Not The Best Choice For Maintainable Code
on
Going Dynamic with PHP
·
· Score: 3, Insightful
From the article:
Dynamic objects have a lot of power, but they also carry significant risk. First, the complexity of classes increases tremendously when you start writing magic methods. These classes are harder to understand, debug, and maintain. In addition, as integrated development environments (IDEs) become more intelligent, they may experience problems with dynamic classes such as these because when they look up methods on a class, they won't find them.
This is what I was thinking the entire time I was reading the article. I mean, it's one thing to have to whip up some small project for yourself, it's another to build a project that is maintainable by a group of people.
I'd bet that Brian W. Kernighan and Rob Pike (The Practice of Programming) would probably recommend against using it. It doesn't provide for clarity, nor does it simplify, it just makes things "easier" for the guy that writes the original code.
Compiling isn't the problem, it's when the application you're attempting to compile requires later versions of the libraries than you have installed & your vendor prefers to patch older versions than provide new versions (read RHEL).
AutoPackage, that's cool. I hadn't seen that before. But it seems to me that requesting modifications from developers is not a very stable way of ensuring that you get what you need.
Your arguments against the Mac UI are purely personal preference. You like blue better than red, you like menu bars on bottom, you like 28 virtual desktops... who cares. It's your decision.
Sorry, but screwing around with the "look and feel" of my desktop just isn't something I have time for; if that's really important to you, then you may be able to think hard enough to find applications that do the things you've said you "can't" do.
Keep trying to figure out the best "look and feel" for your desktop. The rest of us just have better things to do.
I work with people who work with Linux all day. About 85% of us use OS X as our OS of choice. There's still just too much fray in KDE or Gnome to consider them.
When I get home, the last thing I want is to figure out why my network card isn't working after the latest upgrade. I'm not interested in finding out that I need to upgrade 15 different libraries before I can upgrade KDE--because I've been doing it all day long. I just want to get on my computer and do what I've got to do. That's the difference. "Customization" usually means petty UI skins, and as much as I've used KDE I've never seen much that was really customizable that wasn't customizable in OS X (except UI skins).
Don't get me wrong KDE is awesome. Believe me, if KDE provided everything that OS X does, I'd be there in a heartbeat. But it's not about a pretty GUI, it's about the fact that everything "just works." I don't need to figure out how to cut and paste out of this particular application, or the quirks of any individual application, for that matter; that's because they all work the same. Maybe that's not a boon to you, but I just want to get stuff done. I don't have time anymore to screw around with quirks and idiosyncrasies of a desktop environment that is just incomplete.
it's not like anyone is trying to pretend that this is the first time it's been done.
if it's so. fucking. old. then move on--there's nothing to see here. Let the kids have fun and get on to stuff that doesn't tire you so.
Don't take this personal--because you and I both know I don't know a thing about you--but your post makes you sound like one of those bitter old men who can't stand to see kids having fun either a) because they can't do it anymore, or b) because they never had the chance to do it themselves.
The article you link to indicates that the sensors work just fine, but even when they do detect something, there is no ready reaction, or use of the data by government officials for nearly a week.
I'd say it's not the sensors that are useless, but rather the government.
Virtuozzo has vzmigrate, but in a real production environment where you're migrating customers between hardware nodes, the arp-cache is your enemy much more than the few seconds of down-time that a product's migration tool incurs.
Thank you for your response. I'd like to elucidate my point of view for a moment as it appears that I've done a poor job of explaining myself.
1) Microsoft licensing and Microsoft software can not be distinguished from one another.
2) There is a glut of technical problems with Microsoft software that I touched on as it pertains to Exchange; it is well documented (Google "Microsoft vulnerabilty"). Although part of my post focuses on the inherent problems that Microsoft licensing proves to be, I believe it's unfair to characterize it as if it is exclusively focused on the licensing problems.
It seems that my mention of numerous technical problems was not enough, or was not focused upon in my response. I will work to provide a more robust set of technical as well as licensing issues. I took it for granted that the copious documented technical problems that Microsoft software has historically suffered would provide an analog to my focus on licensing.
There is at least one well known (at least in my circles) alternative to Exchange. Take a look at this:
It's in early Alpha, and requires a whole lot more than Sendmail (as the original poster mentioned, but hey, it's Microsoft bashing, so it's OK not to read the OP right?)
No genius, the point is that Sendmail is not the same type of product as Exchange. If you want to toot about how great Exchange is because it has "features" that Sendmail doesn't have, then be ready when someone bursts your bubble and lets you know that Sendmail isn't supposed to have those "features." Nobody in their right mind wants Sendmail to be a calendaring agent, but there are other alternatives available that will do the job.
I've been managing Exchange since 5.0. I can count the number of times I've had to rescue anything from a corrupt data store on two fingers (in 12 years)...
So you let me know how 2005 - 1997 = 12, then maybe I'll start believing your best practices stories.
It's also readily apparent you've never used Exchange before, because moving mailboxes is simple...
Sorry fella, but last time I worked with Exchange, there was no individual data store for each user; everyone had everything in a single database... just about the single most idiotic thing I've ever seen. And keep in mind, the data store couldn't go beyond the size of a single partition. So you keep on believing that you know what I've done, but I'm damn glad I left the button clicking to you MCSE monkeys long ago.
Licensing is part of the Microsoft world, it's not that difficult. Nor does it take much time. Most companies that use MS products know how licensing work...
Shucks, maybe that's why the OP (Original Post, you remember that, right?) mentioned licensing as one of the main reasons why people aren't upgrading (read: paying again for bug fixes in earlier versions of Exchange). People know how it works, and when you've been screwed into client access licenses that no longer work on your new version of SBS or you're not able to upgrade from Exchange SBS to the next full version of Exchange without buying the previous full version of Exchange, you learn it's probably better to tuck that Win NT4 box running Exchange 5.5 far behind a couple of packet filtering OpenBSD routers and a DMZ FreeBSD or Debian server running Sendmail/Postfix and clamav. And the best part for that small business is that they can keep on using the same old hardware that is no longer fit to run the latest version of Windows; just repurpose machines that would be thrown-out anyway.
You know how easy it is to add new Exchange servers to an SBS Exchange environment? Very. Buy a new copy of Exchange an add to the SBS Exchange org.
I guess I wasn't very clear; there are Microsoft Tech Documents for upgrading from Exchange 5.5 to Exchange 2000 that say that you can not do this from SBS server to full version of Win2K Server and Exchange 2000. You just can't. You must upgrade from SBS to NT4/Exchange 5.5 and then upgrade to Win2k and Exchange 2000. In a world where everyone can afford to pay just for the right to access the server that you've already paid money for, maybe that's an acceptible answer, but for small businesses, this kind of classic Microsoft double-dipping makes following the upgrade treadmill impossible.
If that really was the case, why are not more people moving to [open source Exchange alternatives]...
And try to remember where this post all began... nobody is upgrading... nobody is changing the version of Exchange that they have---and likely when they do, it will be for an open source alternative.
Using Sendmail does not imply that calendaring is not available.
One quick google search using "outlook calendaring open source" yielded this among other items: http://openconnector.org/
Who needs wireless email?
Hmmm... I guess need Exchange to read email on my wireless phone. Guess I'll have to tell my people that they can't send emails to me any longer because we use Sendmail as our MTA.
Who needs signle instance storage?
Not me.
Who needs to resurrect messages from a corrupt data store?
Not me.
Who needs to figure out how to keep the mail server running once you've filled the disks with a massive file that you can't move to a larger disk (because it's being accessed)?
Not me.
Who needs to figure out why people intermittently can't connect to the Exchange server anymore when all the licenses are used?
Not me.
Who wants to deal with departments of employees calling with the same question while you wait for more client access licenses to be purchased?
Not me.
Who wants to figure out how to upgrade from SBS to an even more expensive version of Exchange (only to find out that you can't "upgrade")?
Not me.
I can go on and on.
Exchange is a fine product for some limited settings. For the rest of us, there are feature-for-feature open source alternatives that will work with Outhouse. They don't entail rediculous licensing problems inherent in Exchange and are engineered better.
Although other dialects are far more prevalent in China, Mandarin is what is used by the government.
Learning the language is sound advice for anyone who wants to invest in an economy that has the potential to burgeon far beyond the capaciy of the US and EU; but learning the wrong dialect will have you talking with farmers living in the stone age rather than the elite few with the power to create wealth.
See unattended for an open source project with all these features as well as some good community support & the ability to integrate with a build/update system that actually does more than Windows.
You forgot the , "Embrace, Extend, Extinguish" posts, which hit the nail on the head.
History shows that if Microsoft intends to deploy PDF it will not follow the standard. Like any other MS "solution," this is not a triumph for the user, it is a tool for Microsoft to leverage against anything that threatens Microsoft's income.
The worst part about it is, it's going to take another version of Office, while Microsoft breaks another standard to get back to where we were before this announcement--I mean, we were almost done with the nightmare Word is. Now it's going to take another cycle of PHBs believing that the next version of Office will solve their problem, with all the trickle-down frustration that it means for us. It means that it will be years to get back to where people realize that Microsoft promises mean only one thing: Microsoft profits. Of course, by that time there will be more glitter and flashing lights to distract them--and decades of disappointment and frustration will be forgotten.
That you don't understand why people would greet this announcement with skepticism is strange to me; I don't understand how you can be so naive regarding this when we have a history of this kind of action laid out behind us.
Disclaimer: I'm a Mac OS X convert as of two months ago
Welcome. As an OS X convert that followed much the same path you have (just about 5 years ago), I think you'll find that you'll finally stop hopping from OS to OS, from distribution to distribution trying to find something that actually works.
Linux / FreeBSD: Works well. If you know your stuff, it is easy to fix stuff and set up. I've had issues with upgrades however, after some time it will eventually mess up. Desktop applications are a mishmash of good and bad, or poorly thought out in a single crucial aspect whilst being very powerful.
I run Linux servers for a living. I have run a smattering of Linux disros, FreeBSD and OpenBSD as desktop OS's. I currently have an Ubuntu X server at the house, and although I'm not saying you're wrong, I am saying that upgrades have been a relatively minor concern over the years.
I don't think it can be beaten as an end-user operating system, however I can see that it lacks certain things that corporations would like in a desktop computer.
Please elucidate; I see much more offered by OS X than Windows to a corporate environment .
Okay all you web designers/coders, next time your boss tells you he wants something on the site that uses some crappy activeX component, or uses some of Microsoft's built in "Java"script (because the competitor is using it), you just tell him, "piss off, that's not going to work for all browsers if I use a single code base."
tell him jwhitener said so.
oh and if you can't do it with one code base, y00 suck lam3r... jwhitener says you can ALWAYS do it with one code base. RTFM d00d!!!!!!1111!!!111!!11oneone
and if you'd ever read any of the dialogues of Plato you might think better of stating it as if it were an attribute exclusively exhibited by peoples of the present and our "declining education standards."
if you're so worried about the standards of education, then do something. otherwise, shut your pie hole.
My personal bet is that if the latter happens (OS X on standard machines), within 10 years we'll see a 50% Windows, 30% OS X, and 15% Linux, 5% other varients in the desktop market - in the server market it may be much as it is now, maybe with OS X and Linux overtaking the bulk of the traditional Unix route.
With its new version of Windows, Microsoft will be leaving behind the old model of trying to cater to everyone. Making the OS cost more, and making the piracy prevention harder to circumvent, they intend to "take the high road" and sell only to those who can afford their price.
This leaves an ever-easier-to-use-and-install Linux poised to take much of the "low end" market.
The next generation Apple computers will compete on both levels, mostly against Microsoft's offerings, but with the use of Intel chips, Apple has finally broken the last psychological and system related hurdle keeping the masses from switching to a better platform.
If you will, I believe Apple will be competing for the "middle class" of end-user computing. This was a market created by Microsoft. Microsoft itself is leaving it behind believing that the middle class is going to move on from current general computing on to a more appliance-based environment (i.e. game consoles that are created for more general tasks than the consoles of the past, but more task-specific than the PC of yesteryear).
So anyway, that's my take on it. I think this is a great time to be a Cocoa developer. I think this will break-open the confined user base just about the time that XP users will really start looking at their next computer.
Re:Microsoft doesn't deserve this criticism
on
Korean MSN Site Hacked
·
· Score: 3, Insightful
The news here is that it wasn't just a vulnerability published, nor a proof of concept, it was a full fledged crack attack against one of the sites that represent the corporation itself. The news here is that it's the same old Microsoft. The news here is that "Trustworthy Computing" is just another marketing buzzword.The news here is that if you can't even manage to secure your own servers, how do you expect the rest of the world to do it?
Microsoft deserves every bit of blame that they get. They want to pretend like security is something that can be applied like a coat of paint, but in the end, incidents like this prove that it's the same old crap rolling out of Redmond.
But why should it be necessary to read, man pages for different systems? Why are tools like Cfengine necessary? (Shout out to CFengine devel team, you kick butt!)
see profuse documentation for different systems basically doing the same thing as a necessary evil rather than a inherent benefit.
And believe me, I think monolithic programs are just wrong (see RPM for a good example of how not to create a utility). But we're not talking about different functionality here, were' talking about how often that functionality is invoked.
It doesn't sound like you've never really had to deal with supporting many different daemons in an environment supporting different *nixen.
The real problem with the "UNIX way" is not that there are small, separate services that act as baffles; we all agree that's a good thing. The problem is that each service requires a separate individualized configuration syntax; that each one is controlled by a separate program or super-daemon, each one has its own unique problems and attributes, and each require a different means of control. Next time you have to deal with UCSPI and its simple, but absolutely different zeitgeist, you let me know how easy it is to admin. Next time you need to figure out the runlevel or S and K order of a System V style rc script, or figure out how to to configure rc.conf on FreeBSD vs OpenBSD style rc.conf, tell me how easy and beneficial the UNIX way is. Tell me how nice it is to have apachectl running httpd on some boxes, and/etc/init.d/httpd on others. And if that isn't enough, then tell me how nice it is to restart xinetd whenever you're ready to reload a single service configuration.
Don't get me wrong; I hate dealing with the absolute idiocy that is Windows Server, but uniform daemon control is nothing like trying to amalgamate them into a single (more vulnerable, slower, more complex and in the end, less-useful) service.
Sure slick, let me know how well that firewall works next time you click on the wrong link with IE and the next root level IE exploit installs an app on your machine that starts opening up connections from behind the firewall.
Well that's funny. I feel exactly the opposite about the whole thing. When I want to get work done, I fire up the Mac. Things just work right. It stays out of my way while I complete my work.
When I want to experiment, play and tinker endlessly with the system (it seems most often to get it working the way it already should, or find another project that works better) I fire up the Linux box. When I want to spend time learning how to get something working or the elements of a project I fire up the Linux box.
Not saying that one is better than the other, it's just interesting that you find you're more productive on a Linux box.
Slashdot Mirror
This is what I was thinking the entire time I was reading the article. I mean, it's one thing to have to whip up some small project for yourself, it's another to build a project that is maintainable by a group of people.
I'd bet that Brian W. Kernighan and Rob Pike (The Practice of Programming) would probably recommend against using it. It doesn't provide for clarity, nor does it simplify, it just makes things "easier" for the guy that writes the original code.
Compiling isn't the problem, it's when the application you're attempting to compile requires later versions of the libraries than you have installed & your vendor prefers to patch older versions than provide new versions (read RHEL).
AutoPackage, that's cool. I hadn't seen that before. But it seems to me that requesting modifications from developers is not a very stable way of ensuring that you get what you need.
Alright flameboy.
Your arguments against the Mac UI are purely personal preference. You like blue better than red, you like menu bars on bottom, you like 28 virtual desktops... who cares. It's your decision.
Sorry, but screwing around with the "look and feel" of my desktop just isn't something I have time for; if that's really important to you, then you may be able to think hard enough to find applications that do the things you've said you "can't" do.
Keep trying to figure out the best "look and feel" for your desktop. The rest of us just have better things to do.
And just what do you do when the application that you need to install isn't in your distribution's binary repository?
compile.
And what do you do when you when the application you need to compile requires versions of libraries that aren't distributed?
"yum update" is just not an option.
You're the exception.
I work with people who work with Linux all day. About 85% of us use OS X as our OS of choice. There's still just too much fray in KDE or Gnome to consider them.
When I get home, the last thing I want is to figure out why my network card isn't working after the latest upgrade. I'm not interested in finding out that I need to upgrade 15 different libraries before I can upgrade KDE--because I've been doing it all day long. I just want to get on my computer and do what I've got to do. That's the difference. "Customization" usually means petty UI skins, and as much as I've used KDE I've never seen much that was really customizable that wasn't customizable in OS X (except UI skins).
Don't get me wrong KDE is awesome. Believe me, if KDE provided everything that OS X does, I'd be there in a heartbeat. But it's not about a pretty GUI, it's about the fact that everything "just works." I don't need to figure out how to cut and paste out of this particular application, or the quirks of any individual application, for that matter; that's because they all work the same. Maybe that's not a boon to you, but I just want to get stuff done. I don't have time anymore to screw around with quirks and idiosyncrasies of a desktop environment that is just incomplete.
christ man, lighten up.
it's not like anyone is trying to pretend that this is the first time it's been done.
if it's so. fucking. old. then move on--there's nothing to see here. Let the kids have fun and get on to stuff that doesn't tire you so.
Don't take this personal--because you and I both know I don't know a thing about you--but your post makes you sound like one of those bitter old men who can't stand to see kids having fun either a) because they can't do it anymore, or b) because they never had the chance to do it themselves.
The article you link to indicates that the sensors work just fine, but even when they do detect something, there is no ready reaction, or use of the data by government officials for nearly a week.
I'd say it's not the sensors that are useless, but rather the government.
Virtuozzo has vzmigrate, but in a real production environment where you're migrating customers between hardware nodes, the arp-cache is your enemy much more than the few seconds of down-time that a product's migration tool incurs.
Thank you for your response. I'd like to elucidate my point of view for a moment as it appears that I've done a poor job of explaining myself.
:) ).
1) Microsoft licensing and Microsoft software can not be distinguished from one another.
2) There is a glut of technical problems with Microsoft software that I touched on as it pertains to Exchange; it is well documented (Google "Microsoft vulnerabilty"). Although part of my post focuses on the inherent problems that Microsoft licensing proves to be, I believe it's unfair to characterize it as if it is exclusively focused on the licensing problems.
It seems that my mention of numerous technical problems was not enough, or was not focused upon in my response. I will work to provide a more robust set of technical as well as licensing issues. I took it for granted that the copious documented technical problems that Microsoft software has historically suffered would provide an analog to my focus on licensing.
There is at least one well known (at least in my circles) alternative to Exchange. Take a look at this:
http://www.novell.com/products/openexchange/
Feature for feature it has everything Exchange has (except the licensing
No genius, the point is that Sendmail is not the same type of product as Exchange. If you want to toot about how great Exchange is because it has "features" that Sendmail doesn't have, then be ready when someone bursts your bubble and lets you know that Sendmail isn't supposed to have those "features." Nobody in their right mind wants Sendmail to be a calendaring agent, but there are other alternatives available that will do the job.
I hate to break it to you d00d, but your calculator is broke.
The following is from Microsoft's Support documentation
Exchange 5.0 5.0.1457 March 1997
So you let me know how 2005 - 1997 = 12, then maybe I'll start believing your best practices stories.
Sorry fella, but last time I worked with Exchange, there was no individual data store for each user; everyone had everything in a single database... just about the single most idiotic thing I've ever seen. And keep in mind, the data store couldn't go beyond the size of a single partition. So you keep on believing that you know what I've done, but I'm damn glad I left the button clicking to you MCSE monkeys long ago.
Shucks, maybe that's why the OP (Original Post, you remember that, right?) mentioned licensing as one of the main reasons why people aren't upgrading (read: paying again for bug fixes in earlier versions of Exchange). People know how it works, and when you've been screwed into client access licenses that no longer work on your new version of SBS or you're not able to upgrade from Exchange SBS to the next full version of Exchange without buying the previous full version of Exchange, you learn it's probably better to tuck that Win NT4 box running Exchange 5.5 far behind a couple of packet filtering OpenBSD routers and a DMZ FreeBSD or Debian server running Sendmail/Postfix and clamav. And the best part for that small business is that they can keep on using the same old hardware that is no longer fit to run the latest version of Windows; just repurpose machines that would be thrown-out anyway.
I guess I wasn't very clear; there are Microsoft Tech Documents for upgrading from Exchange 5.5 to Exchange 2000 that say that you can not do this from SBS server to full version of Win2K Server and Exchange 2000. You just can't. You must upgrade from SBS to NT4/Exchange 5.5 and then upgrade to Win2k and Exchange 2000. In a world where everyone can afford to pay just for the right to access the server that you've already paid money for, maybe that's an acceptible answer, but for small businesses, this kind of classic Microsoft double-dipping makes following the upgrade treadmill impossible.
Check this out d00d:
http://www.novell.com/products/openexchange/
And try to remember where this post all began... nobody is upgrading... nobody is changing the version of Exchange that they have---and likely when they do, it will be for an open source alternative.
Using Sendmail does not imply that calendaring is not available.
One quick google search using "outlook calendaring open source" yielded this among other items:
http://openconnector.org/
Hmmm... I guess need Exchange to read email on my wireless phone. Guess I'll have to tell my people that they can't send emails to me any longer because we use Sendmail as our MTA.
Not me.
Who needs to resurrect messages from a corrupt data store?
Not me.
Who needs to figure out how to keep the mail server running once you've filled the disks with a massive file that you can't move to a larger disk (because it's being accessed)?
Not me.
Who needs to figure out why people intermittently can't connect to the Exchange server anymore when all the licenses are used?
Not me.
Who wants to deal with departments of employees calling with the same question while you wait for more client access licenses to be purchased?
Not me.
Who wants to figure out how to upgrade from SBS to an even more expensive version of Exchange (only to find out that you can't "upgrade")?
Not me.
I can go on and on.
Exchange is a fine product for some limited settings. For the rest of us, there are feature-for-feature open source alternatives that will work with Outhouse. They don't entail rediculous licensing problems inherent in Exchange and are engineered better.
Yep.
Although other dialects are far more prevalent in China, Mandarin is what is used by the government.
Learning the language is sound advice for anyone who wants to invest in an economy that has the potential to burgeon far beyond the capaciy of the US and EU; but learning the wrong dialect will have you talking with farmers living in the stone age rather than the elite few with the power to create wealth.
See unattended for an open source project with all these features as well as some good community support & the ability to integrate with a build/update system that actually does more than Windows.
You forgot the , "Embrace, Extend, Extinguish" posts, which hit the nail on the head.
History shows that if Microsoft intends to deploy PDF it will not follow the standard. Like any other MS "solution," this is not a triumph for the user, it is a tool for Microsoft to leverage against anything that threatens Microsoft's income.
The worst part about it is, it's going to take another version of Office, while Microsoft breaks another standard to get back to where we were before this announcement--I mean, we were almost done with the nightmare Word is. Now it's going to take another cycle of PHBs believing that the next version of Office will solve their problem, with all the trickle-down frustration that it means for us. It means that it will be years to get back to where people realize that Microsoft promises mean only one thing: Microsoft profits. Of course, by that time there will be more glitter and flashing lights to distract them--and decades of disappointment and frustration will be forgotten.
That you don't understand why people would greet this announcement with skepticism is strange to me; I don't understand how you can be so naive regarding this when we have a history of this kind of action laid out behind us.
Welcome. As an OS X convert that followed much the same path you have (just about 5 years ago), I think you'll find that you'll finally stop hopping from OS to OS, from distribution to distribution trying to find something that actually works.
I run Linux servers for a living. I have run a smattering of Linux disros, FreeBSD and OpenBSD as desktop OS's. I currently have an Ubuntu X server at the house, and although I'm not saying you're wrong, I am saying that upgrades have been a relatively minor concern over the years.
Please elucidate; I see much more offered by OS X than Windows to a corporate environment .
Thanks
Okay all you web designers/coders, next time your boss tells you he wants something on the site that uses some crappy activeX component, or uses some of Microsoft's built in "Java"script (because the competitor is using it), you just tell him, "piss off, that's not going to work for all browsers if I use a single code base."
tell him jwhitener said so.
oh and if you can't do it with one code base, y00 suck lam3r... jwhitener says you can ALWAYS do it with one code base. RTFM d00d!!!!!!1111!!!111!!11oneone
more likely:
it's just a lot easier to develop for the mac.
and if you'd ever read any of the dialogues of Plato you might think better of stating it as if it were an attribute exclusively exhibited by peoples of the present and our "declining education standards."
if you're so worried about the standards of education, then do something. otherwise, shut your pie hole.
What's disconcerting is the attempt by you and others to label anyone you don't agree with a "terrorist."
Hussein was a despot, a criminal, a tyrant and a murderer, but he was not a terrorist.
Using the "terrorist" label so blithely--whether it be intentional distortion or just ignorance--is a problem.
With its new version of Windows, Microsoft will be leaving behind the old model of trying to cater to everyone. Making the OS cost more, and making the piracy prevention harder to circumvent, they intend to "take the high road" and sell only to those who can afford their price.
This leaves an ever-easier-to-use-and-install Linux poised to take much of the "low end" market.
The next generation Apple computers will compete on both levels, mostly against Microsoft's offerings, but with the use of Intel chips, Apple has finally broken the last psychological and system related hurdle keeping the masses from switching to a better platform.
If you will, I believe Apple will be competing for the "middle class" of end-user computing. This was a market created by Microsoft. Microsoft itself is leaving it behind believing that the middle class is going to move on from current general computing on to a more appliance-based environment (i.e. game consoles that are created for more general tasks than the consoles of the past, but more task-specific than the PC of yesteryear).
So anyway, that's my take on it. I think this is a great time to be a Cocoa developer. I think this will break-open the confined user base just about the time that XP users will really start looking at their next computer.
The news here is that it wasn't just a vulnerability published, nor a proof of concept, it was a full fledged crack attack against one of the sites that represent the corporation itself. The news here is that it's the same old Microsoft. The news here is that "Trustworthy Computing" is just another marketing buzzword.The news here is that if you can't even manage to secure your own servers, how do you expect the rest of the world to do it?
Microsoft deserves every bit of blame that they get. They want to pretend like security is something that can be applied like a coat of paint, but in the end, incidents like this prove that it's the same old crap rolling out of Redmond.
Agreed. (Choice is good. Documentation is good).
But why should it be necessary to read, man pages for different systems? Why are tools like Cfengine necessary? (Shout out to CFengine devel team, you kick butt!)
see profuse documentation for different systems basically doing the same thing as a necessary evil rather than a inherent benefit.
And believe me, I think monolithic programs are just wrong (see RPM for a good example of how not to create a utility). But we're not talking about different functionality here, were' talking about how often that functionality is invoked.
It doesn't sound like you've never really had to deal with supporting many different daemons in an environment supporting different *nixen.
/etc/init.d/httpd on others. And if that isn't enough, then tell me how nice it is to restart xinetd whenever you're ready to reload a single service configuration.
The real problem with the "UNIX way" is not that there are small, separate services that act as baffles; we all agree that's a good thing. The problem is that each service requires a separate individualized configuration syntax; that each one is controlled by a separate program or super-daemon, each one has its own unique problems and attributes, and each require a different means of control. Next time you have to deal with UCSPI and its simple, but absolutely different zeitgeist, you let me know how easy it is to admin. Next time you need to figure out the runlevel or S and K order of a System V style rc script, or figure out how to to configure rc.conf on FreeBSD vs OpenBSD style rc.conf, tell me how easy and beneficial the UNIX way is. Tell me how nice it is to have apachectl running httpd on some boxes, and
Don't get me wrong; I hate dealing with the absolute idiocy that is Windows Server, but uniform daemon control is nothing like trying to amalgamate them into a single (more vulnerable, slower, more complex and in the end, less-useful) service.
Sure slick, let me know how well that firewall works next time you click on the wrong link with IE and the next root level IE exploit installs an app on your machine that starts opening up connections from behind the firewall.
Well that's funny. I feel exactly the opposite about the whole thing. When I want to get work done, I fire up the Mac. Things just work right. It stays out of my way while I complete my work.
When I want to experiment, play and tinker endlessly with the system (it seems most often to get it working the way it already should, or find another project that works better) I fire up the Linux box. When I want to spend time learning how to get something working or the elements of a project I fire up the Linux box.
Not saying that one is better than the other, it's just interesting that you find you're more productive on a Linux box.