The article starts out detailing the alleged history of RFC822 and how it "laid the foundation of SMTP". Problem is, 822 doesn't have anything to do with SMTP. That's covered in RFC821. It's downhill from there. Pseudo-technical details that lack even the most basic understanding of fundamental technologies. Please take away these people's word processors (and their MUAs too, I bet they do most of their damage via email)... maybe they can still be trusted with crayons (though in Dvorak's case, I don't think he should be given that either).
Gee, what could those "de facto" standards be? Say cmd-c to copy, cmd-x to cut and cmd-v to paste? Odd, that's exactly what Macs *do*. Of course, before anyone says Apple copied MS, it's worth noting the these standards are de jure, not de factor... their part of what's called CUA (common user architecture), and they predate Window 1.0.
OTOH, one thing that *does* bother me about the Mac is that you can't pull down individual menus with Alt-F/Alt-E/etc. The commands within the menus tend to have shortcuts, but if you forget exactly what the set of "Edit" options are (for the particular program), it's less straightforward to see them (except by using the mouse, which I like to avoid where unnecessary). My own solution is to enable Alt-M to focus on the menu in the accessibility settting. But that still makes me press right-arrow several times to get to the menu of interest.
This story has made me think that it would be nice if the Wikipedia software kept a count of readership per article too. A complete list of connecting IPs would be unwieldy and a bit pointless; but a simple counter would be a nice feature.
Obviously, given that content is GFDL'd, this doesn't automatically tell you how many people read something that is syndicated. But it would be interesting when reading an article to know whether 3 people or 3 million people had read that article.
_The Register_ doesn't quite seem to understand that prank articles aren't good topics 365 days a year. Or maybe they should hire some writers from _The Onion_ who have more talent.
I had never used the "move" function, I confess (the fetal hormones thing I did as a redirect). But my point turns out still to be correct. For example, I created a page called/sandbox1 at:
maxwell's demon is incorrect. For example I started the article "Fetal hormones and sexual orientation". Disclaimer: it's not a very good article, and I only copied the text proposed by someone else in a discussion on a different article (but that's how Wikipedia works its magic, start with *something* and let it improve from there). The page I first created is:
Over time, editors felt that the title "Prenatal hormones and sexual orientation" was a better one (and in the meanwhile, the article has got at least a bit less bad). A recent edit can be found at:
However, if you just look for the old title, you will be redirected to the new title. But the old title still retains its independent edit history, and the ability to directly URL reference each old revision.
Ah... it just makes me miss the days when I had slacker college students:-).
Certainly I entirely concur that an encyclopedia is not a primary source for academic research, especially not at a college level. I was just trying to help with the narrow technical question of citing specific known texts (as one might with a specific edition of a printed work).
This has a rather painfully simple technical solution. Just demand that your students cite specific version of an article by URL. For example, the *current* article on Slashdot at Wikipedia is:
Someone will probably modify the Slashdot article itself in the next five minutes. But the URL I give will retain unchanged forever. Of course, if the edit someone does in five minutes improves the information, my snapshot URL won't present the improvement. But then the future improvement isn't what your students read either.
I happen to know that BWJones really is a genuine expert in some specialized medical areas. I'm pretty sure he's not just talking about arcana of his favorite musical group. So in a sense, he really should be shown a certain deference in edits. Not an unlimited deference, of course: what goes in an encyclopedia like Wikipedia is not just the "best knowledge" but also the best writing, and content that is neutral point-of-view, verifiable, reflects consensus, and so on. "Experts" are not automatically the best deciders of all these things.
Still, there's no doubt that people who know a little are overly hasty to change the edits of people who know a lot. In my experience, as a trend this gets ironed out. Editors who consistently provide evidence and show expertise are, over time, granted proper deference (but not a fixed-in-stone sort). And making a clear case on Wikipedia talk pages before introducing substantial or contentious material tends to make the addition happen more smoothly.
However, more traditional "hired gun" authority suffers from an equally pernicious problem, probably worse even. While BWJones is an expert is some things, he may not be the *particular* expert a traditional encyclopedia hires. And specialized fields have their own contentious issues, and differing professional opinions at the margins of the topics. If someone with a competing theory to that BWJones holds is hired for a standard medical encyclopedia, BWJones gets ZERO say in which content goes in. Whoever is hired (and is indeed another expert in the relevant area), decides the whole of what goes there. Sure that one (or a couple) paid editor acts in good faith, and tries to be neutral. But it's almost impossible not to present your favorite theory as more likely than the competing approach. The anarchy of Wikipedia tends to smooth out that bias.
But ESR isn't at all a rabid proponent of Linux (heck, he disses the GPL nowadays), but rather a proponent of "the greatness of Eric Raymond". I confess Theo de Raadt comes closer to the characterization I said was absent. But even there it's different: Murphy, O'Gara, and Dvorak are such foolish figures because they don't know a thing about actual programming or technology; whatever excesses de Raadt might engage in, he *is* the principle author of an excellent operating system.
Interestingly, neither Linux, *BSD, nor Apple/OSX seem to have doltish lapdogs in the same way. Sure, proponents of Linux, BSD and OSX sometimes say foolish things. But there's no single figure from those communities who can be so relied upon to unthinking chant trollish drivel.
Ummm... where can I get QuickView Plus and Windows 95, exactly?:-) (actually, I think I *tried* QV+ about 5 years ago, and it didn't support that old format, even that far back). Of course, if you're volunteering, email me.
100 years is a LONG time! I'm sure OASIS won't introduce something really incompatible in 2010; and many tools will almost certainly support OpenDocument v.1 in 2010. But 2105 is... well, what were autos, trains, electrical outlets, light bulbs, telephones, sound recordings, typewriters, etc. like in 1905? You can recognize all those things over the century, but a lot of details have changed.
I should have written "OpenDocument" to be more exact. But the idea is the same. Quite possibly, 100 years from now only OpenDocument v.17+ will be supported in widespread existing tools. Version 1 isn't necessarily the end point, and it may be the features become incompatible over a number of versions. But WhiteWorlf666 is exactly correct that since the v.1 *standard* will still be available, there's nothing stopping future programmers from writing an import filter for current tools (which may be really easy by then, using some kind of high-level XML interface that might exist then).
Fetchmail is/was a fairly useful program. And his two famous essays, "Cathedral and Bazaar" and "Homesteading the Noosphere" are reasonably interesting (not entirely accurate, but also not entirely without some reasonable analysis). But 1998 was the last time ESR did ANYTHING even remotely of public interest.
In a recent interview, Eric Raymond shocked many Free Software developers. Interviewed developers commented "What? Is he still around? Now there is someone you don't think about anymore." Raymond's contentious statements were uniformly viewed as desperate cries for attention by a once notable software developer. A sympathetic developer commented, "I certainly hope ESR gets the professional counciling he so obviously needs; he just needs to learn that you can lead a productive life outside the limelight. I know of a good support group for people like him and Dvorak."
I have some old papers I wrote, back in the MS-Word/DOS days. Yeah, I know I'm old:-) (and didn't know better yet, in the late 1980s); but there are people and documents that matter which are even older. I'd genuinely like to make them available along with various other writing of mine. Unfortunately, absolutely nothing appears able to read these old formats, especially not anything made by Microsoft.
On the other hand, when I wrote papers in WordPerfect 4.2 or so, not much later, those formats are still happily read by recent WordPerfect versions. Not to say WP is a perfectly open format either, but at least it's not quite so breaky as MS formats are.
Fortunately for me, the old Word/DOS format is like all those wordprocessor formats of the 1980s: mostly ASCII text, with a smattering of control codes in it. Ultimately, with a little bit more work than should be needeed, I can extract the text and reformat the document. Had this been Word97 or something, I'd just have a steaming blob of undocumented binary data. Which exactly what all MS formats will be 20 years from now. I'm not so happy that WordPerfect moved to a more binary-ish format around WP6/7; they were also quite discernable using 'less' in the WP5.1/2 timeframe. They've retained better compatibility (even filters for the older versions to read the newer thing). But "readable in 'less'" is an important quality.
Luckily OpenOffice format is now, and will always remain "readable in less". XML is like that. Not that it's quite effortlessly readable; but I don't see XML parsers as going away, so extracting data will still be quite reasonable in 50 years, even if the format itself is no longer implemented in current products. Not effortless, but not outrageously difficult.
Definitely pompous ass material, but I think we knew that.
Maybe an even better analogy would be getting spam for penis enlargement, then writing an angry letter in return boasting of how large your penis already is without their product (and how dare they suppose otherwise).
The MS recruitment letter is just that: spam. Other than the name at the top, they probably sent out 5000 of them the same day. And no, MS didn't check each other *potential candidate's* biography either, just that their resume (or some other document) said they were a computer programmer or similar.
I'm hardly pro-spam, but when I get it, I don't think I'm being *personally* insulted either.
"Begging the question is the term for a type of fallacy occuring in deductive reasoning in which the proposition to be proved is assumed implicitly or explicitly in one of the premises"
Sun really has contributed some significant things to FOSS. OpenOffice is hugely important, for example. And not as code per se, but as standards, things like NFS are wonderful.
Probably much of the problem comes from the erratic behavior of Sun itself. Johnathan Schwartz in particular has, on alternating weeks, either sucked up to the FOSS community or belligerently insulted FOSS. And their patent issues are still looming, and not at all FOSS friendly, e.g.:
Plus the whole matter of Sun being one of the major semi-backdoor funder of the SCO lawsuits really hasn't won them any love among FOSS users. A few million dollars spent on trying to kill Linux through deception and spurious lawsuits doesn't spread the love.
Slashdot Mirror
The article starts out detailing the alleged history of RFC822 and how it "laid the foundation of SMTP". Problem is, 822 doesn't have anything to do with SMTP. That's covered in RFC821. It's downhill from there. Pseudo-technical details that lack even the most basic understanding of fundamental technologies. Please take away these people's word processors (and their MUAs too, I bet they do most of their damage via email)... maybe they can still be trusted with crayons (though in Dvorak's case, I don't think he should be given that either).
Gee, what could those "de facto" standards be? Say cmd-c to copy, cmd-x to cut and cmd-v to paste? Odd, that's exactly what Macs *do*. Of course, before anyone says Apple copied MS, it's worth noting the these standards are de jure, not de factor... their part of what's called CUA (common user architecture), and they predate Window 1.0.
OTOH, one thing that *does* bother me about the Mac is that you can't pull down individual menus with Alt-F/Alt-E/etc. The commands within the menus tend to have shortcuts, but if you forget exactly what the set of "Edit" options are (for the particular program), it's less straightforward to see them (except by using the mouse, which I like to avoid where unnecessary). My own solution is to enable Alt-M to focus on the menu in the accessibility settting. But that still makes me press right-arrow several times to get to the menu of interest.
sudo rm -rf
A really nasty one too. Naturally, you'll need to enter your admin password to operate it... but that's the rule for Mac viruses, after all.
This story has made me think that it would be nice if the Wikipedia software kept a count of readership per article too. A complete list of connecting IPs would be unwieldy and a bit pointless; but a simple counter would be a nice feature.
Obviously, given that content is GFDL'd, this doesn't automatically tell you how many people read something that is syndicated. But it would be interesting when reading an article to know whether 3 people or 3 million people had read that article.
_The Register_ doesn't quite seem to understand that prank articles aren't good topics 365 days a year. Or maybe they should hire some writers from _The Onion_ who have more talent.
You think? Senile at 55 y.o.? I'd definitely believe "not quite right in the head", but too young for senility, I'd reckon.
I had never used the "move" function, I confess (the fetal hormones thing I did as a redirect). But my point turns out still to be correct. For example, I created a page called /sandbox1 at:
l k:Lulu_of_the_Lotus-Eaters/sandbox1&oldid=25078799
/sandbox2. A recent edit is:
l k:Lulu_of_the_Lotus-Eaters/sandbox2&oldid=25078899
http://en.wikipedia.org/w/index.php?title=User_ta
I performed some edits, then moved the page to
http://en.wikipedia.org/w/index.php?title=User_ta
The old URL continues to work happily, and refers to the same snapshot of content as it did before the move.
maxwell's demon is incorrect. For example I started the article "Fetal hormones and sexual orientation". Disclaimer: it's not a very good article, and I only copied the text proposed by someone else in a discussion on a different article (but that's how Wikipedia works its magic, start with *something* and let it improve from there). The page I first created is:
o rmones_and_sexual_orientation&oldid=21401415
l _hormones_and_sexual_orientation&oldid=23504212
http://en.wikipedia.org/w/index.php?title=Fetal_h
Over time, editors felt that the title "Prenatal hormones and sexual orientation" was a better one (and in the meanwhile, the article has got at least a bit less bad). A recent edit can be found at:
http://en.wikipedia.org/w/index.php?title=Prenata
However, if you just look for the old title, you will be redirected to the new title. But the old title still retains its independent edit history, and the ability to directly URL reference each old revision.
Ah... it just makes me miss the days when I had slacker college students :-).
Certainly I entirely concur that an encyclopedia is not a primary source for academic research, especially not at a college level. I was just trying to help with the narrow technical question of citing specific known texts (as one might with a specific edition of a printed work).
This has a rather painfully simple technical solution. Just demand that your students cite specific version of an article by URL. For example, the *current* article on Slashdot at Wikipedia is:
t &oldid=24939679
http://en.wikipedia.org/w/index.php?title=Slashdo
Someone will probably modify the Slashdot article itself in the next five minutes. But the URL I give will retain unchanged forever. Of course, if the edit someone does in five minutes improves the information, my snapshot URL won't present the improvement. But then the future improvement isn't what your students read either.
I happen to know that BWJones really is a genuine expert in some specialized medical areas. I'm pretty sure he's not just talking about arcana of his favorite musical group. So in a sense, he really should be shown a certain deference in edits. Not an unlimited deference, of course: what goes in an encyclopedia like Wikipedia is not just the "best knowledge" but also the best writing, and content that is neutral point-of-view, verifiable, reflects consensus, and so on. "Experts" are not automatically the best deciders of all these things.
Still, there's no doubt that people who know a little are overly hasty to change the edits of people who know a lot. In my experience, as a trend this gets ironed out. Editors who consistently provide evidence and show expertise are, over time, granted proper deference (but not a fixed-in-stone sort). And making a clear case on Wikipedia talk pages before introducing substantial or contentious material tends to make the addition happen more smoothly.
However, more traditional "hired gun" authority suffers from an equally pernicious problem, probably worse even. While BWJones is an expert is some things, he may not be the *particular* expert a traditional encyclopedia hires. And specialized fields have their own contentious issues, and differing professional opinions at the margins of the topics. If someone with a competing theory to that BWJones holds is hired for a standard medical encyclopedia, BWJones gets ZERO say in which content goes in. Whoever is hired (and is indeed another expert in the relevant area), decides the whole of what goes there. Sure that one (or a couple) paid editor acts in good faith, and tries to be neutral. But it's almost impossible not to present your favorite theory as more likely than the competing approach. The anarchy of Wikipedia tends to smooth out that bias.
But ESR isn't at all a rabid proponent of Linux (heck, he disses the GPL nowadays), but rather a proponent of "the greatness of Eric Raymond". I confess Theo de Raadt comes closer to the characterization I said was absent. But even there it's different: Murphy, O'Gara, and Dvorak are such foolish figures because they don't know a thing about actual programming or technology; whatever excesses de Raadt might engage in, he *is* the principle author of an excellent operating system.
Murphy:Sun == Dvorak:Microsoft == O'Gara:SCOX
Interestingly, neither Linux, *BSD, nor Apple/OSX seem to have doltish lapdogs in the same way. Sure, proponents of Linux, BSD and OSX sometimes say foolish things. But there's no single figure from those communities who can be so relied upon to unthinking chant trollish drivel.
Ummm... where can I get QuickView Plus and Windows 95, exactly? :-) (actually, I think I *tried* QV+ about 5 years ago, and it didn't support that old format, even that far back). Of course, if you're volunteering, email me.
100 years is a LONG time! I'm sure OASIS won't introduce something really incompatible in 2010; and many tools will almost certainly support OpenDocument v.1 in 2010. But 2105 is... well, what were autos, trains, electrical outlets, light bulbs, telephones, sound recordings, typewriters, etc. like in 1905? You can recognize all those things over the century, but a lot of details have changed.
I should have written "OpenDocument" to be more exact. But the idea is the same. Quite possibly, 100 years from now only OpenDocument v.17+ will be supported in widespread existing tools. Version 1 isn't necessarily the end point, and it may be the features become incompatible over a number of versions. But WhiteWorlf666 is exactly correct that since the v.1 *standard* will still be available, there's nothing stopping future programmers from writing an import filter for current tools (which may be really easy by then, using some kind of high-level XML interface that might exist then).
Fetchmail is/was a fairly useful program. And his two famous essays, "Cathedral and Bazaar" and "Homesteading the Noosphere" are reasonably interesting (not entirely accurate, but also not entirely without some reasonable analysis). But 1998 was the last time ESR did ANYTHING even remotely of public interest.
In a recent interview, Eric Raymond shocked many Free Software developers. Interviewed developers commented "What? Is he still around? Now there is someone you don't think about anymore." Raymond's contentious statements were uniformly viewed as desperate cries for attention by a once notable software developer. A sympathetic developer commented, "I certainly hope ESR gets the professional counciling he so obviously needs; he just needs to learn that you can lead a productive life outside the limelight. I know of a good support group for people like him and Dvorak."
I have some old papers I wrote, back in the MS-Word/DOS days. Yeah, I know I'm old :-) (and didn't know better yet, in the late 1980s); but there are people and documents that matter which are even older. I'd genuinely like to make them available along with various other writing of mine. Unfortunately, absolutely nothing appears able to read these old formats, especially not anything made by Microsoft.
On the other hand, when I wrote papers in WordPerfect 4.2 or so, not much later, those formats are still happily read by recent WordPerfect versions. Not to say WP is a perfectly open format either, but at least it's not quite so breaky as MS formats are.
Fortunately for me, the old Word/DOS format is like all those wordprocessor formats of the 1980s: mostly ASCII text, with a smattering of control codes in it. Ultimately, with a little bit more work than should be needeed, I can extract the text and reformat the document. Had this been Word97 or something, I'd just have a steaming blob of undocumented binary data. Which exactly what all MS formats will be 20 years from now. I'm not so happy that WordPerfect moved to a more binary-ish format around WP6/7; they were also quite discernable using 'less' in the WP5.1/2 timeframe. They've retained better compatibility (even filters for the older versions to read the newer thing). But "readable in 'less'" is an important quality.
Luckily OpenOffice format is now, and will always remain "readable in less". XML is like that. Not that it's quite effortlessly readable; but I don't see XML parsers as going away, so extracting data will still be quite reasonable in 50 years, even if the format itself is no longer implemented in current products. Not effortless, but not outrageously difficult.
Can't we finally have a /. standard of NOT listing articles by Dvorak on the front page (no I didn't RTFA, not do I care to).
DO NOT FEED THE TROLLS!
Definitely pompous ass material, but I think we knew that.
Maybe an even better analogy would be getting spam for penis enlargement, then writing an angry letter in return boasting of how large your penis already is without their product (and how dare they suppose otherwise).
The MS recruitment letter is just that: spam. Other than the name at the top, they probably sent out 5000 of them the same day. And no, MS didn't check each other *potential candidate's* biography either, just that their resume (or some other document) said they were a computer programmer or similar.
I'm hardly pro-spam, but when I get it, I don't think I'm being *personally* insulted either.
Well, a bat (the mammal kind) isn't a rodent. But other than that, amen. :-)
Apparently IQ is of no use in finding the actual meaning of "begs the question"!
God do I hate that misuse.
From http://en.wikipedia.org/wiki/Beg_the_question:
"Begging the question is the term for a type of fallacy occuring in deductive reasoning in which the proposition to be proved is assumed implicitly or explicitly in one of the premises"
In a legal contract, changing the right (or wrong) 5% of the language can make a REALLY BIG difference.
Sun really has contributed some significant things to FOSS. OpenOffice is hugely important, for example. And not as code per se, but as standards, things like NFS are wonderful.
1 40308780
Probably much of the problem comes from the erratic behavior of Sun itself. Johnathan Schwartz in particular has, on alternating weeks, either sucked up to the FOSS community or belligerently insulted FOSS. And their patent issues are still looming, and not at all FOSS friendly, e.g.:
http://www.groklaw.net/article.php?story=20050202
Plus the whole matter of Sun being one of the major semi-backdoor funder of the SCO lawsuits really hasn't won them any love among FOSS users. A few million dollars spent on trying to kill Linux through deception and spurious lawsuits doesn't spread the love.