I'd rather have a motherboard manufacturer that properly supports www.coreboot.org - then we'd really have the freedom to quick boot linux, or use openfirmware, or jump straight into grub without touching the disk, or...
Right now, it seems some manufacturers (Tyan, MSI) help out a bit for some types of motherboards, because it might help them sell some clusters. But it would be nice to have the option on the latest/greatest desktop boards too.
Take an image of the harddrive, before making any changes. There's just so many ways to lose information (browser cache etc) when you're trying to do something like this.
Well, I would claim coding in Flex is relatively easy. Just use the SDK with the text editor of your choice. All open source, no worries about bloat/crashing/etc.
As for the ui components, I find them quite well done. I just wish they'd used the same set of UI components in Flash CS3 - the two different sets of UI components don't mesh well.
And yea, sometimes some documented features need workarounds, but on the whole I feel the quality is about the same as a lot of other runtimes/languages I've coded for. I also think they've come a long way since the old Flash days - they cleaned up the runtime libs quite a bit.
The only thing I regret slightly is that AS1.0 used to be a nice little prototype based language, and AS2.0/3.0 being class-based (not bad, but very different - much more like Java). But that's Javascript/ECMAScript's doing, not really Adobe's.
I feel another distributed computing project coming up: after SETI@home, and Folding@home, maybe this would make for an interesting way to get statistics on cosmic rays?
You don't need a router to block your printer/filesharing. You need a firewall. Better get used to it too - with ipv6 we'll probably need to change our habits to use firewalls to delineate our "trusted networks" instead of using ip's in a private address space.
But seriously, what alternatives do people have instead of Vista for a pre-installed system. Seriously?
Apple maybe? Every week I hear from another client, friend or relative who's switching because of Vista. Not that I think MacOSX or Apple is flawless though, they've got their own set of issues. And in a few isolated cases, switching is impossible because an app isn't available for the Mac platform (I've heard from accountants and architects with AutoCad).
But apart from those, the majority is saying goodbye to Windows and not looking back.
It's true that there is no "default user" but that shouldn't stop projects for finding really, really good defaults.
Some random reasons: a) when people first start using a new system, the habits they form usually correspond with the defaults - the better the defaults are, the more likely it is that the users won't want to deviate from them. The point is that users start tweaking their system in lots of different ways because they're searching for good defaults (for them), when the original defaults fail them.
b) sometimes it really is better to have the user adapt to the system: see Palm's Graffiti handwriting recognition vs the Newton's. While the former was a lot more restrictive, and while there is no "default handwriting" shared by people all over the world, it still worked better than the latter. You could argue that's because the Newton and a lot of similar systems since then were really crap, but you have to realize it's a trade-off - sometimes simplicity is better than flexibility - especially when creating a "perfect" system is very very difficult.
c) every choice you offer forces the user to think about it, hesitate, etc. If you see screenfulls of radio buttons, checkboxes etc it becomes difficult to find the setting you need. If most of the settings are something you want to change maybe once or twice in your "desktop lifetime", and some of the settings you need more regularly, you're always going to be bothered with the rarely used ones when trying to find the commonly used ones. I've always thought adding checkboxes and dialogs to be laziness on the part of the programmer. If you can avoid asking the question altogether, don't bother the user - get out of the way of the user's focus.
d) I'm advocating simplicty in UI, not in the underlying apps and libraries. I still believe that if you really really want to tweak something, you should be able to do it. I don't understand why gconf and its editor gets so much hate: you can easily tweak it there, and there's even a "short" and "long description" of the keys in there. For the one or two times I really want tweak something really specific, I can find it fast enough in there.
You wrote the comment I was going to write. I echo the feeling this "reclaiming" is revisionist. The only time I heard the term "cracker" really being used was wrt "software cracking" to bypass its (copy) protection of some sort. And the people that did it happily referred to themselves as crackers, too.
Like you say, the ESR "cracker" is really an ex post facto invention - and basically a pejorative term. Well, RMS used "cracker" in 1983 too, iirc - but there too it was a conscious and artificial attempt at changing the vocabulary to distinguish the "good" (ie. themselves), from the "bad" (ie. people with different morals/ideals or different/less (?) skills). I guess my problem with the word is that it's mainly used by people who like to feel they're better than other people.
I feel "hacker" is a neutral term completely orthogonal to the malevolent or benevolent intentions of the hacker. This is also why I like the terms white/gray/black hat, because these conversely are only about ethics/intentions and not about skill.
I think the point of the Yes Men's "performances" is to make the people present realize that they can easily go over their own ethical boundaries. Maybe they'll pay more attention to their own ethics after that.
Only, it's not "their code" - it's derived (well, if they made changes to the dlls, that is) from other people's code, and so must respect those people's copyright. They always had the option to start from scratch, too.
One major thing to remember is that in the open source world, developers are almost always users first. So if you don't have any users, you're going to have a really hard time attracting developers.
So limit your scope for your first release, and get something working and usable ready first. Only once things are sort-of working for a first generation of users should you advertise it a bit: first impressions do count.
While it's not "ready" they clearly seem to have plans to have it running on linux.
It probably still won't run on Linux/PPC or ARM or even Windows2000 (funny that), but I guess it's as close to "cross platform" as I see any commercial company get.
I had good experiences with Gandi. Last I checked, they had the best Terms of Service, and while they weren't exactly the cheapest, they weren't expensive either.
They're quite straight forward, no ad-bombardement or spamming; they just do what they say they would do. The few times I had a problem (not because of them), they were always quite helpful. I reserve the right to change my mind, but up till now, I've been nothing but happy (based on my experience with them for many years)
The problem isn't that developers think they're good UI designers too, the problem is that EVERYONE thinks they're good UI designers.
And yes, I've seen quite a few Graphic Designers, or Industrial Designers think they can be a good UI designer too, when in actuality they were just so-so.
What's needed is a willingness to just invest a bit more time thinking about the UI, and prototyping different UIs. Discussing the different trade-offs with different people helps a lot too.
Swedes pay dearly in taxes for all the human services they have available, even if, differently from Brazil or Colombia, those services are available to everyone who is entitled to them.
So are you saying those services aren't worth the money?
Haven't seen Linus say this - he's consistently been saying that while the very big (clusters) and the very small (embedded) can be very interesting, he's mostly interested in off-the-shelf systems where all the volume is, whatever that currently is.
Just to let you know: I am actually a European. So when I say Europeans probably know more about US laws than their own countries, it's because I've actually tried to find out and understand the laws in my own country, and had a lot more trouble to get a useful analysis than the US equivalents.
Slashdot Mirror
I'd rather have a motherboard manufacturer that properly supports www.coreboot.org - then we'd really have the freedom to quick boot linux, or use openfirmware, or jump straight into grub without touching the disk, or ...
Right now, it seems some manufacturers (Tyan, MSI) help out a bit for some types of motherboards, because it might help them sell some clusters. But it would be nice to have the option on the latest/greatest desktop boards too.
Take an image of the harddrive, before making any changes. There's just so many ways to lose information (browser cache etc) when you're trying to do something like this.
Well, I would claim coding in Flex is relatively easy. Just use the SDK with the text editor of your choice. All open source, no worries about bloat/crashing/etc.
As for the ui components, I find them quite well done. I just wish they'd used the same set of UI components in Flash CS3 - the two different sets of UI components don't mesh well.
And yea, sometimes some documented features need workarounds, but on the whole I feel the quality is about the same as a lot of other runtimes/languages I've coded for. I also think they've come a long way since the old Flash days - they cleaned up the runtime libs quite a bit.
The only thing I regret slightly is that AS1.0 used to be a nice little prototype based language, and AS2.0/3.0 being class-based (not bad, but very different - much more like Java). But that's Javascript/ECMAScript's doing, not really Adobe's.
I feel another distributed computing project coming up: after SETI@home, and Folding@home, maybe this would make for an interesting way to get statistics on cosmic rays?
You don't need a router to block your printer/filesharing. You need a firewall. Better get used to it too - with ipv6 we'll probably need to change our habits to use firewalls to delineate our "trusted networks" instead of using ip's in a private address space.
Maybe tracker shouldn't do initial indexing of preexisting files while it's on battery... could be an idea for a feature request?
But seriously, what alternatives do people have instead of Vista for a pre-installed system. Seriously?
Apple maybe? Every week I hear from another client, friend or relative who's switching because of Vista. Not that I think MacOSX or Apple is flawless though, they've got their own set of issues. And in a few isolated cases, switching is impossible because an app isn't available for the Mac platform (I've heard from accountants and architects with AutoCad).
But apart from those, the majority is saying goodbye to Windows and not looking back.
It's true that there is no "default user" but that shouldn't stop projects for finding really, really good defaults.
Some random reasons:
a) when people first start using a new system, the habits they form usually correspond with the defaults - the better the defaults are, the more likely it is that the users won't want to deviate from them. The point is that users start tweaking their system in lots of different ways because they're searching for good defaults (for them), when the original defaults fail them.
b) sometimes it really is better to have the user adapt to the system: see Palm's Graffiti handwriting recognition vs the Newton's. While the former was a lot more restrictive, and while there is no "default handwriting" shared by people all over the world, it still worked better than the latter. You could argue that's because the Newton and a lot of similar systems since then were really crap, but you have to realize it's a trade-off - sometimes simplicity is better than flexibility - especially when creating a "perfect" system is very very difficult.
c) every choice you offer forces the user to think about it, hesitate, etc. If you see screenfulls of radio buttons, checkboxes etc it becomes difficult to find the setting you need. If most of the settings are something you want to change maybe once or twice in your "desktop lifetime", and some of the settings you need more regularly, you're always going to be bothered with the rarely used ones when trying to find the commonly used ones. I've always thought adding checkboxes and dialogs to be laziness on the part of the programmer. If you can avoid asking the question altogether, don't bother the user - get out of the way of the user's focus.
d) I'm advocating simplicty in UI, not in the underlying apps and libraries. I still believe that if you really really want to tweak something, you should be able to do it. I don't understand why gconf and its editor gets so much hate: you can easily tweak it there, and there's even a "short" and "long description" of the keys in there. For the one or two times I really want tweak something really specific, I can find it fast enough in there.
Maybe look at Vala then, though it's probably still at "toy-language" stage.
Hey, if Saudi Arabia and China can implement "Great Firewalls", I'm sure the US can too, don't you agree?
You wrote the comment I was going to write. I echo the feeling this "reclaiming" is revisionist. The only time I heard the term "cracker" really being used was wrt "software cracking" to bypass its (copy) protection of some sort. And the people that did it happily referred to themselves as crackers, too.
Like you say, the ESR "cracker" is really an ex post facto invention - and basically a pejorative term. Well, RMS used "cracker" in 1983 too, iirc - but there too it was a conscious and artificial attempt at changing the vocabulary to distinguish the "good" (ie. themselves), from the "bad" (ie. people with different morals/ideals or different/less (?) skills). I guess my problem with the word is that it's mainly used by people who like to feel they're better than other people.
I feel "hacker" is a neutral term completely orthogonal to the malevolent or benevolent intentions of the hacker. This is also why I like the terms white/gray/black hat, because these conversely are only about ethics/intentions and not about skill.
One doesn't exclude the other.
I think the point of the Yes Men's "performances" is to make the people present realize that they can easily go over their own ethical boundaries. Maybe they'll pay more attention to their own ethics after that.
Only, it's not "their code" - it's derived (well, if they made changes to the dlls, that is) from other people's code, and so must respect those people's copyright. They always had the option to start from scratch, too.
One major thing to remember is that in the open source world, developers are almost always users first. So if you don't have any users, you're going to have a really hard time attracting developers.
So limit your scope for your first release, and get something working and usable ready first. Only once things are sort-of working for a first generation of users should you advertise it a bit: first impressions do count.
Maybe you should read this:
o perFAQ#Does_Apollo_support_Linux
http://labs.adobe.com/wiki/index.php/Apollo:Devel
While it's not "ready" they clearly seem to have plans to have it running on linux.
It probably still won't run on Linux/PPC or ARM or even Windows2000 (funny that), but I guess it's as close to "cross platform" as I see any commercial company get.
http://righttocreate.blogspot.com/2005/12/why-drug -companies-dont-need-patents.html
I had good experiences with Gandi. Last I checked, they had the best Terms of Service, and while they weren't exactly the cheapest, they weren't expensive either.
They're quite straight forward, no ad-bombardement or spamming; they just do what they say they would do. The few times I had a problem (not because of them), they were always quite helpful. I reserve the right to change my mind, but up till now, I've been nothing but happy (based on my experience with them for many years)
The problem isn't that developers think they're good UI designers too, the problem is that EVERYONE thinks they're good UI designers.
And yes, I've seen quite a few Graphic Designers, or Industrial Designers think they can be a good UI designer too, when in actuality they were just so-so.
What's needed is a willingness to just invest a bit more time thinking about the UI, and prototyping different UIs. Discussing the different trade-offs with different people helps a lot too.
Haven't seen Linus say this - he's consistently been saying that while the very big (clusters) and the very small (embedded) can be very interesting, he's mostly interested in off-the-shelf systems where all the volume is, whatever that currently is.
a host of filesystems, maybe? truckloads of drivers?
I'm neither. I liked Resident Evil and Event Horizon, but thought the others were ... meh.
... it's still Vampire Killer to me. That's what happens if you memorize what's hidden behind every bloody candle and chest in the entire game.
Just to let you know: I am actually a European. So when I say Europeans probably know more about US laws than their own countries, it's because I've actually tried to find out and understand the laws in my own country, and had a lot more trouble to get a useful analysis than the US equivalents.