I didn't said that. I mean Sun purpose is to be profitable (to live) its main goal is not to gain open source community trust (but it is one of other goals). Especially when you look at it from the perspective of their stance on Linux - usually they spread useless FUD and stuff around Linux, but on the other hand they want the community to trust them, they wish not to opensource Java, but they want your trust. Is it not developers input to Solaris the thing they really want? Trust is secondary thing - this is not "for fun", this is for profit...:) I don't mean this is bad.
But I'am not bashing them. I'am just being sceptic and wish to wait some time to look at how it will roll... Also I don't think we (OS community) owe much thanks for Sun, of course OOo is great etc. but they (Sun) also benefit from OOo selling their SO so they own the same amunt of "thank you" as we do. This is busines combined with respect and cooperating, quite sleep stuff.:)
What a lot of Slashdotters might not realize is that Sun has spent literally millions of hours over the last couple of years "unencumbering" Solaris from patented code that was owned by other companies opposed to the open sourcing of their intellectual property. They did this for no reason other than to prove to the open source community that they are serious about open sourcing Solaris, and hopefully to sell some good Sun iron in the process.
No. They didn't explictly done this to prove anything. Without such process (verifing sourcecode copyrights/patents) they could not have opened their code (well they could just with huge risk of being sued). This (building trust) is not Sun main goal - their goal is to go to open source develpement model as it can benefit them. Pure business...
8. If the distribution and/or use of the Program is restricted in certain countries either by patents or by copyrighted interfaces, the original copyright holder who places the Program under this License may add an explicit geographical distribution limitation excluding those countries, so that distribution is permitted only in or among countries not thus excluded. In such case, this License incorporates the limitation as if written in the body of this License.
Java (and JavaScript dialects such as XUL and so on) and PHP... Well PHP is *server* *side* interpreted language so your point is fault. Problem with technologies like ActiveX and similar are due to the fact the programs (sucked from web site) are run onto users machine. With PHP program is run on server and server just passes the output to users machine. So a bug in PHP program won't affect users machine since the bug runs *on* *the* *server*.
Java is of course great example of ActiveX alternative. With Java you can create applets/programs that run fully on user's machine but still Java is known to be relatively secure.
Also with Firefox/Mozilla/Gecko you get a set of totally new technologies based on JavaScript (note that this is something different than Java)... Such technology is called XUL. If you actually have used Firefox and it extensions (addons) - these are writen in XUL language.
XUL has no wide adoption (despite of Firefox internal projects like its addons) but it is an interesting technology. If you want to know more please Google after XUL.
Also another problem with ActiveX is that we can't simply ditch it. Loads of applications use AX already - so we must use it also, at least in some peroid of time and we must to secure it somehow (since we must use it because we are dependant on it). Lack of some feature that is still present is always an lack. Well since it will be no AX apps on Earth we can not state that lack of it support is in did a feature since it is not... But it still does not imply that we can't diss AX - AX sucks but what? Windows also sucks but we can't have Corel on Linux so all this windows-sucking looses a point since Linux is OK but you can't do certain jobs with it. It is OK to keep the users informed what they can and can not do with Firefox.
MS made XP to automagically update itself. I don't know if it requires MSIE (but I can not deinstall it in fact, Add/Remove dialog just removes MSIEs shortcuts but it leaves mshtml.dll intact) but all my Windows machines update themselves using just Active Directory...
This is plain stupid. You just didn't get me right. Look the topic is about *organizations* (like companies) having *loads* of servers running NT - only problem with NT is that they cannot fix it. They could if they have source code - f.e. they could hire a programmer that can do it - probably it will be cheaper than moving all of the boxes to 2003 Server. Or maybe if somebody was able to fix those NT bugs - like a company hiring loads of programmers - they could make fixes and sell them to others. But nobody despite MS can do it since nobody has the code.
So please stop bitching about that you can't fix Linux as it is totally different topic. Don't look at it from perspective of your one and only desktop PC system running Linux which you obviously don't quite understand...
As for documentation - well nobody in open source community likes to write it (I actually like it - but I am not coder). Coders like to code but not to document. eZ Publish was initially a closed product - it was opened somewhere near 4.x version (so prior three versions were closed) - so it looks like (sadly) only most popular projects get good documentation (and docs are a basical thing - a program is not valuable to me if I don't know how it works (I don't have docs)).
As for PHP - well PHP is now quite advanced and really good engine for Web. The problem with PHP is that it is too easy - everybody can code with PHP so it ends up with having loads of ugly code without proper habits and so on. Without system logic/visualisation logic separation, without cacheing, without templates etc. but PHP got all that features you just need to know how to use them...
It is not just CMS - it is also entire framework to make modules that suit you. And it is easy to make such modules. With standard CMS systems you usualy have some prededifned obvious stuff like news, gallery, poll etc. but it is hard (it forces you to code in PHP) to add custom ones.
Also eZ Publish comes with best of breed features like caching, templates, XML, url-rewrites etc.
Check out its site for details, case studies and so on...
http://ez.no/ez_publish/info/web_publishing_func ti onality
Also if your zine/newspaper is published on paper and different media you probably want to simplify the proces and reuse content on both (electronic and paper) media. So maybe you will be interested in integrating these things with XML. FreeSoftwareMagazine has an interesting article/introduction on this topic:
http://www.freesoftwaremagazine.com/free_issues/ is sue_01/focus_format_xml/
3 - yes maybe you may have lot of memory and disk space - this may be true. But consider other things like CPU cycles (please don't tell me that "modern computers" talk with CPU because I simply wont belive it) - and having such setup that every app comes with its own libs (so no shared libs causes):
* More diskspace usage (we have settled on that). * More RAM usage (I can settle on fact disks being cheap, RAM is not cheap IMHO and it is never too much of it for me, I don't waste my RAM on crap). * More disk operation - this is something you don't have a lot huh? Think - how fast is your I/O? You have plenty of that also? * More CPU usage (since it comes from I/O operations).
Now how does it look? I think we are far from the point where hardware is free, hardware costs so we have to spare it using well designed software.
> # You can easily know which files belong to which > software packages
Well that was handled already. Just dont put everything to system like "make install" but add another layer to this proces and use package management system. F.e. RPM (I know there are others and do the same, I'll focus RPM only) - if you want to know which files came with package do: rpm -ql foo, if you have a file in FS and want to know which package it belongs to do rpm -qf/somepath/somefile - so this is not an advantage since we can do this with both setups...
> # You can easily remove the entire package by > using simple rm -r command
Same as with RPM - just issue rpm -e...
I see other advantages of bundling binaries with libraries - like it is easy to install and deploy on various incompatible systems (think Linux versions) with incompatible ABIs... Also you won't get library dependency error since everys such bulk-package comes with its own libs so no chance of such error.
But I prefer to use packages prepared for my system, if there is no package for my system I build it myself. If the thing is small enough (one binary) I just compile it, strip it and put into my ~/bin... If the thing is obscure and I don't want to play with packaging it I just compile it (or not with binary only software), make it static with statifier and put *one* binary into my ~/bin... But I usualy have RPMs for anything I am using since it is quite common format and Fedora is quite popular so chances are big that even if given program is not in distribution it is in one of extras repositories or homepage hosts Fedora (or generic) RPMs anyway...
I know this is just HTML (and some others...) editor but IMHO it would be great to have embeded Gecko view on the code output from Bluefish (maybe as tab) inside... This just begs to embed Gecko into it (like mozilla-bonobo? it lives?). Or maybe make something that Bluefish can control Firefox window. F.e. when you switch focus from Bluefish to Firefox the page automagically redraws...
Heh, my girlfriend is totally untechnical - she manages to surf the web, read email (gmail via browser), chat on IM, listen to music, watch movie but she has problems f.e. with copy&paste - she kinda don't get it.:) She also tends to not read anything but click (or doubleclick - totally random) on anything near cursor if something pops up. Well she is just artistic minded and treats computer as something like phone/tv/tool/hifi whatever - totally ungeekish.:)
I've set her account with GNOME, locked most of stuff (with gconf) so she can't accidentally remove panel or important icon. Also I've set permissions for files so she cannot accidentaly delete something shared...:) I did it becuase I am lazy and don't like to fix everything over and over again...
It looks like this: http://oceanic.wsisiz.edu.pl/~kosmowsk/misc /k8_des k.jpeg
In her work she uses Windows but guess what - she finds Linux more colourful (girls like this) and more convinient than Windoze.:) Also she find it cool that she can login to her grapchical desktop from any machine in my flat (X11 does that). And (the nice part) - she uses terminal since she likes way talk (you know - BSD talkd) works.:)
I can't understand why it does not index OOo/SO documents? Those formats are *open* and well documented. Or FireFox/Mozilla bookmarks/mail/history - it is also open and documented - I bet the community is also willing to help when they (Yahoo developers) have some issues with that. Also probably it is more common than some obscure DOS editors...
In my office we use only OOo (but on Windows) FireFox and Thunderbird - we have crafted some rather nice services including central databases with LDAP export to email clients, custom web apps running exclusively with FireFox (XUL-based), OpenOffice.org is connected to databases also, all server infrastructure is running Linux (Fedora) and lowlevel stuff (DNS, routing, FW etc.) is working on OpenBSD...
So - having desktop search tool that will allow to index that (OOo/Mozilla) will be usefull to us. Todays offering simply suck as they go indexing only some expensive and crappy formats that some expensive and ureliable software produces...
I always wonder why those blokes in service shops (you know - places where you service used computers and stuff) usually never have even heard of Linux (not to mention Knoppix or Live CD). Every time I visit them (hey I live in Poland - we have a lot of old computers here:) - somehow, I like it) and boot Linux (f.e. from CD because my drive bay went off or smth.) they are shocked...
LiveCDs are great for diagnosing - but I would not recommend Knoppix - it is bloated, Knoppix is rather for presentation of Linux to newbies. Systemrescuecd (use Google) is quite good. It contains loads of usefull utilities like AIDA (hardware detection/report) for DOS, dban (for shredding all data on disk - I like it), full blown Linux 2.6 based distro - only console/fb apps - but most of essential stuff. Look if you are going to rescue anything you should have a clue what you are doing.:) When you need Knoppix (KDE and stuff - I am just to lazy to pass parameters on command line or to master the disc sysrescd does that for me) and GUI to rescue anything don't bother... Rescue is always last resort of keeping the data safe.
Try AVAst - it has free "home" version for home/uncomercial use. Free version goes without some advanced stuff (like vorking with MS Active Directory, reporting etc.) suitable mostly in networks of installations not single computer. AVAst interface is a bit ugly (but can be skinned), but at least it does not depend on retarded mshtml.dll controls. And it's detection engine is one of fastest (I cannot remember article with test right know, just belive me:))...
I would use at (sheduling demon aviable on most unices) and streamripper (for MP3) streams. Or for saving stream I would use mplayer (since it can probably handle most streams well) with pcm as output and {insert your favourite audio encoder here} for audio encoding... And probably glue it all together with some little shell script. Maybe run it on server which is always online and has proper storage and processing power... This is actually an easy thing to do, no need to make it complicated running in GUI.:)
That depends in which direction you wish to go. Right know I find it good to have website built by few people. Usualy there should be one for website code/logic, second for database design and third for layout/visualisation/layout-specific-logic - for properly designed webpage (and such have future) it is extremely important to separate those things.
So as graphics designer you probably need to have good knowledge of: 1. Basics: HTML, CSS, graphics programs (loads of, depends on what is your artistic direction), maybe Flash (but I don't like it). 2. More advanced stuff: template engines like smarty or common CMS systems, W3C specifications, accesibity, interfaces design methodology, typography.
As database designer you should be aware of things like: 1. Basics: SQL basics, popular RDBMS like MySQL, SQL Server, PostgreSQL (go with this one), SQLite etc. 2. Advanced: database optimalization, RDBMS specific issues, DB abstraction layers (to make code portable between RDBMS), DB server administration etc.
As code/logic designer: 1. Basics: PHP, Perl, Python - you should at least know few such languages. 2. Advanced: internals of servers (like Apache), advanced programming (probably object oriented), use of popular extensions, systems (like Zope, Pear common CMS systems), templating engines (to work with graphics guy).
Also there is another side of Web which is *content* - it is also wide area how to manage your site, analyze visits, clickpaths, focus on best stuff and forget not visited parts, marketing and so on.
There is really lot to learn to be good webdesigner, you should know at least basics of those topics to be self contained designer. Otherwise I sugesst you focus on things you like the best (graphics/coding/managing etc.) and work in a team of narrow specialists.
* You are certainly sure that you need to plug it directly in your notebook?
I mean if yes than you are focused on bandwith from/to this device, but you've mentioned about firewire and USB - they do not have awfully big bandwith as PCI so this would be a bottleneck...
IMHO for $1000 you surely can get equivalent of PCI device (what is this device?) on either USB, firewire or PCMCIA. If not you can get small case (mATX?) with two PCI slots as external device and export this PCI device via network (you surely can get it below $1000 including entire small PC system and Gb ethernet conectivity)...
Yes but wait. (I am reffering to your 1st paragraph). I understand that 9x was different and now we have XP. OK but this is like something that is easly predictible is it? Like when you design modern multiuser system, where users are separated for system and you know that older legacy programs will have trouble with that (due to different permissions) than you can perdict and manage this also. I don't know Windows much but XP has something like compatibility mode where you run 9x programs in somekind of emulation. This emulation should take care of things like that... I understand that some apps are broken by their authors (there is no MS fault here) but loads of others don't - they are just designed for 9x and using standard 9x system procedures (storing registry, trying to write some files etc.) - so IMHO XP should fix that for these applications giving them some additional layer of automagically intercepting such calls and translating them to proper way... Or maybe MS does not care about it?
Well you could give an option - where to store the configs.:) My friends from highschool are doing IM client from some time. It is actually good and upon first run it asks where to store user profile ("Program Files" or "Application Data" in user profile dir). They made it so maybe because of backward compatibility with Win9x - I don't really know - I don't use Windows (other than administering few XP boxen) since three years... Maybe that is it? If they complained about storing configs in $HOME on *nix - just ignore them, they are stupid:)... But in *nix you could also give a choice where to store configs. Examining some variable maybe? Or smth.?
I can't agree with that... Linux is kernel, what you are reffering to is distribution (like Fedora or Debian) - so in Linux (meaning distro) software is installed completely different than in Windows - in Windows you get installers from all over the Web and push them to system (they usualy work). In Linux you download packages with given software (here Firefox) from less or more official repository of software *customized* *for* *your* *Linux* *flavour (aka distribution). So the proper way of installing Firefox is either - grab binaries from firefox site and unpack them (but keep in mind that in unix system you will need to tweak it bit) or go to your distro package site, grab firefox package and install it. Second way is proper - and if you will find that second way gives you broken instalation you're are allowed to go complaining to packager...:) I assure you that it will be fixed soon. In my Fedora I have no problems with Firefox or whatever user application. I have few glitches with system services (needed tweaking) but this was due to introduction of SELinux and I usualy have very customized configs of my servers so I've had to tweak a little.
I haven't mean you specificaly. Parent post was about Linux and GPLed software going into commercial area (think: alternative software for iPod) - well there is no way to do so (FIXME as I don't know if iPod's EULA is against it - it probably is). You as one person and owner of an iPod can do probably whatever you want with it, you can load up whatever software you wish, you can hit it with a hammer and pin to your wall - that is because such behaviour does not conflict with Apple interest.
Now when some commertial firm starts to offer alternative (be it Linux or GPL, it does not matter here) software for iPod and it conflicts with Apple interest (whatever you will want to do with iPod will conflict with Apple, at least when it is not something like wool coating for your iPod) - then Apple is after you, and that is (FIXME, FIXME) what EULA is all about. Not end user but concurent firms.
IMHO such project has no chances for commercial adoption. First of all iPod is not only hardware - it is also part of a bigger system (think Mac, think iTunes etc.). Apple now is in quite big control over this product, and it simply works for normal user. And also probably iPod EULA (or something similar - I don't really own one) forbids you to use something different than Apple software on iPod (FIXME here).
Linux version is nice thing to have. If somebody really want to go with this overpriced hardware (you can get gizmos more powerful than iPod, like iRiver, but they are probably not mass-market trendy) you can go with iPod and Linux on it. But I don't see larger audience for that. Just few hackers that like to play with hardware, like to be free with it (most of users don't care about it - and I understand it), use rsync to upload your songs, load self compiled/developed/custom/share it... software to _your_ hardware etc. - this is generally what Linux is for:) - to give great computing experience for people who can reach for it.
Slashdot Mirror
I didn't said that. I mean Sun purpose is to be profitable (to live) its main goal is not to gain open source community trust (but it is one of other goals). Especially when you look at it from the perspective of their stance on Linux - usually they spread useless FUD and stuff around Linux, but on the other hand they want the community to trust them, they wish not to opensource Java, but they want your trust. Is it not developers input to Solaris the thing they really want? Trust is secondary thing - this is not "for fun", this is for profit... :) I don't mean this is bad.
:)
But I'am not bashing them. I'am just being sceptic and wish to wait some time to look at how it will roll... Also I don't think we (OS community) owe much thanks for Sun, of course OOo is great etc. but they (Sun) also benefit from OOo selling their SO so they own the same amunt of "thank you" as we do. This is busines combined with respect and cooperating, quite sleep stuff.
What a lot of Slashdotters might not realize is that Sun has spent literally millions of hours over the last couple of years "unencumbering" Solaris from patented code that was owned by other companies opposed to the open sourcing of their intellectual property. They did this for no reason other than to prove to the open source community that they are serious about open sourcing Solaris, and hopefully to sell some good Sun iron in the process.
No. They didn't explictly done this to prove anything. Without such process (verifing sourcecode copyrights/patents) they could not have opened their code (well they could just with huge risk of being sued). This (building trust) is not Sun main goal - their goal is to go to open source develpement model as it can benefit them. Pure business...
8. If the distribution and/or use of the Program is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Program under this License
may add an explicit geographical distribution limitation excluding
those countries, so that distribution is permitted only in or among
countries not thus excluded. In such case, this License incorporates
the limitation as if written in the body of this License.
Java (and JavaScript dialects such as XUL and so on) and PHP... Well PHP is *server* *side* interpreted language so your point is fault. Problem with technologies like ActiveX and similar are due to the fact the programs (sucked from web site) are run onto users machine. With PHP program is run on server and server just passes the output to users machine. So a bug in PHP program won't affect users machine since the bug runs *on* *the* *server*.
Java is of course great example of ActiveX alternative. With Java you can create applets/programs that run fully on user's machine but still Java is known to be relatively secure.
Also with Firefox/Mozilla/Gecko you get a set of totally new technologies based on JavaScript (note that this is something different than Java)... Such technology is called XUL. If you actually have used Firefox and it extensions (addons) - these are writen in XUL language.
XUL has no wide adoption (despite of Firefox internal projects like its addons) but it is an interesting technology. If you want to know more please Google after XUL.
Also another problem with ActiveX is that we can't simply ditch it. Loads of applications use AX already - so we must use it also, at least in some peroid of time and we must to secure it somehow (since we must use it because we are dependant on it). Lack of some feature that is still present is always an lack. Well since it will be no AX apps on Earth we can not state that lack of it support is in did a feature since it is not... But it still does not imply that we can't diss AX - AX sucks but what? Windows also sucks but we can't have Corel on Linux so all this windows-sucking looses a point since Linux is OK but you can't do certain jobs with it. It is OK to keep the users informed what they can and can not do with Firefox.
MS made XP to automagically update itself. I don't know if it requires MSIE (but I can not deinstall it in fact, Add/Remove dialog just removes MSIEs shortcuts but it leaves mshtml.dll intact) but all my Windows machines update themselves using just Active Directory...
This is plain stupid. You just didn't get me right. Look the topic is about *organizations* (like companies) having *loads* of servers running NT - only problem with NT is that they cannot fix it. They could if they have source code - f.e. they could hire a programmer that can do it - probably it will be cheaper than moving all of the boxes to 2003 Server. Or maybe if somebody was able to fix those NT bugs - like a company hiring loads of programmers - they could make fixes and sell them to others. But nobody despite MS can do it since nobody has the code.
So please stop bitching about that you can't fix Linux as it is totally different topic. Don't look at it from perspective of your one and only desktop PC system running Linux which you obviously don't quite understand...
As for documentation - well nobody in open source community likes to write it (I actually like it - but I am not coder). Coders like to code but not to document. eZ Publish was initially a closed product - it was opened somewhere near 4.x version (so prior three versions were closed) - so it looks like (sadly) only most popular projects get good documentation (and docs are a basical thing - a program is not valuable to me if I don't know how it works (I don't have docs)).
As for PHP - well PHP is now quite advanced and really good engine for Web. The problem with PHP is that it is too easy - everybody can code with PHP so it ends up with having loads of ugly code without proper habits and so on. Without system logic/visualisation logic separation, without cacheing, without templates etc. but PHP got all that features you just need to know how to use them...
http://ez.no/ez_publish
c ti onality
/ is sue_01/focus_format_xml/
It is not just CMS - it is also entire framework to make modules that suit you. And it is easy to make such modules. With standard CMS systems you usualy have some prededifned obvious stuff like news, gallery, poll etc. but it is hard (it forces you to code in PHP) to add custom ones.
Also eZ Publish comes with best of breed features like caching, templates, XML, url-rewrites etc.
Check out its site for details, case studies and so on...
http://ez.no/ez_publish/info/web_publishing_fun
Also if your zine/newspaper is published on paper and different media you probably want to simplify the proces and reuse content on both (electronic and paper) media. So maybe you will be interested in integrating these things with XML. FreeSoftwareMagazine has an interesting article/introduction on this topic:
http://www.freesoftwaremagazine.com/free_issues
3 - yes maybe you may have lot of memory and disk space - this may be true. But consider other things like CPU cycles (please don't tell me that "modern computers" talk with CPU because I simply wont belive it) - and having such setup that every app comes with its own libs (so no shared libs causes):
* More diskspace usage (we have settled on that).
* More RAM usage (I can settle on fact disks being cheap, RAM is not cheap IMHO and it is never too much of it for me, I don't waste my RAM on crap).
* More disk operation - this is something you don't have a lot huh? Think - how fast is your I/O? You have plenty of that also?
* More CPU usage (since it comes from I/O operations).
Now how does it look? I think we are far from the point where hardware is free, hardware costs so we have to spare it using well designed software.
> # You can easily know which files belong to which
/somepath/somefile - so this is not an advantage since we can do this with both setups...
> software packages
Well that was handled already. Just dont put everything to system like "make install" but add another layer to this proces and use package management system. F.e. RPM (I know there are others and do the same, I'll focus RPM only) - if you want to know which files came with package do: rpm -ql foo, if you have a file in FS and want to know which package it belongs to do rpm -qf
> # You can easily remove the entire package by
> using simple rm -r command
Same as with RPM - just issue rpm -e...
I see other advantages of bundling binaries with libraries - like it is easy to install and deploy on various incompatible systems (think Linux versions) with incompatible ABIs... Also you won't get library dependency error since everys such bulk-package comes with its own libs so no chance of such error.
But I prefer to use packages prepared for my system, if there is no package for my system I build it myself. If the thing is small enough (one binary) I just compile it, strip it and put into my ~/bin... If the thing is obscure and I don't want to play with packaging it I just compile it (or not with binary only software), make it static with statifier and put *one* binary into my ~/bin... But I usualy have RPMs for anything I am using since it is quite common format and Fedora is quite popular so chances are big that even if given program is not in distribution it is in one of extras repositories or homepage hosts Fedora (or generic) RPMs anyway...
I know this is just HTML (and some others...) editor but IMHO it would be great to have embeded Gecko view on the code output from Bluefish (maybe as tab) inside... This just begs to embed Gecko into it (like mozilla-bonobo? it lives?). Or maybe make something that Bluefish can control Firefox window. F.e. when you switch focus from Bluefish to Firefox the page automagically redraws...
Heh, my girlfriend is totally untechnical - she manages to surf the web, read email (gmail via browser), chat on IM, listen to music, watch movie but she has problems f.e. with copy&paste - she kinda don't get it. :) She also tends to not read anything but click (or doubleclick - totally random) on anything near cursor if something pops up. Well she is just artistic minded and treats computer as something like phone/tv/tool/hifi whatever - totally ungeekish. :)
:) I did it becuase I am lazy and don't like to fix everything over and over again...
c /k8_des k.jpeg
:) Also she find it cool that she can login to her grapchical desktop from any machine in my flat (X11 does that). And (the nice part) - she uses terminal since she likes way talk (you know - BSD talkd) works. :)
I've set her account with GNOME, locked most of stuff (with gconf) so she can't accidentally remove panel or important icon. Also I've set permissions for files so she cannot accidentaly delete something shared...
It looks like this:
http://oceanic.wsisiz.edu.pl/~kosmowsk/mis
In her work she uses Windows but guess what - she finds Linux more colourful (girls like this) and more convinient than Windoze.
When you will be able to fix NT sourcecode yourself and compile your own hotfixes, let us know.
I can't understand why it does not index OOo/SO documents? Those formats are *open* and well documented. Or FireFox/Mozilla bookmarks/mail/history - it is also open and documented - I bet the community is also willing to help when they (Yahoo developers) have some issues with that. Also probably it is more common than some obscure DOS editors...
In my office we use only OOo (but on Windows) FireFox and Thunderbird - we have crafted some rather nice services including central databases with LDAP export to email clients, custom web apps running exclusively with FireFox (XUL-based), OpenOffice.org is connected to databases also, all server infrastructure is running Linux (Fedora) and lowlevel stuff (DNS, routing, FW etc.) is working on OpenBSD...
So - having desktop search tool that will allow to index that (OOo/Mozilla) will be usefull to us. Todays offering simply suck as they go indexing only some expensive and crappy formats that some expensive and ureliable software produces...
I always wonder why those blokes in service shops (you know - places where you service used computers and stuff) usually never have even heard of Linux (not to mention Knoppix or Live CD). Every time I visit them (hey I live in Poland - we have a lot of old computers here :) - somehow, I like it) and boot Linux (f.e. from CD because my drive bay went off or smth.) they are shocked...
:) When you need Knoppix (KDE and stuff - I am just to lazy to pass parameters on command line or to master the disc sysrescd does that for me) and GUI to rescue anything don't bother... Rescue is always last resort of keeping the data safe.
LiveCDs are great for diagnosing - but I would not recommend Knoppix - it is bloated, Knoppix is rather for presentation of Linux to newbies. Systemrescuecd (use Google) is quite good. It contains loads of usefull utilities like AIDA (hardware detection/report) for DOS, dban (for shredding all data on disk - I like it), full blown Linux 2.6 based distro - only console/fb apps - but most of essential stuff. Look if you are going to rescue anything you should have a clue what you are doing.
Try AVAst - it has free "home" version for home/uncomercial use. Free version goes without some advanced stuff (like vorking with MS Active Directory, reporting etc.) suitable mostly in networks of installations not single computer. AVAst interface is a bit ugly (but can be skinned), but at least it does not depend on retarded mshtml.dll controls. And it's detection engine is one of fastest (I cannot remember article with test right know, just belive me :))...
http://www.avast.com/eng/down_home.html
I would use at (sheduling demon aviable on most unices) and streamripper (for MP3) streams. Or for saving stream I would use mplayer (since it can probably handle most streams well) with pcm as output and {insert your favourite audio encoder here} for audio encoding... And probably glue it all together with some little shell script. Maybe run it on server which is always online and has proper storage and processing power... This is actually an easy thing to do, no need to make it complicated running in GUI. :)
That depends in which direction you wish to go. Right know I find it good to have website built by few people. Usualy there should be one for website code/logic, second for database design and third for layout/visualisation/layout-specific-logic - for properly designed webpage (and such have future) it is extremely important to separate those things.
So as graphics designer you probably need to have good knowledge of:
1. Basics: HTML, CSS, graphics programs (loads of, depends on what is your artistic direction), maybe Flash (but I don't like it).
2. More advanced stuff: template engines like smarty or common CMS systems, W3C specifications, accesibity, interfaces design methodology, typography.
As database designer you should be aware of things like:
1. Basics: SQL basics, popular RDBMS like MySQL, SQL Server, PostgreSQL (go with this one), SQLite etc.
2. Advanced: database optimalization, RDBMS specific issues, DB abstraction layers (to make code portable between RDBMS), DB server administration etc.
As code/logic designer:
1. Basics: PHP, Perl, Python - you should at least know few such languages.
2. Advanced: internals of servers (like Apache), advanced programming (probably object oriented), use of popular extensions, systems (like Zope, Pear common CMS systems), templating engines (to work with graphics guy).
Also there is another side of Web which is *content* - it is also wide area how to manage your site, analyze visits, clickpaths, focus on best stuff and forget not visited parts, marketing and so on.
There is really lot to learn to be good webdesigner, you should know at least basics of those topics to be self contained designer. Otherwise I sugesst you focus on things you like the best (graphics/coding/managing etc.) and work in a team of narrow specialists.
* It has to be portable?
* You are certainly sure that you need to plug it directly in your notebook?
I mean if yes than you are focused on bandwith from/to this device, but you've mentioned about firewire and USB - they do not have awfully big bandwith as PCI so this would be a bottleneck...
IMHO for $1000 you surely can get equivalent of PCI device (what is this device?) on either USB, firewire or PCMCIA. If not you can get small case (mATX?) with two PCI slots as external device and export this PCI device via network (you surely can get it below $1000 including entire small PC system and Gb ethernet conectivity)...
I think you can do anything you whish :) you have full source code access so you probably can compile a one static lib with everything in it...
Yes but wait. (I am reffering to your 1st paragraph). I understand that 9x was different and now we have XP. OK but this is like something that is easly predictible is it? Like when you design modern multiuser system, where users are separated for system and you know that older legacy programs will have trouble with that (due to different permissions) than you can perdict and manage this also. I don't know Windows much but XP has something like compatibility mode where you run 9x programs in somekind of emulation. This emulation should take care of things like that... I understand that some apps are broken by their authors (there is no MS fault here) but loads of others don't - they are just designed for 9x and using standard 9x system procedures (storing registry, trying to write some files etc.) - so IMHO XP should fix that for these applications giving them some additional layer of automagically intercepting such calls and translating them to proper way... Or maybe MS does not care about it?
Well you could give an option - where to store the configs. :) My friends from highschool are doing IM client from some time. It is actually good and upon first run it asks where to store user profile ("Program Files" or "Application Data" in user profile dir). They made it so maybe because of backward compatibility with Win9x - I don't really know - I don't use Windows (other than administering few XP boxen) since three years... Maybe that is it? If they complained about storing configs in $HOME on *nix - just ignore them, they are stupid :)... But in *nix you could also give a choice where to store configs. Examining some variable maybe? Or smth.?
I can't agree with that... Linux is kernel, what you are reffering to is distribution (like Fedora or Debian) - so in Linux (meaning distro) software is installed completely different than in Windows - in Windows you get installers from all over the Web and push them to system (they usualy work). In Linux you download packages with given software (here Firefox) from less or more official repository of software *customized* *for* *your* *Linux* *flavour (aka distribution). So the proper way of installing Firefox is either - grab binaries from firefox site and unpack them (but keep in mind that in unix system you will need to tweak it bit) or go to your distro package site, grab firefox package and install it. Second way is proper - and if you will find that second way gives you broken instalation you're are allowed to go complaining to packager... :) I assure you that it will be fixed soon. In my Fedora I have no problems with Firefox or whatever user application. I have few glitches with system services (needed tweaking) but this was due to introduction of SELinux and I usualy have very customized configs of my servers so I've had to tweak a little.
I haven't mean you specificaly. Parent post was about Linux and GPLed software going into commercial area (think: alternative software for iPod) - well there is no way to do so (FIXME as I don't know if iPod's EULA is against it - it probably is). You as one person and owner of an iPod can do probably whatever you want with it, you can load up whatever software you wish, you can hit it with a hammer and pin to your wall - that is because such behaviour does not conflict with Apple interest.
Now when some commertial firm starts to offer alternative (be it Linux or GPL, it does not matter here) software for iPod and it conflicts with Apple interest (whatever you will want to do with iPod will conflict with Apple, at least when it is not something like wool coating for your iPod) - then Apple is after you, and that is (FIXME, FIXME) what EULA is all about. Not end user but concurent firms.
IMHO such project has no chances for commercial adoption. First of all iPod is not only hardware - it is also part of a bigger system (think Mac, think iTunes etc.). Apple now is in quite big control over this product, and it simply works for normal user. And also probably iPod EULA (or something similar - I don't really own one) forbids you to use something different than Apple software on iPod (FIXME here).
:) - to give great computing experience for people who can reach for it.
Linux version is nice thing to have. If somebody really want to go with this overpriced hardware (you can get gizmos more powerful than iPod, like iRiver, but they are probably not mass-market trendy) you can go with iPod and Linux on it. But I don't see larger audience for that. Just few hackers that like to play with hardware, like to be free with it (most of users don't care about it - and I understand it), use rsync to upload your songs, load self compiled/developed/custom/share it... software to _your_ hardware etc. - this is generally what Linux is for
Just my 2p.