> MS Office is definitely better than OpenOffice...
MS Office is only better than OpenOffice if you grew up using MS Office and are so used to it that you take a lot of its quirks for granted and consider them to be normal. It's not as bad in this regard as Word Perfect has historically been, but on the other hand Word Perfect is apparently trying to become (gradually) more like normal software; whereas, MSO is apparently trying to become more gratuitously weird and unusable with each new version. Schenanighans like taking away the menus in the latest version (as if menu *items* that hide themselves if you don't use them every day weren't user-hostile enough) are *not* my idea of how to make software better.
But some proprietary software *is* better than the available open-source alternatives. The example I usually give is Pegasus Mail, but Mathematica is certainly in this category.
And yeah, going the other way, Apache is one of the best examples, because it's not just a little better than the proprietary alternatives; it's a *WHOLE LOT* better.
Yeah, well, in my world, the phrase "phone call" usually goes into the same sentence with words like "interruption", "annoyance", and "accursed", and phrases such as "modern work environment", "noise pollution", and "Scott Adams".
Somewhere, in a completely unrelated sentence, which probably takes place in an idyllic setting, one might find the word "perfect", sitting alongside such words as "quiet" and "peaceful" and "relaxing". In this context, one supposes that it's probably snowing gently, a dry powdery snow. Something Bach wrote for violins is playing softly in the background, and I'm reading a good book, or taking a hot bath, or perhaps both.
Trying to combine the two concepts is an exercise in surrealism, as far as I'm concerned.
> if implemented commercially could keep lead out of landfills and the ecosystem
Okay, I can sort of see how using less lead in products would keep it out of landfills. That makes sense. But I'm a bit confused on the subject of keeping lead out of the environment. Where is the lead coming from that they're talking about *not* putting in products? Are manufacturers currently transmuting gold into lead or something, and I just don't know about it? Because I was under the impression that it came, ultimately, *from* the environment, in which case the total amount of lead in the world is not changing. It's just a question of exactly where it's located: in ores, in the stuff on the store shelves, in the stuff in homes and businesses, in the landfill, those are all just locations. It's still the same amount of lead.
Talking of keeping stuff out of the environment makes sense if the stuff in question is man-made, like plastics. But lead is an element. You might as well talk about keeping iron out of the environment, or nitrogen for that matter. It doesn't make sense.
> Out of interest, what do you consider the smallest possible user > base that any concession should be made with regard to support?
I don't know that I could put a number to that, but I sure wouldn't expend any great effort supporting Itanium. The only people who moved to Itanium in the first place were people who absolutely had to be on the latest up-and-coming platform (which Intel *assured* them Itanium was going to be), and almost all of that userbase has long since moved on now. I think the Vax *still* has a larger userbase than Itanium had at its peak, to say nothing of now.
Sparc is more worthy of consideration. I realize it's been a platform in decline since before Itanium was announced, but it's declining from a position of significance, so its userbase is holding onto something that was once very much a major platform. Also, the Sparc userbase has always tended more toward the "Upgrade? Never!" end of the scale. I mean, if it took Sun fifteen years to get all their users migrated from the old SunOS to Solaris, which was *supposedly* a straightforward upgrade, how long do you think it's going to take to get them all moved over to x86 and/or x64, which is clearly a whole different platform?
All of that is to say, putting Sparc and Itanium in the same category is *weird* to my way of thinking. They're very different cases.
> Is an open source project obliged to provide support for its users?
I don't know about legal obligation, but ethically I would say you should make a serious effort to provide the support for as long as you said you were going to provide it for. So for instance if you say that a certain release is "supported until January of 2009", you should try very hard to support it until then.
> Attempts to remove yourself from the mailing list may only result in more mailings from the site of ill repute
Will wonders never cease? I suppose next you're going to tell me that the luck hasn't really traveled around the world seven times?
The headline makes it sound like classmates.com is a real, more-or-less legitimate website that happens to have some shady practices. I was under the impression it was nothing more than a source of spam, just like bluemountain.com and that African outfit with the all-uppercase keyboards.
> KOHA (along with Dynix, Sirsi, Gaylord, VTLS, and a few others) provides [ILS software]
Actually, Gaylord doesn't do that anymore. All their former ILS stuff is handled by PLS now. Gaylord continues to exist and to sell other stuff (like book covers and book carts and whatnot), but they no longer have anything directly to do with automation systems.
> The thing is, these catalog systems pretty much only accept MARC-formatted records.
Of course. We're talking about library catalogs here, right? What *else* would they accept?
*Is* there another standard format for bibliographic records? If so, I've never heard of it.
> The MARC format is kind of obscure,
It's certainly not obscure in the sense of being unusual. It is the *only* format for bibliographic records, so far as I am aware. As you point out, *all* library software, regardless of vendor, uses it exclusively, at least for input and output, and often for internal storage as well.
> and it's nothing we want to generate ourselves
Now, there I don't blame you. Despite the exclusive ubiquitousness of MARC, it is a truly depraved and horrible format to have to work with directly. It was developed by librarians who didn't understand computer systems very well, in an era when computer systems filled entire rooms and typically used punched cards, so it's quite painful to have to work with.
Nonetheless, bib records in any other format would be worthless because no software would be able to read them.
> so we provide CSV data to OCLC and they convert it to MARC format for us. The amazing part > of the racket they're running is that we have to *pay* OCLC to make these records for us
You thought they'd do it for free, out of the goodness of their hearts?
> and then they turn around and require *another* payment from anyone who wants to use the records.
Ah, yes, I see. It's the charge-you-both-ways aspect of things that bothers you.
Well, you could hire someone *else*, other than OCLC, to make the records for you, someone who would allow you to then distribute them free of charge. It would have to be someone who has experience doing library cataloging, of course, but that just sort of goes with the territory, since it is, basically, library cataloging that you want to have done for you. You don't have to use OCLC. They're not the only people who know how to do cataloging.
So far, to my knowledge, there hasn't been any noticeable decline in library usage. Most library directors track their circulation statistics religiously and expect them to go up every year without fail.
Library usage *details* have shifted over the years, in terms of what exactly people get out of libraries... Reference materials usage, for instance, is *way* down (unless you count the internet as a reference material, which many libraries do... this seems a bit disingenuous to me, not because the internet can't be used as a reference material but because in practice approximately 0.0000000437% of library patrons use it that way... but it's a fairly common position for libraries to hold on paper, that the internet is primarily a reference source or service). Biographies are down too. But other things are way up, not least audio-visual materials. (DVD circulation has absolutely *exploded* in the last five years, you simply would not believe. I don't know how people find the time to watch so much video content, unless most of the population either never sleeps or doesn't work for a living.)
But OCLC is not synonymous with libraries, and libraries will not rise or fall with OCLC. Many libraries don't even use their service.
> The way I see it, it's definitely in OCLC's best interest to embrace > the internet and help libraries gain some popularity.
I don't think libraries need OCLC's help with popularity. That would be kind of like Singapore getting financial advice from Bangladesh.
> If this keeps up, it looks like they're liable to be replaced by > something smaller, faster, and free-er that uses the Internet.
It's called Z39.50. It's not a centralized one-big-database source of records. It's a protocol (err, or a suite of protocols; Z39.50 itself is technically one protocol, but it's often used in conjunction with SIP2 and NCIP...) that libraries use to conveniently share catalog records with other libraries. You can have multiple Z39.50 sources, so you don't need One Big Source For All Records; you just need a number of small and medium-sized sources that collectively have a lot of records. Mutual sharing agreements are fairly common, and some library systems just freely share with everyone unconditionally.
> It's the fact that OCLC wants to be the only records database out there
They can want that as much as they want, it ain't gonna happen. There are too many other sources.
Granted, OCLC is probably the largest single centralized source. But Z39.50 completely obviates the advantages of centralization anyway, because your cataloging software, if your ILS is even vaguely modern, automatically queries your various sources in turn until it either finds the record or runs off the end of the list. Actually, some of them query all your sources in parallel and aggregate the results into a list of records you can choose from. The one we use does that.
So you just put twenty or thirty large library systems with open-access catalogs on your list of Z39.50 sources, and Bob is your uncle.
> [an encyclopedia] owns the "articles," true, but it can't prevent the "facts" from being transfered. > > In other words: no, OCLC doesn't own the books, or the facts about them, but they do own the database. > Sort of true, the copyright in this case *only* applies to when original work incorporated > in to the collection of facts, making the "collection" a copyrighted entity.
In order to clear up your confusion here, I'd have to explain to you what a MARC record is, but believe me, you do *NOT* want to know about MARC records if you can possibly avoid it. (I certainly wish I could erase the knowledge from *my* memory.) So just trust me: OCLC, although they're being pretty overbearing, are nonetheless probably within their rights here.
But the article summary is blowing the ramifications all out of proportion, because most libraries don't use the OCLC service anyway (though many do), so it's not like they're the only place to get bib records. Far from it. A lot of libraries just use mutual Z39.50 catalog sharing agreements with other library systems and maybe get the occasional record from the LOC catalog. Also, a lot of book vendors these days will send you MARC records along with your order. Baker & Taylor, for instance, provides this service. For them, it's a complementary service that helps them sell books, so they don't need to make money on the records the way OCLC does.
Yeah, I was previously unaware that OCLC was non-profit, and I've worked in a library for 8+ years. (Granted, we don't use any of their services in the library where I work. But I was very much aware of their existence and what some of their services were, and very much unaware that they were non-profit. Certainly we generally think of them as a vendor.)
It could be worse. They could be all up in arms about the whole "Library 2.0" thing again, and how evil it is to restrict flagrantly inappropriate behaviors (such as skateboarding) in the library.
> The contract apparently says, or the data provider wants people to believe > that it says, that libs can use the data themselves, but cannot transfer it.
My understanding of how this works with the OCLC MARC record service is that you can use the records in your own library catalog, which means you can show the records to people who are looking up books on your catalog, but you can't allow others (e.g., other libraries) to copy them into another database (e.g., another library catalog) that's also going to be publicly available. Something like a home user's browser cache would not get you into any trouble, but mutual Z39.50 catalog-sharing agreements with other library systems (a very common practice, at least as common as using the OCLC records) is verboten if you use the OCLC MARC record service.
That's one reason a lot of libraries don't use their service. Cost is another. The widespread availability of free records from other sources (including the LOC, book vendors like Baker & Taylor, many of whom can just bundle the records in with your book orders these days, and the aforementioned mutual sharing agreements with other libraries) is a third reason.
> The main source of the bibliographic records that are carried in > library databases is a non-profit organization called OCLC.
That's absolutely not true. Many (possibly most) libraries don't use OCLC MARC records at *all*, and even most of the libraries that do only use them when they can't find a MARC record somewhere else (e.g., from the LOC) for free. I don't have any formal statistics to cite for this, but I've been on several library-related mailing lists for several years, as part of my job, and followed numerous conversation threads about OCLC records, so I'm not just guessing out of ignorance, either.
They *are* a fairly major and widely used service, but they're nothing like the majority/monopoly provider that the article summary implies.
> Over the weekend OCLC 'leaked' its new policy that claims contractual > rights in the subsequent uses of the data, uses such as downloading > book information into Zotero or other bibliographic software. The > policy explicitly forbids any use that would compete with OCLC.
This is not a very big change, really, in the scheme of things. They've always considered the MARC records they provide to be copyrighted and all rights reserved except those specifically granted. For instance, any library that uses their records cannot then make the resulting catalog generally available via Z39.50/NCIP for other libraries to freely borrow from, because that would violate the OCLC copyright. Since mutual-catalog-sharing agreements are a *major* (perhaps *the* major) source of bib records for a many libraries, especially libraries that use a modern ILS, this is a fairly onerous restriction.
As I mentioned earlier, a lot of libraries don't use OCLC records, partly because of these issues, and partly because of the cost.
> What about release critical bugs? Currently there are 173 of them:
Most of those bugs are not in the installer. If you read the summary (not even the actual article, just the summary) carefully, you'll note that this is a release candidate of the *installer*, not of the whole distro. I would imagine the distribution as a whole is probably still on schedule for the originally promised timeframe of "when it's ready".
Yes. I'm pretty sure every other regular reader of slashdot was already very much aware, before seeing this story, that Lenny is the codename for the upcoming release of Debian. In fact, given that the words "Lenny" and especially "Debian" are a good deal more important to most of us, and are words we see and use much more often, than the words "Lesbian" and "Denny", I imagine that if the article had been misprinted as "Lesbian Denny", a lot of us probably would have read it as "Debian Lenny" without noticing the typo.
Pagers cost too much for the abilities they give you. They cost as much or more than a cellphone (well, depending on how much you use the cellphone, but it only costs more if you use it for things the pager won't do), but you can't make outgoing calls, can't receive very much information incoming either (typically just a short numeric code), can't store information (like phone numbers) on them, can't open them up and use them as a light when you need to see to put your key in the keyhole, and so on and so forth.
So most of the people who used to use pages have gone over to other solutions, mainly cellphones.
If you must be reachable all the time, then these days you can just get a cellphone that supports per-caller ringtones and set a custom (possibly loud and annoying) one for when certain numbers (e.g., work) call you. If you don't want to be annoyed by other stuff, you can set everybody else's ringtone to a quiet hum, or even a famous John Cage number, and then check your messages when you feel like it.
There *are* people who still use pagers. The local obstetrician where I live uses one, for instance. So I know you can still get them. But are they worth it, that's another question.
Personally I'm glad that I don't need to be reachable absolutely all the time, but that's a separate issue.
> if they have a list of all the kiddie porn sites on the web, why don't they just go after the site owners?
Kiddie porn is like warez and fraudulent spam-marketed junk products (e.g., online prescription drugstores without a prescription requirement). The sites migrate so frequently, by the time you call the ISP to complain the site's already moved on somewhere else. Most of it's hosted without permission from the owner or operator of the system it's hosted on. It's like trying to stop spam by calling the spammer's ISP and asking them to shut down the mail server.
Not that filtering is going to work a whole lot better.
> Of course, the police don't tell you about the Spectre IV, which has > a radar-detector-detector-detector detector circuit in them.
Meh. That can be defeated with a Spectre IV Detector, or, generically, a radar detector detector detector detector detector. If they want to detect you undetectably, they'll have to watch you on satellite video and check your speed that way. Of course, that only works if you're not using a cloaking device...
My personal favorite regex trick is the zero-width assertion. I'm particularly fond of zero-width negative lookbehind assertions. Backreferences are also cool.
Slashdot Mirror
> MS Office is definitely better than OpenOffice...
MS Office is only better than OpenOffice if you grew up using MS Office and are so used to it that you take a lot of its quirks for granted and consider them to be normal. It's not as bad in this regard as Word Perfect has historically been, but on the other hand Word Perfect is apparently trying to become (gradually) more like normal software; whereas, MSO is apparently trying to become more gratuitously weird and unusable with each new version. Schenanighans like taking away the menus in the latest version (as if menu *items* that hide themselves if you don't use them every day weren't user-hostile enough) are *not* my idea of how to make software better.
But some proprietary software *is* better than the available open-source alternatives. The example I usually give is Pegasus Mail, but Mathematica is certainly in this category.
And yeah, going the other way, Apache is one of the best examples, because it's not just a little better than the proprietary alternatives; it's a *WHOLE LOT* better.
Yeah, well, in my world, the phrase "phone call" usually goes into the same sentence with words like "interruption", "annoyance", and "accursed", and phrases such as "modern work environment", "noise pollution", and "Scott Adams".
Somewhere, in a completely unrelated sentence, which probably takes place in an idyllic setting, one might find the word "perfect", sitting alongside such words as "quiet" and "peaceful" and "relaxing". In this context, one supposes that it's probably snowing gently, a dry powdery snow. Something Bach wrote for violins is playing softly in the background, and I'm reading a good book, or taking a hot bath, or perhaps both.
Trying to combine the two concepts is an exercise in surrealism, as far as I'm concerned.
> if implemented commercially could keep lead out of landfills and the ecosystem
Okay, I can sort of see how using less lead in products would keep it out of landfills. That makes sense. But I'm a bit confused on the subject of keeping lead out of the environment. Where is the lead coming from that they're talking about *not* putting in products? Are manufacturers currently transmuting gold into lead or something, and I just don't know about it? Because I was under the impression that it came, ultimately, *from* the environment, in which case the total amount of lead in the world is not changing. It's just a question of exactly where it's located: in ores, in the stuff on the store shelves, in the stuff in homes and businesses, in the landfill, those are all just locations. It's still the same amount of lead.
Talking of keeping stuff out of the environment makes sense if the stuff in question is man-made, like plastics. But lead is an element. You might as well talk about keeping iron out of the environment, or nitrogen for that matter. It doesn't make sense.
> Out of interest, what do you consider the smallest possible user
> base that any concession should be made with regard to support?
I don't know that I could put a number to that, but I sure wouldn't expend any great effort supporting Itanium. The only people who moved to Itanium in the first place were people who absolutely had to be on the latest up-and-coming platform (which Intel *assured* them Itanium was going to be), and almost all of that userbase has long since moved on now. I think the Vax *still* has a larger userbase than Itanium had at its peak, to say nothing of now.
Sparc is more worthy of consideration. I realize it's been a platform in decline since before Itanium was announced, but it's declining from a position of significance, so its userbase is holding onto something that was once very much a major platform. Also, the Sparc userbase has always tended more toward the "Upgrade? Never!" end of the scale. I mean, if it took Sun fifteen years to get all their users migrated from the old SunOS to Solaris, which was *supposedly* a straightforward upgrade, how long do you think it's going to take to get them all moved over to x86 and/or x64, which is clearly a whole different platform?
All of that is to say, putting Sparc and Itanium in the same category is *weird* to my way of thinking. They're very different cases.
> Is an open source project obliged to provide support for its users?
I don't know about legal obligation, but ethically I would say you should make a serious effort to provide the support for as long as you said you were going to provide it for. So for instance if you say that a certain release is "supported until January of 2009", you should try very hard to support it until then.
> Attempts to remove yourself from the mailing list may only result in more mailings from the site of ill repute
Will wonders never cease? I suppose next you're going to tell me that the luck hasn't really traveled around the world seven times?
The headline makes it sound like classmates.com is a real, more-or-less legitimate website that happens to have some shady practices. I was under the impression it was nothing more than a source of spam, just like bluemountain.com and that African outfit with the all-uppercase keyboards.
> KOHA (along with Dynix, Sirsi, Gaylord, VTLS, and a few others) provides [ILS software]
Actually, Gaylord doesn't do that anymore. All their former ILS stuff is handled by PLS now. Gaylord continues to exist and to sell other stuff (like book covers and book carts and whatnot), but they no longer have anything directly to do with automation systems.
There is a Perl module for working with MARC records. I've not used it personally, so I can't comment on how easy it is to use.
> The thing is, these catalog systems pretty much only accept MARC-formatted records.
Of course. We're talking about library catalogs here, right? What *else* would they accept?
*Is* there another standard format for bibliographic records? If so, I've never heard of it.
> The MARC format is kind of obscure,
It's certainly not obscure in the sense of being unusual. It is the *only* format for bibliographic records, so far as I am aware. As you point out, *all* library software, regardless of vendor, uses it exclusively, at least for input and output, and often for internal storage as well.
> and it's nothing we want to generate ourselves
Now, there I don't blame you. Despite the exclusive ubiquitousness of MARC, it is a truly depraved and horrible format to have to work with directly. It was developed by librarians who didn't understand computer systems very well, in an era when computer systems filled entire rooms and typically used punched cards, so it's quite painful to have to work with.
Nonetheless, bib records in any other format would be worthless because no software would be able to read them.
> so we provide CSV data to OCLC and they convert it to MARC format for us. The amazing part
> of the racket they're running is that we have to *pay* OCLC to make these records for us
You thought they'd do it for free, out of the goodness of their hearts?
> and then they turn around and require *another* payment from anyone who wants to use the records.
Ah, yes, I see. It's the charge-you-both-ways aspect of things that bothers you.
Well, you could hire someone *else*, other than OCLC, to make the records for you, someone who would allow you to then distribute them free of charge. It would have to be someone who has experience doing library cataloging, of course, but that just sort of goes with the territory, since it is, basically, library cataloging that you want to have done for you. You don't have to use OCLC. They're not the only people who know how to do cataloging.
> I mean, how popular are libraries anymore?
So far, to my knowledge, there hasn't been any noticeable decline in library usage. Most library directors track their circulation statistics religiously and expect them to go up every year without fail.
Library usage *details* have shifted over the years, in terms of what exactly people get out of libraries... Reference materials usage, for instance, is *way* down (unless you count the internet as a reference material, which many libraries do... this seems a bit disingenuous to me, not because the internet can't be used as a reference material but because in practice approximately 0.0000000437% of library patrons use it that way... but it's a fairly common position for libraries to hold on paper, that the internet is primarily a reference source or service). Biographies are down too. But other things are way up, not least audio-visual materials. (DVD circulation has absolutely *exploded* in the last five years, you simply would not believe. I don't know how people find the time to watch so much video content, unless most of the population either never sleeps or doesn't work for a living.)
But OCLC is not synonymous with libraries, and libraries will not rise or fall with OCLC. Many libraries don't even use their service.
> The way I see it, it's definitely in OCLC's best interest to embrace
> the internet and help libraries gain some popularity.
I don't think libraries need OCLC's help with popularity. That would be kind of like Singapore getting financial advice from Bangladesh.
> If this keeps up, it looks like they're liable to be replaced by
> something smaller, faster, and free-er that uses the Internet.
It's called Z39.50. It's not a centralized one-big-database source of records. It's a protocol (err, or a suite of protocols; Z39.50 itself is technically one protocol, but it's often used in conjunction with SIP2 and NCIP...) that libraries use to conveniently share catalog records with other libraries. You can have multiple Z39.50 sources, so you don't need One Big Source For All Records; you just need a number of small and medium-sized sources that collectively have a lot of records. Mutual sharing agreements are fairly common, and some library systems just freely share with everyone unconditionally.
> It's the fact that OCLC wants to be the only records database out there
They can want that as much as they want, it ain't gonna happen. There are too many other sources.
Granted, OCLC is probably the largest single centralized source. But Z39.50 completely obviates the advantages of centralization anyway, because your cataloging software, if your ILS is even vaguely modern, automatically queries your various sources in turn until it either finds the record or runs off the end of the list. Actually, some of them query all your sources in parallel and aggregate the results into a list of records you can choose from. The one we use does that.
So you just put twenty or thirty large library systems with open-access catalogs on your list of Z39.50 sources, and Bob is your uncle.
> [an encyclopedia] owns the "articles," true, but it can't prevent the "facts" from being transfered.
> > In other words: no, OCLC doesn't own the books, or the facts about them, but they do own the database.
> Sort of true, the copyright in this case *only* applies to when original work incorporated
> in to the collection of facts, making the "collection" a copyrighted entity.
In order to clear up your confusion here, I'd have to explain to you what a MARC record is, but believe me, you do *NOT* want to know about MARC records if you can possibly avoid it. (I certainly wish I could erase the knowledge from *my* memory.) So just trust me: OCLC, although they're being pretty overbearing, are nonetheless probably within their rights here.
But the article summary is blowing the ramifications all out of proportion, because most libraries don't use the OCLC service anyway (though many do), so it's not like they're the only place to get bib records. Far from it. A lot of libraries just use mutual Z39.50 catalog sharing agreements with other library systems and maybe get the occasional record from the LOC catalog. Also, a lot of book vendors these days will send you MARC records along with your order. Baker & Taylor, for instance, provides this service. For them, it's a complementary service that helps them sell books, so they don't need to make money on the records the way OCLC does.
Yeah, I was previously unaware that OCLC was non-profit, and I've worked in a library for 8+ years. (Granted, we don't use any of their services in the library where I work. But I was very much aware of their existence and what some of their services were, and very much unaware that they were non-profit. Certainly we generally think of them as a vendor.)
Actually, the way I heard it, Neitzche is dead.
(No, I don't just mean physically. I mean it the way he meant it. He has far fewer remaining followers than God.)
It could be worse. They could be all up in arms about the whole "Library 2.0" thing again, and how evil it is to restrict flagrantly inappropriate behaviors (such as skateboarding) in the library.
> The contract apparently says, or the data provider wants people to believe
> that it says, that libs can use the data themselves, but cannot transfer it.
My understanding of how this works with the OCLC MARC record service is that you can use the records in your own library catalog, which means you can show the records to people who are looking up books on your catalog, but you can't allow others (e.g., other libraries) to copy them into another database (e.g., another library catalog) that's also going to be publicly available. Something like a home user's browser cache would not get you into any trouble, but mutual Z39.50 catalog-sharing agreements with other library systems (a very common practice, at least as common as using the OCLC records) is verboten if you use the OCLC MARC record service.
That's one reason a lot of libraries don't use their service. Cost is another. The widespread availability of free records from other sources (including the LOC, book vendors like Baker & Taylor, many of whom can just bundle the records in with your book orders these days, and the aforementioned mutual sharing agreements with other libraries) is a third reason.
> The main source of the bibliographic records that are carried in
> library databases is a non-profit organization called OCLC.
That's absolutely not true. Many (possibly most) libraries don't use OCLC MARC records at *all*, and even most of the libraries that do only use them when they can't find a MARC record somewhere else (e.g., from the LOC) for free. I don't have any formal statistics to cite for this, but I've been on several library-related mailing lists for several years, as part of my job, and followed numerous conversation threads about OCLC records, so I'm not just guessing out of ignorance, either.
They *are* a fairly major and widely used service, but they're nothing like the majority/monopoly provider that the article summary implies.
> Over the weekend OCLC 'leaked' its new policy that claims contractual
> rights in the subsequent uses of the data, uses such as downloading
> book information into Zotero or other bibliographic software. The
> policy explicitly forbids any use that would compete with OCLC.
This is not a very big change, really, in the scheme of things. They've always considered the MARC records they provide to be copyrighted and all rights reserved except those specifically granted. For instance, any library that uses their records cannot then make the resulting catalog generally available via Z39.50/NCIP for other libraries to freely borrow from, because that would violate the OCLC copyright. Since mutual-catalog-sharing agreements are a *major* (perhaps *the* major) source of bib records for a many libraries, especially libraries that use a modern ILS, this is a fairly onerous restriction.
As I mentioned earlier, a lot of libraries don't use OCLC records, partly because of these issues, and partly because of the cost.
> What about release critical bugs? Currently there are 173 of them:
Most of those bugs are not in the installer. If you read the summary (not even the actual article, just the summary) carefully, you'll note that this is a release candidate of the *installer*, not of the whole distro. I would imagine the distribution as a whole is probably still on schedule for the originally promised timeframe of "when it's ready".
> Ok, am I the only one who read Lesbian Denny?
Yes. I'm pretty sure every other regular reader of slashdot was already very much aware, before seeing this story, that Lenny is the codename for the upcoming release of Debian. In fact, given that the words "Lenny" and especially "Debian" are a good deal more important to most of us, and are words we see and use much more often, than the words "Lesbian" and "Denny", I imagine that if the article had been misprinted as "Lesbian Denny", a lot of us probably would have read it as "Debian Lenny" without noticing the typo.
> I should cease my attempts at speed reading...
HTH.HAND.
Pagers cost too much for the abilities they give you. They cost as much or more than a cellphone (well, depending on how much you use the cellphone, but it only costs more if you use it for things the pager won't do), but you can't make outgoing calls, can't receive very much information incoming either (typically just a short numeric code), can't store information (like phone numbers) on them, can't open them up and use them as a light when you need to see to put your key in the keyhole, and so on and so forth.
So most of the people who used to use pages have gone over to other solutions, mainly cellphones.
If you must be reachable all the time, then these days you can just get a cellphone that supports per-caller ringtones and set a custom (possibly loud and annoying) one for when certain numbers (e.g., work) call you. If you don't want to be annoyed by other stuff, you can set everybody else's ringtone to a quiet hum, or even a famous John Cage number, and then check your messages when you feel like it.
There *are* people who still use pagers. The local obstetrician where I live uses one, for instance. So I know you can still get them. But are they worth it, that's another question.
Personally I'm glad that I don't need to be reachable absolutely all the time, but that's a separate issue.
> if they have a list of all the kiddie porn sites on the web, why don't they just go after the site owners?
Kiddie porn is like warez and fraudulent spam-marketed junk products (e.g., online prescription drugstores without a prescription requirement). The sites migrate so frequently, by the time you call the ISP to complain the site's already moved on somewhere else. Most of it's hosted without permission from the owner or operator of the system it's hosted on. It's like trying to stop spam by calling the spammer's ISP and asking them to shut down the mail server.
Not that filtering is going to work a whole lot better.
> Of course, the police don't tell you about the Spectre IV, which has
> a radar-detector-detector-detector detector circuit in them.
Meh. That can be defeated with a Spectre IV Detector, or, generically, a radar detector detector detector detector detector. If they want to detect you undetectably, they'll have to watch you on satellite video and check your speed that way. Of course, that only works if you're not using a cloaking device...
You're all in violation of my patent on filing patents.
My personal favorite regex trick is the zero-width assertion. I'm particularly fond of zero-width negative lookbehind assertions. Backreferences are also cool.