First of all, Arthur wasn't the descendant of Khan, the guy overseeing the demolition of his house is.
In fact, I'm referring to Mr. Prosser, who is convinced by Ford to lie in the mud in place of Arthur to block bulldozers, so Ford and Arthur may go to the pub.
1) Do you have a predilection for little fur hats?
2) Do you prefer axes to climbing roses on the door of your cottage?
3) Have you ever suffered of inexplicable but terrbly attractive visions of houses consumed by fire?
4) Have you ever felt your brain filled with a thousand hairy horsemen shouting at you?
If the answer is "yes" to all questions, you are a Genghis Khan descendant, unfortunately you are too busy lying in the mud to go to the pub -- pardon, restaurant.
As for "stored procedures, views, triggers,..." these fall under the category of --USELESS FEATURES--.
...useless until you realize you have to connect to the DB from several applications written in several different languages for which you have either to reimplement your way to manipulate the data everywhere, or you have to put in a middle layer of some sort which is able to talk multiple languages (via CORBA, SOAP, plain XML RPC, custom protocol, whatever) and ensure that everyone is accessing data exclusively through it.
Reimplementing logic everywhere across different languages is usually a bad approach because it doubles the development and testing effort.
Middle layers are usually hard to get right the first time, much harder than using stored procedures and triggers, since the typical procedural language is not so at ease at manipulating relational data.
On MySQL: it has a somewhat bad reputation in the field because of the people abusing it: it's fast, it's free, it's easy to set up, but living with these gotchas is definitively too painful for developers more concerned about correctness than speed.
In the end: firing up Firebird/Oracle/PostgreSQL/SAPDB for simple data is plainly stupid, but often it is done anyways since they do a decent job even in that cases; firing up MySQL for your 30+ GB DB containing your whole network topology which is used by everyone for billing, service assurance, troubleshooting, network planning and whatever is stupid as well. MySQL AB knows that, and in fact it now proposes MaxDB (was: SAPDB) as well.
Hint: in absence of a single authority having the power to dictate how thing should be done "OR ELSE", this is the only sane way to do it: define reasonable guidelines and tell developers about them: many will follow.
we could soon have X servers everywhere, so that we could run applications from POSIX servers whenever HTTP didn't suffice.
Uhm, X protocol is quite "chatty". NoMachine MX literally does miracles to circumvent network latency problems, still I think this makes X protocol merely usable on WANs and MANs (i.e. the Internet), not "a good choice".
OTOH, X protocol is just fine on LANs.
Re:Most important technology not on the roadmap?
on
The GNOME Roadmap
·
· Score: 2, Informative
What about the vector graphics plans?
Is a SVG based window manager so far away?
On Windows, I simply up the screen size by changing from 1024 768 to 800 600. (I wished linux could do this.)
You have two options:
Ctrl+Alt+ plus/minus on the numerical keypad, to switch between video modes. It doesn't resize your desktop, but it offers an enlarged view that you can scroll with the mouse pointer. It has been in XFree86 since day zero.
XFree86 4.3 introduces the RandR extension, allowing both to change the video mode AND the desktop size, effectively changing resolution on the fly. There's a simple applet for Gnome 2 (it's gnome-randr-applet on Debian unstable) that offers access to that, don't know about KDE.
Doesn't the GPL say you cannot discriminate against any group?
Yes.
Or is their license being revoked because they are in violation of the GPL?
IANAL, but you can't violate the GPL: you can only accept it or not.
If you adhere to the GPL requests, you automatically obtain permission to redistribute a GPL'd workd, without even emailing the copyright holders. It is basically the easiest way to obtain permission.
If you don't accept it, and you have no other permissions to distribute said work (for example, by contracting directly with the copyright holders), copyright law states that you are not allowed to redistribute (except in case of fair use?)
If you distribute something without permission from the copyright holders, you are violating copyright law.
IMHO, Fyodor can't revoke anything because there is nothing that can be revoked in the first place.
Just s/GNOME/Gimp/g, and that's what's actually happened (except that The Gimp is quite useful).
Early versions of The Gimp were Motif-based (and useful enough to draw Tux), then people decided to write their own Motif replacement, and thus GTK was born.
AFAIK, the copyright on Tux belongs to its author, Larry Lewing (he doesn't explicitly states that, but art works are copyrighted by default, right?), who then granted permission to everyone to use/modify it, provided he and The Gimp are acknowledged.
Apparently not, the translation contains errors, but it attempts to be faithful: I may assure that the Italian version sounds pleasingly clear as well (which is not bad for a language where the same word is used both for "advanced" and for "leftover":-).
And yes, that's exactly the word being used: "ridiculous" (about shrink-wrap notices). Let's hope the meme is spread among other judges.
The copyright holder legally holds all the rights on their work, and they are (and should be!) free to restrict how you use their work.
Are you saying that because there's a law saying so, or do you actually believe there's some sort of "natural right" for anybody to dictate the fate of their work after it gets into the public's hands?
If the latter, do you realize that said law was put there with the intent to stimulate authors to actually produce new works and distribute them to the public by temporarily granting them a right that otherwise they wouldn't have at all, and that right now is used exactly for the opposite (redistributing the same content over and over for decades)?
The Gimp has had for some time (since version 1.2 IIRC) some support for vectorial drawing: you can define paths using bezier curves, which may be adjusted, saved and restored, and drawn on the current layer using the current brush. But drawing (and selecting the layer) must be done manually.
The next version of The Gimp adds the ability to save and restore paths as SVG paths (before, it used an ad-hoc simple textual format), and also the ability to import an SVG image by rendering it on a bitmap (like it did with PS images).
That's it: a useful thing to have, but it has little to do with vectorial drawing.
There was a GNU project (which apparently failed) that was trying to create a vector art authoring tool. I can't remember the name of it.
You are talking about GYVE: its developement has stopped in 2002.
OTOH, for Free vectorial drawing programs, check out sodipodi (and its IMHO nicer branch Inkscape) and the good ol' Sketch (now called Skencil).
One of the nicest aspects of Tcl is that is it seriously multi-platform.
Another nice aspect of Tcl is that it can easily evaluate code in a different stack frame (example: in the caller's context), and source code can be easily passed to procedures as strings between braces (as it is usually done), so extending/reimplementing the language control structures is as simple as writing a new procedure (and no special/ugly syntax is required).
My experience is that government clerks are not the brightest users anyway and they tend to use a limited range of features they have been tought.
Even the brightest users tend to use just a small set of all available features, it's just they use a completely different set of features.
If you want to please them all, a product should have lots and lots of features. Unfortunately, trying to please everyone usually results in pissing off everyone...
You might as well, at that point, make bash setuid
Just a note: making Bash suid root won't work: if the effective user ID (the one affected by the setuid bit) is 0 (read: root), Bash simply resets the effective user ID to the real user ID (the one inherithed from the parent process). Other interpreters probably do that as well.
OTOH, making Bash setuid any other user works as expected.
Of course this doesn't prevent a suid root wrapper to change its real user id before forking a shell (otherwise su, sudo and friends couldn't work...)
Re:Why not interpreted C++, instead?
on
Perl is Sweet Sixteen
·
· Score: 3, Informative
C++ didn't even exist when Perl was first invented.
According to wikipedia, C++ dates back to 1979, with the first commercial compiler in 1985.
Slashdot Mirror
Well, GStreamer is steadily getting there (this project deserves more awareness also among end users, so that's why I keep posting about it).
Well, GStreamer is probably going to be it.
First of all, Arthur wasn't the descendant of Khan, the guy overseeing the demolition of his house is.
In fact, I'm referring to Mr. Prosser, who is convinced by Ford to lie in the mud in place of Arthur to block bulldozers, so Ford and Arthur may go to the pub.
1) Do you have a predilection for little fur hats?
2) Do you prefer axes to climbing roses on the door of your cottage?
3) Have you ever suffered of inexplicable but terrbly attractive visions of houses consumed by fire?
4) Have you ever felt your brain filled with a thousand hairy horsemen shouting at you?
If the answer is "yes" to all questions, you are a Genghis Khan descendant, unfortunately you are too busy lying in the mud to go to the pub -- pardon, restaurant.
...useless until you realize you have to connect to the DB from several applications written in several different languages for which you have either to reimplement your way to manipulate the data everywhere, or you have to put in a middle layer of some sort which is able to talk multiple languages (via CORBA, SOAP, plain XML RPC, custom protocol, whatever) and ensure that everyone is accessing data exclusively through it.
Reimplementing logic everywhere across different languages is usually a bad approach because it doubles the development and testing effort.
Middle layers are usually hard to get right the first time, much harder than using stored procedures and triggers, since the typical procedural language is not so at ease at manipulating relational data.
On MySQL: it has a somewhat bad reputation in the field because of the people abusing it: it's fast, it's free, it's easy to set up, but living with these gotchas is definitively too painful for developers more concerned about correctness than speed.
In the end: firing up Firebird/Oracle/PostgreSQL/SAPDB for simple data is plainly stupid, but often it is done anyways since they do a decent job even in that cases; firing up MySQL for your 30+ GB DB containing your whole network topology which is used by everyone for billing, service assurance, troubleshooting, network planning and whatever is stupid as well. MySQL AB knows that, and in fact it now proposes MaxDB (was: SAPDB) as well.
Well, the GNOME Human Interface gudelines and the KDE User Interface Guidelines are there exactly for this. GNUStep probably uses the OpenStep ones, since it is an OpenStep replica.
Hint: in absence of a single authority having the power to dictate how thing should be done "OR ELSE", this is the only sane way to do it: define reasonable guidelines and tell developers about them: many will follow.
Conectiva also made a damn fine graphical front-end to apt-get, which works both with Conectiva and Debian: Synaptic.
we could soon have X servers everywhere, so that we could run applications from POSIX servers whenever HTTP didn't suffice.
Uhm, X protocol is quite "chatty". NoMachine MX literally does miracles to circumvent network latency problems, still I think this makes X protocol merely usable on WANs and MANs (i.e. the Internet), not "a good choice".
OTOH, X protocol is just fine on LANs.
Well, you asked for it.
No problem with that. The point is: can you be sued over trademark infringiment just because people call it the way they want? Probably not. QED.
As has been said one million times...
IIRC, X dates back to mid '87 (perhaps earlier, but not that much)
You have two options:
Yes.
Or is their license being revoked because they are in violation of the GPL?
IANAL, but you can't violate the GPL: you can only accept it or not.
IMHO, Fyodor can't revoke anything because there is nothing that can be revoked in the first place.
Ok, but the real problem is not what, it is why.
Early versions of The Gimp were Motif-based (and useful enough to draw Tux), then people decided to write their own Motif replacement, and thus GTK was born.
Pardon for the typo, the correct name is Larry Ewing.
AFAIK, the copyright on Tux belongs to its author, Larry Lewing (he doesn't explicitly states that, but art works are copyrighted by default, right?), who then granted permission to everyone to use/modify it, provided he and The Gimp are acknowledged.
Tim Berners-Lee got the knighthood less than a month ago.
Apparently not, the translation contains errors, but it attempts to be faithful: I may assure that the Italian version sounds pleasingly clear as well (which is not bad for a language where the same word is used both for "advanced" and for "leftover" :-).
And yes, that's exactly the word being used: "ridiculous" (about shrink-wrap notices). Let's hope the meme is spread among other judges.
Are you saying that because there's a law saying so, or do you actually believe there's some sort of "natural right" for anybody to dictate the fate of their work after it gets into the public's hands?
If the latter, do you realize that said law was put there with the intent to stimulate authors to actually produce new works and distribute them to the public by temporarily granting them a right that otherwise they wouldn't have at all, and that right now is used exactly for the opposite (redistributing the same content over and over for decades)?
Yes and no.
The Gimp has had for some time (since version 1.2 IIRC) some support for vectorial drawing: you can define paths using bezier curves, which may be adjusted, saved and restored, and drawn on the current layer using the current brush. But drawing (and selecting the layer) must be done manually.
The next version of The Gimp adds the ability to save and restore paths as SVG paths (before, it used an ad-hoc simple textual format), and also the ability to import an SVG image by rendering it on a bitmap (like it did with PS images).
That's it: a useful thing to have, but it has little to do with vectorial drawing.
There was a GNU project (which apparently failed) that was trying to create a vector art authoring tool. I can't remember the name of it.
You are talking about GYVE: its developement has stopped in 2002.
OTOH, for Free vectorial drawing programs, check out sodipodi (and its IMHO nicer branch Inkscape) and the good ol' Sketch (now called Skencil).
Another nice aspect of Tcl is that it can easily evaluate code in a different stack frame (example: in the caller's context), and source code can be easily passed to procedures as strings between braces (as it is usually done), so extending/reimplementing the language control structures is as simple as writing a new procedure (and no special/ugly syntax is required).
That's as close to Lisp macros as you can get.
Even the brightest users tend to use just a small set of all available features, it's just they use a completely different set of features.
If you want to please them all, a product should have lots and lots of features. Unfortunately, trying to please everyone usually results in pissing off everyone...
Just a note: making Bash suid root won't work: if the effective user ID (the one affected by the setuid bit) is 0 (read: root), Bash simply resets the effective user ID to the real user ID (the one inherithed from the parent process). Other interpreters probably do that as well.
OTOH, making Bash setuid any other user works as expected.
Of course this doesn't prevent a suid root wrapper to change its real user id before forking a shell (otherwise su, sudo and friends couldn't work...)
According to wikipedia, C++ dates back to 1979, with the first commercial compiler in 1985.
C++ is a terrible language for interpretation
Right. At least someone tried
Better to have a language which is designed for interpretation from the start.
Completely agree.