I've found myself in the same boat. What pulled me in.. consistency across platforms and KDE integration. Opera works basically the same way on Windows, Mac, and Linux. Firefox doesn't.. the Mac build is made to be more Safari-like. This annoys me.. I would actually prefer Fox over Safari if this wasn't the case. Firefox is somewhat handicaped in KDE, lacking any type of reasonable integration. I use OSX at work and Gentoo+KDE at home, so I used Safari and Konqueror in the past. I always used Fox for the rare occasion of windows browsing.
I can understand the KDE problems with Firefox, I would even help w/ QT integration if I had enough free time. The Mac issue is really annoying though.. at least to me. I like to use favicons on the toolbar for my most common sites. I can do this w/ Fox in Windows or Linux, but not on Mac. It sounds odd, but it's a problem for me. Apparantly most people like it that way, but I don't. When Opera came out, I had the chance to use the same browser for both OSX and KDE. It was consistent and integrated well with both (especially with KDE). I can use Favicons the way I like and I actually like the layout more.
A few annoyances remain, but Opera is very nice. I hope they fix some of the rendering issues for version 9. It's very fast but has more rendering bugs than any of the other browsers I've used. That hasn't been enough to make turn away though. It's the only negative for me.
I didn't even notice the amount of controversy this post created. Usually I don't respond to my own post, but it seems like I need to explain my reasoning...
Hear is why I think pointer arithmetic is bad(tm) by example:
I used to work for a defense contractor. The project was large and had a large amount of legacy C code that needed to be integrated w/ newer C++ code. The legacy code had already been deployed in a number of places for a number of years. Performance and stability were both of top importance. It had to run in real time (at least the C portion did) and had to be completely stable. The legacy code had a large amount of pointer arithmetic. While profiling, we found hundreds of instances where incorrect pointer arithmetic would cause memory leaks and nasty bounds problems. Very few problems were found in the code that used arrays. It took so much time to fix these problems that it added months to the development time of the product. The pointer arithmetic was extremely confusing and we ended up re-writing much of it using arrays. The result was something that was rock stable and ran a few milliseconds slower. The speed difference was not significant and the stability meant that we could be confident that the system would be dependable when delployed. Which was _extremely_ important.
Would you like to fly on an airplane that had software written with excessive pointer arithmetic on it? It wouldn't surprise me if this was the case on some planes, but the thought would be unsettling to me. The reality is that most people get confused when doing pointer arithmetic and make more mistakes. For people that don't get confused and can write perfect code like this, more power to you. I would be worried about adding code like this to any project though, because other project members will most likely not be nearly as good and will make mistakes. That's why I would ask the person to rewrite it. (BTW, I wouldn't be saying "this teh sucks" but "future maintainers may have a hard time understanding this"). I am very big on performance and optimizations, but not at the sake of stability and readability (because less readable code will end up being less stable down the line).
The array version is neither clearer nor less error-prone. And I dare you to prove differently!
It seems like this challenge is issued and then you go on to say it can't be done... If you're saying it can't be proved, I agree. I'm not sure that anyone can "prove" that pointer arithmetic is less clear. This is subjective. Perhaps if pointer arithmetic were the standard way of doing things, it would be easy for everyone to use. Arrays are far more common though, and most people are much more comfortable with them as a result. The reason I think it's more error-prone is coupled to this. In practice, people tend to make more mistakes with them,because they're less familiar with how they should be used.
I wouldn't want to work for you though. Your competitors will write a faster program that uses less memory and you will loose the contract/job.
I'm not sure how often a contract would be lost due to array related performance issues. On the occassions where I've had to fix performance issues, the problem was usually a poorly written algorithm and not anything like "array overhead".
The pointer version is more portable across the chasm of compiler quality. That is, it will run well even on poorer compilers.
It's pretty hard to find hardware not supported by a major compiler these days. GCC works on some amazingly obscure hardware. I suppose in instances where the hardware is extremely rare and doesn't have a decent compiler, pointer arithmetic may be the best option. This would be a very rare circumstance though.
Thinking like this.. Is why we have big, ugly, bloated programs that require overpowered CPUs.
Imagine if everyone said that back when all we had was assembly... Also, I took a job for PHP/Perl experience and we're now considering moving to Ruby/Rails. I know a lot of people are in the same situation. The fact that it's not in the job postings is actually nice. Do you really want to see "2 years of Ruby on Rails experience"? Being a developer also means that you continue to evaluate available technologies and how they may help you and your organization. Rails is definitely such a technology.
The problem with datagnomes is that there are only so many of them. Most sites, for example only have 1 or 2 database servers and many more web servers. The problem here is that the datagnome population grows at a low rate, whereas the webgnome population seems to grow exponenentially. This makes it easy to overwork the datagnomes. Which is bad for overall gnome productivity. So.. in most cases it's better to make the webgnomes do as much of the logic work as possible. So all the datagnomes have to do is scoop up the data and throw it back to the webgnomes.. they usually stay pretty busy just doing the scooping. If you make them run around too much, then your webgnomes will be sitting around smokin cigs waiting for the datagnomes to get their work done.
These types of opimizations are virtually pointless on modern machines. The increased readability and lower likelihood of programming errors on the array option far outway any speed increase for the pointer option. Plus, as you noticed, the resulting assembly is basically the same. Most likely, both will run at virtually the same speed with modern compilers. Not only would the speed difference be unnoticeable, it would be utterly inconsequential.
IMHO there is no place for pointer arithmetic in modern software. If someone working for me wrote something like the second option, I would ask them to rewrite it.
Maybe the reason it took two weeks is because they had windows admins set up linux servers. Maybe that's _also_ why it didn't work well. Maybe they need to do some training or hire someone with the right skillsets?
The whole concept of Mono is somewhat confounding to me. Why not just implement something new? Why spend so much time recoding the.NET API?? When I first heard of Mono I thought that it was cool that existing.NET apps could potentially run in Linux. Then I realized that I don't give a ____ about these apps anyway. And I gaurantee I won't bother running some GPhoto type application in Windows. Anyone who uses this is just going to be using Linux anyway. On the server end.. I have to wonder what the ROI would be for porting apps from Windows-IIS/.NET to Linux/Mono.NET. It seems like a complete waste of time. Also, most people that write cross platform apps simply don't care about.NET and won't bother learning it, even if it really is as cool as Miguel says. Most of the Mac/Linux types quit coding on Windows a long time ago. Those that still do code on Windows usually do it as their job, which is usually in a position they've had for 5+ years.
There are so many other ways to create cross platform code in a non-MS API. QT and WxWidgets are both quite nice. I do admit that I might consider Mono over Java though. If only just to avoid the "Java trap." Of course this only matters if I want to do some GPL type coding.
Sooner or later MS is going to put out.NETv2, complete with a whole new API and super-duper-mega-active server pages. Of course all of the IIS users (which will remain 99% of the.NET users) will pretty much have to "upgrade." Then the Mono developers get to do it all over again when they could have just started from an existing cross-platform kit or just created their own. Mono seems so much like a ship going way off course whith no one on board willing to question the path.
I hope I'm wrong though. If what the Mono developers say is true than it will be really exciting down the line. I've become pretty skeptical over the years though.. It's already been over a year since Mono 1.0 and I can't name one commonly used cross platform Mono/.NET app. There certainly seems to be way more resources getting dumped into it than results coming out.
Chances are a lot of that voltage was lost by discharging to the floor and such. So it was probably higher previously. That does mean that 30/40,000 is still well short of "spontaneous combustion" though..
How do you figure that things aren't getting worse? We may or may not be heading toward some eminent ecological catastrophe, but it seems fairly obvious that things are getting (at least a little) worse. Both the frequency and severity (as in strength) of hurricanes has risen over the past century. Previous hurricane seasons have been much more active than usual and this year will almost definitely be the worst one on record. How is that not getting worse? Whether the cause be global warming or an increase in the Sun's acitivity, or something else, something is causing the situation to worsen. It seems painfully obvious.
Despite the fact that I have a BS in Computer Science from a top ranked school, I've been turned down for several jobs due to the lack of "Certification." In one instance I was told that I was underqualified due to the lack of "A+ Certification." Somehow my CS degree just didn't prepare me for installing drivers and such.
The way I see it, I probably didn't want those jobs anyway. It comes from ignorance on the part of management. There are still many CIOs that don't really understand IT and certifications are a bit of a security blanket for them. These people also tend to hire individuals with a business management background as their managers rather than IT people (they tend to think of IT people as managerial challenged). In then end I don't want to work under management that has no IT experience, so it's usually just a good indication to move on. Unfortunately this still seems very common, especially for medium sized business and large non-technology corporations.
I guess if that's the type of job you want, then get certifications. Otherwise, just take classes and let your work/knowledge speak for itself. A competent manager should know whether or not the person they're talking to is qualified (and should also know how to write a decent HR request).
I think PHP is great, but I don't think it's quite ready for a robust content management system.
If PHP isn't ready by now, then it never will be ready enough for you.. PHP has matured a great deal. Most of the problems that people blame on PHP are due to web designers developing applications. By web designers I mean people who traditionally designed web pages using somthing like homesite and have little to no coding experience. PHP came along and was straightforward enough for them to slap some dynamic code together. The problem is that without much coding experience, things would get ugly very quick and the solutions were far from scalable. When in the right hands, PHP is highly effective for web programming. I don't think _anything_ is more mature for this task than PHP. One reason for PHP's existense is that Perl was less suited for these tasks. I personally don't feel that JSP or ASP are any further along. Now.. that doesn't necessarily mean PHP is always the best solution, it's just more mature.
Not only that there were some serious PHP security issues.
This is another thing unfairly blamed on PHP. These problems come up because 99% of the PHP developers out there use it as an Apache module. This means the code runs in a trusted environment and is inherintly unsecure. There is a reason why CGI was not set up this way. Unfortunately CGI was way to slow and everyone pretty much had to start using the modules. Now we have FastCGI. It's secure and runs code even faster than Apache modules. Concerned about security? mod_php, mod_perl, mod_py, and even mod_ruby all have MAJOR security problems. Use FastCGI.
I personally prefer using Ruby/Rails + FastCGI over PHP. So, I'm not exactly a PHP fanboy. PHP has it's place though, and I wouldn't just dismiss it because so many people write bad code with it. There's plenty of poor python code out there and there are truckloads of _awful_ perl code.
How exactly do you propose going about feeding all of the starving people in the world? How long will it take for this to happen so we can move on with things? Do you really think that as humans we'll _EVER_ actually accomplish this? Even if we did have the means (and that's unlikely), I actually doubt that people would be willing to sacrifice they're own tax money that goes to programs that they need to feed people in other countrys. Now I'm not saying that's right, but it is true.
Second, the space program spends a few billion a year to explore and research space, the military spends a few HUNDRED billion a year to kill people (and research how to do it better). Now.. I'm not saying we shouldn't have a military, but this is also a fact. We're all killing each other right and left right now and you want to take money away from space to feed starving people?? Maybe we could just stop killing each other and free up the few hundred billion a year, maybe this is a better first step before feeding the hungry of the world.
Re:A bit too enthusiastic IMO..
on
NASA's Shuttle Plans
·
· Score: 2, Informative
The only reason the foam is there is to keep ice from forming on the side of the tank. Since the fuel is cryogenic, it is obviously quite cold and causes large chunks of ice to form on the side of the tank. On launch this ice would simply destroy the Shuttle's tile system, it would be much much worse than the foam. If there is nothing for the ice to damage, then you don't really need the foam anyway. Many rockets have no such foam at all. Ice just builds up on the rocket and sheds during launch. You can see this in a lot of launch footage from the gemeni/apollo days. The ice would fall to the pad as the launch clears the tower. The absence of foam is not likely a significant safety risk, but it is probably still usefull enough to keep. I'm sure it has several side benefits, such as allowing the tank to stay fueled and on the pad for longer periods of time.
It's not extortion or illegal when the chips are sold below market value. Basically, Intel is saying: "If you buy AMD you won't get our DISCOUNT" not: "If you buy AMD you will get charged more than others." Essentially they would charge the same amount, but this would hurt Dell's ability to undercut the prices of its competitors.
Since selling below market price is considered a "gift" under the law, then refusing to do so isn't considered illegal. It may be of questionable morality, but it is legal and happens all the time (unfortunately).
AMD is probably not planning on winning this, but rather just trying to shed light on Intel's business practices and win PR points by pointing fingers. The sad reality is that if the tables were turned AMD would almost certainly do the same thing. Their business practices are equally (if not more) aggressive than Intel's. They have to be.
You seem to contradict yourself by saying that you don't like having OS's install everything in/usr/bin, but then say you wan't/usr/local/ clear of anything used by the OS. Either the distro puts all apps in/usr/bin and keeps/usr/local (and opt) clean or puts core utilities in/usr/bin and apps in/usr/local or/opt.
If you use apt-get or emerge you don't have to worry much about this stuff.
Re:why not use the $ for universal healthcare?
on
Back to Moon in 2015?
·
· Score: 1
why not use the $ for travelling to the moon, instead of wasting it on universal health care?
For someone who is obviously a proud contributor to the FOSS community, it's pretty sad to see him blasting the very work he was a large part of. I usually hate the: "If you don't like it, create a patch" attitude, but for someone that's done so much FOSS coding, I think this motto applies. I use OS X at work and it's great.. but this is just mob mentaility at it's finest. I'm tired of hearing about how much better OS X is than Linux. It's only better on the desktop and it SHOULD be. It's perfomance sucks compared to Linux. It's memory management sucks compared to Linux. It's threading capability sucks compared to Linux. Try running OS X as a server and you will be underwelmed. All Apple does is try to make something that's pretty, usable, and works. Apple would really be doing a piss poor job if it wasn't better at these things.
Linux is better in so many ways. So your stupid sound card doesn't multiplex. Cry some more. If it's so bad, help fix the problem. He obviously has the skillset. This guy just seems to like to b*tch.
Yeah I agree, that's why I refer to Apple as "somewhat" giving. That meaning that they don't hate it (I don't see how they could since their entire OS is based on BSD) but don't really love it. They just take it for what it's worth (to them). Intel actually truly wants open source to succeed because (despite pubic opinion) they don't like being a part of "Wintel." As it is now, Intel's fate is tied to that of M$. That's what they want to get away from.
Despite the hype surrounding Apple, Linux is still the chief competitor to M$. Apple has always (and will always) targeted the niche "premium computer" market. Even the Mini is a premium price for it's performance level. Linux is ideal for the more lucrative low cost market. It also gives them a beating in developing nations, which is about the only way for M$ to expand on the desktop. So it's natural that M$ does all it can do to squash Linux.
I see these preconceptions a lot though. It's sad to see strong supporters of Linux (like Intel and Nvidia) get dogged by open source supporters when they do so much more than their competition. Just because the P4 is terd and still has so much market share doesn't mean Intel hates Linux.
Actually Intel is big supporter of Linux and one of the founders of OSDL. Apple has contributed somewhat to he OSS community as well, so I wouldn't say they're trying to get rid of "linuxers." They probably wouldn't exist anymore without relying on open source.
I don't think the average user will know or care that it runs Linux. I'm sure Sony will set up something that's simple and straightforward to use. This would allow the PS3 to essentially act as a full scale computer. Is this not usefull? You can check your email, do word processing, etc. You could access PS3 game sites and download game expansions or get on forums. Better yet, anyone with a HDD could presumably develop games for the PS3. You're thinking of Linux as an OS tinkering toy, which is the wrong way to think of it in this instance. True, that's what it is for the PS2 and other homebrew console hacks. It's safe to assume that this will be a fully usable OS.
I don't understand why everyone seems to think KDE and Gnome are so unusable. They both seem very usable to me. In fact, I get around more easily in KDE than OS X. OS X is just more "pretty." I prefer KDE to Gnome, but I can certainly use Gnome just fine. Why do so many people buy into the OS X hype? Not that OS X sucks.. not by a long shot. I use it every day and enjoy it (love would be an exagerartion). Really though.. what makes desktop linux so bad? Everybody I know who has tried and taken the time to learn it ends up enjoying it. It's only those not willing to try or don't want to learn something new that say it sucks.
As a user of both Mac and Linux I can tell you that Mac has a severe shortage of OSS when compared to Linux. In terms of OSS apps, it doesn't really have any more than Windows. Trying to get native builds of my fav Linux programs running on my Mac has largely been an excersice in futility. Sure I can _somewhat_ run them via Apple's X Server, but this is _far_ from the quality I experience w/ these apps in Linux. For the most part, there is very little interest in OSS development for the Mac. The Darwin projects move along at a snail's pace (Just check their website, it's a little sad really). Fink is the only one that has any type of momentum, and it's consistently well behind the top Linux distros. It's not their fault, they're just understaffed.. due to the lack of interest.
Even though it is built on OSS, OS X is very much a proprietary OS that is based on the proprietary software model. It is not a completely open environment and thus will _never_ attract as much attention from OSS developers as pure BSD's and Linux. Trying to build OSS on OS X is like working with one hand behind your back. You're not completely blind, as you are with Windows, but still handicapped. The most difficult problem w/ creating OSS on the Mac is it's outdated library set and the lack of a native package manager. Need a newer version of a library that's on OS X to build your Open Source app? Sure you can package it as.app, but then it won't work correctly on Linux or standard BSDs. In the end, most OSS devs choose ease of distribution on Linux and BSD over OS X.
Yeah there are a few great open source apps out for OS X/cocoa (Adium, Cyberduck, etc.), but there aren't that many. Plus they are mostly OS X specific and won't run on anything else, something OSS devs usually like to avoid as much as possible. Even though I use my Mac more often, I tend to do all of my OSS development in Linux because it's simply better suited for that.
Ever try running Open Office on OS X? It's slow and painful. It takes a spinoff like NeoOffice to get it working reasonably. No disrespect to the NeoOffice crew, I admire what they do. In the end though, you'll probably use MS Office.
If you get a Mac, don't expect to get all free software. Expect to pay for it.
Hey all you Gentoo fan boys. I've been using Debian or Red Hat or (whatever distro) for years and I say Gentoo sux. I was up late compiling crap with Gentoo once and said.. "Hey this SUCKS!!" I got soooo tired of searching for pr0n while waitng for that stupid compiler. I really just wanted to play games, but I couldn't reboot into Windows cos it took soooo long to compile. Gentoo users are stupid idiots. I don't know why they want to compile this crap. They should build an LFS system and be real men. Gentoo sucks because apt-get is cool. I don't know why you would even bother using apt-get in Gentoo though, what a waste. I'm just going to run around saying how dumb I think Gentoo is.
(sarcasm)
Ever try downloading a major Java app and run it on an unusual platform? It seems like it rarely works right on anything that isn't Sun or M$. I've tried running several Java apps in Linux just to give up in frustration. They ran fine in Windows. I've also tried running several Java apps on OS X with the same results.
Free or not free has nothing to do with it. It does not seem well tested across platforms. It may "work" but with varying degrees of success. Linux and OS X are much more common than AIX et al and it still is very flakey.
A few of the best Java apps have worked around this, but they have to work around platform specific bugs.
Slashdot Mirror
I've found myself in the same boat. What pulled me in.. consistency across platforms and KDE integration. Opera works basically the same way on Windows, Mac, and Linux. Firefox doesn't.. the Mac build is made to be more Safari-like. This annoys me.. I would actually prefer Fox over Safari if this wasn't the case. Firefox is somewhat handicaped in KDE, lacking any type of reasonable integration. I use OSX at work and Gentoo+KDE at home, so I used Safari and Konqueror in the past. I always used Fox for the rare occasion of windows browsing.
I can understand the KDE problems with Firefox, I would even help w/ QT integration if I had enough free time. The Mac issue is really annoying though.. at least to me. I like to use favicons on the toolbar for my most common sites. I can do this w/ Fox in Windows or Linux, but not on Mac. It sounds odd, but it's a problem for me. Apparantly most people like it that way, but I don't. When Opera came out, I had the chance to use the same browser for both OSX and KDE. It was consistent and integrated well with both (especially with KDE). I can use Favicons the way I like and I actually like the layout more.
A few annoyances remain, but Opera is very nice. I hope they fix some of the rendering issues for version 9. It's very fast but has more rendering bugs than any of the other browsers I've used. That hasn't been enough to make turn away though. It's the only negative for me.
I didn't even notice the amount of controversy this post created. Usually I don't respond to my own post, but it seems like I need to explain my reasoning...
,because they're less familiar with how they should be used.
Hear is why I think pointer arithmetic is bad(tm) by example:
I used to work for a defense contractor. The project was large and had a large amount of legacy C code that needed to be integrated w/ newer C++ code. The legacy code had already been deployed in a number of places for a number of years. Performance and stability were both of top importance. It had to run in real time (at least the C portion did) and had to be completely stable. The legacy code had a large amount of pointer arithmetic. While profiling, we found hundreds of instances where incorrect pointer arithmetic would cause memory leaks and nasty bounds problems. Very few problems were found in the code that used arrays. It took so much time to fix these problems that it added months to the development time of the product. The pointer arithmetic was extremely confusing and we ended up re-writing much of it using arrays. The result was something that was rock stable and ran a few milliseconds slower. The speed difference was not significant and the stability meant that we could be confident that the system would be dependable when delployed. Which was _extremely_ important.
Would you like to fly on an airplane that had software written with excessive pointer arithmetic on it? It wouldn't surprise me if this was the case on some planes, but the thought would be unsettling to me. The reality is that most people get confused when doing pointer arithmetic and make more mistakes. For people that don't get confused and can write perfect code like this, more power to you. I would be worried about adding code like this to any project though, because other project members will most likely not be nearly as good and will make mistakes. That's why I would ask the person to rewrite it. (BTW, I wouldn't be saying "this teh sucks" but "future maintainers may have a hard time understanding this"). I am very big on performance and optimizations, but not at the sake of stability and readability (because less readable code will end up being less stable down the line).
The array version is neither clearer nor less error-prone. And I dare you to prove differently!
It seems like this challenge is issued and then you go on to say it can't be done... If you're saying it can't be proved, I agree. I'm not sure that anyone can "prove" that pointer arithmetic is less clear. This is subjective. Perhaps if pointer arithmetic were the standard way of doing things, it would be easy for everyone to use. Arrays are far more common though, and most people are much more comfortable with them as a result. The reason I think it's more error-prone is coupled to this. In practice, people tend to make more mistakes with them
I wouldn't want to work for you though. Your competitors will write a faster program that uses less memory and you will loose the contract/job.
I'm not sure how often a contract would be lost due to array related performance issues. On the occassions where I've had to fix performance issues, the problem was usually a poorly written algorithm and not anything like "array overhead".
The pointer version is more portable across the chasm of compiler quality. That is, it will run well even on poorer compilers.
It's pretty hard to find hardware not supported by a major compiler these days. GCC works on some amazingly obscure hardware. I suppose in instances where the hardware is extremely rare and doesn't have a decent compiler, pointer arithmetic may be the best option. This would be a very rare circumstance though.
Thinking like this.. Is why we have big, ugly, bloated programs that require overpowered CPUs.
As far as standard arrays causing ex
Imagine if everyone said that back when all we had was assembly... Also, I took a job for PHP/Perl experience and we're now considering moving to Ruby/Rails. I know a lot of people are in the same situation. The fact that it's not in the job postings is actually nice. Do you really want to see "2 years of Ruby on Rails experience"? Being a developer also means that you continue to evaluate available technologies and how they may help you and your organization. Rails is definitely such a technology.
The problem with datagnomes is that there are only so many of them. Most sites, for example only have 1 or 2 database servers and many more web servers. The problem here is that the datagnome population grows at a low rate, whereas the webgnome population seems to grow exponenentially. This makes it easy to overwork the datagnomes. Which is bad for overall gnome productivity. So.. in most cases it's better to make the webgnomes do as much of the logic work as possible. So all the datagnomes have to do is scoop up the data and throw it back to the webgnomes.. they usually stay pretty busy just doing the scooping. If you make them run around too much, then your webgnomes will be sitting around smokin cigs waiting for the datagnomes to get their work done.
These types of opimizations are virtually pointless on modern machines. The increased readability and lower likelihood of programming errors on the array option far outway any speed increase for the pointer option. Plus, as you noticed, the resulting assembly is basically the same. Most likely, both will run at virtually the same speed with modern compilers. Not only would the speed difference be unnoticeable, it would be utterly inconsequential.
IMHO there is no place for pointer arithmetic in modern software. If someone working for me wrote something like the second option, I would ask them to rewrite it.
Yeah that's one thing that really could improve.
Maybe the reason it took two weeks is because they had windows admins set up linux servers. Maybe that's _also_ why it didn't work well. Maybe they need to do some training or hire someone with the right skillsets?
The whole concept of Mono is somewhat confounding to me. Why not just implement something new? Why spend so much time recoding the .NET API?? When I first heard of Mono I thought that it was cool that existing .NET apps could potentially run in Linux. Then I realized that I don't give a ____ about these apps anyway. And I gaurantee I won't bother running some GPhoto type application in Windows. Anyone who uses this is just going to be using Linux anyway. On the server end.. I have to wonder what the ROI would be for porting apps from Windows-IIS/.NET to Linux/Mono.NET. It seems like a complete waste of time. Also, most people that write cross platform apps simply don't care about .NET and won't bother learning it, even if it really is as cool as Miguel says. Most of the Mac/Linux types quit coding on Windows a long time ago. Those that still do code on Windows usually do it as their job, which is usually in a position they've had for 5+ years.
.NETv2, complete with a whole new API and super-duper-mega-active server pages. Of course all of the IIS users (which will remain 99% of the .NET users) will pretty much have to "upgrade." Then the Mono developers get to do it all over again when they could have just started from an existing cross-platform kit or just created their own. Mono seems so much like a ship going way off course whith no one on board willing to question the path.
There are so many other ways to create cross platform code in a non-MS API. QT and WxWidgets are both quite nice. I do admit that I might consider Mono over Java though. If only just to avoid the "Java trap." Of course this only matters if I want to do some GPL type coding.
Sooner or later MS is going to put out
I hope I'm wrong though. If what the Mono developers say is true than it will be really exciting down the line. I've become pretty skeptical over the years though.. It's already been over a year since Mono 1.0 and I can't name one commonly used cross platform Mono/.NET app. There certainly seems to be way more resources getting dumped into it than results coming out.
*prepares for the flames*
Chances are a lot of that voltage was lost by discharging to the floor and such. So it was probably higher previously. That does mean that 30/40,000 is still well short of "spontaneous combustion" though..
How do you figure that things aren't getting worse? We may or may not be heading toward some eminent ecological catastrophe, but it seems fairly obvious that things are getting (at least a little) worse. Both the frequency and severity (as in strength) of hurricanes has risen over the past century. Previous hurricane seasons have been much more active than usual and this year will almost definitely be the worst one on record. How is that not getting worse? Whether the cause be global warming or an increase in the Sun's acitivity, or something else, something is causing the situation to worsen. It seems painfully obvious.
Despite the fact that I have a BS in Computer Science from a top ranked school, I've been turned down for several jobs due to the lack of "Certification." In one instance I was told that I was underqualified due to the lack of "A+ Certification." Somehow my CS degree just didn't prepare me for installing drivers and such.
The way I see it, I probably didn't want those jobs anyway. It comes from ignorance on the part of management. There are still many CIOs that don't really understand IT and certifications are a bit of a security blanket for them. These people also tend to hire individuals with a business management background as their managers rather than IT people (they tend to think of IT people as managerial challenged). In then end I don't want to work under management that has no IT experience, so it's usually just a good indication to move on. Unfortunately this still seems very common, especially for medium sized business and large non-technology corporations.
I guess if that's the type of job you want, then get certifications. Otherwise, just take classes and let your work/knowledge speak for itself. A competent manager should know whether or not the person they're talking to is qualified (and should also know how to write a decent HR request).
I think PHP is great, but I don't think it's quite ready for a robust content management system.
If PHP isn't ready by now, then it never will be ready enough for you.. PHP has matured a great deal. Most of the problems that people blame on PHP are due to web designers developing applications. By web designers I mean people who traditionally designed web pages using somthing like homesite and have little to no coding experience. PHP came along and was straightforward enough for them to slap some dynamic code together. The problem is that without much coding experience, things would get ugly very quick and the solutions were far from scalable. When in the right hands, PHP is highly effective for web programming. I don't think _anything_ is more mature for this task than PHP. One reason for PHP's existense is that Perl was less suited for these tasks. I personally don't feel that JSP or ASP are any further along. Now.. that doesn't necessarily mean PHP is always the best solution, it's just more mature.
Not only that there were some serious PHP security issues.
This is another thing unfairly blamed on PHP. These problems come up because 99% of the PHP developers out there use it as an Apache module. This means the code runs in a trusted environment and is inherintly unsecure. There is a reason why CGI was not set up this way. Unfortunately CGI was way to slow and everyone pretty much had to start using the modules. Now we have FastCGI. It's secure and runs code even faster than Apache modules. Concerned about security? mod_php, mod_perl, mod_py, and even mod_ruby all have MAJOR security problems. Use FastCGI.
I personally prefer using Ruby/Rails + FastCGI over PHP. So, I'm not exactly a PHP fanboy. PHP has it's place though, and I wouldn't just dismiss it because so many people write bad code with it. There's plenty of poor python code out there and there are truckloads of _awful_ perl code.
How exactly do you propose going about feeding all of the starving people in the world? How long will it take for this to happen so we can move on with things? Do you really think that as humans we'll _EVER_ actually accomplish this? Even if we did have the means (and that's unlikely), I actually doubt that people would be willing to sacrifice they're own tax money that goes to programs that they need to feed people in other countrys. Now I'm not saying that's right, but it is true.
Second, the space program spends a few billion a year to explore and research space, the military spends a few HUNDRED billion a year to kill people (and research how to do it better). Now.. I'm not saying we shouldn't have a military, but this is also a fact. We're all killing each other right and left right now and you want to take money away from space to feed starving people?? Maybe we could just stop killing each other and free up the few hundred billion a year, maybe this is a better first step before feeding the hungry of the world.
The only reason the foam is there is to keep ice from forming on the side of the tank. Since the fuel is cryogenic, it is obviously quite cold and causes large chunks of ice to form on the side of the tank. On launch this ice would simply destroy the Shuttle's tile system, it would be much much worse than the foam. If there is nothing for the ice to damage, then you don't really need the foam anyway. Many rockets have no such foam at all. Ice just builds up on the rocket and sheds during launch. You can see this in a lot of launch footage from the gemeni/apollo days. The ice would fall to the pad as the launch clears the tower. The absence of foam is not likely a significant safety risk, but it is probably still usefull enough to keep. I'm sure it has several side benefits, such as allowing the tank to stay fueled and on the pad for longer periods of time.
It's not extortion or illegal when the chips are sold below market value. Basically, Intel is saying: "If you buy AMD you won't get our DISCOUNT" not: "If you buy AMD you will get charged more than others." Essentially they would charge the same amount, but this would hurt Dell's ability to undercut the prices of its competitors.
Since selling below market price is considered a "gift" under the law, then refusing to do so isn't considered illegal. It may be of questionable morality, but it is legal and happens all the time (unfortunately).
AMD is probably not planning on winning this, but rather just trying to shed light on Intel's business practices and win PR points by pointing fingers. The sad reality is that if the tables were turned AMD would almost certainly do the same thing. Their business practices are equally (if not more) aggressive than Intel's. They have to be.
You seem to contradict yourself by saying that you don't like having OS's install everything in /usr/bin, but then say you wan't /usr/local/ clear of anything used by the OS. Either the distro puts all apps in /usr/bin and keeps /usr/local (and opt) clean or puts core utilities in /usr/bin and apps in /usr/local or /opt.
If you use apt-get or emerge you don't have to worry much about this stuff.
why not use the $ for travelling to the moon, instead of wasting it on universal health care?
For someone who is obviously a proud contributor to the FOSS community, it's pretty sad to see him blasting the very work he was a large part of. I usually hate the: "If you don't like it, create a patch" attitude, but for someone that's done so much FOSS coding, I think this motto applies. I use OS X at work and it's great.. but this is just mob mentaility at it's finest. I'm tired of hearing about how much better OS X is than Linux. It's only better on the desktop and it SHOULD be. It's perfomance sucks compared to Linux. It's memory management sucks compared to Linux. It's threading capability sucks compared to Linux. Try running OS X as a server and you will be underwelmed. All Apple does is try to make something that's pretty, usable, and works. Apple would really be doing a piss poor job if it wasn't better at these things.
Linux is better in so many ways. So your stupid sound card doesn't multiplex. Cry some more. If it's so bad, help fix the problem. He obviously has the skillset. This guy just seems to like to b*tch.
Yeah I agree, that's why I refer to Apple as "somewhat" giving. That meaning that they don't hate it (I don't see how they could since their entire OS is based on BSD) but don't really love it. They just take it for what it's worth (to them). Intel actually truly wants open source to succeed because (despite pubic opinion) they don't like being a part of "Wintel." As it is now, Intel's fate is tied to that of M$. That's what they want to get away from.
Despite the hype surrounding Apple, Linux is still the chief competitor to M$. Apple has always (and will always) targeted the niche "premium computer" market. Even the Mini is a premium price for it's performance level. Linux is ideal for the more lucrative low cost market. It also gives them a beating in developing nations, which is about the only way for M$ to expand on the desktop. So it's natural that M$ does all it can do to squash Linux.
I see these preconceptions a lot though. It's sad to see strong supporters of Linux (like Intel and Nvidia) get dogged by open source supporters when they do so much more than their competition. Just because the P4 is terd and still has so much market share doesn't mean Intel hates Linux.
Actually Intel is big supporter of Linux and one of the founders of OSDL. Apple has contributed somewhat to he OSS community as well, so I wouldn't say they're trying to get rid of "linuxers." They probably wouldn't exist anymore without relying on open source.
I don't think the average user will know or care that it runs Linux. I'm sure Sony will set up something that's simple and straightforward to use. This would allow the PS3 to essentially act as a full scale computer. Is this not usefull? You can check your email, do word processing, etc. You could access PS3 game sites and download game expansions or get on forums. Better yet, anyone with a HDD could presumably develop games for the PS3. You're thinking of Linux as an OS tinkering toy, which is the wrong way to think of it in this instance. True, that's what it is for the PS2 and other homebrew console hacks. It's safe to assume that this will be a fully usable OS.
I don't understand why everyone seems to think KDE and Gnome are so unusable. They both seem very usable to me. In fact, I get around more easily in KDE than OS X. OS X is just more "pretty." I prefer KDE to Gnome, but I can certainly use Gnome just fine. Why do so many people buy into the OS X hype? Not that OS X sucks.. not by a long shot. I use it every day and enjoy it (love would be an exagerartion). Really though.. what makes desktop linux so bad? Everybody I know who has tried and taken the time to learn it ends up enjoying it. It's only those not willing to try or don't want to learn something new that say it sucks.
This is absolutely not true.
.app, but then it won't work correctly on Linux or standard BSDs. In the end, most OSS devs choose ease of distribution on Linux and BSD over OS X.
As a user of both Mac and Linux I can tell you that Mac has a severe shortage of OSS when compared to Linux. In terms of OSS apps, it doesn't really have any more than Windows. Trying to get native builds of my fav Linux programs running on my Mac has largely been an excersice in futility. Sure I can _somewhat_ run them via Apple's X Server, but this is _far_ from the quality I experience w/ these apps in Linux. For the most part, there is very little interest in OSS development for the Mac. The Darwin projects move along at a snail's pace (Just check their website, it's a little sad really). Fink is the only one that has any type of momentum, and it's consistently well behind the top Linux distros. It's not their fault, they're just understaffed.. due to the lack of interest.
Even though it is built on OSS, OS X is very much a proprietary OS that is based on the proprietary software model. It is not a completely open environment and thus will _never_ attract as much attention from OSS developers as pure BSD's and Linux. Trying to build OSS on OS X is like working with one hand behind your back. You're not completely blind, as you are with Windows, but still handicapped. The most difficult problem w/ creating OSS on the Mac is it's outdated library set and the lack of a native package manager. Need a newer version of a library that's on OS X to build your Open Source app? Sure you can package it as
Yeah there are a few great open source apps out for OS X/cocoa (Adium, Cyberduck, etc.), but there aren't that many. Plus they are mostly OS X specific and won't run on anything else, something OSS devs usually like to avoid as much as possible. Even though I use my Mac more often, I tend to do all of my OSS development in Linux because it's simply better suited for that.
Ever try running Open Office on OS X? It's slow and painful. It takes a spinoff like NeoOffice to get it working reasonably. No disrespect to the NeoOffice crew, I admire what they do. In the end though, you'll probably use MS Office.
If you get a Mac, don't expect to get all free software. Expect to pay for it.
Hey all you Gentoo fan boys. I've been using Debian or Red Hat or (whatever distro) for years and I say Gentoo sux. I was up late compiling crap with Gentoo once and said.. "Hey this SUCKS!!" I got soooo tired of searching for pr0n while waitng for that stupid compiler. I really just wanted to play games, but I couldn't reboot into Windows cos it took soooo long to compile. Gentoo users are stupid idiots. I don't know why they want to compile this crap. They should build an LFS system and be real men. Gentoo sucks because apt-get is cool. I don't know why you would even bother using apt-get in Gentoo though, what a waste. I'm just going to run around saying how dumb I think Gentoo is. (sarcasm)
Ever try downloading a major Java app and run it on an unusual platform? It seems like it rarely works right on anything that isn't Sun or M$. I've tried running several Java apps in Linux just to give up in frustration. They ran fine in Windows. I've also tried running several Java apps on OS X with the same results.
Free or not free has nothing to do with it. It does not seem well tested across platforms. It may "work" but with varying degrees of success. Linux and OS X are much more common than AIX et al and it still is very flakey.
A few of the best Java apps have worked around this, but they have to work around platform specific bugs.