Or Glade. Glade's been around for a while, and it works. It doesn't output code (well, it can, but that's deprecated), so it isn't quite what the GP was looking for.
Put aside antitrust law for a minute and then tell me what law MS has broken. If you think it was bundling browsers, pass a law saying "OS makers cannot bundle browsers" and apply it to everyone.
The web browser situation doesn't make much sense to me. Firefox's success is measured in downloads, for instance, not the number of people who call the Mozilla Foundation requesting an installation CD. If I want to install Opera, I need a web browser to get to the Opera website.
I suppose Microsoft could set IE's home page to a Microsoft site describing alternative browsers. Then there are concerns of fairness -- Firefox is listed above Dillo, for instance, oh noes! But that would reduce antitrust concerns.
I think anyone with good knowledge of KDE and GNOME would adapt to Windows quickly. They would be less able to change settings, but otherwise they could work effectively right away.
Sun has different priorities. If they refuse a contribution, there's no guarantee that they will duplicate that functionality. Also, if they took the original contribution, they could offer more new features in a given amount of time, with more QA effort available.
A fork would merge in Sun's code directly after review, so that's still a net gain.
Something like that. US nuclear submarines were out of commission for some time when Microsoft force-upgraded to IE7. That made a lot of people very angry.
For some games, you can rent them to try them out. If it's a bad game, then you lose five or ten dollars. If it's good, you lose five or ten dollars. And the developers don't even see an appreciable portion of this money.
This to me sounds like laziness. "But parallel programming is HARD!"
That's probably a better argument than fighting the CPU designers. If parallel programming is hard, it's more expensive to create parallel programs. They'll take longer to write. It'll take longer to implement new features for them. It'll take more money to maintain them. All that seems like a good reason to avoid parallel programming.
On the other hand, if someone comes up with a new parallel programming paradigm that's slightly more difficult than procedural/object-oriented programming, but offers these benefits -- or if this exists already -- it'll make sense to switch to that paradigm as your performance needs increase.
"Actually, I can see a whole lot of potential parallelism in compiling C. (Think parallelism)."
That's about what you said.
To give some more useful examples:
- Each compilation unit can be compiled in isolation, assuming you have certain information about the referenced code. (Granted, this is C, so the compilation units are pretty large.) - You can assemble a fair bit of information about each section of code -- enough to parallelize much of the conversion to IR -- in parallel. You will need to block on other processes at times, though. - Each function can be optimized in isolation.
Preprocessing, I think, hasn't been easy to parallelize -- I don't think distcc does it.
By "virtual memory" we mean "non-shared address spaces". For example: let's say you have a web browser and a text editor open. They both allocate a chunk of memory. The pointers they get back both happen to be 0xDEADBEEF. These pointers point to different sections of memory.
It just so happens that, with virtual memory, I can allocate more memory than I have RAM for. If I try accessing memory that doesn't have RAM backing it, the OS gets called. It will either swap out memory from other applications, get rid of caches that it controls, or kill some random process and steal its memory.
Your complaint is not about virtual memory -- it's about swapping to disk.
I've seen things in Linux that are just now making their way into competing operating systems. Mainly window manager features, because that's one application that I'm using constantly. Edge detection, always-on-top, virtual desktops, rolling up windows...
I haven't had X11 crash on me, ever.[1] It's rare to get the window manager to crash. I use e17, and when it crashes, it pops up a dialog box saying "Oops, I just crashed. Click here to restart, or here to quit." When an application crashes, it takes out that application, not X11.
[1] Okay, this is an outright lie. I haven't had it crash on me since I stopped using Gentoo -- since I stopped editing xorg.conf manually -- and I haven't ever had it start successfully and then crash. I have, however, given it a bad configuration and had it error out immediately.
How much money would Disney lose if out-of-print works lost copyright? They'd have to make old Steamboat Willie cartoons available for sale. They might lose a few million a year by not being able to release limited edition reprintings of their older films.
More likely, there'd be a delay between going out of print and going into the public domain -- ten or twenty years. That's long enough that Disney could still clean up on the limited editions and Steamboat Willie would remain in copyright. But why would Disney accept this when they can get more for free?
If you expect to get any useful information from/., you're sorely mistaken.
That said, I'm a Gen Yer and would be reasonably willing to download, but I'd be more willing to find a different book. If I knew of a particular out-of-print book that I wanted... I certainly wouldn't buy used books online. I'd prefer a reasonable quality used book to an ebook, but I'd not hesitate to download if the physical copy was not readily available.
From a scientific point of view, "it just is" and "god made it" are idempotent.
From what you are saying, there are two functions, "it just is" and "god made it", such that: "it just is" ("it just is"(x)) = "it just is"(x) forall x "god made it"("god made it"(x)) = "god made it"(x) forall x
Sections 7.2 and 7.3 deal with this. Google is, it seems, following OpenID 2.0, as far as indirection is concerned.
It was implied, though, that Google's allowing "username@example.com" rather than the typical "username.example.com". If they want to accept that, fine; but if they want to require that from other people, that's not so great.
1. Do they make it possible for everyone else to implement exactly what they are doing, on both the producer and consumer end, without any patent restrictions, royalties, or discriminatory licensing?
I'm assuming you can. Their stuff is a thin wrapper over OpenId; it'd probably take a week or less to implement it for your code.
2. How close is what they are doing to the latest version of the standard, not 1.0?
There is nothing similar in the 2.0 OpenId standard.
3. Do they try to get what they are doing into version 2.1 (or whatever) of the standard?
4. Do they really have a reason for doing this? Like making the login easier for normal nontechnical people rather than you and I?
It allows you to log in with "example@gmail.com" rather than "example.gmail.com". On the other hand, this could be implemented on the client side easily enough.
Other than that, this is just another layer of indirection. It's as if Google had a modified version of the C standard library in which "free" took a pointer to a pointer to the memory you want to free -- there might be a reason, but in practice it's going to be too confusing for everyone to start using it.
This means that you could potentially have a government get into power which enjoyed less than the support of at least 50% of the citizens! That situation would be intolerable in a democracy such as Australia.
That candidate would also enjoy the opposition of less than 50% of the citizens. Not voting is equivalent to giving every candidate an equal level of support -- something that the Australian voting system doesn't directly allow.
Have you heard of scantron? You can use paper ballots and get very fast, accurate results.
The only thing such a system won't count automatically is writing in a name. And you could simply mark a lack of machine-readable vote in one category; if there is a sufficient number of them, you can go through that manually.
Or you could do a machine-mediated vote: if none of the candidates appeal to you, you can type in the name of your candidate manually. OCR for a single font is relatively simple and accurate. There'll still be typos, but only for writeins, and most candidates with a chance at winning will be on the ballot. But here, you can even count the writeins in a very short period. And all it requires is a slightly modified typewriter, an LCD (to offer the default choices for each election), and a scanner.
Slashdot Mirror
Or Glade. Glade's been around for a while, and it works. It doesn't output code (well, it can, but that's deprecated), so it isn't quite what the GP was looking for.
Or MonoDevelop.
Put aside antitrust law for a minute and then tell me what law MS has broken. If you think it was bundling browsers, pass a law saying "OS makers cannot bundle browsers" and apply it to everyone.
The web browser situation doesn't make much sense to me. Firefox's success is measured in downloads, for instance, not the number of people who call the Mozilla Foundation requesting an installation CD. If I want to install Opera, I need a web browser to get to the Opera website.
I suppose Microsoft could set IE's home page to a Microsoft site describing alternative browsers. Then there are concerns of fairness -- Firefox is listed above Dillo, for instance, oh noes! But that would reduce antitrust concerns.
I think anyone with good knowledge of KDE and GNOME would adapt to Windows quickly. They would be less able to change settings, but otherwise they could work effectively right away.
How many of these have a competing commercial offering based on the same code?
Sun has different priorities. If they refuse a contribution, there's no guarantee that they will duplicate that functionality. Also, if they took the original contribution, they could offer more new features in a given amount of time, with more QA effort available.
A fork would merge in Sun's code directly after review, so that's still a net gain.
Something like that. US nuclear submarines were out of commission for some time when Microsoft force-upgraded to IE7. That made a lot of people very angry.
Interestingly, the D programming language is getting pure functions, which are automatically parallelizable.
For some games, you can rent them to try them out. If it's a bad game, then you lose five or ten dollars. If it's good, you lose five or ten dollars. And the developers don't even see an appreciable portion of this money.
This to me sounds like laziness. "But parallel programming is HARD!"
That's probably a better argument than fighting the CPU designers. If parallel programming is hard, it's more expensive to create parallel programs. They'll take longer to write. It'll take longer to implement new features for them. It'll take more money to maintain them. All that seems like a good reason to avoid parallel programming.
On the other hand, if someone comes up with a new parallel programming paradigm that's slightly more difficult than procedural/object-oriented programming, but offers these benefits -- or if this exists already -- it'll make sense to switch to that paradigm as your performance needs increase.
"Actually, I can see a whole lot of potential parallelism in compiling C. (Think parallelism)."
That's about what you said.
To give some more useful examples:
- Each compilation unit can be compiled in isolation, assuming you have certain information about the referenced code. (Granted, this is C, so the compilation units are pretty large.)
- You can assemble a fair bit of information about each section of code -- enough to parallelize much of the conversion to IR -- in parallel. You will need to block on other processes at times, though.
- Each function can be optimized in isolation.
Preprocessing, I think, hasn't been easy to parallelize -- I don't think distcc does it.
By "virtual memory" we mean "non-shared address spaces". For example: let's say you have a web browser and a text editor open. They both allocate a chunk of memory. The pointers they get back both happen to be 0xDEADBEEF. These pointers point to different sections of memory.
It just so happens that, with virtual memory, I can allocate more memory than I have RAM for. If I try accessing memory that doesn't have RAM backing it, the OS gets called. It will either swap out memory from other applications, get rid of caches that it controls, or kill some random process and steal its memory.
Your complaint is not about virtual memory -- it's about swapping to disk.
Linux has 0.8% market share!
Though that's counting me and my beard of unusual size, so take it as you wish.
I've seen things in Linux that are just now making their way into competing operating systems. Mainly window manager features, because that's one application that I'm using constantly. Edge detection, always-on-top, virtual desktops, rolling up windows...
I haven't had X11 crash on me, ever.[1] It's rare to get the window manager to crash. I use e17, and when it crashes, it pops up a dialog box saying "Oops, I just crashed. Click here to restart, or here to quit." When an application crashes, it takes out that application, not X11.
[1] Okay, this is an outright lie. I haven't had it crash on me since I stopped using Gentoo -- since I stopped editing xorg.conf manually -- and I haven't ever had it start successfully and then crash. I have, however, given it a bad configuration and had it error out immediately.
And Ubuntu doesn't prevent you from choosing. I run E17 on Ubuntu, for instance. It just gives you a good default.
How much money would Disney lose if out-of-print works lost copyright? They'd have to make old Steamboat Willie cartoons available for sale. They might lose a few million a year by not being able to release limited edition reprintings of their older films.
More likely, there'd be a delay between going out of print and going into the public domain -- ten or twenty years. That's long enough that Disney could still clean up on the limited editions and Steamboat Willie would remain in copyright. But why would Disney accept this when they can get more for free?
If you expect to get any useful information from /., you're sorely mistaken.
That said, I'm a Gen Yer and would be reasonably willing to download, but I'd be more willing to find a different book. If I knew of a particular out-of-print book that I wanted... I certainly wouldn't buy used books online. I'd prefer a reasonable quality used book to an ebook, but I'd not hesitate to download if the physical copy was not readily available.
astyle --random can help.
From a scientific point of view, "it just is" and "god made it" are idempotent.
From what you are saying, there are two functions, "it just is" and "god made it", such that:
"it just is" ("it just is"(x)) = "it just is"(x) forall x
"god made it"("god made it"(x)) = "god made it"(x) forall x
I concur. Lotus Notes is just too large to be a reasonable solution for this problem.
There is nothing similar in the 2.0 OpenId standard.
HAHA DISREGARD THAT, I DON'T READ STANDARDS
The standard: http://openid.net/specs/openid-authentication-2_0.html
Sections 7.2 and 7.3 deal with this. Google is, it seems, following OpenID 2.0, as far as indirection is concerned.
It was implied, though, that Google's allowing "username@example.com" rather than the typical "username.example.com". If they want to accept that, fine; but if they want to require that from other people, that's not so great.
1. Do they make it possible for everyone else to implement exactly what they are doing, on both the producer and consumer end, without any patent restrictions, royalties, or discriminatory licensing?
I'm assuming you can. Their stuff is a thin wrapper over OpenId; it'd probably take a week or less to implement it for your code.
2. How close is what they are doing to the latest version of the standard, not 1.0?
There is nothing similar in the 2.0 OpenId standard.
3. Do they try to get what they are doing into version 2.1 (or whatever) of the standard?
4. Do they really have a reason for doing this? Like making the login easier for normal nontechnical people rather than you and I?
It allows you to log in with "example@gmail.com" rather than "example.gmail.com". On the other hand, this could be implemented on the client side easily enough.
Other than that, this is just another layer of indirection. It's as if Google had a modified version of the C standard library in which "free" took a pointer to a pointer to the memory you want to free -- there might be a reason, but in practice it's going to be too confusing for everyone to start using it.
This means that you could potentially have a government get into power which enjoyed less than the support of at least 50% of the citizens! That situation would be intolerable in a democracy such as Australia.
That candidate would also enjoy the opposition of less than 50% of the citizens. Not voting is equivalent to giving every candidate an equal level of support -- something that the Australian voting system doesn't directly allow.
Have you heard of scantron? You can use paper ballots and get very fast, accurate results.
The only thing such a system won't count automatically is writing in a name. And you could simply mark a lack of machine-readable vote in one category; if there is a sufficient number of them, you can go through that manually.
Or you could do a machine-mediated vote: if none of the candidates appeal to you, you can type in the name of your candidate manually. OCR for a single font is relatively simple and accurate. There'll still be typos, but only for writeins, and most candidates with a chance at winning will be on the ballot. But here, you can even count the writeins in a very short period. And all it requires is a slightly modified typewriter, an LCD (to offer the default choices for each election), and a scanner.