> I generally save as Doc files for them and > try to verify on another person's machine > that it will open correctly.
I generally send a.zip containing three formats: sxw, pdf, plain ASCII txt. (I will eventually replace sxw with odt, but 2.0 was only *just* released.) If they just want to print the thing, they've got the PDF, so that works. *EVERYONE* can open the plain text version, no sweat, and it leaves them the choice of installing the OpenOffice.org software if they want to try for the best possible interoperability and edit the document while preserving the formatting.
Fundamentally there's no such thing as *full* interoperability from one computer to another with word processing formats, anyway. I've found that when going between my main workstation (which runs Gnome on Linux), my Windows Me system, and my family's Windows 98 system, even after ensuring that all of them have all the fonts the document uses (e.g., by using only the corefonts or only the Bitstream Vera fonts, and installing them on all the systems) and the same version of the OpenOffice.org software, the same document will nevertheless consume different numbers of pages on the different systems, presumably because of minor differences in the font rendering or somesuch. The situation with Microsoft software is not significantly better, especially taking into account that people have different versions of it. Besides that, if you're assuming everyone you might exchange documents with has all the same fonts you do, you're living in serious delusion.
Fundamentally, word processing formats were not (any of them, AFAIK) designed for online distribution; they were designed for creating something that's destined to be printed on paper. If you want online distribution, well, that's what we've got the web for, isn't it?
It's about customizeability. I know, most people don't care; they just leave all the settings they way they come from the factory, never change so much as the wallpaper, much less any actual functionality. Hey, if that makes you happy, whatever floats your boat. Me, I prefer to have the system set up to behave the way I want it to behave, manages windows the way I want them managed, and so on. When I try to use somebody else's system, that isn't set up my way, it's a real pain, because nothing works the way *I* want it to work, and the default behaviors are highly suboptimal for the way I work.
For me, it's all about that, about the ability to teach the computer to work with me, rather than the other way around. After all, who owns whom? The computer belongs to *me*, it should work *my* way; why should I train myself to do things its way? If you were an employer, would you train yourself to do things your employees' way?
> Could you please explain what's "Ma Bell" for us foreigners?
Ma Bell was what people in the US called the phone company, when it was a total monopoly. Then in the twentieth century the US government split up the phone company into a lot of smaller local phone companies, which we now call Baby Bells -- but they're not really babies at all; a better term would perhaps be Bell Brothers or simply The Phone Cabal.
Because of the way the split-up worked, everyone now gets to choose which Baby Bell to buy long distance service from, so they compete on that, but for local phone service you either buy from your local baby bell or else you do without phone service. So where before Ma Bell grossly overcharged for long distance service, now instead we pay ridiculous and exorbitant rates just to *have* a phone line, with all kinds of additional surcharges and fees (which vary unpredictably from month to month), and every little extra feature we might want is extra. Long distance calls, however, cost only a little. It costs more to have call waiting for a month than to talk to someone 500 miles away for three hours. You want a touch-tone line, instead of pulse? It's extra. Caller ID? Extra. Anonymous Call Block? Extra. You want DLS? Sorry, your line isn't approved for that. You want to call France? No problem, same low rate as France.
> The new OpenDocument Fellowship is working with a petition > to get Microsoft to implement the format.
Petitions are meaningless. Nobody pays them any heed. You really want Microsoft (or anyone else) to do something? Write them. Personally. Preferably on business letterhead, in Microsoft's case. That's *MUCH* more likely to have an impact than signing any petition. In all history I am not aware of a single documented instance of anyone being presented with a petition, seeing how many people signed it, and, as a result, changing their mind and doing as it suggests where they previously were planning otherwise. The primary effect of a petition is to make the people who are collecting the signatures feel placated, feel as if they are "doing" something. A secondary effect is to waste the time of the people doing the signing.
> I don't want to sound too much like flamebait but how on earth is this a Laptop?
Bear in mind, screen sizes are measured on the diagonal. One could easily fit a 20"-diagonal laptop on a 17"-across lap, which is not an unusually large lap, at least around here.
It would certainly still be more readily portable than a typical desktop, and yet (especially if the keyboard takes advantage of that extra space) rather more usable, in day-to-day desktop-replacement terms, than a typical small notebook system. If you have a desktop system for regular use and want something small to carry to every meeting and class to take notes, this is probably not what you're looking for. Some people, however, do not want to deal with having two computers to maintain, and they'd like to be able to take the computer with them easily when they go someplace for the weekend or whatever; that is the niche these large laptops are designed to fill. Yes, "table-top" might be a more accurate term taken literally, but "laptop" is the industry standard term for folding all-in-one portable computers.
> The only thing the CTP games had in common with the real Civ > franchise was this word: "Civilization".
Oh. This was not made clear to me. I just picked up the CTP demo because it was the only demo of a commerical Civilization game that I could find available for my operating system. I have too much experience with software in general and games in particular to shell out money for a game without seeing a demo first; the only commercial games I've *ever* purchased without playing a demo first are the Lost Treasures of Infocom, and that's a little bit different.
> Either your command is wrong (it should do that formatting as its > output) or you should have used [g]awk instead of cut.
If you have to go to that much trouble, just open the command's output as a pipe in a Perl script: open FILEHANDLE, "command |"; then just read from the filehandle and process the data in whatever way you like.
> What makes MSH rock is that it's a python-like programming languaje > PLUS a user-oriented (user=administrator) shell like bash.
Oh. And here all along I'd been lead to believe it was a Haskell-like programming language with a shell interface.
But OOP-versus-PFP paradigm issues aside, your basic point, that the interesting thing about Monad is the combination of the scripting language with the shell interface, is essentially right on the money. As a programming language, it's not that innovative -- we've already got great programming languages, including great PFP languages like Haskell, and in terms of multiparadigmatic languages, Perl6 may very likely beat Monad to release at this point (though it's going to be close), and *as a programming language* would blow its doors off, in terms of practical and useful innovation. And for people who prefer to stick with an OOP approach, Python and others have considerable mindshare already, so it would be an uphill battle *if* Monad can even compete there, which from what I know of it it may not, as it leans, from what I am given to understand, more toward the functional side of things, paradigm-wise. But Monad does two things these others do not: first, it's built into a command-line interface, and second, it's *eventually* supposed to get bundled into the OS so that you can just count on its being there on every (Windows) system, which will be real nice. I think it's a significant shame that they scrapped the plans to include it in the Longhorn desktop (now called Vista), and I hope it does make it into the Longhorn server offering (whatever the marketing department ends up calling that).
The only thing cooler than a great language built into a great command-line interface is a great command-line interface and a great language built together into a great editor, a la emacs/elisp/eshell, which absolutely rocks. With Monad replacing cmd, the text editor is IMO the next place Microsoft needs to look, in terms of core improvements to their basic toolset. Notepad is *LONG* overdue to be replaced with something rather a lot more powerful. Wouldn't it be great if the editor they replaced it with was fully scriptable with Monad and had a Monad shell built right into it? Talk about getting developers on board! This would be worth a hundred thousand conferences and symposiums with pep talks and CEOs chanting "Developers, Developers, Developers!"
Only thing is, in some cases the clones are *better*. I tried to like the Civ CTP demo enough to buy the game, really I did. If it had been even *close* to as much fun to play as Freeciv, I would have bought it just for the variety, but I found myself hating it, because the stupid UI kept getting in the way of the game. After that I never looked at a demo for another commercial civ game. (I did buy Railroad Tycoon II, though. That demo was actually fun, so I bought the game.)
And yeah, the message he's sending to the Freeciv developers (notice especially the preface he slaps on it; he's implying that how he views any given "clone" might be contingent on something related to what follows) appears to be basically "Don't let me catch you doing anything questionable in its legality, because we will defend what we legally can." Okay, that's fair.
The beer and hot dogs and bacon are minor details. When it says "most distinguishable traits", it's probably talking about things like Bart's fabled disrespect for authority, which for cultural reasons Badr wouldn't be able to have. Also bear in mind that Omar will necessarily be the undisputed ruler of his household.
In other words, they're basically adapting the drawing style of The Simpsons to a completely different cast of characters in a show with a completely different focus and appeal.
> Microsoft has claimed the cost of software is not an important issue > in the developing world.
This is true, but Microsoft's reason ("they won't have the expertise") is irrelevant. If they have computers, they'll develop the ability to use them (for the few hours a day when they have power). The real reason software licensing costs are a non-issue in the developing world is because copyright law is (viewed as) a non-issue there. They think nothing of installing one OEM copy of Windows 98 on every computer they can get their hands on. Any phone-home stuff Microsoft might build in doesn't make any difference either, in a world where getting to a phone line means an all-day journey to the city.
> that will surely be a victory for Oracle, who will secure a strong lead
Umm, Oracle *has* a strong lead. They're the big boy in this market. What they'd allegedly be doing (*if* they do what you suggest, which at this point is speculation, however likely it seems) is [attempting to] squash a much smaller competitor. MySQL AB is certainly not the competitor they'd most like to squash, nor as far as that goes the second, but those companies have strong businesses in other markets (markets where Oracle has no product) and are overall larger companies than Oracle, so there really isn't a whole lot that Oracle can do there. Whether they would consider it worth the potential negative PR (and other costs, but that's probably the largest cost and certainly the most difficult to quantify) to squash this much smaller player is an open question. Personally I think it would be unwise, but I don't make decisions for Oracle.
> Cause more OSS FUD: "What will happen to your open source vendor? It > could evaporate tomorrow. Stick with Oracle, who will be there for you."
This one could backfire. What can happen to InnoDB or MySQL can happen also to Oracle. There's always a bigger fish. There are, after all, larger companies with their hands in the database market (two of them, the way I count). Sure, Oracle's offering is in every way superior to a certain larger company's database (probably to that of both larger companies, but especially to the one that's a larger concern, which I am going out of my way to avoid naming), but when has that ever stopped a business from being shut down or otherwise incapacitated by a larger competitor using devious means?
> Many companies were just waiting for the 5.0 release to try it out > I'm sure
I know of a company that officially cites stored procedures as a critical reason why they can't make their product work with MySQL. This is presumably just talk, as their product doesn't work with Oracle either, or anything else that doesn't come out of Redmond (well, okay, the web service does work with other browsers, at least), but it is nevertheless the excuse they currently are using. One of their engineers actually asked me, when he found out I'd used MySQL a little, if I knew how soon their new release was coming out, since he'd heard they were implementing stored procedures. (I guess management was feeding the engineers the same line they were feeding customers.)
This is neither here nor there for the company in question, partly because they're a Microsoft shop anyway and so I can't imagine they'd actually support MySQL, and partly for another reason I won't state here. Nevertheless, it seems to me that your observation above may have some merit; surely this ISV I'm talking about cannot be the only one that considers stored procedures an important feature.
> The mysql vs. postgres thing gets so out of hand.
You want to see out of hand? Start a discussion functional programming versus object-oriented programming with a Python programmer and a Scheme programmer.
The significance of this statement cannot be overstated. The CPAN with its wealth of modules is such an enormous benfit, this alone would make Perl five times better than any of the languages the other poster listed, even if it were otherwise nothing special (which it is).
> Perl gives one the ability to write very very obsfucated code.
Most of it only looks obfuscated to people who don't speak Perl. Once you become fluent, you read code that others would consider obfuscated and ask, "What's the big deal?"
(There are exceptions, of course; code that's used ACME::Eyedrops and whatnot really is too obfuscated to be clearly legible, and then there's Matt's Script Archive. But nobody plays those games in production code.)
Re:Might have taken a while....
on
Vim 6.4 Released
·
· Score: 1
> The CVS version is very usable.
Yeah, Emacs 21.0 was fairly usable too, but like I said, the slow release cycle isn't, IMO, a big deal with text editors. The major ones are a much more mature product than most other kinds of software these days, so we're not sitting here going, "I just can't go another *day* without having [some feature it doesn't have yet]", like we still are with e.g. OpenOffice (which *still* doesn't have parenthesis matching, for instance, a sore omission).
> You may argue that nobody willingly uses variable names like > ll1O0, l1100 and lllOO
I would argue that no programmer worth his salt willingly uses fonts that don't make those characters easy to distinguish. I certainly don't. However, unicode characters are another matter; I only care about how my fonts display safe and sane characters (i.e., ASCII characters with decimal values from 32 through 126). Unicode characters can show up as a character-sized box as far as I'm concerned, because they don't belong in my source code, and the editor shows control characters as \nnn escapes anyhow (except for line endings (which it displays as newlines) and horizontal tabs (which it replaces with spaces in most modes, because that's how I have it configured)).
The idea that the Dvorak layout is "better" comes from the assertion (which has never been conclusively established, as far as I am aware) that it allowed for a faster typing rate, i.e., a larger words-per-minute number. That *may* have been relevant in the 1950s, when a large percentage of typing consisted of retyping existing material, because typewriters didn't have save and restore, copy and paste, or print merge capability (among other things). This is no longer anything like relevant, because these days almost all typing is on-the-fly, coming right out of the brain, which only goes so fast. I used to be able to type 50wpm or so, back when I used to type in things that I'd first written out on paper (e.g., compositions that I'd worked on at the library or in study hall). It's been so long since I've done that, I don't think I can type faster than about 30wpm now, yet I find myself pausing here and there, because I'm still typing faster than I can compose.
Now, there *are* problems with the QWERTY layout, problems having to do with certain keys (notably shift) being way too far out of the way, other keys (notably CAPSLOCK and the left window key) too easy to hit by mistake, and with one entire digit going unused, because there's one big fat spacebar where both thumbs go; however, Dvorak makes these same mistakes in pretty much exactly the same way.
What's that got to do with anything? Fermat expressly stated what sets the numbers in question had to belong to. The slashdot poster just used variables from the letter ranges normally associated with those sets (e.g., n for a natural number) and assumed that anyone who's had any math would understand. Either way, if you remove (or ignore) the constraints on the numbers, it's trivial to define new groups of numbers that "solve" the thing, but that was not the intention.
> I think typing "I, for one" is redundant and annoying. Stop it.
I, for one, welcome our new redundantly-welcomed "I, for one"-typing redundant overlords.
Re:Might have taken a while....
on
Vim 6.4 Released
·
· Score: 1
The major text editors all use a slower release cycle. Do you remember how long we waited for Emacs 21, with its wealth of promised new features? That seems like a long time now, and yet Emacs 22 is not on anyone's radar yet, as far as I am aware. I'm not a vim fan, but the slow release cycle is not one of the reasons.
You know, in all the *many* discussions I've read about vim over the years, one thing I've never heard is "Well, it has lots of features, but the problem is, it's so bloated and buggy!" Have you ever heard that criticism leveled at vim? Does it have a problem in that regard? If so, this is news to me.
I have, numerous times, asked vim afficianados, "Does it support foo, bar, and baz?" (where foo, bar, and baz are any three arbitrarily chosen advanced features I use on a regular basis in Emacs and would not want to live without), and the answer has NEVER been "Oh, yes, it does all three of those things." It's been a couple of years since the last time, so we can try this again: does it support automatic mode-dependent (i.e., sensitive to the type of file you're editing) grouping-symbol matching (wherein when you type a closing grouping symbol it highlights the opening one)? Does it support correct automatic rewrapping of nested quotations in email and usenet messages? Can it be customized so that when editing a certain type of file, inserting certain characters has additional effects?
It seems to me that a critical lack of important features is vim's one major shortcoming, and that's what development really ought to focus on. Bugs in vim? I've never heard of that being a serious problem.
Slashdot Mirror
> I generally save as Doc files for them and
.zip containing three formats: sxw, pdf, plain ASCII txt. (I will eventually replace sxw with odt, but 2.0 was only *just* released.) If they just want to print the thing, they've got the PDF, so that works. *EVERYONE* can open the plain text version, no sweat, and it leaves them the choice of installing the OpenOffice.org software if they want to try for the best possible interoperability and edit the document while preserving the formatting.
> try to verify on another person's machine
> that it will open correctly.
I generally send a
Fundamentally there's no such thing as *full* interoperability from one computer to another with word processing formats, anyway. I've found that when going between my main workstation (which runs Gnome on Linux), my Windows Me system, and my family's Windows 98 system, even after ensuring that all of them have all the fonts the document uses (e.g., by using only the corefonts or only the Bitstream Vera fonts, and installing them on all the systems) and the same version of the OpenOffice.org software, the same document will nevertheless consume different numbers of pages on the different systems, presumably because of minor differences in the font rendering or somesuch. The situation with Microsoft software is not significantly better, especially taking into account that people have different versions of it. Besides that, if you're assuming everyone you might exchange documents with has all the same fonts you do, you're living in serious delusion.
Fundamentally, word processing formats were not (any of them, AFAIK) designed for online distribution; they were designed for creating something that's destined to be printed on paper. If you want online distribution, well, that's what we've got the web for, isn't it?
It's about customizeability. I know, most people don't care; they just leave all the settings they way they come from the factory, never change so much as the wallpaper, much less any actual functionality. Hey, if that makes you happy, whatever floats your boat. Me, I prefer to have the system set up to behave the way I want it to behave, manages windows the way I want them managed, and so on. When I try to use somebody else's system, that isn't set up my way, it's a real pain, because nothing works the way *I* want it to work, and the default behaviors are highly suboptimal for the way I work.
For me, it's all about that, about the ability to teach the computer to work with me, rather than the other way around. After all, who owns whom? The computer belongs to *me*, it should work *my* way; why should I train myself to do things its way? If you were an employer, would you train yourself to do things your employees' way?
> Could you please explain what's "Ma Bell" for us foreigners?
Ma Bell was what people in the US called the phone company, when it was a total monopoly. Then in the twentieth century the US government split up the phone company into a lot of smaller local phone companies, which we now call Baby Bells -- but they're not really babies at all; a better term would perhaps be Bell Brothers or simply The Phone Cabal.
Because of the way the split-up worked, everyone now gets to choose which Baby Bell to buy long distance service from, so they compete on that, but for local phone service you either buy from your local baby bell or else you do without phone service. So where before Ma Bell grossly overcharged for long distance service, now instead we pay ridiculous and exorbitant rates just to *have* a phone line, with all kinds of additional surcharges and fees (which vary unpredictably from month to month), and every little extra feature we might want is extra. Long distance calls, however, cost only a little. It costs more to have call waiting for a month than to talk to someone 500 miles away for three hours. You want a touch-tone line, instead of pulse? It's extra. Caller ID? Extra. Anonymous Call Block? Extra. You want DLS? Sorry, your line isn't approved for that. You want to call France? No problem, same low rate as France.
> The new OpenDocument Fellowship is working with a petition
> to get Microsoft to implement the format.
Petitions are meaningless. Nobody pays them any heed. You really want Microsoft (or anyone else) to do something? Write them. Personally. Preferably on business letterhead, in Microsoft's case. That's *MUCH* more likely to have an impact than signing any petition. In all history I am not aware of a single documented instance of anyone being presented with a petition, seeing how many people signed it, and, as a result, changing their mind and doing as it suggests where they previously were planning otherwise. The primary effect of a petition is to make the people who are collecting the signatures feel placated, feel as if they are "doing" something. A secondary effect is to waste the time of the people doing the signing.
> I don't want to sound too much like flamebait but how on earth is this a Laptop?
Bear in mind, screen sizes are measured on the diagonal. One could easily fit a 20"-diagonal laptop on a 17"-across lap, which is not an unusually large lap, at least around here.
It would certainly still be more readily portable than a typical desktop, and yet (especially if the keyboard takes advantage of that extra space) rather more usable, in day-to-day desktop-replacement terms, than a typical small notebook system. If you have a desktop system for regular use and want something small to carry to every meeting and class to take notes, this is probably not what you're looking for. Some people, however, do not want to deal with having two computers to maintain, and they'd like to be able to take the computer with them easily when they go someplace for the weekend or whatever; that is the niche these large laptops are designed to fill. Yes, "table-top" might be a more accurate term taken literally, but "laptop" is the industry standard term for folding all-in-one portable computers.
> The only thing the CTP games had in common with the real Civ
> franchise was this word: "Civilization".
Oh. This was not made clear to me. I just picked up the CTP demo because it was the only demo of a commerical Civilization game that I could find available for my operating system. I have too much experience with software in general and games in particular to shell out money for a game without seeing a demo first; the only commercial games I've *ever* purchased without playing a demo first are the Lost Treasures of Infocom, and that's a little bit different.
> Either your command is wrong (it should do that formatting as its
> output) or you should have used [g]awk instead of cut.
If you have to go to that much trouble, just open the command's output as a pipe in a Perl script: open FILEHANDLE, "command |"; then just read from the filehandle and process the data in whatever way you like.
> What makes MSH rock is that it's a python-like programming languaje
> PLUS a user-oriented (user=administrator) shell like bash.
Oh. And here all along I'd been lead to believe it was a Haskell-like programming language with a shell interface.
But OOP-versus-PFP paradigm issues aside, your basic point, that the interesting thing about Monad is the combination of the scripting language with the shell interface, is essentially right on the money. As a programming language, it's not that innovative -- we've already got great programming languages, including great PFP languages like Haskell, and in terms of multiparadigmatic languages, Perl6 may very likely beat Monad to release at this point (though it's going to be close), and *as a programming language* would blow its doors off, in terms of practical and useful innovation. And for people who prefer to stick with an OOP approach, Python and others have considerable mindshare already, so it would be an uphill battle *if* Monad can even compete there, which from what I know of it it may not, as it leans, from what I am given to understand, more toward the functional side of things, paradigm-wise. But Monad does two things these others do not: first, it's built into a command-line interface, and second, it's *eventually* supposed to get bundled into the OS so that you can just count on its being there on every (Windows) system, which will be real nice. I think it's a significant shame that they scrapped the plans to include it in the Longhorn desktop (now called Vista), and I hope it does make it into the Longhorn server offering (whatever the marketing department ends up calling that).
The only thing cooler than a great language built into a great command-line interface is a great command-line interface and a great language built together into a great editor, a la emacs/elisp/eshell, which absolutely rocks. With Monad replacing cmd, the text editor is IMO the next place Microsoft needs to look, in terms of core improvements to their basic toolset. Notepad is *LONG* overdue to be replaced with something rather a lot more powerful. Wouldn't it be great if the editor they replaced it with was fully scriptable with Monad and had a Monad shell built right into it? Talk about getting developers on board! This would be worth a hundred thousand conferences and symposiums with pep talks and CEOs chanting "Developers, Developers, Developers!"
Papers: "Google will do everything!"
Balmer: "Microsoft will do more!"
> who take my ideas and produce half-assed clones
Only thing is, in some cases the clones are *better*. I tried to like the Civ CTP demo enough to buy the game, really I did. If it had been even *close* to as much fun to play as Freeciv, I would have bought it just for the variety, but I found myself hating it, because the stupid UI kept getting in the way of the game. After that I never looked at a demo for another commercial civ game. (I did buy Railroad Tycoon II, though. That demo was actually fun, so I bought the game.)
And yeah, the message he's sending to the Freeciv developers (notice especially the preface he slaps on it; he's implying that how he views any given "clone" might be contingent on something related to what follows) appears to be basically "Don't let me catch you doing anything questionable in its legality, because we will defend what we legally can." Okay, that's fair.
The beer and hot dogs and bacon are minor details. When it says "most distinguishable traits", it's probably talking about things like Bart's fabled disrespect for authority, which for cultural reasons Badr wouldn't be able to have. Also bear in mind that Omar will necessarily be the undisputed ruler of his household.
In other words, they're basically adapting the drawing style of The Simpsons to a completely different cast of characters in a show with a completely different focus and appeal.
> Microsoft has claimed the cost of software is not an important issue
> in the developing world.
This is true, but Microsoft's reason ("they won't have the expertise") is irrelevant. If they have computers, they'll develop the ability to use them (for the few hours a day when they have power). The real reason software licensing costs are a non-issue in the developing world is because copyright law is (viewed as) a non-issue there. They think nothing of installing one OEM copy of Windows 98 on every computer they can get their hands on. Any phone-home stuff Microsoft might build in doesn't make any difference either, in a world where getting to a phone line means an all-day journey to the city.
> that will surely be a victory for Oracle, who will secure a strong lead
Umm, Oracle *has* a strong lead. They're the big boy in this market. What they'd allegedly be doing (*if* they do what you suggest, which at this point is speculation, however likely it seems) is [attempting to] squash a much smaller competitor. MySQL AB is certainly not the competitor they'd most like to squash, nor as far as that goes the second, but those companies have strong businesses in other markets (markets where Oracle has no product) and are overall larger companies than Oracle, so there really isn't a whole lot that Oracle can do there. Whether they would consider it worth the potential negative PR (and other costs, but that's probably the largest cost and certainly the most difficult to quantify) to squash this much smaller player is an open question. Personally I think it would be unwise, but I don't make decisions for Oracle.
> Cause more OSS FUD: "What will happen to your open source vendor? It
> could evaporate tomorrow. Stick with Oracle, who will be there for you."
This one could backfire. What can happen to InnoDB or MySQL can happen also to Oracle. There's always a bigger fish. There are, after all, larger companies with their hands in the database market (two of them, the way I count). Sure, Oracle's offering is in every way superior to a certain larger company's database (probably to that of both larger companies, but especially to the one that's a larger concern, which I am going out of my way to avoid naming), but when has that ever stopped a business from being shut down or otherwise incapacitated by a larger competitor using devious means?
> Many companies were just waiting for the 5.0 release to try it out
> I'm sure
I know of a company that officially cites stored procedures as a critical reason why they can't make their product work with MySQL. This is presumably just talk, as their product doesn't work with Oracle either, or anything else that doesn't come out of Redmond (well, okay, the web service does work with other browsers, at least), but it is nevertheless the excuse they currently are using. One of their engineers actually asked me, when he found out I'd used MySQL a little, if I knew how soon their new release was coming out, since he'd heard they were implementing stored procedures. (I guess management was feeding the engineers the same line they were feeding customers.)
This is neither here nor there for the company in question, partly because they're a Microsoft shop anyway and so I can't imagine they'd actually support MySQL, and partly for another reason I won't state here. Nevertheless, it seems to me that your observation above may have some merit; surely this ISV I'm talking about cannot be the only one that considers stored procedures an important feature.
> The mysql vs. postgres thing gets so out of hand.
You want to see out of hand? Start a discussion functional programming versus object-oriented programming with a Python programmer and a Scheme programmer.
> and the majority of CPAN is missing.
The significance of this statement cannot be overstated. The CPAN with its wealth of modules is such an enormous benfit, this alone would make Perl five times better than any of the languages the other poster listed, even if it were otherwise nothing special (which it is).
> Perl gives one the ability to write very very obsfucated code.
Most of it only looks obfuscated to people who don't speak Perl. Once you become fluent, you read code that others would consider obfuscated and ask, "What's the big deal?"
(There are exceptions, of course; code that's used ACME::Eyedrops and whatnot really is too obfuscated to be clearly legible, and then there's Matt's Script Archive. But nobody plays those games in production code.)
> The CVS version is very usable.
Yeah, Emacs 21.0 was fairly usable too, but like I said, the slow release cycle isn't, IMO, a big deal with text editors. The major ones are a much more mature product than most other kinds of software these days, so we're not sitting here going, "I just can't go another *day* without having [some feature it doesn't have yet]", like we still are with e.g. OpenOffice (which *still* doesn't have parenthesis matching, for instance, a sore omission).
> You may argue that nobody willingly uses variable names like
> ll1O0, l1100 and lllOO
I would argue that no programmer worth his salt willingly uses fonts that don't make those characters easy to distinguish. I certainly don't. However, unicode characters are another matter; I only care about how my fonts display safe and sane characters (i.e., ASCII characters with decimal values from 32 through 126). Unicode characters can show up as a character-sized box as far as I'm concerned, because they don't belong in my source code, and the editor shows control characters as \nnn escapes anyhow (except for line endings (which it displays as newlines) and horizontal tabs (which it replaces with spaces in most modes, because that's how I have it configured)).
> I mean, they're better, right?
No, Dvorak is not better than QWERTY.
The idea that the Dvorak layout is "better" comes from the assertion (which has never been conclusively established, as far as I am aware) that it allowed for a faster typing rate, i.e., a larger words-per-minute number. That *may* have been relevant in the 1950s, when a large percentage of typing consisted of retyping existing material, because typewriters didn't have save and restore, copy and paste, or print merge capability (among other things). This is no longer anything like relevant, because these days almost all typing is on-the-fly, coming right out of the brain, which only goes so fast. I used to be able to type 50wpm or so, back when I used to type in things that I'd first written out on paper (e.g., compositions that I'd worked on at the library or in study hall). It's been so long since I've done that, I don't think I can type faster than about 30wpm now, yet I find myself pausing here and there, because I'm still typing faster than I can compose.
Now, there *are* problems with the QWERTY layout, problems having to do with certain keys (notably shift) being way too far out of the way, other keys (notably CAPSLOCK and the left window key) too easy to hit by mistake, and with one entire digit going unused, because there's one big fat spacebar where both thumbs go; however, Dvorak makes these same mistakes in pretty much exactly the same way.
> It's Fermat's Last Theorem...
What's that got to do with anything? Fermat expressly stated what sets the numbers in question had to belong to. The slashdot poster just used variables from the letter ranges normally associated with those sets (e.g., n for a natural number) and assumed that anyone who's had any math would understand. Either way, if you remove (or ignore) the constraints on the numbers, it's trivial to define new groups of numbers that "solve" the thing, but that was not the intention.
> I think typing "I, for one" is redundant and annoying. Stop it.
I, for one, welcome our new redundantly-welcomed "I, for one"-typing redundant overlords.
The major text editors all use a slower release cycle. Do you remember how long we waited for Emacs 21, with its wealth of promised new features? That seems like a long time now, and yet Emacs 22 is not on anyone's radar yet, as far as I am aware. I'm not a vim fan, but the slow release cycle is not one of the reasons.
> but death would be too good for...
Huh? Nobody said anything about shooting him to *death*. Think in terms of a double-barrel sawed-off shotgun loaded with rock salt.
You know, in all the *many* discussions I've read about vim over the years, one thing I've never heard is "Well, it has lots of features, but the problem is, it's so bloated and buggy!" Have you ever heard that criticism leveled at vim? Does it have a problem in that regard? If so, this is news to me.
I have, numerous times, asked vim afficianados, "Does it support foo, bar, and baz?" (where foo, bar, and baz are any three arbitrarily chosen advanced features I use on a regular basis in Emacs and would not want to live without), and the answer has NEVER been "Oh, yes, it does all three of those things." It's been a couple of years since the last time, so we can try this again: does it support automatic mode-dependent (i.e., sensitive to the type of file you're editing) grouping-symbol matching (wherein when you type a closing grouping symbol it highlights the opening one)? Does it support correct automatic rewrapping of nested quotations in email and usenet messages? Can it be customized so that when editing a certain type of file, inserting certain characters has additional effects?
It seems to me that a critical lack of important features is vim's one major shortcoming, and that's what development really ought to focus on. Bugs in vim? I've never heard of that being a serious problem.