 |
 | Home | Contact Us | FAQ | Search & Site Map | Link to Us |
 | Sign In | Join | Other 45 Sites in Network |
Mac Forum / Programming / CodeWarrior / December 2005CW10 is final - caveat emptor
The following text is a flame that I let sit overnight. Wanted to see if I really wanted to post it. Wanted to cool off. I didnt. I had just received "resolution" to my CW10 complaints about certain errors/warnings (see prior complaints about TYPE_BOOL, void*, and func ptr errors/warnings). It appears that I need to modify *many* files to get around errors passing UPP ptrs to Apple library calls (must add extern "C" to the func decl). And all the in-line assignments warnings inside an if () stmt must be unrolled. Its amazing how often that happens. And then there are the void* casts that the compiler no longer allows. Gee. Why. And, I was told that CW 10 is the "final release ... these behaviors will not change". What was that claim about a one year support in prior CW10 threads? Will any bugs/defects get fixed? I now have doubts... Maybe my expectations were simply naive. Ok, so now to test the suggested changes with my test CW10 project - my license has *EXPIRED*! Dang. This is what I paid for?! Ok. So maybe I got caught in the registration server fiasco of a few weeks ago. So I registered (again?) and am now awaiting a license to allow me to run again. - License just arrived, and CW10 will still not run. Something now about the license file being "altered"... Contact with Metrowerks, and another license.dat file arrives. And it now works... So do I want CW10? I dont think so... Can I get a refund on my 3 CW10 copies that will not be useful? We'll see... I will plan on reverting to CW9.5 and continue development there until Xcode catches up... I am sad, ticked, and tired. Sorry for the flame, but I need to vent somewhere. I love CW - the product is a testimony to how integrated build/debug systems can work. I am sorry we are at the end of a road. -harry > And, I was told that CW 10 is the "final release ... these behaviors > will not change". What was that claim about a one year support in prior > CW10 > threads? Will any bugs/defects get fixed? I now have doubts... Maybe my > expectations were simply naive. Well, Ron told me in another thread: > >That's entirely reasonable. I have to ask, though: Will there be > >maintenance updates of CodeWarrior 10? If not, I suspect people will [quoted text clipped - 3 lines] > normal after any release. I do not expect major updates would be > required nor would they be done for any OS System change. If true, this development is disturbing. I bought CodeWarrior on the understanding that what Ron said was going to be true. Ron, any comment? Alwyn In article <dt015a1979-F326E8.20395002122005@text.news.blueyonder.co.uk>, >> And, I was told that CW 10 is the "final release ... these behaviors >> will not change". What was that claim about a one year support in prior [quoted text clipped - 16 lines] > >Ron, any comment? My expectations were not realized. I did authorize a full refund of your money. Metrowerks just released this product to make things easier for the developers community. Ron  SignatureCodeWarrior Community Forum is a free online resource for developers to discuss CodeWarrior topics with other users and our staff -- http://www.codewarrior.com/community -- Ron Liechty - ron.liechty@freescale.com - http://www.codewarrior.com Exactly what "expectation" was not realized? Are there going to be no bug fixes? If not, how is a product that crashes supposed to make things easier for the developer community? FWIW, I'm getting some crashes while editing: Exception: EXC_BAD_ACCESS (0x0001) Codes: KERN_INVALID_ADDRESS (0x0001) at 0x0c59a000 Thread 0 Crashed: 0 IDE_MSL 0x02a3a7b4 0x2a299b0 + 0x10e04 1 CodeWarrior IDE 0x023fff54 0x2008000 + 0x3f7f54 2 CodeWarrior IDE 0x023f8258 0x2008000 + 0x3f0258 3 CodeWarrior IDE 0x023f69d8 0x2008000 + 0x3ee9d8 4 CodeWarrior IDE 0x022bf4f4 0x2008000 + 0x2b74f4 5 CodeWarrior IDE 0x023e9fdc 0x2008000 + 0x3e1fdc 6 CodeWarrior IDE 0x023ebc84 0x2008000 + 0x3e3c84 7 CodeWarrior IDE 0x023ebb68 0x2008000 + 0x3e3b68 8 CodeWarrior IDE 0x02163584 0x2008000 + 0x15b584 9 CodeWarrior IDE 0x023fb6cc 0x2008000 + 0x3f36cc 10 CodeWarrior IDE 0x023fb4dc 0x2008000 + 0x3f34dc 11 CodeWarrior IDE 0x02719d7c 0x2008000 + 0x711d7c 12 CodeWarrior IDE 0x0203a704 RawKeyInputNeeded__10LCommanderFv + 0xfffffd98 13 CodeWarrior IDE 0x0271baa8 SetBuffer__11LDataStreamFPvl + 0xffffed28 14 CodeWarrior IDE 0x0203991c RawKeyInputNeeded__10LCommanderFv + 0xffffefb0 15 CodeWarrior IDE 0x0203ae20 0x2008000 + 0x32e20 16 CodeWarrior IDE 0x02038454 0x2008000 + 0x30454 17 CodeWarrior IDE 0x02047400 0x2008000 + 0x3f400 Larry > Exactly what "expectation" was not realized? Are there going to be no > bug fixes? If not, how is a product that crashes supposed to make > things easier for the developer community? It's not like CW9 was rock-solid, either. So half of one, six dozen of another. Nor was anyone buying CW for the IDE anyway -- CW's strengths were mainly in the C++ compiler and linker, and the C++ library, and also in the plc compiler and some aspects of the debugger. And let me just say that having changed over to Xcode in the last six months, Xcode totally blows when it comes to writing any serious C++ code. This is mainly the fault of gcc and Apple's linker, both of which have caused me misery that is unparalleled by anything other than the misery inflicted by prereleases of ProjectBuilder I had to use while working on a part of the OS for Apple, back in the day. (I am not going into the details of what the problems are, because they are lengthy and I don't feel like getting into a rant. I will, however, say that my complaints have been acknowledged by people at Apple and elsewhere to be legitimate problems with Apple's toolchain so severe that there are major software companies that are *completely unable* to produce a working executable with Apple's toolchain, while their CW-compiled executables have been used by probably everyone on this newsgroup at one time or another.) So, yeah, boo-freakin-hoo if you are clueless enough to still let your source of income depend on MW's dead technology, but also boo-freakin-hoo if you are foolish enough to believe that Xcode is a mature development environment. Sucks either way. Pick your poison. Ben SignatureIf this message helped you, consider buying an item from my wish list: <http://artins.org/ben/wishlist> I changed my name: <http://periodic-kingdom.org/People/NameChange.php> > In article > <dt015a1979-F326E8.20395002122005@text.news.blueyonder.co.uk>, [quoted text clipped - 17 lines] > your money. Metrowerks just released this product to make things easier > for the developers community. Ron, that is very gratifying, as I would not have bought the product had it been known that there would be no bugfixes ever. Do I need to do anything further to claim my refund, or will it happen automatically? Alwyn I don't want a refund, but I wouldn't mind understanding exactly what Ron's connection is to his employer. It seems like he's just a guy outside the company who manages to be on their payroll. I've long since lost track of all the times he's said something as a so-called representative of MW/FS that turned out not to be true. Then we hear some apology because what he thought, hoped, or expected turned out not to be true. In the past two years he's been wrong more often than he's been right. Are there any more openings for company representatives that just offer their best personal guesses instead of actually representing the company? How much does that kind of work pay? I know there are a lot of people here who think Ron is just the greatest thing since sliced bread, but honestly, as far as I'm concerned, a person with real integrity wouldn't regularly mislead people, nor would he continue to work for a company that regularly put him in a position where he would be misleading people. Larry > I know there are a lot of people here who think Ron is just the > greatest thing since sliced bread, but honestly, as far as I'm > concerned, a person with real integrity wouldn't regularly mislead > people, nor would he continue to work for a company that regularly put > him in a position where he would be misleading people. Having been at a university in Chicago and knowing several people who work(ed) for Motorola (aka Freescale), I have heard stories about a general chaos management I heard from *no* other company. In fact, I got a taste of it once, many years ago when they were showing senior students around who worked on research they were funding, and one mid-level manager didn't like an another (the one was showing us around). So we had to use backdoors to avoid the first manager's zone of influence. That was when they were still desperately seeking employees (late 1990s), but it looked more like a turf war zone too me and other students they were showing around. As such, I give Ron a lot of credit sticking around such a mess, assuming it is as bad in the unit he is working for... Thorsten >I don't want a refund, but I wouldn't mind understanding exactly what >Ron's connection is to his employer. It seems like he's just a guy [quoted text clipped - 6 lines] >that just offer their best personal guesses instead of actually >representing the company? How much does that kind of work pay? During the last 3 years Metrowerks (Developers Technology) has had 5 different president / directors. We took a big hit in Motorola's and then Freescale's manpower cuts and went from a company of over 720 to about 300. So yes, I said a lot of things 3 years ago that were later proven wrong. I was promised things and they were changed when the next president came. However, I believed then and I believe now that Metrowerks has always been a partner to the developers community and fundamentally did what was right for their customers. In hindsight we should have closed up with CodeWarrior 9 a year ago when we didn't go to WWDC. But I'm happy to be a part of what I consider an honorable closure of the Macintosh product. Ron SignatureCodeWarrior Community Forum is a free online resource for developers to discuss CodeWarrior topics with other users and our staff -- http://www.codewarrior.com/community -- Ron Liechty - ron.liechty@freescale.com - http://www.codewarrior.com > However, I believed then and I believe now that Metrowerks has always > been a partner to the developers community and fundamentally did what > was right for their customers. I would say rather that Metrowerks *employees* have always attempted to do right by their customers. I've often seen employees bend over backwards to help customers, often despite official policies. > In hindsight we should have closed up with CodeWarrior 9 a year ago > when we didn't go to WWDC. But I'm happy to be a part of what I > consider an honorable closure of the Macintosh product. The writing has been on the wall for some time. Were I still using Codewarrior, I know I'd be pleased to have the epilogue, I just wouldn't be *relying* on it... -- Steve >> I would say rather that Metrowerks *employees* have always >> attempted to do right by their customers. I've often seen >> employees bend over backwards to help customers, often >> despite official policies. I wholeheartedly agree. At this year's MacHack, Ron & Josef Wankerl showed up to tell us the woe about dropping support. These guys have taken a lot of sh*t for their company (a lot of it from me, too), and they were just as disillusioned by this dropping-the-ball-on-Intel thing as we were. After their session, they bought beers for everyone there (on Metrowerks' dime), just to say thanks for the past 12 years. A class act. In fact it was Josef who pushed try to to get one final version of CodeWarrior out the door, without a lot of support from Freescale management. A final inexpensive version, even without all the pieces they were hoping to put in, was better than stopping at an overpriced CW 9. Metrowerks will always remain in my mind the "good" company, in which Greg Galanos' original dream of a great Mac compiler saw its fruition, and guys like Josef, Ron, Greg and others tried to give back to the Mac development community. The "evil" came with Freescale, as they sucked the soul out of the company, disheartened their employees and customers, and took a virtual compiler monopoly into cancellation in a mere two years. I doubt Freescale management can cast a reflection in the mirror anymore. Jonathan Hoyle Eastman Kodak > In fact it was Josef who pushed try to to get one final version of > CodeWarrior out the door, without a lot of support from Freescale > management. A final inexpensive version, even without all the pieces > they were hoping to put in, was better than stopping at an overpriced > CW 9. But CW 10 is perhaps not such a good bargain when you consider that bugs will never be remedied and the level of support for it is questionable, to put it mildly. > Metrowerks will always remain in my mind the "good" company, in which > Greg Galanos' original dream of a great Mac compiler saw its fruition, And it was Mr Galanos who sold Metrowerks to Motorola at a very good price. I've never understood why Motorola should have been so interested in producing development systems for processors and hosted systems not manufactured by them. > and guys like Josef, Ron, Greg and others tried to give back to the Mac > development community. The "evil" came with Freescale, as they sucked > the soul out of the company, disheartened their employees and > customers, and took a virtual compiler monopoly into cancellation in a > mere two years. You can blame the loss of monopoly fairly and squarely on Apple and the open source movement. It has been shown over and over again that it is virtually impossible to make a profit on development tools, especially in an environment where a comparable toolset is offered for free. > I doubt Freescale management can cast a reflection in the mirror > anymore. I believe CodeWarrior users have a lot to thank Freescale's management for. From a profitability point of view, the obvious thing to do would have been to cancel development of CodeWarrior for Mac as soon as the free Xcode appeared. It would not have been good for developers or their users, but it is almost always sensible corporate policy to get rid of non-profitable activities that are not part of the core business. Alwyn > It has been shown over and over again that it is > virtually impossible to make a profit on development tools, especially > in an environment where a comparable toolset is offered for free. Except that XCode is no comparable toolset, but a significantly inferior one (for professional developers aka those who were 'still' paying full price for CW). Thorsten >> Except that XCode is no comparable toolset, but a >> significantly inferior one (for professional developers >> aka those who were 'still' paying full price for CW). Excellent point. Consider that throughout CodeWarrior's life, MPW was being given away freely. To blame Apple for the death of CodeWarrior is unfair. Had Metrowerks continued to show up at WWDC's, continued to evolve their tools set to keep up with Apple, kept the x86 compilers so it could produce Universal Binaries as well, I believe CodeWarrior would be alive and strong today. I firmly believe it was Freescale, not Apple, who played the biggest part in CodeWarrior's death. Jonathan Hoyle Eastmnan Kodak In article <1134251318.078407.234490@g44g2000cwa.googlegroups.com>, "Jonathan Hoyle" <jonhoyle@mac.com> wrote: > To blame Apple for the death of CodeWarrior is unfair. It would have died anyway. Apple merely hastened its demise by offering stiff competition, especially with Cocoa. > Had Metrowerks > continued to show up at WWDC's, continued to evolve their tools set to > keep up with Apple, kept the x86 compilers so it could produce > Universal Binaries as well, I believe CodeWarrior would be alive and > strong today. And still losing money. I believe the only time CodeWarrior was a money spinner was when Metrowerks were getting generous subsidies from Apple. Apple clearly no longer need it. > I firmly believe it was Freescale, not Apple, who played the biggest > part in CodeWarrior's death. CodeWarrior for Mac never made any sense from a Freescale perspective. If they could have sold it, as they sold the x80 stuff, I'm sure they would have done so. But there were no buyers. Eventually it had to be given the *coup de grace*; I'm only surprised it took so long. I'm not saying that is is a good thing; on the contrary, lack of choice is always bad, and a good case can be made for the superiority of various aspects of CodeWarrior. But its demise was a foregone conclusion ever since the release of OS X with Xcode. Alwyn Are all of the developers of Mac OS X at Apple 'unprofessional'? They all use Xcode. It seems you're being a little dramatic... > You can blame the loss of monopoly fairly and squarely on Apple and the > open source movement. It has been shown over and over again that it is > virtually impossible to make a profit on development tools, especially > in an environment where a comparable toolset is offered for free. Sadly this seems to be the case. Personally, it seems pretty stupid to me. If my livelihood depends on using using a tool 40 hours a week most weeks of the year, it just seems crazy to me to use an inferior tool to save $200-$300/year. That's about a dollar a day, and I'm not going to use something inferior to save a dollar a day. If you're making $60/hour, a minute of your time is worth $1, and I can easily save a minute a day with the faster compiles. And yet as soon as Project Builder was available, scads of people were dumping CW because they hating paying for it. Even now Xcode is notorious for its sluggish compile speed, and I haven't heard anyone claim that its STL is as good as CW's. Xcode has no cross-platform option either. If MW had really pushed to keep CW cutting edge, who knows, maybe people would have stuck with it a lot longer. But any chance of that really went down the tubes when all they really did for years was keep it useable. Larry > > You can blame the loss of monopoly fairly and squarely on Apple and the > > open source movement. It has been shown over and over again that it is [quoted text clipped - 8 lines] > $60/hour, a minute of your time is worth $1, and I can easily save a > minute a day with the faster compiles. True enough, but you're missing the whole picture. When Mac OS X was released, only Xcode supported the native Mach-O format, and, more important still, it was quite a while before one could develop Cocoa programs with CodeWarrior. CodeWarrior was thus perceived by many to be hopelessly outmoded. > And yet as soon as Project Builder was available, scads of people were > dumping CW because they hating paying for it. Even now Xcode is > notorious for its sluggish compile speed, and I haven't heard anyone > claim that its STL is as good as CW's. Xcode has no cross-platform > option either. I agree about the compilation speed and the C++ standard library, but I would imagine that the cross-platform thing was a minority interest. (And today, of course, CW is anything but cross-platform.) > If MW had really pushed to keep CW cutting edge, who knows, maybe > people would have stuck with it a lot longer. But any chance of that > really went down the tubes when all they really did for years was keep > it useable. My argument has been that Motorola and Freescale were hardware companies, who could hardly have regarded software development tools as their core business. The contract for Motorola to take over Metrowerks was the death warrant for CodeWarrior for Mac, and I would have thought that execution would have followed immediately on the release of Xcode. But it took many more years for it to be killed off, and I wonder why. Inertia (i.e. bad management) is the only reason I can think of. My point to Jon was that, rather than blast Freescale managers for their incompetence, he should thank them for it, paradoxical as it may seem. Alwyn > True enough, but you're missing the whole picture. When Mac OS X was > released, only Xcode supported the native Mach-O format, and, more > important still, it was quite a while before one could develop Cocoa > programs with CodeWarrior. CodeWarrior was thus perceived by many to be > hopelessly outmoded. Oh no, I'm not missing that part of the picture. I covered that part of the picture in my comment about not keeping CW cutting edge. >> And yet as soon as Project Builder was available, scads of people were >> dumping CW because they hating paying for it. Even now Xcode is [quoted text clipped - 4 lines] > I agree about the compilation speed and the C++ standard library, but I > would imagine that the cross-platform thing was a minority interest. Probably, but the more it offers the bigger its market (especially if the platform you stop supporting is Windows), and, perhaps more important, the more it positions itself as *the* preeminent tool of its kind. Kill Pascal support, kill Java support, kill cross-platform support, and not only do you end up with a tool that appeals to fewer people, but you have a tool that's perceived differently by every potential customer, and in this case one of those perceptions is that it's withering on the vine. I was talking to a guy in the beef department of my local Whole Foods store the other day, and he was telling me that they don't make anything on their high end cuts of beef. They even lose money on some of them. But if they don't carry them it will cost them customers. So they lose money on every filet mignon they sell, but they sell them because if they didn't, it would negatively affect people's perception of the store as a whole. > (And today, of course, CW is anything but cross-platform.) Today CW on track to be no-platform. >> If MW had really pushed to keep CW cutting edge, who knows, maybe >> people would have stuck with it a lot longer. But any chance of that [quoted text clipped - 8 lines] > But it took many more years for it to be killed off, and I wonder why. > Inertia (i.e. bad management) is the only reason I can think of. I honestly don't know what was going on there, so I resist passing any kind of judgement on it. I'm just sad. I'm sad I'll have to stop using something I've enjoyed using for nearly 10 years. Sad that we're essentially stuck with no options for our primary tools. And sad that said tools (Xcode and Interface Builder) are both less than they should be. I shutter to think what Xcode would be like today if CW had been dropped four or five years ago. Larry > > My argument has been that Motorola and Freescale were hardware > > companies, who could hardly have regarded software development tools as [quoted text clipped - 6 lines] > I honestly don't know what was going on there, so I resist passing any > kind of judgement on it. I don't know either; it's just my best guess, based on snippets I have read from supposedly authentic sources, and my first-hand experience of how big business works. I'm sure Ron could put me right if he wanted to (but of course he won't!). > I'm just sad. I'm sad I'll have to stop using > something I've enjoyed using for nearly 10 years. Sad that we're > essentially stuck with no options for our primary tools. And sad that > said tools (Xcode and Interface Builder) are both less than they should > be. I shutter to think what Xcode would be like today if CW had been > dropped four or five years ago. For what it's worth, I completely share your sentiments. Alwyn > > (And today, of course, CW is anything but cross-platform.) > > Today CW on track to be no-platform. CW does many platforms, just not Mac. :( > And yet as soon as Project Builder was available, scads of people were > dumping CW because they hating paying for it. Even now Xcode is > notorious for its sluggish compile speed, and I haven't heard anyone > claim that its STL is as good as CW's. Xcode has no cross-platform > option either. I have been told that if you run CW10 (which is slower than CW9) and Xcode 2.2 side-by-side on a quad G5 they are actually of comparable speed (because Xcode, unlike CW, takes advantage of multiple processors). Whether you think that's wonderful or sad is your call :-) Also, Apple hired Howard Hinnant to work on STL, and we all know that he's the reason why MW STL is awesome. Ben SignatureIf this message helped you, consider buying an item from my wish list: <http://artins.org/ben/wishlist> I changed my name: <http://periodic-kingdom.org/People/NameChange.php> > have been told that if you run CW10 (which is slower than CW9) and Xcode 2.2 > side-by-side on a quad G5 they are actually of comparable speed (because > Xcode, unlike CW, takes advantage of multiple processors). Yes, why on earth did CW never do this? > Whether you think that's wonderful or sad is your call :-) Since I don't have a quad G5, definitely sad. LOL > Also, Apple hired Howard Hinnant to work on STL, and we all know that he's the > reason why MW STL is awesome. Well, at least that's encouraging. I'd love to hear what he's muttering. LOL Larry > > have been told that if you run CW10 (which is slower than CW9) and Xcode > > 2.2 side-by-side on a quad G5 they are actually of comparable speed > > (because Xcode, unlike CW, takes advantage of multiple processors). > > Yes, why on earth did CW never do this? It's really hard to convert a gigantic source base to be MP-safe, I am sure you understand that. MP started becoming relevant when Mac OS X came out, just around the time that CW started its slide into irrelevance. While Ron has in the past claimed that they looked into it and that parallelism made CW slower, I am not inclined to take that statement particularly seriously -- certainly, I would expect the compilation to be faster than on two CPUs than on one, but not by a factor of two due to the overhead of locking and messaging between threads. Ben SignatureIf this message helped you, consider buying an item from my wish list: <http://artins.org/ben/wishlist> I changed my name: <http://periodic-kingdom.org/People/NameChange.php> > > > have been told that if you run CW10 (which is slower than CW9) and Xcode > > > 2.2 side-by-side on a quad G5 they are actually of comparable speed [quoted text clipped - 4 lines] > It's really hard to convert a gigantic source base to be MP-safe, I am sure > you understand that. Not many people do. A carefully considered single-threaded implementation can both be more correct and higher performance than a poorly considered MP implementation. Threading correctly is hard... > > > (because Xcode, unlike CW, takes advantage of multiple processors). > > > > Yes, why on earth did CW never do this? > > It's really hard to convert a gigantic source base to be MP-safe The IDE was designed with some thought to multi-threading and even in its current state holds together amazingly well on multi-processor machines. (I.e. it is possible to simultaneously edit, debug, have multiple finds running, and build with multiple compiles running.) For the Mac OS IDE the challenge is finding compiler plugins that are willing to go concurrent. (When one or more concurrent-willing plugin is present the Concurrent Compiles preference panel is visible.) Dave > > > > (because Xcode, unlike CW, takes advantage of multiple processors). > > > [quoted text clipped - 10 lines] > willing to go concurrent. (When one or more concurrent-willing plugin is > present the Concurrent Compiles preference panel is visible.) That's interesting to know, though I think it only changes my "hard to convert a gigantic source base" (the IDE) to "hard to convert an enormous source base" (the compiler) :-) Ben SignatureIf this message helped you, consider buying an item from my wish list: <http://artins.org/ben/wishlist> I changed my name: <http://periodic-kingdom.org/People/NameChange.php> > > > > > (because Xcode, unlike CW, takes advantage of multiple processors). > > > > [quoted text clipped - 14 lines] > gigantic source base" (the IDE) to "hard to convert an enormous source base" > (the compiler) :-) Fair enough! Indeed, the popular approach for concurrent compiles has each run in a separate process which lets the compiler sidestep the whole threading issue. Concurrency introduces a whole new set of GUI issues though--the Build Progress window gets a bit unwieldy with a couple of dozen rows.. Perhaps someone will try the Apple gcc under the CodeWarrior IDE and see how many gears they can keep spinning. (Here, having more processors really helps.) Dave > > Also, Apple hired Howard Hinnant to work on STL, and we all know that he's > > the > > reason why MW STL is awesome. > > Well, at least that's encouraging. I'd love to hear what he's > muttering. LOL After being so busy I hadn't had a chance to read the newsgroups for a week, I'm muttering: Thank you Ben. That was very nice to say. -Howard > I would say rather that Metrowerks *employees* have always attempted to > do right by their customers. I've often seen employees bend over > backwards to help customers, often despite official policies. Correct. As one insider put it a few months ago problems at MW were entrenched for several years: http://macslash.org/comments.pl?sid=5447&op=Reply&threshold=0&commentsort=0&mode =thread&pid=95635 Freescale's apathy for MW and Apple was a direct result of losing billions shortly after the take over of MW which was small change by comparison. Well, it seems Metrowerks dev tools are sinking. And many people from the "real" Metrowerks seem to be jumping the Titanic called Freescale. In August I asked Ron: >> will >>there be continued interested from MW in the evolution of C++? What >>about your C++ library developers, will they go elsewhere? And his answer was: >They will continue to be at Metrowerks. Today, on the Boost mailing list I found: >>>"John Maddock" <john@johnmaddock.co.uk> wrote in message news:000501c5fb50$d1515150$d2071452@fuji... > I'm trying to figure out what configuration magic is needed to get the > Boost.TR1 library working with CW-9.4. John, Howard Hinnant used to be the Metrowerks library guy, but he is now working for Apple. You might contact him privately and ask if he has any idea who to direct Metrowerks questions to nowadays. <snip> <<< > Howard Hinnant used to be the Metrowerks library guy, but he is now > working for Apple. Howard Hinnant is a world expert on the C++ standard library. His appointment bodes well for Apple and perhaps also libstdc++. Alwyn > Well, it seems Metrowerks dev tools are sinking. Gosh, are you sure? I think there is still some room for doubt. Ben SignatureIf this message helped you, consider buying an item from my wish list: <http://artins.org/ben/wishlist> I changed my name: <http://periodic-kingdom.org/People/NameChange.php> Well, it seems Metrowerks dev tools are sinking. And many people from the "real" Metrowerks seem to be jumping the Titanic called Freescale. In August I asked Ron: >> will >>there be continued interested from MW in the evolution of C++? What >>about your C++ library developers, will they go elsewhere? And his answer was: >They will continue to be at Metrowerks. Today, on the Boost mailing list I found: >>>"John Maddock" <john@johnmaddock.co.uk> wrote in message news:000501c5fb50$d1515150$d2071452@fuji... > I'm trying to figure out what configuration magic is needed to get the > Boost.TR1 library working with CW-9.4. John, Howard Hinnant used to be the Metrowerks library guy, but he is now working for Apple. You might contact him privately and ask if he has any idea who to direct Metrowerks questions to nowadays. <snip> <<< In article <dt015a1979-B3486B.14594406122005@text.news.blueyonder.co.uk>, >> In article >> <dt015a1979-F326E8.20395002122005@text.news.blueyonder.co.uk>, [quoted text clipped - 21 lines] >it been known that there would be no bugfixes ever. Do I need to do >anything further to claim my refund, or will it happen automatically? Send me your acknowledgement letter from Digital river Ron SignatureCodeWarrior Community Forum is a free online resource for developers to discuss CodeWarrior topics with other users and our staff -- http://www.codewarrior.com/community -- Ron Liechty - ron.liechty@freescale.com - http://www.codewarrior.com > I had just received "resolution" to my CW10 complaints about certain > errors/warnings (see prior complaints about TYPE_BOOL, void*, and func > ptr errors/warnings). As was pointed out to you in the thread you started in thiy very group about this, you are getting warnings in your code about illegal assumptions in your code. Are the warnings new? Yes. Is this a bug in CW? No. Is this a potential bug in your code? Yes! Your code is poor. - Face it, and turn off the warning. If you don't like turning off warnings, well, then you will have to fix your poor code. You should not be complaining, but rather thank MW that this warning was added to CW 10. Thorsten > It appears that I need to modify *many* files to get around errors > passing UPP ptrs to Apple library calls (must add extern "C" to the > func decl). I use lots of UPPs and I never use extern "C" declarations. > And all the in-line assignments warnings inside an if () stmt > must be unrolled. Its amazing how often that happens. And then there > are the void* casts that the compiler no longer allows. Gee. Why. I don't know why you're having these problems. I have about 700 files in my project and I only had to do a minor amount of tweaking to get it building in CW10. I haven't seen any problems with void* casts. > And, I was told that CW 10 is the "final release ... these behaviors > will not change". What was that claim about a one year support in prior > CW10 threads? I don't recall any promise of a full year of support. > Will any bugs/defects get fixed? I now have doubts... Maybe my > expectations were simply naive. At this point I'm not convinced the problems you're seeing are bugs in CW. It sounds more like issues with your code. > So do I want CW10? I dont think so... Can I get a refund on my 3 CW10 > copies that will not be useful? We'll see... > > I will plan on reverting to CW9.5 and continue development there until > Xcode catches up... That could be a while. LOL Larry > > And, I was told that CW 10 is the "final release ... these behaviors > > will not change". What was that claim about a one year support in prior > > CW10 threads? > > I don't recall any promise of a full year of support. Well, this is what Ron said at the time: > You may ask how long will CodeWarrior for Mac OS be supported > We will have e-mail support until the end of the year just as it > currently is provided. If release slips we will extend this. > We will have support on newsgroups, and CodeWarrior Forums at least for > a full year after CodeWarrior 10 is released <news:ron.liechty-BE1299.12150101082005@news.newsguy.com> So it's probably true that the OP had a right to expect a year's support when he bought the product. However, they now have on their CodeWarrior 10 Product Page (in small print, nota bene): > This product is sold without support and will not be sold after the first > quarter of 2006. <http://www.metrowerks.com/MW/Develop/Desktop/Mac10.htm> Caveat emptor indeed! Alwyn |
Reply to this Message |
 Sign In
 Join
 My Latest Posts My Monitored Threads My Blog My Photo Gallery My Profile My Homepage Start New Thread Enable EMail Alerts Rate this Thread
|
©2008 Advenet LLC Privacy Policy - Terms of Use
This website includes both content owned or controlled by Advenet as well as content owned or controlled by third parties.