Without wanting to beat the "security through obscurity" dead horse any further, I don't see what's to stop the Nurk developers from doing exactly what Google does: changing the algorithm every month or so, so that pages which are "nurkbombed" in one release don't count as high in the next one. Admittedly, nurkbombers might have a slight time advantage over googlebombers (since they'd have access to the source), but the principle that the algorithm evolves based on attempts to exploit it is independent of whether the algorithm is open or closed-source.
I understand what you're trying to say, but I don't think you used a very good example to back up your point. The article you cite is written by a completely anonymous journalist (no byline). The reporter of the article cites unnamed "Palestinian sources" who are in turn talking about unnamed "Fatah agents." And the article as a whole turns out to be entirely impossible to confirm: the camera crews who were allegedly threatend are never named, nor are the agencies they belong to. Indeed, it is stated that the journalists "refused to publicly acknowledge the pressure," so even if they had been identified there would be no easy way to back the story up.
All of this would be one thing if the reporting had come from the Wall Street Journal or a similarly well-known source, but "worldtribune.com?" It's an internet-only newspaper...
I agree that we ought to be extremely wary of what we read on anonymous internet websites. But nobody reads slashdot for factual reporting, they read it for opinions. We should expect a much higher standard of reporting in our news sources, and for me the article you cited doesn't cut it.
Yes, I especially enjoyed this bit near the top, where Gates is talking about using time-shared equipment back in junior high:
You would type the programs off-line on this yellow paper tape and then put it into the tape reader, dial up the computer, and very quickly feedin [sic] the paper tape and run your program. They charged you not only for the connect time, but also for storage units and CPU time. So, if you had a program that had a loop in it of some type you could spend a lot of money very quickly. And so we went through the money that the Mothers Club had given very rapidly. It was a little awkward for the teachers, because it was just students sitting there and zoom! -- the money was gone.
...is the obvious fact that demoting Pluto to a mere Kuiper Belt object runs the grave risk of offending its namesake, the dread Lord of the Underworld. I have yet to see any respectable scientist propose the obvious solution to this problem: to ameliorate Pluto, several fattened heifers should be slaughered in sacrifice on the steps of the Rose Center. I'm shocked, shocked that these scientists would risk angering the god; the last time someone (namely, Persephone) did so, Demeter plunged the Earth into famine, and ulimately the seasons were split in two.
Wow, this is great. It could be hooked up to emacs, and used whenever somebody wrote the words <FLASH> or window.open, the user would experience a sensation of nausea. This would simulate the end-user's experience quite nicely...
cat's interface is simplicity itself. Simply list the files, and voila, concatenation. Many (most non-GNU?) unices don't even have those confusing command-line options.
I was going to nominate echo, but I decided against it because most shells force you to escape apostrophes and the like. Arguably this is a problem with shells, not echo, but nonetheless, the user experience is less than perfect. Not so with cat.
I wish I'd seen this discussion sooner, since I recently had a brief email from a Broadcast.com employee about this very matter. Of course, there's no telling whether the employee was either authorized to speak for the company or correctly informed about the situation...
Anyhow, here it is (editied slightly but with the content unchanged). Perhaps this will shed a little light on the validity of the rumors Slashdot reported on.
From: xxxx@broadcast.com To: [myself] Subject: RE: BCST0000250179 Broadcast.com Help Submission Form Date: Tue, 7 Dec 1999 11:36:31 -0600
Good afternoon.
I just found out that we are planning on using both Real and WMP formats in the future. Thanks for listening!
-----Original Message----- From: [myself] Sent: Friday, December 03, 1999 2:01 AM To: xxxx@broadcast.com Subject: RE: BCST0000250179 Broadcast.com Help Submission Form
Thanks for the reply; sorry to hear about the philosophy books going offline. Whilst I've got your attention, could you tell me whether you plan to offer more selections in RealPlayer format? As a Linux user, I can't hear any WinMP files at all.
This would seem to indicate that they will not be dumping RealAudio altogether. As far as I'm concerned, they should offer their audio in streaming MP3, as there exist quite good players for this format for all (useful) platforms I know of offhand.
It's worth noting that there are already a lot of selections on Broadcast.com that are already available only in WinMP format. Check the AudioBooks section if you don't believe me.
I'm with you there. I guess my point is that they should store the books as XML, and produce plain-text versions from it for public consumption. I certainly agree with you (and PG) that plain-text is a necessary format to have available (for end users). The conversion from XML to HTML (or, say, PDF) would be a lot easier than converting from text, though.
I couldn't agree more with your and jaso's statements. The point is not to mark up the books so that they look pretty on the web, the point is to add the structural elements that plain-text doesn't preserve back into them. HTML is a bad idea (as I've argued above somewhere), but it is trivial to produce it for presentation on the web if need be. Likewise, from an XML source, it is trivial to produce a flat text file. Going from a flat text file to a format that includes structural information, though, is decidedly non-trivial.
Don't get me wrong, I hate M$ just as much as the next slashdotter, but converting the PG books to the openebook format is a good idea. The fact that M$ will doubtless try to co-opt this format is beside the point; the point is that it is a published standard, and more importantly that it is XML-based.
It really doesn't matter whether the PG texts can be plugged directly into a web browser or not, the important thing is to make the texts into structured texts. Really it doesn't matter what XML format is used, but since one already exists for books, it might as well be the format that is used, if it is technically sufficient.
From an XML format, it is trivial to produce the plain-ASCII format that PG seems to be so fond of (one might even say irrationally fond of). It is also trivial to produce a web-ready HTML document for online publishing, one that works with all browsers. Heck, we could even make it work for lynx, for people without a text editor...:-) Storing the documents in HTML is a bad idea, for a few reasons:
HTML is hard to parse. Ok, it isn't that hard these days, but it's still harder than XML. Besides, formatting should not be stored with the text (see below).
Who knows where HTML will be in 10 years? Odds are that the spec will change considerably as browsers and the needs of the web evolve. I think it's a pretty good bet that XML itself will stay relatively stable, though.
PG was right not to use particular markup schemes in their documents, chiefly because of the volatility of markup-scheme systems. What an XML scheme does is to store the structure of documents. PG's flat-text format clearly loses that structure.
The thing is, when someone types/scans/edits a work and submits it to PG, they ought to have a way of specifying the things that are lost in the flat-file format: footnotes, chapter and section headings, bibliographies, etc. Once this happens, the works can be linked to with more granularity than as ftp://.../book.txt. This is critical (as several posters previously have stated) if books are to become full citizens of the noosphere.
Hmm, well when I write a sentence like that last one I know it's time for bed. I hope my arguments are clear. Re openebooks, my main point is that M$'s ownership of the standard is irrelevant as far as its usefulness is concerned. If anyone is interested in discussing these issues and maybe taking a look at what a good XML format would be, please email me.
Of course, in Athens only non-slave men could vote. Obviously I'm not saying that women voting is what's wrong modern democracy, just that Athens wasn't as great for some as for others (and thus wasn't really a democracy in the most literate sense of the word).
The big problem is campaign finance reform, but underlying that is a much more complicated quetion of how people get their information about candidates and use it to make their choices about who to vote for.
Slashdot Mirror
David Foster Wallace wrote an excellent article for Harper's a while back which described the prescriptivist / descriptivist split in more detail.
Without wanting to beat the "security through obscurity" dead horse any further, I don't see what's to stop the Nurk developers from doing exactly what Google does: changing the algorithm every month or so, so that pages which are "nurkbombed" in one release don't count as high in the next one. Admittedly, nurkbombers might have a slight time advantage over googlebombers (since they'd have access to the source), but the principle that the algorithm evolves based on attempts to exploit it is independent of whether the algorithm is open or closed-source.
Ha, good ones. How about...
More Stupid Database Exploits
I understand what you're trying to say, but I don't think you used a very good example to back up your point. The article you cite is written by a completely anonymous journalist (no byline). The reporter of the article cites unnamed "Palestinian sources" who are in turn talking about unnamed "Fatah agents." And the article as a whole turns out to be entirely impossible to confirm: the camera crews who were allegedly threatend are never named, nor are the agencies they belong to. Indeed, it is stated that the journalists "refused to publicly acknowledge the pressure," so even if they had been identified there would be no easy way to back the story up.
All of this would be one thing if the reporting had come from the Wall Street Journal or a similarly well-known source, but "worldtribune.com?" It's an internet-only newspaper...
I agree that we ought to be extremely wary of what we read on anonymous internet websites. But nobody reads slashdot for factual reporting, they read it for opinions. We should expect a much higher standard of reporting in our news sources, and for me the article you cited doesn't cut it.
Yes, I especially enjoyed this bit near the top, where Gates is talking about using time-shared equipment back in junior high:
Rented software, anyone?...is the obvious fact that demoting Pluto to a mere Kuiper Belt object runs the grave risk of offending its namesake, the dread Lord of the Underworld. I have yet to see any respectable scientist propose the obvious solution to this problem: to ameliorate Pluto, several fattened heifers should be slaughered in sacrifice on the steps of the Rose Center. I'm shocked, shocked that these scientists would risk angering the god; the last time someone (namely, Persephone) did so, Demeter plunged the Earth into famine, and ulimately the seasons were split in two.
Wow, this is great. It could be hooked up to emacs, and used whenever somebody wrote the words <FLASH> or window.open, the user would experience a sensation of nausea. This would simulate the end-user's experience quite nicely...
cat's interface is simplicity itself. Simply list the files, and voila, concatenation. Many (most non-GNU?) unices don't even have those confusing command-line options.
I was going to nominate echo, but I decided against it because most shells force you to escape apostrophes and the like. Arguably this is a problem with shells, not echo, but nonetheless, the user experience is less than perfect. Not so with cat.
speck
I wish I'd seen this discussion sooner, since I recently had a brief email from a Broadcast.com employee about this very matter. Of course, there's no telling whether the employee was either authorized to speak for the company or correctly informed about the situation...
Anyhow, here it is (editied slightly but with the content unchanged). Perhaps this will shed a little light on the validity of the rumors Slashdot reported on.
From: xxxx@broadcast.comTo: [myself]
Subject: RE: BCST0000250179 Broadcast.com Help Submission Form
Date: Tue, 7 Dec 1999 11:36:31 -0600
Good afternoon.
I just found out that we are planning on using both Real and WMP formats in the future. Thanks for listening!
-----Original Message-----
From: [myself]
Sent: Friday, December 03, 1999 2:01 AM
To: xxxx@broadcast.com
Subject: RE: BCST0000250179 Broadcast.com Help Submission Form
Thanks for the reply; sorry to hear about the philosophy books going offline. Whilst I've got your attention, could you tell me whether you plan to offer more selections in RealPlayer format? As a Linux user, I can't hear any WinMP files at all.
This would seem to indicate that they will not be dumping RealAudio altogether. As far as I'm concerned, they should offer their audio in streaming MP3, as there exist quite good players for this format for all (useful) platforms I know of offhand.
It's worth noting that there are already a lot of selections on Broadcast.com that are already available only in WinMP format. Check the AudioBooks section if you don't believe me.
speck
I'm with you there. I guess my point is that they should store the books as XML, and produce plain-text versions from it for public consumption. I certainly agree with you (and PG) that plain-text is a necessary format to have available (for end users). The conversion from XML to HTML (or, say, PDF) would be a lot easier than converting from text, though.
I couldn't agree more with your and jaso's statements. The point is not to mark up the books so that they look pretty on the web, the point is to add the structural elements that plain-text doesn't preserve back into them. HTML is a bad idea (as I've argued above somewhere), but it is trivial to produce it for presentation on the web if need be. Likewise, from an XML source, it is trivial to produce a flat text file. Going from a flat text file to a format that includes structural information, though, is decidedly non-trivial.
Don't get me wrong, I hate M$ just as much as the next slashdotter, but converting the PG books to the openebook format is a good idea. The fact that M$ will doubtless try to co-opt this format is beside the point; the point is that it is a published standard, and more importantly that it is XML-based.
It really doesn't matter whether the PG texts can be plugged directly into a web browser or not, the important thing is to make the texts into structured texts. Really it doesn't matter what XML format is used, but since one already exists for books, it might as well be the format that is used, if it is technically sufficient.
From an XML format, it is trivial to produce the plain-ASCII format that PG seems to be so fond of (one might even say irrationally fond of). It is also trivial to produce a web-ready HTML document for online publishing, one that works with all browsers. Heck, we could even make it work for lynx, for people without a text editor... :-) Storing the documents in HTML is a bad idea, for a few reasons:
The thing is, when someone types/scans/edits a work and submits it to PG, they ought to have a way of specifying the things that are lost in the flat-file format: footnotes, chapter and section headings, bibliographies, etc. Once this happens, the works can be linked to with more granularity than as ftp://.../book.txt. This is critical (as several posters previously have stated) if books are to become full citizens of the noosphere.
Hmm, well when I write a sentence like that last one I know it's time for bed. I hope my arguments are clear. Re openebooks, my main point is that M$'s ownership of the standard is irrelevant as far as its usefulness is concerned. If anyone is interested in discussing these issues and maybe taking a look at what a good XML format would be, please email me.
Of course, in Athens only non-slave men could vote. Obviously I'm not saying that women voting is what's wrong modern democracy, just that Athens wasn't as great for some as for others (and thus wasn't really a democracy in the most literate sense of the word).
The big problem is campaign finance reform, but underlying that is a much more complicated quetion of how people get their information about candidates and use it to make their choices about who to vote for.