1. PowerPlant is complicated in comparison to Objective C - Cocoa. Granted this is somewhat subjective. Also, most people that have used GUI Frameworks will find PowerPlant more than a bit foreign. PowerPlant's use of mix and match classes is quite a bit different than coding in GTK+ or Motif. Not worse, just different and that difference is a bit of a hurdle. I agree that the using the Mac's old native toolbox feels like going back to assembler.
2. I am looking forward to using Cocoa. PowerPlant and Carbon are another option, but at this point going this way is more "beta" than going with Cocoa, which is well established.
3. I agree that Java is an option for lightweight apps. But the key word is _option_. There is still a speed issue with larger apps. Apple still makes use of Object C internally for a reason.
On a more person note, having used PowerPlant and native toolkit and Motif and GTK, I think it is going to be fun to mess around with Objective C and the old Next APIs (Cocoa).
Good God. Object C is not a bit deal at all. It is just like C with two additional statements. I don't think it would take any time to pick up. Must less than what it takes to pick up C++. Besides, you can always just code in C++ using the CodeWarrior tools if that is your thing. You would be giving up the Cocoa environment and some of those productivity gains, but you could stick with C++. I've been playing around with CodeWarrior's C++ Framework and I have to say it is pretty damn complicated. Very different for people that are use to Motif or GTK or QT. There is more multiple inheritance in there than you can shake a stick at.
The thing that worries me is that Apple seems to be pushing people to use Java to develop under Cocoa. Now that is not something that I am looking forward too and I would worry about to what extent they are going to support Objective C going forward. One of the negatives about Apple is that no matter what they say you never know where they are going be in two years.
I think that the attraction of OSX to persons with a Unix background stem from Yellow Box programming environment along with the Quartz 2D GUI environment--all of this sitting on top of regular Unix.
In a real sense this solves many of the issues that exist with the current XWindows based GUIs that sit on Unix systems. Would it not be cool to have anti-aliasing and transparancy? Also, the whole display model built on top of EPS is really nice. It opens up a lot of possibilities.
All of this fun GUI stuff reachable via the Objective-C OpenStep APIs. After hearing about the reputation of these APIs, I know that I am anxious to play with them.
And all of this sitting on top of familiar Unix. I am guessing that moving things over and putting a new front end on, while being able to keep much, will be pretty darn attractive.
Now if they could only tone done the interface (the Aqua stuff) and get a little more friendly towards GNUStep (I like this as insurance in case Apple does something crazy, and I'd like to easily move stuff over to other versions of Unix) I'd be pretty happy.
An alternative solution that I believe may be the answer is to create an open community equivalent to the W3C (is that right, the body that maintains the HTML standard, whether or not it is followed, (it's late and my brain is getting foggy)) for office document formats.
This is the only answer to the problem and W3C is a great example. It takes the control away from MicroSoft, a company that uses the spec as a means of driving upgrade sales and maintain their monopoly (the real purpose of.DOC these days) and places control with back with the consumer.
The idea for a common doc format could be marketed successfully based on two points.
First, a common doc format would allow companys and individuals could save large amounts of money by not having to upgrade to the latest verion of Word every two years. This would impact a company's bottom line.
Second, a common doc format would provide companies and individuals with a level of "insurance" that older document types that hold important data would not at some point in the future become obsolete.
Neither of these points even brings up the obvious benefit to the rest of us that use non-MS systems. It would increase competition in the Word Processing arena and would probably move use towards a world where.DOC and.HTML could be interchangeable. Based on the above points, many companies would require that their employees maintain company data using the new open standard.
How is this post informative? It looks like pandering to me.
It is informative because it shows the sort of feelings that SGI is engendering in the Linux community by not playing lip service to Open Source but making real contributions.
Their current strategy is right on. They are the only large company that really understands the whole Open Source thing and they believe that they have found a way to be profitable in the new environment. I think it will work.
Does this mean I would be able to just mount the drive from the 200 onto a Linux (i386) box and pretty much be all set?
I think this is exactly what it means. We are also looking forward to being able to do this. I'd love to buy a lower cost Linux box from SGI (although I can't say enough good things about the Origin.) Or put one together myself.
Re:What window manager are they going to use?
on
SGIs Linux Future
·
· Score: 1
I really hope they port 4dwm and the tools that go along with it. Having that for a desktop is really neat.
You and me both! I have started using Linux at home and it is just not as smooth as the SGI O2 I have at work. Even though my Linux box is two or three times faster, it still is "chunky". The UI is not as smooth or consistent. The more SGI can make Linux look and act like the default IRIX environment, the happier I will be.
The Oregon court has made a very smart, forward looking decision. It is absolutely critical that ISPs have access to both the Cable and DSL infrastructure. Bell and the rest of them dream of the day that they will *force* their cable and dsl subscribers to look at _their_ portal... and _their_ ads. They will force everyone through _their_ proxy server. They will choose what sorts of email servers they will support. And which sort of IP packets they will let through.
It is important that people wake up and realize the value of competition. Persons using dial-up access now have CHOICE. If you don't like your ISP, you can choose one of the other 15 that service your area. This competition has resulted in improved service, as each ISP strives to add the bullet points that another ISP may currently have and to fix problems that may be causing customers to complain.
Local Loop (the physical medium that connects a home or business to an ISP) and the IP Services provided by the ISP are two different animals. Bell is using their monopoly on the one (high speed local loop) to gain a monopoly on the other (IP Services.)
Let's hope that other states follow the lead of Oregon.
Slashdot Mirror
1. PowerPlant is complicated in comparison to Objective C - Cocoa. Granted this is somewhat subjective. Also, most people that have used GUI Frameworks will find PowerPlant more than a bit foreign. PowerPlant's use of mix and match classes is quite a bit different than coding in GTK+ or Motif. Not worse, just different and that difference is a bit of a hurdle. I agree that the using the Mac's old native toolbox feels like going back to assembler.
2. I am looking forward to using Cocoa. PowerPlant and Carbon are another option, but at this point going this way is more "beta" than going with Cocoa, which is well established.
3. I agree that Java is an option for lightweight apps. But the key word is _option_. There is still a speed issue with larger apps. Apple still makes use of Object C internally for a reason.
On a more person note, having used PowerPlant and native toolkit and Motif and GTK, I think it is going to be fun to mess around with Objective C and the old Next APIs (Cocoa).
Good God. Object C is not a bit deal at all. It is just like C with two additional statements. I don't think it would take any time to pick up. Must less than what it takes to pick up C++. Besides, you can always just code in C++ using the CodeWarrior tools if that is your thing. You would be giving up the Cocoa environment and some of those productivity gains, but you could stick with C++. I've been playing around with CodeWarrior's C++ Framework and I have to say it is pretty damn complicated. Very different for people that are use to Motif or GTK or QT. There is more multiple inheritance in there than you can shake a stick at.
The thing that worries me is that Apple seems to be pushing people to use Java to develop under Cocoa. Now that is not something that I am looking forward too and I would worry about to what extent they are going to support Objective C going forward. One of the negatives about Apple is that no matter what they say you never know where they are going be in two years.
I think that the attraction of OSX to persons with a Unix background stem from Yellow Box programming environment along with the Quartz 2D GUI environment--all of this sitting on top of regular Unix.
In a real sense this solves many of the issues that exist with the current XWindows based GUIs that sit on Unix systems. Would it not be cool to have anti-aliasing and transparancy? Also, the whole display model built on top of EPS is really nice. It opens up a lot of possibilities.
All of this fun GUI stuff reachable via the Objective-C OpenStep APIs. After hearing about the reputation of these APIs, I know that I am anxious to play with them.
And all of this sitting on top of familiar Unix. I am guessing that moving things over and putting a new front end on, while being able to keep much, will be pretty darn attractive.
Now if they could only tone done the interface (the Aqua stuff) and get a little more friendly towards GNUStep (I like this as insurance in case Apple does something crazy, and I'd like to easily move stuff over to other versions of Unix) I'd be pretty happy.
An alternative solution that I believe may be the answer is to create an open community equivalent to the W3C (is that right, the body that maintains the HTML standard, whether or not it is followed, (it's late and my brain is getting foggy)) for office document formats.
.DOC these days) and places control with back with the consumer.
.DOC and .HTML could be interchangeable. Based on the above points, many companies would require that their employees maintain company data using the new open standard.
This is the only answer to the problem and W3C is a great example. It takes the control away from MicroSoft, a company that uses the spec as a means of driving upgrade sales and maintain their monopoly (the real purpose of
The idea for a common doc format could be marketed successfully based on two points.
First, a common doc format would allow companys and individuals could save large amounts of money by not having to upgrade to the latest verion of Word every two years. This would impact a company's bottom line.
Second, a common doc format would provide companies and individuals with a level of "insurance" that older document types that hold important data would not at some point in the future become obsolete.
Neither of these points even brings up the obvious benefit to the rest of us that use non-MS systems. It would increase competition in the Word Processing arena and would probably move use towards a world where
It is informative because it shows the sort of feelings that SGI is engendering in the Linux community by not playing lip service to Open Source but making real contributions.
Their current strategy is right on. They are the only large company that really understands the whole Open Source thing and they believe that they have found a way to be profitable in the new environment. I think it will work.
200 onto a Linux (i386) box and pretty much be all set?
I think this is exactly what it means. We are also looking forward to being able to do this. I'd love to buy a lower cost Linux box from SGI (although I can't say enough good things about the Origin.) Or put one together myself.
I really hope they port 4dwm and the tools that go along with it. Having that for a desktop is
really neat.
You and me both! I have started using Linux at home and it is just not as smooth as the SGI O2 I have at work. Even though my Linux box is two or three times faster, it still is "chunky". The UI is not as smooth or consistent. The more SGI can make Linux look and act like the default IRIX environment, the happier I will be.
The Oregon court has made a very smart, forward looking decision. It is absolutely critical that ISPs have access to both the Cable and DSL infrastructure. Bell and the rest of them dream of the day that they will *force* their cable and dsl subscribers to look at _their_ portal ... and _their_ ads. They will force everyone through _their_ proxy server. They will choose what sorts of email servers they will support. And which sort of IP packets they will let through.
It is important that people wake up and realize the value of competition. Persons using dial-up access now have CHOICE. If you don't like your ISP, you can choose one of the other 15 that service your area. This competition has resulted in improved service, as each ISP strives to add the bullet points that another ISP may currently have and to fix problems that may be causing customers to complain.
Local Loop (the physical medium that connects a home or business to an ISP) and the IP Services provided by the ISP are two different animals. Bell is using their monopoly on the one (high speed local loop) to gain a monopoly on the other (IP Services.)
Let's hope that other states follow the lead of Oregon.