To replicate that functionality I would have to backup my entire email pretty much every time I got any, which is completely impractical.
Not really -- backing up from Gmail is as simple as IMAP. If you're talking about new email, that's just as simple -- and there are tools to keep IMAP mailboxes in sync. I imagine Gmail will happily forward incoming email.
Not to mention tedious. Yeah, shell scripts and all that crap
Who said anything about shell scripts? It's pretty much a builtin feature of at least one email client I know of, and there are third-party programs you can get dedicated to the purpose.
And if you've got a program to do it, where's the tedium?
why bother when Gmail does it all for me
Mostly because you have no idea if Gmail is going to be there tomorrow, or if your Internet connection is going to be there tomorrow.
Now, granted, I'm lazy too -- I haven't backed up my work Gmail ever, and it would be pretty inconvenient if that was lost.
Point is, it's not nearly as huge a difference as you're making it out to be. There are real reasons to use Gmail -- labels, fulltext index, nice web interface for when that matters -- yours just seems (ahem) retarded.
Free software could allow you to encrypt your data automatically from the service provider, to migrate yourself and your data to a compatible competitor, or to implement "premium" features yourself. But what's in it for the service provider?
Depends how it's done.
First, consider that many services have made your data somewhat portable, either intentionally or by accident. Gmail has been hot on this thread -- but Gmail supports IMAP. And most of the services being discussed here are web-based, which implies that unless the provider takes explicit measures to prevent it, you could always write a program to scrape their site for the information.
But consider the implications of data portability, by itself.
Done right, it means that you get encryption (if relevant) -- GnuPG is supported in both of the IMAP clients I've used lately -- as well as additional features, by way of Firefox extensions and third-party services -- think Facebook apps. (Or think of any service with an open API -- I just implemented Twitter's, took maybe an hour, from knowing nothing about Twitter.)
The only thing left is "precise control of their software" -- they certainly control the server-side software, but again, Firefox extensions and APIs mean that most things (ads, etc) can be stripped out by third parties.
Now, what's in it for the service provider?
Well, read all of the above -- that's a lot more service you're providing to an end-user. And it's a lot of features you get for free. Take YouTube -- you can embed it in any webpage, any random blog or whatever. It means free publicity for YouTube, and more people posting videos to YouTube, even if they only want them back on their own site.
Take Twitter -- I just added a feature to a project at work which pulls in a user's latest status update from Twitter and shows it on their profile. That's free publicity for Twitter, and quite possibly some new users because of it.
Gmail's popular because it's free. Try monetising it.
Like Google's already done, with the ads, and the premium accounts?
If you charge less, you're eating your desktop division's lunch.
If you don't have a desktop division, that means you're eating your competition's desktop division's lunch.
If you charge more, you're providing less service for more money, because the company ultimately doesn't own its data.
That assumes an identical featureset.
Consider Exchange vs Google Apps.
Exchange is going to have tighter integration with desktop apps like MS Office, and Office has more features and better support for legacy formats. It's also very likely you do backup inhouse, which may be required by your industry, and is probably going to give you more peace of mind. And it's not dependent on any third party, save for Microsoft, and there only for updates -- if Microsoft.com went down, your exchange server would still be up.
Google Apps has tighter integration with Google services, and, generally, better functionality for sharing documents and collaborating online. It has the additional advantage of outsourcing backup (Google can do it) and distribution (none needed), while being available anywhere. And depending on what you need in the way of desktop hardware, it's entirely possible you might not need any desktop software besides a browser -- which gives you many of the maintenance advantages of thin clients.
It is, in other words, apples and oranges. For some, Google Apps is a compelling alternative -- even worth paying for, possibly worth paying more for.
If I had to guess, I would say that a single-threaded filesystem is maybe a step up from Win98, if that. If you remember, reading from a floppy on Win98 would bring the entire system to a crawl, as any disk IO would make the rest of the OS block until it was done.
I didn't use Minix at the time, but I would guess it was like that, at least for the filesystem itself, if not for the entire OS -- in that, if the filesystem was blocked doing disk IO, all attempts to talk to the filesystem would also be blocked.
Minix is a full, Unix-like operating system, whereas Linux is a kernel.
Really? I was under the impression that it used GNU tools to fill in the rest of the system, just like Linux.
In fact, it's also credited to just one person -- Andy Tanenbaum.
I could be wrong, though. If so, that would be another point against Minix -- Linux was able to leverage existing, portable tools. Minix was not yet POSIX-compliant at the time, making portability between Minix and anything else a bit harder.
When I went to interview for my current job, I was wearing some nice pants, a buttoned shirt, probably even a tie. I was freshly showered and shaved.
Now, I haven't shaved in probably a year of working here, and I've never worn those pants or that shirt since. (I'm not even sure I fit into the pants anymore.) I'm a decided Unix longhair, one of my coworkers wears sandals...
The point is, you dress up for the interview as a show of respect, and to make a first impression. It has nothing to do with the job, and everything to do with the interview. (That said -- check the dress code, afterwards...)
- Have I thought about real answers to the typical questions and not just canned responses
I think a better question would be, can you do this on the fly?
Really, they're asking those questions to get to know you better -- it's a way of opening a conversation, and trying to get a feel for you as a person. If you've actually thought out a response, and you know it by heart, it becomes a canned response, and you can tell.
- During tech interviews, can I provide real world examples or am I spitting out algorithms and examples from text books?
Not really particularly important, I would think.
I've been on the other side of the (metaphorical) interviewing table a few times, and it's much better to provide an actual problem to solve than to ask for an example. You can start with FizzBuzz if you like, but give them something actually challenging, to show you how they think on the fly.
One I especially liked was a small snippet of JavaScript with a bug in it, which you had to understand closures in order to get. Less than ten lines -- find the bug. Obvious if you had any Javascript experience or skill, impossible if you'd just read a book.
And as a candidate, pay attention to when you fail a question. They won't tell you, but you should know. And when you know, go research that and be ready to ace it on your next interview.
It's a bad thing for several reasons, which have already been laid out elsewhere.
Using 4.2.2.2 means I abuse whatever poor bastard owns that IP now, and using my own DNS server means I'm paying for the same service twice -- once for my ISP to run their servers, and once to run my own.
And you can get the same functionality, for web browsing, in your browser. There's pretty much no app other than web browsing which would benefit to this, and several which would be hurt by this.
That unfortunately doesn't defend against GET requests. This means that if your script allows anything vulnerable to be requested via GET, then the 'image' attack vector is not protected.
Correct. That's why I mentioned "stupid things with GET requests", and why the page I linked to also links to some documentation about idempotence.
Building an app on top of this, I don't think I've ever written a GET request that modifies anything it shouldn't.
It does defend against a form in a hidden iframe, however, you have to actually *use* it for your application to be protected.
True, but it is enabled by default in the new skeleton ApplicationController, and the whole secret key thing is negated by the new default cookie-based session store.
So, you have to pay attention when porting an old app, but you should be paying attention when porting anything anyway. But when building a new app, you almost have to go out of your way to get this wrong. Either way, it's very easy to fix if you find yourself not using it.
Closed source, for now. There's always the chance it will be opened in the future, but even if it was, there are other licenses to consider -- for example, we link against plugins which are MIT-licensed, and that's not GPL compatible.
And we do contribute back to open source projects we use. So, really, EXT's loss is jQuery's gain.
Now, the commercial license would be worth it. We're not against paying for things -- most of us use TextMate, for example, with legitimate licenses, and we have paid for Photoshop, and on other projects, Windows XP and Visual Studio.
The problem is, EXT doesn't have anything we need that jQuery doesn't, and they've already made one major license change. It's pretty clear we can't trust them to be consistent, or to interpret the GPL correctly. Can't build on a platform that shaky, and don't really want to work with someone who can't tell the difference between client/server communication and library linking.
That not only prevents against the image hack, it also prevents against things like a hidden form in an iframe.
Granted, it's still possible for you to do stupid things with GET requests, and it's possible you could turn it off entirely. But it's pretty trivial to stay safe here.
And no, there's not really going to be a sane way for browsers to protect you from this, unless you've left on all the annoying "You are about to send data over the internet!!!1!one" warnings. This is really going to be up to site admins to fix.
Granted, not all of Quake was written by John Carmack, but he is credited with quite a lot he's done by himself. He's got a shadowing trick named after himself, after all -- Carmack's Reverse.
So, given something like Word or Oracle, it's plausible that the first version, or even the first prototype, was written by exactly one guy. Take Linus Torvalds -- say what you will, but the original Linux was entirely his, complete with 386 support and a multithreaded filesystem (already giving it an edge over Minix).
Oh, and I doubt any actual paid publicists were used. Seriously, how would that actually work, and how would you justify the expense? I'm sure you were joking, but actually think about this -- for better or worse, these people are famous through word of mouth, among their peers. I'm guessing most have done something worth mentioning to earn that fame.
In fact, this has already happened to a Javascript library: EXTJS. Not quite in the sense you're talking about -- it was GPL'd -- but we still had to port away from it.
We might've been willing to release some of the Javascript source -- after all, GP is right, it's not like we can hide it -- but the author was claiming it applied to the web app serving the Javascript, also.
Although that's patently absurd, it's also untested in court, and it proved that he's exactly the kind of assmunch we don't want to work with. We've just finished porting everything to jQuery and MooTools. Probably better off for it.
If they do browse around the net, and bother to ask why I do what I do, then not playing their game hurts them more than piracy. It means they don't really have an excuse to start calling me a villain, and it means they may actually hear how much I wanted to play their game.
I don't care about spore, but damnit, Bioshock looked awesome, and like exactly my kind of game -- and I haven't played it yet, because of the DRM. All their hard work gone to waste because of that -- at least, developers will hear this part.
On the other hand, if they just look at statistics, they should note how much more Spore was pirated than games before it. It's a longshot, but in this case, piracy may actually help, because then they'd actually realize that DRM drives people to piracy, not away from it.
there are some DRM systems that don't control the user in this way, and might be said to be genuinely moderate; but none of them are effective.
Actually, "none of them are effective" applies to all DRM schemes.
In fact, that does kind of turn your argument on its head -- it's nonsensical to think that any one kind of DRM is more effective than another, either in restricting the user or in preventing actual piracy, because ultimately, all of them have been cracked, many before release, and pretty much all of the games that employ such cracked DRM are available, crack included, via BitTorrent.
So, no DRM can be more effective than any other DRM, because all DRM is absolutely ineffective.
The only way effectiveness could be a scale here is if you count actual profit from the game. In this case, it's hard to measure, because most of the industry is doing it...
Oh wait. That's right -- the whole industry is pretty much imploding, and people are going to consoles. At least with a console, the DRM is known, predictable, and isn't going to fuck up the rest of your multi-thousand-dollar gaming machine.
And please note my sig. I'm not endorsing piracy here (it actually hurts them more if you don't even play their game), and I'm not endorsing consoles as a solution. I want the PC to survive as a platform -- for that matter, PC as in "Personal Computer" (I like Linux). And if it's going to survive as a platform, the DRM madness has to stop.
Electronic voting machines have, time and time again, been shown to be inaccurate, insecure, and untrustworthy. In particular, those from Premiere Election Solutions (formerly Diebold) do not leave a verifiable paper trail, making a recount impossible.
What will you do to ensure that my vote will be counted in this election, or the next?
Pay attention to your choice here. I've used Bazaar and Mercurial for some personal projects, and will be migrating to Git at work. (Notice a common theme? DVCS! (Distributed Version Control System))
Version control has a profound impact on how you work. On small/personal projects, I recommend DVCS simply because it's easier to setup, and quicker to work with. Why waste time building a Subversion server, or signing up for one on Google Code? And if it's on Google Code, do you enjoy not being able to work when your Internet is out? Never mind the difference between commits happening instantly, and taking several minutes.
On larger projects, DVCS is helpful simply because merging is easier. It's gotten to the point where I avoid branching in Subversion for anything that'll take less than a day, because it could easily take 20 minutes or more to merge those changes back in, assuming no one else is touching trunk in the meantime -- and I'll have to resolve a few dozen conflicts. With Git, merging is much quicker and easier, so I branch whenever I feel like it.
* Test everything
Hard to say whether the parent menant human or automated testing. You need both.
In particular, read up on test-driven development, and behavior-driven development.
Once you understand that, unless your company has a profound need for speed, use the language which most cleanly allows you to specify intent, not mechanism. (If working with a larger project, this may not be up to you, but there's always the occasional script.)
Oh, and chances are, your company does not have a profound need for speed. Premature optimization is the root of all evil. Optimize after you have a performance issue, not before. (Given the choice between two equally pretty ways of doing something, you might choose the faster -- but if the faster-executing way is going to take 20 times more code, don't do it until you need it.)
For me, it drives me nuts when I'm in some MS app and the Home/End keys don't take me to the beginning/end of the document.
I'm fine with ctrl+home/end for that.
The sad, hilarious fact here is that in years of using Linux, everything from GNOME/KDE to WindowMaker, Fluxbox, Enlightenment, and brief brushes with TWM and Ratpoison, the home/end paradigm is always pretty much the same: In a viewer, they go to the beginning/end of a document. In an editor, they go to the beginning/end of a line, and in all GUI edit boxes I can remember using, they go to the beginning/end of a document.
Recently, I'm forced to use OS X as I wait for my new laptop to come in. And here, I have command+right/left in apps which follow the Apple model, home/end in the Terminal (changed from the default of shift+home/end), except in apps like less -- and, for that matter, most Terminal apps seem fairly broken about this -- vim, less... irb gets that right, but delete-forward doesn't work (called "del" elsewhere).
Except, as you say, in some apps -- Firefox might support both, I'm not sure.
It's possible it's just what I'm used to, but it really seems like, on OS X, my UI is even less consistent than on Linux, and that's just sad.
There's never a need to maximize an application to full screen unless you're one of those braindead Windows users who can't multitask
Bullshit, of course. Granted, there are plenty of apps which really don't need to go fullscreen, and I've started training myself to think this way on other OSes. (This browser isn't fullscreen, and there's no need for it to be -- easier to read when the line length is shorter anyway.)
The green "+" button works exactly as it should, always enlarging the window only as much as necessary to fit all content
But this is the part that really doesn't work -- see, this doesn't happen. The green '+' works exactly the way they felt like making it work on that app, in that document, on this particular Thursday.
Whereas here on Linux, I have one button to maximize the window, and if the window can't be resized, I simply don't have that button.
They still suck. Spaces is still a big, steaming pile of crap.
As long as we're talking about Spaces, there doesn't seem to be a keyboard shortcut (or I just haven't found it) for "move this window to another space". Instead, the closest I get is click+hold the window, then ctrl+arrow to switch spaces, dragging it with me.
When I switch to another space, focus should be on the last window used within that space.
I generally click on windows to focus them now, given that focus in general doesn't work the way I want on the Mac. I got sloppy focus to work -- in Terminal -- provided the Terminal app itself already has focus. That's probably the single biggest thing I miss from Linux.
Incidentally, the single biggest thing I miss from OS X is the fact that, in Terminal, at least, I could cycle forward/backward through terminal windows with a keystroke. I don't want tabbed everything -- I have a big enough monitor (and another on the way) that I can see all the windows I want. I just want more convenient ways to cycle through them -- ideally, one keystroke to switch apps, and one to switch windows within an app.
My window manager obviously knows enough to do this anyway -- it groups like windows when I get enough of them.
they still work better on average for my needs/desires than Linux and certainly Windows.
Unfortunately, the lack of sloppy focus, among many other things, has made it downright hostile. I can't really use OS X any more than I can use Windows for anything non-trivial.
With Linux, my main problem is application support, which isn't a huge deal -- it has everything I need to work (Rails, Firefox+Firebug, svn, git, Kate) and most of what I need to play (BitTorrent, mp
Go back and review this article, the purpose of this entire discussion in the first place is about the masses of people that struggle to adopt Mobile telephone platforms, not specialized tasks or hobbies.
Go back to the post where I first started talking about usability. All I said was:
usability is about more than just discoverability and learning curve.
Offtopic? Maybe. But, if I may be so bold, not wrong.
If the mobile phone industry cannot make the usability of those platforms easier
Nonsensical statement. How do you make "usability" easier?
Perhaps you meant to say, make these platforms easier to use. Or perhaps, more usable.
I would say:
If the mobile phone industry cannot improve the discoverability of those platforms, then people will not adopt them.
And this is true, and indeed, it is what the article was talking about. Nowhere in the article or the summary is the word "usability" mentioned.
I am not arguing that discoverability isn't important, or shouldn't be addressed. My point was, it should never be the only concern, even when you're only talking about "usability". You have to consider, once people do start to learn how to use phones in general, what's going to make them want to use your phone, in particular?
Discoverability is the difference between an iPhone and a Windows Mobile device. Usability is the difference between an iPhone and an iPod Touch -- or between an iPhone and a subnotebook ("netbook").
There are premium versions of Gmail, which can be paid for. I suppose the contract still says they can do whatever they want, but it's worth considering.
That said, I suspect the reason there are actually apologists is:
Because they make cool *looking* equipment?
No, because they make stuff that actually works, and works well, at least for its intended purpose. If you do exactly what Apple has designed for you to do, it will be the most seamless, painless, beautiful experience you've ever had.
And, because you're actually Apple's customer -- you're never Microsoft's customer -- you're much more likely to stay on that golden path of "exactly what Jobs thought of."
But let me be clear: I'm no fanboy, and I'm no apologist.
The trouble is, when you get off that path, things become very painful indeed. I want to use this keyboard I'm typing on as a PC keyboard, on my Linux machine, and I want a goddamned Insert key, but Apple has turned this into an 'fn' key, which is interpreted in hardware (or maybe firmware), placing it strictly outside my control.
Or take OS X -- I would like my home and end keys to go to the beginning and end of a line. I would like Terminal to have things like the del key actually work. I would love sloppy focus, and the ability to make stuff other than video actually go fullscreen, not just kinda-sorta-maximized-maybe.
And I get these things, kinda, sorta, maybe, with third-party apps -- but it'll never work well at all until it's blessed by Apple. Virtual desktops/workspaces sucked on OS X until Leopard implemented Spaces.
Slashdot Mirror
I tell people not to send those, and I train my spamfilter on them. Problem solved.
To replicate that functionality I would have to backup my entire email pretty much every time I got any, which is completely impractical.
Not really -- backing up from Gmail is as simple as IMAP. If you're talking about new email, that's just as simple -- and there are tools to keep IMAP mailboxes in sync. I imagine Gmail will happily forward incoming email.
Not to mention tedious. Yeah, shell scripts and all that crap
Who said anything about shell scripts? It's pretty much a builtin feature of at least one email client I know of, and there are third-party programs you can get dedicated to the purpose.
And if you've got a program to do it, where's the tedium?
why bother when Gmail does it all for me
Mostly because you have no idea if Gmail is going to be there tomorrow, or if your Internet connection is going to be there tomorrow.
Now, granted, I'm lazy too -- I haven't backed up my work Gmail ever, and it would be pretty inconvenient if that was lost.
Point is, it's not nearly as huge a difference as you're making it out to be. There are real reasons to use Gmail -- labels, fulltext index, nice web interface for when that matters -- yours just seems (ahem) retarded.
Free software could allow you to encrypt your data automatically from the service provider, to migrate yourself and your data to a compatible competitor, or to implement "premium" features yourself. But what's in it for the service provider?
Depends how it's done.
First, consider that many services have made your data somewhat portable, either intentionally or by accident. Gmail has been hot on this thread -- but Gmail supports IMAP. And most of the services being discussed here are web-based, which implies that unless the provider takes explicit measures to prevent it, you could always write a program to scrape their site for the information.
But consider the implications of data portability, by itself.
Done right, it means that you get encryption (if relevant) -- GnuPG is supported in both of the IMAP clients I've used lately -- as well as additional features, by way of Firefox extensions and third-party services -- think Facebook apps. (Or think of any service with an open API -- I just implemented Twitter's, took maybe an hour, from knowing nothing about Twitter.)
The only thing left is "precise control of their software" -- they certainly control the server-side software, but again, Firefox extensions and APIs mean that most things (ads, etc) can be stripped out by third parties.
Now, what's in it for the service provider?
Well, read all of the above -- that's a lot more service you're providing to an end-user. And it's a lot of features you get for free. Take YouTube -- you can embed it in any webpage, any random blog or whatever. It means free publicity for YouTube, and more people posting videos to YouTube, even if they only want them back on their own site.
Take Twitter -- I just added a feature to a project at work which pulls in a user's latest status update from Twitter and shows it on their profile. That's free publicity for Twitter, and quite possibly some new users because of it.
Gmail's popular because it's free. Try monetising it.
Like Google's already done, with the ads, and the premium accounts?
If you charge less, you're eating your desktop division's lunch.
If you don't have a desktop division, that means you're eating your competition's desktop division's lunch.
If you charge more, you're providing less service for more money, because the company ultimately doesn't own its data.
That assumes an identical featureset.
Consider Exchange vs Google Apps.
Exchange is going to have tighter integration with desktop apps like MS Office, and Office has more features and better support for legacy formats. It's also very likely you do backup inhouse, which may be required by your industry, and is probably going to give you more peace of mind. And it's not dependent on any third party, save for Microsoft, and there only for updates -- if Microsoft.com went down, your exchange server would still be up.
Google Apps has tighter integration with Google services, and, generally, better functionality for sharing documents and collaborating online. It has the additional advantage of outsourcing backup (Google can do it) and distribution (none needed), while being available anywhere. And depending on what you need in the way of desktop hardware, it's entirely possible you might not need any desktop software besides a browser -- which gives you many of the maintenance advantages of thin clients.
It is, in other words, apples and oranges. For some, Google Apps is a compelling alternative -- even worth paying for, possibly worth paying more for.
Is their software significantly better than MythTV?
If so, how?
If not, why don't they just sell a MythTV box?
*blink* Multithreaded filesystem?? What do you mean?
Just what I said. Here, further reading:
http://www.dina.kvl.dk/~abraham/Linus_vs_Tanenbaum.html
If I had to guess, I would say that a single-threaded filesystem is maybe a step up from Win98, if that. If you remember, reading from a floppy on Win98 would bring the entire system to a crawl, as any disk IO would make the rest of the OS block until it was done.
I didn't use Minix at the time, but I would guess it was like that, at least for the filesystem itself, if not for the entire OS -- in that, if the filesystem was blocked doing disk IO, all attempts to talk to the filesystem would also be blocked.
Minix is a full, Unix-like operating system, whereas Linux is a kernel.
Really? I was under the impression that it used GNU tools to fill in the rest of the system, just like Linux.
In fact, it's also credited to just one person -- Andy Tanenbaum.
I could be wrong, though. If so, that would be another point against Minix -- Linux was able to leverage existing, portable tools. Minix was not yet POSIX-compliant at the time, making portability between Minix and anything else a bit harder.
Good advice, mostly:
- Am I dressed and groomed appropriately?
This is important.
When I went to interview for my current job, I was wearing some nice pants, a buttoned shirt, probably even a tie. I was freshly showered and shaved.
Now, I haven't shaved in probably a year of working here, and I've never worn those pants or that shirt since. (I'm not even sure I fit into the pants anymore.) I'm a decided Unix longhair, one of my coworkers wears sandals...
The point is, you dress up for the interview as a show of respect, and to make a first impression. It has nothing to do with the job, and everything to do with the interview. (That said -- check the dress code, afterwards...)
- Have I thought about real answers to the typical questions and not just canned responses
I think a better question would be, can you do this on the fly?
Really, they're asking those questions to get to know you better -- it's a way of opening a conversation, and trying to get a feel for you as a person. If you've actually thought out a response, and you know it by heart, it becomes a canned response, and you can tell.
- During tech interviews, can I provide real world examples or am I spitting out algorithms and examples from text books?
Not really particularly important, I would think.
I've been on the other side of the (metaphorical) interviewing table a few times, and it's much better to provide an actual problem to solve than to ask for an example. You can start with FizzBuzz if you like, but give them something actually challenging, to show you how they think on the fly.
One I especially liked was a small snippet of JavaScript with a bug in it, which you had to understand closures in order to get. Less than ten lines -- find the bug. Obvious if you had any Javascript experience or skill, impossible if you'd just read a book.
And as a candidate, pay attention to when you fail a question. They won't tell you, but you should know. And when you know, go research that and be ready to ace it on your next interview.
Did NOT LIE about their experience or lack thereof
In what way is not listing a particular job a lie?
I never heard that a resume was supposed to list every job you ever had. Should I put "paperboy" on there, too?
The rest is worth knowing, and does kind of say something about the general quality of applicants.
It's a bad thing for several reasons, which have already been laid out elsewhere.
Using 4.2.2.2 means I abuse whatever poor bastard owns that IP now, and using my own DNS server means I'm paying for the same service twice -- once for my ISP to run their servers, and once to run my own.
And you can get the same functionality, for web browsing, in your browser. There's pretty much no app other than web browsing which would benefit to this, and several which would be hurt by this.
That's not "don't run", that's the "beta" part.
The point is, if this happens to you, too bad. Unless your hardware warranty covers it, you're SOL.
That unfortunately doesn't defend against GET requests. This means that if your script allows anything vulnerable to be requested via GET, then the 'image' attack vector is not protected.
Correct. That's why I mentioned "stupid things with GET requests", and why the page I linked to also links to some documentation about idempotence.
Building an app on top of this, I don't think I've ever written a GET request that modifies anything it shouldn't.
It does defend against a form in a hidden iframe, however, you have to actually *use* it for your application to be protected.
True, but it is enabled by default in the new skeleton ApplicationController, and the whole secret key thing is negated by the new default cookie-based session store.
So, you have to pay attention when porting an old app, but you should be paying attention when porting anything anyway. But when building a new app, you almost have to go out of your way to get this wrong. Either way, it's very easy to fix if you find yourself not using it.
My boast wasn't about Ruby, it was pointing out the trivialty of the problem.
If Microsoft has actually done this, and done it right, for five and a half years, great! It means even less of an excuse for anyone to get it wrong.
Closed source, for now. There's always the chance it will be opened in the future, but even if it was, there are other licenses to consider -- for example, we link against plugins which are MIT-licensed, and that's not GPL compatible.
And we do contribute back to open source projects we use. So, really, EXT's loss is jQuery's gain.
Now, the commercial license would be worth it. We're not against paying for things -- most of us use TextMate, for example, with legitimate licenses, and we have paid for Photoshop, and on other projects, Windows XP and Visual Studio.
The problem is, EXT doesn't have anything we need that jQuery doesn't, and they've already made one major license change. It's pretty clear we can't trust them to be consistent, or to interpret the GPL correctly. Can't build on a platform that shaky, and don't really want to work with someone who can't tell the difference between client/server communication and library linking.
Ruby On Rails has prevented this, by default, for almost a year:
http://api.rubyonrails.org/classes/ActionController/RequestForgeryProtection/ClassMethods.html
That not only prevents against the image hack, it also prevents against things like a hidden form in an iframe.
Granted, it's still possible for you to do stupid things with GET requests, and it's possible you could turn it off entirely. But it's pretty trivial to stay safe here.
And no, there's not really going to be a sane way for browsers to protect you from this, unless you've left on all the annoying "You are about to send data over the internet!!!1!one" warnings. This is really going to be up to site admins to fix.
Granted, not all of Quake was written by John Carmack, but he is credited with quite a lot he's done by himself. He's got a shadowing trick named after himself, after all -- Carmack's Reverse.
So, given something like Word or Oracle, it's plausible that the first version, or even the first prototype, was written by exactly one guy. Take Linus Torvalds -- say what you will, but the original Linux was entirely his, complete with 386 support and a multithreaded filesystem (already giving it an edge over Minix).
Oh, and I doubt any actual paid publicists were used. Seriously, how would that actually work, and how would you justify the expense? I'm sure you were joking, but actually think about this -- for better or worse, these people are famous through word of mouth, among their peers. I'm guessing most have done something worth mentioning to earn that fame.
In fact, this has already happened to a Javascript library: EXTJS. Not quite in the sense you're talking about -- it was GPL'd -- but we still had to port away from it.
We might've been willing to release some of the Javascript source -- after all, GP is right, it's not like we can hide it -- but the author was claiming it applied to the web app serving the Javascript, also.
Although that's patently absurd, it's also untested in court, and it proved that he's exactly the kind of assmunch we don't want to work with. We've just finished porting everything to jQuery and MooTools. Probably better off for it.
It's a tricky problem.
If they do browse around the net, and bother to ask why I do what I do, then not playing their game hurts them more than piracy. It means they don't really have an excuse to start calling me a villain, and it means they may actually hear how much I wanted to play their game.
I don't care about spore, but damnit, Bioshock looked awesome, and like exactly my kind of game -- and I haven't played it yet, because of the DRM. All their hard work gone to waste because of that -- at least, developers will hear this part.
On the other hand, if they just look at statistics, they should note how much more Spore was pirated than games before it. It's a longshot, but in this case, piracy may actually help, because then they'd actually realize that DRM drives people to piracy, not away from it.
there are some DRM systems that don't control the user in this way, and might be said to be genuinely moderate; but none of them are effective.
Actually, "none of them are effective" applies to all DRM schemes.
In fact, that does kind of turn your argument on its head -- it's nonsensical to think that any one kind of DRM is more effective than another, either in restricting the user or in preventing actual piracy, because ultimately, all of them have been cracked, many before release, and pretty much all of the games that employ such cracked DRM are available, crack included, via BitTorrent.
So, no DRM can be more effective than any other DRM, because all DRM is absolutely ineffective.
The only way effectiveness could be a scale here is if you count actual profit from the game. In this case, it's hard to measure, because most of the industry is doing it...
Oh wait. That's right -- the whole industry is pretty much imploding, and people are going to consoles. At least with a console, the DRM is known, predictable, and isn't going to fuck up the rest of your multi-thousand-dollar gaming machine.
And please note my sig. I'm not endorsing piracy here (it actually hurts them more if you don't even play their game), and I'm not endorsing consoles as a solution. I want the PC to survive as a platform -- for that matter, PC as in "Personal Computer" (I like Linux). And if it's going to survive as a platform, the DRM madness has to stop.
Electronic voting machines have, time and time again, been shown to be inaccurate, insecure, and untrustworthy. In particular, those from Premiere Election Solutions (formerly Diebold) do not leave a verifiable paper trail, making a recount impossible.
What will you do to ensure that my vote will be counted in this election, or the next?
* Version control everything
Pay attention to your choice here. I've used Bazaar and Mercurial for some personal projects, and will be migrating to Git at work. (Notice a common theme? DVCS! (Distributed Version Control System))
Version control has a profound impact on how you work. On small/personal projects, I recommend DVCS simply because it's easier to setup, and quicker to work with. Why waste time building a Subversion server, or signing up for one on Google Code? And if it's on Google Code, do you enjoy not being able to work when your Internet is out? Never mind the difference between commits happening instantly, and taking several minutes.
On larger projects, DVCS is helpful simply because merging is easier. It's gotten to the point where I avoid branching in Subversion for anything that'll take less than a day, because it could easily take 20 minutes or more to merge those changes back in, assuming no one else is touching trunk in the meantime -- and I'll have to resolve a few dozen conflicts. With Git, merging is much quicker and easier, so I branch whenever I feel like it.
* Test everything
Hard to say whether the parent menant human or automated testing. You need both.
In particular, read up on test-driven development, and behavior-driven development.
Once you understand that, unless your company has a profound need for speed, use the language which most cleanly allows you to specify intent, not mechanism. (If working with a larger project, this may not be up to you, but there's always the occasional script.)
Oh, and chances are, your company does not have a profound need for speed. Premature optimization is the root of all evil. Optimize after you have a performance issue, not before. (Given the choice between two equally pretty ways of doing something, you might choose the faster -- but if the faster-executing way is going to take 20 times more code, don't do it until you need it.)
For me, it drives me nuts when I'm in some MS app and the Home/End keys don't take me to the beginning/end of the document.
I'm fine with ctrl+home/end for that.
The sad, hilarious fact here is that in years of using Linux, everything from GNOME/KDE to WindowMaker, Fluxbox, Enlightenment, and brief brushes with TWM and Ratpoison, the home/end paradigm is always pretty much the same: In a viewer, they go to the beginning/end of a document. In an editor, they go to the beginning/end of a line, and in all GUI edit boxes I can remember using, they go to the beginning/end of a document.
Recently, I'm forced to use OS X as I wait for my new laptop to come in. And here, I have command+right/left in apps which follow the Apple model, home/end in the Terminal (changed from the default of shift+home/end), except in apps like less -- and, for that matter, most Terminal apps seem fairly broken about this -- vim, less... irb gets that right, but delete-forward doesn't work (called "del" elsewhere).
Except, as you say, in some apps -- Firefox might support both, I'm not sure.
It's possible it's just what I'm used to, but it really seems like, on OS X, my UI is even less consistent than on Linux, and that's just sad.
There's never a need to maximize an application to full screen unless you're one of those braindead Windows users who can't multitask
Bullshit, of course. Granted, there are plenty of apps which really don't need to go fullscreen, and I've started training myself to think this way on other OSes. (This browser isn't fullscreen, and there's no need for it to be -- easier to read when the line length is shorter anyway.)
The green "+" button works exactly as it should, always enlarging the window only as much as necessary to fit all content
But this is the part that really doesn't work -- see, this doesn't happen. The green '+' works exactly the way they felt like making it work on that app, in that document, on this particular Thursday.
Whereas here on Linux, I have one button to maximize the window, and if the window can't be resized, I simply don't have that button.
They still suck. Spaces is still a big, steaming pile of crap.
As long as we're talking about Spaces, there doesn't seem to be a keyboard shortcut (or I just haven't found it) for "move this window to another space". Instead, the closest I get is click+hold the window, then ctrl+arrow to switch spaces, dragging it with me.
When I switch to another space, focus should be on the last window used within that space.
I generally click on windows to focus them now, given that focus in general doesn't work the way I want on the Mac. I got sloppy focus to work -- in Terminal -- provided the Terminal app itself already has focus. That's probably the single biggest thing I miss from Linux.
Incidentally, the single biggest thing I miss from OS X is the fact that, in Terminal, at least, I could cycle forward/backward through terminal windows with a keystroke. I don't want tabbed everything -- I have a big enough monitor (and another on the way) that I can see all the windows I want. I just want more convenient ways to cycle through them -- ideally, one keystroke to switch apps, and one to switch windows within an app.
My window manager obviously knows enough to do this anyway -- it groups like windows when I get enough of them.
they still work better on average for my needs/desires than Linux and certainly Windows.
Unfortunately, the lack of sloppy focus, among many other things, has made it downright hostile. I can't really use OS X any more than I can use Windows for anything non-trivial.
With Linux, my main problem is application support, which isn't a huge deal -- it has everything I need to work (Rails, Firefox+Firebug, svn, git, Kate) and most of what I need to play (BitTorrent, mp
Go back and review this article, the purpose of this entire discussion in the first place is about the masses of people that struggle to adopt Mobile telephone platforms, not specialized tasks or hobbies.
Go back to the post where I first started talking about usability. All I said was:
usability is about more than just discoverability and learning curve.
Offtopic? Maybe. But, if I may be so bold, not wrong.
If the mobile phone industry cannot make the usability of those platforms easier
Nonsensical statement. How do you make "usability" easier?
Perhaps you meant to say, make these platforms easier to use. Or perhaps, more usable.
I would say:
If the mobile phone industry cannot improve the discoverability of those platforms, then people will not adopt them.
And this is true, and indeed, it is what the article was talking about. Nowhere in the article or the summary is the word "usability" mentioned.
I am not arguing that discoverability isn't important, or shouldn't be addressed. My point was, it should never be the only concern, even when you're only talking about "usability". You have to consider, once people do start to learn how to use phones in general, what's going to make them want to use your phone, in particular?
Discoverability is the difference between an iPhone and a Windows Mobile device. Usability is the difference between an iPhone and an iPod Touch -- or between an iPhone and a subnotebook ("netbook").
you get what you pay for.
There are premium versions of Gmail, which can be paid for. I suppose the contract still says they can do whatever they want, but it's worth considering.
Why does Apple get a free pass?
They don't, as you can see here.
If M$ did this, people would be all over them.
I would say we're "all over them" here.
That said, I suspect the reason there are actually apologists is:
Because they make cool *looking* equipment?
No, because they make stuff that actually works, and works well, at least for its intended purpose. If you do exactly what Apple has designed for you to do, it will be the most seamless, painless, beautiful experience you've ever had.
And, because you're actually Apple's customer -- you're never Microsoft's customer -- you're much more likely to stay on that golden path of "exactly what Jobs thought of."
But let me be clear: I'm no fanboy, and I'm no apologist.
The trouble is, when you get off that path, things become very painful indeed. I want to use this keyboard I'm typing on as a PC keyboard, on my Linux machine, and I want a goddamned Insert key, but Apple has turned this into an 'fn' key, which is interpreted in hardware (or maybe firmware), placing it strictly outside my control.
Or take OS X -- I would like my home and end keys to go to the beginning and end of a line. I would like Terminal to have things like the del key actually work. I would love sloppy focus, and the ability to make stuff other than video actually go fullscreen, not just kinda-sorta-maximized-maybe.
And I get these things, kinda, sorta, maybe, with third-party apps -- but it'll never work well at all until it's blessed by Apple. Virtual desktops/workspaces sucked on OS X until Leopard implemented Spaces.
I'll second Slicehost. Never used their DNS, but with how right they've gotten everything else, I suspect the DNS will rock.