Do we really want our scientific community to be comprised of people who are in it for the money and attention? Given the choice between the guy looking for financial success and the geek looking to keep scratchin' that curiosity itch, I'm betting all my chips on the curious geek.....every time.
WTF people?!? How'd this even get on Slashdot? With all of the elitist attitudes espoused around here, I'm surprised you'd even consider encouraging the acceptance of bourgeois pricks into a field that should be filled with guys who are doing it because they're fucking CURIOUS!
My landlord owns a small local business that has about 7 workstations and a server. For about the first year and a half, I never paid rent. I would just go in and provide tech support, do a new install, or whatever every once in a while and we'd call it even. I tried as hard as I could to keep this up for as long as possible, but alas, I've gotten their network running so smoothly that there's not as much of a need for me there these days.
When I was in the 3rd grade (~1980) they decided to teach us BASIC in the new computer lab at the elementary school. I was excited about showing off my skillz to the teacher and my fellow classmates; I was already familiar with BASIC since I had regulary been going to the mall and used the trs-80's on display at Radio Shack to try out the things I had learned in books. On occasion, I would go down to Video Concepts and use their Commodore 64, but I didn't like it because it wouldn't tell you about any sytax errors until you tried to run the program. The trs-80 would notify you of any syntax errors as soon as you completed a line.
Unfortunately, and to my demise, this taught me poor programming habits (I relied on those damned line-by line syntax checks, causing me to be less careful). Sure enough, the third grade computer lab was comprised of Commodore 64's. "No problem," I remember thinking to myself, "I'll just be careful."
The problem is, our first assignment was to type a program (~20 lines) from a sheet of paper to the computer. We weren't allowed to run it until the teacher came to grade us. It was a pass-fail assignment/test. If the program ran, you passed. Thanks to the sloppy habits I had developed on those fucking trs-80's I failed.
This is not unique to OO; one can do the same in procedural programming. I do it all the time in my (non OO) PHP scripts. Just use regex to check the safety of the user's input.
One possibility would be an annual award program which could include - for example - a "best documentation" category. The combination of a cash prize (it needn't be large) plus the bragging rights for having won could provide the necessary nudge to improving open-source code.
I am intrigued by your ideas and wish to subscribe to your newsletter.
But seriously, that's not a bad idea. Maybe we could get the FSF or somebody to do something like that.
I agree with you that the author has some good points, but she generalizes a bit. For example, the lack of documentation issue also applies to proprietary software in some (many ?) cases. Back in the day, I finally decided to bite the bullet, stop using Win 3.x, and learn Win 95. Where was the fucking documentation for that!? I had to go buy Windows 95 for Dummies (don't knock it....it's a damned good series; anybody who disagrees has hubris issues), not because the new UI was a complete non-intuitive mystery, but because I wanted a good summary of all the little things. Like, what were all those icons in the Documents folder under the Start menu? And why were some of the documents still listed there even after I deleted them? etc....
I'm not saying I altogether disagree with the author on the documentation point (or the "it's my project and I'll code what I want to problem"). There are some cases where these two issues combined piss me off more that either of the problems alone (AA: 1 + 1 = 3):
For example, I recently decided to give Litestep a try on my Win2K laptop. I was generally impressed with it. It installed fine and worked (more or less) as advertised. I started experimenting with different skins/themes, and finally settled on what I liked. There were a few annoying glitches. The GUI search for files function is now broken, but I can deal with that (use DOS or Cygwin). Maximized windows still show the window border (no biggie), and there was no clear way to change the desktop wallpaper (quick hack: I put the image I wanted to use in the appropriate directory and renamed it to what the shell expected to be there).
Overall, these glitches were only slightly annoying (after all, it's free). I eventually ran accross an issue I couldn't work around, so I checked out their support on line. On the home page, they list recent questions submitted by users and provide a link to an answer. One of the questions was "How do I change my desktop wallpaper?" I figured I'd click on the link to see what the "official" answer was. After all, there doesn't seem to be an intuitive/simple way to do this in the GUI. You know what the official answer was? (paraphrased) "If you can't figure this out, then stop using Litestep. It's already in the FAQ."
What the fuck!?? I started to look at some of this assholes other "answers" to his users' questions and many of them conveyed a similar attitude. I mean, they might as well drop the veneer and let out their full hostility with shit like, "yOuR sUcH a FsCkInG n00B...kw1t n0w B4 wE 0wN0rz j00r B0xEn!!!11!!"
So now I say, "Fuck Litestep." They're just a bunch of fucking juveniles that are hostile to their users. I'd rather deal with Microsoft where at least they don't shit talk their users on their web site.
If you go back and read some of Marsh's statements after this became public knowledge, you'll see that he expresses grave concern about the publicity because EV1's licensing terms with SCO WERE supposed to be confidential.
There are two simple ways to get around this; both are browser independent. The first way (not user friendly) is to offer a non-clickable link (i.e. plaintext). This forces the user to copy and paste the link into the address bar, which is seen by the target server as a direct request (with no referer).
The second method (slightly less user unfriendly) is to make use of a meta-refresh tag. The clickable link that you provide should point to a local page that is comprised of nothing but a meta-refresh tag that redirects you to the target server immediately (i.e. using a 0 second delay). This also strips the referer header and is seen by the target server as a direct request. This is still somewhat user unfriendly because it breaks the back button; however, this can be remedied by giving the meta-refresh tag a 2 or 3 second delay and offering an ad or picture as content for that brief moment.
Slashdot Mirror
Do we really want our scientific community to be comprised of people who are in it for the money and attention? Given the choice between the guy looking for financial success and the geek looking to keep scratchin' that curiosity itch, I'm betting all my chips on the curious geek.....every time.
WTF people?!? How'd this even get on Slashdot? With all of the elitist attitudes espoused around here, I'm surprised you'd even consider encouraging the acceptance of bourgeois pricks into a field that should be filled with guys who are doing it because they're fucking CURIOUS!
[/end rant]
Anyway. Yeah, what's up with that?
My landlord owns a small local business that has about 7 workstations and a server. For about the first year and a half, I never paid rent. I would just go in and provide tech support, do a new install, or whatever every once in a while and we'd call it even. I tried as hard as I could to keep this up for as long as possible, but alas, I've gotten their network running so smoothly that there's not as much of a need for me there these days.
When I was in the 3rd grade (~1980) they decided to teach us BASIC in the new computer lab at the elementary school. I was excited about showing off my skillz to the teacher and my fellow classmates; I was already familiar with BASIC since I had regulary been going to the mall and used the trs-80's on display at Radio Shack to try out the things I had learned in books. On occasion, I would go down to Video Concepts and use their Commodore 64, but I didn't like it because it wouldn't tell you about any sytax errors until you tried to run the program. The trs-80 would notify you of any syntax errors as soon as you completed a line.
Unfortunately, and to my demise, this taught me poor programming habits (I relied on those damned line-by line syntax checks, causing me to be less careful). Sure enough, the third grade computer lab was comprised of Commodore 64's. "No problem," I remember thinking to myself, "I'll just be careful."
The problem is, our first assignment was to type a program (~20 lines) from a sheet of paper to the computer. We weren't allowed to run it until the teacher came to grade us. It was a pass-fail assignment/test. If the program ran, you passed. Thanks to the sloppy habits I had developed on those fucking trs-80's I failed.
http://www.xwt.org/
This is not unique to OO; one can do the same in procedural programming. I do it all the time in my (non OO) PHP scripts. Just use regex to check the safety of the user's input.
I sent my insecure scripts to counseling and they feel much better about themselves now.
One possibility would be an annual award program which could include - for example - a "best documentation" category. The combination of a cash prize (it needn't be large) plus the bragging rights for having won could provide the necessary nudge to improving open-source code.
I am intrigued by your ideas and wish to subscribe to your newsletter.
But seriously, that's not a bad idea. Maybe we could get the FSF or somebody to do something like that.
I agree with you that the author has some good points, but she generalizes a bit. For example, the lack of documentation issue also applies to proprietary software in some (many ?) cases. Back in the day, I finally decided to bite the bullet, stop using Win 3.x, and learn Win 95. Where was the fucking documentation for that!? I had to go buy Windows 95 for Dummies (don't knock it....it's a damned good series; anybody who disagrees has hubris issues), not because the new UI was a complete non-intuitive mystery, but because I wanted a good summary of all the little things. Like, what were all those icons in the Documents folder under the Start menu? And why were some of the documents still listed there even after I deleted them? etc....
I'm not saying I altogether disagree with the author on the documentation point (or the "it's my project and I'll code what I want to problem"). There are some cases where these two issues combined piss me off more that either of the problems alone (AA: 1 + 1 = 3):
For example, I recently decided to give Litestep a try on my Win2K laptop. I was generally impressed with it. It installed fine and worked (more or less) as advertised. I started experimenting with different skins/themes, and finally settled on what I liked. There were a few annoying glitches. The GUI search for files function is now broken, but I can deal with that (use DOS or Cygwin). Maximized windows still show the window border (no biggie), and there was no clear way to change the desktop wallpaper (quick hack: I put the image I wanted to use in the appropriate directory and renamed it to what the shell expected to be there).
Overall, these glitches were only slightly annoying (after all, it's free). I eventually ran accross an issue I couldn't work around, so I checked out their support on line. On the home page, they list recent questions submitted by users and provide a link to an answer. One of the questions was "How do I change my desktop wallpaper?" I figured I'd click on the link to see what the "official" answer was. After all, there doesn't seem to be an intuitive/simple way to do this in the GUI. You know what the official answer was? (paraphrased) "If you can't figure this out, then stop using Litestep. It's already in the FAQ."
What the fuck!?? I started to look at some of this assholes other "answers" to his users' questions and many of them conveyed a similar attitude. I mean, they might as well drop the veneer and let out their full hostility with shit like, "yOuR sUcH a FsCkInG n00B...kw1t n0w B4 wE 0wN0rz j00r B0xEn!!!11!!"
So now I say, "Fuck Litestep." They're just a bunch of fucking juveniles that are hostile to their users. I'd rather deal with Microsoft where at least they don't shit talk their users on their web site.
The objective, moral truth is that there is no objective moral truth.
If you go back and read some of Marsh's statements after this became public knowledge, you'll see that he expresses grave concern about the publicity because EV1's licensing terms with SCO WERE supposed to be confidential.
If you REALLY have everything, that would include a container to hold it.
There are two simple ways to get around this; both are browser independent. The first way (not user friendly) is to offer a non-clickable link (i.e. plaintext). This forces the user to copy and paste the link into the address bar, which is seen by the target server as a direct request (with no referer).
The second method (slightly less user unfriendly) is to make use of a meta-refresh tag. The clickable link that you provide should point to a local page that is comprised of nothing but a meta-refresh tag that redirects you to the target server immediately (i.e. using a 0 second delay). This also strips the referer header and is seen by the target server as a direct request. This is still somewhat user unfriendly because it breaks the back button; however, this can be remedied by giving the meta-refresh tag a 2 or 3 second delay and offering an ad or picture as content for that brief moment.
You don't know wha the fuck you're talking about. Read Feist, moron.