Damn, I hadn't thought of that. You didn't mention the fact that if you have an analogue watch, by the time you've wound it round to the correct time, a few seconds have passed and it's no longer the correct time! Not to mention getting that molecular-level precision; getting out the microscope and setting it up is such a bind.
Seriously, you've never thought that correcting a watch once a year isn't that much of a struggle? Versus the fact that as someone else said, there are already auto-syncing watches. "Clocks with everything" is a very odd mentality, but still most of my friends don't wear watches. It's amusing when they have to take out their phone to see the time.
That's not consolidation, that's taking one existing system and removing support for all others. Now, I personally would have no problem with removing support for rpm, or ebuild, or deb or whatever and focusing on just one type, since in the end, they all do pretty much the same thing. In fact, people often complain about the development of rpm when deb already existed.
If one were picked above all others (Which is unlikely, and mostly unworkable) then to be successful and to have all advantages of a unified system, it would need to take on the features of the other systems anyway, so all but the die-harders would probably be pleased anyway.
This may be my last reply, since my head is about to explode from repeated ignorance of content put forward.
I'm not advocating restriction of choice, I'm advocating the development of a GRUP or OTPS or whatever. I expect, however, that if this were worked on cooperatively between the distributions, we would end up with something far better than what we have now. Nowhere have I said that I am going to build such a system and force it on people, mainly because that is impossible - you can take your words back out of my mouth, thanks.
How would anyone silence proponents of other approaches? While we're at the underhand remarks about Nazism, I assume you're implying I have an SS, as well? Since I don't, there will always be some amount of controversy. Perhaps the proponents of other approaches will actually want to put their approaches into the unified system, then (shock horror) it can be customised to the user's taste, if they wish.
There is nothing I see currently that is mutually exclusive about the systems, nothing to say that the good points of each cannot be incorporated into an "OPTS." In which case, how do we have a worse situation than we have now? Yes you are forced to use one system, but since it includes the features of all the ones before, (which is likely, since it would be built with knowledge of existing systems) there's no disadvantage.
Choosing later means first improving the system, and then there will presumably be forks and changes and (shock horror) ugrades. If you don't fit into the 'one size fits all' (which it will probably accomplish far better than anything windows has to offer, since it will have input from all sizes) there will always be alternatives.
Uh, I think the GUI bit was referring to someone else's post about source...
It was actually this situation that made me realise how glad I am to have made the switch. When I want a piece of software, if it's not in a repository (unlikely) it is on freshmeat or sourceforge or something, availabe for free, for my distribution, and it does what I want.
AFAIAC, a tarball that can be uncompressed, compiled, linked, and installed is a perfectly workable packaging system.
My I in that case rephrase it to "package management system?" I don't think tarballs can fit that description, unless you extend the package manager to the user, which is like saying counting in your head is a calculator application.
That's a pretty poor example. How widespread is Firefox usage among Linux users?
Unfortunately irrelevant. It's impossible to prove that the best system is always chosen. If you want another example, how about VCR/Betamax? Just because there is no consensus is no indication whatsoever of there being one clear winner, and nor is it an indication that amalgamating features will be a bad result.
...what I said was that those who cannot follow the instructions for the tools they are using should not switch from Windows to Linux.
Like I said, installing from source is more confusing than installing using synaptic or portage or whatever, hence package management systems are more popular than source.
That's like going to a buffet where there is only one dish.
By which token, what we have now is something like having a buffet with different dishes, but every dish is in a different house, at a different party. With a consolidated system, or at the very least an abstraction system, you have one main dish in your own house, then others in others.
There's nothing to say that the One True Packaging System would be one of the current ones; it would likely incorporate all the best features of them all. If you still want to be backward, you can use another one, but to be honest, there's not a vast amount of difference. SuSE, FC, Mandriva and Redhat already use RPMs, Debian and anything derived therefrom uses.debs, et al. There's not a lot of choice at the moment, and if the OTPS were adopted, others would be available, at least for a while
What you appear to be saying is that, because this would probably restrict choice, we mustn't improve. A consolidated system would provide benefits, and as a side effect could limit choice. If you really wanted three different systems that worked on all distros, I'm sure you could help develop one, and there's nothing saying that the users themselves couldn't suggest/submit improvements to the OTPS.
Who is qualified to decide which distros should be terminated
Uh, excuse me? Firstly, some distros probably could do with "extermination," as you rather emotively put it, but as TFA says, the linux world does spread itself thinly, to its detriment.
Who put you in charge?
Me, when I decided to use Linux. I elected to use a form of computing whereby I could take charge, make suggestions (which is all I'm doing) and, frankly, what are you doing but the same as I, except advocating stand-still?
What if I don't like the fact that there is no unified way of doing things?
The long and short of it is that the only disadvantage you've put forward so far is that you wouldn't be able to choose something else, which is no reason not to improve what we have. We can deal with choice later, and, like I said, in an open source world it's easier to cater to more people's tastes, even if not everybody's.
You're quite right. What's needed is a way of verifying that the address in the from field is really who sent it, some kind of PGP signature or the like would probably work.
While we all appreciate the fact that some people aren't able to partake in outdoor activities, possibly including those who are chronically indisposed to sunlight, just because one subset of people cannot enjoy them doesn't mean the rest don't.
Well, first off, for many newbies it's less confusing for it all to be wrapped in GUI fluff (not necessarily for complete newbies, but that a whole different kettle-o-fish)
Then of course, have you ever tried to install decent free software for windows? It's horrific! I tried to find a free CD ripper for windows, but there was not a single good quality one in sight, and it took an hour or two just to ascertain that. The endless trawling through google results of time limited registration and so on was sickening to me, and no doubt confusing and frustrating for anyone else.
"No,"
You mean, "or," surely? In effect you have the same end result, except that a new system would allow the introduction of new features which would be very useful in the building of single way of packaging. Because not all systems allow, for example, build flags, you either have to have a lowest-common-denominator system, or go with a system which is not as well abstracted (in terms of the user experience) as it otherwise could be. For example, portage provides its use flags, but there is no equivalent for debs.
While an abstraction layer would bring many and most of the outlined benefits, it would miss out on a few that, in my opinion, were quite crucial. If it were to happen it would also discourage the consolidation of all the systems, and we still have to have many many overworked packagers and maintainers. The article mentioned the linux hackers being spread to thinly - consolidation rather than abstraction would help this aspect.
Remember a "packaging system" consists of a way of wrapping up the source code/binary files and also a "package manager." Source code on its own provides nothing of the sort, and doesn't even have the metadata necessary to create one.
"The fact that we have different distros with different package installers is a clear sign that there is no consensus on the best method. Obviously, if one method becomes superior, most distros will adopt it. Until then, let the competition continue."
I think you have too much faith in the likelihood of a consensus being reached without something designed to bridge all distributions. The fact that firefox is better than IE has not yet led to widespread adoption of it, and you're relying in the major distros all a) swallowing any pride and b) completely overhauling the way their system is maintained and installed, while one distribution gets none of the hassle.
Saying that confused new users should be guided towards Mandrake or suchlike is simply ignoring the benefits a consolidation could afford. Likewise saying those who cannot install from source should switch. While linux ought not to sacrifice usability from the point of view of the average and up, it should not sacrifice it on the low end, because even hackers can use GUI tools. Consolidation still brings the advantages mentioned, gets rid of most confusion, provides a warm fuzzy GUI to cradle the confused, and if all goes to plan, will allow all the control of "manual transmission."
I know it's what gentoo does, and it's probably a system such as portage/emerge that will eventually become a global packaging system, I would have thought, because it doesn't require a package for every architecture and distribution.
I think we'll probably see the --use options turning into GUI installer types, though, with checkboxes for./configure type flags.
This is not a packaging system. There is no consolidated repository for source "packages," there is no way of externally mapping dependencies or handling them automatically. There is no hard-and-fast way of installing them that works for every source package, and above all, it's time consuming and potentially confusing for a first-time user. Not necessarily even a "Joe Average" user, but anyone that hasn't seen linux before. I know I was confused when first faced with this arcane method of installing stuff, and royally screwed my first system up with it.
Certainly, it would be possible to create a package system, that containe the source along with necessary metadata, but don't try and pass it off as anything like meeting the requirements a packaging system fulfills.
But there are huge benefits to be gained in usability with a consolidated packaging system. Think what it means - projects need to be ported to ONE distribution system. Proprietary drivers need support ONE more system, make sure it works with ONE system. Users would be nearly garunteed to be able to find the program they want if there were consolidated repositories. The repos could be divided a la debian into stable, free, nonfree, testing, unstable - although preferably in my view another that's in between testing and unstable. Thataway every distro could be as stable as debian, as bleeding edge as fedora and whatehaveyou. Also preferably organised by flags as opposed to directories, thatway you can have "nonfree, stable."
You've gotta admit that the massive web of differing packagin systems is often high up on the list of criticisms against usability, and if there were consolidated repos, then you needn't worry about getting software from dodgy sites as most stuff would be right there in repo.
Hopefully you wouldn't be using an unpatched version of firefox, anyway. Nothing will protect people from their own stupidity except teaching, and it is likely that firefox will be targeted soon anyway. With any luck, the mozilla people will release fixes in a more timely fashion and, more importantly, in a more accessible fashion. You won't have to reboot your computer, just restart the browser, and things like that.
This, partially, is the difference between Firefox and most other OSS products. There's a huge effort to encourage widespread adoption. It's really admirable - look at http://www.getfirefox.com/, http://www.spreadfirefox.com/ (the one that was hacked, shh,) the NYtimes advert. It's all far more organised and well executed than any "advertising" others put out. And it's working, along with the efforts of its faithful users. Eventually, even the "no way" group will switch over, if people keep trying. There will be no incentive whatsoever to stay, as there will be so many other people already adopted it.
Or perhaps recommend a site that works/renders in FF, but not in IE. You'd probably have to rig it yourself, but it would show IE's shoddiness. You could even put your little W3C labels on there, and say, "well, the site's correct, that's what the labels say. It must be IE's fault."
You can be pretty sure quite easily. If you are mailed your password from the "forgotten..." section, it's in plaintext for certain. If not, it's likely to be hashed.
I once registered for a site that required you give your full address, including postcode. If you gave an invalid postcode, or it didn't match your address, they didn't let you register. Of course, it's easily bypassable by saying you live in Antarctica. Or the Democratic Republic of Congo.
Slashdot Mirror
Damn, I hadn't thought of that. You didn't mention the fact that if you have an analogue watch, by the time you've wound it round to the correct time, a few seconds have passed and it's no longer the correct time! Not to mention getting that molecular-level precision; getting out the microscope and setting it up is such a bind.
Tell me, did you actually run the script? The worst it does is set the background colour for the terminal you run it from.
Seriously, you've never thought that correcting a watch once a year isn't that much of a struggle? Versus the fact that as someone else said, there are already auto-syncing watches. "Clocks with everything" is a very odd mentality, but still most of my friends don't wear watches. It's amusing when they have to take out their phone to see the time.
Don't run this!! It will delete all your files!! And mail your penis to SCO!1
If one were picked above all others (Which is unlikely, and mostly unworkable) then to be successful and to have all advantages of a unified system, it would need to take on the features of the other systems anyway, so all but the die-harders would probably be pleased anyway.
I'm not advocating restriction of choice, I'm advocating the development of a GRUP or OTPS or whatever. I expect, however, that if this were worked on cooperatively between the distributions, we would end up with something far better than what we have now. Nowhere have I said that I am going to build such a system and force it on people, mainly because that is impossible - you can take your words back out of my mouth, thanks.
How would anyone silence proponents of other approaches? While we're at the underhand remarks about Nazism, I assume you're implying I have an SS, as well? Since I don't, there will always be some amount of controversy. Perhaps the proponents of other approaches will actually want to put their approaches into the unified system, then (shock horror) it can be customised to the user's taste, if they wish.
There is nothing I see currently that is mutually exclusive about the systems, nothing to say that the good points of each cannot be incorporated into an "OPTS." In which case, how do we have a worse situation than we have now? Yes you are forced to use one system, but since it includes the features of all the ones before, (which is likely, since it would be built with knowledge of existing systems) there's no disadvantage.
Choosing later means first improving the system, and then there will presumably be forks and changes and (shock horror) ugrades. If you don't fit into the 'one size fits all' (which it will probably accomplish far better than anything windows has to offer, since it will have input from all sizes) there will always be alternatives.
It was actually this situation that made me realise how glad I am to have made the switch. When I want a piece of software, if it's not in a repository (unlikely) it is on freshmeat or sourceforge or something, availabe for free, for my distribution, and it does what I want.
My I in that case rephrase it to "package management system?" I don't think tarballs can fit that description, unless you extend the package manager to the user, which is like saying counting in your head is a calculator application.
That's a pretty poor example. How widespread is Firefox usage among Linux users?
Unfortunately irrelevant. It's impossible to prove that the best system is always chosen. If you want another example, how about VCR/Betamax? Just because there is no consensus is no indication whatsoever of there being one clear winner, and nor is it an indication that amalgamating features will be a bad result.
Like I said, installing from source is more confusing than installing using synaptic or portage or whatever, hence package management systems are more popular than source.
That's like going to a buffet where there is only one dish.
By which token, what we have now is something like having a buffet with different dishes, but every dish is in a different house, at a different party. With a consolidated system, or at the very least an abstraction system, you have one main dish in your own house, then others in others.
There's nothing to say that the One True Packaging System would be one of the current ones; it would likely incorporate all the best features of them all. If you still want to be backward, you can use another one, but to be honest, there's not a vast amount of difference. SuSE, FC, Mandriva and Redhat already use RPMs, Debian and anything derived therefrom uses .debs, et al. There's not a lot of choice at the moment, and if the OTPS were adopted, others would be available, at least for a while
What you appear to be saying is that, because this would probably restrict choice, we mustn't improve. A consolidated system would provide benefits, and as a side effect could limit choice. If you really wanted three different systems that worked on all distros, I'm sure you could help develop one, and there's nothing saying that the users themselves couldn't suggest/submit improvements to the OTPS.
Who is qualified to decide which distros should be terminated Uh, excuse me? Firstly, some distros probably could do with "extermination," as you rather emotively put it, but as TFA says, the linux world does spread itself thinly, to its detriment.
Who put you in charge?
Me, when I decided to use Linux. I elected to use a form of computing whereby I could take charge, make suggestions (which is all I'm doing) and, frankly, what are you doing but the same as I, except advocating stand-still?
What if I don't like the fact that there is no unified way of doing things?
The long and short of it is that the only disadvantage you've put forward so far is that you wouldn't be able to choose something else, which is no reason not to improve what we have. We can deal with choice later, and, like I said, in an open source world it's easier to cater to more people's tastes, even if not everybody's.
You're quite right. What's needed is a way of verifying that the address in the from field is really who sent it, some kind of PGP signature or the like would probably work.
You can set the from: field to your own website, however. This is legitimate, and the right thing to do, since it is originating from your own server.
Pleasure?
Specifically, playing the Toronto subset, meaning that the Marmelade Sandwich rule for unblocking Paddington is replaced with the Beer Drinking Rule.
While we all appreciate the fact that some people aren't able to partake in outdoor activities, possibly including those who are chronically indisposed to sunlight, just because one subset of people cannot enjoy them doesn't mean the rest don't.
Well, first off, for many newbies it's less confusing for it all to be wrapped in GUI fluff (not necessarily for complete newbies, but that a whole different kettle-o-fish)
Then of course, have you ever tried to install decent free software for windows? It's horrific! I tried to find a free CD ripper for windows, but there was not a single good quality one in sight, and it took an hour or two just to ascertain that. The endless trawling through google results of time limited registration and so on was sickening to me, and no doubt confusing and frustrating for anyone else.
While an abstraction layer would bring many and most of the outlined benefits, it would miss out on a few that, in my opinion, were quite crucial. If it were to happen it would also discourage the consolidation of all the systems, and we still have to have many many overworked packagers and maintainers. The article mentioned the linux hackers being spread to thinly - consolidation rather than abstraction would help this aspect.
"The fact that we have different distros with different package installers is a clear sign that there is no consensus on the best method. Obviously, if one method becomes superior, most distros will adopt it. Until then, let the competition continue."
I think you have too much faith in the likelihood of a consensus being reached without something designed to bridge all distributions. The fact that firefox is better than IE has not yet led to widespread adoption of it, and you're relying in the major distros all a) swallowing any pride and b) completely overhauling the way their system is maintained and installed, while one distribution gets none of the hassle.
Saying that confused new users should be guided towards Mandrake or suchlike is simply ignoring the benefits a consolidation could afford. Likewise saying those who cannot install from source should switch. While linux ought not to sacrifice usability from the point of view of the average and up, it should not sacrifice it on the low end, because even hackers can use GUI tools. Consolidation still brings the advantages mentioned, gets rid of most confusion, provides a warm fuzzy GUI to cradle the confused, and if all goes to plan, will allow all the control of "manual transmission."
I think we'll probably see the --use options turning into GUI installer types, though, with checkboxes for ./configure type flags.
This is not a packaging system. There is no consolidated repository for source "packages," there is no way of externally mapping dependencies or handling them automatically. There is no hard-and-fast way of installing them that works for every source package, and above all, it's time consuming and potentially confusing for a first-time user. Not necessarily even a "Joe Average" user, but anyone that hasn't seen linux before. I know I was confused when first faced with this arcane method of installing stuff, and royally screwed my first system up with it.
Certainly, it would be possible to create a package system, that containe the source along with necessary metadata, but don't try and pass it off as anything like meeting the requirements a packaging system fulfills.
But there are huge benefits to be gained in usability with a consolidated packaging system. Think what it means - projects need to be ported to ONE distribution system. Proprietary drivers need support ONE more system, make sure it works with ONE system. Users would be nearly garunteed to be able to find the program they want if there were consolidated repositories. The repos could be divided a la debian into stable, free, nonfree, testing, unstable - although preferably in my view another that's in between testing and unstable. Thataway every distro could be as stable as debian, as bleeding edge as fedora and whatehaveyou. Also preferably organised by flags as opposed to directories, thatway you can have "nonfree, stable."
You've gotta admit that the massive web of differing packagin systems is often high up on the list of criticisms against usability, and if there were consolidated repos, then you needn't worry about getting software from dodgy sites as most stuff would be right there in repo.
Hopefully you wouldn't be using an unpatched version of firefox, anyway. Nothing will protect people from their own stupidity except teaching, and it is likely that firefox will be targeted soon anyway. With any luck, the mozilla people will release fixes in a more timely fashion and, more importantly, in a more accessible fashion. You won't have to reboot your computer, just restart the browser, and things like that.
This, partially, is the difference between Firefox and most other OSS products. There's a huge effort to encourage widespread adoption. It's really admirable - look at http://www.getfirefox.com/, http://www.spreadfirefox.com/ (the one that was hacked, shh,) the NYtimes advert. It's all far more organised and well executed than any "advertising" others put out. And it's working, along with the efforts of its faithful users. Eventually, even the "no way" group will switch over, if people keep trying. There will be no incentive whatsoever to stay, as there will be so many other people already adopted it.
Or perhaps recommend a site that works/renders in FF, but not in IE. You'd probably have to rig it yourself, but it would show IE's shoddiness. You could even put your little W3C labels on there, and say, "well, the site's correct, that's what the labels say. It must be IE's fault."
You can be pretty sure quite easily. If you are mailed your password from the "forgotten..." section, it's in plaintext for certain. If not, it's likely to be hashed.
AFAIK, they become far more difficult when salted, is that not correct?
I once registered for a site that required you give your full address, including postcode. If you gave an invalid postcode, or it didn't match your address, they didn't let you register. Of course, it's easily bypassable by saying you live in Antarctica. Or the Democratic Republic of Congo.