Actually, the X11 version provided by Apple can run rootless OR fullscreen. Personally I use it mainly fullscreen with WindowMaker as WM -- it's quite puzzling for people that see me using WM on my ibook, thinking I'm under linux, when with a mouse stroke I reveal the OSX desktop;-)
I also used it in Xnest, that can be useful too. Here is the small script I have to run it in Xnest:
are you kidding ? Here's a hello world in Smalltalk:
Transcript show: 'Hello World !'.
Effectively, that looks extremely complex. You send a message "show:" to the object Transcript, with a string 'Hello World!' as a parameter.
Btw, you can use Smalltalk for scripting (although I personally didn't tried it;-) -- try GNU Smalltalk.
But better, you should have a look on Squeak, and to be on-topic, you *absolutely* need to try Seaside if you're interested by extraordinary technologies to create websites. Ruby On Rails is quite nice, but I was blown away by Seaside. It's inspired by WebObjects, and it's really fantastic (as it leverage the fact that it uses Smalltalk;-)
You would be surprised how many times we bid on a project, lose it to the lowest bidder, and then a few years later hear from that same company again because the relationship went south and the person's code is completely unmaintainable. Many times they end up paying us to rebuild the app at that point...
Funny and a bit OT, but I had the exact opposite the other day. A few years ago, somebody asked me to do a proposition for a website (small database, some system to publish shared documents, but nothing so complex, really). As I was still a part-time student at the time, it was a good opportunity. With a friend we wrote a fairly complete proposal and demo, and set our prices raisonnably (or so we thought).
It turns out that they finally choose.. the most expensive company (most expensive as in around $30k -- which considering the project's complexity was frankly a robbery). Well, at the time I just thought, they preffered a probably bigger company instead of free-lances, too bad for them (hm and for us;-).
Last week the friend that originally asked me the proposal called me back, and told me that 1) the thing was bugged and never finished, and they handled it in the meantime to another company, that also b0rked it 2) that the company that got the contract originally *never* gave a proposal (THAT is fishy... I wonder who they bribed;-) 3) and in the end, they will probably need to redo everything, so there will be perhaps an opportunity to step in again.
I will never cease to be amazed at the way some companies handles projects...
GNUstep has compatibility problems with what InterfaceBuilder generates on the Macs ? well, yes, sortof.
What IB and GORM (the GNUstep equivalent to IB) do, is to *serialize* a bunch of objects. GNUstep does that in a cross-platform way (if I create a.gorm on linux/ppc it will work on windows/x86), but of course it's not byte-by-byte compatible with what IB does -- it's impossible, as the serialisation depends directly on the platform/runtime.
BUT. InterfaceBuilder can now generates XML files instead of the previous binaries, and there are some efforts to make GNUstep read them. So we'll probably have the hability to read directly InterfaceBuilder files (nibs) in the coming year. If you want to help:-)
Additionally, you could also use Renaissance, an XML framework to generate interfaces on GNUstep and on OSX (so, just one file for both platforms).
it only implements a subset of the Mac Toolkit anyway.
Er, well, it's technically true -- GNUstep only implements the Cocoa part, not Quicktime, for example. But frankly, GNUstep cover nearly everything in Cocoa that you don't have too much difficulties to port a Cocoa program to GNUstep in general (and vice-versa).
On the other hand Sun has access to the old OpenStep implementation since they co-developed it with NeXT.. IIRC it is more complete and more stable than GNUStep, although it relies on PS rather than PDF for drawing. It would be interesting if it could be open-sourced and like you suggest made the basis of their suite.
Well, I'm not so sure that OPENSTEP/Solaris is "more complete" than GNUstep, as we have nearly everything implemented (the missing things are classes barely used). What GNUstep miss is some new classes added by Apple recently, that's all. And GNUstep implements partly theses new classes, while OPENSTEP/Solaris doesn't. On the other hand, yes, OPENSTEP/Solaris is probably more stable than GNUstep/Solaris;-) (although GNUstep/linux works quite well now).
Anyway, as you say, it won't happen. Sun is a strange company -- having OPENSTEP available (and moreover, the fantastic development tools !) and a bunch of OPENSTEP applications (they bought LighthouseDesign!!),... they choose to can the project and burry thoses apps, and go along with Java to bring an "Object Oriented Desktop". We still wait for it with Java.. Talk about a heavy Not Invented Here syndrom... (it's even worse as they participated in the API creation..)
Re:The key purchase: Jobs and Unix
on
NeXTSTEP To Mac OS X
·
· Score: 2, Insightful
Problem was that BeOS was not as polished, by far, than OPENSTEP. It used DisplayPostScript, perfect choice for printing (and thus perfect for the mac DTP market), it was based on a Mach kernel with a BSD personality, it existed since years, was a proven platform with huge deployment on critical apps, and had wonderful and ahead of their time development tools. Compared to BeOS, seriously, there was no comparison.
BeOS was a neat OS, but it was years behind OPENSTEP technically, on many levels. BeOS multi-user support was lacking, the POSIX compatibility layer was not perfect either, etc.
It made perfect sense to buy NeXT rather than BeOS. The thing is, many people at the time knew BeOS and even tried it on their desktop, while few used NeXTSTEP/OPENSTEP. Thus people didn't understood well the (wise) choice. But frankly, it had more to do with the technical side than it had to do with Steve Jobs.
Steve Jobs back at Apple was a bonus, but I'm not even sure it was considered as a bonus by G.Amelio anyway;-) (and we can only say it was a bonus, because it turned out that way...)
Not quite. ObjectiveC requires the programmer to spend more effort handling memory management and distinguishes between objects and primitives.
Yes and no... Yes, it needs a bit more effort than Smalltalk, but the reference counting garbage collector thingy is not really hard to use, and you mostly use autorelease anyway... and yes, you distinguishes between objects and primitives like int or float, because Objective-C is "just" an OOP superset of the C language. But if you want you can just use NSNumber... quite similar to jave on that point of view. It's mostly a pragmatic decision.
Also, the typical Smalltalk class system seems much more orthogonal. ObjectiveC frameworks (including Cocoa, and I assume OpenStep) don't seem to really take advantage of the capabilities of dynamic classes... (...)
Uh ? theses four methods are here for convenient use. You can always initialize the NSString object using a NSData object..
(...) Why are these four separate methods with separate names? Why is "string" part of the method name? In Smalltalk, these would probably all be "toString" methods on the objects. If ObjectiveC was Smalltalk-like, you'd at least have something like this:
It's just that Smalltalk allow you to not use parenthesis when the interpreter can remove ambiguity, while it's necessary to use the brackets in Objective-C...
The main difference between Smalltalk and Objective-C (apart that *everything is an object* in Smalltalk..) is the existence of blocks, that you don't have in Objective-C (hm, actually.. there are some implementations...). But blocks are obligatory in Smalltalk due to its OO-ness, while they are not in Objective-C as you can just reuse C construct (if/while/for..).
Though I agree, blocks are extremely nice and that's generally something that Objective-C programmers would like gcc to include:-)
I am actually generally a fan of Smalltalk, but Objective-C is the closest thing you can find, and it has some nice things too. And the "image" approach of Smalltalk, while often nice, has also its problems. Although my dream development system would probably be Smalltalk+OpenStep (while keeping the possibility of using Objective-C code if needed for speed..).
There is actually a Squeak-Objective-C bridge that nearly works (it has a problem with delegates, mostly..), and that nicely send -release messages to objects automatically, using the Squeak GC...;-)
Hm..I thinkg you misunderstand how things are related in fact.. under X-Window, you need a special program, in charge of the window management (ie, to move them, etc.). It's called.. a window manager. WindowMaker is a X11 window manager.
Then you have X11 programs, that are in charge of their window's content. As programming directly using XLib is not fun, there is a lot of X11 "toolkits". Qt and Gtk are examples of toolkits.
GNUstep is an implementation of the OpenStep API. Basically, it's a programming library, a toolkit like Qt and Gtk if you want -- not only for graphic apps, but also for non-graphic apps. In fact, the OpenStep API is divided in two frameworks: Foundation (which deals with non-graphical things such as threads, files, unicode strings, etc.) and AppKit (which provides all the nifty widgets and how you use them). But, additionally to that, GNUstep *also* provides development applications: Gorm, a graphical interface builder, and ProjectBuilder, an IDE.
Actually, GNUstep runs mainly on X11, but the way it is architectured, it's not that complex to use other drawing display. For example, there is 3 backends for X11 -- one using xlib, the other using libart, and the third using Cairo. And there is a backend for Windows GDI. So in fact, it's not tied at all to a X11, and the notion of an independant window manager is specific to X11 (actually, GNUstep apps don't really need a window manager even under X11 -- they can manage themselves..).
But, if you're under X11, chances are that you want to run other programs alongside GNUstep programs -- KDE/GNOME programs for example. You then *need* a window manager. WindowMaker is the "default" window manager recommanded by the GNUstep project, mainly because its look and feel match the GNUstep look and feel. But you can use others window manager.
And WindowMaker itself doesn't use GNUstep.
Not sure if I understood well your questions, I tried to explain how things are related, hopefully more clearly.
At any rate, yes GNUstep is the framework that WindowMaker and its tools are built on to be completely accurate.
Well.. actually, no, that was the point. WindowMaker doesn't use GNUstep and it's not even programmed in Objective-C ! But, it's considered as the "default" WM for GNUstep applications, as it implements some support for GNUstep apps, but that's all.
Actually, the "import" versus "include + ifdefs" is not a problem. It's not even really a GNUstep problem -- more a gcc one. At one point, they deprecated the "import" (but the support was still there..). Now, the "import" is no more deprecated in the current gcc. So... if you want to use import, go on, it works with the apple gcc and the fsf gcc..
The automatic garbage collection support (using boehm library) in GNUstep is, afaik, only available for -base (Foundation), not -gui (AppKit), although I could be wrong. I must confess that I never tried it, as the retain/release/autorelease garbage collector scheme works well enough (and is very flexible) as soon as you understand it (a very good article about it is this one).
er, there is a gui bundle for gnustep that actually change that -- it moves the menu on the top, like on MacOS. So you can choose to not use the vertical menus, if you want(personally, I'm a big fan of vertical menus, but on a big screen - on a small screen, I tend to prefer horizontal menus:-)
Actually, you probably want to correct your screen's gamma -- with a proper gamma, the default color scheme of GNUstep looks much better.
It's also not very difficult to change the color scheme using Preferences.app, as you can see on this screenshot.
Plus, the look can be modified with themes
Actually, you probably want to correct your screen's gamma -- with a proper gamma, the default color scheme of GNUstep looks much better.
It's also not very difficult to change the color scheme using Preferences.app, as you can see on this screenshot: http://www.nongnu.org/backbone/images/screenshots/ preferences-4.png
Plus, the look can be modified with themes : http://www.roard.com/screenshots/screenshot_theme3 4.png
- GNUStep looks like something that I shat out last night after eating at taco bell
Themes exists for that reason. Correctly setting your display's gamma helps, too.
- GNUStep is GPL'd, which means it is anti-business and useless to those of us who have work to do, not religion
GNUstep is LGPL'd, which means that it is business-friendly and useful to those of us who have work to do, not religion. Bonus: it's also useful for those of us who want to do CS religion.
- Being "pure" open source, GNUStep is vastly inferior to OS X, which is maintained by a cadre of well paid professional American programmers, not $2.99/hour indian high school dropouts
Taking out Saddam changed the world. Libya has forsaken terrorism as an instrument of state policy. Saudi Arabia has hardened its stance against the Wahhabists. Jordan has become more closely aligned with the West. Iran is liberalizing, although we've still got a long way to go there. North Korea is rattling its saber, which historically is a sign of desperation and weakness from that regime. Support for Palestinian terrorism in Lebanon and Egypt is at an all-time low; the intifada is effectively over, since there hasn't been a successful attack against Israel in three months now. Even Syria, last holdout of the Baathists, is moving toward a more civilized position.
Excuse me, but what's the relation with Iraq and all thoses events ? Libya had just not forsaken terrorism the day after Iraq was invaded by US troop -- it's the result of a much longer process which began many years ago. The "liberalization" of Iran has absolutely NOTHING to do with the Iraq war, again, it's a process that began years ago (and hopefully that will continue). North Korea is in a pity state since a long time ago, and that's not the first time they issue treats; anyway I fail to see why it's such a good news for you. Palestinian terrorism continue, and there is actually an increase of violence in Iraq as well. And Syria's position is slowly evolving, but the credit could also be assigned to Bashar al-Assad (yet for once you're not that wrong, the actual international situation probably helped too, but that's not the only explanation).
In any cases, attacking Iraq wasn't such a great decision, and had far less effects than what you describe (most of them were trends already in place). Iraq links to terrorisms and bin laden are quite weak, contrary to what you think. Sure, there was some connexions, but frankly not something that absolutely mandated an US expedition (against the whole world). Saddam was NOT a threat to US, militarily and on terrorism.
But yes, Saddam was a threat to USA. Actually, it could have become a quite big threat. And yes, it involves oil. Not in the way that USA wanted Saddam's oil; no, it's a bit more subtil. USA fund its deficit with the dollar, as it is the international currency. And furthermore, the dollar is the oil's currency. Do you know what Saddam wanted to do ? sell is oil using Euro. *That* would have been desastrous for USA -- because it would have created a precedent and other oil countries could have very well started to deal using Euro. If it had happend, forget about financing the US deficit with the dollar...
In fact, this explanation is surely not the only one; but it's a quite important one and generally one that is forgotten. I think Bush attacked Iraq for plenty of reasons; oil was definitely in balance, but the strong feeling against Iraq by Bush and his men was another. I think they just used 9/11 as an excuse for their own personal plans. The occasion was just too good.
Is it so things like the bombing of Dresdon or Hiroshima and Nagasaki, or American crimes during Viet Nam, won't be able to be called 'terrorism'?
Well, those things AREN'T terrorism, obviously. Remember the definition of the term: covert attacks against noncombatants with the intention of influencing political of social policy through fear. Bombing during wartime is carried out with the intent of destroying the enemy's means to fight or will to fight. That's not terrorism. And war crimes, when they occur, are not terrorism either, for the same reason.
Actually, the use of the atomic bomb could quite easily be mapped on your definition of terrorism.
The two attacks against Nagasaki and Hiroshima weren't exactly advertised before done and obviously the planes were covert...
Sure, you could say that, as it is war, it is expected to have bombs and thus it's not covert attack. Well, same for terrorism. We *expect* terrorist attacks.
And obviously, you can't say that theses bombs weren't an attack against non-combattants. They erased two *cities*.
Lastly, while the direct effect of it while Japan surrendering, one other clear impact (and goal) was definitely to show the world what kind of power USA now had with nuclear bomb. So you could very well argue that was done "with the intention of influencing political of social policy through fear".
We could possibly admit that bombing Hiroshima could fall under "war" (or war crime, even), but bombing Nagasaki is way more borderline to your definition of terrorism than what you seem to think.
I don't know for other countries, but in France nearly everybody has a credit card (visa, mastercard, etc) with a chip. After all we invented it;-) [well, Roland Moreno did it]
\begin{rant}
In GB though my bank refused to give me a credit card (yes, you know, thoses 25yo PhD students could leave the country with our money ! moron...) and only allowed me a cash card (can't pay with it). Which is quite strange for me, as I got one in France since I'm 15. Doh.
Make's you wonder what what OS X would have been like had Apples plan to by BeOS not fallen through. BeOS had a lot of features NeXT did not have and some that are just being implemented now, such as journaled file systems found in Panther.
Well, the reverse is true;-) -- the NeXT development environment was way ahead (and still is). Even if BeOS dev was quite nice, for sure. Plus, OPENSTEP used vector display system (DisplayPostScript), which then permitted true wysiwyg, and leads to DisplayPDF.
Actually, you can't imagine what OSX would have been like, because frankly, OSX is quite different from OPENSTEP, and not always in a good way. The need of supporting legacy -- software and UI -- would have modified BeOS the same way it has modified OPENSTEP. I personally much prefer OPENSTEP UI (I'm speaking about the feel, not the look -- although I also happend to like the clean look of OPENSTEP over the in-your-face look in OSX... but commercially (marketing) it's more useful to have OSX look:-)
The possibility of acquiring a patent, and thereby a guaranteed source of revenue, is what spurs innovation.
Wrong. It's the idea, but it's not what actually happend. In fact, patents were NOT created to incite inventors to invent, because, face it, inventor invented things well before patents. Patents were created as a tradeoff with inventors, because at the time (industrial revolution), there was a big problem : inventors invented things, but kept the recipe secret, to have an edge over the competition. Thus, patents was a deal with governments : "give us the secret (the patent) and you'll have a state monopoly for a certain period". Patents were created so inventors explained their invention. Patents were NOT necessary as an incentive to innovate, but as an incentive to SHARE the discovery with the society.
Now, in our modern world, you should perhaps understand that it's quite rare to have people looking for patents in order to understand a discovery. Nineteen century patents were wonderfully done, for most of them. Now patents are legal giberrish totally encrypted for a normal engieneer. In fact, in most companies, the motto is to NOT look at patents, for fear of legals battels. Explain to me then what's the goal of patents !
Plus, in the case that interests us, it's not "normal" patents, but Software patents. With software patents we're in a totally different realm, the one of pure thought. Instead of describing a method of doing something, people now describe the idea of doing something. This shift is absolutely scary. Add that the fact that the application process is incredibly bad (many patents are accepted dispite prior art, or dispite an innovative part), and the fact that the software "industry" innovations works at the scale of months or a few year, not the twenty-something years awarded by a patent... And you have a BIG problem as a software engineer. Frankly, look at the actual history of software engineering and show me how many cases where a patent actually profited to the whole industry ? (and even if it could happend, is it worth the price ?) And, will you argue that the software industry perfectly innovated before software patents ?
In fact, if software patents were 1) awarded by real experts 2) limited to say, 4-5 years 3) and submarine patents were illegal 4) and describe actually a method and not an idea, then, yes, perhaps software patents would be a good incentive to innovation (even if we innovate perfectly without them). But face it, it won't happend, and with their current state, they actually do more harm than good. I also fear something, if SP pass in EU, big companies will have a clear field, and I think that's when the situation will run amok.
But for the most part, the profit motive is what drives innovation. Patents are essential to that process.
However, of the two of us, I seem to be the one who's far more aware of the real world pressures put on politicians and lawmakers: these people didn't suddenly wake up one day and say "Hey, software patents are a good thing!" of their own accord, someone led them to believe that, and that someone (clearly) were businesses and their lobbyists.
I'm entirely aware of the lobbyist pressures. And in my opinion, the most pressures comes from US companies, not EU companies (95% of the EU companies were actually *against* the SP, but according to the EU commission, that poll was dismissed because thoses 95% weren't "economically signifiant" (which is a bit stupid -- a startup won't be economically signifiant, but that's where the innovation comes from. And most of the EU software companies are small companies...). Anyway, sorry if my reaction was a bit harsh, but what made me jump isn't the fact that you claim there's lobby (it should be obvious for anybody), but the fact that you stated :
"In the real world, where companies and countries have to compete against one another in business, not recognising software patents in the EU whilst they are being handed out like hot cakes in the US is the quickest way to destroy software development within the EU. I don't like it - in fact, I hate it - but those are the reasons behind it."
So, I understand that you don't like the Software Patents et all, but saying what you say is just wrong. "not recognising software patents in the EU whilst they are being handed out like hot cakes in the US" IS NOT the quickest way to destroy software development within the EU, to the contrary !
If you wanted to say that, the FACT that the US allowed the SP leads today to a push of SP in the EU, yes, that's true. But what should also consider is that most of the push is the result of US companies. And frankly, if that what you wanted to express, it was badly worded. To me, I just read it like "if SP exists in US, EU needs to have it too" -- the exact argument of the SP lobbyist in EU. And that argument is deep wrong, as I explained it in my post (nothing prevents a EU company to get a US patent if that's so necessary, but two wrong won't made a good, and allowing EU SP won't balance US SP, I even believe that it will just have the nice result of turning lawyers firms amok).
In the real world, where companies and countries have to compete against one another in business, not recognising software patents in the EU whilst they are being handed out like hot cakes in the US is the quickest way to destroy software development within the EU. I don't like it - in fact, I hate it - but those are the reasons behind it.
Excuse me, sir, but... what are you smoking ?
The Patent System is grossly abused in general, and particularly in the software area. It's widely acknowledged (check the economical studies available on the ffii site for example) that Software Patents doesn't increase inovation (and it's not particularly a difficult thing to understand). I won't be long here, suffice to say that programming is inherently an incremental work, based on top of others ideas, and moreover, it's one of the most complex creation that could be done (hell, it's so complex that we can't manage to produce 100% sure non-bugged software for any complex procedure). Pushing Software Patents is so deeply wrong that it would be funny if not real. How could you think a second that the patent system, with submarine patents, looooong submission delay, innovation-challenged patents, and incredibly inaccurate verification, could work in a field that would use dozens of theses "patents" without even realizing it, all that in a period not even sufficient to apply a patent ?
Why do the EU want theses ? it's totally insane. Not having patents is actually the BEST thing that could happens to the EU (and to all others countries). Plus, Software Patents aren't bringing added value at all, only $$$ for lawyers. Not one engineer read software patents (and for good reason -- not only because of the judicial risk).
The only reason I could imagine, sadly, is that some EU bureaucrats get big dollars by US companies. The fact that theses bureaucrats just choose to overrulle the European Commission is so incredible that my hope is it will create an enormous indignation (because, face it, the average EU citizen doesn't care about patents, but perhaps the beahvior shown by the bureaucrat (total irrespect to the elected representants) will trigger something).
Slashdot Mirror
Given that it's WindowMaker, I'm surprised they don't just think you're running a really old version of OS X. :-)
:-))
(For anyone who doesn't get the joke, just take the next step....
Indeed :-) -- furthermore as I'm running GNUstep programs, so the imitation is complete...
Actually, the X11 version provided by Apple can run rootless OR fullscreen. Personally I use it mainly fullscreen with WindowMaker as WM -- it's quite puzzling for people that see me using WM on my ibook, thinking I'm under linux, when with a mouse stroke I reveal the OSX desktop ;-)
I also used it in Xnest, that can be useful too. Here is the small script I have to run it in Xnest:
are you kidding ? Here's a hello world in Smalltalk:
Transcript show: 'Hello World !'.
Effectively, that looks extremely complex. You send a message "show:" to the object Transcript, with a string 'Hello World!' as a parameter.
Btw, you can use Smalltalk for scripting (although I personally didn't tried it ;-) -- try GNU Smalltalk.
But better, you should have a look on Squeak, and to be on-topic, you *absolutely* need to try Seaside if you're interested by extraordinary technologies to create websites. Ruby On Rails is quite nice, but I was blown away by Seaside. It's inspired by WebObjects, and it's really fantastic (as it leverage the fact that it uses Smalltalk ;-)
Definitely a good idea to check, imho..
Funny and a bit OT, but I had the exact opposite the other day. A few years ago, somebody asked me to do a proposition for a website (small database, some system to publish shared documents, but nothing so complex, really). As I was still a part-time student at the time, it was a good opportunity. With a friend we wrote a fairly complete proposal and demo, and set our prices raisonnably (or so we thought).
It turns out that they finally choose.. the most expensive company (most expensive as in around $30k -- which considering the project's complexity was frankly a robbery). Well, at the time I just thought, they preffered a probably bigger company instead of free-lances, too bad for them (hm and for us ;-).
Last week the friend that originally asked me the proposal called me back, and told me that 1) the thing was bugged and never finished, and they handled it in the meantime to another company, that also b0rked it 2) that the company that got the contract originally *never* gave a proposal (THAT is fishy... I wonder who they bribed ;-) 3) and in the end, they will probably need to redo everything, so there will be perhaps an opportunity to step in again.
I will never cease to be amazed at the way some companies handles projects...
GNUstep has compatibility problems with what InterfaceBuilder generates on the Macs ? well, yes, sortof.
What IB and GORM (the GNUstep equivalent to IB) do, is to *serialize* a bunch of objects. GNUstep does that in a cross-platform way (if I create a .gorm on linux/ppc it will work on windows/x86), but of course it's not byte-by-byte compatible with what IB does -- it's impossible, as the serialisation depends directly on the platform/runtime.
BUT. InterfaceBuilder can now generates XML files instead of the previous binaries, and there are some efforts to make GNUstep read them. So we'll probably have the hability to read directly InterfaceBuilder files (nibs) in the coming year. If you want to help :-)
Additionally, you could also use Renaissance, an XML framework to generate interfaces on GNUstep and on OSX (so, just one file for both platforms).
it only implements a subset of the Mac Toolkit anyway.
Er, well, it's technically true -- GNUstep only implements the Cocoa part, not Quicktime, for example. But frankly, GNUstep cover nearly everything in Cocoa that you don't have too much difficulties to port a Cocoa program to GNUstep in general (and vice-versa).
On the other hand Sun has access to the old OpenStep implementation since they co-developed it with NeXT.. IIRC it is more complete and more stable than GNUStep, although it relies on PS rather than PDF for drawing. It would be interesting if it could be open-sourced and like you suggest made the basis of their suite.
Well, I'm not so sure that OPENSTEP/Solaris is "more complete" than GNUstep, as we have nearly everything implemented (the missing things are classes barely used). What GNUstep miss is some new classes added by Apple recently, that's all. And GNUstep implements partly theses new classes, while OPENSTEP/Solaris doesn't. On the other hand, yes, OPENSTEP/Solaris is probably more stable than GNUstep/Solaris ;-) (although GNUstep/linux works quite well now).
Anyway, as you say, it won't happen. Sun is a strange company -- having OPENSTEP available (and moreover, the fantastic development tools !) and a bunch of OPENSTEP applications (they bought LighthouseDesign!!), ... they choose to can the project and burry thoses apps, and go along with Java to bring an "Object Oriented Desktop". We still wait for it with Java..
Talk about a heavy Not Invented Here syndrom... (it's even worse as they participated in the API creation..)
Problem was that BeOS was not as polished, by far, than OPENSTEP. It used DisplayPostScript, perfect choice for printing (and thus perfect for the mac DTP market), it was based on a Mach kernel with a BSD personality, it existed since years, was a proven platform with huge deployment on critical apps, and had wonderful and ahead of their time development tools. Compared to BeOS, seriously, there was no comparison. BeOS was a neat OS, but it was years behind OPENSTEP technically, on many levels. BeOS multi-user support was lacking, the POSIX compatibility layer was not perfect either, etc. It made perfect sense to buy NeXT rather than BeOS. The thing is, many people at the time knew BeOS and even tried it on their desktop, while few used NeXTSTEP/OPENSTEP. Thus people didn't understood well the (wise) choice. But frankly, it had more to do with the technical side than it had to do with Steve Jobs. Steve Jobs back at Apple was a bonus, but I'm not even sure it was considered as a bonus by G.Amelio anyway ;-) (and we can only say it was a bonus, because it turned out that way...)
Not quite. ObjectiveC requires the programmer to spend more effort handling memory management and distinguishes between objects and primitives.
Yes and no... Yes, it needs a bit more effort than Smalltalk, but the reference counting garbage collector thingy is not really hard to use, and you mostly use autorelease anyway... and yes, you distinguishes between objects and primitives like int or float, because Objective-C is "just" an OOP superset of the C language. But if you want you can just use NSNumber ... quite similar to jave on that point of view. It's mostly a pragmatic decision.
Also, the typical Smalltalk class system seems much more orthogonal. ObjectiveC frameworks (including Cocoa, and I assume OpenStep) don't seem to really take advantage of the capabilities of dynamic classes... (...)
Uh ? theses four methods are here for convenient use. You can always initialize the NSString object using a NSData object..
(...) Why are these four separate methods with separate names? Why is "string" part of the method name? In Smalltalk, these would probably all be "toString" methods on the objects. If ObjectiveC was Smalltalk-like, you'd at least have something like this:
I really don't understand your example. How is it different than:
in Objective-C ?
just to show you, you can write your Smalltalk example like that:
It's just that Smalltalk allow you to not use parenthesis when the interpreter can remove ambiguity, while it's necessary to use the brackets in Objective-C ...
The main difference between Smalltalk and Objective-C (apart that *everything is an object* in Smalltalk..) is the existence of blocks, that you don't have in Objective-C (hm, actually.. there are some implementations...). But blocks are obligatory in Smalltalk due to its OO-ness, while they are not in Objective-C as you can just reuse C construct (if/while/for ..).
Though I agree, blocks are extremely nice and that's generally something that Objective-C programmers would like gcc to include :-)
I am actually generally a fan of Smalltalk, but Objective-C is the closest thing you can find, and it has some nice things too. And the "image" approach of Smalltalk, while often nice, has also its problems. Although my dream development system would probably be Smalltalk+OpenStep (while keeping the possibility of using Objective-C code if needed for speed..).
There is actually a Squeak-Objective-C bridge that nearly works (it has a problem with delegates, mostly..), and that nicely send -release messages to objects automatically, using the Squeak GC... ;-)
Hm..I thinkg you misunderstand how things are related in fact.. under X-Window, you need a special program, in charge of the window management (ie, to move them, etc.). It's called.. a window manager. WindowMaker is a X11 window manager.
Then you have X11 programs, that are in charge of their window's content. As programming directly using XLib is not fun, there is a lot of X11 "toolkits". Qt and Gtk are examples of toolkits.
GNUstep is an implementation of the OpenStep API. Basically, it's a programming library, a toolkit like Qt and Gtk if you want -- not only for graphic apps, but also for non-graphic apps. In fact, the OpenStep API is divided in two frameworks: Foundation (which deals with non-graphical things such as threads, files, unicode strings, etc.) and AppKit (which provides all the nifty widgets and how you use them). But, additionally to that, GNUstep *also* provides development applications: Gorm, a graphical interface builder, and ProjectBuilder, an IDE.
Actually, GNUstep runs mainly on X11, but the way it is architectured, it's not that complex to use other drawing display. For example, there is 3 backends for X11 -- one using xlib, the other using libart, and the third using Cairo. And there is a backend for Windows GDI. So in fact, it's not tied at all to a X11, and the notion of an independant window manager is specific to X11 (actually, GNUstep apps don't really need a window manager even under X11 -- they can manage themselves..).
But, if you're under X11, chances are that you want to run other programs alongside GNUstep programs -- KDE/GNOME programs for example. You then *need* a window manager. WindowMaker is the "default" window manager recommanded by the GNUstep project, mainly because its look and feel match the GNUstep look and feel. But you can use others window manager.
And WindowMaker itself doesn't use GNUstep.
Not sure if I understood well your questions, I tried to explain how things are related, hopefully more clearly.
At any rate, yes GNUstep is the framework that WindowMaker and its tools are built on to be completely accurate.
Well.. actually, no, that was the point. WindowMaker doesn't use GNUstep and it's not even programmed in Objective-C ! But, it's considered as the "default" WM for GNUstep applications, as it implements some support for GNUstep apps, but that's all.
"a development framework that hasn't needed significant modification for 10 years"
That's not because it's stable, etc. It's because nobody uses it.
Tell that to Cocoa developers on Mac... check facts the next time.
Actually, the "import" versus "include + ifdefs" is not a problem. It's not even really a GNUstep problem -- more a gcc one. At one point, they deprecated the "import" (but the support was still there..). Now, the "import" is no more deprecated in the current gcc. So... if you want to use import, go on, it works with the apple gcc and the fsf gcc..
The automatic garbage collection support (using boehm library) in GNUstep is, afaik, only available for -base (Foundation), not -gui (AppKit), although I could be wrong. I must confess that I never tried it, as the retain/release/autorelease garbage collector scheme works well enough (and is very flexible) as soon as you understand it (a very good article about it is this one).
er, there is a gui bundle for gnustep that actually change that -- it moves the menu on the top, like on MacOS. So you can choose to not use the vertical menus, if you want(personally, I'm a big fan of vertical menus, but on a big screen - on a small screen, I tend to prefer horizontal menus :-)
Actually, you probably want to correct your screen's gamma -- with a proper gamma, the default color scheme of GNUstep looks much better.
It's also not very difficult to change the color scheme using Preferences.app, as you can see on this screenshot.
Plus, the look can be modified with themes
Actually, you probably want to correct your screen's gamma -- with a proper gamma, the default color scheme of GNUstep looks much better. It's also not very difficult to change the color scheme using Preferences.app, as you can see on this screenshot: http://www.nongnu.org/backbone/images/screenshots/ preferences-4.png
Plus, the look can be modified with themes : http://www.roard.com/screenshots/screenshot_theme3 4.png
You could want to install/use Backbone or Garma, which are GNUstep-based Desktops...
Other links, Objective-C and Apple Cocoa
Themes exists for that reason. Correctly setting your display's gamma helps, too.
- GNUStep is GPL'd, which means it is anti-business and useless to those of us who have work to do, not religionGNUstep is LGPL'd, which means that it is business-friendly and useful to those of us who have work to do, not religion. Bonus: it's also useful for those of us who want to do CS religion.
- Being "pure" open source, GNUStep is vastly inferior to OS X, which is maintained by a cadre of well paid professional American programmers, not $2.99/hour indian high school dropoutsWhat a troll. Moron.
Taking out Saddam changed the world. Libya has forsaken terrorism as an instrument of state policy. Saudi Arabia has hardened its stance against the Wahhabists. Jordan has become more closely aligned with the West. Iran is liberalizing, although we've still got a long way to go there. North Korea is rattling its saber, which historically is a sign of desperation and weakness from that regime. Support for Palestinian terrorism in Lebanon and Egypt is at an all-time low; the intifada is effectively over, since there hasn't been a successful attack against Israel in three months now. Even Syria, last holdout of the Baathists, is moving toward a more civilized position.
Excuse me, but what's the relation with Iraq and all thoses events ? Libya had just not forsaken terrorism the day after Iraq was invaded by US troop -- it's the result of a much longer process which began many years ago. The "liberalization" of Iran has absolutely NOTHING to do with the Iraq war, again, it's a process that began years ago (and hopefully that will continue). North Korea is in a pity state since a long time ago, and that's not the first time they issue treats; anyway I fail to see why it's such a good news for you. Palestinian terrorism continue, and there is actually an increase of violence in Iraq as well. And Syria's position is slowly evolving, but the credit could also be assigned to Bashar al-Assad (yet for once you're not that wrong, the actual international situation probably helped too, but that's not the only explanation).
In any cases, attacking Iraq wasn't such a great decision, and had far less effects than what you describe (most of them were trends already in place). Iraq links to terrorisms and bin laden are quite weak, contrary to what you think. Sure, there was some connexions, but frankly not something that absolutely mandated an US expedition (against the whole world). Saddam was NOT a threat to US, militarily and on terrorism.
But yes, Saddam was a threat to USA. Actually, it could have become a quite big threat. And yes, it involves oil. Not in the way that USA wanted Saddam's oil; no, it's a bit more subtil. USA fund its deficit with the dollar, as it is the international currency. And furthermore, the dollar is the oil's currency. Do you know what Saddam wanted to do ? sell is oil using Euro. *That* would have been desastrous for USA -- because it would have created a precedent and other oil countries could have very well started to deal using Euro. If it had happend, forget about financing the US deficit with the dollar...
In fact, this explanation is surely not the only one; but it's a quite important one and generally one that is forgotten. I think Bush attacked Iraq for plenty of reasons; oil was definitely in balance, but the strong feeling against Iraq by Bush and his men was another. I think they just used 9/11 as an excuse for their own personal plans. The occasion was just too good.
Is it so things like the bombing of Dresdon or Hiroshima and Nagasaki, or American crimes during Viet Nam, won't be able to be called 'terrorism'?
Well, those things AREN'T terrorism, obviously. Remember the definition of the term: covert attacks against noncombatants with the intention of influencing political of social policy through fear. Bombing during wartime is carried out with the intent of destroying the enemy's means to fight or will to fight. That's not terrorism. And war crimes, when they occur, are not terrorism either, for the same reason.
Actually, the use of the atomic bomb could quite easily be mapped on your definition of terrorism.
The two attacks against Nagasaki and Hiroshima weren't exactly advertised before done and obviously the planes were covert ...
Sure, you could say that, as it is war, it is expected to have bombs and thus it's not covert attack. Well, same for terrorism. We *expect* terrorist attacks.
And obviously, you can't say that theses bombs weren't an attack against non-combattants. They erased two *cities*.
Lastly, while the direct effect of it while Japan surrendering, one other clear impact (and goal) was definitely to show the world what kind of power USA now had with nuclear bomb. So you could very well argue that was done "with the intention of influencing political of social policy through fear".
We could possibly admit that bombing Hiroshima could fall under "war" (or war crime, even), but bombing Nagasaki is way more borderline to your definition of terrorism than what you seem to think.
5-What does the expression proferred by the Defendant "strategy of fear, uncertainty, and doubt" referred in the article mean?
Microsoft asking what's FUD ? Priceless !
I don't know for other countries, but in France nearly everybody has a credit card (visa, mastercard, etc) with a chip. After all we invented it ;-) [well, Roland Moreno did it]
\begin{rant}
In GB though my bank refused to give me a credit card (yes, you know, thoses 25yo PhD students could leave the country with our money ! moron...) and only allowed me a cash card (can't pay with it). Which is quite strange for me, as I got one in France since I'm 15. Doh.
\end{rant}Make's you wonder what what OS X would have been like had Apples plan to by BeOS not fallen through. BeOS had a lot of features NeXT did not have and some that are just being implemented now, such as journaled file systems found in Panther. Well, the reverse is true ;-) -- the NeXT development environment was way ahead (and still is). Even if BeOS dev was quite nice, for sure. Plus, OPENSTEP used vector display system (DisplayPostScript), which then permitted true wysiwyg, and leads to DisplayPDF.
Actually, you can't imagine what OSX would have been like, because frankly, OSX is quite different from OPENSTEP, and not always in a good way. The need of supporting legacy -- software and UI -- would have modified BeOS the same way it has modified OPENSTEP. I personally much prefer OPENSTEP UI (I'm speaking about the feel, not the look -- although I also happend to like the clean look of OPENSTEP over the in-your-face look in OSX ... but commercially (marketing) it's more useful to have OSX look :-)
Wrong. It's the idea, but it's not what actually happend. In fact, patents were NOT created to incite inventors to invent, because, face it, inventor invented things well before patents. Patents were created as a tradeoff with inventors, because at the time (industrial revolution), there was a big problem : inventors invented things, but kept the recipe secret, to have an edge over the competition. Thus, patents was a deal with governments : "give us the secret (the patent) and you'll have a state monopoly for a certain period". Patents were created so inventors explained their invention. Patents were NOT necessary as an incentive to innovate, but as an incentive to SHARE the discovery with the society.
Now, in our modern world, you should perhaps understand that it's quite rare to have people looking for patents in order to understand a discovery. Nineteen century patents were wonderfully done, for most of them. Now patents are legal giberrish totally encrypted for a normal engieneer. In fact, in most companies, the motto is to NOT look at patents, for fear of legals battels. Explain to me then what's the goal of patents !
Plus, in the case that interests us, it's not "normal" patents, but Software patents. With software patents we're in a totally different realm, the one of pure thought. Instead of describing a method of doing something, people now describe the idea of doing something. This shift is absolutely scary. Add that the fact that the application process is incredibly bad (many patents are accepted dispite prior art, or dispite an innovative part), and the fact that the software "industry" innovations works at the scale of months or a few year, not the twenty-something years awarded by a patent... And you have a BIG problem as a software engineer. Frankly, look at the actual history of software engineering and show me how many cases where a patent actually profited to the whole industry ? (and even if it could happend, is it worth the price ?) And, will you argue that the software industry perfectly innovated before software patents ?
In fact, if software patents were 1) awarded by real experts 2) limited to say, 4-5 years 3) and submarine patents were illegal 4) and describe actually a method and not an idea, then, yes, perhaps software patents would be a good incentive to innovation (even if we innovate perfectly without them). But face it, it won't happend, and with their current state, they actually do more harm than good. I also fear something, if SP pass in EU, big companies will have a clear field, and I think that's when the situation will run amok.
But for the most part, the profit motive is what drives innovation. Patents are essential to that process.Sure, but protifs rarelly involves patents.
I'm entirely aware of the lobbyist pressures. And in my opinion, the most pressures comes from US companies, not EU companies (95% of the EU companies were actually *against* the SP, but according to the EU commission, that poll was dismissed because thoses 95% weren't "economically signifiant" (which is a bit stupid -- a startup won't be economically signifiant, but that's where the innovation comes from. And most of the EU software companies are small companies...).
Anyway, sorry if my reaction was a bit harsh, but what made me jump isn't the fact that you claim there's lobby (it should be obvious for anybody), but the fact that you stated : "In the real world, where companies and countries have to compete against one another in business, not recognising software patents in the EU whilst they are being handed out like hot cakes in the US is the quickest way to destroy software development within the EU. I don't like it - in fact, I hate it - but those are the reasons behind it."
So, I understand that you don't like the Software Patents et all, but saying what you say is just wrong. "not recognising software patents in the EU whilst they are being handed out like hot cakes in the US" IS NOT the quickest way to destroy software development within the EU, to the contrary !
If you wanted to say that, the FACT that the US allowed the SP leads today to a push of SP in the EU, yes, that's true. But what should also consider is that most of the push is the result of US companies. And frankly, if that what you wanted to express, it was badly worded. To me, I just read it like "if SP exists in US, EU needs to have it too" -- the exact argument of the SP lobbyist in EU. And that argument is deep wrong, as I explained it in my post (nothing prevents a EU company to get a US patent if that's so necessary, but two wrong won't made a good, and allowing EU SP won't balance US SP, I even believe that it will just have the nice result of turning lawyers firms amok).
Excuse me, sir, but... what are you smoking ?
The Patent System is grossly abused in general, and particularly in the software area. It's widely acknowledged (check the economical studies available on the ffii site for example) that Software Patents doesn't increase inovation (and it's not particularly a difficult thing to understand). I won't be long here, suffice to say that programming is inherently an incremental work, based on top of others ideas, and moreover, it's one of the most complex creation that could be done (hell, it's so complex that we can't manage to produce 100% sure non-bugged software for any complex procedure). Pushing Software Patents is so deeply wrong that it would be funny if not real. How could you think a second that the patent system, with submarine patents, looooong submission delay, innovation-challenged patents, and incredibly inaccurate verification, could work in a field that would use dozens of theses "patents" without even realizing it, all that in a period not even sufficient to apply a patent ?
Why do the EU want theses ? it's totally insane. Not having patents is actually the BEST thing that could happens to the EU (and to all others countries). Plus, Software Patents aren't bringing added value at all, only $$$ for lawyers. Not one engineer read software patents (and for good reason -- not only because of the judicial risk).
The only reason I could imagine, sadly, is that some EU bureaucrats get big dollars by US companies. The fact that theses bureaucrats just choose to overrulle the European Commission is so incredible that my hope is it will create an enormous indignation (because, face it, the average EU citizen doesn't care about patents, but perhaps the beahvior shown by the bureaucrat (total irrespect to the elected representants) will trigger something).