I mean, really, what was the last 'innovation' that occured in the *nix/world?
Jeez... are you serious? Come on, Unix is one of the more important platforms for research, if not the most important. It is flexible, it is reliable, most of the scientific community is familiar with it. And these days it is also free!
Just talking about Linux I could point you to Berlin,
some guys with rather interesting ideas for building user interfaces. Or the Beowulf Project, for massive distributed computing. Or RTLinux (and KURT), for full featured real-time operating systems. How about ReiserFS, that takes database-like balanced trees to the filesystem level. Or SELinux, a research prototype of a high-security operating system.
And the list goes on and on (forgive me for not looking up links, go Google for these ones): SPIN (a dynamically extensible operating system written in Modula-3, runs on Linux), all the research stuff at Mosix (including distributed shared memory, grid management, network RAM and more), the Hello Project (an operating system in Standard ML atop Linux), all the emulation stuff which hardly needs to be introduced, and all the kernel work for supporting different processor architectures.
Also note BDS's Kame Project, an advanced implementation of IPV6 and IPSec; the evolutionary scheduler for Linux; the networking kernel stuff, including the QoS work; OpenBIOS; the User-mode Linux kernel. Look up also the "C10K problem" for an interesting paper on server performance, (and while you're on that, khttpd and TUX kernel webservers).
Unix gave you the Internet, for root's sake. How much more "innovative" does it needs to get?
Re:What I would like to know is...
on
The End Of Minix?
·
· Score: 1
Well, the entire UNIX culture is "loathe to change things" and is engaged in an almost paranoid nostiaga about their teletypes and VTxxx's, but I wouldn't extend that to "People" as a whole.
UNIX's attitudes towards the past nearly killed it, and are still a major impediment to adoption, so let's not put it on a pedestal.
You mean that having stable, well known user and development interfaces for decades is a bad thing? That dropping support for good, tested and functional facilities, such as the teletype (meaning tty devices), and instead having everyone redesign applications and relearn interfaces every couple of years, is the way to go? Next time you're going to tell us that it's justifiable to make everyone throw their hardware in the trash and buy a bigger box, just to run a new release of an operating system.
I'm pretty comfortable with the teletypes on my fancy new laptop, thank you very much. I don't think a GUI will substitute them, for me at least, any time soon (if ever).
That said, I think Mozilla does leave too big a footprint. I remember back in the Good Old Days you could get Netscape Navigator and Communicator as separate packages. I'd actually like a lean Mozilla browser and a separate Mozilla mail app. No webpage creation, no messenger, no chat/irc. I'll definitely keep an eye on Phoenix.
FWIW, Debian packages mozilla in separate components. You have mozilla-browser (around 9.5M) and mozilla-mailnews (1.9M), as well as several smaller packages for other components (psm, chatzilla, xmlterm, js-debugger, dom-inspector) and the mozilla-dev package with headers and stuff. So right now, Debian's Mozilla is about the same size as Phoenix.
That being said, I'm fetching Phoenix right now. Lets see how it fares against Galeon...
Right now, copying a file between two computers involves: 1. launch an ftp server, 2. copy the file to ftproot, 3. run my script which automatically logs into said server, 4 type "get ". It's not super-inconvenient, but definitely not as easy as a laser pointer drag and drop.
Why not just scp file host:/dir ?
It's easy, it's secure, and it even works for copying files between two hosts from a third one.
I'm tired of people screaming around about the scary DRM/Palladium/whatever stupid-ass control device the big companies are trying to push this week. Geez, why waste time fighting them? Let them have their toy, who cares? Are you really so upset that you won't be able to listen Britney's crap without giving up rights? Is like being afraid of Microsoft, for root's sake: if you know about computers and electronics, and use Linux, Microsoft is irrelevant. There's nothing Microsoft can do that could possibly restrict you in any way. Not even their "Palladiums" and "safe audio paths" and computers with integrated DRM... Shit, maybe I'm not ready right now to make my own computer from spare chips and stuff, but I know I could do it, if I needed to. And would have a lot of fun in the process, too.
My point is that we're the guys that create technology. They can lock everybody but us. If new soundcards only play their stuff, on their crippled systems, then we'll build our own systems and soundcards.
Now, as for stuff to listen to, here's a plan. Artists, the guys that make the music and the movies, the ones that really can give the finger at the DMCA and the rest by making their own stuff, could use some help from us technical people. Say, the problem with independent artists is that we don't know them. Try going to iuma.org. The problem is not lack of material, but the fact that, from all that big heap of music, you probably won't know what to pick.
So, how about building a MovieCritic-like database that can give you hints about what independent music you're probably going to like? Or software and documentation to make it easy for artists to record OGGs or setup Internet radio stations? Or an open PayPal-type of system to let them get money from their fans while cutting the greedy middle-man out? Or just help a local band getting online. You'll have a lot of fun, and you'll be way more damaging to the RIAA than bitching about it on Slashdot.
I don't know, I think those are cool projects, and that we should try spending more time in that kind of stuff, and less moaning about Hollywood, the RIAA or Microsoft.
I can recommend "Dreamhost". I had a couple of websites with them for almost 3 years, until very recently (moved the sites to my own servers). They are not the cheapest, I think, but they have fast pipes and very stable servers. In all that time I only remember one long downtime, of some hours, because of a hardware failure (when they fixed it and reported/apologized for it, they included a picture of the faulty component --just geeky enough). They also have a decent account administration web system, and the support guys actually have blood flowing through their brains. And yes, they're a Debian shop.
Having said that, the distro probably doesn't matter if you're not the admin. Yes, choosing Debian over RH may speak well about their skill and knowledge... but to be completely honest, I didn't notice it was Debian until a year or so.
Comparing RH to Microsoft sounds to me a bit harsh for the hats, but...
My company has recently been involved in several projects using Linux -- some from quite unlikely customers, such as the long-time Microsoft buddy that has The Way Out... but that's another story. Anyway, their consultors were pretty much learning to use Linux on the fly, and they have zero Unix background. They of course use RedHat, and they did succeed in installing a couple of Linux systems, which ended up being, well... somewhat imperfect.
Alright, yes, they were ugly and insecure and just crap overall. But then again, they were learning, so I don't blame them. I just think that they shouldn't be able to install such systems. Or at least believe that the machines were tip-top and running smoothly.
To rant even further, the thing that bothers me most about Microsoft is the idiotizing effect that has on their users. I'm sick of people mailing me 2MB worth of word documents every other day, given that my net link is rather small and I don't use Windows -- but they don't even know what they're doing. They just pressed a colourful and friendly button and poof, off it went. I just stopped trying to explain that I don't even run Windows, which makes reading their docs a pita for me.
It's like the people that just double-click on executable attachments in their mail, to get the cute sheep on their desktop (and the nasty trojan on their disk). Filtering content and babysitting software for such users is, imho, a battle lost before it starts. Fighting this requires only common sense and a bit of computer knowledge --surely no more that the bit of training you need for operating a car. If using a computer required even a small bit of computer knowledge, most of these things wouldn't happen.
But anyway, I don't blame computer-illiterate users for this state of things. I do blame companies such as Microsoft that actually encourage this ignorance by struggling to build software that even an idiot can use.
And on that account, yes, I do consider RedHat as the Microsoft of Linux, and I do hold a certain amount of disgust and resentment for their practices.
Close. The best HTML/SGML/XML editor, ever, is Emacs (with psgml).
It is also the best text/code/email editor ever made.
Yes, this sounds like a troll, I know. Sorry. about that. In all seriousness, I honestly believe what I just wrote --Emacs is the one application I can't live without).
Not trying to correct you here, or anything, I'm just a coder that uses the GIMP from time to time for web graphics and application icons and other imagery... But I'm a bit suprised from your list of features lacking in the GIMP. I'm obviously missing something, probably because I'm no graphics artist, but I'm hoping you can enlighten me (and hopefully other coders that have the time for adding the GIMP whatever is missing).
Can GIMP do ant-aliasing as well? NO
Just fired up my GIMP (v1.2.3), opened the text tool, and there it was: a checkbox for antialiasing. I also tried "stroking" a couple of selections and paths, and the results were most certainly antialiased. So what's the antialiasing that's missing?
Can GIMP handle Vector shapes? NO
There's a whole bunch of stuff used to manipulate "paths", which are nothing but vector shapes. I never used them before, but now I gave them a try, and I got beziers rather easily. Is there more to vector shapes in Photoshop?
Can GIMP handle the new advanced brush textures in PS7? NO
Haven't used Photoshop in a long time. What are they like?
Can GIMP handle print graphics (CMYK)? NO
Used Image -> Mode -> Decompose, and there it was: a dialog offering me to extract channels: RGB, HSV, CMY, CMYK and alpha. What am I missing?
Can GIMP smoothly interoperate with other powerfull graphics apps? NO
Well, I have made it work with Povray once, when I wanted 3D images from blueprints of an office floor for a monitoring system that we're cooking. I used a couple of perl scripts for that. Went rather nice. Probably that's not what you meant by "interoperate" or "powerful graphics apps", but then, could you elaborate on what you did mean?
Can GIMP automagically generate SANE reading HTML for quick web prototyping (don't diss the WYSIWYG in photoshop till you see it, it's not even close to front page)? NO
I have no idea of what you're saying here. I mean, generating HTML for showing a picture? Like an <img> tag? Isn't that just typing four lines in emacs?
The list goes on and on.
Care to elaborate on that?
The GIMP is good at what it was meant to do, be a simple tool for editing RGB images, but to compare it to photoshop is incorrect. Even the GIMP (who have done a great job) acknoledges that.
Well... I always though of Photoshop as a tool for editing RGB images. I think Adobe has Illustrator or something to handle vector drawing, something else for DTP, etc.
You'll get the coolest desktop background. The "land surface, ocean color, and sea ice" image is beautiful; the the "city lights" image looks rather false, though. Don't get the clouds image, you can get "live" clouds with a bit more hackery --think of it as a light version of Hiro's "Earth" widget in Snow Crash:-)
I wrote a script for getting a view from an viewpoint that goes round the earth every 6 hours and update the screen every 5-10 min. I wrote another to fetch an image of the clouds from the Xplanet page, which is updated every 3 hours, IIRC, from real data from the weather satellites (details and tips on the Xplanet page, or you can mail me if you want my scripts).
Check out a
screenshot
of my laptop's background, using this images (low quality scaled pic, sorry, we're very bandwidth challenged). Anyway, it sure beats a picture of the cats:-)
If you are already converting to CIELAB, try using the L component alone, not the full Delta E. That yields very good results. Probably color-blind people will have less problems with the generated colors, too, since this is about perceived brightness, not hue (but I'm not too sure about this).
I just made a quick Perl hack to test this. It generates 500 pairs of random colors, and outputs them sorted by "distance". It does so converting to LAB and then computing "distance" as abs(L1-L2). Check the output here, mail me if you want the script.
My point is that repackaging something in a nonstandard format is a massive amount of unnecessary effort. Debian won't always include that latest apps (neither can any distro), and that people should be able to create Linux packages the same way they create Solaris or OpenBSD packages. To think that different Linux distros are different OS saddens me, as it means there's a massive fragmentation that's occured on the platform ala Unix and the individual Linux OS's will never have the same impact on competitive platforms as a single Linux OS would.
The thing is, I believe, that if the package format were the only difference, the RPM-or-DEB issue would be moot. The reason why Debian use its own packaging scheme, and IMHO what makes it far superior than RedHat and others when it comes to install or upgrade things, is the Debian Policy.
When some maintainer makes a DEB, it does it according to policy. And this doesn't mean only where things get installed, it means lots of depends/conflict/suggests information, checking that the package is able to upgrade from very old versions, that if it provides an alternative for other package already installed it doesn't clobber it, that the new app gets managed by the Debian menu system... lots of nice things. And it works: dselect upgrade and install works fine, with almost no user intervention, 99% of the time, even on weird potato/woody hybrid systems such as my laptop.
I think that converting and installing DEBs in RPM-based systems should be quite easy. The opposite, however, would most likely be very hard without screwing policy, because most RPMs out there are a far cry from compliance. I've never used alien, and probably won't anytime soon. I think it would be a very bad thing for the integrity of the system as whole.
Because as a sysadmin running mainstream packages is more beneficial to prepackaged things. The combination of Apache+PHP+PostgreSQL/MySQL is far better managed outside of the Debian process. If you can live with PHP 4.0.1 for a couple years than go for it.
I see your point. And yes, I've been able to live with PHP 4.0 for a couple of years, indeed (4.0.3, actually). PostgreSQL and MySQL from potato, too --never had a problem.
Sure, sometimes I need a newer version of some package (say, openldap). For those I usually pull the woody source package and build potato binaries. Check this article, it explains the basics on doing this trick.
The easy solution there is to simply build my own.debs and that is what I'll probably do in the future when I have more time to read the packages guide.
Man, trust me, is far easier to use the woody source packages. Building debs is not trivial.
The funny part of this whole thread is that I'm replying on a machine running unstable. But the fact is I don't think I'd use Debian on a server. As a sysadmin the core release makes sense but the fact that other non-essential packages like Apache are never upgraded in a release does not. I'd rather run the mainstream release of a package with perhaps only a few modifications for install location than the Debianized patched to hell version.
Well I don't know... what exactly is that you need from the mainstream package that is missing in the one in potato? I mean, you could build a potato package from the woody sources (apt-get source apache, with woody as deb-src in sources.list), but... why?
So I currently run FreeBSD and RedHat (sigh) on my servers. I'd love to run Debian but it simply doesn't make any sense.
Funny. The only Linux I allow in my servers is Debian stable.
Personally, I'd think that making the OS easier to use would be a good idea.
Well, I don't know. Sometimes I think that the "ease of use" thing is idiotizing everyone.
I mean, did you know one of them ladies of, say, 20 years ago, working in offices with no word processor, email or the likes? They could pretty much record a conversation word for word in shorthand, and they could type amazingly fast on mechanical typewriters. I suppose it was hard to learn to write in shorthand or type fast, but they were incredibly efficient and productive once they did. Today, our average employee needs like 15 minutes to type a single letter, because he uses backspace as often as the space bar --and is completely hopeless without a spell checker.
Sometimes I wake up in the middle of the night shivering at the vision of people complaining about having to learn to read because it is too hard. Now I'm exaggerating , of course (I hope so). I guess my point is that ease of use isn't everything, and that people can use a "non-intuitive" operating system or whatever, sometimes being far more productive than the easy way would allow them to be. I mean, geez, how hard is to learn to type "ls" instead of clicking on the picture of a folder? And I think it is much faster to type "ifconfigeth0192.168.1.32up" than browsing through the intuitive GUI to the appropriate dialog to change your IP address.
I think English has much better chances than Latin to become the universal language. Perhaps not the English that we know today, but some sort of hybrid based on English.
And I'm ok with that, even though my primary language is Spanish, and I like it much more than English. Perhaps we won't live to see it, but I'm pretty sure that we'll end up with one universal language. As people all around the world get more and more connected (yes, through the Internet, that will eventually take over TV and telephone as the primary means for electronic communication), a single unified language will very probably appear, if only for practical reasons.
The only feature of Spanish that I'd really, really hope will catch up is the clean spoken-to-written, and back to spoken, transformation rules. I mean, what's the matter with you English-speaking people? Why is that you can't write your words as they sound? Your language makes it so damn hard to code speech synthesis and recognition...:-P
He says "The Unix euphoria of the early days was gone by the end of the decade, which had seen vendors choose sides and celebrate the forking of Unix into so many vendor-specific mutants."
I suppose that's what competition does. Let's not let this happen to linux. Or is it just going to happen naturally? Is it already happening?
It don't think it will ever happen. As I see it, the reason for the forking of Unix into different and incompatible flavors was that most vendors could not share their code. So everyone implemented its own version of the same software, which was, of course, not entirely compatible with their neighbor's implementation. And on top of that, each vendor did their best to get ahead of the competition by including cool features that only their flavor of Unix had.
Since Linux is free software, "vendors" don't need to rewrite anything to have a product. They can focus in fixing bugs and adding features. And every "vendor" can incorporate to its flavor of the OS the best features of the other flavors. I think this tends to keep most Linux flavors compatible with each other, if only because they share most of their code.
And my opinion is in no way
tainted by my oft-stated desire to find the man who invented Flash and Shockwave and break his
arms and legs with breezeblocks.
Amen to that, brother. If you find him, I'll be glad to hold the little bastard steady while you work on him.
Slashdot Mirror
cp -r folder /mnt/win
Shouldn't you be using Windows?
Jeez... are you serious? Come on, Unix is one of the more important platforms for research, if not the most important. It is flexible, it is reliable, most of the scientific community is familiar with it. And these days it is also free!
Just talking about Linux I could point you to Berlin, some guys with rather interesting ideas for building user interfaces. Or the Beowulf Project, for massive distributed computing. Or RTLinux (and KURT), for full featured real-time operating systems. How about ReiserFS, that takes database-like balanced trees to the filesystem level. Or SELinux, a research prototype of a high-security operating system.
And the list goes on and on (forgive me for not looking up links, go Google for these ones): SPIN (a dynamically extensible operating system written in Modula-3, runs on Linux), all the research stuff at Mosix (including distributed shared memory, grid management, network RAM and more), the Hello Project (an operating system in Standard ML atop Linux), all the emulation stuff which hardly needs to be introduced, and all the kernel work for supporting different processor architectures.
Also note BDS's Kame Project, an advanced implementation of IPV6 and IPSec; the evolutionary scheduler for Linux; the networking kernel stuff, including the QoS work; OpenBIOS; the User-mode Linux kernel. Look up also the "C10K problem" for an interesting paper on server performance, (and while you're on that, khttpd and TUX kernel webservers).
Unix gave you the Internet, for root's sake. How much more "innovative" does it needs to get?
You mean that having stable, well known user and development interfaces for decades is a bad thing? That dropping support for good, tested and functional facilities, such as the teletype (meaning tty devices), and instead having everyone redesign applications and relearn interfaces every couple of years, is the way to go? Next time you're going to tell us that it's justifiable to make everyone throw their hardware in the trash and buy a bigger box, just to run a new release of an operating system.
I'm pretty comfortable with the teletypes on my fancy new laptop, thank you very much. I don't think a GUI will substitute them, for me at least, any time soon (if ever).
I'm dumping Galeon. At least for a while.
Render is noticeably faster, and the UI feels as fast as GTK. Can't believe this thing is XUL. Amazing.
FWIW, Debian packages mozilla in separate components. You have mozilla-browser (around 9.5M) and mozilla-mailnews (1.9M), as well as several smaller packages for other components (psm, chatzilla, xmlterm, js-debugger, dom-inspector) and the mozilla-dev package with headers and stuff. So right now, Debian's Mozilla is about the same size as Phoenix.
That being said, I'm fetching Phoenix right now. Lets see how it fares against Galeon...
Why not just scp file host:/dir ?
It's easy, it's secure, and it even works for copying files between two hosts from a third one.
I'm tired of people screaming around about the scary DRM/Palladium/whatever stupid-ass control device the big companies are trying to push this week. Geez, why waste time fighting them? Let them have their toy, who cares? Are you really so upset that you won't be able to listen Britney's crap without giving up rights? Is like being afraid of Microsoft, for root's sake: if you know about computers and electronics, and use Linux, Microsoft is irrelevant. There's nothing Microsoft can do that could possibly restrict you in any way. Not even their "Palladiums" and "safe audio paths" and computers with integrated DRM... Shit, maybe I'm not ready right now to make my own computer from spare chips and stuff, but I know I could do it, if I needed to. And would have a lot of fun in the process, too.
My point is that we're the guys that create technology. They can lock everybody but us. If new soundcards only play their stuff, on their crippled systems, then we'll build our own systems and soundcards.
Now, as for stuff to listen to, here's a plan. Artists, the guys that make the music and the movies, the ones that really can give the finger at the DMCA and the rest by making their own stuff, could use some help from us technical people. Say, the problem with independent artists is that we don't know them. Try going to iuma.org. The problem is not lack of material, but the fact that, from all that big heap of music, you probably won't know what to pick.
So, how about building a MovieCritic-like database that can give you hints about what independent music you're probably going to like? Or software and documentation to make it easy for artists to record OGGs or setup Internet radio stations? Or an open PayPal-type of system to let them get money from their fans while cutting the greedy middle-man out? Or just help a local band getting online. You'll have a lot of fun, and you'll be way more damaging to the RIAA than bitching about it on Slashdot.
I don't know, I think those are cool projects, and that we should try spending more time in that kind of stuff, and less moaning about Hollywood, the RIAA or Microsoft.
I can recommend "Dreamhost". I had a couple of websites with them for almost 3 years, until very recently (moved the sites to my own servers). They are not the cheapest, I think, but they have fast pipes and very stable servers. In all that time I only remember one long downtime, of some hours, because of a hardware failure (when they fixed it and reported/apologized for it, they included a picture of the faulty component --just geeky enough). They also have a decent account administration web system, and the support guys actually have blood flowing through their brains. And yes, they're a Debian shop.
Having said that, the distro probably doesn't matter if you're not the admin. Yes, choosing Debian over RH may speak well about their skill and knowledge... but to be completely honest, I didn't notice it was Debian until a year or so.
Hope this helps.
IIRC, Solaris 8 is actually Solaris 2.8. Solaris 9 is 2.9. Not to be confused with the SunOS version, which would be 5.8 for Solaris 8...
Don't ask. I never really understood Sun's versioning.
Comparing RH to Microsoft sounds to me a bit harsh for the hats, but...
My company has recently been involved in several projects using Linux -- some from quite unlikely customers, such as the long-time Microsoft buddy that has The Way Out... but that's another story. Anyway, their consultors were pretty much learning to use Linux on the fly, and they have zero Unix background. They of course use RedHat, and they did succeed in installing a couple of Linux systems, which ended up being, well... somewhat imperfect.
Alright, yes, they were ugly and insecure and just crap overall. But then again, they were learning, so I don't blame them. I just think that they shouldn't be able to install such systems. Or at least believe that the machines were tip-top and running smoothly.
To rant even further, the thing that bothers me most about Microsoft is the idiotizing effect that has on their users. I'm sick of people mailing me 2MB worth of word documents every other day, given that my net link is rather small and I don't use Windows -- but they don't even know what they're doing. They just pressed a colourful and friendly button and poof, off it went. I just stopped trying to explain that I don't even run Windows, which makes reading their docs a pita for me.
It's like the people that just double-click on executable attachments in their mail, to get the cute sheep on their desktop (and the nasty trojan on their disk). Filtering content and babysitting software for such users is, imho, a battle lost before it starts. Fighting this requires only common sense and a bit of computer knowledge --surely no more that the bit of training you need for operating a car. If using a computer required even a small bit of computer knowledge, most of these things wouldn't happen.
But anyway, I don't blame computer-illiterate users for this state of things. I do blame companies such as Microsoft that actually encourage this ignorance by struggling to build software that even an idiot can use.
And on that account, yes, I do consider RedHat as the Microsoft of Linux, and I do hold a certain amount of disgust and resentment for their practices.
The best HTML editor, ever, is BareBone's BBEdit
Close. The best HTML/SGML/XML editor, ever, is Emacs (with psgml).
It is also the best text/code/email editor ever made.
Yes, this sounds like a troll, I know. Sorry. about that. In all seriousness, I honestly believe what I just wrote --Emacs is the one application I can't live without).
Not trying to correct you here, or anything, I'm just a coder that uses the GIMP from time to time for web graphics and application icons and other imagery... But I'm a bit suprised from your list of features lacking in the GIMP. I'm obviously missing something, probably because I'm no graphics artist, but I'm hoping you can enlighten me (and hopefully other coders that have the time for adding the GIMP whatever is missing).
Just fired up my GIMP (v1.2.3), opened the text tool, and there it was: a checkbox for antialiasing. I also tried "stroking" a couple of selections and paths, and the results were most certainly antialiased. So what's the antialiasing that's missing?
There's a whole bunch of stuff used to manipulate "paths", which are nothing but vector shapes. I never used them before, but now I gave them a try, and I got beziers rather easily. Is there more to vector shapes in Photoshop?
Haven't used Photoshop in a long time. What are they like?
Used Image -> Mode -> Decompose, and there it was: a dialog offering me to extract channels: RGB, HSV, CMY, CMYK and alpha. What am I missing?
Well, I have made it work with Povray once, when I wanted 3D images from blueprints of an office floor for a monitoring system that we're cooking. I used a couple of perl scripts for that. Went rather nice. Probably that's not what you meant by "interoperate" or "powerful graphics apps", but then, could you elaborate on what you did mean?
I have no idea of what you're saying here. I mean, generating HTML for showing a picture? Like an <img> tag? Isn't that just typing four lines in emacs?
The list goes on and on.
Care to elaborate on that?
The GIMP is good at what it was meant to do, be a simple tool for editing RGB images, but to compare it to photoshop is incorrect. Even the GIMP (who have done a great job) acknoledges that.
Well... I always though of Photoshop as a tool for editing RGB images. I think Adobe has Illustrator or something to handle vector drawing, something else for DTP, etc.
So what am I missing here?
You're quite right. Fluoridation, the most evil commie conspiracy to sap and impurify all of our precious bodily fluids...
Go get Xplanet, from xplanet.sourceforge.net, and the 2048x1024 JPEGs from visibleearth.nasa.gov, and hack a couple of scripts to run the thing from cron, on the root window.
You'll get the coolest desktop background. The "land surface, ocean color, and sea ice" image is beautiful; the the "city lights" image looks rather false, though. Don't get the clouds image, you can get "live" clouds with a bit more hackery --think of it as a light version of Hiro's "Earth" widget in Snow Crash :-)
I wrote a script for getting a view from an viewpoint that goes round the earth every 6 hours and update the screen every 5-10 min. I wrote another to fetch an image of the clouds from the Xplanet page, which is updated every 3 hours, IIRC, from real data from the weather satellites (details and tips on the Xplanet page, or you can mail me if you want my scripts).
Check out a screenshot of my laptop's background, using this images (low quality scaled pic, sorry, we're very bandwidth challenged). Anyway, it sure beats a picture of the cats :-)
If you are already converting to CIELAB, try using the L component alone, not the full Delta E. That yields very good results. Probably color-blind people will have less problems with the generated colors, too, since this is about perceived brightness, not hue (but I'm not too sure about this).
I just made a quick Perl hack to test this. It generates 500 pairs of random colors, and outputs them sorted by "distance". It does so converting to LAB and then computing "distance" as abs(L1-L2). Check the output here, mail me if you want the script.
The thing is, I believe, that if the package format were the only difference, the RPM-or-DEB issue would be moot. The reason why Debian use its own packaging scheme, and IMHO what makes it far superior than RedHat and others when it comes to install or upgrade things, is the Debian Policy.
When some maintainer makes a DEB, it does it according to policy. And this doesn't mean only where things get installed, it means lots of depends/conflict/suggests information, checking that the package is able to upgrade from very old versions, that if it provides an alternative for other package already installed it doesn't clobber it, that the new app gets managed by the Debian menu system... lots of nice things. And it works: dselect upgrade and install works fine, with almost no user intervention, 99% of the time, even on weird potato/woody hybrid systems such as my laptop.
I think that converting and installing DEBs in RPM-based systems should be quite easy. The opposite, however, would most likely be very hard without screwing policy, because most RPMs out there are a far cry from compliance. I've never used alien, and probably won't anytime soon. I think it would be a very bad thing for the integrity of the system as whole.
I see your point. And yes, I've been able to live with PHP 4.0 for a couple of years, indeed (4.0.3, actually). PostgreSQL and MySQL from potato, too --never had a problem.
Sure, sometimes I need a newer version of some package (say, openldap). For those I usually pull the woody source package and build potato binaries. Check this article, it explains the basics on doing this trick.
Man, trust me, is far easier to use the woody source packages. Building debs is not trivial.
Well I don't know... what exactly is that you need from the mainstream package that is missing in the one in potato? I mean, you could build a potato package from the woody sources (apt-get source apache, with woody as deb-src in sources.list), but... why?
Funny. The only Linux I allow in my servers is Debian stable.
Te luciste, maestro taco. Toda la felicidad del mundo, para ambos, y un tequila a su salud!
Personally, I'd think that making the OS easier to use would be a good idea.
Well, I don't know. Sometimes I think that the "ease of use" thing is idiotizing everyone.
I mean, did you know one of them ladies of, say, 20 years ago, working in offices with no word processor, email or the likes? They could pretty much record a conversation word for word in shorthand, and they could type amazingly fast on mechanical typewriters. I suppose it was hard to learn to write in shorthand or type fast, but they were incredibly efficient and productive once they did. Today, our average employee needs like 15 minutes to type a single letter, because he uses backspace as often as the space bar --and is completely hopeless without a spell checker.
Sometimes I wake up in the middle of the night shivering at the vision of people complaining about having to learn to read because it is too hard. Now I'm exaggerating , of course (I hope so). I guess my point is that ease of use isn't everything, and that people can use a "non-intuitive" operating system or whatever, sometimes being far more productive than the easy way would allow them to be. I mean, geez, how hard is to learn to type "ls" instead of clicking on the picture of a folder? And I think it is much faster to type "ifconfigeth0192.168.1.32up" than browsing through the intuitive GUI to the appropriate dialog to change your IP address.
I think English has much better chances than Latin to become the universal language. Perhaps not the English that we know today, but some sort of hybrid based on English.
And I'm ok with that, even though my primary language is Spanish, and I like it much more than English. Perhaps we won't live to see it, but I'm pretty sure that we'll end up with one universal language. As people all around the world get more and more connected (yes, through the Internet, that will eventually take over TV and telephone as the primary means for electronic communication), a single unified language will very probably appear, if only for practical reasons.
The only feature of Spanish that I'd really, really hope will catch up is the clean spoken-to-written, and back to spoken, transformation rules. I mean, what's the matter with you English-speaking people? Why is that you can't write your words as they sound? Your language makes it so damn hard to code speech synthesis and recognition... :-P
No mames.
La red es el protector? En todo caso, "la red es la protectora".
La lengua de castellano? No, no. La "lengua castellana", o "el castellano" simplemente.
No tiene qué preocupar sobre nada? Tú decir cosas bien intencionadas, pero tú estructurar tus frases con las nalgas.
It don't think it will ever happen. As I see it, the reason for the forking of Unix into different and incompatible flavors was that most vendors could not share their code. So everyone implemented its own version of the same software, which was, of course, not entirely compatible with their neighbor's implementation. And on top of that, each vendor did their best to get ahead of the competition by including cool features that only their flavor of Unix had.
Since Linux is free software, "vendors" don't need to rewrite anything to have a product. They can focus in fixing bugs and adding features. And every "vendor" can incorporate to its flavor of the OS the best features of the other flavors. I think this tends to keep most Linux flavors compatible with each other, if only because they share most of their code.
Amen to that, brother. If you find him, I'll be glad to hold the little bastard steady while you work on him.
Now, that is a radical thought :-)
But then again, I have to say that the A3000, when it first came out, was about the sexiest thing I've ever seen in my life... :-)