...was half a vote against Gore, and half a vote against Bush. Some of us honestly do not care which of the two win, you know, since both have made a living of fellating big corporations for ad money.
My feeling is that if Nader really does prevent Gore from getting elected -- and from the looks of Florida that may actually happen -- maybe the Dems will be humiliated enough to put up someone worth voting for next time.
There's also a commercial Mac/MSDOS emulator available for 30 bucks at www.intellivisionlives.com, which comes with about 50 game images and the rest are pretty easy to come by on Usenet or the web.
I'm sure someone wearing a tie would find a reason to object anyway, though.
I'm one of the "culprits" but I'm OK with it.
on
Shortcomings Of OSS?
·
· Score: 1
Not that I've released text editor #174*, but I did release a GPL'ed CD cataloging program, probably about the 50th of its kind. I wrote it because all the existing Linux CD cataloging programs relied on the user typing an identifier for each CD they used, rather than just pulling the ISO9660 disk label off the CD.
Could I have submitted a patch to someone else's project? Probably. But replacing the primary key in a CD database would have made for a pretty big patch. Besides, I wanted to prove to myself that you could write a graphical app using Perl and Qt. I wanted to prove you could write a database app that didn't require a database server. I wanted a graphical app with a command line version available. And I also hoped that other CD catalog authors would take my disk ID concept and include it in their real cataloger programs, since that feature is pretty common on "other platforms."
So far (AFAIK) that hasn't happened, but I'm still hopeful. Meanwhile I've gotten a few new-feature patches emailed to me, but haven't incorporated them yet. Why not? Well, the program does what I need it to, after all, and the patches would have extended the functionality further beyond that of CD cataloger. I no more want to browse directories with a CD cataloger, for example, than I want to read news with a web browser.
I also want to revamp it, probably rewrite it in C++ as a fullblown KDE app, include MySQL support because text files are way too slow for million-record databases, RPMify it the way you can't easily do with perl programs with dependencies, but all that is Real Soon Now.
This makes me part of the problem, I know, but I hope I've at least explained why it doesn't bother me.
*And I did actually write a text editor - my very first Tcl experiment was a Notepad clone. I also had the sense to never release it;)
Whether or not I'm on drugs, I run Mozilla M17 (not Netscape 6 PRs) as my main browser at home now.
Mozilla is way faster than Netscape 4.7x, renders fonts a lot better, and doesn't crash a couple times a day like Netscape 4 does (and Netscape 6PR2 did.) It has trouble loading some Java applets and it was a bitch to get SSL working, but I'm looking forward to a nicely packaged 1.0. But I downloaded this one because the Nautilus "preview" wanted it, and I'm glad I did.
Even more so, I'm looking forward to someone writing a browser that uses Gecko and supports Java/JS/SSL, but doesn't do all that lame non-standard XML user interface garbage that looks nothing like my other windows. I'd be running Galeon right now if it even did JS and SSL.
I haven't tried Konqueror yet because I can't get KDE2 to install on my Mandrake 7.0 box - too old a kernel version and I haven't had time to download the 7.2 beta ISO's. One of these weekends.
Xfce really reminds me of the OS/2 Warp 3.0 app starter panel. I always thought CDE's panel was an imitation of that panel, but I suppose CDE could have come first. I suppose they both came from HP Dashboard.
Anyway, I don't see Xfce as "competition" for Gnome/KDE, both of which I see primarily as transition technologies to make users coming from other OSes a little more comfortable. I think there are going to be more Windows and MacOS users coming over than Warp3 and CDE users.
For my own part, I almost never run things out of a panel, but if I do, I much prefer the "hidden slide-out menu at the top of the screen" variety like the old AmigaDOS, or a configurable desktop root menu like everything else used to use. But hey, more choice is better. There are a million Explorer replacements for Windows too, and you don't see too many holy wars on their turf.
Maybe to be REALLY lightweight what we need is "bwm", the bash window manager, whose desktop looks just like a console except you can bring up X apps in front of it, and have all the widgets rendered in ASCII;)
Actually, that seems like the traditional big-free-software-project way of doing things - look at (GNU/)Linux itself.
Speaking of the article in general, it seems ironic to fault the Gnome guys for not turning down a corporate foundation's support when KDE itself exists more or less at the whim of Trolltech. But I don't want to write too much since if there were an article moderation system (hint hint) I would have marked this one as flamebait.
I use KDE now, and will move to 2.0 when there are final Mandrake RPM's for me to install on the machines I support. Gnome will be awesome when all this stuff comes to pass stably (Nautilus, Gecko embedding, GnomeStarOffice, etc) but just as with KDE before 2.0 goes final, it doesn't mean a whole lot to the end user. I hope it does happen quickly, because I like the gnome ethos (and look and feel) a lot better than KDE and would prefer to switch once it comes of age.
Meanwhile, if a few of the KDE folks feel threatened by this for some reason, perhaps they should concentrate on making KOffice the real deal since they've already got a jump on the browser, shell and integration side of things. Being able to import/export Word documents would be a good thing to focus on. While they're at it maybe they can put pressure on Troll somehow to fix their license rather than hoping to move the mountain to Moses.
Deja's not a free service at all. Deja gets compensated for their service by having its users look at advertisements. You might call it a trade of services. But you can't call it "free", nor even "without cost." With this service, they've added to that cost but shifted some of it to the content providers themselves, many (most?) of whom aren't even aware of Deja's existence. For them the cost is the potential of a tarnished reputation - I certainly couldn't post a review of a hardware product on Usenet, for example, if I knew that it would be archived permanently with links to that product or a competitor. I wonder if a.signature license would help.
I know this is way too late in the thread to matter but...
Why a single patron? Here is an example of a group with a big back catalog and established fan base, leveraging that fanbase as one big virtual patron to avoid having to sign another record deal. I guess we won't know whether it worked until December (I'm not a fan myself,) but it seems like an interesting experiment.
There always seems to be the assumption that if you make something free-software or open-source, you're required to throw it on your web site, announce it on freshmeat, and whatever else. There seems to be a suggestion that if anyone comes knocking at your door asking for your software, you've got to hand it out to them.
I've looked at the GPL in particular, and I don't think this is the case. It seems to me that the GPL prevents you from restricting *further* distribution of your code, once you actually distribute it. Meaning you have to give the source code to your clients and they, at their option, may distribute it further without paying you as long as they respect the same restrictions.
But so what? If you're selling something into an obviously vertical market, who says you can't charge $2,000 or $10,000 or $500,000 for the GPL'ed code? Do you really think Client A is going to turn around and give what they paid for to Potential Client B, who happens to also be Client A's competition? Do you think they're going to put it up on their website with a big notice saying "Here's what we blew our IT budget on this year, knock yourselves out?"
It seems like the GPL is better suited to this sort of thing than other licenses because it prevents your clients from turning the product into a proprietary package which it then uses to compete against you. You could write a non-compete clause into the support contract easily enough, though, which would probably be a good idea even if you do GPL the code.
There are the other XPL's too (Netscape et al), which seem to provide some of the benefits of open source to the squeamish, and you could always roll your own since everyone else seems to anyway.
Actually, Americans do continue to have constitutional rights at work - regardless of what they sign - except perhaps those in the military. I don't know whether privacy is actually a constitutional right, though, or what it entails if so. Certainly some elements of privacy were legally removed during the corporate-happy legal environment of the 80's and 90's, but by no means all of them. That there are slashdotters who believe this demonstrates that the "if you don't like it, find another job" argument needs some serious rethinking. Then again, that's exactly what I myself did.
Slashdot Mirror
It's about who can be the first to port MAME!
My feeling is that if Nader really does prevent Gore from getting elected -- and from the looks of Florida that may actually happen -- maybe the Dems will be humiliated enough to put up someone worth voting for next time.
http://www.primenet.com/~im14u2c/intv/
There's also a commercial Mac/MSDOS emulator available for 30 bucks at www.intellivisionlives.com, which comes with about 50 game images and the rest are pretty easy to come by on Usenet or the web.
I'm sure someone wearing a tie would find a reason to object anyway, though.
Could I have submitted a patch to someone else's project? Probably. But replacing the primary key in a CD database would have made for a pretty big patch. Besides, I wanted to prove to myself that you could write a graphical app using Perl and Qt. I wanted to prove you could write a database app that didn't require a database server. I wanted a graphical app with a command line version available. And I also hoped that other CD catalog authors would take my disk ID concept and include it in their real cataloger programs, since that feature is pretty common on "other platforms."
So far (AFAIK) that hasn't happened, but I'm still hopeful. Meanwhile I've gotten a few new-feature patches emailed to me, but haven't incorporated them yet. Why not? Well, the program does what I need it to, after all, and the patches would have extended the functionality further beyond that of CD cataloger. I no more want to browse directories with a CD cataloger, for example, than I want to read news with a web browser.
I also want to revamp it, probably rewrite it in C++ as a fullblown KDE app, include MySQL support because text files are way too slow for million-record databases, RPMify it the way you can't easily do with perl programs with dependencies, but all that is Real Soon Now.
This makes me part of the problem, I know, but I hope I've at least explained why it doesn't bother me.
*And I did actually write a text editor - my very first Tcl experiment was a Notepad clone. I also had the sense to never release it ;)
Mozilla is way faster than Netscape 4.7x, renders fonts a lot better, and doesn't crash a couple times a day like Netscape 4 does (and Netscape 6PR2 did.) It has trouble loading some Java applets and it was a bitch to get SSL working, but I'm looking forward to a nicely packaged 1.0. But I downloaded this one because the Nautilus "preview" wanted it, and I'm glad I did.
Even more so, I'm looking forward to someone writing a browser that uses Gecko and supports Java/JS/SSL, but doesn't do all that lame non-standard XML user interface garbage that looks nothing like my other windows. I'd be running Galeon right now if it even did JS and SSL.
I haven't tried Konqueror yet because I can't get KDE2 to install on my Mandrake 7.0 box - too old a kernel version and I haven't had time to download the 7.2 beta ISO's. One of these weekends.
Anyway, I don't see Xfce as "competition" for Gnome/KDE, both of which I see primarily as transition technologies to make users coming from other OSes a little more comfortable. I think there are going to be more Windows and MacOS users coming over than Warp3 and CDE users.
For my own part, I almost never run things out of a panel, but if I do, I much prefer the "hidden slide-out menu at the top of the screen" variety like the old AmigaDOS, or a configurable desktop root menu like everything else used to use. But hey, more choice is better. There are a million Explorer replacements for Windows too, and you don't see too many holy wars on their turf.
Maybe to be REALLY lightweight what we need is "bwm", the bash window manager, whose desktop looks just like a console except you can bring up X apps in front of it, and have all the widgets rendered in ASCII ;)
Actually, that seems like the traditional big-free-software-project way of doing things - look at (GNU/)Linux itself.
Speaking of the article in general, it seems ironic to fault the Gnome guys for not turning down a corporate foundation's support when KDE itself exists more or less at the whim of Trolltech. But I don't want to write too much since if there were an article moderation system (hint hint) I would have marked this one as flamebait.
I use KDE now, and will move to 2.0 when there are final Mandrake RPM's for me to install on the machines I support. Gnome will be awesome when all this stuff comes to pass stably (Nautilus, Gecko embedding, GnomeStarOffice, etc) but just as with KDE before 2.0 goes final, it doesn't mean a whole lot to the end user. I hope it does happen quickly, because I like the gnome ethos (and look and feel) a lot better than KDE and would prefer to switch once it comes of age.
Meanwhile, if a few of the KDE folks feel threatened by this for some reason, perhaps they should concentrate on making KOffice the real deal since they've already got a jump on the browser, shell and integration side of things. Being able to import/export Word documents would be a good thing to focus on. While they're at it maybe they can put pressure on Troll somehow to fix their license rather than hoping to move the mountain to Moses.
Deja's not a free service at all. Deja gets compensated for their service by having its users look at advertisements. You might call it a trade of services. But you can't call it "free", nor even "without cost." With this service, they've added to that cost but shifted some of it to the content providers themselves, many (most?) of whom aren't even aware of Deja's existence. For them the cost is the potential of a tarnished reputation - I certainly couldn't post a review of a hardware product on Usenet, for example, if I knew that it would be archived permanently with links to that product or a competitor. I wonder if a .signature license would help.
Why a single patron? Here is an example of a group with a big back catalog and established fan base, leveraging that fanbase as one big virtual patron to avoid having to sign another record deal. I guess we won't know whether it worked until December (I'm not a fan myself,) but it seems like an interesting experiment.
"I suppose you'd prefer a little animated clown who would juggle over to the little file cabinet, wink at you and point to the right drawer?"
There always seems to be the assumption that if you make something free-software or open-source, you're required to throw it on your web site, announce it on freshmeat, and whatever else. There seems to be a suggestion that if anyone comes knocking at your door asking for your software, you've got to hand it out to them.
I've looked at the GPL in particular, and I don't think this is the case. It seems to me that the GPL prevents you from restricting *further* distribution of your code, once you actually distribute it. Meaning you have to give the source code to your clients and they, at their option, may distribute it further without paying you as long as they respect the same restrictions.
But so what? If you're selling something into an obviously vertical market, who says you can't charge $2,000 or $10,000 or $500,000 for the GPL'ed code? Do you really think Client A is going to turn around and give what they paid for to Potential Client B, who happens to also be Client A's competition? Do you think they're going to put it up on their website with a big notice saying "Here's what we blew our IT budget on this year, knock yourselves out?"
It seems like the GPL is better suited to this sort of thing than other licenses because it prevents your clients from turning the product into a proprietary package which it then uses to compete against you. You could write a non-compete clause into the support contract easily enough, though, which would probably be a good idea even if you do GPL the code.
There are the other XPL's too (Netscape et al), which seem to provide some of the benefits of open source to the squeamish, and you could always roll your own since everyone else seems to anyway.
Guess "document contains no data" doesn't mean it wasn't posted ;)
Actually, Americans do continue to have constitutional rights at work - regardless of what they sign - except perhaps those in the military. I don't know whether privacy is actually a constitutional right, though, or what it entails if so. Certainly some elements of privacy were legally removed during the corporate-happy legal environment of the 80's and 90's, but by no means all of them. That there are slashdotters who believe this demonstrates that the "if you don't like it, find another job" argument needs some serious rethinking. Then again, that's exactly what I myself did.