Objective C is not a hard language to learn: it's a sibling to C++ in that both tried to add OOP to C. ObjC as used on the Mac combines the best of both worlds -- you get pointers for low level control, *and* a nice OO framework/API and niceties like garbage collection. And of course OS X is beautifully designed, none of the back compat cruft that makes one want to stay away from Win32.
Where C# wins over ObjC though is its similarity to Java (which in turn is fairly comfortable to C++/C programmers).
Ok, yeah, I know a lot of/.-ers won't be happy about Mono. But Mono is free, and you get to target the iPhone/iPad without owning a Mac. Macs are unaffordable for a lot of folk, and this should give open-source projects a way into the App Store*. (The other non-Apple alternative is Adobe's Creative Suite and that costs big $$$.)
*Yeah, they still have to pony up $99 to Apple to join the dev programme. But at least that's a one-time cost and could be sponsored by any OSS-supporting person/organization.
or not. Obama or not, remember that Hollywood greases Republican and Democrat pockets alike. Many of the big guys at the MPAA and RIAA are Democrats too, which must surely help.
As long as Hollywood gives politicians glamour by attending fundraisers, and actual cold hard cash, you won't find anyone in the government willing to speak out against Big Content. The only thing that can change this is public opinion.
> we buy devices that easily could be reprogrammable, but aren't?
DVD players are practically zero maintenance. My grandmum can use it. They're cheap ($30 or so) and also available in portable form factors ($70 or so), so I don't have a problem giving one to my niece for a road trip. Many of these devices support DivX. In fact, most solid-state MP4/video players from Asia support DivX too.
The closest programmable alternative to a DVD player's a Mac Mini-form factor device in the home, and a MP4 player/iPhone type device for the road. Those cost a lot more and require a fair bit of care and feeding.
Mark parent +1 insightful. Most DVD player imports from Asia support DivX (in fact, the only ones who don't support it are the big-brand names like Sony). I've taken random.avi files, burnt 'em on a DVD-R and these have played on a $30 DVD player. Yeah, I know it's not very high-tech, but it works.
I do hope they'll keep the old version (which supported Xvid/Divx) around for download.
A lot of people dislike the awesome bar because they expect a URL bar to do only what Netscape 4's URL bar did: a case-insensitive string prefix search of previously typed URLs. They don't like that it now matches words within URLs, Page titles and bookmarks. Thing is, IE, Chrome, Opera -- all do this now. The market has spoken(tm).
I personally love the feature and access it via the keyboard (it's fully keyboard accessible), I just don't get how it forces someone to use a mouse.
But in Firefox's specific implementation of the awesomebar, I've noticed some slowdowns, especially on older machines (Celerons & even older Pentium HTs with slow 5400rpm hard drives) when all you see is a spinning graphic to the left of the URL bar, if you type in a combination of words that don't exist, and Firefox has a lot of data to look through (it remembers 90 days worth of history by default IIRC). Other browsers (Chrome/Opera) haven't had this problem (but then I use Firefox very heavily and the others casually) so maybe folk are annoyed at the perceived lack of responsiveness.
I used the early Firefox 3.6 betas which had this feature. It was distracting once you had more than 3 tabs. I was glad when they killed this feature in one of the later betas.
I agree this is not a good practise, but I can see why they did it -- it was commercially necessary if they want to keep up with Chrome. Personally what they should have done is adopted Chrome's stable/beta channel strategy, with automatic updates for both channels by default. Who knows, maybe that's exactly what they'll do.
(I know they release betas already, but the notion of a Chrome beta channel is that you're permanently on the beta, trying out new features. If you're more adventuresome you can be on the developer channel, which essentially gives you very frequent updates.)
I don't notice insane slowness either, but UI responsiveness does go down when a lot of tabs are open (say 20+, on a fairly ordinary 2GB XPSP3 system). Switching between tabs is sluggish. This doesn't happen with Chrome, the UI feels responsive. The worst thing that happens with Chrome is that when you click a tab, it may have been caching the tab's contents (i.e. the rendered HTML page) on disk, so it paints the contents of the tab top-to-bottom like an old picture-display program. But the UI is never sluggish.
You're absolutely right. The heuristic for tab -> process is quite convoluted(although it's documented). I wrote per-site because I wanted to be concise and I recollected reading that under many conditions different tabs from the same site (especially if opened with a "Open as new tab" command) would actually render in the same process.
> Sadly, I don't think you even understand why its bloated.
To be honest, as a developer, I've been trying to understand it myself. Firefox feels snappy on low-end machines (even VMs) for light browsing (few tabs open) and only a couple of extensions loaded. It becomes sluggish with loads of tabs open, esp if kept open for a long time. My guess is that despite the improvements to the garbage collector, the one-process-for-all-tabs architecture is to blame.
Mark parent +1 insightful. Compare Chrome's adblocking vs Firefox's, for example. Firefox wins. And there are lots of cool, useful addons, like TabHunter, which is a cool way to navigate through lots of tabs. Or FireFTP -- an FTP client that works wherever Firefox does. Or DownThemAll, a download manager that works wherever Firefox does. And so on.
I think what Firefox _really_ needs is a Chrome-like Task Manager that shows you exactly how much memory/CPU/network your add-on is consuming. For example, on Chrome I know that the Gmail checker add-on takes 10MB memory, and ~0 CPU/network. I can always uninstall it if I think that's too much. Maybe when Firefox's Electrolysis project for per-process tabs goes mainstream, this feature will be implemented.
IE only appears light because it has a habit of stuffing most of its weight into a pile of processes hidden under the catch-all name of "svchost.exe", with additional chunks hidden in the OS itself.
This is exactly why sysadmins shouldn't pretend to be developers, and vice-versa. I don't use IE a lot (only if Firefox and Chrome both fail) but this statement is just wrong, a lazy repeating of a tech 'urban legend'. Go run Process Explorer and it'll show you what the svchosts are doing (hint: hosting services like DNS clients, etc). As for "additional chunks hidden in the OS itself", where exactly is this hidden, especially now that modern IEs don't even have any filesystem-browsing capability?
IE (like Mozilla, like Chrome) uses a lot of DLLs, but memory use etc is counted per process, and what IE reported upto IE7 was actually a fair representation of what each process used. With IE8 on, there are per-site processes like Google Chrome (not per-tab for both browsers as usually thought -- in fact, IE8 released this feature before Chrome) and you can get a better idea of how much memory a site is consuming.
Someone should make a version of Ubuntu or openSuse or Fedora or whatever that is designed for Seniors. Large Fonts, easy to use, very little duplication of apps, no problems.
The Eldy project does something like that; it's a shell over Linux (or Windows). The BBC had a story about it some time back about Eldy-based PCs being sold to the elderly.
One thing I've noticed about using third party DNS services like OpenDNS is that location aware sites that serve up content from different servers depending where you are (like YouTube) don't work well.
OpenDNS is distributed too; for many users they're in the same geographic vicinity, so this really should not happen. Further, most servers that serve up geotargeted content (like Youtube) use geo-ip mapping to target their response to the client IP, not the server IP that was hit.
Do you have any specific examples of geotargeting gone wrong because of OpenDNS use? If you could list them here, I'm sure folk from the sites in question (or OpenDNS itself) would love to investigate the problem.
>...and something which we cannot prove beyond a shadow of a > doubt, namely, what happened 5 to 10 thousand years ago.
Assuming you're refering to evolution, we *can* prove it beyond a shadow of a doubt, because evolution happens *today*. Mendel's work on plants, gray moth speciation in England, etc. And others too
But to go from "Everyone knows the worlds weather is changing" (your words) to "we need to spend trillions of dollars in restructuring the world economy and changing human behaviour" (my paraphrase of most of the green movement's message) is ridiculous, especially on the basis of the evidence presented. The key point is that no one's denying that climate changes. That is a truism. The question is-- is industrial activity the culprit, and to what extent is there warming? Frankly, Nobel prizes to Al Gore notwithstanding, there's still plenty up for debate. In fact, one of the sources of frustration among the CRU lot was that they couldn't show warming from '98 onwards.
But hey, it's hard to show doubt when you're on a moral crusade. Which is what most of global warming activism is all about.
The claims of evolution skeptics and round-earth skeptics is not backed up by observation and evidence. On the other hand, the more extreme claims of anthropogenic global warming _proponents_ are not backed up with sufficient observation and are extrapolated from very small datasets.
Given all of this, to say the "science is settled" is a travesty, and all those who said so fully deserve what's come so far and is undoubtedly coming as there's greater public and scientific scrutiny of their methods:
a) the Yamal tree-ring data - data from 10 trees is extrpolated into a 'trend' and finds its way into a number of papers b) CRU emails - won't say much more, too much said about this already. c) New Zealand average temperature graphs - high-school style 'cooking the graph' to match expectations
At this point, climate scientists who don't open up their raw data, modelling code and assumptions/decision-making are going to look as sleazy as PHB managers who forecast self-serving weird shit to make themselves look good to their bosses.
AGW true believers and "denialists" should be able to agree on this: the data get the last word, because without them theory is groundless. The only way for the CRU researchers to clear themselves of the imputation of serious error or fraud is full disclosure of the measurement techniques, the raw primary data sets, the code used to reduce them, and of their decisions during the process of interpretation. They should have nothing to hide; let them so demonstrate by hiding nothing.
In short, if computer models are the primary tool in making all sorts of climate predictions, then let's have transparency in building the models and getting conclusions from them.
> Will the BBC join? No! So international news is hopeless. Do people care about local news?
Actually, in America, people care primarily for local news. But TV affiliates aren't threatening to delist from Google and those cover local news too. But yes, for world news, there's the BBC, Deutsche Welle, Al-Jazeera, NPR, CNN, Xinhua, IBN... ol' Rupert has delusions of grandeur if he thinks himself indispensable. Hell, betcha foxnews.com and skynews.com *won't* delist, because they're not a newspaper.
> What if google endowed a nonprofit news organization? Or just bought wikinews the rights to use AP feeds?
Great point. In fact, the Christian Science Monitor (good newspaper, no Christian Science bias) is a non-profit org, is online-only 5 days a week, and has correspondents all over the world (doesn't only rely on AP/Reuters). And in fact Google pays AP to host AP news.
Mod parent +1 insightful. I can't believe idiots are crawling out of the woodwork to defend this.
Folks, Microsoft OSes require confirmation of all software (included signed packages) since 2007. It's now 2009 and you're developing *Linux*. Please don't repeat past follies.
There's nothing to discuss here. Your problem is that pretending asking for root authentication for *local* users in *active* sessions... when installing *trusted* software adds security is... well.. only a sign of dogma, snakeoil and ignorance when it comes to providing a secure system.
There's a superficial kind of sense in what he's saying. After all, if someone has console access, he could already pwn the machine, right? Well, the keyword here is defense-in-depth. There are lots of reasons non-root, non-trusted users might want to run from the console. Linux on random net-cafe machines is one example, where all kinds of people use the console. A family PC running Linux (hey, not as farfetched as it sounds) might have different user accounts for each family member.
Sure, it's trivial to fix this with pklocalauthority, but should users have to? It's about as lame as telling folk, "hey, you can just switch off IIS if you don't need it."
It's sad that while Microsoft and Windows has made so many strides towards providing a secure-by-default configuration, Red Hat seems to be losing the plot. Cf another gem of a comment on Bugzilla: I don't particularly care how UNIX has always worked. Sigh.
NT 3.x supported user-space drivers and was criticized by reviewers for poor graphics performance (especially those who wanted to run visualisation/CAD apps on it). But it was rock-solid, as you can imagine.
NT 4 introduced kernel-mode display drivers, which helped it become very popular with engineers who needed these apps (remember, the only other 'mainstream' OS on the market at this time was Win95/98 and System 8/9; NT was rock-solid by comparison and Linux didn't have many commercial apps at this time).
Given that stats show that 3rd-party drivers are the #1 reason behind Windows blue-screens, starting with Windows Vista, Microsoft started to use the mini-driver approach for drivers. They've expanded and refined the use of such drivers with Windows 7.
Essentially, vendors write a user mode client driver that executes in user-space, with some basic functionality being implemented in kernel-space by a Microsoft-written and Microsoft-QA'd driver. So you get crash resistance without losing speed. See Layered Driver Architecture on MSDN for more.
Slashdot Mirror
Objective C is not a hard language to learn: it's a sibling to C++ in that both tried to add OOP to C. ObjC as used on the Mac combines the best of both worlds -- you get pointers for low level control, *and* a nice OO framework/API and niceties like garbage collection. And of course OS X is beautifully designed, none of the back compat cruft that makes one want to stay away from Win32.
Where C# wins over ObjC though is its similarity to Java (which in turn is fairly comfortable to C++/C programmers).
Ok, yeah, I know a lot of /.-ers won't be happy about Mono. But Mono is free, and you get to target the iPhone/iPad without owning a Mac. Macs are unaffordable for a lot of folk, and this should give open-source projects a way into the App Store*. (The other non-Apple alternative is Adobe's Creative Suite and that costs big $$$.)
*Yeah, they still have to pony up $99 to Apple to join the dev programme. But at least that's a one-time cost and could be sponsored by any OSS-supporting person/organization.
or not. Obama or not, remember that Hollywood greases Republican and Democrat pockets alike. Many of the big guys at the MPAA and RIAA are Democrats too, which must surely help.
As long as Hollywood gives politicians glamour by attending fundraisers, and actual cold hard cash, you won't find anyone in the government willing to speak out against Big Content. The only thing that can change this is public opinion.
> we buy devices that easily could be reprogrammable, but aren't?
DVD players are practically zero maintenance. My grandmum can use it. They're cheap ($30 or so) and also available in portable form factors ($70 or so), so I don't have a problem giving one to my niece for a road trip. Many of these devices support DivX. In fact, most solid-state MP4/video players from Asia support DivX too.
The closest programmable alternative to a DVD player's a Mac Mini-form factor device in the home, and a MP4 player/iPhone type device for the road. Those cost a lot more and require a fair bit of care and feeding.
Mark parent +1 insightful. Most DVD player imports from Asia support DivX (in fact, the only ones who don't support it are the big-brand names like Sony). I've taken random .avi files, burnt 'em on a DVD-R and these have played on a $30 DVD player. Yeah, I know it's not very high-tech, but it works.
I do hope they'll keep the old version (which supported Xvid/Divx) around for download.
I could be wrong, but afaik Windows 7 has DivX built-in. It also plays most Quicktime .MOV files out of the box.
A lot of people dislike the awesome bar because they expect a URL bar to do only what Netscape 4's URL bar did: a case-insensitive string prefix search of previously typed URLs. They don't like that it now matches words within URLs, Page titles and bookmarks. Thing is, IE, Chrome, Opera -- all do this now. The market has spoken(tm).
I personally love the feature and access it via the keyboard (it's fully keyboard accessible), I just don't get how it forces someone to use a mouse.
But in Firefox's specific implementation of the awesomebar, I've noticed some slowdowns, especially on older machines (Celerons & even older Pentium HTs with slow 5400rpm hard drives) when all you see is a spinning graphic to the left of the URL bar, if you type in a combination of words that don't exist, and Firefox has a lot of data to look through (it remembers 90 days worth of history by default IIRC). Other browsers (Chrome/Opera) haven't had this problem (but then I use Firefox very heavily and the others casually) so maybe folk are annoyed at the perceived lack of responsiveness.
I used the early Firefox 3.6 betas which had this feature. It was distracting once you had more than 3 tabs. I was glad when they killed this feature in one of the later betas.
I agree this is not a good practise, but I can see why they did it -- it was commercially necessary if they want to keep up with Chrome. Personally what they should have done is adopted Chrome's stable/beta channel strategy, with automatic updates for both channels by default. Who knows, maybe that's exactly what they'll do.
(I know they release betas already, but the notion of a Chrome beta channel is that you're permanently on the beta, trying out new features. If you're more adventuresome you can be on the developer channel, which essentially gives you very frequent updates.)
They also use http://wau.tynt.com/javascripts/TyntLite.js for some pages, so I'd recommend adding http://*.tynt.com/* if your blocking system supports multiple wildcards.
I don't notice insane slowness either, but UI responsiveness does go down when a lot of tabs are open (say 20+, on a fairly ordinary 2GB XPSP3 system). Switching between tabs is sluggish. This doesn't happen with Chrome, the UI feels responsive. The worst thing that happens with Chrome is that when you click a tab, it may have been caching the tab's contents (i.e. the rendered HTML page) on disk, so it paints the contents of the tab top-to-bottom like an old picture-display program. But the UI is never sluggish.
You're absolutely right. The heuristic for tab -> process is quite convoluted(although it's documented). I wrote per-site because I wanted to be concise and I recollected reading that under many conditions different tabs from the same site (especially if opened with a "Open as new tab" command) would actually render in the same process.
> Sadly, I don't think you even understand why its bloated.
To be honest, as a developer, I've been trying to understand it myself. Firefox feels snappy on low-end machines (even VMs) for light browsing (few tabs open) and only a couple of extensions loaded. It becomes sluggish with loads of tabs open, esp if kept open for a long time. My guess is that despite the improvements to the garbage collector, the one-process-for-all-tabs architecture is to blame.
Mark parent +1 insightful. Compare Chrome's adblocking vs Firefox's, for example. Firefox wins. And there are lots of cool, useful addons, like TabHunter, which is a cool way to navigate through lots of tabs. Or FireFTP -- an FTP client that works wherever Firefox does. Or DownThemAll, a download manager that works wherever Firefox does. And so on.
I think what Firefox _really_ needs is a Chrome-like Task Manager that shows you exactly how much memory/CPU/network your add-on is consuming. For example, on Chrome I know that the Gmail checker add-on takes 10MB memory, and ~0 CPU/network. I can always uninstall it if I think that's too much. Maybe when Firefox's Electrolysis project for per-process tabs goes mainstream, this feature will be implemented.
This is exactly why sysadmins shouldn't pretend to be developers, and vice-versa. I don't use IE a lot (only if Firefox and Chrome both fail) but this statement is just wrong, a lazy repeating of a tech 'urban legend'. Go run Process Explorer and it'll show you what the svchosts are doing (hint: hosting services like DNS clients, etc). As for "additional chunks hidden in the OS itself", where exactly is this hidden, especially now that modern IEs don't even have any filesystem-browsing capability?
IE (like Mozilla, like Chrome) uses a lot of DLLs, but memory use etc is counted per process, and what IE reported upto IE7 was actually a fair representation of what each process used. With IE8 on, there are per-site processes like Google Chrome (not per-tab for both browsers as usually thought -- in fact, IE8 released this feature before Chrome) and you can get a better idea of how much memory a site is consuming.
The Eldy project does something like that; it's a shell over Linux (or Windows). The BBC had a story about it some time back about Eldy-based PCs being sold to the elderly.
OpenDNS is distributed too; for many users they're in the same geographic vicinity, so this really should not happen. Further, most servers that serve up geotargeted content (like Youtube) use geo-ip mapping to target their response to the client IP, not the server IP that was hit.
Do you have any specific examples of geotargeting gone wrong because of OpenDNS use? If you could list them here, I'm sure folk from the sites in question (or OpenDNS itself) would love to investigate the problem.
How can a post with score=1 (no karma bonus) get a -1, overrated mod? To the idiot mod who did this: /. doesn't have a "disagree" mod.
> ...and something which we cannot prove beyond a shadow of a
> doubt, namely, what happened 5 to 10 thousand years ago.
Assuming you're refering to evolution, we *can* prove it beyond a shadow of a doubt, because evolution happens *today*. Mendel's work on plants, gray moth speciation in England, etc. And others too
But to go from "Everyone knows the worlds weather is changing" (your words) to "we need to spend trillions of dollars in restructuring the world economy and changing human behaviour" (my paraphrase of most of the green movement's message) is ridiculous, especially on the basis of the evidence presented. The key point is that no one's denying that climate changes. That is a truism. The question is-- is industrial activity the culprit, and to what extent is there warming? Frankly, Nobel prizes to Al Gore notwithstanding, there's still plenty up for debate. In fact, one of the sources of frustration among the CRU lot was that they couldn't show warming from '98 onwards.
But hey, it's hard to show doubt when you're on a moral crusade. Which is what most of global warming activism is all about.
The claims of evolution skeptics and round-earth skeptics is not backed up by observation and evidence. On the other hand, the more extreme claims of anthropogenic global warming _proponents_ are not backed up with sufficient observation and are extrapolated from very small datasets.
Given all of this, to say the "science is settled" is a travesty, and all those who said so fully deserve what's come so far and is undoubtedly coming as there's greater public and scientific scrutiny of their methods:
a) the Yamal tree-ring data - data from 10 trees is extrpolated into a 'trend' and finds its way into a number of papers
b) CRU emails - won't say much more, too much said about this already.
c) New Zealand average temperature graphs - high-school style 'cooking the graph' to match expectations
At this point, climate scientists who don't open up their raw data, modelling code and assumptions/decision-making are going to look as sleazy as PHB managers who forecast self-serving weird shit to make themselves look good to their bosses.
Open-sourcing the Global Warming Debate:
In short, if computer models are the primary tool in making all sorts of climate predictions, then let's have transparency in building the models and getting conclusions from them.
> Will the BBC join? No! So international news is hopeless. Do people care about local news?
Actually, in America, people care primarily for local news. But TV affiliates aren't threatening to delist from Google and those cover local news too. But yes, for world news, there's the BBC, Deutsche Welle, Al-Jazeera, NPR, CNN, Xinhua, IBN... ol' Rupert has delusions of grandeur if he thinks himself indispensable. Hell, betcha foxnews.com and skynews.com *won't* delist, because they're not a newspaper.
> What if google endowed a nonprofit news organization? Or just bought wikinews the rights to use AP feeds?
Great point. In fact, the Christian Science Monitor (good newspaper, no Christian Science bias) is a non-profit org, is online-only 5 days a week, and has correspondents all over the world (doesn't only rely on AP/Reuters). And in fact Google pays AP to host AP news.
Mod parent +1 insightful. I can't believe idiots are crawling out of the woodwork to defend this.
Folks, Microsoft OSes require confirmation of all software (included signed packages) since 2007. It's now 2009 and you're developing *Linux*. Please don't repeat past follies.
Interesting and wrong, I should say:
There's a superficial kind of sense in what he's saying. After all, if someone has console access, he could already pwn the machine, right? Well, the keyword here is defense-in-depth. There are lots of reasons non-root, non-trusted users might want to run from the console. Linux on random net-cafe machines is one example, where all kinds of people use the console. A family PC running Linux (hey, not as farfetched as it sounds) might have different user accounts for each family member.
Sure, it's trivial to fix this with pklocalauthority, but should users have to? It's about as lame as telling folk, "hey, you can just switch off IIS if you don't need it."
It's sad that while Microsoft and Windows has made so many strides towards providing a secure-by-default configuration, Red Hat seems to be losing the plot. Cf another gem of a comment on Bugzilla: I don't particularly care how UNIX has always worked. Sigh.
NT 3.x supported user-space drivers and was criticized by reviewers for poor graphics performance (especially those who wanted to run visualisation/CAD apps on it). But it was rock-solid, as you can imagine.
NT 4 introduced kernel-mode display drivers, which helped it become very popular with engineers who needed these apps (remember, the only other 'mainstream' OS on the market at this time was Win95/98 and System 8/9; NT was rock-solid by comparison and Linux didn't have many commercial apps at this time).
Given that stats show that 3rd-party drivers are the #1 reason behind Windows blue-screens, starting with Windows Vista, Microsoft started to use the mini-driver approach for drivers. They've expanded and refined the use of such drivers with Windows 7.
Essentially, vendors write a user mode client driver that executes in user-space, with some basic functionality being implemented in kernel-space by a Microsoft-written and Microsoft-QA'd driver. So you get crash resistance without losing speed. See Layered Driver Architecture on MSDN for more.