It is correct to say we all knew about and used the individual techniques (principles) for decades. BUT no one before Kent Beck bundled them all into one coherent methodology, and explained the synergies. THAT is the genius of XP!
Sorry, but I disagree. XP is pretty much how a lot of Lisp, Smalltalk, and similar communities have worked for a long time, using pretty much the whole of what is now called XP, not just bits and pieces. That's not surprising: those communities face rapidly changing requirements, they often have easy access to their end users, and they are usually highly qualified and trained. And that's the community Kent Beck came from; he is just trying to popularize those practices and ideas among non-Smalltalk users.
I have my doubts that that is such a good idea. Languages like C++ and Java were designed for non-XP development processes. Neither the languages nor the tools are up to the needs of XP. Trying to use them for XP is like trying to use a wrench as a hammer. But, then, industry has been trying to use wrenches as hammers for as long as we have had programming as a profession.
compared with teams in the same industries, XP teams are extreme in: number of tests, test coverage, frequency of code changes, frequency of design changes, speed of response to requirements changes, amount of time spent coding (vs in meetings), amount of documentation, and bug counts.
Industry is not the only place where programming happens. In fact, industry is far behind the curve when it comes to new methodologies.
XP is very close to the entire approach taken to programming by people in the Smalltalk and Lisp communities, practiced at least since the early 1980's in numerous academic and research labs.
Sales have been hurt largely by a surge in piracy,
Apparently, Mr. Green doesn't know the difference between a hypothesis and established facts. A professional journalist ought to know the difference and clearly indicate it in his writing. (His "according to the music industry" qualification only applies to the subsequent dollar amount.)
I'd propose an alternative hypothesis to the music industry's self-serving pronouncements, so uncritically cited as fact by Mr. Green: there is a limit to how interested people are in getting new music, and they can get the standard commercial stuff through more-and-more radio stations, on air, on cable, and on line.
By way of example, I know I have largely stopped buying CDs. I have all the stuff I really care about in legally purchased CDs (a few hundred), and for the rest, I mostly listen to the radio. Why would I want to pay $15-$20 for music CDs? If the prices came down to $3-$5 per CD, maybe I'd find it more convenient than the radio again. Until then, no thanks.
OSX's HFS+ and UFS seem a bit antiquated, and they are a pain to read from other machines as well. It would be great if OSX could support ReiserFS. Ext3 would also be a reasonable choice.
There is really nothing particularly "extreme" about "extreme programming". People have used the technique for decades. But in the 1990's, a lot of people made a name for themselves by giving new names to decades-old techniques ("generic programming", "patterns", "anti-patterns", "extreme programming", to name just a few).
There is some benefit to the naming and the book writing that surrounds it: more programmers get exposed to the ideas. But don't expect a magic bullet: people have had a lot more experience with those techniques than their recent new clothes suggest. For some problems and environment, they work well, for others they don't. Talk to experienced programmers from the 1980's for more advice:-)
Even if Lindows has done their homework technically and everything works smoothly (and that's a big if), I think Microsoft is not going to let this pass. If they perceive Lindows and similar systems as a threat, they are going to change the licensing agreements for MS Office, and they are going to add little tests here and there to make sure that they are running on genuine Microsoft Windows. It doesn't have to break the applications badly--just a few little inconveniences and problems (can't print in MS Office, can't find its plug-ins in IE) would be sufficient. And Lindows would get blamed.
Let's just hope that Linux doesn't go down in flames alongside with Lindows.
If you have the offer, what you can do is negotiate with your employer for a higher salary without telling them that you actually have another offer. That will give you a more reliable gauge of how much you are actually worth to them in the long term. If they are not meeting your expectations accept the other job and resign a few days later (it is both in bad taste and counterproductive to throw your resignation at them right after failed negotiations for a higher salary). If they make a counter-offer at that point, you then know that it's because of their short-term panic, not because you are worth it to them.
I hope that you aren't actually working for ADC (you don't show up anywhere on the Apple web site). If your opinions reflect Apple's strategy--trying to move people from open toolkits to Apple proprietary toolkits--it's doomed to failure. Apple got a lot of OSX users in academia and research recently because OSX is BSD compatible. But Apple can't take these people for granted; they will switch away from Apple before they start developing using proprietary Apple/NeXT APIs.
As for your other opinions, the resume on your web site lists no experience with X11. And you are just giving us the typical uninformed hot air of NeXT zealots about the supposed evils of X11, mistakenly implying even that X11 is a user interface.
Like any large, mature software system, X11 has its flaws, but X11 has a lot of strengths, too. X11 is a perfectly good substrate on which to build high quality user interfaces and it is the standard on which UNIX toolkits and applications are built. If Apple wants long-term buy-in from UNIX users and the open source community, they should make X11 a standard, transparently available component of OSX, alongside Carbon and Cocoa.
Your clairvoyance needs a bit of work. Let me just say that you have no idea who I'm talking to.
Neither you nor I know for certain whether the sample of developers you talk to is biased or not. But I do know that there is a large number of people developing software for OSX that never talk to you. I also know that most of the software I and my coworkers use on OSX was developed by such people.
"X11 is the equivalent of Quartz." Not hardly!
You are missing the point. I simply pointed out that X11 and Quartz are graphics libraries. The presence or absence of X11 on a platform has nothing to do with how user friendly it is because X11 is not a user interface, it's little more than a graphics library.
So, yes, it looks like we can use X for Quartz. All we need do is [...]
I think you are completely wrong in your assessment (most of the features you mention are already in X servers), but that's water under the bridge. For better or for worse, Apple invested lots of effort in building stuff on top of Quartz.
My point is that what you should do now is integrate X11 as another graphics API into the system, alongside Quartz. People will develop applications for OSX using X11 toolkits whether you want it or not. But by taking control and making this work well, you can improve the user experience.
Come back next year, and tell me if you still think so. In the meantime, have a look at the Mac OS X projects at sourceforge, and check out how many of them use Cocoa.
We don't have to guess about what Cocoa is like, we already know it. I don't think it has much of a future against systems based on Java or C# and their APIs and toolsets.
In any case, there are about 10 projects using Cocoa as far as I can tell, all of the minor ones (Google).
Most OSX related open source projects seem to be about adapting cross-platform libraries to work on OSX.
In all those cases, people should pay proportional to the costs (opportunity and otherwise) they impose on others. The fact that some people get a free ride and the public is paying for it in some cases doesn't make it right.
The best thing software vendors can do now is to keep EULAs really simple; otherwise, end users will distrust them. It also may strengthen EULAs associated with open source. Altogether, just like strict enforcements of copyrights, this is a win for open source.
I didn't claim they were being shut down. They are, however, being put under "Homeland Security", which makes absolutely no sense.
2. Defense Contractors don't fit into the Bush administration's philosophy? Wtf are you smoking?
Yes, the Bush administration likes defense contractors. But LLNL is not a "defense contractor", it's a national lab (as in "llnl.gov"), associated with a university. Conservatives probably would like to privatize it completely, or, even better, just funnel the research money to existing defense contractors.
Why do you think Chiao is a crackpot? Have you even read his paper?
As far as I can tell, Chiao's paper is pure speculation. If it doesn't work, it wouldn't falsify any of our current physical theories, it merely means that one of Chiao's guesses or assumptions is wrong.
I think both kinds of work are out on the fringe (which is what I mean by echoing the word "crackpot"). Both really just need more credible experimental evidence before one can take them seriously.
They suggest a screening of a static gravitational field, what Chiao calls a "gravitoelectric" field, which is what we observe in everyday life. [...] But it has nothing to do with Podkletnov's claims and I don't think he's obliged to cite them at all.
There are maybe a handful of published papers that claim observation gravitational effects from superconductivity. I think it's appropriate to mention them, even if it is just to make the point you did. (And if Podkletnov's effect were real, perhaps it could be explained using mechanisms other than shielding.)
What kind of skill do you think allows a 2 foot figher get within striking range of a 6 foot opponent? Jumping and spinning, of course.
And how do you suppose a 50 pound fighter is supposed to block a strike from a 200 pound fighter? If he just stands there and tries to block, he'll fly across the room like a golf ball.
Jumping may or may not be dignified, but it's what small opponents have to do against large opponents. Jumping constitutes skill.
People in Washington are just using terrorism as an excuse to push an agenda that has nothing to do with terrorism and existed long before 9/11. LLNL, for example, has been an irritation for the Bush administration, and that kind of government funded research doesn't fit too well into their philosophy anyway; that's why they like to play football with it.
No. I was there. NeWS was a buggy, unreliable, slow, poorly thought out piece of software. When it was shipping with Suns, just about everybody still ran MIT X11 even though that was a lot more effort. NeWS stopped shipping long after it was already a clear failure.
From where I sit (in Apple Worldwide Developer Relations), that's not what I'm seeing.
You are getting a biased sample because you talk to the traditional Mac developers and maybe some OpenStep developers. People who don't use Carbon or Cocoa for applications development have no need to talk to you.
I don't think you realize how much of a drag X windows has been on UNIX, despite the heroic efforts of SGI and others to make it usable.
That statement makes no sense. X11 is the equivalent of Quartz. You could put the current Mac UI on top of X11 and the only user-visible difference would be that it would run a whole lot faster than Cocoa on Quartz and that it would be network transparent. Furthermore, X11 won the UNIX market because of end user preferences; if it had been up to the workstation vendors, we'd be using DisplayPostscript, OpenLook, or something similar.
Nothing but the quality of the UI, which after all is a principal competitive advantage of the platform..
The quality of the UI doesn't depend on Quartz, Carbon, or Cocoa. The quality of the UI depends on user interface guidelines that people follow no matter what graphics API or toolkit they use.
There won't be a mass conversion to Cocoa. It's just not going to happen. Even assuming for the sake of argument that Cocoa is a good API, people just don't have the time or interest to develop to such a niche platform if they can just as easily use a toolkit that will work on all the major platforms. A large fraction of OSX applications, commercial, open source, and in-house, will be developed using cross-platform toolkits or X11, whether Apple likes it or not. The only choice Apple has in the matter is to help those toolkits and X11 to look their very best on the OSX desktop.
X11 and UNIX toolkits are crucial to the future of OSX. The more and the better you support them, the better the end user experience will be. If, on the other hand, you try to force people onto Cocoa, you'll just lose again many of the recent converts to OSX. As an OSX developer, I can only hope Apple won't make that mistake.
Mac OSX printing currently can be a headache, too. Many USB printer drivers can't print to remote printers. Some supported local USB printers stop working haphazardly and require power cycling the printer and restarting the print queue in some obscure little application in the Utilities folder.
Still, on the whole, OSX does work a lot better than Windows in my experience. Let's hope Apple will fix this pox, too. I believe Apple has licensed CUPS and is working on replacing the current printing system.
Apple should integrate X11 tightly into the OS, putting it on equal footing with Quartz and QuickDraw/Carbon. Apple should also sponsor an OSX look-and-feel for Gtk+ and ship Gtk+.
If Apple doesn't do this, they people won't magically become devoted Apple Cocoa developers. Rather, Apple will only create unnecessary porting headaches for their newest developers--UNIX developers. Those developers won't switch to Cocoa, they will simply continue using the same toolkits they have always been using (Gtk+, wxWindows, etc.), but with substandard and poorly maintained OSX-backends. That only hurts Apple.
I know it's tough medicine to swallow for Apple. But I really don't see any alternative. Hoping that the world will switch to Cocoa is a pipe dream--whether it is technically good or not, Cocoa is a niche product. Only a small, dedicated core of Mac developers will spend time on it.
We already get a bunch of shallow Carbon and QuickDraw ports. And we will be getting a bunch of shallow Win32 ports (via compatibility libraries), shallow Qt ports, shallow Gtk+, shallow wx ports, shallow Xlib, shallow Motif, and shallow FLTK ports. In different words, all the existing UNIX applications will be ported by putting a minimal Quartz backend onto their toolkits. UNIX and Linux developers have neither time nor interest investing a lot of effort in Cocoa/Quarts development (which isn't such a hot toolkit anyway).
Apple loses nothing by making X11 a standard component of every shipping Mac. Quite to the contrary, they make their machines easier to use by their fastest growing user segment: UNIX/Linux users and scientists/engineers. And even the rather primitive X11 implementation we are getting right now is faster than Quartz. Imagine how much more performance the Mac could gain if X11 were tightly integrated into the system.
If, on the other hand, Apple thinks they can grow a new community of developers devoted to their proprietary APIs, they are sadly mistaken. They are only losing sales that way.
This is science, not a sandbox brawl. You don't get to pick and choose who you cite based on who you don't like. Podkletnov's experimental results are published, peer reviewed, and seem related. If Chiao is aware of them (and he must be), he has an obligation to cite them. If he thinks Podkletnov's results have no bearing on the effects predicted by his own theory, he can explain so in his citation. If he thinks there are fundamental flaws with Pokletnov's experimental setup, he can explain them.
Personally, I think both are crackpots. But if crackpots publish scientific papers, they still should follow the rules of academic conduct, because the rules of academic conduct ultimately are what helps us sort out the real crackpots from the forward thinkers.
I think it's still an open question whether black holes (singularities in space-time) actually exist. All the current observations are compatible with many different kinds of theories. And there is no experimental evidence available at all, only observational evidence.
Slashdot Mirror
Sorry, but I disagree. XP is pretty much how a lot of Lisp, Smalltalk, and similar communities have worked for a long time, using pretty much the whole of what is now called XP, not just bits and pieces. That's not surprising: those communities face rapidly changing requirements, they often have easy access to their end users, and they are usually highly qualified and trained. And that's the community Kent Beck came from; he is just trying to popularize those practices and ideas among non-Smalltalk users.
I have my doubts that that is such a good idea. Languages like C++ and Java were designed for non-XP development processes. Neither the languages nor the tools are up to the needs of XP. Trying to use them for XP is like trying to use a wrench as a hammer. But, then, industry has been trying to use wrenches as hammers for as long as we have had programming as a profession.
Industry is not the only place where programming happens. In fact, industry is far behind the curve when it comes to new methodologies.
XP is very close to the entire approach taken to programming by people in the Smalltalk and Lisp communities, practiced at least since the early 1980's in numerous academic and research labs.
Seems like a discussion between two people who are a bit out of it...
Apparently, Mr. Green doesn't know the difference between a hypothesis and established facts. A professional journalist ought to know the difference and clearly indicate it in his writing. (His "according to the music industry" qualification only applies to the subsequent dollar amount.)
I'd propose an alternative hypothesis to the music industry's self-serving pronouncements, so uncritically cited as fact by Mr. Green: there is a limit to how interested people are in getting new music, and they can get the standard commercial stuff through more-and-more radio stations, on air, on cable, and on line.
By way of example, I know I have largely stopped buying CDs. I have all the stuff I really care about in legally purchased CDs (a few hundred), and for the rest, I mostly listen to the radio. Why would I want to pay $15-$20 for music CDs? If the prices came down to $3-$5 per CD, maybe I'd find it more convenient than the radio again. Until then, no thanks.
OSX's HFS+ and UFS seem a bit antiquated, and they are a pain to read from other machines as well. It would be great if OSX could support ReiserFS. Ext3 would also be a reasonable choice.
There is some benefit to the naming and the book writing that surrounds it: more programmers get exposed to the ideas. But don't expect a magic bullet: people have had a lot more experience with those techniques than their recent new clothes suggest. For some problems and environment, they work well, for others they don't. Talk to experienced programmers from the 1980's for more advice :-)
Let's just hope that Linux doesn't go down in flames alongside with Lindows.
Your geek and nerd cards are revoked immediately. Someone who doesn't know what an IMSAI or S-100 bus is just cannot be a geek or nerd.
If you have the offer, what you can do is negotiate with your employer for a higher salary without telling them that you actually have another offer. That will give you a more reliable gauge of how much you are actually worth to them in the long term. If they are not meeting your expectations accept the other job and resign a few days later (it is both in bad taste and counterproductive to throw your resignation at them right after failed negotiations for a higher salary). If they make a counter-offer at that point, you then know that it's because of their short-term panic, not because you are worth it to them.
As for your other opinions, the resume on your web site lists no experience with X11. And you are just giving us the typical uninformed hot air of NeXT zealots about the supposed evils of X11, mistakenly implying even that X11 is a user interface.
Like any large, mature software system, X11 has its flaws, but X11 has a lot of strengths, too. X11 is a perfectly good substrate on which to build high quality user interfaces and it is the standard on which UNIX toolkits and applications are built. If Apple wants long-term buy-in from UNIX users and the open source community, they should make X11 a standard, transparently available component of OSX, alongside Carbon and Cocoa.
Neither you nor I know for certain whether the sample of developers you talk to is biased or not. But I do know that there is a large number of people developing software for OSX that never talk to you. I also know that most of the software I and my coworkers use on OSX was developed by such people.
"X11 is the equivalent of Quartz." Not hardly!
You are missing the point. I simply pointed out that X11 and Quartz are graphics libraries. The presence or absence of X11 on a platform has nothing to do with how user friendly it is because X11 is not a user interface, it's little more than a graphics library.
So, yes, it looks like we can use X for Quartz. All we need do is [...]
I think you are completely wrong in your assessment (most of the features you mention are already in X servers), but that's water under the bridge. For better or for worse, Apple invested lots of effort in building stuff on top of Quartz.
My point is that what you should do now is integrate X11 as another graphics API into the system, alongside Quartz. People will develop applications for OSX using X11 toolkits whether you want it or not. But by taking control and making this work well, you can improve the user experience.
Come back next year, and tell me if you still think so. In the meantime, have a look at the Mac OS X projects at sourceforge, and check out how many of them use Cocoa.
We don't have to guess about what Cocoa is like, we already know it. I don't think it has much of a future against systems based on Java or C# and their APIs and toolsets.
In any case, there are about 10 projects using Cocoa as far as I can tell, all of the minor ones (Google). Most OSX related open source projects seem to be about adapting cross-platform libraries to work on OSX.
In all those cases, people should pay proportional to the costs (opportunity and otherwise) they impose on others. The fact that some people get a free ride and the public is paying for it in some cases doesn't make it right.
The best thing software vendors can do now is to keep EULAs really simple; otherwise, end users will distrust them. It also may strengthen EULAs associated with open source. Altogether, just like strict enforcements of copyrights, this is a win for open source.
I didn't claim they were being shut down. They are, however, being put under "Homeland Security", which makes absolutely no sense.
2. Defense Contractors don't fit into the Bush administration's philosophy? Wtf are you smoking?
Yes, the Bush administration likes defense contractors. But LLNL is not a "defense contractor", it's a national lab (as in "llnl.gov"), associated with a university. Conservatives probably would like to privatize it completely, or, even better, just funnel the research money to existing defense contractors.
As far as I can tell, Chiao's paper is pure speculation. If it doesn't work, it wouldn't falsify any of our current physical theories, it merely means that one of Chiao's guesses or assumptions is wrong.
I think both kinds of work are out on the fringe (which is what I mean by echoing the word "crackpot"). Both really just need more credible experimental evidence before one can take them seriously.
They suggest a screening of a static gravitational field, what Chiao calls a "gravitoelectric" field, which is what we observe in everyday life. [...] But it has nothing to do with Podkletnov's claims and I don't think he's obliged to cite them at all.
There are maybe a handful of published papers that claim observation gravitational effects from superconductivity. I think it's appropriate to mention them, even if it is just to make the point you did. (And if Podkletnov's effect were real, perhaps it could be explained using mechanisms other than shielding.)
And how do you suppose a 50 pound fighter is supposed to block a strike from a 200 pound fighter? If he just stands there and tries to block, he'll fly across the room like a golf ball.
Jumping may or may not be dignified, but it's what small opponents have to do against large opponents. Jumping constitutes skill.
People in Washington are just using terrorism as an excuse to push an agenda that has nothing to do with terrorism and existed long before 9/11. LLNL, for example, has been an irritation for the Bush administration, and that kind of government funded research doesn't fit too well into their philosophy anyway; that's why they like to play football with it.
No. I was there. NeWS was a buggy, unreliable, slow, poorly thought out piece of software. When it was shipping with Suns, just about everybody still ran MIT X11 even though that was a lot more effort. NeWS stopped shipping long after it was already a clear failure.
Tcl/Tk uses a kind of fake Xlib library. You can expect more complete Xlib emulations on OSX in the future.
You are getting a biased sample because you talk to the traditional Mac developers and maybe some OpenStep developers. People who don't use Carbon or Cocoa for applications development have no need to talk to you.
I don't think you realize how much of a drag X windows has been on UNIX, despite the heroic efforts of SGI and others to make it usable.
That statement makes no sense. X11 is the equivalent of Quartz. You could put the current Mac UI on top of X11 and the only user-visible difference would be that it would run a whole lot faster than Cocoa on Quartz and that it would be network transparent. Furthermore, X11 won the UNIX market because of end user preferences; if it had been up to the workstation vendors, we'd be using DisplayPostscript, OpenLook, or something similar.
Nothing but the quality of the UI, which after all is a principal competitive advantage of the platform..
The quality of the UI doesn't depend on Quartz, Carbon, or Cocoa. The quality of the UI depends on user interface guidelines that people follow no matter what graphics API or toolkit they use.
There won't be a mass conversion to Cocoa. It's just not going to happen. Even assuming for the sake of argument that Cocoa is a good API, people just don't have the time or interest to develop to such a niche platform if they can just as easily use a toolkit that will work on all the major platforms. A large fraction of OSX applications, commercial, open source, and in-house, will be developed using cross-platform toolkits or X11, whether Apple likes it or not. The only choice Apple has in the matter is to help those toolkits and X11 to look their very best on the OSX desktop.
X11 and UNIX toolkits are crucial to the future of OSX. The more and the better you support them, the better the end user experience will be. If, on the other hand, you try to force people onto Cocoa, you'll just lose again many of the recent converts to OSX. As an OSX developer, I can only hope Apple won't make that mistake.
Still, on the whole, OSX does work a lot better than Windows in my experience. Let's hope Apple will fix this pox, too. I believe Apple has licensed CUPS and is working on replacing the current printing system.
If Apple doesn't do this, they people won't magically become devoted Apple Cocoa developers. Rather, Apple will only create unnecessary porting headaches for their newest developers--UNIX developers. Those developers won't switch to Cocoa, they will simply continue using the same toolkits they have always been using (Gtk+, wxWindows, etc.), but with substandard and poorly maintained OSX-backends. That only hurts Apple.
I know it's tough medicine to swallow for Apple. But I really don't see any alternative. Hoping that the world will switch to Cocoa is a pipe dream--whether it is technically good or not, Cocoa is a niche product. Only a small, dedicated core of Mac developers will spend time on it.
Apple loses nothing by making X11 a standard component of every shipping Mac. Quite to the contrary, they make their machines easier to use by their fastest growing user segment: UNIX/Linux users and scientists/engineers. And even the rather primitive X11 implementation we are getting right now is faster than Quartz. Imagine how much more performance the Mac could gain if X11 were tightly integrated into the system.
If, on the other hand, Apple thinks they can grow a new community of developers devoted to their proprietary APIs, they are sadly mistaken. They are only losing sales that way.
Personally, I think both are crackpots. But if crackpots publish scientific papers, they still should follow the rules of academic conduct, because the rules of academic conduct ultimately are what helps us sort out the real crackpots from the forward thinkers.
I think it's still an open question whether black holes (singularities in space-time) actually exist. All the current observations are compatible with many different kinds of theories. And there is no experimental evidence available at all, only observational evidence.