Boost is a very powerful solution in search of a problem. It's fun to see what the experts can do by hacking C++ "features" in unintended ways, but it's usually not something I want in our production code. I allow STL containers, but other parts of the library have to be discussed; boost is pretty much off the table, as is 99% of template metaprogramming-type stuff, particularly expression templates. For that matter, template support in VC++ is quirky enough (and changes enough between versions) that we're pretty cautious about it beyond the basics.
If you really need GC or boost features, most functional and/or scripting languages do it far better and simpler than any bolted on kludge to C++ ever will, and are easier to learn and more fun to use. Don't try to make C++ into something it isn't, just because more and more people don't need the kind of power it offers any more, and don't want to take the time to learn it properly. Proficiency does take time, but once you know the syntax the most important skill required to use modern C++ effectively could be summed up as:
Given the choice between 50 lines of dead simple code that anyone can understand, and 5 lines of gibberish that often takes just as long to write, can't be debugged, can't be maintained, and takes noticeably longer to compile, go with the simple solution every time (for an example, look for boost's swanky Spirit parser on USENET).
We call it "Aim low and SHIP!" And did I mention, I use MACROS?
When I first read about Scratch, I thought of Prograph.... Looking at the video, I thought of a fancied up version of iShell or one of the presentation builders.
Problem I've seen w/ all of the above is that they make programming excruciatingly slow and clumsy if you know what you're doing, and don't help if you don't: there is no evidence that programming can be taught in the absence of a (probably genetic) talent for it. iShell tried hard to empower content producers to script their work, and failed like all the others: the target users were still helpless, and the programmers they hired usually preferred the much more complicated, non-graphical Lingo because they got their work done faster.
For small children starting from ground zero, it might make sense. Users old enough to grasp Python (for example) will quickly either get frustrated and move on to a "real" productivity-oriented programming environment, or find out they can't understand what's going on no matter how many tutorials they do, and give up. Those who do have the gift, though, may as well skip the first step and the rest will find out where they stand quicker if they try a "hard" language to start with. Beginning programming classes are (or should be) about separating the sheep from the goats as efficiently as possible.
I've followed Bussard's work in this area, and we're darn lucky to have him. Working for the Navy makes me nervous, though.
It's always been rumored that the Farnsworth fusor was buried (and it was, big time and deliberately) because it looked like it might work. While that device would probably never have become economically feasible as a power generator, there's not much likelihood the current Tokamak-based designs will either, and they're getting billions for research worldwide. One theory is that Farnsworth's method (a direct ancestor of Bussard's) was too easy to downscale to town or neighborhood size, where a working Tokamak would require an enormous plant that only government or big industry can build (and control.)
If the bad guys want to do the same thing again, it would be awfully easy to just classify Bussard's work (which is not yet practical for anything, and may never be), say it failed, and let it be forgotten. Or maybe just hide it until we're up against the wall (fossil fuel and uranium getting too expensive, breeders still won't work, other fusion research still going nowhere) when it would be the last hope of staving off the apocalypse. Maybe the governments will be sufficiently in Control by then that they'd risk releasing such disruptive technology.
I bought one for my girlfriend for Christmas. They were long since out of stock in the catalogs by then, but I found one for sale at a little gas station. The owner told me he had ordered one for himself, using his business address, and had promptly gotten a letter: "We have no dealers in your area, would you be interested...." It was the last one he had, he ordered 3 at a time and kept them under his counter.
I set the thing up (it was the week before Christmas, and I wrap fast) on my parents' kitchen counter using a 9 inch B&W TV they had. I made a display of 3 linked rotating rings (they each had a gap that made the "rotation" visible) that completely blew my father's mind -- the idea that something like that was possible. Compared to the punched cards and half day delays I was used to at college, it was a darned nice system.
BSD: Anyone can use your code, any way they want, as long as they mention your name. GPL: Anyone can use your code, as long as they mention your name and allow others to use the code as you have let them use it. LGPL: Same as GPL, but with some exceptions allowing others not to be bound to your licence if they don't actually touch your code in their project, but just use it.
Additional thoughts (assuming that you are interested in profit from your work):
Unless forced (i.e., you used someone else's GPL code), free software licenses are only useful if you're prepared to release your work publicly. If you own 100% of the code and it contains no algorithms or data that you can't / don't want to reveal, there are good reasons to use a free license even if you don't buy into the Free software philosophy (and the definition of that is debatable, controversial, and not intuitive.)
GPL If you want to profit from your work, use the GPL. This gains you a user base as fast as possible, respect from the FOSS community, and forces anyone who uses your code to also release their own, or pay you for a different license. GPL provides fair protection from piracy (everyone knows what it is, and there is substantial recourse against violators) but allows you "sell" your work on terms acceptable to commercial developers. You may even be able to "reverse license" GPL'ed improvements from customers, allowing you distribute and sell improved versions you didn't have to write in exchange for a reduced license fee.
A prime example is http://trolltech.com/. Their Qt product contains little of a "trade secret" nature, but is far too big and complex to recreate for the price of a license. Trolltech has contributed greatly to Linux (in particular) by providing a high end UI system for free. This costs them nothing (no sales were lost), and has resulted in thousands of coders becoming both experienced with and enthusiastic about Qt, some of whom may need the commercial version some day.
LGPL forces any changes to your work to be open, but work that uses your work can still be closed, if the developer jumps through the right legal hoops to draw a line between your code and theirs. This tends to make commercial users nervous, while not encouraging them to buy a regular license (even though it's probably cheaper than their legal cost to feel safe.) Given that your source code is going to be released publicly in either case, LGPL has no real advantage over the GPL.
BSD is the "real" free license. It's more honest than the GPL in that you don't proclaim Freedom while forcing your license terms on everyone else, but it's pretty unlikely that you'll get paid other than from contract work for improvements.
How does this work when the standard is "first to file?" It sounds like prior art is irrelevant, which seems to be at the root of most of the fears I see on/.
My understanding of "technical disclosures" is that they place information in the public domain - does that make a difference in this case?
If it does, then we do have a weapon of sorts. It's probably much cheaper to publish than pay patent filing fees, let alone lawyer fees.
So if I invent something, can't afford to pay thousands of dollars for a patent, and someone else finds out and patents it, I'm screwed? I can either keep my invention a secret and hope no one thinks of it, or let it become known and wind up paying a lawyer for the right to use it?
The new procedure for challenging validity (whatever it might be, and if it addresses such cases at all) is likely to cost even more (requires a lawyer, where it's at least theoretically possible to get a patent without one for only a few thousand in filing fees.)
Unless they also get rid of software patents (ha!) this sounds from the article like it simply hands the whole system over to the large companies, rather than the current arrangement that forces them to at least pay some legal fees to rip inventors off.
One of MS's talking points was that there's nothing binary or proprietary: it's all plain text XML. That might be slightly easier to work with than binary flash files -- but it also makes work easier for visitors to "borrow." Decompiling even protected flash files isn't hard either, but it's enough to slow down casual moochers and stop most corporate ones.
Of course, it's kind of silly to brag about openness when the whole thing is based on a closed source plugin. My big problem with the whole thing is that I fully expect support for "unfavored" platforms and browsers to start slipping as soon as there's some market share. I don't want to become a MS henchman, and pay for the privilege too.
Maybe that's not what MS has in mind this time, but with their record the burden of proof is on them. Not to mention that it's common sense to tread carefully with first releases of any new technology, even from companies with a track record (unlike MS) of producing quality products.
I make a point of spending almost *NO* time doing this. There's a few things that have to be done (turn off Windows automatic update; copy.vimrc on Linux; install QuickSilver on OSX ) but they don't take long. I can learn to live with most things, given a day or so; over the course of the first few days I may tweak mouse tracking or the like. An awful lot of this (on OSX / Linux at least) can be handled automatically by copying my home directory.
I've always preferred Gnome for the same reason the gear heads don't: being less configurable means less time and trouble finding and fixing the few things that I can't get used to. The rest I don't care about and rarely change unless I need to. I just want things to work correctly with as little fuss as possible. I didn't enjoy breaking into systems in college, either.
I observe that the parent is tagged "troll, idiot, dvorak". I propose a new tag, "trolidvo", which both honors ol' John as the industry spokesperson he is, and saves us all some typing.
Seriously, this reads more like a satire than the usual Dvo-troll. I wonder if he's considered writing for The Onion.
My understanding is that the "standardization" rules explicitly allow MS to charge a ("reasonable") license fee for stuff like the CLR.
If they decided to ask one cent per application that uses the CLR on non-Windows platforms:
no one could call it unreasonable or exclusionary: many would say that MS deserves to be paid for their research
10 cents extra / copy of the average distro is too much for most makers to absorb, and would cost several dollars to collect in addition to the user outrage involved. Likewise, how about the people who write and distribute a couple of titles?
a lot of the programs involved are GPL, which would cause major legal problems in addition to the price
.NET may be the most effective weapon MS has ever had, with the least public relations liability. The last thing they want to do is cooperate with or even officially recognize a project like Mono: that's like making a pet of a cow you plan to eat in the fall.
Predictably, a goodly percentage of those replying think Chris Tolkien is a hack / grave robber / etc. At the end of the day, he could have taken his father's papers and:
done the best editing job he was able to and publish them (did he have help? Presumably he has an editor at the publisher's, and I wouldn't hand this project to just anyone.)
released them as-is (where they would have been ignored as unreadable or mined for ever worse atrocities by ever worse (screen)writers. Better a RingWraith than to be fouled the way the later Dune books did Frank Herbert.)
shredded them all to avoid being accused of daring to profit from his father's estate (if JRRT had left cash, that would be fine, but writings...!)
Which would you prefer? I'm very happy to get the chance to see Tolkien's remaining writings, but I don't have the time to study the originals and no reason to think anyone else would be better at editing it for me.
I ran Ubuntu 5 for a while and liked it. That version couldn't be upgraded without a lot of trouble, so I dumped it and planned to clean install v6 (originally 6.06, IIRC.)
I'm still trying! I've downloaded ISO's twice and the MD5's match; trying to burn one reliably crashes XP's builtin software and several others I've tried on XP and Vista. I was able to make a disc on OSX, and I eventually found a Windows freeware that worked a few times. The CD's (I've made several) self-check OK, but invariably lock up every computer I've tried them on: I've never gotten more than one step past partitioning, or gotten anything else to run for more than a few minutes when booted from the CD. I've tried on several different computers / burners / drives, I've used the GParted live CD to avoid partitioning from the Ubuntu installer, and I've haunted the forums. I just bought a ready made CD and had the exact same result.
I'd love to try Edgy, but I've given up for now (I don't even have a working Linux box these days.) I guess I'll give Feisty a go when it comes out, but this experience has made me leery of the whole thing: I don't know if I can trust something whose install is that touchy.
I had an almost year old Wall Street Powerbook that went bad. It went back at least 6 times before it finally kept working long enough to go out of warranty.
Good: Apple paid for overnight shipping every time and replaced just about every part in it. They were never other than pleasant over the phone (this was long before there were Apple stores.)
Bad: I finally paid $350 to a 3rd party repair shop, who discovered that the first part Apple replaced was an unrepairably failed refurb part. Apple probably spent more on shipping than the used (but tested) part that finally fixed the problem.
So... Apple has problems like everyone else. They may not have know what they were doing, but they tried to do something.
A few years ago the NY Times got a new crossword editor, who started making puzzles with new words in them. At the time the clues and words had become so standardized that anyone with a few years experience could mechanically fill in most of the Sunday puzzle. The in-crowd wrote furiously to the paper and complained about being forced to think (like noobs!)
An editor with a big publisher, speaking at Purgatory (writers' workshop), made the comment that the best way to get published was to pick a book that was selling well -- and write something exactly like it.
Come on guys... the Romans at least had those virile barbarians to fight. We're sinking into the muck voluntarily, and as fast as we can.
I remember SUSTAR (SUper STAR trek) with a lot of fondness. It was especially cool when I got to college and could play without going through a 300 baud modem and printing the entire board on a teletype for each turn. A friend got an $5 / hour (under the table), 84 hour / week summer job at a local greasy spoon to buy a 48K TRS-80 with a cassette drive: porting Super Star Trek from DEC to Radio Shack BASIC was my first "group" project, and still the most fun.
I confess I didn't bother reading the entire article. There are just too many fatal problems:
Point-and-click programming has failed catastrophically every time it's been tried. My experience (e.g., iShell) taught me that it's too slow and cumbersome for programmers, and still useless for the non-programmers (defined as that vast majority who can't design program logic by any means, graphical or otherwise.) Inform 7 is a recent attempt by IF authors to help others, NOT by making programming unnecessary, but by attempting to impedance match the Author mindset. (I wish them luck, but recognize that most prospective users (the existing IF community) are competent programmers.)
Storytron's basic idea ("thinking, feeling virtual characters" ) is very similar to a system that died several years ago. I can't remember the name, but IIRC it used facial expressions on talking heads as a clue to computed emotional state; "embarrassingly bad" was one of the kinder descriptions I saw. I suspect it hasn't been forgotten by the rest of the IF community either.
Admittedly a lot of IF, with its "guess the magic phrase" problem, can be very frustrating. Their new interaction language, though, has "excruciating user experience" written all over it. If interaction actually requires reverse-diagramming a sentence by point and click (as the article seemed to suggest), I for one wouldn't bother to even try it.
The required commitment (both of time and intellect) for IF works for the small, passionate and closeknit IF community where no money changes hands; since the initial novelty wore off with Infocom, it hasn't ever worked as a commercial product. This is a fundamental disconnect that no imaginable development system or user UI even starts to address. I see zero hope for their business model.
I have a 20" monitor at work and it does make me a lot more productive than the 17" at home - but that's mostly a testimonial to Visual Studio's lousy UI design: 1280x1024 just isn't enough space. Sticking other stuff on a second screen wouldn't make VS itself any more useful.
I generally prefer 2 monitors to one large one. I don't think I'd go larger than 20", especially if I was paying for it; I use a 15" and a 17" at home. Of course, 2 monitors isn't ideal with all programs: Visual Studio can't be spread out as easily as CodeWarrior, for example.
Gartner has zero credibility (or less: they're a Microsoft publicity outlet with a sideline in clueless pronouncements that get taken seriously by clueless "analysts" and executives.)
Building applications from components sounded good back when VB was getting started, before even management figured out that VB crapware built from 3rd party controls was big, slow, buggy/unfixable, and couldn't (by definition) do anything that hadn't been done before. With one exception (a function required by marketing but used by no one, so it didn't have to work perfectly) our every attempt to "leverage" 3rd party ActiveX controls has been a disaster that wound up consuming more time and effort than doing it right to begin with. The one bright side to my management's current infatuation with the preposterous U3 platform is that I've been able to formally ban ActiveX and most other code requiring registration.
Don't get me wrong, I love the idea of reusable components: the state of the art just doesn't make them practical for most uses beyond standard library level. Now that Gartner has spoken, I'm just waiting for some overpaid pinhead to substitute his judgement for mine again and create another fiasco I'll have to fix to save his bonus. I'm still recovering from the latest attempt at outsourcing, where we wound up discarding and rewriting most of an application in much less time than our contractors needed to hose it in the first place.
Apple's quality control started slipping long before Intel, reportedly because Apple discovered they could fix 5% cheaper than testing 100%. It was no joke: out of the box problems were rare before the first iMac's (I never heard of one from 1986 to 1994; I have 15 year old machines that still run w/ no issues.) Starting with the first iMac, Apple quality deteriorated drastically. I have a (dead) Powerbook that went back 6 times before managing to expire its warranty; I sent it to a 3rd party shop that replaced a defective used part dating to the first Apple repair.
At their worst, though, Mac's are better across the board than any Wintel PC's I've ever had: I never buy preassembled Wintel's any more, while I'd probably not bother building a Mac even if it were possible.
That said, I'm in no hurry to buy a MacTel. They obviously need a little more shaking out to be practical. Once they do, though, I can't wait to have Linux, XP, and OSX all available without a reboot.
The closest thing to an "elite" game I know of is interactive fiction (start with http://en.wikipedia.org/wiki/Interactive_fiction ). It has a small but passionate community that produces most of its own games, because, as you may have guessed, they don't sell well enough to be commercially viable, and community standards are too high for hack jobs.
Most IF'ers disdain any graphics, sound, etc: the sentiment has a religious as well as aesthetic side. This keeps production costs low but requires vastly greater quality (in terms of story, program stability) than most mass market games achieve (more than they could achieve, probably, by simply hiring programmers.) Passion is the basic qualification, before technical ability even matters.
Well, let me be the first to enter the, "Guess I'm not using Objective-C" camp
I'd love to find a way to make a living with Objective-C, if I can turn GC totally off (I assume that's possible.)
I've tried C# and managed C++ and hated both: one reason was the atrocious coding style they encourage. Objective-C is a particularly elegant and powerful C, which makes the idea even uglier, like mud wrestling in a tuxedo.
Kludging a GC library into any version of C is like pulling a U-Haul with a race car, not because it slows things down (it doesn't, in most cases) but because it doesn't fit. C++ is not about being sloppy because a runtime lets you avoid learning how to use the language. Leave GC where it's needed and serves a purpose: functional languages, scripting languages, lisp. Bolting it on to C does nothing but encourage incompetents to remain incompetent instead of moving on where they belong, be it a higher level language or McDonald's. Whoever said C++ (or any other language) has to be accessible to everybody? They tried that with VB and Lingo, and in both cases we got a few more bad programmers, the same crowd who still didn't get it, and a crappy new language to sell books.
Yell troll if you will, but given enough experience to be paid for coding (less than it takes to play most musical instruments or learn a natural language well), C++ memory management just isn't that hard, and not getting it right or refusing to try suggests a the kind of attitude I wouldn't hire. Qualified coders are going to be rarer every day the way things are going.
There was once a game called Iron Helix (Macromedia Director; marketed for Mac only IIRC.) It involved scurrying through an abandoned spaceship hunting DNA samples and trying to evade a robot that instantly killed your (robot) character on sight: no combat of any sort, only flight.) It was slow and had crude graphics (stop motion except for segues and a few special effects), but I still remember it fondly for the level of tension and suspense it created, sort of like the first Alien movie. Modern gamers would laugh, but Iron Helix was a whole lot more fun than anything I've tried since (including most recently Half Life 2 on AMD64) and the whole thing was pure adrenaline.
Anyway, I got a job this summer programming in a C++.NET and ObjectARX [autodesk.com] environment using Visual Studio,
Long story short, it took me about a week to figure it all out. It was cake.
Next time, try learning Haskell. Shouldn't take you more than day or two, and I'll spot you Arrows. I'd have said Common Lisp if you hadn't already studied Scheme.
If you really think you "figured out" C++ well enough to be paid for it in a week (based on classes in C, Java, and SmallTalk), heaven help your users and maintainers.
For a C++ person, learning C# is about a day and a half job: mostly memorizing a few new constructs and forcing yourself to be sloppy. Of course, doing anything interesting in C#, like Java, requires learning its framework, and I doubt you can even read that in a day and a half.
Slashdot Mirror
If you really need GC or boost features, most functional and/or scripting languages do it far better and simpler than any bolted on kludge to C++ ever will, and are easier to learn and more fun to use. Don't try to make C++ into something it isn't, just because more and more people don't need the kind of power it offers any more, and don't want to take the time to learn it properly. Proficiency does take time, but once you know the syntax the most important skill required to use modern C++ effectively could be summed up as:
We call it "Aim low and SHIP!" And did I mention, I use MACROS?
When I first read about Scratch, I thought of Prograph.... Looking at the video, I thought of a fancied up version of iShell or one of the presentation builders.
Problem I've seen w/ all of the above is that they make programming excruciatingly slow and clumsy if you know what you're doing, and don't help if you don't: there is no evidence that programming can be taught in the absence of a (probably genetic) talent for it. iShell tried hard to empower content producers to script their work, and failed like all the others: the target users were still helpless, and the programmers they hired usually preferred the much more complicated, non-graphical Lingo because they got their work done faster.
For small children starting from ground zero, it might make sense. Users old enough to grasp Python (for example) will quickly either get frustrated and move on to a "real" productivity-oriented programming environment, or find out they can't understand what's going on no matter how many tutorials they do, and give up. Those who do have the gift, though, may as well skip the first step and the rest will find out where they stand quicker if they try a "hard" language to start with. Beginning programming classes are (or should be) about separating the sheep from the goats as efficiently as possible.
I've followed Bussard's work in this area, and we're darn lucky to have him. Working for the Navy makes me nervous, though.
It's always been rumored that the Farnsworth fusor was buried (and it was, big time and deliberately) because it looked like it might work. While that device would probably never have become economically feasible as a power generator, there's not much likelihood the current Tokamak-based designs will either, and they're getting billions for research worldwide. One theory is that Farnsworth's method (a direct ancestor of Bussard's) was too easy to downscale to town or neighborhood size, where a working Tokamak would require an enormous plant that only government or big industry can build (and control.)
If the bad guys want to do the same thing again, it would be awfully easy to just classify Bussard's work (which is not yet practical for anything, and may never be), say it failed, and let it be forgotten. Or maybe just hide it until we're up against the wall (fossil fuel and uranium getting too expensive, breeders still won't work, other fusion research still going nowhere) when it would be the last hope of staving off the apocalypse. Maybe the governments will be sufficiently in Control by then that they'd risk releasing such disruptive technology.
I bought one for my girlfriend for Christmas. They were long since out of stock in the catalogs by then, but I found one for sale at a little gas station. The owner told me he had ordered one for himself, using his business address, and had promptly gotten a letter: "We have no dealers in your area, would you be interested...." It was the last one he had, he ordered 3 at a time and kept them under his counter.
I set the thing up (it was the week before Christmas, and I wrap fast) on my parents' kitchen counter using a 9 inch B&W TV they had. I made a display of 3 linked rotating rings (they each had a gap that made the "rotation" visible) that completely blew my father's mind -- the idea that something like that was possible. Compared to the punched cards and half day delays I was used to at college, it was a darned nice system.
BSD: Anyone can use your code, any way they want, as long as they mention your name.
GPL: Anyone can use your code, as long as they mention your name and allow others to use the code as you have let them use it.
LGPL: Same as GPL, but with some exceptions allowing others not to be bound to your licence if they don't actually touch your code in their project, but just use it.
Additional thoughts (assuming that you are interested in profit from your work):
Unless forced (i.e., you used someone else's GPL code), free software licenses are only useful if you're prepared to release your work publicly. If you own 100% of the code and it contains no algorithms or data that you can't / don't want to reveal, there are good reasons to use a free license even if you don't buy into the Free software philosophy (and the definition of that is debatable, controversial, and not intuitive.)
GPL If you want to profit from your work, use the GPL. This gains you a user base as fast as possible, respect from the FOSS community, and forces anyone who uses your code to also release their own, or pay you for a different license. GPL provides fair protection from piracy (everyone knows what it is, and there is substantial recourse against violators) but allows you "sell" your work on terms acceptable to commercial developers. You may even be able to "reverse license" GPL'ed improvements from customers, allowing you distribute and sell improved versions you didn't have to write in exchange for a reduced license fee.
A prime example is http://trolltech.com/. Their Qt product contains little of a "trade secret" nature, but is far too big and complex to recreate for the price of a license. Trolltech has contributed greatly to Linux (in particular) by providing a high end UI system for free. This costs them nothing (no sales were lost), and has resulted in thousands of coders becoming both experienced with and enthusiastic about Qt, some of whom may need the commercial version some day.
LGPL forces any changes to your work to be open, but work that uses your work can still be closed, if the developer jumps through the right legal hoops to draw a line between your code and theirs. This tends to make commercial users nervous, while not encouraging them to buy a regular license (even though it's probably cheaper than their legal cost to feel safe.) Given that your source code is going to be released publicly in either case, LGPL has no real advantage over the GPL.
BSD is the "real" free license. It's more honest than the GPL in that you don't proclaim Freedom while forcing your license terms on everyone else, but it's pretty unlikely that you'll get paid other than from contract work for improvements.
How does this work when the standard is "first to file?" It sounds like prior art is irrelevant, which seems to be at the root of most of the fears I see on /.
My understanding of "technical disclosures" is that they place information in the public domain - does that make a difference in this case?
If it does, then we do have a weapon of sorts. It's probably much cheaper to publish than pay patent filing fees, let alone lawyer fees.
So if I invent something, can't afford to pay thousands of dollars for a patent, and someone else finds out and patents it, I'm screwed? I can either keep my invention a secret and hope no one thinks of it, or let it become known and wind up paying a lawyer for the right to use it?
The new procedure for challenging validity (whatever it might be, and if it addresses such cases at all) is likely to cost even more (requires a lawyer, where it's at least theoretically possible to get a patent without one for only a few thousand in filing fees.)
Unless they also get rid of software patents (ha!) this sounds from the article like it simply hands the whole system over to the large companies, rather than the current arrangement that forces them to at least pay some legal fees to rip inventors off.
One of MS's talking points was that there's nothing binary or proprietary: it's all plain text XML. That might be slightly easier to work with than binary flash files -- but it also makes work easier for visitors to "borrow." Decompiling even protected flash files isn't hard either, but it's enough to slow down casual moochers and stop most corporate ones.
Of course, it's kind of silly to brag about openness when the whole thing is based on a closed source plugin. My big problem with the whole thing is that I fully expect support for "unfavored" platforms and browsers to start slipping as soon as there's some market share. I don't want to become a MS henchman, and pay for the privilege too.
Maybe that's not what MS has in mind this time, but with their record the burden of proof is on them. Not to mention that it's common sense to tread carefully with first releases of any new technology, even from companies with a track record (unlike MS) of producing quality products.
I make a point of spending almost *NO* time doing this. There's a few things that have to be done (turn off Windows automatic update; copy .vimrc on Linux; install QuickSilver on OSX ) but they don't take long. I can learn to live with most things, given a day or so; over the course of the first few days I may tweak mouse tracking or the like. An awful lot of this (on OSX / Linux at least) can be handled automatically by copying my home directory.
I've always preferred Gnome for the same reason the gear heads don't: being less configurable means less time and trouble finding and fixing the few things that I can't get used to. The rest I don't care about and rarely change unless I need to. I just want things to work correctly with as little fuss as possible. I didn't enjoy breaking into systems in college, either.
I observe that the parent is tagged "troll, idiot, dvorak". I propose a new tag, "trolidvo", which both honors ol' John as the industry spokesperson he is, and saves us all some typing.
Seriously, this reads more like a satire than the usual Dvo-troll. I wonder if he's considered writing for The Onion.
If they decided to ask one cent per application that uses the CLR on non-Windows platforms:
- done the best editing job he was able to and publish them (did he have help? Presumably he has an editor at the publisher's, and I wouldn't hand this project to just anyone.)
- released them as-is (where they would have been ignored as unreadable or mined for ever worse atrocities by ever worse (screen)writers. Better a RingWraith than to be fouled the way the later Dune books did Frank Herbert.)
- shredded them all to avoid being accused of daring to profit from his father's estate (if JRRT had left cash, that would be fine, but writings...!)
Which would you prefer? I'm very happy to get the chance to see Tolkien's remaining writings, but I don't have the time to study the originals and no reason to think anyone else would be better at editing it for me.I ran Ubuntu 5 for a while and liked it. That version couldn't be upgraded without a lot of trouble, so I dumped it and planned to clean install v6 (originally 6.06, IIRC.)
I'm still trying! I've downloaded ISO's twice and the MD5's match; trying to burn one reliably crashes XP's builtin software and several others I've tried on XP and Vista. I was able to make a disc on OSX, and I eventually found a Windows freeware that worked a few times. The CD's (I've made several) self-check OK, but invariably lock up every computer I've tried them on: I've never gotten more than one step past partitioning, or gotten anything else to run for more than a few minutes when booted from the CD. I've tried on several different computers / burners / drives, I've used the GParted live CD to avoid partitioning from the Ubuntu installer, and I've haunted the forums. I just bought a ready made CD and had the exact same result.
I'd love to try Edgy, but I've given up for now (I don't even have a working Linux box these days.) I guess I'll give Feisty a go when it comes out, but this experience has made me leery of the whole thing: I don't know if I can trust something whose install is that touchy.
I had an almost year old Wall Street Powerbook that went bad. It went back at least 6 times before it finally kept working long enough to go out of warranty.
Good: Apple paid for overnight shipping every time and replaced just about every part in it. They were never other than pleasant over the phone (this was long before there were Apple stores.)
Bad: I finally paid $350 to a 3rd party repair shop, who discovered that the first part Apple replaced was an unrepairably failed refurb part. Apple probably spent more on shipping than the used (but tested) part that finally fixed the problem.
So... Apple has problems like everyone else. They may not have know what they were doing, but they tried to do something.
A few years ago the NY Times got a new crossword editor, who started making puzzles with new words in them. At the time the clues and words had become so standardized that anyone with a few years experience could mechanically fill in most of the Sunday puzzle. The in-crowd wrote furiously to the paper and complained about being forced to think (like noobs!)
An editor with a big publisher, speaking at Purgatory (writers' workshop), made the comment that the best way to get published was to pick a book that was selling well -- and write something exactly like it.
Come on guys... the Romans at least had those virile barbarians to fight. We're sinking into the muck voluntarily, and as fast as we can.
I remember SUSTAR (SUper STAR trek) with a lot of fondness. It was especially cool when I got to college and could play without going through a 300 baud modem and printing the entire board on a teletype for each turn. A friend got an $5 / hour (under the table), 84 hour / week summer job at a local greasy spoon to buy a 48K TRS-80 with a cassette drive: porting Super Star Trek from DEC to Radio Shack BASIC was my first "group" project, and still the most fun.
I confess I didn't bother reading the entire article. There are just too many fatal problems:
Point-and-click programming has failed catastrophically every time it's been tried. My experience (e.g., iShell) taught me that it's too slow and cumbersome for programmers, and still useless for the non-programmers (defined as that vast majority who can't design program logic by any means, graphical or otherwise.) Inform 7 is a recent attempt by IF authors to help others, NOT by making programming unnecessary, but by attempting to impedance match the Author mindset. (I wish them luck, but recognize that most prospective users (the existing IF community) are competent programmers.)
Storytron's basic idea ("thinking, feeling virtual characters" ) is very similar to a system that died several years ago. I can't remember the name, but IIRC it used facial expressions on talking heads as a clue to computed emotional state; "embarrassingly bad" was one of the kinder descriptions I saw. I suspect it hasn't been forgotten by the rest of the IF community either.
Admittedly a lot of IF, with its "guess the magic phrase" problem, can be very frustrating. Their new interaction language, though, has "excruciating user experience" written all over it. If interaction actually requires reverse-diagramming a sentence by point and click (as the article seemed to suggest), I for one wouldn't bother to even try it.
The required commitment (both of time and intellect) for IF works for the small, passionate and closeknit IF community where no money changes hands; since the initial novelty wore off with Infocom, it hasn't ever worked as a commercial product. This is a fundamental disconnect that no imaginable development system or user UI even starts to address. I see zero hope for their business model.
I have a 20" monitor at work and it does make me a lot more productive than the 17" at home - but that's mostly a testimonial to Visual Studio's lousy UI design: 1280x1024 just isn't enough space. Sticking other stuff on a second screen wouldn't make VS itself any more useful.
I generally prefer 2 monitors to one large one. I don't think I'd go larger than 20", especially if I was paying for it; I use a 15" and a 17" at home. Of course, 2 monitors isn't ideal with all programs: Visual Studio can't be spread out as easily as CodeWarrior, for example.
Gartner has zero credibility (or less: they're a Microsoft publicity outlet with a sideline in clueless pronouncements that get taken seriously by clueless "analysts" and executives.)
Building applications from components sounded good back when VB was getting started, before even management figured out that VB crapware built from 3rd party controls was big, slow, buggy/unfixable, and couldn't (by definition) do anything that hadn't been done before. With one exception (a function required by marketing but used by no one, so it didn't have to work perfectly) our every attempt to "leverage" 3rd party ActiveX controls has been a disaster that wound up consuming more time and effort than doing it right to begin with. The one bright side to my management's current infatuation with the preposterous U3 platform is that I've been able to formally ban ActiveX and most other code requiring registration.
Don't get me wrong, I love the idea of reusable components: the state of the art just doesn't make them practical for most uses beyond standard library level. Now that Gartner has spoken, I'm just waiting for some overpaid pinhead to substitute his judgement for mine again and create another fiasco I'll have to fix to save his bonus. I'm still recovering from the latest attempt at outsourcing, where we wound up discarding and rewriting most of an application in much less time than our contractors needed to hose it in the first place.
Apple's quality control started slipping long before Intel, reportedly because Apple discovered they could fix 5% cheaper than testing 100%. It was no joke: out of the box problems were rare before the first iMac's (I never heard of one from 1986 to 1994; I have 15 year old machines that still run w/ no issues.) Starting with the first iMac, Apple quality deteriorated drastically. I have a (dead) Powerbook that went back 6 times before managing to expire its warranty; I sent it to a 3rd party shop that replaced a defective used part dating to the first Apple repair.
At their worst, though, Mac's are better across the board than any Wintel PC's I've ever had: I never buy preassembled Wintel's any more, while I'd probably not bother building a Mac even if it were possible.
That said, I'm in no hurry to buy a MacTel. They obviously need a little more shaking out to be practical. Once they do, though, I can't wait to have Linux, XP, and OSX all available without a reboot.
Most IF'ers disdain any graphics, sound, etc: the sentiment has a religious as well as aesthetic side. This keeps production costs low but requires vastly greater quality (in terms of story, program stability) than most mass market games achieve (more than they could achieve, probably, by simply hiring programmers.) Passion is the basic qualification, before technical ability even matters.
I'd love to find a way to make a living with Objective-C, if I can turn GC totally off (I assume that's possible.)
I've tried C# and managed C++ and hated both: one reason was the atrocious coding style they encourage. Objective-C is a particularly elegant and powerful C, which makes the idea even uglier, like mud wrestling in a tuxedo.
Kludging a GC library into any version of C is like pulling a U-Haul with a race car, not because it slows things down (it doesn't, in most cases) but because it doesn't fit. C++ is not about being sloppy because a runtime lets you avoid learning how to use the language. Leave GC where it's needed and serves a purpose: functional languages, scripting languages, lisp. Bolting it on to C does nothing but encourage incompetents to remain incompetent instead of moving on where they belong, be it a higher level language or McDonald's. Whoever said C++ (or any other language) has to be accessible to everybody? They tried that with VB and Lingo, and in both cases we got a few more bad programmers, the same crowd who still didn't get it, and a crappy new language to sell books.
Yell troll if you will, but given enough experience to be paid for coding (less than it takes to play most musical instruments or learn a natural language well), C++ memory management just isn't that hard, and not getting it right or refusing to try suggests a the kind of attitude I wouldn't hire. Qualified coders are going to be rarer every day the way things are going.
There was once a game called Iron Helix (Macromedia Director; marketed for Mac only IIRC.) It involved scurrying through an abandoned spaceship hunting DNA samples and trying to evade a robot that instantly killed your (robot) character on sight: no combat of any sort, only flight.) It was slow and had crude graphics (stop motion except for segues and a few special effects), but I still remember it fondly for the level of tension and suspense it created, sort of like the first Alien movie. Modern gamers would laugh, but Iron Helix was a whole lot more fun than anything I've tried since (including most recently Half Life 2 on AMD64) and the whole thing was pure adrenaline.
Long story short, it took me about a week to figure it all out. It was cake.
Next time, try learning Haskell. Shouldn't take you more than day or two, and I'll spot you Arrows. I'd have said Common Lisp if you hadn't already studied Scheme.
If you really think you "figured out" C++ well enough to be paid for it in a week (based on classes in C, Java, and SmallTalk), heaven help your users and maintainers.
For a C++ person, learning C# is about a day and a half job: mostly memorizing a few new constructs and forcing yourself to be sloppy. Of course, doing anything interesting in C#, like Java, requires learning its framework, and I doubt you can even read that in a day and a half.
... is a respectable Linux community "front man" to negotiate with. Someone like Neville Chamberlain.