Most job agencies, upon seeing anything what is not.doc would send you a request to resubmit your resume in.doc.
But at least in my experience they are not asking questions about formatting: most of the time.docs are imported into their database and post processed as to extract your info into searchable format.
But generally people are OK with (and prefer - due to possible viruses in.docs) PDF when you apply directly to prospective employer.
I was also during past two job huntings sending both.doc (.rtf actually) and.pdf. At the times, OO.o had poor supports for binary.doc, and renaming produced from sxw.rtf to.doc was doing the trick.
I don't know what features pugdk is missing, but I did have an experience with an Excel file from a client where a macro I needed to run wouldn't work
Insert rant about relying on macros instead of running your business reliably on a database.
Are you idiot? And you work in IT? But am repeating myself.
Learning how to use DB and importing data into it - takes weeks. And in the end you would need at least part time DB admin to take care of the database(s).
Picking up Excel, making primitive VB macro containing desired business logic - takes hours. No extra resources are required.
OO.o is utter failure as scripting goes: language is overly complicated; model object is explicit (you need to init all stuff by your self) and (final *HUGE* nail in the coffin) it is utterly underdocumented.
VB sucks for anything containing logic or longer than 5 lines. But for office applications, as primitive scripting mechanism it is far superior to anything OO.o has to offer.
Add here VB documentation (terrible, but) containing copy-pastable example for literally everything and you have something what even idiot (like me) can use.
Last time I have tried to write trivial macro in OO.o, I had to dig OO.o source code itself. That sucked big time.
If VBA sucks and OO.o is better, why then making script in it takes 5-15 minutes (first record it as macro; then adjust to your needs), while in OO.o after couple of days of digging, you end up doing it manually anyway.
P.S. Thanks to good style support in OO.o, generally I need scripts seldom, yet, when you need one it often turns into disaster with days wasted doing something by hand.
P.P.S. Disclaimer: Linux user for past 10+ years. OO.o user since 1.1.x times.
It's still not as usable as XP, it still doesn't run ALL the software you want, it still doesn't talk to ALL the devices you buy. People opt for XP because it works. Familiarity is a bonus.
Frankly I haven't seen that. And all that rosy even in XP land: there are piles of crap drivers out there easily bringing any OS down.
Though it is hard to disagree with your point: XP (after SP2) is best and most compatible Windows M$ ever produced.
I have still after all this time never actually met a Linux user. Every developer I know has a Windows desktop at home. And I know 100s of developers from all over the world.
If you happen to be in business software, then yes. But again this is the inertia the whole thread is talking about.
If you looks at server application developers or network developers or GUI app developers, they all have tried once or run at home Linux just for sake of (1) fun and (2) novelty. (e.g. my friends working or.Net consultancy use Kubuntu at home just to experiment with Mono and in fact over past year they made their software compatible with Mono.)
Atm I run dual boot system at home and most of the time spend under Linux. Windows I need only to run MPC (Media Player Classic) and even that might be fixed as has VLC now improved considerably support for new video formats (MKVs with embedded fonts and ASS subtitles). But I'm *nix guy for more that 10 years now and Linux is not that strange beast to me.
In fact, what was most surprising to me is how many end users are using Linux. Developers are often bound to platform they develop to (I'm originally Windows developer myself and my past OWL experience still attracts many job offers). But end users who now shifted completely to on-line presence (e.g. Yahoo or Google services) have no dependency on Windows anymore. And it was quite surprise to me (who always dismisses Linux as toy for engineers) that many people actually switch to Ubuntu because it has FireFox and OpenOffice.org. On most of the PCs produced in past 3 years, Ubuntu works flawlessly and many people apparently are not against trying something new for sake of trying something new.
Setting religion aside... Well, nothing is left of the Novell/M$ partnership case once you set religion aside.
Both Novell and M$ are businesses and they do what they need to do to keep their business alive.
I think 99% of/. posters, combined, contributed less than some Novell paid Linux developers what - along with "Boycott Novell" campaign - disqualifies them as involved parties. And since they are not involved parties - they are just bunch of bloggers or using outdated term - whiners.
And also, GPL - v2 and especially v3 - is pretty clear on the issues. If you ever used SUSE, you would know that the OS was all about pragmatism and never about RMS's religion. They were among the first who started including commercial software into distro, thus on one side allowing users to have feature-full system and on another side helping Linux ecosystem grow by reselling software.
Dismissing M$ is plain arrogant - arrogant to users who are stuck in business software locked to M$ platform. Novell helping the users escape the M$ built cage - forcing M$ to open up - is only commendable in my eyes.
P.S. Frankly IMNSHO RH does much damage to FLOSS compared to SUSE/Novell. The amount of code they forked and left for dead is really incomprehensible. From recent events, they even fork now X Window System by starting new project. Instead of participating in community process, RH simply forks. That is more damaging IMHO to FLOSS in general, than commercial deals, since it fragments user and developer communities without actually bringing something new to the table.
Frankly, I do not think that it has something to do with money saving.
I have seen that many times in past when people quoted costs only as justification to switch to better system.
And frankly, systems like Bugzilla, Mantis are magnitude better compared to commercial offerings. I have used number of "state of the art" issue tracking systems - both OSS and commercials - and OSS options are plainly better. Commercial tools win against OSS options only because they have simpler management tools what management (who are in charge of purchase decision making) values higher than overall functionality and usability. But you will not find commercial issue tracking systems where engineers are making decisions.
The supposed advantages of netbooks / mininotebook are excellent portability, battery life and cheapness. But they're not actually that cheap - they're priced too high.
Check out Nokia N800/N810. That looks to me precisely like what netbooks should be.
But actually netbooks were pitched as smallest system capable of running Windows... What frankly to me makes absolutely no sense. I still recall times when Windows 3.x started requiring twice more memory because M$ implemented icon cache. They never being good at handling consumer's precious resources, always shifting costs to hardware.
Many people miss the point here: execs are slaves of shareholders (*). If you want to find who is behind the DRM crap, start looking for who own (and rules) the publishing company.
Once piracy was raised as an issue once, as executives of public companies, they are obliged to evaluate the risk and if necessary introduce measures to alleviate it.
And heck, with numbers like '82%' and '92%' - Good Luck explaining to shareholders and SEC that business is doing fine.
(*) Unless of course status of public company is (1) abused to hide true ownership or (2) used as a fence against potential liabilities. And that pretty much sums up state of publishing industry in general, and game publishing industry in particular. What in the end means that small group of people (if not single man) makes all decisions.
Silly thought just crossed my mind: Wikimedia's Wiki also supports LaTeX rendering.
Did anybody tried to put TeX into Wiki? Integration with lyx (or other editor) might be actually done simply, since wiki accepts plain text as input through web form.
I think the last missing link in OOo's suite of tools is an answer to MS Office's SharePoint server.
I know people who use Wiki specifically for the reason.
It has lots of unexpected features for collaboration e.g. RSS feeds for new pages and category updates.
P.S.
Personally, I don't use Google apps, as a JavaScript implementation of notepad.exe doesn't come close to satisfying my document management needs, and I can't imagine any serious business would disagree.
Well, I can't believe that somebody was fooled by Google's pitch.
Google Doc thingy is fine for simple documents but falls flat for any serious purposes like e.g. specification or protocol.
To the OOo team: Give us an answer to SharePoint! (Please).
For that, I would expect sooner KOffice/Kolab integration, rather than something from OO.o.
On other side, Sun is still backs StarOffice, so they as server company might introduce another Java monstrosity as SharePoint analogue.
Can't speak of iPhone - I can only say to BlackBerry.
In my company we have so called "standby support," when people are getting a BlackBerry from company and have to respond to customer calls.
The amount of abuse BlackBerry can survive is really impressive. Generally, BlackBerrys assigned to standby support pool last for 8-14 months. But the phones rarely have a quiet hour in their lives.
So my biased theory would be that BlackBerry and Treo are failing more because they are used in business more and thus are open to more abuse.
I do my development work in Visual Studio / Netbeans / Jdeveloper.
So, where exactly is the place for vi / emacs today and what benefits do they bring??
You are used to IDEs - go with Emacs. [ Disclaimer: huge VIM fan and Emacs hater. ]
VIM shines when you need to do some work in more or less "batch" mode. Senseless coding comes to mind. Shortcuts are relatively short, but there is a gotcha: VIM has two major modes (insert vs. normal) and shortcuts differ in the modes. (Actually, VIM has officially IIRC 6 modes: visual (selection), insert, replace, normal, ex, command - or something like this). Since VIM was born from console, after some practice you can work in it with closed eyes. (Hint: if you do visual programming - VIM isn't for you).
Emacs integrates all things together wrapping (and often replacing) whole OS by its own "modes." ("Mode" in Emacs is a sort of "plug-in" or "specialized shell"). It has everything, starting from version control mode to tetris clone to web browser. If you would manage to comprehend rather incomprehensible documentation, Emacs could be very good tool for everything. Added bonus: among countless utilities it also includes text editor. Kidding. Emacs is by all means full fledged text processor.
Essential difference between the two: Emacs is very interactive, while VIM is batch oriented. You can do everything without ever leaving Emacs, while VIM easily integrates with shell (and you would need to use it for many things). Configuring Emacs is hurdle and like IDEs is often impossible (mainly due to complexity of configuration). VIM has relatively good documentations and configuring it is rather simple.
Both differ from traditional M$ borne IDEs is that they do not employ button clicking interfaces and can provide most utility to touch typists. Everything (like *everything*) can be done from keyboard.
Since both provide unparalleled configuration flexibility (read: scriptable), it's possible to make out of them highly efficient workhorse for any particular workload. Without button-clicking. In the end that means if you do not wish to make out of them the "workhorse" - customizing work environment isn't your thing - probably you can save learning time by staying with whatever editor you already know and skipping both Emacs and VIM.
P.S. I myself have fond memories of working with Turbo/Borland Pascal/C++ IDEs. My fingers still remember all the shortcuts of decade ago. Yet, recently I had a chance to compare how I did things in past (in Borland IDEs under Windows) and I do things now (in VIM under *nix). After 15 minutes correcting quarter of file in Borland C++ Builder editor, I have rebooted into my Debian and within a minute had rest finished in VIM. Yes, I really like how Borland's IDE looks and works. Yet, in VIM and Linux, I can do my work magnitudes faster.
You should also then:set sw=&ts too (set shiftwidth (sw) to size of tabstop (ts)). That way file would be properly indented with tabs. e.g. you can then set ts=2 or ts=3 or ts=8 and file would change its look accordingly.
In past, when I worked with some micro-softies, I had a shortcut like:map:normal ":set sw=".(12-&ts)." ts=".(12-&ts) (can't be sure about syntax, lazy to check). Pressing F8 was toggling tabsize between 4 and 8 spaces (12 - 8 = 4; 12 - 4 = 8; etc). Was really handy.
i7 memory interconnect would help applications which are not hand-crafted to maximize performance. And I expect that games like Crysis already optimized through the nose to utilize all bandwidths to max.
Or to put it in other words: unoptimized code would gain from i7 more than highly optimized code, since in former case CPU would have more opportunities to optimize memory accesses on its own and better fill up the data bus.
But I also can be wrong and hand crafted code of Crysis/etc is simply cannot take advantage of i7 features.
Real workloads are never should be mixed with benchmarks.
Especially since it is well known that benchmarks often optimized specifically for Intel CPUs.
I expect that games like Crysis and Far Cry would stress heavily RAM, meaning that they hit limit on memory bandwidth (e.g. physics data, textures, etc) before they actually hit top CPU performance (e.g. calculations).
Re:A site geared towards Linux user, to learn Open
on
OpenBSD 4.4 Released
·
· Score: 1
Solaris is also heavily criticized by the Linux crowd for not mindlessly following the GNU "standards"
I would be last to push "GNU standards" as GNU tools are also by now quite old. If you would check Savannah where most GNU tools are maintained, you can easily find for any tool bug report with flamewar where maintainers try to fend off people proposing new features.
Problem is that GNU now is also stuck in some past.
If you would look closer to Linux installations, you would notice that only few pieces of GNU software are left in. And the list of the GNU software is pretty constant, while Linux universe is now more and more densely populated by tools developed elsewhere.
And frankly, as much as I do not like GNU, their text-tools and file-tools are far more superior to anything out there. If it makes users more productive - why not to use it??
As to Solaris... My critic of Solaris would be that they do not preinstall decent shell, normal text editor and pager. I actually addressed to one Solaris engineer the question: when Solaris would ship with sh/vi/more which support keyboard? He laughed and answered that actually Sun is revisiting preinstalled software on Solaris 10, because I'm not alone who find default setup useless and counterproductive.
P.S. GCC and glibc are kind of exceptions and due to heavy Linux bias (and fast development cycle) they are more or less all the time on verge of being forked and made independent from GNU/FSF. FSF tried few times intervene with projects but they got immediate response from developers and maintainer "try that again and we all leave." So they do not try anything. GNU/FSF are politicians and religious leadership - they can't without engineers. GCC and glibc are de facto independent from GNU/FSF now.
P.P.S.
Sun has done a much better job with backwards compatibility than Linux.
I'd be careful with that. "Backward compatible" is not too far from "retarded."
I do not think that following steps of M$DOS and M$Wind0ze (which still carry 25yo(!) errors and mistakes for sake of backward compatibility) would get you anywhere.
Re:A site geared towards Linux user, to learn Open
on
OpenBSD 4.4 Released
·
· Score: 1
As much as I wanted to try BSDs, unfortunately, most of them (OpenBSD included) remain kind of toy for basement kids who do not know anything better.
You wouldn't be attempting a poorly researched troll, would you? I think I just wasted my time:-/
You didn't. It is nice to know that BSD though slowly is also moving somewhere.
I tried *BSD last time about 2.5 years ago (thanks to free VM software it is quite easy) and situation wasn't much improved compared to when I tried FreeBSD about 8 years ago: you still had to do too much hand waving to have a working system. By "working system" of course I mean system which I can use to do my work. Out of box, BSD is sort of M$DOS: bare command prompt, where you need to install and configure everything before you can do something with it.
I do not might wasted gigs of Debian/Sidux/Ubuntu installs: I lost few gigs (which are penny per kilos this days) but I save much more time since I do not need to configure everything. Most of the things come with decent default configs - but some require attention.
*BSD might be true Unix, but this is not something to be particularly proud of, as Unix has history of being worst platform for both users and developers.
If you don't like unix, you don't like linux. (note that that DOESN'T affirm the opposite--if you do like unix, that doesn't mean you have to like linux!)
I do not like BSD. I do not like Unix. But I like Linux.
Linux - Thanks God - is not Unix.
Unix now is more or less corporate tool (e.g. Solaris, HP-UX, AIX) which you might end up using since your business runs on it. But as to plain user, they have nothing to offer out of box. And of course, since business runs on Unix, you would refrain from installing all the end-user stuff on the business critical systems. Best experience with Unix is achieved - as it was many times confirmed in practice - by using Linux as a client.
BSD is good as development community. It is good as technology preview. But that's about it. Using it do some work in business is quite complicated due to very steep learning curve. And even after that you would find that Linux is much much more interoperable that BSD or commercial Unix and much easier to use in practically any environment. "Easier to use" of course means "can do much more work on it without distractions."
Re:A site geared towards Linux user, to learn Open
on
OpenBSD 4.4 Released
·
· Score: 1
I'm not that dumb. It is just starts eating into my production time quickly.
I can make out of fresh install of Debian something useful within half of an hour. And I can easily maintain it that way (weekly "apt-get update && apt-get upgrade" rarely takes longer than one minute).
With BSD it never was the case: one or two days are spent on making out of the system something more useful than M$DOS. Later on, patching is also relatively time consuming.
P.S. And broken ports are also not rarity.
Re:A site geared towards Linux user, to learn Open
on
OpenBSD 4.4 Released
·
· Score: 1
Even Linux man pages (more or less all of them) have by now info how does/doesn't particular tool/call conform to POSIX.
As long as Linux insists on shipping tedious GNU info pages instead of normal man pages, I'm NOT interested.
Show me single Linux distro which removed man.
The "info" weirdo is for GNU tool only. All normal software is pretty happy to live in man pages.
Re:A site geared towards Linux user, to learn Open
on
OpenBSD 4.4 Released
·
· Score: 0, Troll
As long as bash is not default shell, less is not default pager and VIM is not default text editor - I'm NOT interested.
I'm working 99% of time in command line and *BSD archaic tools are not really helping to make the switch from Linux.
And find doesn't search by default in current directory. And grep isn't recursive.
And, for **** sake, where is POSIX conformance statement??? Even Linux man pages (more or less all of them) have by now info how does/doesn't particular tool/call conform to POSIX.
As much as I wanted to try BSDs, unfortunately, most of them (OpenBSD included) remain kind of toy for basement kids who do not know anything better.
Mod me down folks, but here is my advice: if you are serious into *nix - as user or developer - pick Linux.
*BSD might be true Unix, but this is not something to be particularly proud of, as Unix has history of being worst platform for both users and developers.
Slashdot Mirror
Most job agencies, upon seeing anything what is not .doc would send you a request to resubmit your resume in .doc.
But at least in my experience they are not asking questions about formatting: most of the time .docs are imported into their database and post processed as to extract your info into searchable format.
But generally people are OK with (and prefer - due to possible viruses in .docs) PDF when you apply directly to prospective employer.
I was also during past two job huntings sending both .doc (.rtf actually) and .pdf. At the times, OO.o had poor supports for binary .doc, and renaming produced from sxw .rtf to .doc was doing the trick.
I don't know what features pugdk is missing, but I did have an experience with an Excel file from a client where a macro I needed to run wouldn't work
Insert rant about relying on macros instead of running your business reliably on a database.
Are you idiot? And you work in IT? But am repeating myself.
Learning how to use DB and importing data into it - takes weeks. And in the end you would need at least part time DB admin to take care of the database(s).
Picking up Excel, making primitive VB macro containing desired business logic - takes hours. No extra resources are required.
See the difference?
What about some decent scripting/macro language?
OO.o is utter failure as scripting goes: language is overly complicated; model object is explicit (you need to init all stuff by your self) and (final *HUGE* nail in the coffin) it is utterly underdocumented.
VB sucks for anything containing logic or longer than 5 lines. But for office applications, as primitive scripting mechanism it is far superior to anything OO.o has to offer.
Add here VB documentation (terrible, but) containing copy-pastable example for literally everything and you have something what even idiot (like me) can use.
Last time I have tried to write trivial macro in OO.o, I had to dig OO.o source code itself. That sucked big time.
If VBA sucks and OO.o is better, why then making script in it takes 5-15 minutes (first record it as macro; then adjust to your needs), while in OO.o after couple of days of digging, you end up doing it manually anyway.
P.S. Thanks to good style support in OO.o, generally I need scripts seldom, yet, when you need one it often turns into disaster with days wasted doing something by hand.
P.P.S. Disclaimer: Linux user for past 10+ years. OO.o user since 1.1.x times.
It's still not as usable as XP, it still doesn't run ALL the software you want, it still doesn't talk to ALL the devices you buy. People opt for XP because it works. Familiarity is a bonus.
Frankly I haven't seen that. And all that rosy even in XP land: there are piles of crap drivers out there easily bringing any OS down.
Though it is hard to disagree with your point: XP (after SP2) is best and most compatible Windows M$ ever produced.
I have still after all this time never actually met a Linux user. Every developer I know has a Windows desktop at home. And I know 100s of developers from all over the world.
If you happen to be in business software, then yes. But again this is the inertia the whole thread is talking about.
If you looks at server application developers or network developers or GUI app developers, they all have tried once or run at home Linux just for sake of (1) fun and (2) novelty. (e.g. my friends working or .Net consultancy use Kubuntu at home just to experiment with Mono and in fact over past year they made their software compatible with Mono.)
Atm I run dual boot system at home and most of the time spend under Linux. Windows I need only to run MPC (Media Player Classic) and even that might be fixed as has VLC now improved considerably support for new video formats (MKVs with embedded fonts and ASS subtitles). But I'm *nix guy for more that 10 years now and Linux is not that strange beast to me.
In fact, what was most surprising to me is how many end users are using Linux. Developers are often bound to platform they develop to (I'm originally Windows developer myself and my past OWL experience still attracts many job offers). But end users who now shifted completely to on-line presence (e.g. Yahoo or Google services) have no dependency on Windows anymore. And it was quite surprise to me (who always dismisses Linux as toy for engineers) that many people actually switch to Ubuntu because it has FireFox and OpenOffice.org. On most of the PCs produced in past 3 years, Ubuntu works flawlessly and many people apparently are not against trying something new for sake of trying something new.
Setting religion aside... Well, nothing is left of the Novell/M$ partnership case once you set religion aside.
Both Novell and M$ are businesses and they do what they need to do to keep their business alive.
I think 99% of /. posters, combined, contributed less than some Novell paid Linux developers what - along with "Boycott Novell" campaign - disqualifies them as involved parties. And since they are not involved parties - they are just bunch of bloggers or using outdated term - whiners.
And also, GPL - v2 and especially v3 - is pretty clear on the issues. If you ever used SUSE, you would know that the OS was all about pragmatism and never about RMS's religion. They were among the first who started including commercial software into distro, thus on one side allowing users to have feature-full system and on another side helping Linux ecosystem grow by reselling software.
Dismissing M$ is plain arrogant - arrogant to users who are stuck in business software locked to M$ platform. Novell helping the users escape the M$ built cage - forcing M$ to open up - is only commendable in my eyes.
P.S. Frankly IMNSHO RH does much damage to FLOSS compared to SUSE/Novell. The amount of code they forked and left for dead is really incomprehensible. From recent events, they even fork now X Window System by starting new project. Instead of participating in community process, RH simply forks. That is more damaging IMHO to FLOSS in general, than commercial deals, since it fragments user and developer communities without actually bringing something new to the table.
Frankly, I do not think that it has something to do with money saving.
I have seen that many times in past when people quoted costs only as justification to switch to better system.
And frankly, systems like Bugzilla, Mantis are magnitude better compared to commercial offerings. I have used number of "state of the art" issue tracking systems - both OSS and commercials - and OSS options are plainly better. Commercial tools win against OSS options only because they have simpler management tools what management (who are in charge of purchase decision making) values higher than overall functionality and usability. But you will not find commercial issue tracking systems where engineers are making decisions.
The supposed advantages of netbooks / mininotebook are excellent portability, battery life and cheapness. But they're not actually that cheap - they're priced too high.
Check out Nokia N800/N810. That looks to me precisely like what netbooks should be.
But actually netbooks were pitched as smallest system capable of running Windows... What frankly to me makes absolutely no sense. I still recall times when Windows 3.x started requiring twice more memory because M$ implemented icon cache. They never being good at handling consumer's precious resources, always shifting costs to hardware.
Many people miss the point here: execs are slaves of shareholders (*). If you want to find who is behind the DRM crap, start looking for who own (and rules) the publishing company.
Once piracy was raised as an issue once, as executives of public companies, they are obliged to evaluate the risk and if necessary introduce measures to alleviate it.
And heck, with numbers like '82%' and '92%' - Good Luck explaining to shareholders and SEC that business is doing fine.
(*) Unless of course status of public company is (1) abused to hide true ownership or (2) used as a fence against potential liabilities. And that pretty much sums up state of publishing industry in general, and game publishing industry in particular. What in the end means that small group of people (if not single man) makes all decisions.
Silly thought just crossed my mind: Wikimedia's Wiki also supports LaTeX rendering.
Did anybody tried to put TeX into Wiki? Integration with lyx (or other editor) might be actually done simply, since wiki accepts plain text as input through web form.
I think the last missing link in OOo's suite of tools is an answer to MS Office's SharePoint server.
I know people who use Wiki specifically for the reason.
It has lots of unexpected features for collaboration e.g. RSS feeds for new pages and category updates.
P.S.
Personally, I don't use Google apps, as a JavaScript implementation of notepad.exe doesn't come close to satisfying my document management needs, and I can't imagine any serious business would disagree.
Well, I can't believe that somebody was fooled by Google's pitch.
Google Doc thingy is fine for simple documents but falls flat for any serious purposes like e.g. specification or protocol.
To the OOo team: Give us an answer to SharePoint! (Please).
For that, I would expect sooner KOffice/Kolab integration, rather than something from OO.o.
On other side, Sun is still backs StarOffice, so they as server company might introduce another Java monstrosity as SharePoint analogue.
The Windows 7 unveiling garnered largely positive coverage, with many hands-on testers praising it for being faster than Vista.
Had M$ provided me a free notebook I would praise it too!
Here is informed RTFM on the topic.
Can't speak of iPhone - I can only say to BlackBerry.
In my company we have so called "standby support," when people are getting a BlackBerry from company and have to respond to customer calls.
The amount of abuse BlackBerry can survive is really impressive. Generally, BlackBerrys assigned to standby support pool last for 8-14 months. But the phones rarely have a quiet hour in their lives.
So my biased theory would be that BlackBerry and Treo are failing more because they are used in business more and thus are open to more abuse.
I do my development work in Visual Studio / Netbeans / Jdeveloper.
So, where exactly is the place for vi / emacs today and what benefits do they bring??
You are used to IDEs - go with Emacs. [ Disclaimer: huge VIM fan and Emacs hater. ]
VIM shines when you need to do some work in more or less "batch" mode. Senseless coding comes to mind. Shortcuts are relatively short, but there is a gotcha: VIM has two major modes (insert vs. normal) and shortcuts differ in the modes. (Actually, VIM has officially IIRC 6 modes: visual (selection), insert, replace, normal, ex, command - or something like this). Since VIM was born from console, after some practice you can work in it with closed eyes. (Hint: if you do visual programming - VIM isn't for you).
Emacs integrates all things together wrapping (and often replacing) whole OS by its own "modes." ("Mode" in Emacs is a sort of "plug-in" or "specialized shell"). It has everything, starting from version control mode to tetris clone to web browser. If you would manage to comprehend rather incomprehensible documentation, Emacs could be very good tool for everything. Added bonus: among countless utilities it also includes text editor. Kidding. Emacs is by all means full fledged text processor.
Essential difference between the two: Emacs is very interactive, while VIM is batch oriented. You can do everything without ever leaving Emacs, while VIM easily integrates with shell (and you would need to use it for many things). Configuring Emacs is hurdle and like IDEs is often impossible (mainly due to complexity of configuration). VIM has relatively good documentations and configuring it is rather simple.
Both differ from traditional M$ borne IDEs is that they do not employ button clicking interfaces and can provide most utility to touch typists. Everything (like *everything*) can be done from keyboard.
Since both provide unparalleled configuration flexibility (read: scriptable), it's possible to make out of them highly efficient workhorse for any particular workload. Without button-clicking. In the end that means if you do not wish to make out of them the "workhorse" - customizing work environment isn't your thing - probably you can save learning time by staying with whatever editor you already know and skipping both Emacs and VIM.
P.S. I myself have fond memories of working with Turbo/Borland Pascal/C++ IDEs. My fingers still remember all the shortcuts of decade ago. Yet, recently I had a chance to compare how I did things in past (in Borland IDEs under Windows) and I do things now (in VIM under *nix). After 15 minutes correcting quarter of file in Borland C++ Builder editor, I have rebooted into my Debian and within a minute had rest finished in VIM. Yes, I really like how Borland's IDE looks and works. Yet, in VIM and Linux, I can do my work magnitudes faster.
Or right in VIM - :E - would list files in :pwd.
My first and foremost thing in .vimrc is ":imap <Insert> <Nop> ".
I hate when out of casual habit I press Ins in insert mode and VIM sends me to replace mode.
Obviously, habit of pressing 'i' or <Insert> all the time comes from the funny effects typing plain text causes in normal (command) mode.
You should also then :set sw=&ts too (set shiftwidth (sw) to size of tabstop (ts)). That way file would be properly indented with tabs. e.g. you can then set ts=2 or ts=3 or ts=8 and file would change its look accordingly.
In past, when I worked with some micro-softies, I had a shortcut like :map :normal ":set sw=".(12-&ts)." ts=".(12-&ts) (can't be sure about syntax, lazy to check). Pressing F8 was toggling tabsize between 4 and 8 spaces (12 - 8 = 4; 12 - 4 = 8; etc). Was really handy.
The trick helps to correct files with broken encoding.
:set ff=unix will not work with files with broken encoding. :%s/^V^M//g with :set ff={dos,unix} makes sure that all lines would have proper ending.
But it doesn't magically increases RAM bandwidth.
i7 memory interconnect would help applications which are not hand-crafted to maximize performance. And I expect that games like Crysis already optimized through the nose to utilize all bandwidths to max.
Or to put it in other words: unoptimized code would gain from i7 more than highly optimized code, since in former case CPU would have more opportunities to optimize memory accesses on its own and better fill up the data bus.
But I also can be wrong and hand crafted code of Crysis/etc is simply cannot take advantage of i7 features.
Real workloads are never should be mixed with benchmarks.
Especially since it is well known that benchmarks often optimized specifically for Intel CPUs.
I expect that games like Crysis and Far Cry would stress heavily RAM, meaning that they hit limit on memory bandwidth (e.g. physics data, textures, etc) before they actually hit top CPU performance (e.g. calculations).
Solaris is also heavily criticized by the Linux crowd for not mindlessly following the GNU "standards"
I would be last to push "GNU standards" as GNU tools are also by now quite old. If you would check Savannah where most GNU tools are maintained, you can easily find for any tool bug report with flamewar where maintainers try to fend off people proposing new features.
Problem is that GNU now is also stuck in some past.
If you would look closer to Linux installations, you would notice that only few pieces of GNU software are left in. And the list of the GNU software is pretty constant, while Linux universe is now more and more densely populated by tools developed elsewhere.
And frankly, as much as I do not like GNU, their text-tools and file-tools are far more superior to anything out there. If it makes users more productive - why not to use it??
As to Solaris... My critic of Solaris would be that they do not preinstall decent shell, normal text editor and pager. I actually addressed to one Solaris engineer the question: when Solaris would ship with sh/vi/more which support keyboard? He laughed and answered that actually Sun is revisiting preinstalled software on Solaris 10, because I'm not alone who find default setup useless and counterproductive.
P.S. GCC and glibc are kind of exceptions and due to heavy Linux bias (and fast development cycle) they are more or less all the time on verge of being forked and made independent from GNU/FSF. FSF tried few times intervene with projects but they got immediate response from developers and maintainer "try that again and we all leave." So they do not try anything. GNU/FSF are politicians and religious leadership - they can't without engineers. GCC and glibc are de facto independent from GNU/FSF now.
P.P.S.
Sun has done a much better job with backwards compatibility than Linux.
I'd be careful with that. "Backward compatible" is not too far from "retarded."
I do not think that following steps of M$DOS and M$Wind0ze (which still carry 25yo(!) errors and mistakes for sake of backward compatibility) would get you anywhere.
As much as I wanted to try BSDs, unfortunately, most of them (OpenBSD included) remain kind of toy for basement kids who do not know anything better.
You wouldn't be attempting a poorly researched troll, would you? I think I just wasted my time :-/
You didn't. It is nice to know that BSD though slowly is also moving somewhere.
I tried *BSD last time about 2.5 years ago (thanks to free VM software it is quite easy) and situation wasn't much improved compared to when I tried FreeBSD about 8 years ago: you still had to do too much hand waving to have a working system. By "working system" of course I mean system which I can use to do my work. Out of box, BSD is sort of M$DOS: bare command prompt, where you need to install and configure everything before you can do something with it.
I do not might wasted gigs of Debian/Sidux/Ubuntu installs: I lost few gigs (which are penny per kilos this days) but I save much more time since I do not need to configure everything. Most of the things come with decent default configs - but some require attention.
*BSD might be true Unix, but this is not something to be particularly proud of, as Unix has history of being worst platform for both users and developers.
If you don't like unix, you don't like linux. (note that that DOESN'T affirm the opposite--if you do like unix, that doesn't mean you have to like linux!)
I do not like BSD. I do not like Unix. But I like Linux.
Linux - Thanks God - is not Unix.
Unix now is more or less corporate tool (e.g. Solaris, HP-UX, AIX) which you might end up using since your business runs on it. But as to plain user, they have nothing to offer out of box. And of course, since business runs on Unix, you would refrain from installing all the end-user stuff on the business critical systems. Best experience with Unix is achieved - as it was many times confirmed in practice - by using Linux as a client.
BSD is good as development community. It is good as technology preview. But that's about it. Using it do some work in business is quite complicated due to very steep learning curve. And even after that you would find that Linux is much much more interoperable that BSD or commercial Unix and much easier to use in practically any environment. "Easier to use" of course means "can do much more work on it without distractions."
I'm not that dumb. It is just starts eating into my production time quickly.
I can make out of fresh install of Debian something useful within half of an hour. And I can easily maintain it that way (weekly "apt-get update && apt-get upgrade" rarely takes longer than one minute).
With BSD it never was the case: one or two days are spent on making out of the system something more useful than M$DOS. Later on, patching is also relatively time consuming.
P.S. And broken ports are also not rarity.
Even Linux man pages (more or less all of them) have by now info how does/doesn't particular tool/call conform to POSIX.
As long as Linux insists on shipping tedious GNU info pages instead of normal man pages, I'm NOT interested.
Show me single Linux distro which removed man.
The "info" weirdo is for GNU tool only. All normal software is pretty happy to live in man pages.
As long as bash is not default shell, less is not default pager and VIM is not default text editor - I'm NOT interested.
I'm working 99% of time in command line and *BSD archaic tools are not really helping to make the switch from Linux.
And find doesn't search by default in current directory. And grep isn't recursive.
And, for **** sake, where is POSIX conformance statement??? Even Linux man pages (more or less all of them) have by now info how does/doesn't particular tool/call conform to POSIX.
As much as I wanted to try BSDs, unfortunately, most of them (OpenBSD included) remain kind of toy for basement kids who do not know anything better.
Mod me down folks, but here is my advice: if you are serious into *nix - as user or developer - pick Linux.
*BSD might be true Unix, but this is not something to be particularly proud of, as Unix has history of being worst platform for both users and developers.