I am not falling into a trap, because I know that C is not object-oriented!
You'll note that not once did I claim that C was object oriented -- it blatantly
isn't. What I did claim was that gtk+ was object oriented,
as others have pointed out in this thread.
Most people
don't need to pipe windows through the net.
Maybe so, maybe not. But what about those of us that do?
Networking is the biggest strength of X as far as I'm concerned.
I couldn't do my job without it. I support an application running on a server
in the USA from here in the UK. More mundanely, it
lets me run apps on a server in our machine room, and display
them on my desktop machine. I do this all day, every day.
You may be correct in saying that most home users don't need X to have networking capabilities, but corporate users certainly do...
I'm well aware of that fact. KMenuBar is implemented using QMenuBar. Thus a KDE app has
two ways of putting up menus. This is exactly what the original editorial was complaining about,
and is the point I was trying to make. To complain that gtk+ and gnome
both have ways to display a menu is a little unfair, given that Qt and KDE do the same.
Actually, as far as I'm aware, Gnome provides no way to display a normal menu of its own (I can't speak for Bonobo), although
it does provide a way to do popup menus (which funnily enough are
implemented using the gtk+ GtkMenu object). Mmmmm, don't you just
love that code reuse that gtk+ supposedly doesn't have...
more KDE apps are developed more quickly
because QT is C++.
Really? I've been programming for nearly 2 decades now, and I've yet to see any
proof that C++ leads to quicker development, despite what it's advocates may claim. Certainly, I see more gtk+ apps being developed than Qt apps.
In an object-oriented system, you can
reuse menu classes -- i.e., you don't have to reinvent the menu implementation with each library spin off.
Indeed. You seem to be falling into the trap, though, of thinking that gtk+ isn't object oriented, just because it's written in C. In fact, it is.
And it has a nice, consistent menu implementation, too. Just call
gtk_menu_new(), add the menu items you require, and plonk it into your application. Simple, really. Just like using the QMenuBar class in Qt. Speaking of which, if your wonderful object oriented system lets you reuse code, rather than reimplement it, why does KDE choose to define its own KMenuBar class, rather than just using the default Qt one?
If I were in his shoes I'd
like to be on record as opposing the migration.
Indeed. That's something that I forgot to say. Put your objections
to management in writing. That way, you know that management
are aware of the problems with their chosen route, and
furthermore, you can prove it. If you have an email system
that supports delivery notification, use that -- and don't
forget to cc a copy to an external address too. I've seen
email systems that conveniently managed to "lose" potentially damaging messages...
It is a bit biased, but the author makes good
points.
It's more than a bit biased. It's utterly, completely
and unfairly biased. It has numerous digs at GNOME, all
conveniently attributed to anonymous sources. It's low on facts,
and its only purpose can be flamebait. And at that, I'm sure
it will succeed. I have never before criticised/. for posting a story,
but there has to be a first tiem for everything. This story crossed the line.
Flame wars don't achieve anything, other than lots of page view. Maybe the conspirancy theorists are right about/. trying to get more ad revenue...
but the point is that the
direction here is likely to be totally beyond our control
And therein lies the problem. Management need to be made forcefully aware
that the agency is not a Windows only shop, and
that proposing Windows only solutions like this is a road to ruin. Sure, you may only be
a minority, but they need to know that you cannot integrate with their solution without (at the very least) significant work. The need to know what the impact of alienating your department will be on the agency as a whole.
Like it or not, management are stupid. Sure there are a few exceptions, but on the whole,
it's a good approximation. I once worked at a company where management decreed that all corporate email should be handled by exchange and outlook.
Only after buying the servers, and doing an initial roll out to some PCs did they realise that 30% of the desktops ran SunOS or Solaris on Sparc hardware... Management don't understand technological issues like these, and they need to have them explained.
And of course, there was Data General's DG/UX for the Motorola 88K series of RISC
processors.
Actually, these days it runs on Intel CPUs. Although
they continue to maintain the m88k version, they haven't
sold any 88k based machines for many years now.
It's actually one of my favourite Unices. It sucked quite badly in early versions, but
later ones are much better. Interestingly, it's the only Unix version
I know of (other than Linux) that doesn't originate from "real" Unix. The kernel was
rewritten from scratch to conform with the specs. It contains none of the original Unix code. The userland was all licensed from SVR4, though.
Why are all these companies changing their names to dances?
Well the SCO name change actually makes complete sense.
Given that they've sold off the Unix business, the only product they have
left is Tarantella, so they may as well name the
company after it. The SCO brand name only has recognition in the
Unix world anyway.
The one I really don't understand is Scriptics, the company in charge of the
Tcl scripting language, changing its name to Ajuba Solutions. Nope, don't get that one. Even if they want to distance themselves from being purely a Tcl company, the new name sucks!
This solves one of my main objections
to traditional laptops -- that they don't have a suitable
graphics chip to drive an external monitor at a decent
resolution (read 1600x1200). I still don't feel a 15" laptop is
sufficiently portable to be worth the effort, but it's a good sign for
the future.
Doom was Wolfenstein with a lot more of the color red used.
In one sense, yes. I thought Wolfenstein 3D was an
amazing game at the time. But the superiority of Doom's gameplay
is fairly evident -- today, over 6 years from its original release, I still go back
and play the odd game of Doom now and again. I can't say the same about
Wolf3D, no matter how much I liked it at the time...
The Sun website
talks glowingly of all the really cool things they will do with Gnome... but those with a
memory (and a web browser pointed towards the Open Group's website) will remember that Sun
said pretty much the same thing for Motif/CDE.. and look where that went.
That's precisely why I'm not entirely convinced by the GNOME Foundation.
The took HP's (already grim) VUE desktop and threw in a bunch of
designed-by-committee stuff to end up with the nightmare that was CDE.
I'm just worried GNOME is going to be heading in the same direction.
I presonally couldn't care less about KDE. I tried both, and absolutely hated KDE. It's just not for me. I'm not convinced that GNOME is either, but time will tell on that one. I wish both groups luck, but I remain unconvinced that I'll be using either of their products in the future.
You've got to admit that the gameplay in something like Quake III Arena (single player,
even) is a whole lot better than from Doom.
Actually, no. In fact, quite the opposite. Doom was pretty much near
the top of the pile in terms of gameplay. Sure, Q3A is by far the best
thing Id have done since Doom, but even then, it's lacking that sense
of urgency that Doom had. But then, that's mostly due to the difference between
having 3 beautifully details polygon models on screen in Q3A compared to the
20 enemies all wanting to kill you that you found in Doom.
games like Civ:CTP && Diablo II blow the old turn-based strategy games
I'm constantly stunned at how many people like the Diablo games. They're graphical rip-offs of the roguelike genre (particularly angband), but without any of depth or the playability.
:-) I can't recall the last time arcade games were
that cheap here in the UK. Most seem to be £1 a go, but I've seen some
at up to £3 a go (that's around $5 US).
Sadly, all the new arcade games are crap anyway, so there's
little incentive to pay the high prices. There is, however, a growing
trend towards having retro machines in arcades, and they're typically priced
at 10-20p (that's about 16-32) per game. I was playing an original space invaders machine
not too long ago (although sadly t had lost its tinted screen overlay). It brought back some wonderful memories...
That'd be me, and pretty much all of my computer
geek friends. Even some of my non-geek friends
have a home network for multi-player gaming under
Win32.
I couldn't use the Tesco Direct site even with Netscape.
I send details to their tech support, explaining the problem,
and they have completely ignored it. Consequently, I don't
go there any more. Annoying, but then life sometime is...
under the usual criteria for judging server OSes (scalability, reliability,
security) Linux is not a particulary good server OS
I'd have to disagree with you there. Yes, NT fails miserably on all of
those, which makes Linux look good. However, that doesn't
mean Linux is lacking in those areas. In fact, it does pretty well
for all of them. I'm not claiming it's the best in any area,
but it's far from the worst. It doesn't scale to the high end as well as
Solaris, DG/UX or Irix, for example. Nor does it have the reliability of Tandem or the security of the various MAC-enabled OSes.
However, for probably 80% of businesses, it's a suitable server OS for their needs. It has sufficient scalability, reliability and security to get the job done.
Had a VAX account once a long time back, but I'd
already seen Unix and wasn't impressed.
While it's true that VAX and VMS have gone together
for a long time, everyone seems to forget that Unix ran on
VAX machines in days of old. Thus the phrase "a VAX account"
doesn't actually mean much...
For changing a nameserver, for example, you could
just rdist or rsync/etc/resolv.conf to all the machines.
However, it gets more complex when you want to modify a file
rather than overwrite it. To change the gateway, as in
your example, you'll want to modify the GATEWAY parameter in/etc/sysconfig/network, while leaving the HOSTNAME parameter untouched.
The only practical way to do this is to write a script, but
then you're doing that already. Writing scripts is The Unix Way (tm):-)
wotsit.org has info on lots of file formats, including JPEG.
Of course, JPEG isn't a file format but an encoding algorithm.
There are currently two main file formats for distributing
encoded JPEG data streams: JFIF and TIFF. Fortunately, wotsit.org
contains both specs for the file formats and specs
for the encoding algorithm. This message was brought to you
courtesy of Pedants-R-Us (tm).
MacOS X server certainly has it's work cut out if it's going to succeed.
It has two main problems. First is lack of applications. It's in a similar
position that of Linux a few years ago. I expect Jobs to push Ellison
quite hard to get an Oracle port. Without that, big businesses just aren't
going to take it seriously for heavy duty server use. Being BSD
based, we can assume ports of Apache and Samba will appear shortly
(if they haven't already), but without a heavy duty database (and in the minds of
most decision makers, that means Oracle, rightly or wrongly), it'll remain as a niche file/print/web server.
Secondly, it has to show significant advantages over the competition, and thanks to Linux, the competition is in pretty
good shape at the moment. I doubt OS X will win on performance, and it can't win on cost, so it has to rely on other areas if it's to succeed. The only way I can see it doing well is if it comes with some stunning easy to use remote management software. Getting the UI right has long been one of Apple's strong points, so perhaps it has a chance after all, but it's going to be a long hard struggle...
Everything has its purpose right? Linux's purpose is to be the
best server operating system available
Actually no. Linux's sole purpose is to give Linus Torvalds a
Unix that he can run on his PC. That's why it was born,
and any other uses are purely coincidental. As it happens, Linux is a pretty
good server OS. It is also (and here's the controversial bit)
the best desktop OS in the world. No, I'm not on drugs -- I seriously believe
that it is the best desktop OS in the world for me. It does pretty much everything I want, and it does it better than all the alternatives I've tried. I'm not foolish enough to claim that it's for everyone yet, but given time it will get there.
The features that make it a good server OS don't preclude desktop use. Or are you going to try and claim that Win9x is a better desktop OS than NT4 or Win2K?
You might want to mention that he said this in 1996.
If you read my post, you'll see that I wrote "a few years ago".
That tends to imply at least 3, so I think it covers 1996
(i.e. 4 years ago) pretty well...
Try carrying that desktop somewhere on a trip (business or otherwise...)
Try using that desktop on a plane.
You're missing my point. What I said was that a traditional
laptop is poorly suited to being either portable or
useful on the desktop. For mobile computing, something
like a Libretto or a Picturebook Vaio is much better
than a full sized laptop. Similarly, a desktop is better for when you're
not travelling.
Slashdot Mirror
You'll note that not once did I claim that C was object oriented -- it blatantly isn't. What I did claim was that gtk+ was object oriented, as others have pointed out in this thread.
Maybe so, maybe not. But what about those of us that do? Networking is the biggest strength of X as far as I'm concerned. I couldn't do my job without it. I support an application running on a server in the USA from here in the UK. More mundanely, it lets me run apps on a server in our machine room, and display them on my desktop machine. I do this all day, every day. You may be correct in saying that most home users don't need X to have networking capabilities, but corporate users certainly do...
I'm well aware of that fact. KMenuBar is implemented using QMenuBar. Thus a KDE app has two ways of putting up menus. This is exactly what the original editorial was complaining about, and is the point I was trying to make. To complain that gtk+ and gnome both have ways to display a menu is a little unfair, given that Qt and KDE do the same. Actually, as far as I'm aware, Gnome provides no way to display a normal menu of its own (I can't speak for Bonobo), although it does provide a way to do popup menus (which funnily enough are implemented using the gtk+ GtkMenu object). Mmmmm, don't you just love that code reuse that gtk+ supposedly doesn't have...
Really? I've been programming for nearly 2 decades now, and I've yet to see any proof that C++ leads to quicker development, despite what it's advocates may claim. Certainly, I see more gtk+ apps being developed than Qt apps.
In an object-oriented system, you can reuse menu classes -- i.e., you don't have to reinvent the menu implementation with each library spin off.
Indeed. You seem to be falling into the trap, though, of thinking that gtk+ isn't object oriented, just because it's written in C. In fact, it is. And it has a nice, consistent menu implementation, too. Just call gtk_menu_new(), add the menu items you require, and plonk it into your application. Simple, really. Just like using the QMenuBar class in Qt. Speaking of which, if your wonderful object oriented system lets you reuse code, rather than reimplement it, why does KDE choose to define its own KMenuBar class, rather than just using the default Qt one?
Indeed. That's something that I forgot to say. Put your objections to management in writing. That way, you know that management are aware of the problems with their chosen route, and furthermore, you can prove it. If you have an email system that supports delivery notification, use that -- and don't forget to cc a copy to an external address too. I've seen email systems that conveniently managed to "lose" potentially damaging messages...
It's more than a bit biased. It's utterly, completely and unfairly biased. It has numerous digs at GNOME, all conveniently attributed to anonymous sources. It's low on facts, and its only purpose can be flamebait. And at that, I'm sure it will succeed. I have never before criticised /. for posting a story,
but there has to be a first tiem for everything. This story crossed the line.
Flame wars don't achieve anything, other than lots of page view. Maybe the conspirancy theorists are right about /. trying to get more ad revenue...
And therein lies the problem. Management need to be made forcefully aware that the agency is not a Windows only shop, and that proposing Windows only solutions like this is a road to ruin. Sure, you may only be a minority, but they need to know that you cannot integrate with their solution without (at the very least) significant work. The need to know what the impact of alienating your department will be on the agency as a whole. Like it or not, management are stupid. Sure there are a few exceptions, but on the whole, it's a good approximation. I once worked at a company where management decreed that all corporate email should be handled by exchange and outlook. Only after buying the servers, and doing an initial roll out to some PCs did they realise that 30% of the desktops ran SunOS or Solaris on Sparc hardware... Management don't understand technological issues like these, and they need to have them explained.
Actually, these days it runs on Intel CPUs. Although they continue to maintain the m88k version, they haven't sold any 88k based machines for many years now. It's actually one of my favourite Unices. It sucked quite badly in early versions, but later ones are much better. Interestingly, it's the only Unix version I know of (other than Linux) that doesn't originate from "real" Unix. The kernel was rewritten from scratch to conform with the specs. It contains none of the original Unix code. The userland was all licensed from SVR4, though.
Well the SCO name change actually makes complete sense. Given that they've sold off the Unix business, the only product they have left is Tarantella, so they may as well name the company after it. The SCO brand name only has recognition in the Unix world anyway.
The one I really don't understand is Scriptics, the company in charge of the Tcl scripting language, changing its name to Ajuba Solutions. Nope, don't get that one. Even if they want to distance themselves from being purely a Tcl company, the new name sucks!
This solves one of my main objections to traditional laptops -- that they don't have a suitable graphics chip to drive an external monitor at a decent resolution (read 1600x1200). I still don't feel a 15" laptop is sufficiently portable to be worth the effort, but it's a good sign for the future.
In one sense, yes. I thought Wolfenstein 3D was an amazing game at the time. But the superiority of Doom's gameplay is fairly evident -- today, over 6 years from its original release, I still go back and play the odd game of Doom now and again. I can't say the same about Wolf3D, no matter how much I liked it at the time...
That's precisely why I'm not entirely convinced by the GNOME Foundation. The took HP's (already grim) VUE desktop and threw in a bunch of designed-by-committee stuff to end up with the nightmare that was CDE. I'm just worried GNOME is going to be heading in the same direction. I presonally couldn't care less about KDE. I tried both, and absolutely hated KDE. It's just not for me. I'm not convinced that GNOME is either, but time will tell on that one. I wish both groups luck, but I remain unconvinced that I'll be using either of their products in the future.
Actually, no. In fact, quite the opposite. Doom was pretty much near the top of the pile in terms of gameplay. Sure, Q3A is by far the best thing Id have done since Doom, but even then, it's lacking that sense of urgency that Doom had. But then, that's mostly due to the difference between having 3 beautifully details polygon models on screen in Q3A compared to the 20 enemies all wanting to kill you that you found in Doom.
games like Civ:CTP && Diablo II blow the old turn-based strategy games
I'm constantly stunned at how many people like the Diablo games. They're graphical rip-offs of the roguelike genre (particularly angband), but without any of depth or the playability.
That'd be me, and pretty much all of my computer geek friends. Even some of my non-geek friends have a home network for multi-player gaming under Win32.
I couldn't use the Tesco Direct site even with Netscape. I send details to their tech support, explaining the problem, and they have completely ignored it. Consequently, I don't go there any more. Annoying, but then life sometime is...
Hmmm. Doesn't sound like much of a party. Get thee some rock music! :-)
I'd have to disagree with you there. Yes, NT fails miserably on all of those, which makes Linux look good. However, that doesn't mean Linux is lacking in those areas. In fact, it does pretty well for all of them. I'm not claiming it's the best in any area, but it's far from the worst. It doesn't scale to the high end as well as Solaris, DG/UX or Irix, for example. Nor does it have the reliability of Tandem or the security of the various MAC-enabled OSes. However, for probably 80% of businesses, it's a suitable server OS for their needs. It has sufficient scalability, reliability and security to get the job done.
While it's true that VAX and VMS have gone together for a long time, everyone seems to forget that Unix ran on VAX machines in days of old. Thus the phrase "a VAX account" doesn't actually mean much...
For changing a nameserver, for example, you could just rdist or rsync /etc/resolv.conf to all the machines.
However, it gets more complex when you want to modify a file
rather than overwrite it. To change the gateway, as in
your example, you'll want to modify the GATEWAY parameter in /etc/sysconfig/network, while leaving the HOSTNAME parameter untouched.
The only practical way to do this is to write a script, but
then you're doing that already. Writing scripts is The Unix Way (tm) :-)
Of course, JPEG isn't a file format but an encoding algorithm. There are currently two main file formats for distributing encoded JPEG data streams: JFIF and TIFF. Fortunately, wotsit.org contains both specs for the file formats and specs for the encoding algorithm. This message was brought to you courtesy of Pedants-R-Us (tm).
MacOS X server certainly has it's work cut out if it's going to succeed. It has two main problems. First is lack of applications. It's in a similar position that of Linux a few years ago. I expect Jobs to push Ellison quite hard to get an Oracle port. Without that, big businesses just aren't going to take it seriously for heavy duty server use. Being BSD based, we can assume ports of Apache and Samba will appear shortly (if they haven't already), but without a heavy duty database (and in the minds of most decision makers, that means Oracle, rightly or wrongly), it'll remain as a niche file/print/web server. Secondly, it has to show significant advantages over the competition, and thanks to Linux, the competition is in pretty good shape at the moment. I doubt OS X will win on performance, and it can't win on cost, so it has to rely on other areas if it's to succeed. The only way I can see it doing well is if it comes with some stunning easy to use remote management software. Getting the UI right has long been one of Apple's strong points, so perhaps it has a chance after all, but it's going to be a long hard struggle...
Actually no. Linux's sole purpose is to give Linus Torvalds a Unix that he can run on his PC. That's why it was born, and any other uses are purely coincidental. As it happens, Linux is a pretty good server OS. It is also (and here's the controversial bit) the best desktop OS in the world. No, I'm not on drugs -- I seriously believe that it is the best desktop OS in the world for me. It does pretty much everything I want, and it does it better than all the alternatives I've tried. I'm not foolish enough to claim that it's for everyone yet, but given time it will get there. The features that make it a good server OS don't preclude desktop use. Or are you going to try and claim that Win9x is a better desktop OS than NT4 or Win2K?
If you read my post, you'll see that I wrote "a few years ago". That tends to imply at least 3, so I think it covers 1996 (i.e. 4 years ago) pretty well...
You're missing my point. What I said was that a traditional laptop is poorly suited to being either portable or useful on the desktop. For mobile computing, something like a Libretto or a Picturebook Vaio is much better than a full sized laptop. Similarly, a desktop is better for when you're not travelling.