Instead of such empty claims, you are always welcome to post list of problems/missing applications and functionality.
Linux developers don't get this, and probably never will.
Unix in general is "system written by engineers for engineers". It is professional OS. It's not about some silly users pretending - but about things get done. I used to run Linux 100% of time. Many of my friends now run 100% Linux - and nobody's complaining. Or rather they complain "feature X of application Y sucks Z way" - no, not blank statements like yours. And the statement is also applicable to any OS, not just Linux. Ask me, I can provide you hundreds examples where Windows and its applications do not live up to their promises. Shall I then claim that Windows devels "don't get this"?
I have a thing against commenting all the pointless cruft flushing on us from ZDNet, yet I find that hilarious. From RTFA:
By the end of the year the OpenSolaris community will be widely recognised as larger and more active than the Linux community -and every competing OS developer community except Microsoft's will have copied the key ideas including its organisational structure, the core provisions in the community development license, and Solaris specific technologies including ZFS and Dtrace.
BUHAHAHHAHAHAHHHAHAHAHAH!!!!!!!!!
That's why I love modern IT journalism: it is self-contained and never really leave one's office desk - with notable exceptions of sponsored trips to important (from pov of sponsor) exhibitions and conferences.
Political merits. No new product/offering can win against established in market products/offerings. If that would be possible - imagine that market worked that way many users would have woken up to system with Windows replaced by BSD/Linux. But. Miracles do not happen. Traditionally, it had been observed on many occasions, new products first have to score new installations - before they can go after established players. Needless to remind that many predicted similar demise of Linux after release of OpenDarwin. No miracles did happen back then too.
Technical merits. First, "ZFS and Dtrace" are engineer's toys. Their real world worth is yet to be identified. (And identification alone costs 100GP. lol.) Second, only total idiot CTO/CIO would replace existing working system with something new. However good new stuff is, as long as existing solution works - nobody would even spend a minute looking into something new. Upgrade cycle will come - viable new options will be investigated. (People prayed for prolonged upgrade cycle - God heard their prayers and granted them the freedom with Linux.) Provided current state of affairs - very stable Linux offerings from RH/Novell/etc, best M$ Windows 2003 server product - I would expect that OpenSolaris would be... unnoticed for the whole next year. Sun might beat some PR drums - but few would consider such a young system for anything important.
To put it generally: people (and market) are very inertial. It takes time to get into new system. And at moment that "new stuff" slot of most IT people's minds is occupied with "Linux, BSD, Vista" items. The list is too long already. I take that most would consider evaluation of Vista to be most important, since that what many customers inevitably would end-up using shortly.
Provided that what I said above bear some sense, no way OpenSolaris would (i) gain enough attention to (2) attract enough (system) developers. Probably some Java developers would find that interesting - but they are very poor when it comes to system programming. OpenSolaris would remain underdog - system for hobbyists. Many talented Solaris developers (Sun discarded not so many years ago lots) went to BSD, Darwin and Linux - I do not expect them to start trusting Sun again all so suddenly nor to jump into active development.
Conclusion: OpenSolaris next year needs to concentrate on improved hardware support, improved software installation, improved desktop, improved preinstalled application package, etc - leaving world domination target to some distant future.
(*) Also I find it absolutely hilarious that system which still shipps with original 20yo "vi" to win anything - least its users hearts. Ppl, first get some decent text editor - world domination comes later.
Provide a scalable, manageable platform for collaboration and the development of Web-based business applications with Windows SharePoint Services 3.0, a versatile technology in Windows Server 2003.
And what that supposed to mean to the CIOs??? Or as developer and admin?
Well, we do not develop web services and heck I'm not sure what those are anyway. Last I heard Google doesn't use them - and it is playground of IBM/M$. From all it looks like: another non-technology crap (made up of buzz-words) created to sell something else. I have experienced many of such creations from M$ (and not only). And I'm sure many are yet to come: it is big companies, they need to sell something... BIG.
As to corporate information sharing my employer suddenly went with Wikimedia's MediaWiki (the same one which powers Wikipedia). Meeting minutes, white papers, tech notes, sales remarks, bug analysis, etc - lands in Wiki were it can be viewed, corrected and extended by others. People especially liked the page audit & version control: that was one of the reasons for the Wiki to win against conventional content management systems. Big plus is of course that it runs on Linux so it can sit on our existing file/print/raid/svn/backup servers. Also, installation was no brainer with basically everything done by already prepared SUSE rpms.
I bought myself LCD from Sharp. Good price. Good resolution. Low power consumption.
In our local store there is side-by-sede exhibit of LCDs and Plasmas (Philips' ones). There are also new LCDs with 3200 dynamic contrast ratio. They look comparable to image quality of plasmas.
General observation for 37". Good plasma & good LCD - of comparable quality, 1080p capable - all cost now +3000€. LCD at 3200 (dynamic) contrast performs very well in side-by-side comparison with plasma. In fact, I wasn't being able to tell one from another. (While normally color reproduction of plasmas stands out.)
Cheap plasmas with high power consumption and low resolution (1024x768 - still performing visually better than LCD's 1366x768) would set you back 1400€+ and improved power bill (200 v. 400 watts (yes, new plasmas are as good as LCDs at power saving)).
Cheap LCDs have now contrast ratio of 1200/1600 (dynamic one). And cost the same 1400€ as cheap plasmas do. Visually less stunning, they still suck less power. But of course, black isn't really black here.
In the end, I went with cheap brand-name LCD and am pretty happy. TV is TV - few movie and lots of TV shows. Movies need good contrast - TV shows don't.
P.S. Piece of advice. Do not buy black LCD TVs. Black frame around screen doesn't bode well with black as seen on LCD screen. Otherwise - it's okay;)
P.P.S. Word about resolutions. TV native resolution is used nowhere. I have 1366x768 and I thought that my PC would be able to take advantage of it - mistake. TVs do really support only 480i/p, 576i, 720i/p & 1080i/p - the actual number of pixels has nothing to do with it since picture is always approximated. Also, I found that actual visible pixel resolution is smaller than advertised for 720/1080 modes: edges are always left out of screen. Probably I am mistaken, but that seems to be a solution to MPEG compression introduced bugs on screen edges. I googled for it - but found no information. (But probably it is mine geforce7800gt just suck - it produces terribly unwatchable 1080i too.)
Ubuntu is claimed to be better for desktop compared to many other distros.
Friend of mine uses Mandriva/Mandrake for about five years now as his desktop. And he is pure user - not admin/developer.
I use Debian for last four years. (Before that was 2 years of SUSE and year of RedHat)
I do not like Ubuntu because it uses that crippled thing called Gnome, but Kubuntu seems to solve the problem. I keep Kubuntu up-to-date live CD just in case I would need rescue system - though I rarely have need for it.
Uhm... Well, I do not know how to respond your question. I use Linux as desktop for many years. Several my friends do the same. None runs (K)Ubuntu. I think Linux desktop in general is Ok for many people. Especially for serious people: people choose outstanding reliability of Linux over bells'n'wistles of Windows.
You may not like the specification, but you have the option to use it or not. If Vista didn't support it, you wouldn't have that option at all. So, again, where is the problem here?
Read the RTFA, it's quite entertaining reading. Many example problems are there.
Basically the RTFA spins around the Vista way of closing "analogue hole": if protected content present in system memory, all channels which do not support content protection will be downgraded. IOW, if you would try to play DRM'ed WMA music during your work - and you work with imaging applications - Vista would downgrade quality of output of (amongst other things) your imaging application. It can be anything - anything protected - and whole system will go into "content protection" mode. The author already sees the problem for sensitive application fields which depend on unfiltered channels. E.g. degradation of image quality produced by medical appliances might threaten lives.
As he states, he is more interested in costs parts of the equation, so he goes into depth to show problem in full. IOW, as user of sensitive imaging software, in order to get some sort of guarantee that Vista will not at any time stop working, one would need to: upgrade display, upgrade video card, upgrade sound card, upgrade motherboard and of course your imaging software. Whoever you are - producer or user - M$ wants you with Vista to rework every part of computing system. (And that all would still not prevent fallouts: compromised driver, some bus fluctuations might still trigger Vista's alarm and it would downgrade performance of system.) Now try to think of all involved costs - for both consumers and producers. That's not any kind of small number.
One of the examples author provides, is security system. Intruder might send piece of malware which will do nothing disruptive but launch piece of protected content, thus potentially downgrading unprotected video and audio quality. After all, security cameras are not intended for protected materials, yet they could be used e.g. for dvd ripping. The effect of the malware will be simple: guards will see distorted picture and will hear lossy audio. Thus helping criminals. All thanks to Vista security - on guard of protected content.
If all the guys who write all the pointless "let's install Linux" reviews/articles actually coordinated and made a write up about using particular distro for let say one year - encountered problems, ease to find solution, user community, security, etc - that info would be welcome by many users and also distro developers.
Throwing idea. Though modern journalism is well known for its "skin deep" nature, so I do not expect miracles.
I soon realized, however, that I generally didn't know the official repository package names for most of the apps that I wanted to download, so using apt-get quickly became a problem.
That's why aptitude's command "search" does exist.
e.g. "aptitude search sudoku" would search package names (and descriptions?) for string "sudoku". "dpkg -l '*sudoku*'" haven't really ever worked.
P.S. RTFA sucks. Judging Linux by ease of installation?? Give me a break. I use Linux precisely because (compared to Windows) I need to install it only once. And then it just works. Many of my friends use Linux precisely because of that stability - that allows people to actually concentrate on my own work. (M$Windows? You just have to reboot XP every week and reinstall it every year - to keep it running normally.)
Just pay no attention to what people here say about GUI v. CLI. GUI/mouse addicted geeks would never understand that.
I can over phone tell guy to fire up terminal and then letter-by-letter tell him commands - and he would read me their output. It takes only minutes to educate him how to do that.
Trying to do the same in GUI - Windows or GNOME - normally ends up in failure. If VNC/etc isn't available - it is next to impossible. Did anybody in helpdesks actually tried to count all the possible error messages Windows' Computer Management can produce? I lost count long time ago. Or for that matter how often GNOME changes layout of their control panel? Ah... Damm just fire up terminal, type few commands and it's done.
And there is even more to it. It is easier to teach somebody to type verbatim (or copy'n'pase) command into terminal - than to explain precisely how user needs to click his way thru the GUI. You can print command (and their output) for later reference. But to precisely describe what was done in GUI to achieve some particular task - is next to impossible. Especially in modern Windows' dynamic GUI. (Actually several my friends were collecting all CLI tips I were giving them - if not to learn, but at least to have in case if problem arises again). And my experience of working with computer-illiterate people was precisely confirmation of that GUI badness: configuration of open windows on screen can be so confusing that normal people never really are getting used to it. And if you are called on phone, how can you get out of user what's happening with his computer? Maximize button is often seen as panacea (or at least as placebo) but not really a solution. Terminal? It's DUMB. Specifically to avoid all the mess.
Does thunderbird 2 have this great feature? (here's where someone tells me its been in Thunderbird 1 for years!... )
I use Netscape^WMozilla^WMessenger^WThunderbird for years now and haven't seen any kind of such feature.
The sorting rules ("Tools" > "Message Filters..." ) are unhelpful too: they are applied only to incoming mails.
Judging from number of critical bugs open, you have little chances to convince its developers to implement something like that. They have their hands full: Tb developer base is much smaller and code base is much dirtier compared to Ff.
Try to file a bug in bugzilla. You can also send message to gmail support forums - feature sounds interesting and Google people don't skip opportunities to improve something;-)
Leave all the GUI/HTML nastiness to Tb/Netscape/Mozilla/Opera/FoxMail/Evolution/etc.
Please, please, please, do not ruin last properly working MUA out there. And do not give anyone (very very bad) ideas.
P.S. I know, my post might look funny, but it is sad reality. Tb has pretty constant number of bugs which are never getting fixed. Many were filed in times of Mozilla Messenger - and yet unfixed. As of now, Tb still cannot properly produce error message when more than single mail box is checked. If it works, it works. If there are problems - Tb would be totally unhelpful/counterproductive in finding them. mutt/fetchmail work all the time and allow to really see if there is a problem and what problem is. With Tb you might just get some error message in background - and would be left wondering for days why mail isn't coming/get mail button does nothing.
I've got nothing against PostgreSQL -- just never used it.
Let me share bit of my experience. Grab any commercial RDBM developer/user and put him to PostgreSQL and MySQL.
In my experience, Pgsql would slowly crunch any query one would throw at it. On other side MySQL would DoS your server by trying to actually run query it cannot optimize (e.g. many ORs or IN in SELECT's WHERE). And it cannot optimize hell a lot of types of queries. RTFA in fact describes MySQL performance that way: "Supports both left and right outer joins using both ANSI and ODBC syntax." Notice that no exceptions about Pgsql queries are made: plain SQL92/99 referred. DoSing server - especially if it is hosted is last one wants while developing application for clients.
Happened to me - as developer - many times. Actually all the time. Though, normally clients end up running dumb queries and are pretty happy that they can actually filter their few thousands of records. Nothing extraordinary in real life (95% of applications) is needed.
End result is good stable performance of MySQL on real applications vs. ease of development with Pgsql. YMMV.
+100. Had I any moderation points, I'd mod you up Insightful.
The point is of course that no society is ideal. Or to put it better that society is based on some ideal. Different ideals - different societies. For European civilizations common ideal is of course money. Talk is cheap so we count all in money. The ideal behind Asian civilizations... I do not know. Notably, unlike us, they value what people are saying and lying (what is perfectly Ok in EU/US) is more or less regarded as crime in Muslim world. Anybody who has friends from e.g. Iran obviously have noticed that.
Hm. I think many fashion people would brown Zune just for sake of its color - brown. Many leather and fur goods are of that color. And nobody did the color before - Apple likes traditional colors, Creative likes colors of Apple, etc.
Now that I think of it, probably it's Mrs Gates has ordered for xmas brown player for her sablefur coat. And husband complied.
My experience of apt-get has been that it has pretty terrible ways of dealing with broken dependencies in the RPM database.
I heard of such problem (though never used the apt-rpm by myself).
Most often the problem was attributed to RH/SUSE love to heavily patched software which leads to requirement to have rpm depend on precise version of the patched software (glibc, perl, python, kernel headers, etc).
Debian's packages normally try to be lax about dependencies: they do not depend on some particular version of other packages, but rather range of versions. e.g. some packaged doesn't depend on "glibc-2.3.99-cvs20041212", but rather on "glibc-2.4 and higher". (I still shiver recalling hell of manually upgrading RH/SUSE servers - after new release of glibc.)
IMHO, your problem is ideological incompatibility of apt and of rpms as provided by vendors.
On Debian with all Debian's repositories + several second party repositories, apt/aptitude does great job of finding compatible combination of packages to install: often it may resort even to downgrading some packages to satisfy dependencies and allow you to install request package.
As soon as you would try to customize installation you would find that it is not as trivial as it might be.
If you do not customize your distro - then it's fine. If you customize rarely - yum would still do fine job.
But if you have too have some latest version (e.g. I need to have latest gcc installed, so that I can test the compiler) rpm/yum become major pain in the ass. E.g. new version needs new library or new version of library and the library has it's own dependencies. Aptitude now handles everything for me nicely - yum in past was barely capable of conflicting packages handling of several repositories. Not to mention that it did that very very slowly.
And do not forget that LD_KERNEL=2.2.5 - or rpm would hang yum. You would have to kill -9 it and then clean/rebuild repository. Or it doesn't hang anymore? Or RedHat set the variable system-wide? Or RedHat finally got down to earth to fix their beloved futexes? [/sarcams]
But why then apt-get works faster than yum? Earlier versions of apt were written I think in Perl (yum is written in Python) but still yum (implemented after apt) seemed like step in wrong direction.
Also, yum is even slower than aptitude and aptitude does one hell of checks and tests - as well as pile of algorithms to solve dependency problems. (Never living with four concurrent repositories was easier in my life.)
That really surprised me when I found out that Debian's apt doesn't use database for its repository - and still worked faster than yum. And the mundane rebuilding of database one has to do for rpm (stuck with us from times of rpm database upgrade) is just not needed with apt.
P.S. Raw dpkg is slower than rpm precisely because rpm uses fast database, while dpkg has to update huge text file.
Worth to mention that apt now deprecated in favor of aptitude. Aptitude marks packages installed as part of dependency resolution and when you later remove the installed software, it would also remove all automatically installed packages.
Yum, a popular RPM-based manager (like apt, but specifically for RPM) was certainly a total piece of shit the last time I tried it.
+100. yum is dumbest and slowest tool I have ever seen. Especially when people try to pitch it against apt-get. And aptitude is ages ahead of any package management RedHat ever implemented.
+100. I've been there too. Forget about simultaneously editing document on more than two computers. (In fact same goes for M$O - though overall compatibility/portability is much better.)
I would love it just to work, but at moment PDF export is only way to have portable (in read-only sense) document.
It's probably 'coded better' with C++ and what-not, creating bloated structures and resource piggishness.
It is not. M$Office is much more optimized (by all means) product. StarOffice itself was based on previous work - so the code base was already split even before Sun acquisition. And then add development of Sun and OO.o which do not perfectly fit each other.
And Sun's following development effort which threw in Java to the backet didn't help either.
The result is buggy bloated mess. Don't argue with me. I use OOo every day. And I had read the source code.
It's free - but there is nothing more to it. ODF compatibility is still far below any usability level so all the PR talk about ODF magic is just what it is - PR talk. IOW, all OOo has now is its free beer's price: $0.00.
Is it only me who thinks that PHP after version 2 started getting so much weight and bloat that I would believe anything about how insecure it had become.
Rate new features/functions added to PHP at times seems to be exponential. Something that points to poor project management: it looks like incapable of handling the exploded PHP popularity and attention it gathered.
Though my opinion might be outdated: I was programming PHP last time when version 3 was getting its first releases.
But since he is always starts brawls, he can eventually become evil. Though I'm sure that game would not allow that.
Well, I haven't finished NWN2 yet, so I don't know. But I thought KOTOR2 was excellent in that regard. Your actions and your influence really could change your companions. So it wouldn't surprise me if NWN2 is the same.
NWN2 forums were already flooded once by that problem: you cannot attack friendly targets. Unlike in NWN1. IOW, game is just not designed to look deep into alignments. In NWN2 it seems to be much harder to become evil: there are bunch of stuff for good alignment, but real bad actions are not there.
Though, yes: just like in NWN1 expansion packs, alignment is really adjusted by your actions. (NWN1' alignment was bonker and did really nothing (besides some magic items limited to particular alignments))
But then, realistically you cannot make any real use of alignment, since that would disrupt game completely. One of the problems I have mentioned that in SP NPC are giving you quests to advance the game. Try to imagine what would happen if you as evil character (if you have decided to become one) wouldn't be able to get quest from good guy. On one side it improves reality of game play. But on other side it is hell for game developers since they would have to triple main quest: for good, for neutral and for evil characters. Though I'm already thrilled with all the prospects.
So just to conclude, NWN2 uses alignments - but not to the extent how they are used in table games. Live DMs are still unreachable by computer simulations.
Slashdot Mirror
Instead of such empty claims, you are always welcome to post list of problems/missing applications and functionality.
Unix in general is "system written by engineers for engineers". It is professional OS. It's not about some silly users pretending - but about things get done. I used to run Linux 100% of time. Many of my friends now run 100% Linux - and nobody's complaining. Or rather they complain "feature X of application Y sucks Z way" - no, not blank statements like yours. And the statement is also applicable to any OS, not just Linux. Ask me, I can provide you hundreds examples where Windows and its applications do not live up to their promises. Shall I then claim that Windows devels "don't get this"?
+10! Bravo!! ZDNet is rich source of (mis)information and (non)analytics.
I have a thing against commenting all the pointless cruft flushing on us from ZDNet, yet I find that hilarious. From RTFA:
BUHAHAHHAHAHAHHHAHAHAHAH!!!!!!!!!
That's why I love modern IT journalism: it is self-contained and never really leave one's office desk - with notable exceptions of sponsored trips to important (from pov of sponsor) exhibitions and conferences.
Political merits. No new product/offering can win against established in market products/offerings. If that would be possible - imagine that market worked that way many users would have woken up to system with Windows replaced by BSD/Linux. But. Miracles do not happen. Traditionally, it had been observed on many occasions, new products first have to score new installations - before they can go after established players. Needless to remind that many predicted similar demise of Linux after release of OpenDarwin. No miracles did happen back then too.
Technical merits. First, "ZFS and Dtrace" are engineer's toys. Their real world worth is yet to be identified. (And identification alone costs 100GP. lol.) Second, only total idiot CTO/CIO would replace existing working system with something new. However good new stuff is, as long as existing solution works - nobody would even spend a minute looking into something new. Upgrade cycle will come - viable new options will be investigated. (People prayed for prolonged upgrade cycle - God heard their prayers and granted them the freedom with Linux.) Provided current state of affairs - very stable Linux offerings from RH/Novell/etc, best M$ Windows 2003 server product - I would expect that OpenSolaris would be ... unnoticed for the whole next year. Sun might beat some PR drums - but few would consider such a young system for anything important.
To put it generally: people (and market) are very inertial. It takes time to get into new system. And at moment that "new stuff" slot of most IT people's minds is occupied with "Linux, BSD, Vista" items. The list is too long already. I take that most would consider evaluation of Vista to be most important, since that what many customers inevitably would end-up using shortly.
Provided that what I said above bear some sense, no way OpenSolaris would (i) gain enough attention to (2) attract enough (system) developers. Probably some Java developers would find that interesting - but they are very poor when it comes to system programming. OpenSolaris would remain underdog - system for hobbyists. Many talented Solaris developers (Sun discarded not so many years ago lots) went to BSD, Darwin and Linux - I do not expect them to start trusting Sun again all so suddenly nor to jump into active development.
Conclusion: OpenSolaris next year needs to concentrate on improved hardware support, improved software installation, improved desktop, improved preinstalled application package, etc - leaving world domination target to some distant future.
(*) Also I find it absolutely hilarious that system which still shipps with original 20yo "vi" to win anything - least its users hearts. Ppl, first get some decent text editor - world domination comes later.
And what that supposed to mean to the CIOs??? Or as developer and admin?
Well, we do not develop web services and heck I'm not sure what those are anyway. Last I heard Google doesn't use them - and it is playground of IBM/M$. From all it looks like: another non-technology crap (made up of buzz-words) created to sell something else. I have experienced many of such creations from M$ (and not only). And I'm sure many are yet to come: it is big companies, they need to sell something ... BIG.
As to corporate information sharing my employer suddenly went with Wikimedia's MediaWiki (the same one which powers Wikipedia). Meeting minutes, white papers, tech notes, sales remarks, bug analysis, etc - lands in Wiki were it can be viewed, corrected and extended by others. People especially liked the page audit & version control: that was one of the reasons for the Wiki to win against conventional content management systems. Big plus is of course that it runs on Linux so it can sit on our existing file/print/raid/svn/backup servers. Also, installation was no brainer with basically everything done by already prepared SUSE rpms.
I bought myself LCD from Sharp. Good price. Good resolution. Low power consumption.
In our local store there is side-by-sede exhibit of LCDs and Plasmas (Philips' ones). There are also new LCDs with 3200 dynamic contrast ratio. They look comparable to image quality of plasmas.
General observation for 37". Good plasma & good LCD - of comparable quality, 1080p capable - all cost now +3000€. LCD at 3200 (dynamic) contrast performs very well in side-by-side comparison with plasma. In fact, I wasn't being able to tell one from another. (While normally color reproduction of plasmas stands out.)
Cheap plasmas with high power consumption and low resolution (1024x768 - still performing visually better than LCD's 1366x768) would set you back 1400€+ and improved power bill (200 v. 400 watts (yes, new plasmas are as good as LCDs at power saving)).
Cheap LCDs have now contrast ratio of 1200/1600 (dynamic one). And cost the same 1400€ as cheap plasmas do. Visually less stunning, they still suck less power. But of course, black isn't really black here.
In the end, I went with cheap brand-name LCD and am pretty happy. TV is TV - few movie and lots of TV shows. Movies need good contrast - TV shows don't.
P.S. Piece of advice. Do not buy black LCD TVs. Black frame around screen doesn't bode well with black as seen on LCD screen. Otherwise - it's okay ;)
P.P.S. Word about resolutions. TV native resolution is used nowhere. I have 1366x768 and I thought that my PC would be able to take advantage of it - mistake. TVs do really support only 480i/p, 576i, 720i/p & 1080i/p - the actual number of pixels has nothing to do with it since picture is always approximated. Also, I found that actual visible pixel resolution is smaller than advertised for 720/1080 modes: edges are always left out of screen. Probably I am mistaken, but that seems to be a solution to MPEG compression introduced bugs on screen edges. I googled for it - but found no information. (But probably it is mine geforce7800gt just suck - it produces terribly unwatchable 1080i too.)
Ubuntu is claimed to be better for desktop compared to many other distros.
Friend of mine uses Mandriva/Mandrake for about five years now as his desktop. And he is pure user - not admin/developer.
I use Debian for last four years. (Before that was 2 years of SUSE and year of RedHat)
I do not like Ubuntu because it uses that crippled thing called Gnome, but Kubuntu seems to solve the problem. I keep Kubuntu up-to-date live CD just in case I would need rescue system - though I rarely have need for it.
Uhm... Well, I do not know how to respond your question. I use Linux as desktop for many years. Several my friends do the same. None runs (K)Ubuntu. I think Linux desktop in general is Ok for many people. Especially for serious people: people choose outstanding reliability of Linux over bells'n'wistles of Windows.
Read the RTFA, it's quite entertaining reading. Many example problems are there.
Basically the RTFA spins around the Vista way of closing "analogue hole": if protected content present in system memory, all channels which do not support content protection will be downgraded. IOW, if you would try to play DRM'ed WMA music during your work - and you work with imaging applications - Vista would downgrade quality of output of (amongst other things) your imaging application. It can be anything - anything protected - and whole system will go into "content protection" mode. The author already sees the problem for sensitive application fields which depend on unfiltered channels. E.g. degradation of image quality produced by medical appliances might threaten lives.
As he states, he is more interested in costs parts of the equation, so he goes into depth to show problem in full. IOW, as user of sensitive imaging software, in order to get some sort of guarantee that Vista will not at any time stop working, one would need to: upgrade display, upgrade video card, upgrade sound card, upgrade motherboard and of course your imaging software. Whoever you are - producer or user - M$ wants you with Vista to rework every part of computing system. (And that all would still not prevent fallouts: compromised driver, some bus fluctuations might still trigger Vista's alarm and it would downgrade performance of system.) Now try to think of all involved costs - for both consumers and producers. That's not any kind of small number.
One of the examples author provides, is security system. Intruder might send piece of malware which will do nothing disruptive but launch piece of protected content, thus potentially downgrading unprotected video and audio quality. After all, security cameras are not intended for protected materials, yet they could be used e.g. for dvd ripping. The effect of the malware will be simple: guards will see distorted picture and will hear lossy audio. Thus helping criminals. All thanks to Vista security - on guard of protected content.
If all the guys who write all the pointless "let's install Linux" reviews/articles actually coordinated and made a write up about using particular distro for let say one year - encountered problems, ease to find solution, user community, security, etc - that info would be welcome by many users and also distro developers.
Throwing idea. Though modern journalism is well known for its "skin deep" nature, so I do not expect miracles.
From RTFA:
That's why aptitude's command "search" does exist.
e.g. "aptitude search sudoku" would search package names (and descriptions?) for string "sudoku". "dpkg -l '*sudoku*'" haven't really ever worked.
P.S. RTFA sucks. Judging Linux by ease of installation?? Give me a break. I use Linux precisely because (compared to Windows) I need to install it only once. And then it just works. Many of my friends use Linux precisely because of that stability - that allows people to actually concentrate on my own work. (M$Windows? You just have to reboot XP every week and reinstall it every year - to keep it running normally.)
Just pay no attention to what people here say about GUI v. CLI. GUI/mouse addicted geeks would never understand that.
I can over phone tell guy to fire up terminal and then letter-by-letter tell him commands - and he would read me their output. It takes only minutes to educate him how to do that.
Trying to do the same in GUI - Windows or GNOME - normally ends up in failure. If VNC/etc isn't available - it is next to impossible. Did anybody in helpdesks actually tried to count all the possible error messages Windows' Computer Management can produce? I lost count long time ago. Or for that matter how often GNOME changes layout of their control panel? Ah... Damm just fire up terminal, type few commands and it's done.
And there is even more to it. It is easier to teach somebody to type verbatim (or copy'n'pase) command into terminal - than to explain precisely how user needs to click his way thru the GUI. You can print command (and their output) for later reference. But to precisely describe what was done in GUI to achieve some particular task - is next to impossible. Especially in modern Windows' dynamic GUI. (Actually several my friends were collecting all CLI tips I were giving them - if not to learn, but at least to have in case if problem arises again). And my experience of working with computer-illiterate people was precisely confirmation of that GUI badness: configuration of open windows on screen can be so confusing that normal people never really are getting used to it. And if you are called on phone, how can you get out of user what's happening with his computer? Maximize button is often seen as panacea (or at least as placebo) but not really a solution. Terminal? It's DUMB. Specifically to avoid all the mess.
mutt - console interface == fetchmail.
I use Netscape^WMozilla^WMessenger^WThunderbird for years now and haven't seen any kind of such feature.
The sorting rules ("Tools" > "Message Filters..." ) are unhelpful too: they are applied only to incoming mails.
Judging from number of critical bugs open, you have little chances to convince its developers to implement something like that. They have their hands full: Tb developer base is much smaller and code base is much dirtier compared to Ff.
Try to file a bug in bugzilla. You can also send message to gmail support forums - feature sounds interesting and Google people don't skip opportunities to improve something ;-)
NO! NO! NO!
Leave all the GUI/HTML nastiness to Tb/Netscape/Mozilla/Opera/FoxMail/Evolution/etc.
Please, please, please, do not ruin last properly working MUA out there. And do not give anyone (very very bad) ideas.
P.S. I know, my post might look funny, but it is sad reality. Tb has pretty constant number of bugs which are never getting fixed. Many were filed in times of Mozilla Messenger - and yet unfixed. As of now, Tb still cannot properly produce error message when more than single mail box is checked. If it works, it works. If there are problems - Tb would be totally unhelpful/counterproductive in finding them. mutt/fetchmail work all the time and allow to really see if there is a problem and what problem is. With Tb you might just get some error message in background - and would be left wondering for days why mail isn't coming/get mail button does nothing.
Let me share bit of my experience. Grab any commercial RDBM developer/user and put him to PostgreSQL and MySQL.
In my experience, Pgsql would slowly crunch any query one would throw at it. On other side MySQL would DoS your server by trying to actually run query it cannot optimize (e.g. many ORs or IN in SELECT's WHERE). And it cannot optimize hell a lot of types of queries. RTFA in fact describes MySQL performance that way: "Supports both left and right outer joins using both ANSI and ODBC syntax." Notice that no exceptions about Pgsql queries are made: plain SQL92/99 referred. DoSing server - especially if it is hosted is last one wants while developing application for clients.
Happened to me - as developer - many times. Actually all the time. Though, normally clients end up running dumb queries and are pretty happy that they can actually filter their few thousands of records. Nothing extraordinary in real life (95% of applications) is needed.
End result is good stable performance of MySQL on real applications vs. ease of development with Pgsql. YMMV.
+100. Had I any moderation points, I'd mod you up Insightful.
The point is of course that no society is ideal. Or to put it better that society is based on some ideal. Different ideals - different societies. For European civilizations common ideal is of course money. Talk is cheap so we count all in money. The ideal behind Asian civilizations... I do not know. Notably, unlike us, they value what people are saying and lying (what is perfectly Ok in EU/US) is more or less regarded as crime in Muslim world. Anybody who has friends from e.g. Iran obviously have noticed that.
Hm. I think many fashion people would brown Zune just for sake of its color - brown. Many leather and fur goods are of that color. And nobody did the color before - Apple likes traditional colors, Creative likes colors of Apple, etc.
Now that I think of it, probably it's Mrs Gates has ordered for xmas brown player for her sable fur coat. And husband complied.
Get a real shell, man!!
I heard of such problem (though never used the apt-rpm by myself).
Most often the problem was attributed to RH/SUSE love to heavily patched software which leads to requirement to have rpm depend on precise version of the patched software (glibc, perl, python, kernel headers, etc).
Debian's packages normally try to be lax about dependencies: they do not depend on some particular version of other packages, but rather range of versions. e.g. some packaged doesn't depend on "glibc-2.3.99-cvs20041212", but rather on "glibc-2.4 and higher". (I still shiver recalling hell of manually upgrading RH/SUSE servers - after new release of glibc.)
IMHO, your problem is ideological incompatibility of apt and of rpms as provided by vendors.
On Debian with all Debian's repositories + several second party repositories, apt/aptitude does great job of finding compatible combination of packages to install: often it may resort even to downgrading some packages to satisfy dependencies and allow you to install request package.
As soon as you would try to customize installation you would find that it is not as trivial as it might be.
If you do not customize your distro - then it's fine. If you customize rarely - yum would still do fine job.
But if you have too have some latest version (e.g. I need to have latest gcc installed, so that I can test the compiler) rpm/yum become major pain in the ass. E.g. new version needs new library or new version of library and the library has it's own dependencies. Aptitude now handles everything for me nicely - yum in past was barely capable of conflicting packages handling of several repositories. Not to mention that it did that very very slowly.
And do not forget that LD_KERNEL=2.2.5 - or rpm would hang yum. You would have to kill -9 it and then clean/rebuild repository. Or it doesn't hang anymore? Or RedHat set the variable system-wide? Or RedHat finally got down to earth to fix their beloved futexes? [/sarcams]
But why then apt-get works faster than yum? Earlier versions of apt were written I think in Perl (yum is written in Python) but still yum (implemented after apt) seemed like step in wrong direction.
Also, yum is even slower than aptitude and aptitude does one hell of checks and tests - as well as pile of algorithms to solve dependency problems. (Never living with four concurrent repositories was easier in my life.)
That really surprised me when I found out that Debian's apt doesn't use database for its repository - and still worked faster than yum. And the mundane rebuilding of database one has to do for rpm (stuck with us from times of rpm database upgrade) is just not needed with apt.
P.S. Raw dpkg is slower than rpm precisely because rpm uses fast database, while dpkg has to update huge text file.
Worth to mention that apt now deprecated in favor of aptitude. Aptitude marks packages installed as part of dependency resolution and when you later remove the installed software, it would also remove all automatically installed packages.
+100. yum is dumbest and slowest tool I have ever seen. Especially when people try to pitch it against apt-get. And aptitude is ages ahead of any package management RedHat ever implemented.
+100. I've been there too. Forget about simultaneously editing document on more than two computers. (In fact same goes for M$O - though overall compatibility/portability is much better.)
I would love it just to work, but at moment PDF export is only way to have portable (in read-only sense) document.
It is not. M$Office is much more optimized (by all means) product. StarOffice itself was based on previous work - so the code base was already split even before Sun acquisition. And then add development of Sun and OO.o which do not perfectly fit each other.
And Sun's following development effort which threw in Java to the backet didn't help either.
The result is buggy bloated mess. Don't argue with me. I use OOo every day. And I had read the source code.
It's free - but there is nothing more to it. ODF compatibility is still far below any usability level so all the PR talk about ODF magic is just what it is - PR talk. IOW, all OOo has now is its free beer's price: $0.00.
Is it only me who thinks that PHP after version 2 started getting so much weight and bloat that I would believe anything about how insecure it had become.
Rate new features/functions added to PHP at times seems to be exponential. Something that points to poor project management: it looks like incapable of handling the exploded PHP popularity and attention it gathered.
Though my opinion might be outdated: I was programming PHP last time when version 3 was getting its first releases.
NWN2 forums were already flooded once by that problem: you cannot attack friendly targets. Unlike in NWN1. IOW, game is just not designed to look deep into alignments. In NWN2 it seems to be much harder to become evil: there are bunch of stuff for good alignment, but real bad actions are not there.
Though, yes: just like in NWN1 expansion packs, alignment is really adjusted by your actions. (NWN1' alignment was bonker and did really nothing (besides some magic items limited to particular alignments))
But then, realistically you cannot make any real use of alignment, since that would disrupt game completely. One of the problems I have mentioned that in SP NPC are giving you quests to advance the game. Try to imagine what would happen if you as evil character (if you have decided to become one) wouldn't be able to get quest from good guy. On one side it improves reality of game play. But on other side it is hell for game developers since they would have to triple main quest: for good, for neutral and for evil characters. Though I'm already thrilled with all the prospects.
So just to conclude, NWN2 uses alignments - but not to the extent how they are used in table games. Live DMs are still unreachable by computer simulations.