I will never understand the mentality that figures it is better off using a halfassed product offered for free by amateurs with no support TODAY than using a full featured product offered for free from a multi million dollar software company that's been around for thirty years but MIGHT not be here in another ten.
I mean, why deal with SHITTY NOW when something that is good MIGHT be SHITTY LATER? Do you go to a restaurant and ask for a raw steak because, if you cooked it, it might go bad in a few weeks, and with the raw steak YOU can be sure that it's kept properly refrigerated and you can cook it exactly how you like? I'm sorry, but I'm not that good a chef. I like my software well done.
Furthermore, with all the easier to use, free as in beer tools that encode and decode MP3, the only reasons to use OGG are a) politics and b) quality. Now that both WMA and iTunes are offering free as in beer codecs with (b) that offer integral CD burning, etc, you have to actively decide to sacrifice easy of use for politics to use OGG. And since the politics of 'free to alter' software aren't something many people care about in the least, you'd have to sacrifice easy of use for nothing.
And that's why I don't give a shit about Vorbis, Theora or anything. I look at it like this: is it a good codec? Yes. Does it have industry support? No. Do good codecs without industry support "make it?" Fairly rarely. Obscure formats limit your possible usage in a much more immediate way than the possibiity of writing your own linear audio editing app using native Virbis ever will. So I don't care.
Well, the reason Microsoft includes codecs for download on their site is that the codec producer paid them to do so. They're not giving away bandwidth just because it's cool.
If you want to get DivX, Theora, Ogg or WHATEVER in WMP, you'd have to pay the same price Intel does. It's not impossible, and if it's important to the community to do so, have a project manager get in touch with Microsoft and we'll start a collection.
VP3 is BETTER for low bitrates and low resolutions. It was one of the big improvements added to the PocketMVP player before I sold my PocketPC and stopped caring.
Check before you comment. It's not that "American Pie" isn't available for "buy the song," it's not available AT ALL. In fact, you can buy tracks 2-10 of the album American Pie...but without that one song, it's not much of a record, you know?
I think they must have changed something for 4.5. Gearsec and iTunesHelper are no longer services...just the iPod Helper. I was wondering why I no longer saw its 8 meg presence in my system. So maybe this whole argument is moot.
Windows swaps unused program memory to disc anyway, so most of that 50 gig is sitting on the hard drive when it's not used.
Hey, if you don't like it, don't pay for it. One could make the argument that $5 is too much to pay for sugared water, a lump of plastic, fried potatoes and seared meat between bread. That doesn't lower the cost of a happy meal.
Plenty of people like iTunes. They've sold 70 million songs in a little over a year. Obviously these people think they're getting something worthwhile. I think it's great that people are willing to pay for what's essentially a bunch of bits. Seeing as I like computers and creating content for them, I'd like to eventually be able to sell some of my bits...and through iTMS, there's now a market for them.
- NOT to support artists, who have no say in whether or not their music is on allofmp3 and get little to no money in return
- NOT to support change in the US by saying "hey, I am willing to buy music for a fair price," but instead demanding an impossibly low price.
- NOT to support the concept of copyright, which is all artists have to protect them from rampant piracy. It's not just there to make corporations money, you know...there are artists whose work is so heavily pirated they've basically become unviable (several hip-hop acts come to mind).
- CHOOSE to support a black market whose "legality" is based on a crooked organization who has assumed control over music it had no part in making and plays no part in supporting. You may think that it should be legal, but in my eyes it's no better than KaZaa.
Uh, actually the services DECREASE the damage from a crash. If iTunes crashes, your music downloads still go through. If the Music Store crashes, your music still plays. I've never had to restart gearsecurity, and it DOES get shut down when it isn't in use. Furthermore, the resources it "sucks down" are paltry compared to Nero (20 meg) or DirectCD (14 meg). Anyhow, it's not that Apple can't HANDLE multithreading, just that they didn't. They chose to modularize the program. Which is kind of commendable...it's nice to know if I'm not burning a CD, it's not loaded into memory. Or that if I didn't use my iPod, I could turn off iPod support. Worthwhile when your music program wants to use 60+ meg.
Anyhow, the reason they "worry" about the music is that iTunes, unlike many other jukebox-type apps, is built to handle massive amounts of music. I have 20,000 files in my database. Doing anything in that database -- adding to it, removing from it, even sorting it -- could take massive resources from the processor and lock the program, stopping the music. Adding a few thousand songs CAN cause the interface to freeze for a minute, but it doesn't stop the music, mostly due to some clever design including the threading model. Furthermore, iTunes is one of the few music programs I've seen that doesn't stop playing while a CD is spinning up on my old Athlon Gig. Although, it does bother me that it has to skip a second to update track info in a song that's currently playing...it doesn't do that on the mac.
I think, at the end of the day, victory will go to whoever can court the most artists. Music stores are really pimping their exclusives, and obviously it'll be harder for Apple to grab exclusives on SONY artists now that Connect is live. But Apple is doing something most labels aren't -- it's courting the independent labels and targeting specific markets with their excellent Jazz, Classical and Audiobooks sections. As such, they have an impressive NUMBER of songs, as well as an impressive DISTRIBUTION of artists. I am more likely to find what I am looking for on iTunes than on any other music service.
However, iTunes does have a major problem, in that by allowing labels to control their songs on a song-by-song basis, a lot of labels have removed one track off an album just so you can't buy the whole thing for $9.99. Usually, this one track is a skit or something, but occasionally, they remove the #1 hit. Try finding Don McClean's "American Pie," for example. This is friggin' annoying...and a lot of other services (with, I'm sure, more slipshod toolsets) don't have this problem.
Incidentally, the technology you're speaking of is background services. iTunes uses them to perform any task that might take processor time away from music playback...they use separate services for managing the iPod, downloading from the music store and burning CDs.
It's kind of an unusual practice in the PC world...generally, services are used for things which run all the time and are related to maintenance of the system. So why does iTunes use services as opposed to threads? Well, I dunno. My theory is that contacting services is similar to contacting the background threads used in OS X, making it easier to adapt the existing codebase.
Anyhow, if you're still using Windows 98, throw it away! God, that operating system is a piece of shit....Windows 2000/XP is worth the price just for the time saved from not having to reboot all the time or worry about bluescreening.
Oh, the same tired tale. You know, if you want to get the record companies to stop promoting shitty bands, maybe you should first get people to stop buying shitty albums. It is a tragedy that Britney Spears has sold ten times as many records as Tom Waits with a quarter of the talent...but them's the breaks.
I mean, it's not like independent labels don't exist, don't sign bands and don't release albums to the mass market INCLUDING over iTunes. In fact, iTunes has more independent labels than any other online music store. You don't think SONY Connect is gonna court Asian Man Records, do you?
I've actually been quite surprised by the number of GOOD artists who get above-the-fold promotion on iTunes. Even the ones who don't are displayed RIGHT NEXT to their corporate shilling brethern. And the selection beats the crap out of the most ecclectic brick & mortars I've seen. I mean, iTunes has the friggin' Kind Geedorah record. I had to literally threaten violence at my local Newbury Comics to get them to even ORDER that shit, and even then it cost me $20.
Some cases will use these drives as internals...but they're a bit thicker than will fit in a standard 2.5" drive bay, so they have to put them into an external enclosure. Some older systems will fit bigger hard drives; in fact, I used to have a Pismo powerbook that I swear could fit TWO laptop drives in it. It was also twice as heavy as my TiBook.
Don't worry....regular 100 gig 2.5" drives are coming soon!
Heh. In college, I was a member of a group that was sharing Japanese CD games that were never offered in the US. Hard drive space was a premium...i think I had MAYBE 10 gig. But I had access to a number of older CD-ROMs...so, I would rip and compress the CDs until they'd fit two-to-a-CDR, burn them, and share them off these old CD-ROMS. I had 6 of them going...
Eh, java's not so bad. It's wordy, but you have a lot more linear control over what's going on. Scripting languages are great for common tasks, but the second you get into the wierd, you start to get inefficient.
For example, I had to reformat email piped in from one process before sending it through a very twitchy server. So I wrote an EmailFilterStream that would do the checking, and "piped" the data through that before writing it to the disc. The code looked something like this:
void writeMail(InputStream is){
EmailFilterStream efs = new EmailFilterStream(new FileOutputStream("filename");
int charIn;
while((charIn = is.read()) != -1) efs.write(charIn);
efs.close(); }
Now, if I wanted to read that in by line, I could wrap the input with a Reader and the output with a Writer. If I wanted to buffer the input, I could wrap the InputStream in a BufferedInputStream. There's all sorts of stuff I could do that would be kind of cumbersome in Python. Best of all, I got a multipurpose stream I could pipe ANY output through and wind up with a valid email.
Yeah, there are about a hundred thousand ways to do this. I like mine.
I especially like his abuse of the static class instantiator to execute code while the interpretter is still searching for a void Main(string[]) method. Very sneaky.
If nobody likes to write in perl, then why write in it?
I guarantee you: there is not a THING you do in perl you can't also do in Java. Meaning that you could easily use Jython.
There's probably also not a thing you can do in perl under Windows that isn't already exposed via COM. So use JScript under WSH!
I mean, who cares how MANY packages there are that do the same thing in fewer/more lines, with more/fewer functions, and better/worse documentation, when you don't like the language in the first place?
Perl is an elitist's tool, the programmatic equivalent of a Ford F150. If all you need is an efficient commuter, STOP USING PERL.
In my experience, it's not that easy. Yes, if you take something from your CLI like "start program A, pipe output to program B, continue," it'll be slower. But the advantage of COM is that if you commonly interoperate the functions of two programs, you DO NOT have to start a new program each time. Instead, you "instantiate" the object, and then just call the function of the other. Oh, and piping can be a bit faster for string ops...but if the COM programmer understood streams at all, it should be faster overall for large operations to use COM.
requires a huge amount of effort and code bloat to create your own.
Now, this simply isn't true. COM has been INVISIBLE to programmers writing in VC since...well, it's been invisible for so long I don't even know when it started. Just write your damned program and compile it, and there is your COM stuff, right there.
requires a messy register/unregister process to distribute or use
COM doesn't REQUIRE that everything be registered or unregistered...the registry allows you to install the object wherever you like without having to update all of the associated programs to tell them where it is. If all of the COM objects you use are in the same directory as your code, you can just reference them by name. This is no different from maintaining a path or references in config files...except that COM's registration is a bit more apocryphal thanks to GUIDs.
even with the huge amount of COM objects MS offers, they still can't match the functionality of common unix tools.
This simply isn't true, and it shows a massive lack of knowledge of how things work in Windows. Just about anything you can do in a program that runs on Windows can be replicated in COM...and it works wonders for hierarchical documents, something I've never had an easy time of in UNIX. Working with Excel documents under COM is easy stuff.
as was pointed out by another...COM requires a full implementation/interface to be loaded and exposed even when the client needs very limited functionality 99% of the time.
But that's really what Windows is about -- doing tasks ALL DAY. Starting a program up and keeping it running until you leave at the end of the day. To to that, you want to speed up execution at the expense of start up. If you want a program to do nothing more than what you want to do with NO outside work, move to Haskell or Ocaml.
COM collections are awful. Particularly since most of MS's languages do not know how to box a primitive into an object.
This is just wild. Collections in COM implement a VERY basic interface which allows you to iterate through them and handle them as strictly typed objects (thus enabling you to do complex hierarchical operations without removing the objects from their context). It doesn't "box primitives to objects" because primitives aren't supposed to be used that way (it would require overhead for recording the type of the primitive, and thus a good deal more memory, a premium when COM was invented, so nobody wanted an extra byte with every byte just to tell the program it's a Byte). If you want to pass an array over COM, you pass it AS an array (which would work fine), or you implement your own Collection interface to box them properly. This should take you about ten minutes, and is the way to do things.
Variants, BTW, are amazingly useful things for people who don't know how to program. They help these people to LEARN how to program syntactically before slapping them with classes and types and primitives and instantiation and pointers. Anybody who writes with variants for more than a year needs to be smacked...unless, of course, they are using them ironically.
COM sucks. Even MS knows it. Why do you think they ditched it?
Well, I think they ditched it because they realized.NET's way of strongly typing EVERYTHING, even primitives, and acc
UH, PHP could be bad, if it was formatted incorrectly, variables named poorly, etc. These are all easy things to clean up.
However, Perl can be bad simply because the syntax allows for so many shortcuts that it's impossible to keep them straight unless you are a full time Perl hacker. And I do mean hacker..."programming" in perl is like "taking a leisurely drive" through the demoliton derby. Besides the shortened syntax for everything from instantiation and object to loading and parsing files, there's always a bunch of different ways to do things and no cogent way of telling when one of these synctactical shortcuts is performing a side effect operation that is essential to the rest of the code.
Surely there is GREAT code written in Perl, but all of it ignores this foolish micronization of syntax. In fact, one could just as easily fault the writers of GOOD perl code for doing everything wrong and not using the shortest, most efficient way of performing operations (thus running up the interpretation time). I know it's kind of a dumb argument, but lanugages like Perl are EXACTLY why I prefer to work in C# and Java. I'd rather take the time to do it self-explanatory, rather than find out I did it wrong and have to remember what -p does again.
The difference is, even as a non programmer...or a programmer who isn't deeply entrenched in C# syntax...you can tell what that c# is doing. The similar perl line would be absolutely inscruitable.
And you don't have to type that much more...with autocomplete in VS.NET. Code that's self teaching as well as self documenting is worth a LOT more to me and my having-to-support-shit-i-wrote-drunk-and-at-the-la st-minute-a-year-ago job then being able to write a few more usefull characters per second.
True, true. Over the past five years or so I've used about a dozen scripting languages, and none of them has really stood out and said, "pick me, I'm the best." For starters, some of the fastest languages to write in are also the most esoteric. And some of the most forgiving languages are among the slowest. Here's a brief breakdown of when *I* use which language.
Oh, and I do not use PERL online, as PERL is WAY too powerful a scripting langauge to allow web visitors to access. Sure, you can tie it down pretty heavily, but why worry about the hassle? JSP has a great security model and I'm used to it, and PHP, being DESIGNED for the web rather than ADAPTED to it, is a nice compromise. These are personal biases and preferences based on my own experiences, please don't flame me because your biases, preferences and experiences WILL be different.
On the web server, when I want to give it to other people: JSP / JScript (under ASP). On the web server, when I just want to get it done: PHP On the web server, when money isn't an option and I want to give it to a web designer and never worry about it again: ColdFusion
On a UN*X machine, when I want to give it to other people: Python On a UN*X machine, when I just want to get it done: Perl On a UN*X machine, when I just might have to contact a Java object: Python
On a Windows machine, when I don't care about accessing COM objects: Perl under WSH On a Windows machine, when I need good drag and drop support, pretty menus, use COM objects, use Windows networking resources, etc: JScript under WSH (I am more comfortable ending likes with semicolons); On a Windows machine, when I might give the code to somebody else who's mind is hopelessly stuck in ASP/VB land: VBScript
Oh, and don't overlook the usefulness of combining these guys! Marshalling output between scripting languages with UNIX shell scripts or WSH containers is not a black art...if you don't have the time, and you know how to do function XX in Perl and function YY in Python, why NOT pipe XX to YY? I have scripts on my webserver that use three or four different scripting languages, merely because adapting pre-existing code was quicker than porting it.
Use, yes. Contribute, no. That's the whole POINT, isn't it? That OSS offers something different because you CAN contribute to it, whereas with commercial software you have absolutely no control over the underlying system.
I mean, that's really the only reason TO use Linux. It's not like it's a more robust product that offers fewer time wasting hassles.
No, it doesn't prove a damned thing. YOU wouldn't have to use it, any more than you have to compile iptables into your kernel. And there's no reason why it should use more overhead than, say, PERL.
Your fear of adding new things is typical among Linux users. They can be added, but if optimizing for them requires modifications to ANYTHING else, opt to slow them down. And so we're stuck trying to patch holes in X. Fuck that, man, I'll stick with Windows...or better still, OS X.
Slashdot Mirror
Excellent post Twirlip of the Mists!
I will never understand the mentality that figures it is better off using a halfassed product offered for free by amateurs with no support TODAY than using a full featured product offered for free from a multi million dollar software company that's been around for thirty years but MIGHT not be here in another ten.
I mean, why deal with SHITTY NOW when something that is good MIGHT be SHITTY LATER? Do you go to a restaurant and ask for a raw steak because, if you cooked it, it might go bad in a few weeks, and with the raw steak YOU can be sure that it's kept properly refrigerated and you can cook it exactly how you like? I'm sorry, but I'm not that good a chef. I like my software well done.
Furthermore, with all the easier to use, free as in beer tools that encode and decode MP3, the only reasons to use OGG are a) politics and b) quality. Now that both WMA and iTunes are offering free as in beer codecs with (b) that offer integral CD burning, etc, you have to actively decide to sacrifice easy of use for politics to use OGG. And since the politics of 'free to alter' software aren't something many people care about in the least, you'd have to sacrifice easy of use for nothing.
And that's why I don't give a shit about Vorbis, Theora or anything. I look at it like this: is it a good codec? Yes. Does it have industry support? No. Do good codecs without industry support "make it?" Fairly rarely. Obscure formats limit your possible usage in a much more immediate way than the possibiity of writing your own linear audio editing app using native Virbis ever will. So I don't care.
Well, the reason Microsoft includes codecs for download on their site is that the codec producer paid them to do so. They're not giving away bandwidth just because it's cool.
If you want to get DivX, Theora, Ogg or WHATEVER in WMP, you'd have to pay the same price Intel does. It's not impossible, and if it's important to the community to do so, have a project manager get in touch with Microsoft and we'll start a collection.
VP3 is BETTER for low bitrates and low resolutions. It was one of the big improvements added to the PocketMVP player before I sold my PocketPC and stopped caring.
Check before you comment. It's not that "American Pie" isn't available for "buy the song," it's not available AT ALL. In fact, you can buy tracks 2-10 of the album American Pie...but without that one song, it's not much of a record, you know?
I think they must have changed something for 4.5. Gearsec and iTunesHelper are no longer services...just the iPod Helper. I was wondering why I no longer saw its 8 meg presence in my system. So maybe this whole argument is moot.
Windows swaps unused program memory to disc anyway, so most of that 50 gig is sitting on the hard drive when it's not used.
Hey, if you don't like it, don't pay for it. One could make the argument that $5 is too much to pay for sugared water, a lump of plastic, fried potatoes and seared meat between bread. That doesn't lower the cost of a happy meal.
Plenty of people like iTunes. They've sold 70 million songs in a little over a year. Obviously these people think they're getting something worthwhile. I think it's great that people are willing to pay for what's essentially a bunch of bits. Seeing as I like computers and creating content for them, I'd like to eventually be able to sell some of my bits...and through iTMS, there's now a market for them.
You also choose
- NOT to support artists, who have no say in whether or not their music is on allofmp3 and get little to no money in return
- NOT to support change in the US by saying "hey, I am willing to buy music for a fair price," but instead demanding an impossibly low price.
- NOT to support the concept of copyright, which is all artists have to protect them from rampant piracy. It's not just there to make corporations money, you know...there are artists whose work is so heavily pirated they've basically become unviable (several hip-hop acts come to mind).
- CHOOSE to support a black market whose "legality" is based on a crooked organization who has assumed control over music it had no part in making and plays no part in supporting. You may think that it should be legal, but in my eyes it's no better than KaZaa.
Uh, actually the services DECREASE the damage from a crash. If iTunes crashes, your music downloads still go through. If the Music Store crashes, your music still plays. I've never had to restart gearsecurity, and it DOES get shut down when it isn't in use. Furthermore, the resources it "sucks down" are paltry compared to Nero (20 meg) or DirectCD (14 meg). Anyhow, it's not that Apple can't HANDLE multithreading, just that they didn't. They chose to modularize the program. Which is kind of commendable...it's nice to know if I'm not burning a CD, it's not loaded into memory. Or that if I didn't use my iPod, I could turn off iPod support. Worthwhile when your music program wants to use 60+ meg.
Anyhow, the reason they "worry" about the music is that iTunes, unlike many other jukebox-type apps, is built to handle massive amounts of music. I have 20,000 files in my database. Doing anything in that database -- adding to it, removing from it, even sorting it -- could take massive resources from the processor and lock the program, stopping the music. Adding a few thousand songs CAN cause the interface to freeze for a minute, but it doesn't stop the music, mostly due to some clever design including the threading model. Furthermore, iTunes is one of the few music programs I've seen that doesn't stop playing while a CD is spinning up on my old Athlon Gig. Although, it does bother me that it has to skip a second to update track info in a song that's currently playing...it doesn't do that on the mac.
I think, at the end of the day, victory will go to whoever can court the most artists. Music stores are really pimping their exclusives, and obviously it'll be harder for Apple to grab exclusives on SONY artists now that Connect is live. But Apple is doing something most labels aren't -- it's courting the independent labels and targeting specific markets with their excellent Jazz, Classical and Audiobooks sections. As such, they have an impressive NUMBER of songs, as well as an impressive DISTRIBUTION of artists. I am more likely to find what I am looking for on iTunes than on any other music service.
However, iTunes does have a major problem, in that by allowing labels to control their songs on a song-by-song basis, a lot of labels have removed one track off an album just so you can't buy the whole thing for $9.99. Usually, this one track is a skit or something, but occasionally, they remove the #1 hit. Try finding Don McClean's "American Pie," for example. This is friggin' annoying...and a lot of other services (with, I'm sure, more slipshod toolsets) don't have this problem.
Incidentally, the technology you're speaking of is background services. iTunes uses them to perform any task that might take processor time away from music playback...they use separate services for managing the iPod, downloading from the music store and burning CDs.
It's kind of an unusual practice in the PC world...generally, services are used for things which run all the time and are related to maintenance of the system. So why does iTunes use services as opposed to threads? Well, I dunno. My theory is that contacting services is similar to contacting the background threads used in OS X, making it easier to adapt the existing codebase.
Anyhow, if you're still using Windows 98, throw it away! God, that operating system is a piece of shit....Windows 2000/XP is worth the price just for the time saved from not having to reboot all the time or worry about bluescreening.
Oh, the same tired tale. You know, if you want to get the record companies to stop promoting shitty bands, maybe you should first get people to stop buying shitty albums. It is a tragedy that Britney Spears has sold ten times as many records as Tom Waits with a quarter of the talent...but them's the breaks.
I mean, it's not like independent labels don't exist, don't sign bands and don't release albums to the mass market INCLUDING over iTunes. In fact, iTunes has more independent labels than any other online music store. You don't think SONY Connect is gonna court Asian Man Records, do you?
I've actually been quite surprised by the number of GOOD artists who get above-the-fold promotion on iTunes. Even the ones who don't are displayed RIGHT NEXT to their corporate shilling brethern. And the selection beats the crap out of the most ecclectic brick & mortars I've seen. I mean, iTunes has the friggin' Kind Geedorah record. I had to literally threaten violence at my local Newbury Comics to get them to even ORDER that shit, and even then it cost me $20.
Some cases will use these drives as internals...but they're a bit thicker than will fit in a standard 2.5" drive bay, so they have to put them into an external enclosure. Some older systems will fit bigger hard drives; in fact, I used to have a Pismo powerbook that I swear could fit TWO laptop drives in it. It was also twice as heavy as my TiBook.
Don't worry....regular 100 gig 2.5" drives are coming soon!
Heh. In college, I was a member of a group that was sharing Japanese CD games that were never offered in the US. Hard drive space was a premium...i think I had MAYBE 10 gig. But I had access to a number of older CD-ROMs...so, I would rip and compress the CDs until they'd fit two-to-a-CDR, burn them, and share them off these old CD-ROMS. I had 6 of them going...
Well, it's your own fault for using Linux. If you used Windows like me, you'd be forced to upgrade every four years or so...
Eh, java's not so bad. It's wordy, but you have a lot more linear control over what's going on. Scripting languages are great for common tasks, but the second you get into the wierd, you start to get inefficient.
For example, I had to reformat email piped in from one process before sending it through a very twitchy server. So I wrote an EmailFilterStream that would do the checking, and "piped" the data through that before writing it to the disc. The code looked something like this:
void writeMail(InputStream is){
EmailFilterStream efs = new EmailFilterStream(new FileOutputStream("filename");
int charIn;
while((charIn = is.read()) != -1) efs.write(charIn);
efs.close();
}
Now, if I wanted to read that in by line, I could wrap the input with a Reader and the output with a Writer. If I wanted to buffer the input, I could wrap the InputStream in a BufferedInputStream. There's all sorts of stuff I could do that would be kind of cumbersome in Python. Best of all, I got a multipurpose stream I could pipe ANY output through and wind up with a valid email.
Yeah, there are about a hundred thousand ways to do this. I like mine.
Oh. You just discovered yet another problem with your average scripting language: absolutly inscruitable error messages.
Say what you will about Java...you can't beat a nice stack trace for PURE context. Except maybe a dotnet app in debug mode.
I especially like his abuse of the static class instantiator to execute code while the interpretter is still searching for a void Main(string[]) method. Very sneaky.
If nobody likes to write in perl, then why write in it?
I guarantee you: there is not a THING you do in perl you can't also do in Java. Meaning that you could easily use Jython.
There's probably also not a thing you can do in perl under Windows that isn't already exposed via COM. So use JScript under WSH!
I mean, who cares how MANY packages there are that do the same thing in fewer/more lines, with more/fewer functions, and better/worse documentation, when you don't like the language in the first place?
Perl is an elitist's tool, the programmatic equivalent of a Ford F150. If all you need is an efficient commuter, STOP USING PERL.
And I'll bite back.
.NET's way of strongly typing EVERYTHING, even primitives, and acc
it's much slower
In my experience, it's not that easy. Yes, if you take something from your CLI like "start program A, pipe output to program B, continue," it'll be slower. But the advantage of COM is that if you commonly interoperate the functions of two programs, you DO NOT have to start a new program each time. Instead, you "instantiate" the object, and then just call the function of the other. Oh, and piping can be a bit faster for string ops...but if the COM programmer understood streams at all, it should be faster overall for large operations to use COM.
requires a huge amount of effort and code bloat to create your own.
Now, this simply isn't true. COM has been INVISIBLE to programmers writing in VC since...well, it's been invisible for so long I don't even know when it started. Just write your damned program and compile it, and there is your COM stuff, right there.
requires a messy register/unregister process to distribute or use
COM doesn't REQUIRE that everything be registered or unregistered...the registry allows you to install the object wherever you like without having to update all of the associated programs to tell them where it is. If all of the COM objects you use are in the same directory as your code, you can just reference them by name. This is no different from maintaining a path or references in config files...except that COM's registration is a bit more apocryphal thanks to GUIDs.
even with the huge amount of COM objects MS offers, they still can't match the functionality of common unix tools.
This simply isn't true, and it shows a massive lack of knowledge of how things work in Windows. Just about anything you can do in a program that runs on Windows can be replicated in COM...and it works wonders for hierarchical documents, something I've never had an easy time of in UNIX. Working with Excel documents under COM is easy stuff.
as was pointed out by another...COM requires a full implementation/interface to be loaded and exposed even when the client needs very limited functionality 99% of the time.
But that's really what Windows is about -- doing tasks ALL DAY. Starting a program up and keeping it running until you leave at the end of the day. To to that, you want to speed up execution at the expense of start up. If you want a program to do nothing more than what you want to do with NO outside work, move to Haskell or Ocaml.
COM collections are awful. Particularly since most of MS's languages do not know how to box a primitive into an object.
This is just wild. Collections in COM implement a VERY basic interface which allows you to iterate through them and handle them as strictly typed objects (thus enabling you to do complex hierarchical operations without removing the objects from their context). It doesn't "box primitives to objects" because primitives aren't supposed to be used that way (it would require overhead for recording the type of the primitive, and thus a good deal more memory, a premium when COM was invented, so nobody wanted an extra byte with every byte just to tell the program it's a Byte). If you want to pass an array over COM, you pass it AS an array (which would work fine), or you implement your own Collection interface to box them properly. This should take you about ten minutes, and is the way to do things.
Variants, BTW, are amazingly useful things for people who don't know how to program. They help these people to LEARN how to program syntactically before slapping them with classes and types and primitives and instantiation and pointers. Anybody who writes with variants for more than a year needs to be smacked...unless, of course, they are using them ironically.
COM sucks. Even MS knows it. Why do you think they ditched it?
Well, I think they ditched it because they realized
UH, PHP could be bad, if it was formatted incorrectly, variables named poorly, etc. These are all easy things to clean up.
However, Perl can be bad simply because the syntax allows for so many shortcuts that it's impossible to keep them straight unless you are a full time Perl hacker. And I do mean hacker..."programming" in perl is like "taking a leisurely drive" through the demoliton derby. Besides the shortened syntax for everything from instantiation and object to loading and parsing files, there's always a bunch of different ways to do things and no cogent way of telling when one of these synctactical shortcuts is performing a side effect operation that is essential to the rest of the code.
Surely there is GREAT code written in Perl, but all of it ignores this foolish micronization of syntax. In fact, one could just as easily fault the writers of GOOD perl code for doing everything wrong and not using the shortest, most efficient way of performing operations (thus running up the interpretation time). I know it's kind of a dumb argument, but lanugages like Perl are EXACTLY why I prefer to work in C# and Java. I'd rather take the time to do it self-explanatory, rather than find out I did it wrong and have to remember what -p does again.
The difference is, even as a non programmer...or a programmer who isn't deeply entrenched in C# syntax...you can tell what that c# is doing. The similar perl line would be absolutely inscruitable.
a st-minute-a-year-ago job then being able to write a few more usefull characters per second.
And you don't have to type that much more...with autocomplete in VS.NET. Code that's self teaching as well as self documenting is worth a LOT more to me and my having-to-support-shit-i-wrote-drunk-and-at-the-l
True, true. Over the past five years or so I've used about a dozen scripting languages, and none of them has really stood out and said, "pick me, I'm the best." For starters, some of the fastest languages to write in are also the most esoteric. And some of the most forgiving languages are among the slowest. Here's a brief breakdown of when *I* use which language.
Oh, and I do not use PERL online, as PERL is WAY too powerful a scripting langauge to allow web visitors to access. Sure, you can tie it down pretty heavily, but why worry about the hassle? JSP has a great security model and I'm used to it, and PHP, being DESIGNED for the web rather than ADAPTED to it, is a nice compromise. These are personal biases and preferences based on my own experiences, please don't flame me because your biases, preferences and experiences WILL be different.
On the web server, when I want to give it to other people: JSP / JScript (under ASP).
On the web server, when I just want to get it done: PHP
On the web server, when money isn't an option and I want to give it to a web designer and never worry about it again: ColdFusion
On a UN*X machine, when I want to give it to other people: Python
On a UN*X machine, when I just want to get it done: Perl
On a UN*X machine, when I just might have to contact a Java object: Python
On a Windows machine, when I don't care about accessing COM objects: Perl under WSH
On a Windows machine, when I need good drag and drop support, pretty menus, use COM objects, use Windows networking resources, etc: JScript under WSH (I am more comfortable ending likes with semicolons);
On a Windows machine, when I might give the code to somebody else who's mind is hopelessly stuck in ASP/VB land: VBScript
Oh, and don't overlook the usefulness of combining these guys! Marshalling output between scripting languages with UNIX shell scripts or WSH containers is not a black art...if you don't have the time, and you know how to do function XX in Perl and function YY in Python, why NOT pipe XX to YY? I have scripts on my webserver that use three or four different scripting languages, merely because adapting pre-existing code was quicker than porting it.
Use, yes. Contribute, no. That's the whole POINT, isn't it? That OSS offers something different because you CAN contribute to it, whereas with commercial software you have absolutely no control over the underlying system.
I mean, that's really the only reason TO use Linux. It's not like it's a more robust product that offers fewer time wasting hassles.
No, it doesn't prove a damned thing. YOU wouldn't have to use it, any more than you have to compile iptables into your kernel. And there's no reason why it should use more overhead than, say, PERL.
Your fear of adding new things is typical among Linux users. They can be added, but if optimizing for them requires modifications to ANYTHING else, opt to slow them down. And so we're stuck trying to patch holes in X. Fuck that, man, I'll stick with Windows...or better still, OS X.