And what is the "political agenda" you are going to claim they are advancing?
Oh, I dunno, how about all the curtailments of civil liberties they've been pushing since 9/11 in the name of preventing terrorism - things like the ID cards and database (which are pretty much universally agreed to be totally useless against terrorism), biometrics, etc.
Are you so cynical that you don't think they might have an 'alert system' just to alert the public?
If the alert system was actually there to alert the public, it would've gone up _before_ the arrests were made and gone down again a bit after the arrests were made (you know, when the threat has been reduced by arresting a bunch of evil terrorists). As it was, they only bothered to 'alert the public' after the matter, and at that point they put it up to it's highest level even though they said they had no evidence to suggest there would be any further attacks.
* critical - an attack is expected imminently
* severe - an attack is highly likely
* substantial - an attack is a strong possibility
* moderate - an attack is possible but not likely
* low - an attack is unlikely
Seems to me that 'expected imminently' means you know that an attack is going to happen. The intelligence services are saying that they have no information suggesting any further attack but they are implementing the extra security as a precaution just incase something they don't know about happens. Wouldn't that constitute 'severe' or 'substantial'?
If you're going to use the highest alert level for a _precautionary_ measure when you have nothing to suggest any further attacks, WTF are you going to use when you know damned well someone's going to try something?
I'm a bit confused about the 'alert levels'... They claim that the attack they have stopped wasn't going to happen today, are aware of no other impending attack and that the security measures are just precautionary and there is nothing to worry about. Yet they have used the highest alert level... Shouldn't that kinda be reserved for when they know an attack is about to happen?
Maybe they should just face the truth and rename the 'alert level' system to 'scare-the-shit-out-of-the-public-to-push-our-poli tical-agenda-o-meter'
I guess this just goes to show that you should be using something like Torpark even when merely conducting an online search.
Whilest protecting your privacy does, on the surface, seem like a good thing, I wonder if it might count against you if you were ever suspected of a crime. We've already seen 'he has some encrypted data' used as evidence (even though the contents of the encrypted file weren't known) in one successful conviction, I suspect 'he's using privacy protection software called Tor' may go down the same way.
Remember, only people who have something to hide care about protecting their privacy.:)
Uh, people still use meta tags? Google certainly doesn't pay attention to them - allowing the website owner to specify keywords is far too unreliable, far better to work them out from the content.
Well, it goes into firmware then, which can also be closed or open source!
But verifying that the compiled firmware that's factory-loaded into the EEPROM of a microcontroller matches the source code that's been distributed is a problem. The device may not let you read the firmware back out of the EEPROM, and even if it does - how do you know it's really sending you the code that it's running and not just telling you what you want to hear?
It would be nice if they had 50% bigger pixels at home, and I bet retirees with the money would readily spend the $$$
If your windowing environment did proper font scaling (i.e. it renders a font at a specific _physical_ size rather than a number of pixels, and allows you to configure the top-level scaling so you can make everything larger/smaller) then you would be best off running in the highest screen resolution they can get.
If you've got bad eyesight you get large-type books - not super-pixelated books.
Personally I'm waiting for 300+ DPI displays to appear because then we could stop caring about the resolution. Although I hesitate to consider the bandwidth requirements for a reasonable size display running in that resolution - a 500x280mm screen at 300dpi would be about 20 megapixels. Refresh that at 50Hz and you've got 24 Gigabits per second of data (24 bits per pixel).
A heavy weight OS such as vista is not good for gaming, all the extra memory, cpu and gpu time consumed by the OS is no longer available for the game.
You know, *exactly* the same thing was said by most people with a clue when Microsoft originally invented DirectX and said that Win95 was suitable for games. And they even bundled their wonderful "Hover" game as a proof of concept, which really just proved that yes, the performance of games under Windows really was as bad as expected.
But you know what? It turned out that people were willing to take the hit just so they could run their game without rebooting into DOS. Expect the same thing to happen with Vista.
The IT infrastructure supporting the system will be down more often than up and the costs will spiral in the tens of billions.
Whilest it would provide a fair amount of amusement to me to watch the government screw up an IT system yet again (or rather, EDS or other complete idiots they decide to contract who have shown on numerous previous occasions to be incapable of running an abacus, let alone a national computer database+network), I can think of better things for my taxes to go on.
And you can guarantee that before there is *any* chance of the system being scrapped, it'll have to have been kludged and band-aided (expensively) a few hundred times over the course of several decades.
I might be slightly more inclined to spend that kind of tax money if the result is the people responsible for building a flawed security hole ridden system ending up in jail, but that's never going to happen - they'll just take a big payoff and wait until the next month when the govenment contracts them to screw up another project for an extortionate sum of money.
Forgive my cynicisim, but I've seen the same companies being contracted and screwing up in fairly major ways time and time again - when will the government learn to blacklist companies who cause major screwups or cost overruns?
Hmm...while I'd guess most schools firewalls (if they have them) would probably by default in most cases have most ports open.
Wrong. I used to do network security for schools - they are *really* paranoid (mainly coz if little Johnny's parents find out he's surfing porn at school it generates a helluva lot of bad press for the school in question). Most schools are rather overzealous with the firewall rules.
I'd guess that most any school, like a business would have ports 80, 443
Most schools use web proxies rather than just allowing the traffic straight out. If they do allow HTTPS it'll be via the proxy. Admittedly it isn't rocket science to hack your SSH client to do a CONNECT through the HTTPS proxy, but it's not quite as easy as just firing up OpenSSH. Also, quite a lot of schools seem to only allow HTTPS connections to known trusted sites because of the problem of proxies delivering banned content over HTTPS.
I kinda doubt the schools can afford a network admin who knows what he's doing...
True, school IT staff generally aren't _that_ clued up (although it seems to be getting better), but they use tools which make it easy for them to be overzealous at blocking stuff and they often contract in third party companies to keep their networks secure - this is basically what my old employer did, and a reasonable proportion of us who worked there _did_ have a good amount of clue when it comes to setting up, securing and maintaining the security of networks.
And it's the STATE that says we have to use this internet service
This sort of crazyness is pretty much the norm here in the UK too. The schools get a grant from the government to pay for their internet connection. The local education authorities (LEA) run their own WAN (or rather, they contract someone like RM to do it). These WANs are almost always a complete mess and clearly designed by people who have no business designing networks. Often providing nothing more than basic web and email access with all sorts of crazy routing schemes that take a rocket scientist to understand (ok, a rocket scientist who's particularly good at networking). Usually also with a good few layers of NAT thrown in for good measure.
The schools do of course get a choice as to whether to use the LEA's crazy WAN or sort out their own Internet connection. Now, here's the catch - in a large number of areas, the LEA mandate that each school pays them their "internet grant" to cover the cost of running the WAN, even if the school opts out of actually using it. This basically means that the schools can't afford to use a third party internet connection since they've already been forced to hand over their grant.
Luckilly in some areas the LEAs only require the schools to pay them if the schools are using the LEA supplied WAN. This model is much better because a large chunk of the schools have enough clue to organise their own connections instead, which are of course a fraction of the price and perform much better. And most importantly allow the school to run their network how they want. For example, the schools can do things like provide remote access to email, etc.
My question really, is what fools decided that it would be a good idea to set up a bunch of massive, badly designed and overpriced WANs rather than just using commercially available ISPs.
But if the users aren't contributors, and the developers aren't either, why should *I* care?
Because increasing the platform's popularity will help things you _do_ care about happen. For example, hardware manufacturers aren't interested in supporting a platform hardly anyone uses, but if it's significantly popular then that will change. Support for hardware benefits everyone using Linux. (I'm not going to go into the evilness of closed drivers that this may promote though).
What I don't understand is why people like this expect the "open source community" to port their apps **for free**
They do? I can't say I've seen any commercial vendors trying to get people to port their software for free.
what kind of habits would they have now if the RIAA never started taking action? After all, free beer is free beer.
I would say there is _more_ chance of me infringing copyright now than there used to be, coz now I might have to download illegal copies of things that I can't buy a CD for (because they have started selling corrupt optical discs instead of CDs).
I don't think you've ever _really_ used ClearCase. SVN is great for fairly straightforward projects, but once you need serious merge/branching capabilities it falls short.
I've had to "really use" ClearCase for a long time - it's one of the worst revision control systems I've ever had the misfortune of using. It doesn't even do atomic commits FFS!
dynamic views (which can sometimes be VERY useful)
Dynamic views _can_ be useful in *very* large and complex projects, but they can also make using it for simple stuff an absolute ordeal. And also in the many years I've had of using CVS and SVN I've never found that I absolutely needed dynamic views - there's always another way of doing it, and it's usually not hard.
IMHO the cons far, far outweigh the pros and the over-complexity makes you want to keep as much _out_ of the revision control system as possible, which is clearly the wrong way to do things. On projects that use SVN, the push is to get as much into the revision control system as possible, and that's a good thing (and SVN makes it extremely easy to read and update anything in the repository without fiddling around with views, et-al)
I'd like to know what the 10-years-ahead OSS counterparts to the following software are: Oracle, Opera, Photoshop, AutoCAD. This was meant to show that yelling "OSS is always teh b3s7es7" is as dumb as assuming that free stuff must be worthless.
Please go and learn the meaning of the word "most" and then reread my reply. Where did I say that all FOSS software is "teh b3s7es7" (or even the best, for those of us who actually have a grasp of the English language)? If you actually bother to read what I said, you'll see that I said that in many cases (note: I didn't say all cases) FOSS software is better but gets dismissed.
In any case, my point was not really about closed vs. open, it was simply that business decisions seem to rarely be made based on suitability for the job in hand. This comment applies whether you are talking about a choice made between a closed solution and an open solution or between two competing closed solutions.
And to answer your question, of the list of software you provided I'd suggest that FireFox is currently (and has always been) slightly ahead of Opera in terms of functionality and user friendlyness. However, I'm aware that this is very much a personal choice and there is no clear-cut "better" - maybe Opera is better for you but FireFox is certainly better for me. And that's kind of what I was talking about - I choose FireFox because it is the most suitable for the job I want to do.
Another example: IBM's ClearCase is most certainly about 10 years behind the FOSS revision control systems such as Subversion. Infact, I'd say that ClearCase is a lot worse than CVS. That is definately not to say that there aren't other closed solutions that are better still. So I would ask - what would be the reason for buying ClearCase? It clearly wouldn't be based on functionality, and it clearly wouldn't be based on price.
I don't particularly care whether they're OSS or not.
Clearly given a toss-up between two pieces of software you should consider what is most suitable for the job in hand and the price. There is no point in paying a huge amount of money for a commercial solution if it has little/no signifnicant advantage over the free solution. Similarly there is no point saving money and going with the free solution if it's wholey unsuitable for the job.
Why can't we judge the merits of software on the basis of its usefulness, effeciency, manageability, security, and support level offered with it rather than by looking at its skin color?
I'm sorry, isn't that what I was saying needed to (but usually doesn't) happen? Please go re-read my original post and come back when you can actually make a sensible comment.
the end result will be that an entire suite of FOSS software will exist for the platform by the time commercial interests start noticing the market.
Whilest I will always to choose open solutions over closed ones where possible, I'm not convinced that having the entire software suite implemented before the closed source people take notice is necessarilly a good thing (if that were even possible). The closed software producers are often large corporations with a lot of marketting muscle and Linux may well benefit from them promoting their Linux versions of their software (and thus promoting the whole OS).
I mean, lets that a theoretical example: Lets say Adobe ported PhotoShop to Linux. They put a bit of marketting behind it and a large chunk of PhotoShop users migrate over to Linux. (Ok, I guess this isn't necessarilly the best example since they're pretty attached to their Macs, but bear with me). A migration to Linux like this would give it quite a boost in the desktop market.
Conversely, let's just imagine that The GIMP gets as good as (or better than) PhotoShop before Adobe start caring about Linux. So now they have a viable free (as in beer) competetor to their rather expensive product. Are they actually going to want to promote Linux (which usually ships as standard with The GIMP)? Of course not, they're going to want to use their marketting weight to push people away from Linux in the hope that they can keep as many people as possible from discovering that they can get something as good as PhotoShop for free.
Yes, ok, so The GIMP is available for other platforms, but people are more likley to discover it if it's already installed when they get their computer.
Given time, any company wishing to compete in the Linux market will have to produce software which is significantly better than the established FOSS tools, and that has to be good for us computer users.
It's worth noting that (in my experience) most commercial decisions are not based on "what's best for the job" or even "what gives the most bang for our buck". I've lost count of the number of times employers have forced me to use some very expensive piece of software that really is nowhere near as good at the job as some FOSS software. In most of these cases, the expensive commercial software is a good 10 years behind the free equivalent. Many of these purchasing decisions seem to basically just be made on the "noone ever got fired for buying IBM" premise (replace "IBM" with any large corporation who has been selling expensive software for a long time).
Totally impervious to war, as resistant to acts of God as one can get
I'm sure hosting your website from inside a mountain would be wonderful - your webservers would be totally impervious to an EMP.... sadly your upstream peers would not be so fortunate so whilest your webserver would stay up it would have no connection to the rest of the world...
In my reference IE, specifically loads the HTML engine COM DLLs into its own 'isolated' process. This is done for a couple of reasons, but the main one being security.
I really can't see what security benefit there could possibly be - executable memory is read-only, so sharing it between processes is not a security risk. Isolating code like that is simply a needlessly inefficient use of resources.
as the versioning of these 3rd Party DLLs (even some from MS, but still considered 3rd party to the OS) caused problems because if Version 2.0 of DLL was loaded, and another application needed Version 3.0 the mapping of the DLL would cause problems.
Under Linux the versioning doesn't come into play at all at this point - both processes separately try to mmap() the file, but the kernel will realise it is the same *file* (it has the same inode number) and therefore will share the memory. The kernel has no concept of a global symbol table of sharable symbols - the only thing the kernel knows (or indeed, needs to know) is that there are 2 processes trying to map the same lump of read-only data into memory.
The idea of sharing code based on a vendor-specified version number seems very flawed, whereas if you just base the sharing off the inode number you really can't go wrong.
Learn how a process operates on Windows or goto www.microsoft.com and see how the IE process is isolated just LIKE any other application
If Windows doesn't automatically share libraries between applications then it's a worse operating system than I originally thought. *Every* other modern OS shares dynamically linked code between applications that use it, to do otherwise would be woefully inefficient (both in memory usage and startup time).
Let's look at how Linux handles shared libs (this is about the same as any other modern OS):
You ask the OS to start an executable
The OS mmap()s the executable into memory and looks at what libs it needs to link against
Those libs get mmap()ed into memory, all the symbols get resolved and execution starts
You now ask the OS to start different executable
The OS mmap()s the executable into memory and looks at what libs it needs
Those libs get mmap()ed... but wait, some of them are already mapped into memory because the first executable is using them, so rather than loading a new copy, mmap() simply references the address of the existing mapped data *and that data is shared*
Since the executable code is read-only, there is no security problem with allowing both processes to access the data. (mmap() basically tells the kernel to treat the file a bit like swap space).
So is it simply that the US comes out on top because we have so damned many computers - as opposed to other nations where they're sometimes uncommon in households and people use internet cafes?
ISTR I saw some statistics a while ago suggesting that the UK had a far higher DSL/cable connections to people ratio than most other countries (I think even more than the US). Yet the UK is pretty low down on the list of spammers. Admittedly the UK population is lower than the US population though - they really need to adjust those figures into "spams per citizen" or "spams per internet connection" to make them meaningful enough to draw conclusions about user cluefulness.
Customers get very irate when they are disconnected, sandboxed, and refused further service until they run (free) anti-virus software on their computer.
In other news, drivers get very irate when they aren't allowed to drive their unsafe car on the road until it's been fixed. However, banning people with unsafe cars makes everyone else safer, so is a Good Thing. Same with infected computers. If a computer is actively attacking other systems then drop it's connection ASAP - this is good for two reasons:
1. It stops the infected system from doing any more damage to any other systems/people (this may be relaying spam, DDoSing someone, trying to infect other systems with a worm, running a phishing site, etc). 2. If someone loses their whole connection every time they get infected they might actually start giving a damn about their system's security.
that causes tremendous expense for them in terms of customer calls.
If all the ISPs started taking these measures then it would surely *reduce* the number of support calls since the number of infected systems would be reduced. Sure, there'll be a short term peak in the number of support calls but the long term picture is much better. Sadly, most businesses these days only seem to care about the short term bottom-line.
Yet more evidence that the universe is just a gigantic computer simulation.
It could be argued that a lot of the weirdness in quantum mechanics is down to programming shortcuts for efficiency reasons. E.g. look at things like the double-slit experiment - maybe it's less expensive to simulate a wave function than a particle and only collapse the wave function when it's measured. Just a thought.
In any case, even if the universe is a giant computer simulation, does that make it any less "real" (whatever "real" means)? And there's no way we'd ever know one way or another - the best we can do is discover the rules that govern the workings of the internals of the universe, not the externals.
Slashdot Mirror
And what is the "political agenda" you are going to claim they are advancing?
Oh, I dunno, how about all the curtailments of civil liberties they've been pushing since 9/11 in the name of preventing terrorism - things like the ID cards and database (which are pretty much universally agreed to be totally useless against terrorism), biometrics, etc.
Are you so cynical that you don't think they might have an 'alert system' just to alert the public?
If the alert system was actually there to alert the public, it would've gone up _before_ the arrests were made and gone down again a bit after the arrests were made (you know, when the threat has been reduced by arresting a bunch of evil terrorists). As it was, they only bothered to 'alert the public' after the matter, and at that point they put it up to it's highest level even though they said they had no evidence to suggest there would be any further attacks.
* critical - an attack is expected imminently
* severe - an attack is highly likely
* substantial - an attack is a strong possibility
* moderate - an attack is possible but not likely
* low - an attack is unlikely
Seems to me that 'expected imminently' means you know that an attack is going to happen. The intelligence services are saying that they have no information suggesting any further attack but they are implementing the extra security as a precaution just incase something they don't know about happens. Wouldn't that constitute 'severe' or 'substantial'?
If you're going to use the highest alert level for a _precautionary_ measure when you have nothing to suggest any further attacks, WTF are you going to use when you know damned well someone's going to try something?
I'm a bit confused about the 'alert levels'... They claim that the attack they have stopped wasn't going to happen today, are aware of no other impending attack and that the security measures are just precautionary and there is nothing to worry about. Yet they have used the highest alert level... Shouldn't that kinda be reserved for when they know an attack is about to happen?
i tical-agenda-o-meter'
Maybe they should just face the truth and rename the 'alert level' system to 'scare-the-shit-out-of-the-public-to-push-our-pol
I guess this just goes to show that you should be using something like Torpark even when merely conducting an online search.
:)
Whilest protecting your privacy does, on the surface, seem like a good thing, I wonder if it might count against you if you were ever suspected of a crime. We've already seen 'he has some encrypted data' used as evidence (even though the contents of the encrypted file weren't known) in one successful conviction, I suspect 'he's using privacy protection software called Tor' may go down the same way.
Remember, only people who have something to hide care about protecting their privacy.
properly using meta tags
Uh, people still use meta tags? Google certainly doesn't pay attention to them - allowing the website owner to specify keywords is far too unreliable, far better to work them out from the content.
Well, it goes into firmware then, which can also be closed or open source!
But verifying that the compiled firmware that's factory-loaded into the EEPROM of a microcontroller matches the source code that's been distributed is a problem. The device may not let you read the firmware back out of the EEPROM, and even if it does - how do you know it's really sending you the code that it's running and not just telling you what you want to hear?
It would be nice if they had 50% bigger pixels at home, and I bet retirees with the money would readily spend the $$$
If your windowing environment did proper font scaling (i.e. it renders a font at a specific _physical_ size rather than a number of pixels, and allows you to configure the top-level scaling so you can make everything larger/smaller) then you would be best off running in the highest screen resolution they can get.
If you've got bad eyesight you get large-type books - not super-pixelated books.
Personally I'm waiting for 300+ DPI displays to appear because then we could stop caring about the resolution. Although I hesitate to consider the bandwidth requirements for a reasonable size display running in that resolution - a 500x280mm screen at 300dpi would be about 20 megapixels. Refresh that at 50Hz and you've got 24 Gigabits per second of data (24 bits per pixel).
A heavy weight OS such as vista is not good for gaming, all the extra memory, cpu and gpu time consumed by the OS is no longer available for the game.
You know, *exactly* the same thing was said by most people with a clue when Microsoft originally invented DirectX and said that Win95 was suitable for games. And they even bundled their wonderful "Hover" game as a proof of concept, which really just proved that yes, the performance of games under Windows really was as bad as expected.
But you know what? It turned out that people were willing to take the hit just so they could run their game without rebooting into DOS. Expect the same thing to happen with Vista.
The IT infrastructure supporting the system will be down more often than up and the costs will spiral in the tens of billions.
Whilest it would provide a fair amount of amusement to me to watch the government screw up an IT system yet again (or rather, EDS or other complete idiots they decide to contract who have shown on numerous previous occasions to be incapable of running an abacus, let alone a national computer database+network), I can think of better things for my taxes to go on.
And you can guarantee that before there is *any* chance of the system being scrapped, it'll have to have been kludged and band-aided (expensively) a few hundred times over the course of several decades.
I might be slightly more inclined to spend that kind of tax money if the result is the people responsible for building a flawed security hole ridden system ending up in jail, but that's never going to happen - they'll just take a big payoff and wait until the next month when the govenment contracts them to screw up another project for an extortionate sum of money.
Forgive my cynicisim, but I've seen the same companies being contracted and screwing up in fairly major ways time and time again - when will the government learn to blacklist companies who cause major screwups or cost overruns?
Hmm...while I'd guess most schools firewalls (if they have them) would probably by default in most cases have most ports open.
Wrong. I used to do network security for schools - they are *really* paranoid (mainly coz if little Johnny's parents find out he's surfing porn at school it generates a helluva lot of bad press for the school in question). Most schools are rather overzealous with the firewall rules.
I'd guess that most any school, like a business would have ports 80, 443
Most schools use web proxies rather than just allowing the traffic straight out. If they do allow HTTPS it'll be via the proxy. Admittedly it isn't rocket science to hack your SSH client to do a CONNECT through the HTTPS proxy, but it's not quite as easy as just firing up OpenSSH. Also, quite a lot of schools seem to only allow HTTPS connections to known trusted sites because of the problem of proxies delivering banned content over HTTPS.
I kinda doubt the schools can afford a network admin who knows what he's doing...
True, school IT staff generally aren't _that_ clued up (although it seems to be getting better), but they use tools which make it easy for them to be overzealous at blocking stuff and they often contract in third party companies to keep their networks secure - this is basically what my old employer did, and a reasonable proportion of us who worked there _did_ have a good amount of clue when it comes to setting up, securing and maintaining the security of networks.
And it's the STATE that says we have to use this internet service
This sort of crazyness is pretty much the norm here in the UK too. The schools get a grant from the government to pay for their internet connection. The local education authorities (LEA) run their own WAN (or rather, they contract someone like RM to do it). These WANs are almost always a complete mess and clearly designed by people who have no business designing networks. Often providing nothing more than basic web and email access with all sorts of crazy routing schemes that take a rocket scientist to understand (ok, a rocket scientist who's particularly good at networking). Usually also with a good few layers of NAT thrown in for good measure.
The schools do of course get a choice as to whether to use the LEA's crazy WAN or sort out their own Internet connection. Now, here's the catch - in a large number of areas, the LEA mandate that each school pays them their "internet grant" to cover the cost of running the WAN, even if the school opts out of actually using it. This basically means that the schools can't afford to use a third party internet connection since they've already been forced to hand over their grant.
Luckilly in some areas the LEAs only require the schools to pay them if the schools are using the LEA supplied WAN. This model is much better because a large chunk of the schools have enough clue to organise their own connections instead, which are of course a fraction of the price and perform much better. And most importantly allow the school to run their network how they want. For example, the schools can do things like provide remote access to email, etc.
My question really, is what fools decided that it would be a good idea to set up a bunch of massive, badly designed and overpriced WANs rather than just using commercially available ISPs.
Hint for Kids: Read about SSH / VPN
:)
I think you'd be hard pushed to find a school that allows SSH, ESP, AH, GRE, etc through their firewall...
their curious minds would want them to check if they can now access porn this way too
Noone seems to have considered that maybe porn is an important educational resource.
But if the users aren't contributors, and the developers aren't either, why should *I* care?
Because increasing the platform's popularity will help things you _do_ care about happen. For example, hardware manufacturers aren't interested in supporting a platform hardly anyone uses, but if it's significantly popular then that will change. Support for hardware benefits everyone using Linux. (I'm not going to go into the evilness of closed drivers that this may promote though).
What I don't understand is why people like this expect the "open source community" to port their apps **for free**
They do? I can't say I've seen any commercial vendors trying to get people to port their software for free.
what kind of habits would they have now if the RIAA never started taking action? After all, free beer is free beer.
I would say there is _more_ chance of me infringing copyright now than there used to be, coz now I might have to download illegal copies of things that I can't buy a CD for (because they have started selling corrupt optical discs instead of CDs).
I don't think you've ever _really_ used ClearCase. SVN is great for fairly straightforward projects, but once you need serious merge/branching capabilities it falls short.
I've had to "really use" ClearCase for a long time - it's one of the worst revision control systems I've ever had the misfortune of using. It doesn't even do atomic commits FFS!
dynamic views (which can sometimes be VERY useful)
Dynamic views _can_ be useful in *very* large and complex projects, but they can also make using it for simple stuff an absolute ordeal. And also in the many years I've had of using CVS and SVN I've never found that I absolutely needed dynamic views - there's always another way of doing it, and it's usually not hard.
IMHO the cons far, far outweigh the pros and the over-complexity makes you want to keep as much _out_ of the revision control system as possible, which is clearly the wrong way to do things. On projects that use SVN, the push is to get as much into the revision control system as possible, and that's a good thing (and SVN makes it extremely easy to read and update anything in the repository without fiddling around with views, et-al)
I'd like to know what the 10-years-ahead OSS counterparts to the following software are: Oracle, Opera, Photoshop, AutoCAD.
This was meant to show that yelling "OSS is always teh b3s7es7" is as dumb as assuming that free stuff must be worthless.
Please go and learn the meaning of the word "most" and then reread my reply. Where did I say that all FOSS software is "teh b3s7es7" (or even the best, for those of us who actually have a grasp of the English language)? If you actually bother to read what I said, you'll see that I said that in many cases (note: I didn't say all cases) FOSS software is better but gets dismissed.
In any case, my point was not really about closed vs. open, it was simply that business decisions seem to rarely be made based on suitability for the job in hand. This comment applies whether you are talking about a choice made between a closed solution and an open solution or between two competing closed solutions.
And to answer your question, of the list of software you provided I'd suggest that FireFox is currently (and has always been) slightly ahead of Opera in terms of functionality and user friendlyness. However, I'm aware that this is very much a personal choice and there is no clear-cut "better" - maybe Opera is better for you but FireFox is certainly better for me . And that's kind of what I was talking about - I choose FireFox because it is the most suitable for the job I want to do.
Another example: IBM's ClearCase is most certainly about 10 years behind the FOSS revision control systems such as Subversion. Infact, I'd say that ClearCase is a lot worse than CVS. That is definately not to say that there aren't other closed solutions that are better still. So I would ask - what would be the reason for buying ClearCase? It clearly wouldn't be based on functionality, and it clearly wouldn't be based on price.
I don't particularly care whether they're OSS or not.
Clearly given a toss-up between two pieces of software you should consider what is most suitable for the job in hand and the price. There is no point in paying a huge amount of money for a commercial solution if it has little/no signifnicant advantage over the free solution. Similarly there is no point saving money and going with the free solution if it's wholey unsuitable for the job.
Why can't we judge the merits of software on the basis of its usefulness, effeciency, manageability, security, and support level offered with it rather than by looking at its skin color?
I'm sorry, isn't that what I was saying needed to (but usually doesn't) happen? Please go re-read my original post and come back when you can actually make a sensible comment.
the end result will be that an entire suite of FOSS software will exist for the platform by the time commercial interests start noticing the market.
Whilest I will always to choose open solutions over closed ones where possible, I'm not convinced that having the entire software suite implemented before the closed source people take notice is necessarilly a good thing (if that were even possible). The closed software producers are often large corporations with a lot of marketting muscle and Linux may well benefit from them promoting their Linux versions of their software (and thus promoting the whole OS).
I mean, lets that a theoretical example:
Lets say Adobe ported PhotoShop to Linux. They put a bit of marketting behind it and a large chunk of PhotoShop users migrate over to Linux. (Ok, I guess this isn't necessarilly the best example since they're pretty attached to their Macs, but bear with me). A migration to Linux like this would give it quite a boost in the desktop market.
Conversely, let's just imagine that The GIMP gets as good as (or better than) PhotoShop before Adobe start caring about Linux. So now they have a viable free (as in beer) competetor to their rather expensive product. Are they actually going to want to promote Linux (which usually ships as standard with The GIMP)? Of course not, they're going to want to use their marketting weight to push people away from Linux in the hope that they can keep as many people as possible from discovering that they can get something as good as PhotoShop for free.
Yes, ok, so The GIMP is available for other platforms, but people are more likley to discover it if it's already installed when they get their computer.
Given time, any company wishing to compete in the Linux market will have to produce software which is significantly better than the established FOSS tools, and that has to be good for us computer users.
It's worth noting that (in my experience) most commercial decisions are not based on "what's best for the job" or even "what gives the most bang for our buck". I've lost count of the number of times employers have forced me to use some very expensive piece of software that really is nowhere near as good at the job as some FOSS software. In most of these cases, the expensive commercial software is a good 10 years behind the free equivalent. Many of these purchasing decisions seem to basically just be made on the "noone ever got fired for buying IBM" premise (replace "IBM" with any large corporation who has been selling expensive software for a long time).
Totally impervious to war, as resistant to acts of God as one can get
I'm sure hosting your website from inside a mountain would be wonderful - your webservers would be totally impervious to an EMP.... sadly your upstream peers would not be so fortunate so whilest your webserver would stay up it would have no connection to the rest of the world...
Remember that these are also the people that 'forgot' to convert metric to standard messurements once.
TBH, given that:
1. The entire scientific world uses standard metric measurements
2. Everyone in the world except the US uses standard metric measurements
I'm really amazed that NASA use imperial units at all... it just seems crazy.
In my reference IE, specifically loads the HTML engine COM DLLs into its own 'isolated' process. This is done for a couple of reasons, but the main one being security.
I really can't see what security benefit there could possibly be - executable memory is read-only, so sharing it between processes is not a security risk. Isolating code like that is simply a needlessly inefficient use of resources.
as the versioning of these 3rd Party DLLs (even some from MS, but still considered 3rd party to the OS) caused problems because if Version 2.0 of DLL was loaded, and another application needed Version 3.0 the mapping of the DLL would cause problems.
Under Linux the versioning doesn't come into play at all at this point - both processes separately try to mmap() the file, but the kernel will realise it is the same *file* (it has the same inode number) and therefore will share the memory. The kernel has no concept of a global symbol table of sharable symbols - the only thing the kernel knows (or indeed, needs to know) is that there are 2 processes trying to map the same lump of read-only data into memory.
The idea of sharing code based on a vendor-specified version number seems very flawed, whereas if you just base the sharing off the inode number you really can't go wrong.
If Windows doesn't automatically share libraries between applications then it's a worse operating system than I originally thought. *Every* other modern OS shares dynamically linked code between applications that use it, to do otherwise would be woefully inefficient (both in memory usage and startup time).
Let's look at how Linux handles shared libs (this is about the same as any other modern OS):
Since the executable code is read-only, there is no security problem with allowing both processes to access the data. (mmap() basically tells the kernel to treat the file a bit like swap space).
So is it simply that the US comes out on top because we have so damned many computers - as opposed to other nations where they're sometimes uncommon in households and people use internet cafes?
ISTR I saw some statistics a while ago suggesting that the UK had a far higher DSL/cable connections to people ratio than most other countries (I think even more than the US). Yet the UK is pretty low down on the list of spammers. Admittedly the UK population is lower than the US population though - they really need to adjust those figures into "spams per citizen" or "spams per internet connection" to make them meaningful enough to draw conclusions about user cluefulness.
Customers get very irate when they are disconnected, sandboxed, and refused further service until they run (free) anti-virus software on their computer.
In other news, drivers get very irate when they aren't allowed to drive their unsafe car on the road until it's been fixed. However, banning people with unsafe cars makes everyone else safer, so is a Good Thing. Same with infected computers. If a computer is actively attacking other systems then drop it's connection ASAP - this is good for two reasons:
1. It stops the infected system from doing any more damage to any other systems/people (this may be relaying spam, DDoSing someone, trying to infect other systems with a worm, running a phishing site, etc).
2. If someone loses their whole connection every time they get infected they might actually start giving a damn about their system's security.
that causes tremendous expense for them in terms of customer calls.
If all the ISPs started taking these measures then it would surely *reduce* the number of support calls since the number of infected systems would be reduced. Sure, there'll be a short term peak in the number of support calls but the long term picture is much better. Sadly, most businesses these days only seem to care about the short term bottom-line.
Yet more evidence that the universe is just a gigantic computer simulation.
It could be argued that a lot of the weirdness in quantum mechanics is down to programming shortcuts for efficiency reasons. E.g. look at things like the double-slit experiment - maybe it's less expensive to simulate a wave function than a particle and only collapse the wave function when it's measured. Just a thought.
In any case, even if the universe is a giant computer simulation, does that make it any less "real" (whatever "real" means)? And there's no way we'd ever know one way or another - the best we can do is discover the rules that govern the workings of the internals of the universe, not the externals.
Landing in one piece is the real trick...
:)
taking off in one piece can be a problem too with that much explosive down your pants