I've been following its development by using Rawhide on my system. Yeah, call me crazy and stupid, but the reason I'm doing it is that I *MUST* use the kernel 2.6 now, since 2.4 does not support the onboard SATA controller on my VIA 8237 southbridge. (It'd be kind of a nuisance not being able to use my hard drive under Linux, you know.)
To me, the 2.6 kernel is really almost the ONLY reason to use FC2. Yeah, GNOME 2.5 might have some nice refinements over 2.4, but they're mere incremental improvements that for the most part, I don't even notice.
The 2.6 kernel also finally has ALSA support built-in, which is another good reason to go with FC2. I find it has *MUCH* better sound support as a result. (I could never get the stock OSS drivers working satisfactorily.)
I consider Google to be a has-been at this point. Let's face it: 1) it no longer has the usefulness it once did, due to all the spam in it. 2) Google has so far failed to do anything constructive about it. 3) Google's new technologies just aren't that good anymore. Look at Froogle--I have yet to see it perform as well as things like PriceWatch. 4) Google is losing business left and right--most prominent example: Yahoo is now planning to dump Google. 5) Although many (including myself) have applauded Google in the past for sticking to its core competency of searching, it also means they have failed to take advantage of the synergies possible in a full-blown portal, like Yahoo.
I think it is high time Google either get its act together quick, or be finished off by someone who wants to do better, like Microsoft, or Alltheweb.
1) I don't see how that prevents Bochs from being used to run two OS's at once the way VMWare does.
2) Perhaps people should start contributing to Plex86, and contribute things they want to see like multiple OS support... Or is it that not enough people CARE about running multiple OS's with virtualization?
Methinks maybe the FUD people are spreading here about VMWare and its potential pricing hike after the buyout is all the more reason for people to look at Bochs.;-)
Do we really need to keep rehashing the SCO thing?
I don't think this is "rehashing," myself. As I see it, this story does merit its own posting, rather than just an in-passing comment.
The fact of the matter is, even though this whole SCO suit may be groundless, even though it will probably go down in flames in the end, and even though it is generally something we tend to laugh at, I think it is still imperative that the whole open source community watch this very carefully. After all, even if SCO loses, it still has enormous and far-reaching legal implications for the open source community, whether we like it or not. If the GPL is tested in court (which looks inevitable right now), and SCO wins (which is unlikely), it could seriously weaken the GPL. Likewise, if SCO loses, it could strengthen the GPL. Either way, that is *HUGE*, if you ask me.
I think I should also point out that Ask Jeeves also own Teoma, which is absolutely nothing to be sneezed at.
Not only that, but Microsoft has a world-class research arm with Microsoft Research. With Microsoft Research's world-class research, and Microsoft's deep pockets, you can bet that any improvements Teoma would need to compete with Google WILL be made.
I know there are a lot of geeks out there who will blast this effort as unnecessary--they are the same people who believe the best HTML editor is really a text-editor with an HTML quick-reference sheet handy... These are some of the very same people who loathe the idea of ANYTHING that might pollute the open source world with Windows-like things--in short: anything that infringes on their idea of Unix-like purity. Sure, I too can edit HTML myself if I really wanted to.
However, I think this effort is a HUGE leap forward, not only because it is all open source, but because it is one more tool in the open source arsenal that can be used to fight back at the Microsoft camp.
The fact of the matter is, there are a LOT of people out there for whom FrontPage is absolutely indispensible. These are some of the same people who will be asking a very pointed and straightforward question about migrating to Linux: "Will Linux run something like Microsoft Office?" Just as we need an Office suite like OpenOffice or StarOffice, I think it is high time we had a complete website authoring tool. People from all walks of life, both those in the professional world as well as those doing it just as a hobby, could benefit.
You state your position very well, and I am gratified that we can converse this way without trolling.:-)
I just wanted to clarify a few points for you: I am NOT trying to bash Debian or its installer just for the sake of it. I am just trying to point out what technical shortcomings I see in it.
There is absolutely *NO* distribution I can think of that's perfect. However, I am a big fan of trying to unify things in the Linux world and to do away with unnecessary differences wherever possible. I think THE single biggest problem the open source community faces is the almost total lack of coordination of technical efforts, and the almost total lack of any cohesive standards. Small wonder that some commercial companies are refusing to support Linux, on the grounds that "there's no standard Linux." This kind of situation, where we have such a plethora of distributions, all with their own packaging standards, their own file hierarchy standards, and their own installers, is well-known for being a NIGHTMARE when it comes to support.
That being said, I think this port of Anaconda to Debian looks like a very promising first step towards creating a unified cross-distro installer. It is my opinion that this has *enormous* potential for further unifying the Linux community (and the LSB).
Wrong. I'll bet anaconda can knock the socks off of the new debian-installer. Anaconda also can be text-mode or GUI, and theoretically, it should be able to run on ANY hardware architecture Python runs on.
I agree apt-get is fairly effective; I used to use it before I discovered yum. But yum is truly superior, in my opinion. (For RPM-based distros, anyway.) Yum can also upgrade the entire distribution; I've done that before, too. (Of course, I think it is a pity packages all have such interwoven dependencies that require such a tool, but that's a different matter altogether.)
As to the problems of hardware autodetection: Sure, there will always be fringe cases, but should we be lowering our standards just to accomodate those fringe cases, or should we be aiming HIGHER, by trying to FIX what's wrong with autodetecting those fringe cases, or at least making it handle those cases more gracefully? After all, should we eliminate GUI's altogether, simply because they don't work well on some hardware? Should we eliminate Debian or Red Hat altogether, since they aren't perfect? Obviously no--that would be absurd. Eliminating hardware autodetection because IT isn't perfect is equally absurd, in my opinion.
As to the Debian installer: I do apologize for incorrectly citing that, but if it truly is regarding the new one, that only helps PROVE my point that the old installer was terrible. Otherwise, why are you now suddenly trying to defend their effort to write a NEW one? And if you agree a new one is better, then what's wrong with Anaconda as opposed to that new installer?
1) The interface seems slow and clumsy. I think there's plenty of room for streamlining.
2) It would be nice to have it autodetect my hardware. Granted, I *could* set it up myself if I really wanted to. (Yes, I *AM* a real geek.) However, I don't see any good reason why I should have to do it myself. After all, if the hardware *CAN* be identified automatically and easily, why not do it, and save the work of having to do it manually?
3) The Debian installer asked too many questions.
4) As I say, MOST of all, it is even difficult for the Debian developers to work with. See this if you don't believe me: http://www.mail-archive.com/debian-boot@lists .debi an.org/msg27830.html
As to the total automation of RPM, I am referring to the fact that with dpkg, it typically pops up configuration screens everytime I install something. That has got to be done away with, in my opinion. It makes installing a batch of things MUCH too slow.
Also, please don't confuse apt-get with dpkg. apt works with RPM as well. In fact, I can do even better than you can, by using yum. With yum, there is even a daemon that works automatically!
Well folks, I don't know about you, but I think this is a welcome development. I have been using Red Hat for a LONG time now (>5 years), and at one point, I decided to give Debian a try, after hearing how it was supposed to be the REAL "power users" distro. Well, Debian's installer truly is horrible. Not only is it inflexible and difficult to use from the end-users' standpoint, but I understand that MOST OF ALL, it is even difficult for the Debian developers to work with.
Personally, I think an Anaconda-based Debian installer could be a *HUGE* boom, for 3 reasons:
1) It would make Debian easier to install. In my opinion, "ease of use" != "less powerful." Even developers and power users can benefit, in my opinion, by having things be difficult just for the sake of difficulty.
2) Having a distribution-neutral Anaconda would be a huge boom for the Linux community, and potentially for strengthening the LSB.
3) I'm not so sure I like the idea of having.debs and.rpms together on all Linux systems, but I think it would be nice to see some of the best features of each packaging system melded together--maybe calling the result "dpm." (lol) Seriously; RPM has some nice features (like its TOTAL automation), and DPKG also has some nice features (like OR'd dependencies. I like the idea of that much better than RPM's file dependencies).
We all know there are plenty of dependency resolvers like APT and YUM and UP2DATE, etc.; I think these should ultimately be melded together, too. Why do we need 2 different package formats, and 3 different dependency resolvers/updaters?
Look people, there is more than one possibility here--sure, maybe SunnComm realized there was no way they could win...
But the other possibility is that maybe they also realized this case WOULD be a precedent-setting case, and that the courts might possibly end up striking down (at least) parts of the DMCA as an unconstitutional violation of the 1st Amendment if this case were brought before the courts.
I see no difference between this and the fear some people have of the GPL being tested in court for the same reason. It makes no difference to me whether it comes from the "open source" crowd OR the "corporate" crowd... Which raises another point: why is it that some people have this warped idea that ANYTHING "corporate" is like a four-letter word? For heavens sake people, it is time to stop whining, and start REALLY innovating.
I'm sorry, but that phrase "good little consumers" is already an indication of an arrogant, condescending attitude toward consumers. This is EXACTLY what causes consumers to revolt in the end, and it is precisely this attitude that I am complaining about. There is no question Verisign broke things, and should never have done it in the first place. There is absolutely no question they showed a wreckless disregard for the DNS admins, but who are you to complain when you are doing EXACTLY the same thing, only with consumers?
I'm sorry, but I think you are a little off-base here. After all, it is *USERS*, NOT syadmins, who are at the end of the computing comsumption chain. If you even think about it for one minute: what is computing technology designed first and foremost for? Answer: To meet computing needs that USERS have. It is NOT designed just to please other geeks. Now I realize I'll probably get modded as a troll, even though I am *NOT* excusing Verisign's attitude by any means... But I do think this attitude that "we geeks demand strict adherence to standards, even if it means disabling features that might make life easier for end users" is a bit of a troll, too.
I actually met Richard Stallman once at the Linuxworld Expo in San Jose. I asked for his autograph on a flyer the GNU folks were distributing there, and guess what? He refused to give it to me, saying he would rather I *BUY* an Emacs manual, and let him sign that, so I would be supporting the GNU project. I was broke at the time, so I could not afford to do this, needless to say. My question is, does this new GDL mean that RMS intends to GDL all of his Emacs documentation? Does that mean I can get an Emacs manual for free by copying someone else's, and getting RMS to sign THAT???
Slashdot Mirror
I've been following its development by using Rawhide on my system. Yeah, call me crazy and stupid, but the reason I'm doing it is that I *MUST* use the kernel 2.6 now, since 2.4 does not support the onboard SATA controller on my VIA 8237 southbridge. (It'd be kind of a nuisance not being able to use my hard drive under Linux, you know.)
To me, the 2.6 kernel is really almost the ONLY reason to use FC2. Yeah, GNOME 2.5 might have some nice refinements over 2.4, but they're mere incremental improvements that for the most part, I don't even notice.
The 2.6 kernel also finally has ALSA support built-in, which is another good reason to go with FC2. I find it has *MUCH* better sound support as a result. (I could never get the stock OSS drivers working satisfactorily.)
Just my 2 cents, for what it's worth.
3 things I can think of:
1) Extra pipeline stages (and all the baggage that entails)
2) Extra L1 and L2 cache
3) 64-bitness
Don't be so sure about Google--people also once said the same things you are saying about Netscape.
I consider Google to be a has-been at this point. Let's face it: 1) it no longer has the usefulness it once did, due to all the spam in it. 2) Google has so far failed to do anything constructive about it. 3) Google's new technologies just aren't that good anymore. Look at Froogle--I have yet to see it perform as well as things like PriceWatch. 4) Google is losing business left and right--most prominent example: Yahoo is now planning to dump Google. 5) Although many (including myself) have applauded Google in the past for sticking to its core competency of searching, it also means they have failed to take advantage of the synergies possible in a full-blown portal, like Yahoo.
I think it is high time Google either get its act together quick, or be finished off by someone who wants to do better, like Microsoft, or Alltheweb.
1) I don't see how that prevents Bochs from being used to run two OS's at once the way VMWare does.
2) Perhaps people should start contributing to Plex86, and contribute things they want to see like multiple OS support... Or is it that not enough people CARE about running multiple OS's with virtualization?
Methinks maybe the FUD people are spreading here about VMWare and its potential pricing hike after the buyout is all the more reason for people to look at Bochs. ;-)
Do we really need to keep rehashing the SCO thing?
I don't think this is "rehashing," myself. As I see it, this story does merit its own posting, rather than just an in-passing comment.
The fact of the matter is, even though this whole SCO suit may be groundless, even though it will probably go down in flames in the end, and even though it is generally something we tend to laugh at, I think it is still imperative that the whole open source community watch this very carefully. After all, even if SCO loses, it still has enormous and far-reaching legal implications for the open source community, whether we like it or not. If the GPL is tested in court (which looks inevitable right now), and SCO wins (which is unlikely), it could seriously weaken the GPL. Likewise, if SCO loses, it could strengthen the GPL. Either way, that is *HUGE*, if you ask me.
I think I should also point out that Ask Jeeves also own Teoma, which is absolutely nothing to be sneezed at.
Not only that, but Microsoft has a world-class research arm with Microsoft Research. With Microsoft Research's world-class research, and Microsoft's deep pockets, you can bet that any improvements Teoma would need to compete with Google WILL be made.
I know there are a lot of geeks out there who will blast this effort as unnecessary--they are the same people who believe the best HTML editor is really a text-editor with an HTML quick-reference sheet handy... These are some of the very same people who loathe the idea of ANYTHING that might pollute the open source world with Windows-like things--in short: anything that infringes on their idea of Unix-like purity. Sure, I too can edit HTML myself if I really wanted to.
However, I think this effort is a HUGE leap forward, not only because it is all open source, but because it is one more tool in the open source arsenal that can be used to fight back at the Microsoft camp.
The fact of the matter is, there are a LOT of people out there for whom FrontPage is absolutely indispensible. These are some of the same people who will be asking a very pointed and straightforward question about migrating to Linux: "Will Linux run something like Microsoft Office?" Just as we need an Office suite like OpenOffice or StarOffice, I think it is high time we had a complete website authoring tool. People from all walks of life, both those in the professional world as well as those doing it just as a hobby, could benefit.
Hello CentrX,
:-)
You state your position very well, and I am gratified that we can converse this way without trolling.
I just wanted to clarify a few points for you: I am NOT trying to bash Debian or its installer just for the sake of it. I am just trying to point out what technical shortcomings I see in it.
There is absolutely *NO* distribution I can think of that's perfect. However, I am a big fan of trying to unify things in the Linux world and to do away with unnecessary differences wherever possible. I think THE single biggest problem the open source community faces is the almost total lack of coordination of technical efforts, and the almost total lack of any cohesive standards. Small wonder that some commercial companies are refusing to support Linux, on the grounds that "there's no standard Linux." This kind of situation, where we have such a plethora of distributions, all with their own packaging standards, their own file hierarchy standards, and their own installers, is well-known for being a NIGHTMARE when it comes to support.
That being said, I think this port of Anaconda to Debian looks like a very promising first step towards creating a unified cross-distro installer. It is my opinion that this has *enormous* potential for further unifying the Linux community (and the LSB).
Wrong. I'll bet anaconda can knock the socks off of the new debian-installer. Anaconda also can be text-mode or GUI, and theoretically, it should be able to run on ANY hardware architecture Python runs on.
I agree apt-get is fairly effective; I used to use it before I discovered yum. But yum is truly superior, in my opinion. (For RPM-based distros, anyway.) Yum can also upgrade the entire distribution; I've done that before, too. (Of course, I think it is a pity packages all have such interwoven dependencies that require such a tool, but that's a different matter altogether.)
As to the problems of hardware autodetection: Sure, there will always be fringe cases, but should we be lowering our standards just to accomodate those fringe cases, or should we be aiming HIGHER, by trying to FIX what's wrong with autodetecting those fringe cases, or at least making it handle those cases more gracefully? After all, should we eliminate GUI's altogether, simply because they don't work well on some hardware? Should we eliminate Debian or Red Hat altogether, since they aren't perfect? Obviously no--that would be absurd. Eliminating hardware autodetection because IT isn't perfect is equally absurd, in my opinion.
As to the Debian installer: I do apologize for incorrectly citing that, but if it truly is regarding the new one, that only helps PROVE my point that the old installer was terrible. Otherwise, why are you now suddenly trying to defend their effort to write a NEW one? And if you agree a new one is better, then what's wrong with Anaconda as opposed to that new installer?
What was horrible about the Debian installer:
s .debi an.org/msg27830.html
1) The interface seems slow and clumsy. I think there's plenty of room for streamlining.
2) It would be nice to have it autodetect my hardware. Granted, I *could* set it up myself if I really wanted to. (Yes, I *AM* a real geek.) However, I don't see any good reason why I should have to do it myself. After all, if the hardware *CAN* be identified automatically and easily, why not do it, and save the work of having to do it manually?
3) The Debian installer asked too many questions.
4) As I say, MOST of all, it is even difficult for the Debian developers to work with. See this if you don't believe me:
http://www.mail-archive.com/debian-boot@list
As to the total automation of RPM, I am referring to the fact that with dpkg, it typically pops up configuration screens everytime I install something. That has got to be done away with, in my opinion. It makes installing a batch of things MUCH too slow.
Also, please don't confuse apt-get with dpkg. apt works with RPM as well. In fact, I can do even better than you can, by using yum. With yum, there is even a daemon that works automatically!
Well folks, I don't know about you, but I think this is a welcome development. I have been using Red Hat for a LONG time now (>5 years), and at one point, I decided to give Debian a try, after hearing how it was supposed to be the REAL "power users" distro. Well, Debian's installer truly is horrible. Not only is it inflexible and difficult to use from the end-users' standpoint, but I understand that MOST OF ALL, it is even difficult for the Debian developers to work with.
.debs and .rpms together on all Linux systems, but I think it would be nice to see some of the best features of each packaging system melded together--maybe calling the result "dpm." (lol) Seriously; RPM has some nice features (like its TOTAL automation), and DPKG also has some nice features (like OR'd dependencies. I like the idea of that much better than RPM's file dependencies).
Personally, I think an Anaconda-based Debian installer could be a *HUGE* boom, for 3 reasons:
1) It would make Debian easier to install. In my opinion, "ease of use" != "less powerful." Even developers and power users can benefit, in my opinion, by having things be difficult just for the sake of difficulty.
2) Having a distribution-neutral Anaconda would be a huge boom for the Linux community, and potentially for strengthening the LSB.
3) I'm not so sure I like the idea of having
We all know there are plenty of dependency resolvers like APT and YUM and UP2DATE, etc.; I think these should ultimately be melded together, too. Why do we need 2 different package formats, and 3 different dependency resolvers/updaters?
Look people, there is more than one possibility here--sure, maybe SunnComm realized there was no way they could win...
But the other possibility is that maybe they also realized this case WOULD be a precedent-setting case, and that the courts might possibly end up striking down (at least) parts of the DMCA as an unconstitutional violation of the 1st Amendment if this case were brought before the courts.
I see no difference between this and the fear some people have of the GPL being tested in court for the same reason. It makes no difference to me whether it comes from the "open source" crowd OR the "corporate" crowd... Which raises another point: why is it that some people have this warped idea that ANYTHING "corporate" is like a four-letter word? For heavens sake people, it is time to stop whining, and start REALLY innovating.
I'm sorry, but that phrase "good little consumers" is already an indication of an arrogant, condescending attitude toward consumers. This is EXACTLY what causes consumers to revolt in the end, and it is precisely this attitude that I am complaining about. There is no question Verisign broke things, and should never have done it in the first place. There is absolutely no question they showed a wreckless disregard for the DNS admins, but who are you to complain when you are doing EXACTLY the same thing, only with consumers?
I'm sorry, but I think you are a little off-base here. After all, it is *USERS*, NOT syadmins, who are at the end of the computing comsumption chain. If you even think about it for one minute: what is computing technology designed first and foremost for? Answer: To meet computing needs that USERS have. It is NOT designed just to please other geeks. Now I realize I'll probably get modded as a troll, even though I am *NOT* excusing Verisign's attitude by any means... But I do think this attitude that "we geeks demand strict adherence to standards, even if it means disabling features that might make life easier for end users" is a bit of a troll, too.
I actually met Richard Stallman once at the Linuxworld Expo in San Jose. I asked for his autograph on a flyer the GNU folks were distributing there, and guess what? He refused to give it to me, saying he would rather I *BUY* an Emacs manual, and let him sign that, so I would be supporting the GNU project. I was broke at the time, so I could not afford to do this, needless to say. My question is, does this new GDL mean that RMS intends to GDL all of his Emacs documentation? Does that mean I can get an Emacs manual for free by copying someone else's, and getting RMS to sign THAT???