i use servlets. i dont have a single line of HTML in my servlets. i generate lots of dynamic content. you just need to rethink your paradigm. JSP is not the answer.
true to the former....i like the name tho.:). it was experimental and mainly showed how the algorithms for reading/writing compressed/uncompress/encrypted DOC files go. it was never meant to be anything more than a programmers toy. i couldnt be bothered to improve the UI since i mainly use it for beaming DOC files back and forth from PCs using the serial port...it does DOCtext conversion on the fly and its DOC handling routines are strictly standards conformant. cspot, aportis, tealdoc are actually geared towards end users. ZDOC never was (and still isnt). its the only one that shows the exact headers in a DOC file tho. and has a good DOC handler function set which can be copied for other GPL projects. plus it can handle PDB files from peanut press. the algorithm implementation & functionality is always (and always has been) more important to me than end user UI for all my free software projects.
try dell. theyre very good at mix and match. i purchased machine from them with scsi drives without a scsi card. their tech called up right away and informed me, i told him to ship it anyway (i had cards) and they did. worked fine, first time. im pretty happy with dell for the university i work at. just recived 60 boxes from them with no problems. they even bundled nt/office with em even though we didnt order it (they were going to be turned into linux boxes).
depends on what you want to do with the cluster. redhats piranha tool isnt bad - im using it and it works fine on no name asus p2bds boxes with dual 650mhz piiis. the adaptec ultrawide controller on board the asus works like a dream.
people skills are VERY important - you cant run a business with dissatisfied customers. its ok that VA charges a lot more than clone makers. its ok if VA bills you $100/hr for support. its NOT ok if VA hardware doesnt work. and its certainly NOT ok if their own techs/support guys cant m,ake the hardware work.
if youre not managing your own boxes whats the big deal about someone else choosing a distro that they themselves will be managing and are familiar with ? it costs money to train admins - theyve been trained on redhat, they might screw up debian. linux is linux. who cares ? if you dont like it put youre own debian box up at their place and run it yourself remotely. you have an admin....right ?
http://linuxvirtualserver.org and also compile CORBA into your kernels. you will be wasting machines tho..that stuff is built for fault tolerance. you might want to have a look at MOSIX if all you want to do is have "one big cluster". MOSIX can be d/led from freshmeat (may screw up your kernel - beware).
and if he's looking for a palm based GPLed DOC encoder/decoder try mine :
http://zurk.sourceforge.net...its called ZDOC. should be d/lable from any of the palm sites too...i have around 4000 users.
yeah, a ftp archive with google search would be REALLY KEWL. please put one up! i've been whining to freshmeat for a loong time about it.
these systems are very important...i wonder if they have thought of the p[rotection against spammers angle ? especially now that the first spammer (flatplanet.net) has surfaced on gnutella networks.
yeah. how bout increasing the frequency of releases so i dont have to wait YEARS to see a new stable debain ? all the hardware on the planet is useless if i cant get a modern stable distro. cmon guys - speed it up !
a big organisation usually will have thousands of users using stuff like M$ office. i hate to tell you this - but converting office staff, retraining , buying copies of (insert favourite word proc here) for linux (most likely corels office suite), buying a distro with support, putting convertors in place to handle attachments and emails from outside, finding an alternative to the point and drool outlook is going to be one hell of a big job. Youre not going to convince them to change - its real unlikely. you might be able to get em off to a macintosh platform with M$ office..but lets face it - UNIX (and linux) isnt there yet in the clueless user department. heck, even windoze is hard to support in that sort of an environment.
Just a few points for you to think about. im working in an all M$ environment, altho my machine runs linux and a few others run BSD, solaris, aixm, hp/ux and macos..the majority of users still run nt4sp3 or win98. even win2k is in short supply . its not easy converting offices.
heh. and i used to code in 8085 assembly where there was no such thing as a printf.:)..or a OS or monitor for that matter. looking up hex codes and typing em into a 16 digit keypad connected to an LCD..fun fun fun.
FORTRAN sucked syntactically, C doesnt have this problem, which is why C++/Java/Modula-2 try and copy it.
VB doesnt qualify as a language.:)
ok..you have a point that the language isnt about libraries, but now that C is around and programmers have used > 1 language, the language has shifted to becoming irrelevant. libraries are all important as is portability, security and performance. drag and drop is *not* the way to go, but neither is the "i'll write volumes of code to manipulate XML/HTML/media files".
The paradigm seems to be shifting to a robust language base with extensive support libraries type thing (GUI/text mode development is the programmers choice - i still code in java with pico) with emphasis on portability, performance and security. which is a *good* thing i might add.
of course theres always M$ doing its usual..but since theyre pushing C# instead of VB, they seem to be learning too.:)
languages ARE collections of libraries, like it or not. and its going to get much better once we have decent libraries. the core language is nothing more than a wrapper to access the library. are you telling me C is useful without stdlib/stdio ? Syntax is now becoming increasingly irrelevant since most languages use a C or C like syntax (java included).
exactly. sun systems dont have a BIOS. in fact, nothing has a BIOS except PCs. the only reason for the BIOS was DOS - which cant do anything without a BIOS since its just a glorified file loader. most modern systems all have boot PROMS..most of em with their own language (forth in the case of sun) which you can actually code in to change the boot process. BIOSes suck - the faster we get rid of em the better. linux/bsd/most nixes/NT dont use em anyway (SGIs NT workstations all use boot PROMs).
the languages now are really CRUDE. programmers have to build infrastructure whenever we code. thats prolly the most annoying part (and sometimes the most fun) of coding. what i'd like to see is a language which would provide basic functions off the shelf. if i want an HTML editor stuck between two translucent animated buttons which pull up a hex editor and a MP3 player..i shouldnt have to *code* all that. three function calls and a few lines of code should do that for me in (insert language of choice). i should have to build stuff to parse files...if i want a XML tag called weather in a file somewhere on the disk, it should be able to retrieve it for me in one functional call. Java & the GTK stuff has been trying to do that but it doesnt go far enough. I want a seamless environment to manipulate all the functions available, plus i should be able to cut and paste bits from a library of examples available. And i should have a choice of writing bits in assembly and controlling the machines registers at the same time. tall order huh ?
no. there will be no license.
most coders i know would do this (myself included) :
[1] go thru anonymiser and create a hotmail account.
[2] go thru anonymiser and create a geocities account using email from step 1.
[3] dump illegal/dangerous software on geocities and put link to software on freshmeat.
[4] cops shut down geocities/hotmail account.
[5] goto step 1.
lather, rinse, repeat.
heck, with sufficient tools like wget and html redirection this could be automated.
i do this anyway if im reverse engineering stuff.
and :
6. install bash.
7. go to the hp/ux freeware page and d/l all the stuff.
8. HP maintains a JDK if youre into java. look on hp.com...
9. ssh ssh ssh. install ASAP and comment out the crap in/etc/inetd.conf
i've got a A9000 HP/UX 11.0 box without a video card..text terminals suck. try and get a video card from hp if you dont have one. also note that ssh cant authenticate via passwords if you set the trusted base option in HP/UX. use RSA authentication. the trusted stuff also breaks things...so make sure you have at least a text terminal when you install it.
ok, youre a troll but let me address your points.:-
under the traditional process marketing muscle is leveraged, publishers have contacts with bookstores to promote the books etc...even if you are an unknown author. advertising budgets are also available to promote. theres a lot of stuff going around behind the scenes - ive been there.
apache, perl etc are all failures. yes, they are great. yes, people use them. ask yourself why the apache software foundation is non profit. why there is no company which sells perl. why GCC etc cant be sold in the quantities M$ visual C++ is..even though its far superior.. why emacs is not sold like M$ word is. why linus torvalds who wrote linux never made any money of it. linus doesnt even work on a pure linux job. Then ask yourself if they are true commercial sucesses. yes, redhat makes money - but the main money is made with support. not by *selling* the product. you cant sell something people download off the web. did you know concorde has been a failure even if its seen as the fastest aircraft there is ? it doesnt turn a profit - the R&D costs were too high. In the real world of companies which HAVE to survive and generate revenues, working products dont count - the bottom line ALWAYS is more important. have there been some success with GPLed/free products - sure. is there more profit with closed source - YUP. thats why M$ has profits in the BILLIONS. wake up kid - this is reality.
Yup. my email address definitely makes me a looser. real mature.
change Linux to Any OS in your above statement and it will hold true. see solaris - dog slow on 4 CPUs. the way you write the kernel becomes inefficient with a barrier at around 4 CPUs if you want low end performance. high end performance is completely opposite to it. its a tradeoff - no OS can go around it. we might have a high end kernel or a low end kernel VM option in linux soon. recompiling it plugs in the VM you want.
ok. in short this is how it works.
1] it sets up a server socket on port 8080 with a handler for that. any java applet can do this.
2] it bypasses the java.io.File stuff which is sandboxed by using the netscape file://c:/ command (which is friggin brilliant if i may say so...) to browse files. since its running in a web browser anyway, it can send commands to the local browser.
Fix :
Simple. expand the security sandbox to applets cant use file:// to exploit their local browser.
it will only work for well known authors. supposing i wanted to write a book, wrote several chapters..gave it to a publisher under this protocol. would he pay me ? unlikely. would anyone bother to look at it ? nope. only well known/established authors could hope to make this fly.
mark me down if you will - but the same is true of code released under the GPL. without an established base of fans/supporters - you go nowhere. and thats the main reason open source doesnt fly in the real world.
Your assumption is VERY wrong. If there were no copyrights and all code was open, whats to stop a programmer working for a company to throw the code out in the open ? After all, the company cant sue him for copyright or ip infringement. and how many programmers, given the choice, would lock their code up ? not many. at least not the intelligent ones. given the trend, how many people would buy closed source software ?
the difference between software and music is that there is no need to duplicate the PROCESS OF WRITING MUSIC. you simply "execute" it to enjoy the full benefits. with software this is not the case. the PROCESS is more important than the final EXECUTION. The GPLNET guy is a troll - i wouldnt give two shits about him. whats he going to do - release linux kernel binaries via napster ? sure - go ahead and everyone will laugh.
As a programmer, i'd love to see copyright/ip disappear. programmers enjoy programming new stuff - they dont enjoy duplication of effort. all code *should* be public domain. people will pay for *systems integration* - which is a lot more important. its not the code - its whether the system i want to build will work. and thats whats really important.
To use an analogy, the blueprints of a bus may be available freely, but people still buy busses from GM or wherever. and its the revenue earned from the poeple actually *using* the bus which is important.
I should point out the kids in the US of A are already FAR more privilidged than kids who grow up out of the US. Should we really care whether some kid who wants to view a porn site at a library is not allowed to ? Or should we be more bothered about the fact that there are NO COMPUTERS in libraries not located in the US ? Heck 99% of the planet doesnt have a computer in their homes let alone the local underfunded library.
And on a related note - the abortion information is far more likely to be found in large quantities in any of the books in the library than on the net. libraries are for reading - lets keep it that way. web surfing should be kept strictly in school computer labs. There nothing more annoying than web surfers clogging up the library.
Slashdot Mirror
i use servlets. i dont have a single line of HTML in my servlets. i generate lots of dynamic content. you just need to rethink your paradigm. JSP is not the answer.
true to the former....i like the name tho. :). it was experimental and mainly showed how the algorithms for reading/writing compressed/uncompress/encrypted DOC files go. it was never meant to be anything more than a programmers toy. i couldnt be bothered to improve the UI since i mainly use it for beaming DOC files back and forth from PCs using the serial port...it does DOCtext conversion on the fly and its DOC handling routines are strictly standards conformant. cspot, aportis, tealdoc are actually geared towards end users. ZDOC never was (and still isnt). its the only one that shows the exact headers in a DOC file tho. and has a good DOC handler function set which can be copied for other GPL projects. plus it can handle PDB files from peanut press. the algorithm implementation & functionality is always (and always has been) more important to me than end user UI for all my free software projects.
try dell. theyre very good at mix and match. i purchased machine from them with scsi drives without a scsi card. their tech called up right away and informed me, i told him to ship it anyway (i had cards) and they did. worked fine, first time. im pretty happy with dell for the university i work at. just recived 60 boxes from them with no problems. they even bundled nt/office with em even though we didnt order it (they were going to be turned into linux boxes).
depends on what you want to do with the cluster. redhats piranha tool isnt bad - im using it and it works fine on no name asus p2bds boxes with dual 650mhz piiis. the adaptec ultrawide controller on board the asus works like a dream.
people skills are VERY important - you cant run a business with dissatisfied customers. its ok that VA charges a lot more than clone makers. its ok if VA bills you $100/hr for support. its NOT ok if VA hardware doesnt work. and its certainly NOT ok if their own techs/support guys cant m,ake the hardware work.
if youre not managing your own boxes whats the big deal about someone else choosing a distro that they themselves will be managing and are familiar with ? it costs money to train admins - theyve been trained on redhat, they might screw up debian. linux is linux. who cares ? if you dont like it put youre own debian box up at their place and run it yourself remotely. you have an admin....right ?
err..i meant CODA not CORBA. shoot me for being stupid. :)
http://linuxvirtualserver.org and also compile CORBA into your kernels. you will be wasting machines tho..that stuff is built for fault tolerance. you might want to have a look at MOSIX if all you want to do is have "one big cluster". MOSIX can be d/led from freshmeat (may screw up your kernel - beware).
and if he's looking for a palm based GPLed DOC encoder/decoder try mine :
http://zurk.sourceforge.net...its called ZDOC. should be d/lable from any of the palm sites too...i have around 4000 users.
yeah, a ftp archive with google search would be REALLY KEWL. please put one up! i've been whining to freshmeat for a loong time about it.
these systems are very important...i wonder if they have thought of the p[rotection against spammers angle ? especially now that the first spammer (flatplanet.net) has surfaced on gnutella networks.
yeah. how bout increasing the frequency of releases so i dont have to wait YEARS to see a new stable debain ? all the hardware on the planet is useless if i cant get a modern stable distro. cmon guys - speed it up !
a big organisation usually will have thousands of users using stuff like M$ office. i hate to tell you this - but converting office staff, retraining , buying copies of (insert favourite word proc here) for linux (most likely corels office suite), buying a distro with support, putting convertors in place to handle attachments and emails from outside, finding an alternative to the point and drool outlook is going to be one hell of a big job. Youre not going to convince them to change - its real unlikely. you might be able to get em off to a macintosh platform with M$ office..but lets face it - UNIX (and linux) isnt there yet in the clueless user department. heck, even windoze is hard to support in that sort of an environment.
Just a few points for you to think about. im working in an all M$ environment, altho my machine runs linux and a few others run BSD, solaris, aixm, hp/ux and macos..the majority of users still run nt4sp3 or win98. even win2k is in short supply . its not easy converting offices.
heh. and i used to code in 8085 assembly where there was no such thing as a printf. :) ..or a OS or monitor for that matter. looking up hex codes and typing em into a 16 digit keypad connected to an LCD..fun fun fun.
:)
:)
FORTRAN sucked syntactically, C doesnt have this problem, which is why C++/Java/Modula-2 try and copy it.
VB doesnt qualify as a language.
ok..you have a point that the language isnt about libraries, but now that C is around and programmers have used > 1 language, the language has shifted to becoming irrelevant. libraries are all important as is portability, security and performance. drag and drop is *not* the way to go, but neither is the "i'll write volumes of code to manipulate XML/HTML/media files".
The paradigm seems to be shifting to a robust language base with extensive support libraries type thing (GUI/text mode development is the programmers choice - i still code in java with pico) with emphasis on portability, performance and security. which is a *good* thing i might add.
of course theres always M$ doing its usual..but since theyre pushing C# instead of VB, they seem to be learning too.
languages ARE collections of libraries, like it or not. and its going to get much better once we have decent libraries. the core language is nothing more than a wrapper to access the library. are you telling me C is useful without stdlib/stdio ? Syntax is now becoming increasingly irrelevant since most languages use a C or C like syntax (java included).
exactly. sun systems dont have a BIOS. in fact, nothing has a BIOS except PCs. the only reason for the BIOS was DOS - which cant do anything without a BIOS since its just a glorified file loader. most modern systems all have boot PROMS..most of em with their own language (forth in the case of sun) which you can actually code in to change the boot process. BIOSes suck - the faster we get rid of em the better. linux/bsd/most nixes/NT dont use em anyway (SGIs NT workstations all use boot PROMs).
the languages now are really CRUDE. programmers have to build infrastructure whenever we code. thats prolly the most annoying part (and sometimes the most fun) of coding. what i'd like to see is a language which would provide basic functions off the shelf. if i want an HTML editor stuck between two translucent animated buttons which pull up a hex editor and a MP3 player..i shouldnt have to *code* all that. three function calls and a few lines of code should do that for me in (insert language of choice). i should have to build stuff to parse files...if i want a XML tag called weather in a file somewhere on the disk, it should be able to retrieve it for me in one functional call. Java & the GTK stuff has been trying to do that but it doesnt go far enough. I want a seamless environment to manipulate all the functions available, plus i should be able to cut and paste bits from a library of examples available. And i should have a choice of writing bits in assembly and controlling the machines registers at the same time. tall order huh ?
no. there will be no license.
most coders i know would do this (myself included) :
[1] go thru anonymiser and create a hotmail account.
[2] go thru anonymiser and create a geocities account using email from step 1.
[3] dump illegal/dangerous software on geocities and put link to software on freshmeat.
[4] cops shut down geocities/hotmail account.
[5] goto step 1.
lather, rinse, repeat.
heck, with sufficient tools like wget and html redirection this could be automated.
i do this anyway if im reverse engineering stuff.
and : /etc/inetd.conf
6. install bash.
7. go to the hp/ux freeware page and d/l all the stuff.
8. HP maintains a JDK if youre into java. look on hp.com...
9. ssh ssh ssh. install ASAP and comment out the crap in
i've got a A9000 HP/UX 11.0 box without a video card..text terminals suck. try and get a video card from hp if you dont have one. also note that ssh cant authenticate via passwords if you set the trusted base option in HP/UX. use RSA authentication. the trusted stuff also breaks things...so make sure you have at least a text terminal when you install it.
ok, youre a troll but let me address your points. :-
under the traditional process marketing muscle is leveraged, publishers have contacts with bookstores to promote the books etc...even if you are an unknown author. advertising budgets are also available to promote. theres a lot of stuff going around behind the scenes - ive been there.
apache, perl etc are all failures. yes, they are great. yes, people use them. ask yourself why the apache software foundation is non profit. why there is no company which sells perl. why GCC etc cant be sold in the quantities M$ visual C++ is..even though its far superior.. why emacs is not sold like M$ word is. why linus torvalds who wrote linux never made any money of it. linus doesnt even work on a pure linux job. Then ask yourself if they are true commercial sucesses. yes, redhat makes money - but the main money is made with support. not by *selling* the product. you cant sell something people download off the web. did you know concorde has been a failure even if its seen as the fastest aircraft there is ? it doesnt turn a profit - the R&D costs were too high. In the real world of companies which HAVE to survive and generate revenues, working products dont count - the bottom line ALWAYS is more important. have there been some success with GPLed/free products - sure. is there more profit with closed source - YUP. thats why M$ has profits in the BILLIONS. wake up kid - this is reality.
Yup. my email address definitely makes me a looser. real mature.
change Linux to Any OS in your above statement and it will hold true. see solaris - dog slow on 4 CPUs. the way you write the kernel becomes inefficient with a barrier at around 4 CPUs if you want low end performance. high end performance is completely opposite to it. its a tradeoff - no OS can go around it. we might have a high end kernel or a low end kernel VM option in linux soon. recompiling it plugs in the VM you want.
ok. in short this is how it works.
1] it sets up a server socket on port 8080 with a handler for that. any java applet can do this.
2] it bypasses the java.io.File stuff which is sandboxed by using the netscape file://c:/ command (which is friggin brilliant if i may say so...) to browse files. since its running in a web browser anyway, it can send commands to the local browser.
Fix :
Simple. expand the security sandbox to applets cant use file:// to exploit their local browser.
it will only work for well known authors. supposing i wanted to write a book, wrote several chapters..gave it to a publisher under this protocol. would he pay me ? unlikely. would anyone bother to look at it ? nope. only well known/established authors could hope to make this fly. mark me down if you will - but the same is true of code released under the GPL. without an established base of fans/supporters - you go nowhere. and thats the main reason open source doesnt fly in the real world.
Your assumption is VERY wrong. If there were no copyrights and all code was open, whats to stop a programmer working for a company to throw the code out in the open ? After all, the company cant sue him for copyright or ip infringement. and how many programmers, given the choice, would lock their code up ? not many. at least not the intelligent ones. given the trend, how many people would buy closed source software ?
the difference between software and music is that there is no need to duplicate the PROCESS OF WRITING MUSIC. you simply "execute" it to enjoy the full benefits. with software this is not the case. the PROCESS is more important than the final EXECUTION. The GPLNET guy is a troll - i wouldnt give two shits about him. whats he going to do - release linux kernel binaries via napster ? sure - go ahead and everyone will laugh.
As a programmer, i'd love to see copyright/ip disappear. programmers enjoy programming new stuff - they dont enjoy duplication of effort. all code *should* be public domain. people will pay for *systems integration* - which is a lot more important. its not the code - its whether the system i want to build will work. and thats whats really important.
To use an analogy, the blueprints of a bus may be available freely, but people still buy busses from GM or wherever. and its the revenue earned from the poeple actually *using* the bus which is important.
hell no. GSM phones can be cloned and eavesdropped since the encryption was broken since 98. http://www.isaac.cs.berkeley.edu/isaac/gsm-faq.htm l
I should point out the kids in the US of A are already FAR more privilidged than kids who grow up out of the US. Should we really care whether some kid who wants to view a porn site at a library is not allowed to ? Or should we be more bothered about the fact that there are NO COMPUTERS in libraries not located in the US ? Heck 99% of the planet doesnt have a computer in their homes let alone the local underfunded library.
And on a related note - the abortion information is far more likely to be found in large quantities in any of the books in the library than on the net. libraries are for reading - lets keep it that way. web surfing should be kept strictly in school computer labs. There nothing more annoying than web surfers clogging up the library.
(-1, Offensive)