Actually there are some DRM schemes that have not been broken, typically because they are implemented in well protected hardware which very few people have the resources/money/skill to break.
P4 satellite encryption is one such scheme, used by I think DirecTV in the states these days (different to the previous system they used where the smart cards were vulnerable to glitching).
Likewise the system used by Sky in the UK has not yet been cracked to my knowledge.
Game console security is likewise rather good, piracy for console games is typically much lower than it is for PCs.
Basically when you control the hardware it's a lot easier to control piracy, hence the focus on "trusted" PCs.
Back in the 80s CDs weren't made the same way they are today, and they did indeed rust due to manufacturing defects. Once identified the problem was rapidly corrected and I believe modern CDs do not rust.
Lots of people have replied saying "I do! I do!". In terms of ripping CDs to my computer to make playing them more convenient, I do as well, but this wasn't ScottGants point (I suspect).
I have seen friends of mine bring in entire stacks of recordable CDs containing ripped movies they downloaded from the net. By "stack" I don't mean 3 or 4 but more like 100. They chat about it and trade them openly - there is no guilt there, while they know it's illegal they just don't care. It's easy to be amoral when everybody else is.
I don't download movies and music (I used to download music, back in the Napster days, but reformed), instead I just rent DVDs and listen to net radio all the time. I hardly ever see films I like so much I'd want to watch them over and over. Needless to say, I'm the odd one out.
So we have this conundrum - piracy is a huge problem, especially so because the vast majority of people doing it just couldn't care less that it's illegal. Yet, people do make backups, rip CDs (maybe even movies) to their computer for portability and so on.
How do you balance these two? That is the problem.
Zed is not the total copyright holder on X-Chat. If he was, you'd be right, but he's not, so you're wrong. I know at least one person who has written patches for X-Chat that were accepted and were never consulted about this move (and they're pretty pissed off).
So unless he gets permission from every single person he can't do it. Or he can try and rewrite all the patches, but even that's probably not legal as some of the patches may have affected the design, the API etc - what is an isn't a derived work is rather blurred.
Oh yeah, that's a really good idea. In my first year I actually forgot the power cable for my monitor - bit of a pain but turned out to be a great idea. Time I would have wasted fiddling with it was spent exploring the college grounds, chatting to people on the corridors etc. I got it mailed up after about a week and there was no problems but I'd totally recommend leaving it in the box for a bit.
Especially if you do that and you're studying CompSci then it will send a powerful message to people that you're Not A Nerd and not obsessed by computers. People respect that.
Last year I lived in college (actually, this is Durham university in England so "college" means something a bit different than in America, but they're close enough) and so like any other student I'll chip some advice.
Firstly, one of the best things about living in dorms and being in a collegiate environment is that most of the friends you'll make won't be people doing your course. Of all my friends at university, only one is doing CompSci. The rest are a pretty eclectic bunch, and a pretty even mix of girls and boys. In fact I'm living with 3 girls next year (not deliberate, just the way the combinations worked out when we were all finding houses...). This sort of life has been great fun - the friend from CompSci is a True Geek like me (he's into OpenBSD hacking) so I get to talk shop and blow off steam with a fellow geek occasionally and spend the rest of my time away from the machines with friends doing music, history, physics, engineering - contrary to conventional wisdom the engineering friend I'm thinking of is one of the coolest guys in college:)
Yes anyway, on offering to set up computers for girls: if you are studying CompSci you will be asked to do this, if you offer/advertise or not. It's just a part of being a "computer guy", you'll be doing it for the rest of your life most likely so:
(a) Learn to enjoy it. It's a good way to meet new people - especially women. It's unlikely to lead to anything but I know one guy who did in fact meet his girlfriend after fixing her computer, and they've been steady for several years!
but (b) I wouldn't recommend deliberately seeking it out.
Unless you really, really enjoy cleaning spyware off Windows XP laptops you'll find you have enough work just as word spreads that you are the guy who "knows computers" on your floor. Being happy to help, with a good attitude, will take you far but seeking it out or advertising your "services" will simply mark you in peoples minds as a nerd.
Nerd vs geek is a subtle distinction but important. Geek, as long as it's restrained and doesn't dominate, is socially acceptable at university. If people can immediately notice your chosen area of geekdom by talking to you for a few minutes or worse, just by looking at you, it's over the top. However most students - especially if you're at a top university - will be geeky about something, even the girls. Being a computer geek is fine, just don't shout it out. Being a computer nerd is not fine. Walking the line is something everybody learns to do.
Having a good attitude is really, really important. You'll have to do tech support for boring or annoying guys as well as super-nice and sexy girls, just live with it. Worst faux pax you can do is wear one of those stupid "NO I WILL NOT FIX YOUR COMPUTER" t-shirts. You have a skill most people don't and acting like it makes you gods gift will totally piss people off, which is not what you want.
If you find it's taking too much time, or interfering with your work, just be honest about it. I've put off/refused to do tech support before when I've felt I haven't had the time. Especially, if you don't feel like it, just say "Sure, I'm feeling a bit knackered right now, mind if I take a look tomorrow?" - most non-CS people do not care if their computer is out of action for a few days unless they have an assignment coming up or something. Don't fall into the trap of thinking they find a working machine as important as you do. I've seen students with a totally trashed system just say "oh well, I'll take it home at the end of term and get it replaced" when the end of term is still over a month away!
Hmm, what else. Not much really - be nice, help people, but don't try too hard. Be understanding to those who find using technology hard: just being sympathetic and telling some amusing stories ( "hey, it's not so bad, I once knew somebody...." ) can be a go
No, that's total BS. The GPL says that derived works must also be licensed under the GPL (or a compatible license). It also says you cannot arbitrarily impose additional restrictions, which a 30 day cutoff absolutely is.
The key point is, what would a judge think? And in this case there would be no shortage of expert witnesses willing to testify that patches with no explicit license have an implicit license identical to the original codebase as a matter of course (being derived works).
The other key point is that Zed has pissed off pretty much every contributor with this move, as well as almost certainly breaking the law. I don't understand why - if he hated doing Windows binaries, he could just stop and let somebody else do them. If nobody else could do them satisfactorily, pull Windows support entirely. Screwing over contributors like this is totally bogus.
It's a free market. Don't like MS? There's a free alternative. Stop whining.
Linux isn't currently a viable alternative for the vast majority of people due to backwards compatibility issues (ie, the lack of it). Neither is Apple, especially in business.
Something that is free is no use if you can't use it.
This is really basic stuff, it worries me that people don't seem to understand it....
I don't use Gentoo, I was talking about broken apps installed by wrong ebuilds. Wine is an obvious example of a pathologically broken ebuild, despite its popularity (winehq gets >2million hits/month).
Does the new Reiser4 file system support any of these concepts? -- Is WinFS really as new and exciting as the marketing and media says it is?
ReiserFS and WinFS are two competing philosophies.
WinFS is a data store. It's a separate, monolithic database which stores meta-information about files on the traditional filing system.
ReiserFS is about extending existing filing system technology such that data becomes transparent and self describing allowing any kind of querying facility to be layered on top.
Let me try an explain. Let's say you have a hard disk of OpenOffice Writer documents, which you wish to query. This is hard, because the SXW format is a complex beast. To the operating system it's an opaque series of bytes. Let's see how you'd query photos embedded in these documents:
Firstly you need to locate and open the file using the POSIX fs apis. Next, use a zip library to navigate the compressed filing-system-within-a-file that zip files are, to locate one of the XML files contained within. Now load up an XML parser and navigate the XML to one of the image nodes. Unfortunately not all information is easily represented using XML: this is one such piece, so it's actually stored as a JPEG encoded binary file... decode that, and now navigate the structures in memory to arrive at the data you want.
Word documents are not much better.
This is an extreme example, but hopefully you see my point. Look at how many APIs were required to get to the wanted information, yet they are all fundamentally the same. ZIP files are miniature filing systems which add a compression feature (and performance!). XML is a tree of nodes: hmm, kind of like a filing system. Binary structures in memory tend to be trees or graphs of information: a bit like a filing system that supports linking.
What if we could unify these APIs? What if the underlying operating systems filing system was powerful enough to be the superset of all the features these disparate APIs provided? ZIP files are used for compression, for fast access to the contents and because it makes it easy to send them via the internet and manipulate them with modern file managers. XML is used because it's an efficient way to represent a complex tree with many nodes. Binary structures are used because some stuff just can't be easily encoded as text.
But we have a problem - there are sound technical reasons why openoffice documents are not a sparse collection of files. For one, most filing systems are not fast enough: a file is an expensive thing, opening and reading them even moreso. You don't want a file for every cell in a spreadsheet, or every paragraph in a word processor document. The overhead is too high. There is another problem: files cannot be directories and vice-versa. Having each paragraph as a file may be convenient for search engines but it's not so convenient for users.
What if files could simultaneously be directories, and what if we used a filing system designed so that a 3 byte file is not an unacceptably inefficient design? What if we could decompose our elaborate file formats with our chunks and headers and streams and DIRENTs into a tree of files all accessed via the POSIX APIs: open(), read(), close() ?
No longer would the structure of an image embedded in a word processor document be a mysterious and opaque bytestream to other programs. Now it's trivial to trawl the content of files and index them.
You see, this is the genius of Hans Reiser. He realised that writing indexer plugins for every file format under the sun would never work, it'd never scale, it'd never give users consistently good results. The right way is to make the foundations powerful enough that the concept of file formats itself falls away: by minimizing primitives, by unifiying interfaces, the system becomes more powerful.
The technical challenges of such an approach are enormous, it can only be done because ReiserFS is not a "bet the company" move, as
Oh, not to mention there are lots of nice config tools and integration widgets in Fedora that aren't in most other distros. Even little things like a working pam_xauth can reduce the potential tech support burden.
No, the main problem is that it's too unstable. Things break. I've had to recover a few friends Gentoo systems after an emerge update blasted gcc or whatever. It doesn't happen *often* but it does happen.
Gentoo is not a distro I'd ever give to my non-technical family. Neither is Debian. The reasons should hopefully be obvious.
With the desktop comments I was referring to KDE, explaining why I'd not go for SuSE or Mandrake which are very KDE focussed. Hence the fact that Fedora is basically the best choice right now.
In particular while "emerge" might be easy, there are an awful lot of broken packages out there. I've had to deal with random brokenness caused by bad ebuilds a lot and I have no faith in the system whatsoever. So from my perspective, there is a serious problem. If what you have works for you though, then great!
An installer... you mean, like apt-get ? You know, that tool you just tell 'install foobar', and it downloads and installs the program foobar ? Or would you like some graphical thingie like Synaptic, where you just click on the program, then on Install ? Don't tell me you're still building from source without some specific reason on a 2004 distro ?
The fundamental dilemma is simple.
The only distros on which apt-get and friends actually works reliably are the least user friendly distros.
Debian and Gentoo have serious usability problems in many areas for non-technical users. There's only one distro I'd give to my friends/family today, and that's Fedora Core. Why? Because it has a slick Gnome desktop and I don't think I should be giving non-technical users a desktop which has menu items labelled "smbUmount" and a preference in the file manager called "Minimize memory usage". Sorry. Just MHO. So that leaves Fedora or some customized install of Debian. Compiling from source is right out if only because it takes so long to upgrade the system (think security updates).
Unfortunately Fedora doesn't have many packages, and those packages it does have are often out of date. The moment a new Fedora is released, you have to upgrade too because all the packagers are now targetting the newest release. This is a pain. Often packages are broken, out of date, missing, or only in conflicting repositories. These issues are hard-wired into the system and cannot be solved by throwing manpower at it, despite what some think.
Grandparent is exactly right, we need decentralised packaging. That does not exclude apt type "one command installs", that can easily be layered on top once the fundamental infrastructure is in place. But you've got to get the basics right: build a binary once, install it anywhere.
Go read the autopackage ui vision to see what end users might one day use, then check out the screenshots gallery. Try it for yourself if you like.
Be warned though. On very recent distros (fc2, debian sid etc) the Inkscape package installs but the binaries crash very quickly. It seems some library broke ABI and didn't bump the soname.
It's just what you're used to. I debug code without a debugger all the time, because:
a) I work in bash, and up until recently there was no good bash debugger
b) I work on Wine, and a lot of apps change their behaviour when run under a debugger. The Word team seem to have learned this one the hard way. A debugger is a useful tool but very, very invasive. Some apps actually deliberately change their behaviour when run under a debugger, annoyingly enough, usually to try and stop you reverse engineering them. But suffice it to say that debuggers interfere with the apps they are debugging. I don't tend to use them much these days.
c) I have, in the past, worked a fair bit on Java web apps. These days the debugging tools and IDEs are much better but back then I used only emacs, which has very primitive Java debugging support.
So how do I debug code? Simple.
1) Logging. Logging is the lifeblood of Wine debugging. A good logging framework can make hard problems easy. This is especially useful because users can send you logs so you don't have to be on-site to debug the problem, or even able to reproduce it.
2) Look at the code. This just takes practice but these days if I know roughly what area the code is failing in I can often just stare at it for a while and figure it out. The AIX guy said the same thing. It's slow and unreliable but if you practice it, it can work great for those elusive, hard to nail bugs. This is especially true for thread-safety bugs where a debugger and even logging is pretty much guaranteed to change the nature of the bug.
These techniques are far from foolproof and I'm not implying that debuggers are "bad" or for the "weak" or whatever, they aren't, they are a useful tool. Logging in particular can alter the timing profiles of the program significantly so anything involving races or timing can be screwed up. Just thinking about the code is obviously not foolproof.
But... overreliance on interactive debuggers is a bad thing. It's very common in Windows developers because they've always had great interactive debugging available - I remember when I wrote Windows software using Delphi I was the same. Bugs which changed or disappeared under the debugger always caused me huge pain because I was so used to just whacking a breakpoint in there and examining the variables.
Unfortunately interactive debugging on Linux is total crap, gdb tends to go knock-kneed at anything vaguely threaded and the emacs/gdb integration isn't reliable either. Flip side is you learn to do without, which is a useful skill.
The relevant binary code involved video codecs. These are often patented and licensed from 3rd parties. Even if Phillips had invented and created the technology, that doesn't mean it's in their economic advantage to do so.
Wow, that page is wierd. How is a computer more male than female? I don't see anything particularly male about any of the software that's popular today...
What's depressing is the number of people who search for discrimination so hard they see it even where it doesn't exist.
How can a perfectly reasonable question like that be "gender biased"? It sounds pretty likely to me. Most girls I know at university do humanities because they hate maths and/or science. There are girls who do engineering and physics of course, but they are the minority. They don't appear to have suffered any discrimination though, and are perfectly happy on their course.
He already explained why he hasn't published the information. Whether you agree with his reasoning or not is besides the point, he's explained it and it's not that he just gets a kick out of it. Given that the binary code has broken in the past I tend to agree with him.
Linux doesn't need proprietary drivers
"Linux" may not, but "Linux users" certainly do, at least until the fundamental questions of economics and patents are resolved. Until then proprietary drivers will be on the increase, just like proprietary software in userland will. If Linus doesn't recognise this reality I suspect it may eventually fork.
It is a victory for kernel developers who do not have to waste their time with crash dumps from kernels linked to binary modules, for users who benefit from a more stable kernel, and for the advancement of Linux because it is not held back by archaic binary interfaces
I don't have much sympathy. API versioning isn't rocket science, and it does not doom you to "archaic" interfaces when managed correctly. Compatibility between the major releases (2.4 series, 2.6 series) etc alone would be a major improvement. It can be as simple as introducing a new function instead of changing the prototype of a new one, or introducing padding into the structures.
Unfortunately the kernel developers have this idea that somehow the kernel is exempt from the same rules that govern userspace: if you document and expose interfaces to external code, you keep the interfaces stable.
Quite a lot of armchair coders don't understand backwards compatibility. It does not (necessarily) lead to insecurity or unreliability. I would like to be given examples of occasions where it has, in fact. It does mean managing change, but then this is what maintainership is all about.
The idea that making binary development hard increases the likelyhood of source releases is a fantasy based on an incomplete understanding of the economics involved. The most likely outcome is in fact the discontinuing of the drivers altogether, as has happened here.
The USB hook for binary modules was a real detriment to the USB subsystem. It was taken out for technical reasons.
It was not. Re-read Greg KHs emails. It was taken out because he thought it was a GPL violation (which is a gray area).
Talk about immature. He could leave it there until a new maintainer stepped forward but he'd rather have a dummy spit and stamp his feet.
He has worked for years to produce a solid driver for the community, even signed NDAs to get the relevant specs, and has coded on in the face of ridiculous instability in kernel development. It isn't easy doing such job, yet he did it anyway.
When you have contributed for so long, in the face of such a difficult API to work with, then you may have some authority to say whether his behaviour is "childish" or not. In his position I'd be pretty pissed off as well.
Hell, even Linux userspace has severe problems with keeping stable interfaces. I would hate to be a kernel developer.
I was only talking about X. So that leaves out BSD as it has no GUI, at all.
So? It's still loaded, and still taking up resources (both machine and human resources). I don't see how the X-ness of it or not is relevant.
If you don't use legacy (OS 9 or earlier) applications, Carbon is out too.
No, most interesting Mac apps use Carbon. That includes iTunes, QuickTime and the Finder.
. One only needs to look at things like GtkGLArea, the Qt database additions
GtkGLArea isn't a "customized widget" as there is no equivalent in GTK at all. I never said having new widgets was bad, I said creating exact duplicates of already existing widgets that are specific to your app is bad.
Glade isn't a widget toolkit.
There are only two versions of GTK that need to be installed to run pretty much every GTK app ever written, so I'm not sure what your point is here. Certainly GTK 1.2 is not being used by new apps anymore, unlike Carbon which is being used by new apps - typically the large, industrial strength apps use it over Cocoa.
You've confused a lot of different things in your last paragraph : X is not a scripting system much as Quartz is not. Systems like AppleScript do indeed exist - the existance of KDEs DCOP scripting system clearly proves you wrong on that count. iTunes and QuickTime do not use Cocoa. Quite a few apps use custom widgets in a way that the user doesn't notice, however they are still eating up resources to exist - the calendar program does this for instance. So does Safari.
I'm afraid most of your points aren't grounded in technical reality....
Tar.gz gives better ratios only for certain types of data and you can achieve the same effect by creating "solid" zips (the reason they compress source code better is because tarballs are always solid). So I still think ZIP is more flexible, and can achieve better compression.
You can remove that and it'll still start fast. I've actually deleted Windows, reinstalled it, without reinstalling Office and then run it. It gives you 1 complaint on startup about missing registry entries, but seems to work fine. It still starts in about a second.
Not much software starts so fast so people are always inclined to think Microsoft must "cheat", but they don't. It starts in a second on Linux via CrossOver/Wine as well which is clever because Wine itself imposes a hefty startup penalty.
I posted a more detailed summary of exactly what they do a few days ago here.
Want unix without the X hassle, the 4 different environments you need to run all the programs that you use, the different desktops, etc, etc, etc?
Last time I checked MacOS X had at least 3 different environments loaded at once: BSD, Carbon and Cocoa + whatever custom libraries and toolkits individual apps use (and it's been shown that apps like iTunes do stuff like reinvent stock widgets internally!).
So recommending somebody who is complaining about memory usage to buy a Mac is simply mindless advocacy which just makes you look uninformed. After all, this is the OS where the stock media player is about 60MB.
Slashdot Mirror
P4 satellite encryption is one such scheme, used by I think DirecTV in the states these days (different to the previous system they used where the smart cards were vulnerable to glitching).
Likewise the system used by Sky in the UK has not yet been cracked to my knowledge.
Game console security is likewise rather good, piracy for console games is typically much lower than it is for PCs.
Basically when you control the hardware it's a lot easier to control piracy, hence the focus on "trusted" PCs.
Back in the 80s CDs weren't made the same way they are today, and they did indeed rust due to manufacturing defects. Once identified the problem was rapidly corrected and I believe modern CDs do not rust.
I have seen friends of mine bring in entire stacks of recordable CDs containing ripped movies they downloaded from the net. By "stack" I don't mean 3 or 4 but more like 100. They chat about it and trade them openly - there is no guilt there, while they know it's illegal they just don't care. It's easy to be amoral when everybody else is.
I don't download movies and music (I used to download music, back in the Napster days, but reformed), instead I just rent DVDs and listen to net radio all the time. I hardly ever see films I like so much I'd want to watch them over and over. Needless to say, I'm the odd one out.
So we have this conundrum - piracy is a huge problem, especially so because the vast majority of people doing it just couldn't care less that it's illegal. Yet, people do make backups, rip CDs (maybe even movies) to their computer for portability and so on.
How do you balance these two? That is the problem.
So unless he gets permission from every single person he can't do it. Or he can try and rewrite all the patches, but even that's probably not legal as some of the patches may have affected the design, the API etc - what is an isn't a derived work is rather blurred.
Especially if you do that and you're studying CompSci then it will send a powerful message to people that you're Not A Nerd and not obsessed by computers. People respect that.
Last year I lived in college (actually, this is Durham university in England so "college" means something a bit different than in America, but they're close enough) and so like any other student I'll chip some advice.
Firstly, one of the best things about living in dorms and being in a collegiate environment is that most of the friends you'll make won't be people doing your course. Of all my friends at university, only one is doing CompSci. The rest are a pretty eclectic bunch, and a pretty even mix of girls and boys. In fact I'm living with 3 girls next year (not deliberate, just the way the combinations worked out when we were all finding houses ...). This sort of life has been great fun - the friend from CompSci is a True Geek like me (he's into OpenBSD hacking) so I get to talk shop and blow off steam with a fellow geek occasionally and spend the rest of my time away from the machines with friends doing music, history, physics, engineering - contrary to conventional wisdom the engineering friend I'm thinking of is one of the coolest guys in college :)
Yes anyway, on offering to set up computers for girls: if you are studying CompSci you will be asked to do this, if you offer/advertise or not. It's just a part of being a "computer guy", you'll be doing it for the rest of your life most likely so :
(a) Learn to enjoy it. It's a good way to meet new people - especially women. It's unlikely to lead to anything but I know one guy who did in fact meet his girlfriend after fixing her computer, and they've been steady for several years!
but (b) I wouldn't recommend deliberately seeking it out.
Unless you really, really enjoy cleaning spyware off Windows XP laptops you'll find you have enough work just as word spreads that you are the guy who "knows computers" on your floor. Being happy to help, with a good attitude, will take you far but seeking it out or advertising your "services" will simply mark you in peoples minds as a nerd.
Nerd vs geek is a subtle distinction but important. Geek, as long as it's restrained and doesn't dominate, is socially acceptable at university. If people can immediately notice your chosen area of geekdom by talking to you for a few minutes or worse, just by looking at you, it's over the top. However most students - especially if you're at a top university - will be geeky about something, even the girls. Being a computer geek is fine, just don't shout it out. Being a computer nerd is not fine. Walking the line is something everybody learns to do.
Having a good attitude is really, really important. You'll have to do tech support for boring or annoying guys as well as super-nice and sexy girls, just live with it. Worst faux pax you can do is wear one of those stupid "NO I WILL NOT FIX YOUR COMPUTER" t-shirts. You have a skill most people don't and acting like it makes you gods gift will totally piss people off, which is not what you want.
If you find it's taking too much time, or interfering with your work, just be honest about it. I've put off/refused to do tech support before when I've felt I haven't had the time. Especially, if you don't feel like it, just say "Sure, I'm feeling a bit knackered right now, mind if I take a look tomorrow?" - most non-CS people do not care if their computer is out of action for a few days unless they have an assignment coming up or something. Don't fall into the trap of thinking they find a working machine as important as you do. I've seen students with a totally trashed system just say "oh well, I'll take it home at the end of term and get it replaced" when the end of term is still over a month away!
Hmm, what else. Not much really - be nice, help people, but don't try too hard. Be understanding to those who find using technology hard: just being sympathetic and telling some amusing stories ( "hey, it's not so bad, I once knew somebody ...." ) can be a go
The key point is, what would a judge think? And in this case there would be no shortage of expert witnesses willing to testify that patches with no explicit license have an implicit license identical to the original codebase as a matter of course (being derived works).
The other key point is that Zed has pissed off pretty much every contributor with this move, as well as almost certainly breaking the law. I don't understand why - if he hated doing Windows binaries, he could just stop and let somebody else do them. If nobody else could do them satisfactorily, pull Windows support entirely. Screwing over contributors like this is totally bogus.
Linux isn't currently a viable alternative for the vast majority of people due to backwards compatibility issues (ie, the lack of it). Neither is Apple, especially in business.
Something that is free is no use if you can't use it.
This is really basic stuff, it worries me that people don't seem to understand it ....
I don't use Gentoo, I was talking about broken apps installed by wrong ebuilds. Wine is an obvious example of a pathologically broken ebuild, despite its popularity (winehq gets >2million hits/month).
ReiserFS and WinFS are two competing philosophies.
WinFS is a data store. It's a separate, monolithic database which stores meta-information about files on the traditional filing system.
ReiserFS is about extending existing filing system technology such that data becomes transparent and self describing allowing any kind of querying facility to be layered on top.
Let me try an explain. Let's say you have a hard disk of OpenOffice Writer documents, which you wish to query. This is hard, because the SXW format is a complex beast. To the operating system it's an opaque series of bytes. Let's see how you'd query photos embedded in these documents:
Firstly you need to locate and open the file using the POSIX fs apis. Next, use a zip library to navigate the compressed filing-system-within-a-file that zip files are, to locate one of the XML files contained within. Now load up an XML parser and navigate the XML to one of the image nodes. Unfortunately not all information is easily represented using XML: this is one such piece, so it's actually stored as a JPEG encoded binary file ... decode that, and now navigate the structures in memory to arrive at the data you want.
Word documents are not much better.
This is an extreme example, but hopefully you see my point. Look at how many APIs were required to get to the wanted information, yet they are all fundamentally the same. ZIP files are miniature filing systems which add a compression feature (and performance!). XML is a tree of nodes: hmm, kind of like a filing system. Binary structures in memory tend to be trees or graphs of information: a bit like a filing system that supports linking.
What if we could unify these APIs? What if the underlying operating systems filing system was powerful enough to be the superset of all the features these disparate APIs provided? ZIP files are used for compression, for fast access to the contents and because it makes it easy to send them via the internet and manipulate them with modern file managers. XML is used because it's an efficient way to represent a complex tree with many nodes. Binary structures are used because some stuff just can't be easily encoded as text.
But we have a problem - there are sound technical reasons why openoffice documents are not a sparse collection of files. For one, most filing systems are not fast enough: a file is an expensive thing, opening and reading them even moreso. You don't want a file for every cell in a spreadsheet, or every paragraph in a word processor document. The overhead is too high. There is another problem: files cannot be directories and vice-versa. Having each paragraph as a file may be convenient for search engines but it's not so convenient for users.
What if files could simultaneously be directories, and what if we used a filing system designed so that a 3 byte file is not an unacceptably inefficient design? What if we could decompose our elaborate file formats with our chunks and headers and streams and DIRENTs into a tree of files all accessed via the POSIX APIs: open(), read(), close() ?
No longer would the structure of an image embedded in a word processor document be a mysterious and opaque bytestream to other programs. Now it's trivial to trawl the content of files and index them.
You see, this is the genius of Hans Reiser. He realised that writing indexer plugins for every file format under the sun would never work, it'd never scale, it'd never give users consistently good results. The right way is to make the foundations powerful enough that the concept of file formats itself falls away: by minimizing primitives, by unifiying interfaces, the system becomes more powerful.
The technical challenges of such an approach are enormous, it can only be done because ReiserFS is not a "bet the company" move, as
Oh, not to mention there are lots of nice config tools and integration widgets in Fedora that aren't in most other distros. Even little things like a working pam_xauth can reduce the potential tech support burden.
No, the main problem is that it's too unstable. Things break. I've had to recover a few friends Gentoo systems after an emerge update blasted gcc or whatever. It doesn't happen *often* but it does happen.
With the desktop comments I was referring to KDE, explaining why I'd not go for SuSE or Mandrake which are very KDE focussed. Hence the fact that Fedora is basically the best choice right now.
In particular while "emerge" might be easy, there are an awful lot of broken packages out there. I've had to deal with random brokenness caused by bad ebuilds a lot and I have no faith in the system whatsoever. So from my perspective, there is a serious problem. If what you have works for you though, then great!
Wine imposes virtually no speed overhead. I use Office under Wine all the time, it works great.
The fundamental dilemma is simple.
The only distros on which apt-get and friends actually works reliably are the least user friendly distros.
Debian and Gentoo have serious usability problems in many areas for non-technical users. There's only one distro I'd give to my friends/family today, and that's Fedora Core. Why? Because it has a slick Gnome desktop and I don't think I should be giving non-technical users a desktop which has menu items labelled "smbUmount" and a preference in the file manager called "Minimize memory usage". Sorry. Just MHO. So that leaves Fedora or some customized install of Debian. Compiling from source is right out if only because it takes so long to upgrade the system (think security updates).
Unfortunately Fedora doesn't have many packages, and those packages it does have are often out of date. The moment a new Fedora is released, you have to upgrade too because all the packagers are now targetting the newest release. This is a pain. Often packages are broken, out of date, missing, or only in conflicting repositories. These issues are hard-wired into the system and cannot be solved by throwing manpower at it, despite what some think.
Grandparent is exactly right, we need decentralised packaging. That does not exclude apt type "one command installs", that can easily be layered on top once the fundamental infrastructure is in place. But you've got to get the basics right: build a binary once, install it anywhere.
Go read the autopackage ui vision to see what end users might one day use, then check out the screenshots gallery. Try it for yourself if you like.
Be warned though. On very recent distros (fc2, debian sid etc) the Inkscape package installs but the binaries crash very quickly. It seems some library broke ABI and didn't bump the soname.
Hey, nobody said it'd be easy ....
a) I work in bash, and up until recently there was no good bash debugger
b) I work on Wine, and a lot of apps change their behaviour when run under a debugger. The Word team seem to have learned this one the hard way. A debugger is a useful tool but very, very invasive. Some apps actually deliberately change their behaviour when run under a debugger, annoyingly enough, usually to try and stop you reverse engineering them. But suffice it to say that debuggers interfere with the apps they are debugging. I don't tend to use them much these days.
c) I have, in the past, worked a fair bit on Java web apps. These days the debugging tools and IDEs are much better but back then I used only emacs, which has very primitive Java debugging support.
So how do I debug code? Simple.
1) Logging. Logging is the lifeblood of Wine debugging. A good logging framework can make hard problems easy. This is especially useful because users can send you logs so you don't have to be on-site to debug the problem, or even able to reproduce it.
2) Look at the code. This just takes practice but these days if I know roughly what area the code is failing in I can often just stare at it for a while and figure it out. The AIX guy said the same thing. It's slow and unreliable but if you practice it, it can work great for those elusive, hard to nail bugs. This is especially true for thread-safety bugs where a debugger and even logging is pretty much guaranteed to change the nature of the bug.
These techniques are far from foolproof and I'm not implying that debuggers are "bad" or for the "weak" or whatever, they aren't, they are a useful tool. Logging in particular can alter the timing profiles of the program significantly so anything involving races or timing can be screwed up. Just thinking about the code is obviously not foolproof.
But ... overreliance on interactive debuggers is a bad thing. It's very common in Windows developers because they've always had great interactive debugging available - I remember when I wrote Windows software using Delphi I was the same. Bugs which changed or disappeared under the debugger always caused me huge pain because I was so used to just whacking a breakpoint in there and examining the variables.
Unfortunately interactive debugging on Linux is total crap, gdb tends to go knock-kneed at anything vaguely threaded and the emacs/gdb integration isn't reliable either. Flip side is you learn to do without, which is a useful skill.
The relevant binary code involved video codecs. These are often patented and licensed from 3rd parties. Even if Phillips had invented and created the technology, that doesn't mean it's in their economic advantage to do so.
Wow, that page is wierd. How is a computer more male than female? I don't see anything particularly male about any of the software that's popular today ...
How can a perfectly reasonable question like that be "gender biased"? It sounds pretty likely to me. Most girls I know at university do humanities because they hate maths and/or science. There are girls who do engineering and physics of course, but they are the minority. They don't appear to have suffered any discrimination though, and are perfectly happy on their course.
Linux doesn't need proprietary drivers
"Linux" may not, but "Linux users" certainly do, at least until the fundamental questions of economics and patents are resolved. Until then proprietary drivers will be on the increase, just like proprietary software in userland will. If Linus doesn't recognise this reality I suspect it may eventually fork.
I don't have much sympathy. API versioning isn't rocket science, and it does not doom you to "archaic" interfaces when managed correctly. Compatibility between the major releases (2.4 series, 2.6 series) etc alone would be a major improvement. It can be as simple as introducing a new function instead of changing the prototype of a new one, or introducing padding into the structures.
Unfortunately the kernel developers have this idea that somehow the kernel is exempt from the same rules that govern userspace: if you document and expose interfaces to external code, you keep the interfaces stable.
Quite a lot of armchair coders don't understand backwards compatibility. It does not (necessarily) lead to insecurity or unreliability. I would like to be given examples of occasions where it has, in fact. It does mean managing change, but then this is what maintainership is all about.
The idea that making binary development hard increases the likelyhood of source releases is a fantasy based on an incomplete understanding of the economics involved. The most likely outcome is in fact the discontinuing of the drivers altogether, as has happened here.
The USB hook for binary modules was a real detriment to the USB subsystem. It was taken out for technical reasons.
It was not. Re-read Greg KHs emails. It was taken out because he thought it was a GPL violation (which is a gray area).
Talk about immature. He could leave it there until a new maintainer stepped forward but he'd rather have a dummy spit and stamp his feet.
He has worked for years to produce a solid driver for the community, even signed NDAs to get the relevant specs, and has coded on in the face of ridiculous instability in kernel development. It isn't easy doing such job, yet he did it anyway.
When you have contributed for so long, in the face of such a difficult API to work with, then you may have some authority to say whether his behaviour is "childish" or not. In his position I'd be pretty pissed off as well.
Hell, even Linux userspace has severe problems with keeping stable interfaces. I would hate to be a kernel developer.
So? It's still loaded, and still taking up resources (both machine and human resources). I don't see how the X-ness of it or not is relevant.
If you don't use legacy (OS 9 or earlier) applications, Carbon is out too.
No, most interesting Mac apps use Carbon. That includes iTunes, QuickTime and the Finder.
. One only needs to look at things like GtkGLArea, the Qt database additions
GtkGLArea isn't a "customized widget" as there is no equivalent in GTK at all. I never said having new widgets was bad, I said creating exact duplicates of already existing widgets that are specific to your app is bad.
Glade isn't a widget toolkit.
There are only two versions of GTK that need to be installed to run pretty much every GTK app ever written, so I'm not sure what your point is here. Certainly GTK 1.2 is not being used by new apps anymore, unlike Carbon which is being used by new apps - typically the large, industrial strength apps use it over Cocoa.
You've confused a lot of different things in your last paragraph : X is not a scripting system much as Quartz is not. Systems like AppleScript do indeed exist - the existance of KDEs DCOP scripting system clearly proves you wrong on that count. iTunes and QuickTime do not use Cocoa. Quite a few apps use custom widgets in a way that the user doesn't notice, however they are still eating up resources to exist - the calendar program does this for instance. So does Safari.
I'm afraid most of your points aren't grounded in technical reality ....
Tar.gz gives better ratios only for certain types of data and you can achieve the same effect by creating "solid" zips (the reason they compress source code better is because tarballs are always solid). So I still think ZIP is more flexible, and can achieve better compression.
Not much software starts so fast so people are always inclined to think Microsoft must "cheat", but they don't. It starts in a second on Linux via CrossOver/Wine as well which is clever because Wine itself imposes a hefty startup penalty.
I posted a more detailed summary of exactly what they do a few days ago here.
Ironically OpenOffice does preload.
Last time I checked MacOS X had at least 3 different environments loaded at once: BSD, Carbon and Cocoa + whatever custom libraries and toolkits individual apps use (and it's been shown that apps like iTunes do stuff like reinvent stock widgets internally!).
So recommending somebody who is complaining about memory usage to buy a Mac is simply mindless advocacy which just makes you look uninformed. After all, this is the OS where the stock media player is about 60MB.