PHP does everything in its power to make safe and secure software development damn near impossible.
There's a saying about whether good craftsmen blame their tools...
It's not PHP's fault that the designers of WordPress are about as competent as I was a year out of college. Everything is global, global functions, global variables, all over the place. If it was possible to use a global variable or a global function instead of something sane like a class, then by god they're going global. WordPress altogether just reeks of amateurish practices. Hell, in order to embed the thing on an existing page you include a file called "wp_blog_header" or something. But, it's not a header, and may not even result in a "header" being printed, it's basically all of WordPress. There's another include file called "wp_settings", which is great except it doesn't contain a single setting, it contains only function definitions. There are exit and die statements all over the include files, so if you pull up the page and it's blank, good luck finding out which condition in which include file got triggered to make the thing bail.
The global nature of everything makes it nearly impossible to embed in various template engines, and I hope your own applications aren't defining global functions with the same generic names that WordPress uses. One of PHP's more insecure options, register_globals, is also implemented in WordPress. No idea why they think they need that option, but if it's disabled in PHP then they go through and define all of those global variables anyway. The entire application looks like it was conceived by a fresh college graduate who recruited his younger brothers to actually build it. It's like the MySpace of CMS applications, the only reason it got big was because it filled a need when the need was there. Not because it's good, but because it was available. If there was ever an application in need of a ground-up, compatibility-smashing re-write, this is it.
Tabs were first in Firefox (through an extension). Opera copied the idea from the extension. Pot. Kettle.
Damn kids these days.
In 1995 the Opera browser version 2 ("MultiTorg Opera") had a "multi-document interface" where you could view several pages at the same time in the same application window. Opera introduced tabs as we know them today in version 4, in June 2000. Several browsers I haven't heard of had tabs before then, starting in 1988 with a browser for browsing news (not a "web" browser). The Mozilla browser introduced tabs in Oct. 2001, and Phoenix (Firefox) a year later in Oct. 2002. Safari got them in 2003, and IE7 got them in 2006. You seem to think that tabs burst on the scene through a Firefox extension some time after Oct. 2002 when Phoenix got extension management. You, my friend, are wrong. Even the Mozilla browser had them before that. Hell, there was even a shell for IE that had tabs in 1997.
I'm proud to say that the LMS I designed and developed doesn't bother with browser sniffing. The Javascript UI toolkit I used does in fact feed different markup (mostly classes and styles) to different browsers, but the end result looks the same and I didn't have to do anything extra. There's no reason an LMS needs to support a subset of browsers, even my Javascript-only AJAX-heavy beast of an LMS works in everything from IE6 to Opera 11. An LMS isn't doing anything mindblowing that only one or two browsers can support. I've got SCORM tracking, AICC tracking, tree structures with drag/drop nodes, panels expanding and collapsing, grids sorting and filtering both locally and remotely, all kinds of "DHTML" popup divs, embedded media players, etc etc. IE6 (slowly) supports all of that, as does every later version of IE, and any other major browser released in the past several years.
It sounds like this "Angel" was designed by sub-par designers and built by sub-par developers.
I don't need a stupid-proof computer (my XP box doesn't even have a virus scanner), but I also don't necessarily want to break license terms. I've pirated enough OSs for my time. Linux would be a possibility if I could run everything I want to run on it, but as it is a Linux machine for me is a hobby machine, not a primary machine. Since I prefer to spec and build my own systems, that leaves Windows as my only option.
I would consider OSX a possibility, but if Apple is going to go out of their way to tell me I can't run their software on a system of my choice, then I'll oblige them. It's not like their software is so over-and-above everything else that I need to do whatever I can to run it. It's not all that, and like Linux it still can't run a lot of programs I use.
Apple doesn't litigate against folks who install OS X on non-Apple hardware, unless you're trying to make a business out of it and sell the resulting machines
Is that an argument to break the terms of the license? Why buy the software at all? Why not pirate Windows 7 instead if you're just going to break the license anyway?
One of the OP's issues was that he did not want to break the license, and I assume that applies to Apple licenses as well. If he didn't care about what the license said, then he wouldn't be worried about how much 3 licenses cost.
For me personally, I'm more than happy to abide by Apple's terms. If Apple wants to put a clause in their license which says that I am not allowed to run the software on non-Apple hardware, then I'll go ahead and agree to those terms and not run their software. I'm not going to buy their hardware, so if they don't want me to run their OS on my hardware, then I won't. Another reason why my next computer will be a Win7 machine as well, once I finally move on from XP.
Install Mac OS on 3 computers where the hardware totaled $750? I'm pretty sure there's something in the Mac OS terms that says you are not allowed to install it on any hardware that costs under $1000 per machine. There's definitely a clause in section 2 there that says you're only allowed to install it on Apple-branded hardware.
So which is it, are you suggesting he break the terms of the license, or increase the budget by 4 times?
Whatever the cause, something about JavaScript really gets idiots excited.
I'm pretty sure it's that whole "change a page on the fly" thing that works in all modern browsers. If you want to write a spec using a better scripting language and submit it to the W3C and browser vendors, no one is stopping you. A lot of good things can be done with browser scripting, it's not the developers' fault that everyone standardized on Javascript instead of whatever language you think about at night while you're touching yourself.
Are you saying that Flash is never a good idea to put online? For the past 10-plus years there have been several applications for which Flash is perfectly suited and for which there haven't been any decent alternatives. What are all of the lazy and dumb "developers" supposed to use if Flash is so verboten? If you want an example, use online training courses. Explain how I can create an online training course that takes 12 hours to complete and includes narration, video, non-video animation, other audio, interactions (including drag and drop), and third-party tracking either through a Javascript API (SCORM), or calls to a URL (AICC). Once you have your answer to that, explain how we would have done that using the technology that was available 12 years ago.
Face it, there are plenty of good reasons to use Flash. Like everything else online, advertisers have mostly screwed it up for everyone else.
That being said, Macromedia and Adobe are borderline incompetent with their Flash implementations. Some bugs have been around as long as I've been using it.
I've never tried to configure IIS with a command line. As for at least IIS7, it does actually have a good GUI.
Neither Photoshop nor Soundforge is server software.
The statement "a GUI is not progress" was not qualified as only applying to a specific kind of software. I was trying to make the point that it is in fact possible for a GUI to help people learn how to use software, regardless of what kind of software it is.
but many people don't care about replacing the battery and they'd rather not have some rickety plastic door on it that pops open all the time.
Those are your only choices with Apple? You either get a non-serviceable battery, or a rickety plastic door that pops open all the time? Why can't Apple design a battery cover that stays attached to the phone like every other manufacturer has managed to do? Is that one of those copy-and-paste things, where Apple will eventually come out with a feature long after everyone else has it and then say they do it "better"?
Every phone blocks signal based on how you hold it--I can understand why that isn't obvious to you but this is a technical website so you probably shouldn't hang out here if you don't care to understand basic RF.
If you're going to be technical, then the phone doesn't block anything, your hand does. But, only with the iPhone was your hand actually touching and shorting the antenna.
It is really totally okay if you don't want to buy Apple products and you don't understand anything about technology or engineering
Wow. God help us when Apple's customer base are considered the elite technical/engineering crowd. Apple designs their devices to be extremely simple to use, not technical, designed and marketed for simple people. While generally impressive under the hood, they are marketed with simplicity in mind. I don't know if you heard, but Wozniak doesn't work there any more.
I haven't signed up with Apple to be an iOS developer, so I can't answer most of those questions. From what I understand it, the only legitimate way to distribute your application to the public is through Apple's store and, yes, you need their approval to do so. I believe the Apple development kit includes a simulator to test it on the various platforms, rather than a way to load it to your device directly. That's complete speculation on my part though.
Assuming your prices are the same on every device. Or, you could raise prices appropriately for people on iOS devices and make the same amount for any purchase. Just the people using iOS devices would pay more for the same thing (which apparently is not a problem for them anyway, since they already bought the Apple hardware).
Right. Previously, Apple also outlawed that, where they said the price you offer through the app has to be the same price on your website. They took that part out, so now the theoretical "Apple tax" is actually a real thing. If purchasing something through an iOS app, people should expect to see an extra 30% charge, for Apple's share. It's the price people pay for the convenience of buying things through their iOS apps. Vendors can indicate that the price is lower on their site, but they're not allowed to put a link to their site in the app.
Never seen such a beast. If you find one, let me know.
How about any IDE that includes built-in help for the language you're developing in, or something like Intellisense? Something that basic is a huge help between writing code on a command line and doing it in a development environment. I feel like you're trying to argue that Notepad is the pinnacle of a text editor, because you should already know how to do whatever you're sitting down to do and don't need any reference documentation or other resources.
How about an easy to use interface that makes your highly expressive language easier to use? Why do you think "highly expressive language" and "GUI" (or whatever inherently-biased term you prefer) are mutually-exclusive? Are you trying to argue that the current crop of touch interfaces in iOS and Android are actually a regression in the way that people interact with computers? If you're looking for a saved file, isn't it pretty obvious and intuitive to point at the icon and touch it versus looking up the only syntax for a file descriptor that your software understands, and then verifying the path is correct? Can't we say that it represents progress when we are able to simply point at a file and say "that one", and have the GUI fill in the details?
Here's a question: when you're writing up a design document for the piece of custom software that your customer wants to build, do you stop at the functionality and just promise them a command-line application? What if they ask for a GUI? Are you going to try and argue that your software is so pure that a GUI would only make it worse, or would you take up the challenge to design and produce an interface that is as brilliant as your flawless algorithms?
That's not true. What's true is that GUI designers are so far not very good at creating usable GUIs for complex tasks. It's not the fault of the buttons or dropdowns or tabs, it's the fault of the guy who thought those were the solution (or didn't understand the problem).
In reality just because something is new or shiny doesn't necessarily make it better.
I'm pretty sure that's exactly the point of the story with the new version of OSX removing the GUI tools.
Of course, it's very difficult to argue that a GUI is not an improvement over a command line in terms of making it easy for a person to interface with a computer. A GUI obviously has superior capabilities (in fact, your GUI can even have a command line), but like I said above, the GUI is only as good as the designer. There's nothing inherently bad about GUIs, only GUI designers.
For any automated/scripted functionality or generally complicated task the CLI will be much faster.
Assuming you already know how to do it. One of the things that GUIs help with is context, on one screen you can see all of the relevant options for what you're doing, there can be links to help or documentation about what you're looking at, etc. For a person going into a complicated task with little knowledge about how to complete the task, a CLI is going to be much slower because of the extra research time needed to figure out what your options even are, let alone how to use everything. A well-designed GUI will be able to show you exactly what you have to work with. Like anything else with software, a GUI is only as good as the person who designed it.
Yeah. Also, use an abacus for all of your calculations, sell your car and buy and horse, kill or grow all of your own food, and get rid of your shoes altogether. Progress is overrated.
Slashdot Mirror
PHP does everything in its power to make safe and secure software development damn near impossible.
There's a saying about whether good craftsmen blame their tools...
It's not PHP's fault that the designers of WordPress are about as competent as I was a year out of college. Everything is global, global functions, global variables, all over the place. If it was possible to use a global variable or a global function instead of something sane like a class, then by god they're going global. WordPress altogether just reeks of amateurish practices. Hell, in order to embed the thing on an existing page you include a file called "wp_blog_header" or something. But, it's not a header, and may not even result in a "header" being printed, it's basically all of WordPress. There's another include file called "wp_settings", which is great except it doesn't contain a single setting, it contains only function definitions. There are exit and die statements all over the include files, so if you pull up the page and it's blank, good luck finding out which condition in which include file got triggered to make the thing bail.
The global nature of everything makes it nearly impossible to embed in various template engines, and I hope your own applications aren't defining global functions with the same generic names that WordPress uses. One of PHP's more insecure options, register_globals, is also implemented in WordPress. No idea why they think they need that option, but if it's disabled in PHP then they go through and define all of those global variables anyway. The entire application looks like it was conceived by a fresh college graduate who recruited his younger brothers to actually build it. It's like the MySpace of CMS applications, the only reason it got big was because it filled a need when the need was there. Not because it's good, but because it was available. If there was ever an application in need of a ground-up, compatibility-smashing re-write, this is it.
Tabs were first in Firefox (through an extension). Opera copied the idea from the extension. Pot. Kettle.
Damn kids these days.
In 1995 the Opera browser version 2 ("MultiTorg Opera") had a "multi-document interface" where you could view several pages at the same time in the same application window. Opera introduced tabs as we know them today in version 4, in June 2000. Several browsers I haven't heard of had tabs before then, starting in 1988 with a browser for browsing news (not a "web" browser). The Mozilla browser introduced tabs in Oct. 2001, and Phoenix (Firefox) a year later in Oct. 2002. Safari got them in 2003, and IE7 got them in 2006. You seem to think that tabs burst on the scene through a Firefox extension some time after Oct. 2002 when Phoenix got extension management. You, my friend, are wrong. Even the Mozilla browser had them before that. Hell, there was even a shell for IE that had tabs in 1997.
Here you go, fanboy, educate yourself.
I'm proud to say that the LMS I designed and developed doesn't bother with browser sniffing. The Javascript UI toolkit I used does in fact feed different markup (mostly classes and styles) to different browsers, but the end result looks the same and I didn't have to do anything extra. There's no reason an LMS needs to support a subset of browsers, even my Javascript-only AJAX-heavy beast of an LMS works in everything from IE6 to Opera 11. An LMS isn't doing anything mindblowing that only one or two browsers can support. I've got SCORM tracking, AICC tracking, tree structures with drag/drop nodes, panels expanding and collapsing, grids sorting and filtering both locally and remotely, all kinds of "DHTML" popup divs, embedded media players, etc etc. IE6 (slowly) supports all of that, as does every later version of IE, and any other major browser released in the past several years.
It sounds like this "Angel" was designed by sub-par designers and built by sub-par developers.
I don't need a stupid-proof computer (my XP box doesn't even have a virus scanner), but I also don't necessarily want to break license terms. I've pirated enough OSs for my time. Linux would be a possibility if I could run everything I want to run on it, but as it is a Linux machine for me is a hobby machine, not a primary machine. Since I prefer to spec and build my own systems, that leaves Windows as my only option.
I would consider OSX a possibility, but if Apple is going to go out of their way to tell me I can't run their software on a system of my choice, then I'll oblige them. It's not like their software is so over-and-above everything else that I need to do whatever I can to run it. It's not all that, and like Linux it still can't run a lot of programs I use.
Apple doesn't litigate against folks who install OS X on non-Apple hardware, unless you're trying to make a business out of it and sell the resulting machines
Is that an argument to break the terms of the license? Why buy the software at all? Why not pirate Windows 7 instead if you're just going to break the license anyway?
One of the OP's issues was that he did not want to break the license, and I assume that applies to Apple licenses as well. If he didn't care about what the license said, then he wouldn't be worried about how much 3 licenses cost.
For me personally, I'm more than happy to abide by Apple's terms. If Apple wants to put a clause in their license which says that I am not allowed to run the software on non-Apple hardware, then I'll go ahead and agree to those terms and not run their software. I'm not going to buy their hardware, so if they don't want me to run their OS on my hardware, then I won't. Another reason why my next computer will be a Win7 machine as well, once I finally move on from XP.
Install Mac OS on 3 computers where the hardware totaled $750? I'm pretty sure there's something in the Mac OS terms that says you are not allowed to install it on any hardware that costs under $1000 per machine. There's definitely a clause in section 2 there that says you're only allowed to install it on Apple-branded hardware.
So which is it, are you suggesting he break the terms of the license, or increase the budget by 4 times?
That doesn't really seem like a logical upgrade path for an XP user. It makes a lot more sense for an XP user to move to Win7.
Whatever the cause, something about JavaScript really gets idiots excited.
I'm pretty sure it's that whole "change a page on the fly" thing that works in all modern browsers. If you want to write a spec using a better scripting language and submit it to the W3C and browser vendors, no one is stopping you. A lot of good things can be done with browser scripting, it's not the developers' fault that everyone standardized on Javascript instead of whatever language you think about at night while you're touching yourself.
Are you saying that Flash is never a good idea to put online? For the past 10-plus years there have been several applications for which Flash is perfectly suited and for which there haven't been any decent alternatives. What are all of the lazy and dumb "developers" supposed to use if Flash is so verboten? If you want an example, use online training courses. Explain how I can create an online training course that takes 12 hours to complete and includes narration, video, non-video animation, other audio, interactions (including drag and drop), and third-party tracking either through a Javascript API (SCORM), or calls to a URL (AICC). Once you have your answer to that, explain how we would have done that using the technology that was available 12 years ago.
Face it, there are plenty of good reasons to use Flash. Like everything else online, advertisers have mostly screwed it up for everyone else.
That being said, Macromedia and Adobe are borderline incompetent with their Flash implementations. Some bugs have been around as long as I've been using it.
No, no, it must start with "Google" and end with "Beta". Google Execution Beta
I've never tried to configure IIS with a command line. As for at least IIS7, it does actually have a good GUI.
Neither Photoshop nor Soundforge is server software.
The statement "a GUI is not progress" was not qualified as only applying to a specific kind of software. I was trying to make the point that it is in fact possible for a GUI to help people learn how to use software, regardless of what kind of software it is.
but many people don't care about replacing the battery and they'd rather not have some rickety plastic door on it that pops open all the time.
Those are your only choices with Apple? You either get a non-serviceable battery, or a rickety plastic door that pops open all the time? Why can't Apple design a battery cover that stays attached to the phone like every other manufacturer has managed to do? Is that one of those copy-and-paste things, where Apple will eventually come out with a feature long after everyone else has it and then say they do it "better"?
Every phone blocks signal based on how you hold it--I can understand why that isn't obvious to you but this is a technical website so you probably shouldn't hang out here if you don't care to understand basic RF.
If you're going to be technical, then the phone doesn't block anything, your hand does. But, only with the iPhone was your hand actually touching and shorting the antenna.
It is really totally okay if you don't want to buy Apple products and you don't understand anything about technology or engineering
Wow. God help us when Apple's customer base are considered the elite technical/engineering crowd. Apple designs their devices to be extremely simple to use, not technical, designed and marketed for simple people. While generally impressive under the hood, they are marketed with simplicity in mind. I don't know if you heard, but Wozniak doesn't work there any more.
I haven't signed up with Apple to be an iOS developer, so I can't answer most of those questions. From what I understand it, the only legitimate way to distribute your application to the public is through Apple's store and, yes, you need their approval to do so. I believe the Apple development kit includes a simulator to test it on the various platforms, rather than a way to load it to your device directly. That's complete speculation on my part though.
It worked pretty well when I was trying to figure out how to do things in Photoshop, or Soundforge, or even IIS.
GIMP, on the other hand.. GIMP is a great example of a GUI doing nothing for the software.
Assuming your prices are the same on every device. Or, you could raise prices appropriately for people on iOS devices and make the same amount for any purchase. Just the people using iOS devices would pay more for the same thing (which apparently is not a problem for them anyway, since they already bought the Apple hardware).
Correct, all in-app purchases are taxed. Vendors are not allowed to link to their websites, all purchases inside the app must go through the app.
30% is a great deal unless your customer base is already well-developed, which I think includes Amazon.
When your profit margins are already 30% or less, then giving that 30% to Apple is no longer a "great deal".
Right. Previously, Apple also outlawed that, where they said the price you offer through the app has to be the same price on your website. They took that part out, so now the theoretical "Apple tax" is actually a real thing. If purchasing something through an iOS app, people should expect to see an extra 30% charge, for Apple's share. It's the price people pay for the convenience of buying things through their iOS apps. Vendors can indicate that the price is lower on their site, but they're not allowed to put a link to their site in the app.
Never seen such a beast. If you find one, let me know.
How about any IDE that includes built-in help for the language you're developing in, or something like Intellisense? Something that basic is a huge help between writing code on a command line and doing it in a development environment. I feel like you're trying to argue that Notepad is the pinnacle of a text editor, because you should already know how to do whatever you're sitting down to do and don't need any reference documentation or other resources.
Hmm. You've never had to learn on the job, huh? Everything you've ever needed to know was taught in school?
What about new software?
How about an easy to use interface that makes your highly expressive language easier to use? Why do you think "highly expressive language" and "GUI" (or whatever inherently-biased term you prefer) are mutually-exclusive? Are you trying to argue that the current crop of touch interfaces in iOS and Android are actually a regression in the way that people interact with computers? If you're looking for a saved file, isn't it pretty obvious and intuitive to point at the icon and touch it versus looking up the only syntax for a file descriptor that your software understands, and then verifying the path is correct? Can't we say that it represents progress when we are able to simply point at a file and say "that one", and have the GUI fill in the details?
Here's a question: when you're writing up a design document for the piece of custom software that your customer wants to build, do you stop at the functionality and just promise them a command-line application? What if they ask for a GUI? Are you going to try and argue that your software is so pure that a GUI would only make it worse, or would you take up the challenge to design and produce an interface that is as brilliant as your flawless algorithms?
GUIs are not good at very complex configurations
That's not true. What's true is that GUI designers are so far not very good at creating usable GUIs for complex tasks. It's not the fault of the buttons or dropdowns or tabs, it's the fault of the guy who thought those were the solution (or didn't understand the problem).
In reality just because something is new or shiny doesn't necessarily make it better.
I'm pretty sure that's exactly the point of the story with the new version of OSX removing the GUI tools.
Of course, it's very difficult to argue that a GUI is not an improvement over a command line in terms of making it easy for a person to interface with a computer. A GUI obviously has superior capabilities (in fact, your GUI can even have a command line), but like I said above, the GUI is only as good as the designer. There's nothing inherently bad about GUIs, only GUI designers.
For any automated/scripted functionality or generally complicated task the CLI will be much faster.
Assuming you already know how to do it. One of the things that GUIs help with is context, on one screen you can see all of the relevant options for what you're doing, there can be links to help or documentation about what you're looking at, etc. For a person going into a complicated task with little knowledge about how to complete the task, a CLI is going to be much slower because of the extra research time needed to figure out what your options even are, let alone how to use everything. A well-designed GUI will be able to show you exactly what you have to work with. Like anything else with software, a GUI is only as good as the person who designed it.
Yeah. Also, use an abacus for all of your calculations, sell your car and buy and horse, kill or grow all of your own food, and get rid of your shoes altogether. Progress is overrated.