What about a database engine that locks the records a and b for reading?
What about performing a division operation or some write system call?
What if + is overloaded?
Essentially a property is not much different from a method without arguments, and it should be treated as such,
That's the problem. a = a + b doesn't look like a property. In fact, it might not be, YOU DON'T KNOW JUST BY LOOKING AT IT. It might be that + is overloaded, or that a has a write property that does something funny.
You simply don't know just by looking at it, and I'll bet that if we weren't talking about properties right now, you would NEVER give a statement like a = a + b a second thought.
if Microsoft tells us "don't download executables from untrustworthy sources" Is refering to a statement that is more recent than this feature.
No, it's a general mantra- and used explicitly in regards recently to why people shouldn't use Firefox (because it's unsigned).
Like the parent mentiones, it was something that was needed back then.
No it wasn't. It was bad engineering any way you look at it. WMF files could be part of a container, or be an EXE or a DLL file if they have this kind of executable power-- this isn't a buffer overflow that is merely "fixed" this is yet another untrustworthy file format - that lo- Windows will make every effort to load transparently.
but the fact is that the things just works as designed,
Really!? So you're saying Microsoft wants you to send spam and spread viruses? You're saying Microsoft designed Windows to be so sluggish and unstable?
I find it simply easier to believe that they're mentally incompetent, although your suggestion has a certain appeal.
It's a bug because it's something the user doesn't expect. If the software is designed in a way different than how the user expects it to operate, then yes, that's a bug. That's why people file bugs- it didn't do what they thought it should do.
Sometimes that means it doesn't do what the authors thought it should do either, but sometimes not.
Either way, it's still a bug.
If you really want to blame MS for soemthing, blame them for not keeping track of their old features in an evolving world.
Nope.
This was bad engineering from the start. UNIX would've avoided this problem by simply never listing a WMF-like file viewer in/etc/mailcap, which is an entirely different beast.
UNIX also avoided this problem by not making a draw_image system call- that executes from untrusted-space on purpose.
How is this different from a user in Konqueror or Nautilus or any other file manager double clicking on a data file and the file manager attempting to figure out what it is?
How about that's not UNIX?
KDE and GNOME may be desktop environments that run on UNIX, but they are not UNIX.
Moreover: You have to actually double-click on something. Not just get sent an email or a URL (remote!) or a MSN Messanger message.
The helper-launchers on UNIX should be using/etc/mailcap in that situation that explicitly lists handlers that are (well, supposed to be) safe for "looking at" files coming from an untrustworthy source.
It might be better to say: It's the same as Lynx or Mutt launching a viewer for any other file- except that they honor the MIME type (even if discovered through mime.magic) and/etc/mailcap.
And yet unregistering the DLL is one of the widely recommended steps that's completely successful as long as you don't have some other vulnerability that re-enables it.
As long as you don't have any programs that load images using the GDI interface directly.
That's just one program- one attack vector. There are plenty of others- yes that includes reenabling that DLL, but the execution behavior doesn't exist in that DLL, it exists in GDI32.DLL - something you cannot so easily disable.
This is scary stuff. Really scary.
The real issue here was that MS, in it's infinite stupidity, designed a data file format that inherently contained executable code! That's fucking insane. You load a graphics file and in it it sets a callback function in case of error? How was that not screaming "exploit me now!", even back in the Win 3.x days?
Office documents can have executable macros. Microsoft makes this mistake this a lot.
However, getting something into office is a lot harder than getting something into the GDI system: First of all, people can uninstall Office, but having to uninstall a critical system component in order to be safe,
As every tool, they can be abused, badly. That doesn't mean they're *bad*, no more than an oven is inherently bad just because you can burn babies alive in it.
That's an excellent argument for goto. I expect to see more of it in your code.
Unfortunately, it's a bad argument for side effects- One has to actually put babies into the oven to burn them with it. With properties, it's like putting them in the crib and the oven rips itself out of the wall, goes down the stairs, into the baby's room, and proceeds to burn the baby.
When properties are in use, they can make things happen that make no sense.
Properties could be useful if they were restricted to no function or method calls - i.e. set a flag for later, or to forward the member update into a parent class.
Of course, you could get to the magic mode if you'd put:
IM_DOING_SCARY_SHIT_HERE
right above using them from outside the class- failure to do so and the compiler would warn you that dangerous magic is afoot.
the fact that you can hide side effects is totally cool.
Can this fail?
a = a + b;
Of course it's something that must be respected and not used carelessly, but if it's used right it is a very neat feature.
By right, how exactly do you mean?
Lots of things are said to be "bad in almost every case, but good in at least one, so let's put it in"- I'd like to know exactly when properties should be used correctly.
Every single example I've ever seen either indicates seemingly-random exceptions should be thrown, or unexpected delays should occur.
At least with actual functions/messages/methods/etc these things can actually be understood by mortals:)
So, in other words, it does exactly the same thing Unix does for every single executable file.
No, if it did it exactly the same way UNIX did, then there wouldn't be a problem.
UNIX only looks up magic headers with using the execve() system call, and not with open()- and only if the file is marked +x - and only if it's on a filesystem marked exec.
So in other words, you don't know what you're talking about.
One of the problems here is that Windows' rape victims cannot disable WMF support and continue using Windows: It's part of GDI- a critical system component.
Another problem is that programs that can be convinced to let GDI display an untrustworthy image are all attack vectors.
Another problem is that Microsoft is inconsistant with regards to what opens what- ActiveX and COM are designed to hide which program is actually doing work- and it makes it very difficult for regular users to determine if the file they're downloading from an untrustworthy source can be handled safely by a program.
Yes, that sometimes means file extensions (which are invisible by default), and other times that means magic header handling, and still other times that means a MIME header. All of which seems designed to frustrate the user- since while they don't know exactly what will happen if they start MSN messanger, or visit a web page none of them expect their computer to be eaten by the grues.
It's a bug because it doesn't have the.exe extension- if Microsoft tells us "don't download executables from untrustworthy sources" they mean.exe files- they don't mean.jpg files.
Read the Fucking Back Story: This would be almost 0% issue if any of the following were true:
1. MSIE/SHELLDOC used extensions or mime-types (MSIE) in determining what file format something was [[ This flaw is transparent to users: it can be in almost any file extension ]]
2. MSIE/SHELLDOC had a feature like the mailcap file on UNIX which allows us to only list programs that can operate on untrustworthy files(!)
3. The WMF magic was outside of a critical system component (that could simply be unregistered and removed)
As a result, this is a very serious problem, and by playing Microsoft's tune about how "it's not that big of a deal", you're only making the problem worse.
By the way, someone should (quick!) make some WMF files that use the AbortProc routines to disable printscreen and stuff when they're visible so they can sue MS for DCMA (copy protection circumvention) violations...
What would you prefer - Stroustrup does it all single-handed?
It might be nice if he could.
Stroustrup lives in a fantasy world where the only reason C++ isn't as fast as C, or produce as small of assembly as C is because of the compilers- which he conveniently disavowes responsibility for.
Compare to Objective-C: You'll note that these new C++ "concepts" feature are extremely similar to Objective-C's "protocols"- only not only can a moderate programmer produce a fast Objective-C compiler, they'll know exactly where it can be slowed down and why.
It'll also already be able to do these things in C++ that are so new and innovative.
Meanwhile, some C++ compilers can make "cout << 1 << endl;" slow- and others only do it when the programmer tries to make their own "cout" like device.
If the libraries were lumped in with the core language, C++ would be a much less flexible and less appropriate tool for those kind [systems programming] of tasks.
No, it's if those libraries imposed greater responsibilities on the runtime. As it stands, the C++ runtime already has an awful lot to do- albeit less than Java or Objective-C.
Worse still: The C++ runtime isn't a peer to your own code as it is in Objective-C: With Objective-C you can interact with the runtime as if it were regular library calls.
C has a mountain of library code available, and the functionality of that library code drives new extensions to the C core language (TLS extensions, for example)
But then, C has almost zero runtime (and if you reject certain extensions: it actually has no runtime), and that's what makes it suitable for systems programming.
I don't think C++ is now, or ever was (or with the way these "extensions" keep showing up) - or ever will be suitable for Systems Programming.
Because the C++ programmer infrequently can understand what his runtime is doing- and is not encouraged to know the interface by which C++ does it's magic (because nobody knows- they're still trying to figure out how to make some C++ magic work in a way that isn't slow)- a C++ systems programmer needs a C++ runtime. Nobody has one in systems-space, so the C++ programmer (which isn't a programmer of C++) needs to write it.
The inventor of C++ can't even do this, but any moderate programmer could do this for Objective-C.
It is not at all obscene when you consider that a site like microsoft.com has over 100million visitors each day, and serves up mostly asp.net applications. Even if each instance of an ASP app used up 200KB of memory, then 20,000 users would max out the 4GB system.
It most certainly is obscene. That's why nobody can figure out how to support 100 million visits a day on Windows except Microsoft.
Somespecialsomeones have discovered that using UNIX systems, they can use less than 16k per concurrent visitor, or about 320 megs of core to support that 20,000 concurrent users.
Our system is zero non-maintenance downtime in the 6 months since I was hired and installed it. Total uptime is above 99.95%.
That's the problem. AS/400 systems frequently have zero downtime over the course of over 15 years.
I don't quite understand why everyone always comes down so hard on Windows
Because Microsoft believes that 99.5% uptime over the course of a year, or almost a full month of downtime over the course of 15 years is acceptable.
In practice, that 99.5% uptime is only maintained with the efforts of continuous IT monitoring and maintenance- something that isn't needed when you buy a proper server.
Linux confuses the hell out of me.
Its your other inadequecies that are the problem right now: While Linux can run on many pieces of "Big Iron" (where Windows doesn't run at all, by the way), most people still use things like CMS on these things.
I realize you might not know what that is, but shit, my four year old can use Linux very well, and can't manage to use CMS at all, so maybe you're trying to say that you are confused by something as simple as Linux?
I don't really understand.
In my opinion, why use something that is stuck in the past?
Because it works?
You see, a large number of people actually value getting work done, and do not care at all whether or not it's the latest and greatest.
That's part of the appeal of businesses that switch to Linux- it's based on a technology that first matured in the late 1960's or early 1970's (depending on who you ask). That technology is extremely well understood and as a result, it's very easy for a company to deliver a platform based on it that "just works".
Meanwhile, the Windows platform is a moving target- it's changed fundementally no less frequently than every 5 years, and many can argue significant changes even more frequently.
The result? It's harder to deliver a platform that a solution-provider can guarantee a certain level of productivity with: Consider this: If we have AS/400 systems or UNIX systems that can be up for 15 years under heavy load, and along comes something else that says they can stay up for "most of a year" under "some load" with "constant maintenece" - how is it anyone is expected to take them seriously?
GUI is so, like, 1995.
This makes no sense.
Are you saying that everything's been graphical since 1995? Or anything's been graphical since 1995? Or perhaps that computers have been usefully graphical since 1995?
I don't understand.
Are you suggesting AS/400's aren't graphical? That Linux isn't? That UNIX isn't? That CMS isn't?
I don't understand.
Are you suggesting that a graphical system leads to greater productivity? Greater performance? Greater stability? Greater uptime?
I don't understand.
I have no idea what this has to do with anything else in your thread.
I have no idea why you think that other people make decisions on their business, their platform, and their hardware, based on your own inadequecies.
I have no idea what you could be possibly thinking, and am beginning to suspect you don't.
Huh? Prove that it costs more money to spend more time designing something that scales to different resolutions? Cuz testing and re-design is free, right?
Wrong. Using no tags beyond <p>, <h1> will give you a page that scales to different resolutions.
I think you mean "it takes time to develop the way I've been developing and making sites that look as good as I have been making them" -- well no kidding. You clearly think designing sites for multiple resolutions takes multiple processes.
But here's the thing: It doesn't have to!
There are plenty of sites out there that describe- often using very simple terms- how to make your pages look good without distorting content, and without making the content difficult to reach!
It takes more time and money than doing nothing- surely- but understand that if you compromise your content, people won't come back because they won't have anything there.
Okay your boss and some random guy on the street buys stuff via a Treo. I'm not sure if that is a compelling reason to cater to those two people.
I didn't say it was, and I apologize if you didn't understand that.
What I am saying is that everyone is different and has different needs and wants and etc.
When they might be interested in is what your entity has to offer- whether this is a company or something else- if they can't find it on your webiste, it doesn't matter at all how good it looks.
There is no argument to that: You either care if people can get stuff off your website or you don't.
What websites has your boss avoided due to non-Treo support? Where does he go instead? Would like some specific examples.
And what possible good would that do?
You're still under the impression that it has something to do with how they look!
It doesn't!
It has to do with whether he (or anyone else) can get information out of them.
I see what you're trying to get at, that information is key, but trust in that information has a lot to do with the layout of the site.
I don't think you do understand what I'm trying to get at.
Something made for humans has to be asthetically pleasing.
So what?
I'm not suggesting the contrary, and if you got that impression, I apologize but I do not believe the contrary for a minute!
What I'm saying is that just because it is by someone or anyones' measure aesthetically pleasing does NOT mean that it is useful, good, or trustworthy.
I do not believe for a minute that a web site can't "look good" and yet still be useful and functional and good and trustworthy.
I do however, believe that most websites aren't useful, functional, good, or trustworthy- regardless of whether they look good or not.
I do also believe that improving the aesthetics of a site is easier than improving it's functionality, and as a result, I do encourage a focus on its functionality.
If the "web master" has to make a decision as to whether it will look good or work well, they should be encouraged to pick another developer.
You're splitting hairs, here. Safari is factored into an App and a couple of frameworks. The bulk of Safari is webcore.
I completely disagree:
The fact that Objective-C is a strict superset of C, Objective-C++ is a strict superset of C++, and C++ is _NOT_ a strict superset of C is extremely important.
The fact that Apple could leverage work written in C++ by the KHTML people is extremely important.
The fact that so much functionality can be passed through WebCore via Objective-C in so little code is also extremely important.
I think that's interesting that you bring up the "bulk" of anything- the fact that khtml, written in C++ is so gigantic while offering so little in the way of a functional interface whereas the WebKit/WebCore written in Objective-C and Objective-C++ are able to offer so rhobust and flexible an interface is extremely important.
Finally, Safari's actual source size is unknown. What is known is the amount of code necessary to make a minimal web browser using WebKit versus KHTML. What is also known is the amount of code necessary to make a web browser as capable as Safari using WebKit versus the size of the web-browser parts of Konquerer as in KHTML.
These hairs you refer to are extremely important and paint the absolute best light on Objective-C: That it helps people do hard things easier.
Install GNUstep on your system, along with ProjectBuilder.app and Gorm.app - then take a look at the many example programs- the calculator and text editors are really excellent and terse examples.
Look at the Objective-C FAQ for comp.lang.objective-c
If you have any C-background, Objective-C will be a snap- you just see lines like this:
[foo bar:z baz:m];
and mentally translate them into something like:
SEL bar_baz = sel_get_any_uid("bar:baz:");
IMP foo_bar_baz = objc_msg_lookup(foo, bar_baz);
foo_bar_baz(z, m);
and remember: sel_get_any_uid() and objc_msg_lookup() are extremely fast and often cached. Of course the Objective-C code (with the square-brackets) is easier to read.
Eventually, you stop translating them and just start reading as "send the bar:baz: message to foo with the bar=z and baz=m arguments" and things become easier.
Did you know that there are more C/C++ and Java Developers than there will ever be Objective-C developers?
I think you use the term "developer" a lot more loosely than I do.
For example, Objective-C is a strict superset of C, and thusly anyone who writes C can and is writing Objective-C. In contrast, C++ isn't a strict superset of C, and it's barely C-like in anything but it's name. As a result, only people writing C++ are writing C++.
Worse: QT isn't even written in C++, but an incompatible superset of C++ that I believe is called "moc" - a translator that brings some of the features of Objective-C to C++ programmers.
Furthermore, most of these people writing C++ and Java aren't what I would call developers.
Learning an obscure programming language is not everyone's idea of fun.
Fortunately it's enough peoples' idea of fun: Without you wouldn't have C++ or Java which both started as seriously obscure.
You promote QT and yet the QT language (moc) has even less developers than Objective-C?
Plus not every one wants to stick to single platform. For such professionals, Qt is the best bet.
You are confusing the term platform: QT is a single platform that happens to run on more than one operating system. *step is also a single platform that happens to run on more than one operating system- and in fact, runs on more systems than QT.
It's also written using a language (C or Objective-C) that is better understood than moc.
Use Qt and STFU. Qt is available on all major platforms natively and that includes MAC.
Use GNUStep and STFU. GNUstep is available on all the major platforms natively, and MAC users can use Cocoa.
It makes C++ development completely pain free
Nothing can make C++ development pain free. Not with four casts, operator and function overloading, and templates.
By the way, Qt isn't written in C++, and doesn't make it easy to write in C++- it makes you write in "moc" C++, which adds one of the best features in Objective-C to C++, while still managing to make it ugly and difficult to use.
With GNUStep, you'll be writing Objective-C, which is actually very pleseant to use. There's a reason you never hear about C++ users that love C++: They talk about how "it's good for some tasks", or "they use python when not using C++" but there are tonnes of people who write the code you write in python in Objective-C, and do it faster and cleaner than you ever could.
Yes, GNUStep runs on Windows, and with Yellow box (Dharma), you have yet another choice in what kind of cross-platform compatability you like.
What else do you want
How about a complete and coherant environment that works? That's why we want Objective-C. I highly recommend everyone try it. You'll find that there's a good reason people get more done faster with it, and there's an even better reason why people actually love it.
So how about you STFU before you start telling everyone about all the other things they could use as long as they don't use Objective-C, and just realize that they use Objective-C because it is the best.
Actually, Objective-C is actually Object-Oriented, which generally means that the object itself knows what its doing, while C++ is Class oriented, which means that the class knows what its doing. Thanks to data encapsulation, C++ is really just an extremely complicated way to have a "this" variable passed to all the functions.
That's not true.
Overloading operators and "methods" and templates, and the four casts, and the complete lack of coding standards are all available in C++, and not with Objective-C. Whether this is a good thing or not is still in debate....
Actually, Safari is largely written in C++ (the KDE rendering engine)
Actually, Safari IS written in Objective-C. So is WebKit. WebCore on the other hand is written in Objective-C++ (KWQ), and yes, includes a significant amount of C++ code (KHTML).
I think this adequately demonstrates the flexibility of Objective-C: to be able to interoperate with C and C++ code on their terms, while C++ code can only interact with Objective-C the same way C can.
This coming from a guy who's site is "an all text experience"...that satisfies more 40,000 unique visitors a month. And I don't even advertise, update, or do anything at all with it.
Oh, I just provide content that apparently people are interested in.
Between you and the customer, decide what's best for you.
It's clear he cannot. He's asking slashdot about video resolution when he really means monitor size. Neither of these things have anything to do with gaining visitors- they only cost a site visitors.
If you don't care if you shut out the 8% of the viewers who are at a low rez
Sure. Why not all the nonMSIE users too- that's 13% right there! How about robots! They don't buy anything so let's stop robots too!
The problem with this argument is that you're deliberately telling 10-20% of people who have good communication skills, have demonstrated that they have some intelligence, and have already exhibited interest in what you're offerring, and telling them to go away.
And that's why nobody visits your website.
YOU are the designer
So go design. Quit making TV advertisements and brochures and actually learn how to make web pages.
Is it possible to make a web site scale perfectly across all display sizes and browsers? Yes (well, maybe). Is it cost effective? Hell no.
Prove it.
I do web design for a couple small artsy type sites, and their biggest criteria is style
I live in the real world. Companies put up websites to make money. Sometimes they do this by selling things, sometimes they do this by providing information so that they don't have to field phone calls.
They could care less if your employer's treo can display it properly
And that's why they don't make any money because my employer buys things from his Treo. So do real people in the real world. Pixel perfect in MSIE doesn't make customers or happy web browsers.
A big part of distinguishing yourself from the competition is how professional your site appears to be.
Completely agreed. So who's site is more professional? google? Or any of these artsy-fartsy web sites you keep making?
Which ones make more money?
Don't get me wrong, I think there's a lot to be said for not designing for a specific case, but trying to design for ALL cases is simply not feasible.
I think it is feasible, but you put a condition on it that you don't state: You want to make TV advertisements or postcards and the web is not TV advertisements and postcards. People- real regular people don't like having their TV and brochure experiences transferred to the web.
Big surprise: people who make websites for 80% of "the web browsers out there" tend to make it for 0% of the people interested in what they can offer.
I just find the posts that say "You should use standards only!
Read my post again. I didn't say a single things about standards.
I said that making web pages for human beings was good. I said that making web pages for art critics: not so much.
Art critics don't buy anything, and aren't interested in buying anything. They don't want anything from your website except to point out how you could've used a few more rounded corners.
Human beings- that is, real web browsers, are interested in something, and I can guarantee it's not what your site looks like.
In reality you have to go through enormous contortions to make ugly sites- and infact, lots of "web designers" do this every day.
I'm wondering what a webbuilder should do if s/he were making a persentational website.
Show me one that anyone wants to go to.
In general, presentations are better made for programs that are good at viewing them. Some people really like using html viewers as presentation viewers.
These people should not kid themselves for a moment into thinking that they're making a web site.
I have made applications that happen to live inside Safari or Mozilla or even MSIE.
I have even delivered presentations using Opera's wonderful fullscreen mode.
But nobody thinks for a minute that these are web pages, nor should it matter. The web is about content, there simply isn't any such thing as a "presentational" site. It doesn't exist.
If the content of a website is graphical, what's the solution to make it one-size-fit-all?
There isn't one.
You want a fancy picture book? Consider making a PDF and encouraging people to print it. They might enjoy the thought even if they only use Acrobat to view it.
Trying to satisfy everyone on the web with such things is (clearly) difficult, so why not take it out of the equation all together
do keep in mind that decisions such as these are often not made by developers but by bosses and marketing departments.
Your employer is mentally retarded. That's why he hired you.
Tell him that if he wants to design web pages, he can learn how to do it, otherwise he should leave it to professionals.
developers know better
Yeah, see, I've noticed that they don't. read the rest of the arguments in this thread: They don't know better. They're just as mind-blowingly stupid as your employer, and like your employer, would love to have everyone continue doing things their way just so that they can say it's how it's done.
Sites should be usable and viewable with any resolution with any web browser.
We do not want an art exhibit, we want a web page. With stuff on it. Knowledge.
I for example, frequently browse at 320 pixels across. I don't visit sites that don't work at that resolution. My employer uses his Treo frequently and has even worse to say on that.
But my resolution? Well above 2000 pixels across.
See, just because some web browsers (the users, not the programs) browse at full screen doesn't mean everyone does it.
Web pages are not canvases- they do not have a size, and by artificially attempting to create one, you are doing the web a disservice.
On the other hand, by treating them as such, chances are you have so little to say that it isn't useful at all in which case myself, and other web browsers simply won't visit your site.
You will of course think it has something to do with the modernness of your design and make it even less usable.
Slashdot Mirror
I don't see how
What about a database engine that locks the records a and b for reading?
What about performing a division operation or some write system call?
What if + is overloaded?
Essentially a property is not much different from a method without arguments, and it should be treated as such,
That's the problem. a = a + b doesn't look like a property. In fact, it might not be, YOU DON'T KNOW JUST BY LOOKING AT IT. It might be that + is overloaded, or that a has a write property that does something funny.
You simply don't know just by looking at it, and I'll bet that if we weren't talking about properties right now, you would NEVER give a statement like a = a + b a second thought.
if Microsoft tells us "don't download executables from untrustworthy sources" Is refering to a statement that is more recent than this feature.
/etc/mailcap, which is an entirely different beast.
No, it's a general mantra- and used explicitly in regards recently to why people shouldn't use Firefox (because it's unsigned).
Like the parent mentiones, it was something that was needed back then.
No it wasn't. It was bad engineering any way you look at it. WMF files could be part of a container, or be an EXE or a DLL file if they have this kind of executable power-- this isn't a buffer overflow that is merely "fixed" this is yet another untrustworthy file format - that lo- Windows will make every effort to load transparently.
but the fact is that the things just works as designed,
Really!? So you're saying Microsoft wants you to send spam and spread viruses? You're saying Microsoft designed Windows to be so sluggish and unstable?
I find it simply easier to believe that they're mentally incompetent, although your suggestion has a certain appeal.
It's a bug because it's something the user doesn't expect. If the software is designed in a way different than how the user expects it to operate, then yes, that's a bug. That's why people file bugs- it didn't do what they thought it should do.
Sometimes that means it doesn't do what the authors thought it should do either, but sometimes not.
Either way, it's still a bug.
If you really want to blame MS for soemthing, blame them for not keeping track of their old features in an evolving world.
Nope.
This was bad engineering from the start. UNIX would've avoided this problem by simply never listing a WMF-like file viewer in
UNIX also avoided this problem by not making a draw_image system call- that executes from untrusted-space on purpose.
No.
How is this different from a user in Konqueror or Nautilus or any other file manager double clicking on a data file and the file manager attempting to figure out what it is?
/etc/mailcap in that situation that explicitly lists handlers that are (well, supposed to be) safe for "looking at" files coming from an untrustworthy source.
/etc/mailcap.
How about that's not UNIX?
KDE and GNOME may be desktop environments that run on UNIX, but they are not UNIX.
Moreover: You have to actually double-click on something. Not just get sent an email or a URL (remote!) or a MSN Messanger message.
The helper-launchers on UNIX should be using
It might be better to say: It's the same as Lynx or Mutt launching a viewer for any other file- except that they honor the MIME type (even if discovered through mime.magic) and
And yet unregistering the DLL is one of the widely recommended steps that's completely successful as long as you don't have some other vulnerability that re-enables it.
As long as you don't have any programs that load images using the GDI interface directly.
That's just one program- one attack vector. There are plenty of others- yes that includes reenabling that DLL, but the execution behavior doesn't exist in that DLL, it exists in GDI32.DLL - something you cannot so easily disable.
This is scary stuff. Really scary.
The real issue here was that MS, in it's infinite stupidity, designed a data file format that inherently contained executable code! That's fucking insane. You load a graphics file and in it it sets a callback function in case of error? How was that not screaming "exploit me now!", even back in the Win 3.x days?
Office documents can have executable macros. Microsoft makes this mistake this a lot.
However, getting something into office is a lot harder than getting something into the GDI system: First of all, people can uninstall Office, but having to uninstall a critical system component in order to be safe,
well... maybe that's the point...
As every tool, they can be abused, badly. That doesn't mean they're *bad*, no more than an oven is inherently bad just because you can burn babies alive in it.
That's an excellent argument for goto. I expect to see more of it in your code.
Unfortunately, it's a bad argument for side effects- One has to actually put babies into the oven to burn them with it. With properties, it's like putting them in the crib and the oven rips itself out of the wall, goes down the stairs, into the baby's room, and proceeds to burn the baby.
When properties are in use, they can make things happen that make no sense.
Properties could be useful if they were restricted to no function or method calls - i.e. set a flag for later, or to forward the member update into a parent class.
Of course, you could get to the magic mode if you'd put:
IM_DOING_SCARY_SHIT_HERE
right above using them from outside the class- failure to do so and the compiler would warn you that dangerous magic is afoot.
the fact that you can hide side effects is totally cool.
:)
Can this fail?
a = a + b;
Of course it's something that must be respected and not used carelessly, but if it's used right it is a very neat feature.
By right, how exactly do you mean?
Lots of things are said to be "bad in almost every case, but good in at least one, so let's put it in"- I'd like to know exactly when properties should be used correctly.
Every single example I've ever seen either indicates seemingly-random exceptions should be thrown, or unexpected delays should occur.
At least with actual functions/messages/methods/etc these things can actually be understood by mortals
So is operator overloading.
Operator overloading is stupid as well.
I want it in a clean language like C++!
You really think this:
template>Random_access_iterator Ran, Predicate Cmp< where Can_call_with<Cmp,C::value_type> void sort(Ran first, Ran last, Cmp less);
is clean!?
Cleaner than this?
@protocol MySort
-(int)compareTo:(id)x;
@end
NSArray *sort(NSArray *in, id f);
(equivilent Objective-C code)
So, in other words, it does exactly the same thing Unix does for every single executable file.
No, if it did it exactly the same way UNIX did, then there wouldn't be a problem.
UNIX only looks up magic headers with using the execve() system call, and not with open()- and only if the file is marked +x - and only if it's on a filesystem marked exec.
So in other words, you don't know what you're talking about.
One of the problems here is that Windows' rape victims cannot disable WMF support and continue using Windows: It's part of GDI- a critical system component.
Another problem is that programs that can be convinced to let GDI display an untrustworthy image are all attack vectors.
Another problem is that Microsoft is inconsistant with regards to what opens what- ActiveX and COM are designed to hide which program is actually doing work- and it makes it very difficult for regular users to determine if the file they're downloading from an untrustworthy source can be handled safely by a program.
Yes, that sometimes means file extensions (which are invisible by default), and other times that means magic header handling, and still other times that means a MIME header. All of which seems designed to frustrate the user- since while they don't know exactly what will happen if they start MSN messanger, or visit a web page none of them expect their computer to be eaten by the grues.
It's a bug because it doesn't have the .exe extension- if Microsoft tells us "don't download executables from untrustworthy sources" they mean .exe files- they don't mean .jpg files.
Read the Fucking Back Story: This would be almost 0% issue if any of the following were true:
1. MSIE/SHELLDOC used extensions or mime-types (MSIE) in determining what file format something was [[ This flaw is transparent to users: it can be in almost any file extension ]]
2. MSIE/SHELLDOC had a feature like the mailcap file on UNIX which allows us to only list programs that can operate on untrustworthy files(!)
3. The WMF magic was outside of a critical system component (that could simply be unregistered and removed)
As a result, this is a very serious problem, and by playing Microsoft's tune about how "it's not that big of a deal", you're only making the problem worse.
By the way, someone should (quick!) make some WMF files that use the AbortProc routines to disable printscreen and stuff when they're visible so they can sue MS for DCMA (copy protection circumvention) violations...
Properties are worthless and good at hiding side effects.
Interface/Implementation split is good- Objective-C does that. Maybe you should consider that.
Bit arrays are dead simple in C, and boost::'s version is complicated as hell. What exactly did you have in mind?
What would you prefer - Stroustrup does it all single-handed?
It might be nice if he could.
Stroustrup lives in a fantasy world where the only reason C++ isn't as fast as C, or produce as small of assembly as C is because of the compilers- which he conveniently disavowes responsibility for.
Compare to Objective-C: You'll note that these new C++ "concepts" feature are extremely similar to Objective-C's "protocols"- only not only can a moderate programmer produce a fast Objective-C compiler, they'll know exactly where it can be slowed down and why.
It'll also already be able to do these things in C++ that are so new and innovative.
Meanwhile, some C++ compilers can make "cout << 1 << endl;" slow- and others only do it when the programmer tries to make their own "cout" like device.
If the libraries were lumped in with the core language, C++ would be a much less flexible and less appropriate tool for those kind [systems programming] of tasks.
No, it's if those libraries imposed greater responsibilities on the runtime. As it stands, the C++ runtime already has an awful lot to do- albeit less than Java or Objective-C.
Worse still: The C++ runtime isn't a peer to your own code as it is in Objective-C: With Objective-C you can interact with the runtime as if it were regular library calls.
C has a mountain of library code available, and the functionality of that library code drives new extensions to the C core language (TLS extensions, for example)
But then, C has almost zero runtime (and if you reject certain extensions: it actually has no runtime), and that's what makes it suitable for systems programming.
I don't think C++ is now, or ever was (or with the way these "extensions" keep showing up) - or ever will be suitable for Systems Programming.
Because the C++ programmer infrequently can understand what his runtime is doing- and is not encouraged to know the interface by which C++ does it's magic (because nobody knows- they're still trying to figure out how to make some C++ magic work in a way that isn't slow)- a C++ systems programmer needs a C++ runtime. Nobody has one in systems-space, so the C++ programmer (which isn't a programmer of C++) needs to write it.
The inventor of C++ can't even do this, but any moderate programmer could do this for Objective-C.
It most certainly is obscene. That's why nobody can figure out how to support 100 million visits a day on Windows except Microsoft.
Some special someones have discovered that using UNIX systems, they can use less than 16k per concurrent visitor, or about 320 megs of core to support that 20,000 concurrent users.
Our system is zero non-maintenance downtime in the 6 months since I was hired and installed it. Total uptime is above 99.95%.
That's the problem. AS/400 systems frequently have zero downtime over the course of over 15 years.
I don't quite understand why everyone always comes down so hard on Windows
Because Microsoft believes that 99.5% uptime over the course of a year, or almost a full month of downtime over the course of 15 years is acceptable.
In practice, that 99.5% uptime is only maintained with the efforts of continuous IT monitoring and maintenance- something that isn't needed when you buy a proper server.
Linux confuses the hell out of me.
Its your other inadequecies that are the problem right now: While Linux can run on many pieces of "Big Iron" (where Windows doesn't run at all, by the way), most people still use things like CMS on these things.
I realize you might not know what that is, but shit, my four year old can use Linux very well, and can't manage to use CMS at all, so maybe you're trying to say that you are confused by something as simple as Linux?
I don't really understand.
In my opinion, why use something that is stuck in the past?
Because it works?
You see, a large number of people actually value getting work done, and do not care at all whether or not it's the latest and greatest.
That's part of the appeal of businesses that switch to Linux- it's based on a technology that first matured in the late 1960's or early 1970's (depending on who you ask). That technology is extremely well understood and as a result, it's very easy for a company to deliver a platform based on it that "just works".
Meanwhile, the Windows platform is a moving target- it's changed fundementally no less frequently than every 5 years, and many can argue significant changes even more frequently.
The result? It's harder to deliver a platform that a solution-provider can guarantee a certain level of productivity with: Consider this: If we have AS/400 systems or UNIX systems that can be up for 15 years under heavy load, and along comes something else that says they can stay up for "most of a year" under "some load" with "constant maintenece" - how is it anyone is expected to take them seriously?
GUI is so, like, 1995.
This makes no sense.
Are you saying that everything's been graphical since 1995? Or anything's been graphical since 1995? Or perhaps that computers have been usefully graphical since 1995?
I don't understand.
Are you suggesting AS/400's aren't graphical? That Linux isn't? That UNIX isn't? That CMS isn't?
I don't understand.
Are you suggesting that a graphical system leads to greater productivity? Greater performance? Greater stability? Greater uptime?
I don't understand.
I have no idea what this has to do with anything else in your thread.
I have no idea why you think that other people make decisions on their business, their platform, and their hardware, based on your own inadequecies.
I have no idea what you could be possibly thinking, and am beginning to suspect you don't.
Huh? Prove that it costs more money to spend more time designing something that scales to different resolutions? Cuz testing and re-design is free, right?
Wrong. Using no tags beyond <p>, <h1> will give you a page that scales to different resolutions.
I think you mean "it takes time to develop the way I've been developing and making sites that look as good as I have been making them" -- well no kidding. You clearly think designing sites for multiple resolutions takes multiple processes.
But here's the thing: It doesn't have to!
There are plenty of sites out there that describe- often using very simple terms- how to make your pages look good without distorting content, and without making the content difficult to reach!
It takes more time and money than doing nothing- surely- but understand that if you compromise your content, people won't come back because they won't have anything there.
Okay your boss and some random guy on the street buys stuff via a Treo. I'm not sure if that is a compelling reason to cater to those two people.
I didn't say it was, and I apologize if you didn't understand that.
What I am saying is that everyone is different and has different needs and wants and etc.
When they might be interested in is what your entity has to offer- whether this is a company or something else- if they can't find it on your webiste, it doesn't matter at all how good it looks.
There is no argument to that: You either care if people can get stuff off your website or you don't.
What websites has your boss avoided due to non-Treo support? Where does he go instead? Would like some specific examples.
And what possible good would that do?
You're still under the impression that it has something to do with how they look!
It doesn't!
It has to do with whether he (or anyone else) can get information out of them.
I see what you're trying to get at, that information is key, but trust in that information has a lot to do with the layout of the site.
I don't think you do understand what I'm trying to get at.
Something made for humans has to be asthetically pleasing.
So what?
I'm not suggesting the contrary, and if you got that impression, I apologize but I do not believe the contrary for a minute!
What I'm saying is that just because it is by someone or anyones' measure aesthetically pleasing does NOT mean that it is useful, good, or trustworthy.
I do not believe for a minute that a web site can't "look good" and yet still be useful and functional and good and trustworthy.
I do however, believe that most websites aren't useful, functional, good, or trustworthy- regardless of whether they look good or not.
I do also believe that improving the aesthetics of a site is easier than improving it's functionality, and as a result, I do encourage a focus on its functionality.
If the "web master" has to make a decision as to whether it will look good or work well, they should be encouraged to pick another developer.
You're splitting hairs, here. Safari is factored into an App and a couple of frameworks. The bulk of Safari is webcore.
I completely disagree:
The fact that Objective-C is a strict superset of C, Objective-C++ is a strict superset of C++, and C++ is _NOT_ a strict superset of C is extremely important.
The fact that Apple could leverage work written in C++ by the KHTML people is extremely important.
The fact that so much functionality can be passed through WebCore via Objective-C in so little code is also extremely important.
I think that's interesting that you bring up the "bulk" of anything- the fact that khtml, written in C++ is so gigantic while offering so little in the way of a functional interface whereas the WebKit/WebCore written in Objective-C and Objective-C++ are able to offer so rhobust and flexible an interface is extremely important.
Finally, Safari's actual source size is unknown. What is known is the amount of code necessary to make a minimal web browser using WebKit versus KHTML. What is also known is the amount of code necessary to make a web browser as capable as Safari using WebKit versus the size of the web-browser parts of Konquerer as in KHTML.
These hairs you refer to are extremely important and paint the absolute best light on Objective-C: That it helps people do hard things easier.
You really should just dive into it.
Install GNUstep on your system, along with ProjectBuilder.app and Gorm.app - then take a look at the many example programs- the calculator and text editors are really excellent and terse examples.
Look at the Objective-C FAQ for comp.lang.objective-c
If you have any C-background, Objective-C will be a snap- you just see lines like this:
[foo bar:z baz:m];
and mentally translate them into something like:
SEL bar_baz = sel_get_any_uid("bar:baz:");
IMP foo_bar_baz = objc_msg_lookup(foo, bar_baz);
foo_bar_baz(z, m);
and remember: sel_get_any_uid() and objc_msg_lookup() are extremely fast and often cached. Of course the Objective-C code (with the square-brackets) is easier to read.
Eventually, you stop translating them and just start reading as "send the bar:baz: message to foo with the bar=z and baz=m arguments" and things become easier.
Did you know that there are more C/C++ and Java Developers than there will ever be Objective-C developers?
I think you use the term "developer" a lot more loosely than I do.
For example, Objective-C is a strict superset of C, and thusly anyone who writes C can and is writing Objective-C. In contrast, C++ isn't a strict superset of C, and it's barely C-like in anything but it's name. As a result, only people writing C++ are writing C++.
Worse: QT isn't even written in C++, but an incompatible superset of C++ that I believe is called "moc" - a translator that brings some of the features of Objective-C to C++ programmers.
Furthermore, most of these people writing C++ and Java aren't what I would call developers.
Learning an obscure programming language is not everyone's idea of fun.
Fortunately it's enough peoples' idea of fun: Without you wouldn't have C++ or Java which both started as seriously obscure.
You promote QT and yet the QT language (moc) has even less developers than Objective-C?
Plus not every one wants to stick to single platform. For such professionals, Qt is the best bet.
You are confusing the term platform: QT is a single platform that happens to run on more than one operating system. *step is also a single platform that happens to run on more than one operating system- and in fact, runs on more systems than QT.
It's also written using a language (C or Objective-C) that is better understood than moc.
Use Qt and STFU. Qt is available on all major platforms natively and that includes MAC.
Use GNUStep and STFU. GNUstep is available on all the major platforms natively, and MAC users can use Cocoa.
It makes C++ development completely pain free
Nothing can make C++ development pain free. Not with four casts, operator and function overloading, and templates.
By the way, Qt isn't written in C++, and doesn't make it easy to write in C++- it makes you write in "moc" C++, which adds one of the best features in Objective-C to C++, while still managing to make it ugly and difficult to use.
With GNUStep, you'll be writing Objective-C, which is actually very pleseant to use. There's a reason you never hear about C++ users that love C++: They talk about how "it's good for some tasks", or "they use python when not using C++" but there are tonnes of people who write the code you write in python in Objective-C, and do it faster and cleaner than you ever could.
Yes, GNUStep runs on Windows, and with Yellow box (Dharma), you have yet another choice in what kind of cross-platform compatability you like.
What else do you want
How about a complete and coherant environment that works? That's why we want Objective-C. I highly recommend everyone try it. You'll find that there's a good reason people get more done faster with it, and there's an even better reason why people actually love it.
So how about you STFU before you start telling everyone about all the other things they could use as long as they don't use Objective-C, and just realize that they use Objective-C because it is the best.
Actually, Objective-C is actually Object-Oriented, which generally means that the object itself knows what its doing, while C++ is Class oriented, which means that the class knows what its doing. Thanks to data encapsulation, C++ is really just an extremely complicated way to have a "this" variable passed to all the functions.
That's not true.
Overloading operators and "methods" and templates, and the four casts, and the complete lack of coding standards are all available in C++, and not with Objective-C. Whether this is a good thing or not is still in debate....
Actually, Safari is largely written in C++ (the KDE rendering engine)
Actually, Safari IS written in Objective-C. So is WebKit. WebCore on the other hand is written in Objective-C++ (KWQ), and yes, includes a significant amount of C++ code (KHTML).
I think this adequately demonstrates the flexibility of Objective-C: to be able to interoperate with C and C++ code on their terms, while C++ code can only interact with Objective-C the same way C can.
This coming from a guy who's site is "an all text experience" ...that satisfies more 40,000 unique visitors a month. And I don't even advertise, update, or do anything at all with it.
Oh, I just provide content that apparently people are interested in.
Between you and the customer, decide what's best for you.
It's clear he cannot. He's asking slashdot about video resolution when he really means monitor size. Neither of these things have anything to do with gaining visitors- they only cost a site visitors.
If you don't care if you shut out the 8% of the viewers who are at a low rez
Sure. Why not all the nonMSIE users too- that's 13% right there! How about robots! They don't buy anything so let's stop robots too!
The problem with this argument is that you're deliberately telling 10-20% of people who have good communication skills, have demonstrated that they have some intelligence, and have already exhibited interest in what you're offerring, and telling them to go away.
And that's why nobody visits your website.
YOU are the designer
So go design. Quit making TV advertisements and brochures and actually learn how to make web pages.
Is it possible to make a web site scale perfectly across all display sizes and browsers? Yes (well, maybe). Is it cost effective? Hell no.
Prove it.
I do web design for a couple small artsy type sites, and their biggest criteria is style
I live in the real world. Companies put up websites to make money. Sometimes they do this by selling things, sometimes they do this by providing information so that they don't have to field phone calls.
They could care less if your employer's treo can display it properly
And that's why they don't make any money because my employer buys things from his Treo. So do real people in the real world. Pixel perfect in MSIE doesn't make customers or happy web browsers.
A big part of distinguishing yourself from the competition is how professional your site appears to be.
Completely agreed. So who's site is more professional? google? Or any of these artsy-fartsy web sites you keep making?
Which ones make more money?
Don't get me wrong, I think there's a lot to be said for not designing for a specific case, but trying to design for ALL cases is simply not feasible.
I think it is feasible, but you put a condition on it that you don't state: You want to make TV advertisements or postcards and the web is not TV advertisements and postcards. People- real regular people don't like having their TV and brochure experiences transferred to the web.
Big surprise: people who make websites for 80% of "the web browsers out there" tend to make it for 0% of the people interested in what they can offer.
I just find the posts that say "You should use standards only!
Read my post again. I didn't say a single things about standards.
I said that making web pages for human beings was good.
I said that making web pages for art critics: not so much.
Art critics don't buy anything, and aren't interested in buying anything. They don't want anything from your website except to point out how you could've used a few more rounded corners.
Human beings- that is, real web browsers, are interested in something, and I can guarantee it's not what your site looks like.
In reality you have to go through enormous contortions to make ugly sites- and infact, lots of "web designers" do this every day.
I'm wondering what a webbuilder should do if s/he were making a persentational website.
Show me one that anyone wants to go to.
In general, presentations are better made for programs that are good at viewing them. Some people really like using html viewers as presentation viewers.
These people should not kid themselves for a moment into thinking that they're making a web site.
I have made applications that happen to live inside Safari or Mozilla or even MSIE.
I have even delivered presentations using Opera's wonderful fullscreen mode.
But nobody thinks for a minute that these are web pages, nor should it matter. The web is about content, there simply isn't any such thing as a "presentational" site. It doesn't exist.
If the content of a website is graphical, what's the solution to make it one-size-fit-all?
There isn't one.
You want a fancy picture book? Consider making a PDF and encouraging people to print it. They might enjoy the thought even if they only use Acrobat to view it.
Trying to satisfy everyone on the web with such things is (clearly) difficult, so why not take it out of the equation all together
do keep in mind that decisions such as these are often not made by developers but by bosses and marketing departments.
Your employer is mentally retarded. That's why he hired you.
Tell him that if he wants to design web pages, he can learn how to do it, otherwise he should leave it to professionals.
developers know better
Yeah, see, I've noticed that they don't. read the rest of the arguments in this thread: They don't know better. They're just as mind-blowingly stupid as your employer, and like your employer, would love to have everyone continue doing things their way just so that they can say it's how it's done.
Absolutely amazing...
Stop.
Please.
You are the bane of the web browser.
Sites should be usable and viewable with any resolution with any web browser.
We do not want an art exhibit, we want a web page. With stuff on it. Knowledge.
I for example, frequently browse at 320 pixels across. I don't visit sites that don't work at that resolution. My employer uses his Treo frequently and has even worse to say on that.
But my resolution? Well above 2000 pixels across.
See, just because some web browsers (the users, not the programs) browse at full screen doesn't mean everyone does it.
Web pages are not canvases- they do not have a size, and by artificially attempting to create one, you are doing the web a disservice.
On the other hand, by treating them as such, chances are you have so little to say that it isn't useful at all in which case myself, and other web browsers simply won't visit your site.
You will of course think it has something to do with the modernness of your design and make it even less usable.
The cycle will continue.
And nobody will notice.