Why software Sucks and What you Can Do About It Debate

[Tsuroerusu]

Sounds like OpenSuse has the right idea, but I think until you actually force the average user to do something, they will ignore it. This is why I like the Ubuntu method of popping up a box and greying out the background until the user answers it. Otherwise they ignore it and try to hit Cancel or the close button.

When has Ubuntu started to do that? I just did a for-fun installation of Ubuntu today, and the updater just gave me one of those balloon messages.

I sort of like the idea of having KDE be the power user's customizable interface, while Gnome gives one a more simplistic, uncluttered interface. I think having multiple interfaces just makes sense, targeting different users and also allowing them to try different things and learn from one another.

I don't want to get into a pissing match/flamewar over this, I have been in one of those before and it doesn't get us no where, but I just plainly disagree that KDE has a cluttered or bloated interface. I agree that there are a few things they definitely need to organize better, such as their control center and some of the settings in Konqueror needs some better organization, NOT removal.

Tarball, I'm not sure that an interface that's easy to use straight away can be made at all, much less be made easily. There are just too many concepts for which I've been unable to find decent parallels in the real world of Joe User.

Trying to explain the hierarchy of files is quite troublesome, and one that I believe very few people understand. Ask the average user to find a file using Nautilus, Konqueror, Finder, or Explorer, and they'll likely struggle to do so. Even explaining what I would think of as a simple search utility like Deskbar + Beagle or Spotlight proves challenging. Where in the real world does the average user structure folders within folders for many levels or construct a hierarchy or tree system to organize something? Where in the real world does one search for something in the same manner one does so on a computer?

I completely agree, people can argue until they're green in the face about a computer needing to be as easy to use so that a child that just came out of a woman's vagina can use it without needing a single word of instruction. That is just not possible, it's not possible to hop into a car and know how to use it without some basic fundamentals of understanding.
I couldn't use a calculator before I had explain what minus, plus, multiplication and division was. Same thing with a computer, I need to learn what an icon, a mouse, a keyboard and all those things were, mind you that was back in 1994 on Windows 3.1/95, and computers are much easier to use today, but the basic point is the same, you still need a few fundamentals of understanding.

Getting users to think in terms of applications is something I find terribly difficult. I continually ask people at work what program they used to open some old Mac file without an extension, and they never have the slightest idea. Whether it was Word or Pagemaker, they honestly did not know the difference. Sure, the toolbars where different, and the interface was different, but they didn't really know why. They just clicked on the file on the desktop and used whatever windows popped up. I've never seen an interface that was task-oriented, but I doubt that such an interface would be much better. I think an interface has to be learned, because computers are really unlike anything else in the world, and to use them efficiently means learning to think in a different way. Perhaps there is some perfect interface waiting to be developed, but I really doubt it would be easy to create, and I would guess it would take decades of work, by which time kids will have grown up with the wimp interface and not really care to change.

Good to hear these ideas, gets my brain working.

Yeah I was gonna say, in 20 years, everyone have grown up with computers! Even today, even those stupid "get-piss-drunk-on-friday-night-and-fuck-some-random-guy-i-meet-on-a-street-corner" 14 year old girls can easily understand how to even install a freaking game on Windows!!!

Oh yeah, I forgot to mention that I find it very consistent with my ideas that Troels has to basically force people to use Firefox by hiding Internet Explorer or making the big, blue E point to Firefox instead. It illustrates people's resistance to change. I would do that, but I worry about them finding an Internet Explorer only site for work or something, and being left out in the cold. Running IE within a Firefox tab is far too complex a concept, so unless that were pulled off seamlessly for IE-only sites, it would be too difficult to work.

When I said that I meant that the people I couldn't get to accept a change, you know people who shit their pants in fear when a freaking dialog box come up as they click the print button, for those I need to kind of force them to use Firefox. However that is quite rare actually, when I tell somebody at school to use it, they don't think twice about it, they just say to themselves "Oh well, that thing does the same thing, no problem!" and get on with finding some random idiot, through the danish equivalent of Xanga and Myspace, to invite into bed on Friday or Saturday night!

Troels has to basically force people to use Firefox by hiding Internet Explorer or making the big, blue E point to Firefox instead.

That's what Dan did when we visited the Crapple store in Braintree, MA. He installed Firefox on all the machines and had the Safari icon point to Firefox instead.

Dude that's AWESOME, you should seriously go back and check to see if Firefox is still there!

That's beautiful. In that case, not using Safari won't affect people. Pretty much ever site that works on Safari will work fine on Firefox, if not better.

Safari is built around the KHTML rendering engine, although Konqueror is surperior to Safari and Finder by light years, if a comparison could even be made!

I am quite surprised the ipod has not increased sale of mac osx, i guess because the ipod works on windows.

Dude, you really need to turn into one of those symbiotes from the Spider-Man universe and litterally go with me to school! The iPod has increased both awareness and interest in Macintosh (Crap'in'tosh) computers.

The only reason its ported natively to mac osx is because apple as a brand alone is pretty big.

I am glad Norway have banned the ipod because apple think they can just force there drm on anyone and be accepted.

Norway bans ipod link
http://blogs.pcworld.com/digitalworld/a ... tlaws.html

"European legislators have been giving DRM considerable attention for a while, but Norway has actually gone so far as to declare that Apple's iTunes store is illegal under Norwegian law."

Norway has declared iTunes illegal, not the iPod itself, but I can tell you this:

"The Consumer Council believes Apple has only three options: it can license Fairplay to any manufacturer that wants iTunes songs to play on its machines; it can co-develop an open standard with other companies; or it can abandon DRM altogether."
Source: http://www.theregister.com/2007/01/24/a ... in_norway/

[Tsuroerusu]

I disagree with outlawing iTunes, though. I know the intentions are good, but banning something really doesn't change things. Apple's not going to change their ways because they've been banned, and the music industry is really to blame for forcing DRM on people. Apple has lock-in with their DRM, but it's far less onerous than Microsoft Office's format locking their customers in. There doesn't seem to be a principle that's being fairly applied there, unless they've outlawed Office as well. Microsoft's DRM has the exact same problem, too, are they going to get rid of that as well?

Dude, I know that some recording companies have actually approached Apple wanting them to remove the DRM from their music, but Apple refused!!
In the start the recording companies wanted DRM, but now Apple is just forcing it upon everything.
Norway has SEVERAL times asked Apple to license their DRM to other companies so that consumers weren't locked to the iPod, Apple refused every damn time, so now the hammer comes down and outlaws iTunes because Apple is such a bunch of monopolistic bastards!

[CptnObvious999]

Wow, I can't believe this but I actually agree with basically everything Troels said, especially the part where it requires a little bit of prior knowledge to use a computer.

Right now I think KDE and GNOME are fairely easy to use and in the past years I have seen things improve a lot on the linux desktop. During that time I have seen little to no features being removed, if anything even more have been added. Yet usability has gotten better. My conclusion: more usability != less features.

Every desktop has room for improvement including Linux and I don't think that it will lead to lossing features because the Linux community is damn stuborn about keeping them (as they should be). I am simply saying some rearranging, renaming, and categorization could help immensly.

I bet you no one here can list even 10 features that have been dropped in any Linux app because of usability. If that is the case I don't see how people can get so worked up about something that you fear could happen as Linux gains popularity.

[Tsuroerusu]

Wow, I can't believe this but I actually agree with basically everything Troels said, especially the part where it requires a little bit of prior knowledge to use a computer.

OH SH!T, WTF?!?! QUICK CALL THE COPS!!!

Right now I think KDE and GNOME are fairely easy to use and in the past years I have seen things improve a lot on the linux desktop. During that time I have seen little to no features being removed, if anything even more have been added. Yet usability has gotten better. My conclusion: more usability != less features.

Yep, I come to the same conclusion, I have used KDE every day for over 2½ years (By June 21st it will be three years), so I know the environment pretty well, and it's only a few months ago I found out where you can define a default web browser, email client and terminal emulator. It's barried inside the KDE Control Center as "KDE Components" and you'd never expect to tell KDE that you'd wanna use a GTK browser in a section called "KDE Components". GNOME has a setting called "Preferred Appliaction" where you can tell it what browser, email client and terminal emulator to use, much much better on this one. Also one thing that annoys me in Konqueror is that I can't have the bookmarks toolbar only in the web browser part, but it sticks around in file management mode, that's where the context of what I am doing is heavily involved, and this is exactly some of the things the KDE developers will be revisiting with KDE 4.0.

Every desktop has room for improvement including Linux and I don't think that it will lead to lossing features because the Linux community is damn stuborn about keeping them (as they should be). I am simply saying some rearranging, renaming, and categorization could help immensly.

Yep, if KDE would rename that "KDE Components" to like "Default Application" or something like that, that would be a HUGE improvement.

I bet you no one here can list even 10 features that have been dropped in any Linux app because of usability. If that is the case I don't see how people can get so worked up about something that you fear could happen as Linux gains popularity.

Actually I can tell you something, try installing SLED 10 and look at Banshee's preferences dialog, then install openSUSE 10.2 which carries a much newer version of Banshee and look at it's preferences dialog, DAMN has this thing been dumbed down to the point where it's almost depressing. If people can manage to use iTunes or Windows Media Player, I don't think they'll have any problems with that earlier version of Banshee.

[TankCatNinjaFish]

Three years without an update seems like an awful long time just for security concerns. Most security vulnerabilities I see in Linux aren't remote, but all it takes is one to screw up someone's experience. I think a distribution targeting home users should attempt automatic updates on user's computer, but unless it's done with zero user-interaction, and either perfect or easily reversible, it's a bit of a scary thought. One Xorg breakage and people may never trust it again. I still want to hear more about Ulteo, and find out whether it might be great or just a bunch of hype. Most users don't want to administer their computers, and I'd like to see someone figure out a way to do so for them.

Part of the problem is that we've all been programmed to think that 3 years is "an awful long time" for anything. It isn't. IMHO any software, including OS's, which can't survive *at least* three years with it's default settings (no patching or updating at all) and normal intended use is a catastrophic failure and should not be used.

I guess it just demonstrates the general immaturity of our industry that people expect software to routinely fail. I'm not just talking about MS either. How long would an unpatched Mysql or Apache installation survive in the wild? And despite popular belief, it *is* possible to write secure software which can be proven mathematically to do what it's supposed to and nothing else (google for software verification).

[Vogateer]

Sounds like OpenSuse has the right idea, but I think until you actually force the average user to do something, they will ignore it. This is why I like the Ubuntu method of popping up a box and greying out the background until the user answers it. Otherwise they ignore it and try to hit Cancel or the close button.

When has Ubuntu started to do that? I just did a for-fun installation of Ubuntu today, and the updater just gave me one of those balloon messages.

I should be more detailed. Ubuntu doesn't do what I mentioned for software updates, it simply does that for perhaps a few dialog boxes. The only one I can think of right now is when it asks for the password for administrative access. I'd like it to do so whenever updates are ready, forcing the user to take action, and requiring absolutely no action on their part to initiate the update process.

I sort of like the idea of having KDE be the power user's customizable interface, while Gnome gives one a more simplistic, uncluttered interface. I think having multiple interfaces just makes sense, targeting different users and also allowing them to try different things and learn from one another.

I don't want to get into a pissing match/flamewar over this, I have been in one of those before and it doesn't get us no where, but I just plainly disagree that KDE has a cluttered or bloated interface. I agree that there are a few things they definitely need to organize better, such as their control center and some of the settings in Konqueror needs some better organization, NOT removal.

That's obviously an area of subjectivity, but I do believe that KDE has improved quite a bit on this one, and the newer applications don't seem to display the toolbar button mania that Konqueror has. I did go in once and edit some file that changed Konqueror's toolbar to my liking, but it changed at the next update, and I was too lazy to make updates to the changes I'd previously made, and just ignored the toolbar. Even with that complaint, Konqueror is a fantastic app.

As you mentioned, the KDE Control Center has the inappropriately named "KDE Components" instead of "Preferred Applications" which might not be such a big deal, but I do feel like the Control Center is terribly cluttered and generally a mess, so if anything is poorly named it makes it damned difficult to find. I also think keeping everything in the K menu gives a cluttered feel in comparison to the Gnome panel, which splits things into three menus. Sometimes Gnome really pisses me off, but I do think their interface has a lot of good ideas.

Oh yeah, I forgot to mention that I find it very consistent with my ideas that Troels has to basically force people to use Firefox by hiding Internet Explorer or making the big, blue E point to Firefox instead. It illustrates people's resistance to change. I would do that, but I worry about them finding an Internet Explorer only site for work or something, and being left out in the cold. Running IE within a Firefox tab is far too complex a concept, so unless that were pulled off seamlessly for IE-only sites, it would be too difficult to work.

When I said that I meant that the people I couldn't get to accept a change, you know people who shat their pants in fear when a freaking dialog box come up as they click the print button, for those I need to kind of force them to use Firefox. However that is quite rare actually, when I tell somebody at school to use it, they don't think twice about it, they just say to themselves "Oh well, that thing does the same thing, no problem!" and get on with finding some random idiot, through the danish equivalent of Xanga and Myspace, to invite into bed on Friday or Saturday night!

Dude, pretty much every person I help is like that!

And despite popular belief, it *is* possible to write secure software which can be proven mathematically to do what it's supposed to and nothing else (google for software verification).

I'll look it up, but at first glance that's already setting off my skepticism alarm. Even if it's possible to use software verification to somehow perfect an application, it doesn't mean it's practical and if it's not practical it doesn't make it terribly useful. I mean, it sounds like finding a loophole in human fallibility. Anyhow, if it were that practical, it's hard to imagine programmers simply ignoring it for no good reason. If software verification means it takes 10 times longer to develop a similar featureset, I think I'll take my chances and take advantage of the rapid development while making backups.

[dann]

I echo that useability != less features. But I think that we need to be smart with those features.

Anyway, there seems to be a fear of "dumbing down" Linux in order to cater to the masses. I don't think that has to be the case, but I do think there needs to be more emphasis on incorporating what the masses "require" into some distrobutions (not necessarily all).

It has been commented much in the past that it would be nice if OEM offerred pre-installed Linux. Well, that is not going to happen until said pre-installed Linux gets adopted by a wider audience. We may get a few bones with systems sold sans an OS, but forget about Linux support. The only offerring we had from a major OEM was an incredibly expensive laptop geared towards engineers. I guess their market analysis indicated that the largest group of who would buy a laptop with Linux installed were these engineers.

Right now Linux does a damn good job catering to it's users - geeks. I think it can do a damn good job catering to general users too; a better job than Windows can and maybe eventually OS X.

[TankCatNinjaFish]

I'll look it up, but at first glance that's already setting off my skepticism alarm. Even if it's possible to use software verification to somehow perfect an application, it doesn't mean it's practical and if it's not practical it doesn't make it terribly useful. I mean, it sounds like finding a loophole in human fallibility. Anyhow, if it were that practical, it's hard to imagine programmers simply ignoring it for no good reason. If software verification means it takes 10 times longer to develop a similar featureset,

Rapid development and sloppy coding are different. And verification is not a hard concept to understand or apply: verifying a program (or function or procedure or any chunk of code) means taking the program you've already written, and proving that the behavior of the program is, for a certain subset of possible inputs and environments, predictable... for all other inputs and environments, the program fails gracefully. And if you can't verify the behavior of a program, it's considered broken.

For example, if you're grabbing an environment variable, you *cannot* assume that it'll be one of a few certain values. If you accept any sort of input from the outside world, you account for all possible inputs. If you're accessing the n-th element of an array, crashing is cannot be considered an acceptable form of failing.

I think I'll take my chances and take advantage of the rapid development while making backups

It's not you whose taking the chance, it's whoever is using your program.

[greggh]

I get the feeling that alot of this talk about Linux needing to be pre-installed by vendors is shortsighted. Dann, you mentioned on the last show that alot of people expect to be able to buy an OS on a storeshelf. Even Balmer has said that Vista is probably the last shrinkwrapped boxed Windows product. As the years roll by it will become standard practice for consumers to buy, download and install apps through the web. At that point, the idea of downloading and isntalling your OS (even if it is free) won't seem as crazy and foreign as it does today, even to "stupid users". I think time is on Linux's side.

[hellonorman]

Right now Linux does a damn good job catering to it's users - geeks. I think it can do a damn good job catering to general users too; a better job than Windows can and maybe eventually OS X.

What are your thoughts on how this would be accomplished?

[CptnObvious999]

Right now Linux does a damn good job catering to it's users - geeks. I think it can do a damn good job catering to general users too; a better job than Windows can and maybe eventually OS X.

What are your thoughts on how this would be accomplished?

I know you are asking Dann but here's what I think Linux needs:

• Better video editing applications
• More tutorials built into the applications
• Some more general usability fixes (like the example Troels gave)
• More games and well known applications such as WoW and Photoshop
• Fully functional to open almost anything you throw at it including videos, audio, and microsoft documents
• Hardware compatibility, when I plug in a printer it should popup a window asking me if I want to set it up, and various mp3 players and other hardware should be fully functional (this is the area I have seen the most improvement in)
• Better office suite, sure openoffice and koffice are great but they could still use a lot of work.
• Better networking support, exchange, samba, and configuring my wifi card should be a easier task then it is now.

Thats all I can think of right now, I might add more later.

[hellonorman]

I know you are asking Dann but here's what I think Linux needs:

• Better video editing applications
• More tutorials built into the applications
• Some more general usability fixes (like the example Troels gave)
• More games and well known applications such as WoW and Photoshop
• Fully functional to open almost anything you throw at it including videos, audio, and microsoft documents
• Hardware compatibility, when I plug in a printer it should popup a window asking me if I want to set it up, and various mp3 players and other hardware should be fully functional (this is the area I have seen the most improvement in)
• Better office suite, sure openoffice and koffice are great but they could still use a lot of work.
• Better networking support, exchange, samba, and configuring my wifi card should be a easier task then it is now.

Thats all I can think of right now, I might add more later.

I agree that linux needs all those things. But you could have made that same list years ago.

If getting more general users is the goal I believe the solution is more about integration and attitude.

I have too many ideas floating around in my head to write clearly right now but once I sort them out I'll try to explain what I mean.

[Tsuroerusu]

Sounds like OpenSuse has the right idea, but I think until you actually force the average user to do something, they will ignore it. This is why I like the Ubuntu method of popping up a box and greying out the background until the user answers it. Otherwise they ignore it and try to hit Cancel or the close button.

When has Ubuntu started to do that? I just did a for-fun installation of Ubuntu today, and the updater just gave me one of those balloon messages.

I should be more detailed. Ubuntu doesn't do what I mentioned for software updates, it simply does that for perhaps a few dialog boxes. The only one I can think of right now is when it asks for the password for administrative access. I'd like it to do so whenever updates are ready, forcing the user to take action, and requiring absolutely no action on their part to initiate the update process.

This sounds like a bad idea to me, because for example my mom, she literally gets scared when something unfamiliar pops up, because she's afraid that she's gonna "break something". Let alone be the fact that she doesn't know shit about what updates really mean, and then we're back to your point that an interface that even newborns could use about 2 mins. after they've come out of a woman's vagina, just can't be made, to use a computer you need some fundamentals of understanding, just like with even a stupid calculator, you need to know what minus, plus, multiplication and division are in order to use it.

I sort of like the idea of having KDE be the power user's customizable interface, while Gnome gives one a more simplistic, uncluttered interface. I think having multiple interfaces just makes sense, targeting different users and also allowing them to try different things and learn from one another.

I don't want to get into a pissing match/flamewar over this, I have been in one of those before and it doesn't get us no where, but I just plainly disagree that KDE has a cluttered or bloated interface. I agree that there are a few things they definitely need to organize better, such as their control center and some of the settings in Konqueror needs some better organization, NOT removal.

That's obviously an area of subjectivity, but I do believe that KDE has improved quite a bit on this one, and the newer applications don't seem to display the toolbar button mania that Konqueror has. I did go in once and edit some file that changed Konqueror's toolbar to my liking, but it changed at the next update, and I was too lazy to make updates to the changes I'd previously made, and just ignored the toolbar. Even with that complaint, Konqueror is a fantastic app.

As you mentioned, the KDE Control Center has the inappropriately named "KDE Components" instead of "Preferred Applications" which might not be such a big deal, but I do feel like the Control Center is terribly cluttered and generally a mess, so if anything is poorly named it makes it damned difficult to find. I also think keeping everything in the K menu gives a cluttered feel in comparison to the Gnome panel, which splits things into three menus. Sometimes Gnome really pisses me off, but I do think their interface has a lot of good ideas.

I wouldn't really call the KDE Control Center cluttered, just very unorganized and not very well labeled. Because it's not like there's too many things crammed into there, it's just that a lot of them are labeled in a way that new users don't know what means. "Window Behavior" is not something a new user is gonna know what is, but then again, what other label could things such as that one have.

Oh yeah, I forgot to mention that I find it very consistent with my ideas that Troels has to basically force people to use Firefox by hiding Internet Explorer or making the big, blue E point to Firefox instead. It illustrates people's resistance to change. I would do that, but I worry about them finding an Internet Explorer only site for work or something, and being left out in the cold. Running IE within a Firefox tab is far too complex a concept, so unless that were pulled off seamlessly for IE-only sites, it would be too difficult to work.

When I said that I meant that the people I couldn't get to accept a change, you know people who shat their pants in fear when a freaking dialog box come up as they click the print button, for those I need to kind of force them to use Firefox. However that is quite rare actually, when I tell somebody at school to use it, they don't think twice about it, they just say to themselves "Oh well, that thing does the same thing, no problem!" and get on with finding some random idiot, through the danish equivalent of Xanga and Myspace, to invite into bed on Friday or Saturday night!

Dude, pretty much every person I help is like that!

I didn't exactly need to know what you're doing on Friday night!

And despite popular belief, it *is* possible to write secure software which can be proven mathematically to do what it's supposed to and nothing else (google for software verification).

I'll look it up, but at first glance that's already setting off my skepticism alarm. Even if it's possible to use software verification to somehow perfect an application, it doesn't mean it's practical and if it's not practical it doesn't make it terribly useful. I mean, it sounds like finding a loophole in human fallibility. Anyhow, if it were that practical, it's hard to imagine programmers simply ignoring it for no good reason. If software verification means it takes 10 times longer to develop a similar featureset, I think I'll take my chances and take advantage of the rapid development while making backups.

[/quote]
You guys should take a look over at www.openbsd.org, that is actually some of the things that they do to secure their OS, check this out: http://en.wikipedia.org/wiki/Strlcpy

I echo that useability != less features. But I think that we need to be smart with those features.

Anyway, there seems to be a fear of "dumbing down" Linux in order to cater to the masses. I don't think that has to be the case, but I do think there needs to be more emphasis on incorporating what the masses "require" into some distrobutions (not necessarily all).

It has been commented much in the past that it would be nice if OEM offerred pre-installed Linux. Well, that is not going to happen until said pre-installed Linux gets adopted by a wider audience. We may get a few bones with systems sold sans an OS, but forget about Linux support. The only offerring we had from a major OEM was an incredibly expensive laptop geared towards engineers. I guess their market analysis indicated that the largest group of who would buy a laptop with Linux installed were these engineers.

Right now Linux does a damn good job catering to it's users - geeks. I think it can do a damn good job catering to general users too; a better job than Windows can and maybe eventually OS X.

Dann instead of just talking theries please try to give some specific examples, or else this conversation is gonna be meaningless, because I have had a lot of people use both GNOME or KDE, and they have absolutely no problem using the desktop, it's either an app that's missing (Photoshop for example), a dictionary program, or some hardware not working (Newer ATi craphics cards), it's not a matter of usability in most cases. I can put my mom in front of KDE or GNOME, with an icon on the desktop labeled "Firefox" and she can do all she needs to do on a computer. She can check her email through GMail, which is pretty much all she does in a daily basis, right now she uses Google's Picasa to organize photos, but F-Spot or digiKam (Preferably digiKam) would do just fine.

I get the feeling that alot of this talk about Linux needing to be pre-installed by vendors is shortsighted. Dann, you mentioned on the last show that alot of people expect to be able to buy an OS on a storeshelf. Even Balmer has said that Vista is probably the last shrinkwrapped boxed Windows product. As the years roll by it will become standard practice for consumers to buy, download and install apps through the web. At that point, the idea of downloading and isntalling your OS (even if it is free) won't seem as crazy and foreign as it does today, even to "stupid users". I think time is on Linux's side.

I agree. With Vista, Microsoft is gonna enable people to upgrade to a more expensive version over the internet. Even on Home Basic you get the features of Ultimate, they are just disabled, then you can use your creditcard over the web and upgrade, and boom all the feature you just paid for are there. Also I think with programs such as Firefox and Thunderbird being well known these days in the Windows world, it's a lot easier to explain to some users why the programs are gratis. When I need to explain to people why Firefox is gratis, I actually pull something from places like gnu.org and say something like "The people who made this software, wrote it to be used by as many users as possible, and not preventing users from using it, giving it away is just the easiest way of accomplishing that goal. It enables more efficient feedback and development".

Right now Linux does a damn good job catering to it's users - geeks. I think it can do a damn good job catering to general users too; a better job than Windows can and maybe eventually OS X.

What are your thoughts on how this would be accomplished?

Exactly what I mean, theories are great, but practical examples are better at demonstrating a problem or solution.

Right now Linux does a damn good job catering to it's users - geeks. I think it can do a damn good job catering to general users too; a better job than Windows can and maybe eventually OS X.

What are your thoughts on how this would be accomplished?

I know you are asking Dann but here's what I think Linux needs:

[*]Better video editing applications

It's still early-days but check this out: http://vivia-video.org/
I could imagine seeing something pop as we get into KDE 4.0, because by the developers will no longer have to struggle with basic multimedia things such as playing a sound, which with aRts took at least 30 lines of code, with Phonon, it's freaking 4. With the basic stuff dead simple, developers could focus on the more complicated stuff, however Phonon is not gonna be doing advanced stuff, it will eventually get a capture API, however I doubt that that's what is need for a GarageBand/Premiere type of thing.

[*]More tutorials built into the applications

A wizard really wouldn't hurt every now and then.

[*]Some more general usability fixes (like the example Troels gave)

KDE 4 my friend, KDE 4.

[*]More games and well known applications such as WoW and Photoshop

Well, Second Life is there, which is certainly a step in the right direction, since that is something a lot of people are hooked on. However we really need World of Warcrack for Linux.

[*]Fully functional to open almost anything you throw at it including videos,
audio, and microsoft documents

Eventually people will upgrade to Office 2007 and we'll get docx (OpenXML) files from family and friends, which is what Novell figured since they will be adding OpenXML support to OpenOffice.org.

[*]Hardware compatibility, when I plug in a printer it should popup a window asking me if I want to set it up,

Dude, you really need to quit staring at that GCC Screen Saver all day long and try openSUSE, it already does this!

and various mp3 players and other hardware should be fully functional (this is the area I have seen the most improvement in)

When I plug my USB key (Which essentially is the same as the majority of digital music players) into my box with openSUSE, a dialog box opens up and asks me if I want to browse it using Konqueror or do nothing. If I open it in Konqueror, stick in a CD I can just drag and drop the tracks I want ripped to my MP3 player, and it does it. Pretty great. The iPod, Zune and some Creative players are very very tricky because they use proprietary means of storing music on the device. Since the iPod is so popular even folks like Novell has people working on that as part of their job (Such as the folks writing Banshee), and there's that libgpod or whatever it is, that all apps can use to do iPod stuff. Eric Raymond's examples of we needing to give up our freedoms to allow people to use their iPods are flat out bullsh!t, I can use my iPod with Amarok and Banshee without any closed-source binary blob crap.

[*]Better office suite, sure openoffice and koffice are great but they could still use a lot of work.

We need a Microsoft Office clone, not an alternative, but an actual clone, because unless people who have the necessary cash are forced to, they don't want to take the time to learn anything new. A lot of my friends are students, and whenever pirating Office is way too complicated and cumbersome, OpenOffice looks pretty good to them, however to my aunt, it's just not feasible using Microsoft Office at work (Which she has done for years) and come home and use OpenOffice.org.

[*]Better networking support, exchange, samba, and configuring my wifi card should be a easier task then it is now.[/list]

Exchange support is gonna be really tough as that would require a lot of reverse engineering, which I think is better spent at reverse engineering graphics drivers. Samba, well, Novell is working on that. With WiFi we also need a sh!tload of reverse engineering.

Thats all I can think of right now, I might add more later.

GPLv3 my friend, GPLv3!!

[Vogateer]

Rapid development and sloppy coding are different. And verification is not a hard concept to understand or apply: verifying a program (or function or procedure or any chunk of code) means taking the program you've already written, and proving that the behavior of the program is, for a certain subset of possible inputs and environments, predictable... for all other inputs and environments, the program fails gracefully. And if you can't verify the behavior of a program, it's considered broken.

I did some looking around, and found some documents talking about this from 1988. I'm curious, if this is something that's so practical to do and has so many benefits, then why is it not used by many people?

I think I'll take my chances and take advantage of the rapid development while making backups

It's not you whose taking the chance, it's whoever is using your program.

I was mainly thinking of myself as the user of the program taking my chances. My python scripts aren't going to be used by anyone else at the moment.

Troels gave a link to another program, strlcpy, though the article mentioned that sometimes using it introduces even more bugs. I thought the whole thing with OpenBSD was the incredible amount of time they spend doing code auditing.

[Tsuroerusu]

Troels gave a link to another program, strlcpy, though the article mentioned that sometimes using it introduces even more bugs.

Strlcat and strlcpy are not programs, they're actual C functions:

"The strlcpy function, developed by Todd C. Miller and Theo de Raadt for use in the C programming language, is intended to replace the function strcpy and provide a simpler and more robust interface than strncpy. It is designed to copy the contents of a string from a source string to a destination string."

The article mentions this:

"Furthermore, some, including Ulrich Drepper, argue that use of strlcpy and strlcat can introduce more bugs than they remove[2]; consequently, these functions have not been added to the GNU C Library."

I'm not drinking OpenBSD kool-aid or anything, and I'm not a C programmer at all (In fact, I don't even know how to do hello world in C), but I'm inclined to believe what the OpenBSD guys are saying about this topic, particularly because they live, breathe and die in security.

I thought the whole thing with OpenBSD was the incredible amount of time they spend doing code auditing.

You can read about what OpenBSD does in terms of security on these pages:
http://www.openbsd.org/security.html
http://www.openbsd.org/crypto.html

To summarize, this is what they do:

Full Disclosure - Like many readers of the BUGTRAQ mailing list, we believe in full disclosure of security problems. In the operating system arena, we were probably the first to embrace the concept. Many vendors, even of free software*, still try to hide issues from their users.

Audit Process - Our security auditing team typically has between six and twelve members who continue to search for and fix new security holes. We have been auditing since the summer of 1996. The process we follow to increase security is simply a comprehensive file-by-file analysis of every critical software component. We are not so much looking for security holes, as we are looking for basic software bugs, and if years later someone discovers the problem used to be a security issue, and we fixed it because it was just a bug, well, all the better.

New Technologies - As we audit source code, we often invent new ways of solving problems. Sometimes these ideas have been used before in some random application written somewhere, but perhaps not taken to the degree that we do. strlcpy() and strlcat(), Memory protection purify, Chroot jailing ......

"Secure by Default" - To ensure that novice users of OpenBSD do not need to become security experts overnight (a viewpoint which other vendors seem to have), we ship the operating system in a Secure by Default mode. All non-essential services are disabled. As the user/administrator becomes more familiar with the system, he will discover that he has to enable daemons and other parts of the system. During the process of learning how to enable a new service, the novice is more likely to learn of security considerations.

Cryptography - And of course, since the OpenBSD project is based in Canada, it is possible for us to integrate cryptography.

*) I recently ran into this: http://www.centos.org/modules/tinyconte ... .php?id=17
"Before public disclosure of any security issue, concern, or comment regarding CentOS, our website, our public servers, or mirrors, please email it to the following:"
I personally think the CentOS guys' approach is stupid to put it nicely, and deliberately making your users unaware of security problems makes them more vulnerable.

I'd encourage you read the two pages I linked to, because it really is amazing what this quite small group of people are doing. Remember that OpenBSD is not a "distribution" in the sense that we in the GNU/Linux (I use this term to make a clear distiction between the kernel and the entire OS) world know, OpenBSD is an entire operating system, they have their own coreutils (ls, cp, mv, chmod, chown ....), they have their own C library etc. etc. they even maintain their own version of Apache because they don't agree with the attribution clause that's in the license of Apache 2.x, you can find some interesting OpenBSD answers to that

The OpenBSD project are also the developers of OpenSSH, which I think everyone reading this thread has used at least once in their life, I personally use it every single day. OpenSSH is like some sort drug I think, once you learn all the cool things it can do, you're hooked!


Sorry for pimping OpenBSD for a moment there, I use it on a Dell OptiPlex (800 MHz Pentium 3, 256 MB of RAM) that I use to host a wiki that I use to write stuff in. Last year I wrote an article as a school project in English that I intend to publish under a Creative Commons license when I get it "bugfixed" aka properly spell- and gramma checked.
One thing I'd really like to try would be to run Asterisk on OpenBSD.