
by not reading his mail, or not bothering to respond.
Like others have noted, NDR is an important part of the semantics of Internet mail.
So now I'm sending mail to , for $foo in some of the domains
I use frequently. I hope they'll write me back...
The users of our apps are business professionals who are forced to use them, so they are are more tolerant of access times being a second or two slower than they could be.
I realize that this is just background to the question,
but it strikes me as an odd thing so say:
Being forced to do something doesn't mean you tolerate it.
Do you realize how stressful it is to be forced to work with
slow applications? How harmful it is to your work?
I have been the victim of numerous really crappy internal applications.
It makes me mad, because it shows a lack of respect for my work.
What tends to happen in practice, if the users are technically minded,
is that they don't use the applications as intended, and invent
some primitive system on the side. Excel sheets, and so on.
And the people who support the real system
are usually happily unaware of this.
They live in a fantasy land, where things work fairly well and the users are pleased.
Given the size and scope of Windows, Microsoft probably can't easily port Windows and Win32 and Visual Studio's compiler over to another instruction set easily.
Whatever the cause is, it isn't size and scope.
Practically any piece of free software compiles on a dozen architectures.
For example, Debian Gnu/Linux ships around thirteen gigabytes of software
for each of eleven architectures...
Iostreams is big. I really like C++ but that's one part of it that I've never really been a fan of. I prefer the compactness of printf format strings.
printf() is nice -- as long as you know at compile time the types of
all arguments, and never print any types except the handful which printf() supports...
For you.
For normal systems, "Hello, world!" is more like the 5344 bytes
I got when I compiled it on my Debian machine.
Or 4896 bytes on my ppc machine. Hey, that must mean
the PowerPC architecture is 8.4% better than AMD64!
But seriously -- what those 256 kilobytes mean is simply that
noone has bothered to optimize the linker and standard library
of your toolchain for iostreams.
Possibly because too many people think they are too large
and don't use them.
I'm probably going to get modded as either flamebait or troll for saying this, but I really fail to see the attraction here. As I see it, although the original code might be desirable to keep around as a reference for historical purposes, the state of the art in program design has advanced well beyond what that program has implemented. I'm not talking about the lack of any graphics or fancy features, what I mean is that this old code is almost structureless, difficult to understand, and appears virtually impossible to modify (not that one would ever want to, mind you, it's the principle of the thing). I understand that some of the modern program design methodologies we use today had not yet been discovered when this Crowther wrote this, but other than serving as an example of how _not_ to write a computer program today [...]
You miss the point.
It has nothing to do with bloody software engineering.
It's about being curious about how computers work,
and about how some hacker's mind worked back then.
It's about how cool it is that this FORTRAN gibberish translates into
a really good game.
You were excited about all that when you got your first computer, weren't you?
yeah, can't say I'm anything other than a rogue, nethack, moria, umoria fan. the modern games with their "animation" and "pictures" and "sound" are just too easy.
Funny you should say that --
unlike adventure, Nethack has all those features
(ASCII animations, ASCII graphics and -- in at least the Amiga port -- sound).
Re:Article not very informative
on
Hardening Linux
·
· Score: 1
The article isn't very informative and makes several assumptions about the distribution being used. For example, when it tells the reader to "ps aux|grep http" and then "kill -9 [the pid]" it doesn't take into account that Debian systems are running Apache2 as 'apache2', not 'httpd'. Why you would SIGKILL the running process instead of just using apachectl or the appropriate init script is also just as short-sighted.
It triggers me on two other points too:
ps + grep + kill is so 1990s.
pkill from Solaris via procps is the right tool if you
really need to do this.
Sending SIGKILL to a process is rude.
If the process doesn't die from SIGTERM, chances are that
you should think again, not reach for the SIGKILL Wand of Death...
Even when I check "developer tools", I still have to go through the list to include those older, simpler parts of the toolchain that, for some reason, are too "simple" for todays' developers.
Who ever thought that "tree" would be near-forgotten? What next, fgrep?:-)
Tree is hardly a developer tool, is it?
But yes, it's sometimes useful and I install it on all my machines.
We're just the same here in Australia as in any western country like England or America. I've lived in England, and I've heard we're more similar to America than England.
But if you had lived in America instead, don't you think they would have told you
you were kind of english?
In fact, I'd say the programmer who thinks he's the best is most likely to be the worst. The programmer who will say he doesn't know is most likely to be the best.
No... Rather, it's hard for the programmer himself to know how good he is.
You can observe things like
"I'm better at producing solid code than that guy"
or
"that guy botched that task which I found simple"
or
"people seem to trust my abilities"
or
"I failed to understand the Linux networking code"
but you can't get meaningful measurements.
(You can get meaningless measurements;
those are the ones which end up on your CV.)
NAT works for most people because they do simplistic things with their internet connection.
Also, they do simplistic things with their internet connection
because they are behind NAT.
Countless cool technologies and ideas have died,
just because NAT was in the way.
Everyone should be assigned a personal cert by a central authority as part of signing on. For communication where you want to be able to verify your authenticity, you use the personal cert. Everyone can use this cert for anything---encrypted communications, commerce, etc.
I still don't understand why His Majesty the King of Sweden won't sign my GnuPG key.
Or some equivalent national system.
There's something like it in.se: "E-legitimation".
A certificate you can order from the major banks.
However, it appears to be somewhat Windows-specific, and you can only use it to authenticate yourself to the major banks and the Government, i.e. The Man. As far as I can see, it's useless for establishing trust between two humans.
All this is, of course, completely orthogonal to the Internet.
Slashdot Mirror
A lot of things depend on that if A happens before B, it gets timestamped as <= B. Compiling things across a networked file system is enough to make you want highly accurate time.
I don't know what typical C++ code looked like in the early 1990s, but I'm sure it sucked.
But I can tell you I've seen a lot of bad C++ code written by people who did whatever pleased them, and none of it abused operator overloading. If this was what worried Sun, they made the wrong choice.
I think what really happened was that someone was on a Language Purist trip. "We only need one single syntax for method calls", and so on.
Seems to me the following would be good enough for many purposes (assuming C, C++ or similar source). When spellchecking a set of source files:
Or, you could just use aspell:
Have you ever tried to walk someone through a bloody GUI over the phone? Or even by email/IM?
I haven't used it, but it should be pointed out that Eclipse is now so big that it's also used for a lot of stupid reasons -- that it's used for $foo doesn't prove that it's a good fit for $foo, or for anything else.
I used to be in a project where we used Visual Studio 6. In reality, it sucked, so I wrote all my code in Emacs.
Uh, I think that is covered by the phrase "seriously worry about their mental health".
That doesn't match the Computer Sweden article which lists 24 out of 35 organizations in the working group as newcomers, 18 of these 24 being "Microsoft Gold Certified Partner", and cites a SIS representative saying there were eleven members in late June -- and that the big influx started on Thursday (two work days before the vote?!).
It's probably better to simply getenv("HOME") to find $HOME. It's already in your process; why bother asking a file (or possibly a remote database)?
But yes, there are other mundane and valid reasons for reading /etc/passwd.
Just as long as (s)he isn't spending an hour every night looking at girls, or talking to porn.
And all four responded quickly with proper, readable NDRs. One large Microsoft-dominated tech company, one mid-sized Microsoft-dominated tech company, my sucky ex-ISP, and the server in my closet.
Except a backup MX without access to an up-to-date copy of the full user database and rewriting rules of the real MX that does local delivery. Remember that a backup MX can (probably should) be far from the primary: it is supposed to kick in when the primary has lost network, suffered a disk crash, or is down for any kind of maintenance.
Yes. I have mailed people and become seriously offended when the don't reply. That's because I can assume that when I get no feedback, it's the recipient who blew it - by not reading his mail, or not bothering to respond.
Like others have noted, NDR is an important part of the semantics of Internet mail.
So now I'm sending mail to , for $foo in some of the domains I use frequently. I hope they'll write me back ...
I realize that this is just background to the question, but it strikes me as an odd thing so say:
I have been the victim of numerous really crappy internal applications. It makes me mad, because it shows a lack of respect for my work. What tends to happen in practice, if the users are technically minded, is that they don't use the applications as intended, and invent some primitive system on the side. Excel sheets, and so on.
And the people who support the real system are usually happily unaware of this. They live in a fantasy land, where things work fairly well and the users are pleased.
This is offtopic but ... it should be obvious that you do not deserve being raped
just because you are sent to prison.
Ok, it was a joke. It's just that it isn't just a joke.
Whatever the cause is, it isn't size and scope. Practically any piece of free software compiles on a dozen architectures. For example, Debian Gnu/Linux ships around thirteen gigabytes of software for each of eleven architectures ...
printf() is nice -- as long as you know at compile time the types of all arguments, and never print any types except the handful which printf() supports ...
For you. For normal systems, "Hello, world!" is more like the 5344 bytes I got when I compiled it on my Debian machine.
Or 4896 bytes on my ppc machine. Hey, that must mean the PowerPC architecture is 8.4% better than AMD64!
But seriously -- what those 256 kilobytes mean is simply that noone has bothered to optimize the linker and standard library of your toolchain for iostreams. Possibly because too many people think they are too large and don't use them.
You miss the point. It has nothing to do with bloody software engineering. It's about being curious about how computers work, and about how some hacker's mind worked back then. It's about how cool it is that this FORTRAN gibberish translates into a really good game. You were excited about all that when you got your first computer, weren't you?
Funny you should say that -- unlike adventure, Nethack has all those features (ASCII animations, ASCII graphics and -- in at least the Amiga port -- sound).
It triggers me on two other points too:
But I have the feeling we have all been trolled.
Tree is hardly a developer tool, is it? But yes, it's sometimes useful and I install it on all my machines.
But if you had lived in America instead, don't you think they would have told you you were kind of english?
No ... Rather, it's hard for the programmer himself to know how good he is.
You can observe things like
"I'm better at producing solid code than that guy"
or
"that guy botched that task which I found simple"
or
"people seem to trust my abilities"
or
"I failed to understand the Linux networking code"
but you can't get meaningful measurements.
(You can get meaningless measurements; those are the ones which end up on your CV.)
Also, they do simplistic things with their internet connection because they are behind NAT. Countless cool technologies and ideas have died, just because NAT was in the way.
Hell yes.
I still don't understand why His Majesty the King of Sweden won't sign my GnuPG key. Or some equivalent national system.
There's something like it in .se: "E-legitimation".
A certificate you can order from the major banks.
However, it appears to be somewhat Windows-specific, and you can only use it to authenticate yourself to the major banks and the Government, i.e. The Man. As far as I can see, it's useless for establishing trust between two humans.
All this is, of course, completely orthogonal to the Internet.