One of the reasons I'm so worried to see the downward trend towards fascism in the United States is that in many ways Europe is not going in the opposite direction, it is simply lagging behind. Sure, I came to live in Switzerland, but I'm always seeing the same political abuses start to happen here just a few years after they start to happen in the United States, the same pro-corporations laws like the DMCA and the same trampling on people's rights, just a bit delayed.
Somehow this happening in the EU does not really surprise me.:-(
A quick search reveals that this Roy Schestowitz is the same guy behind boycottnovell.com (though the domain's registers don't seem to match those for schestowitz.com).
I don't know about you but for me this removes a lot of credibility from his claims (which, in good Slashdot fashion, I haven't read). I always found boycottnovell.com extremely FUDish (even though I, for many reasons, didn't like the Novell-MS agreement). The whole Novell-MS agreement taught me that some people from the Free Software or Open Source community will spread as much FUD as MS, or try to.
I'm not trying to make an ad hominem argument, just thought I'd mention who the guy was --as even if you disagree with my opinion of boycottnovell.com, I suppose you'll find this info useful.
Disclaimer: I used to work at Novell at the time but now I don't. I've been critical of many decisions by Novell (including the agreement with MS). Nowadays I work for the main competitor the eventual Microhoo would have.
Maybe, but the possibility of there only being two main search engines out there, with the next largest competitor Ask.com at a paltry 4.1%, is fairly scary.
How is that any different from there only being two main search engines out there, with the next largest competitor MSN at a paltry 5.33%?
Could you explain to me how you think I "figured that out in about five seconds" after having "never used that tool before" and decided that it "looks great to me" without trying it? I've been scratching my head over here trying to figure out how you thought I pulled that off.
Well, I think GP is right in claiming that you don't get a nice aliased circle but "you'll get a badly aliased stroke that's fuzzy" instead. It really looks bad, even though it is aliased.
But then again, I'll give you that his "keep in mind here that I've developed software plugins for the gimp since before version 2.0" is entirely irrelevant.
GP: What about "Select > Border" (optionally followed by "Select > Feather") and then just Bucket Fill?
I installed it on my 1st generation nano two weeks ago. Not only you get drag and drop support but lots of other interesting features. An important one for me is support for OGG and FLAC files. They even have a project going to get Wikipedia on it.:-)
> What happens when an advertiser notices that the related article contains material that has a different spin from its marketing communications? [...] I don't see how anyone can ever build a "Chinese wall" between advertising and editorial when any advertiser can be an editor.
I fail to see why this would be any different from now, when a would-be-advertiser (ie. someone that would advertise on wikipedia if this were allowed) notices that an article contains material that has a different spin from what it would communicate in wikipedia (again, if advertisements were allowed). This would-be-advertiser can, right now, become an editor just as well.
I don't see how being able to pay to place an ad on wikipedia would make an advertiser more prone to place non-neutral information in an article about its products, I'd even be inclined to believe that the current situation of being unable to pay to place ads would make them even more prone to place non-neutral information, if any.
Ballmer is going to be making the announcement at 2 p.m. PST. It's going to be transmited as a webcast in online (probably in a crippled patent-encoumbered file format). It seems a transcript will be posted 2 hours after it ends.
You say you are a "small email service provider" with 75 million email addresses? A simple google search shows that an email service provider with even 10 million email addresses can't be called "small" by any stretch of the word.
The sbl-xbl list stops ~ 80% of our spam. [...] Bayesian doesn't stop spam. It just flags stuff as possible spam.
Both Bayesian algorithms as well as sbl-xbl checks are just methods to flag a message/SMTP-transaction as possible spam. What you do after those checks is (eg. drop the message altogether, flag it somehow so the MUA shows it differently) could be claimed to be the difference between "stopping" and "flagging" spam. But make no mistake, sbl-xbl and Bayesian methods are the same kind of things: methods to check a message/SMTP-transaction and see if it is likely spam. How would, for example, automatically dropping messages with a 80% probability of being spam not be "stopping spam"?
That said, I think the best approach is the one used by SpamAssassin, of associating different scores to each of the tests that can be used to distinguish spam from ham. The scores are found using neural networks. While this has the minor disadvantage of requiring the message to be transported to your server (instead of being able to reject it as part of the SMTP transaction), it should have the highest reliability as far as picking spam from ham does. Usually I drop messages with a really high score, store messages with a high score in a separate per-user "Spam" folder and store the rest in the users' regular inbox.
Hmm, in case anyone is interested, I'm starting to write down a draft about how I setup my servers to do this using Debian and free software.
Even Turing failed miserably, predicting machines would be passing his Turing test around 2000. It doesn't sound like we've progressed that much from the time when he made his prediction; machines keep getting faster but as far as mathematics and theory is concerned, it doesn't seem we've come that far from where we were 50 years ago.
in around 2015-2020, you could say that we won't need people to write software, because you just explain what you want to a computer and it will write it for you, and there's no reason then to have people working in that job.
Uh, I thought that, explaining what you want to a computer, is precissely what programming is all about. Isn't source code a program's best specification? What are programmers doing if not explaining what they want from the computer?
When someone says "I want a programming language in which I need only say what I wish done," give him a lollipop.
The way the Netware stuff is currently being run in GNU/Linux is *not* using XEN. Novell has ported most of the user-space applications that make part of Netware (eg. VirtualOffice, eDirectory, NCP/NSS) to GNU/Linux. These are distributed as Open Enterprise Server, which is, basically, SUSE + Netware Applications. So I don't see how having Netware would make Novell more motivated to support XEN.
I would suggest using svnwiki, a wiki system that stores its whole contents in a Subversion repository (Disclaimer: I am the main author of svnwiki). That allows you to use the usual svn commands (svn diff, svn log, svn update, etc.) to work with your wiki as well as using the web interface.
You can see an example wiki (in spanish) and its associated svn repository (login as anonymous, password is the empty string; Slashdot seems to strip out this auth information from my URL) to get an idea of what the repository looks like.
One thing I really like about IM is that I don't have to even care about the input format, so when the users upload their random collection of gifs, jpegs, tiffs, 30 meg bmps and picts, I can just give the whole indiscriminate mess to IM and get back a set of nice, shiny, sized and formatted pngs.
I just want to point out that's something to like about NetPBM as well. You'd use anytopnm.
Note that your bar.tif file in your example is in PNM format (not TIFF, which is probably what you expect, seeing that you're using bar.tif rather than bar.pnm). PNM files can include multiple images. You probably want this:
According to the tifftopnm and pnmtotiff man pages, support for reading/writing multi-image TIF/PNM files was added in version 10.27 from March 2005. So either identify is not recognizing that your PNM file or you could use a newer version (or, perhaps, gasp, pamscale does not support multi-image PNM files?).
Hmm, I tend to find the approach of NetPBM easier to work with: having lots of separate utilities each doing a single simple thing and making it easy to use them together by piping the output of one to the next. Besides, reading/writing PBM files is trivial so you can very easily use these tools from your programs (by piping your image through them) or you can very easily create new filters that integrate well with the rest. I recommend you check out NetPBM.
If you need automatic processing of many images (the sort of thing ImageMagick is being praised for), I recommend you check it out.
the global 500m composites stored in the world big directory are raw binary files with the
dimensions 3 x 86400 x 43200 (channels x columns x rows); data type is unsigned byte, with no
header. They can be used for direct file access by data processing software (e.g. for subsetting,
web-streaming etc.)
Novell has very similar communities. You can read a little bit about them in this article of Novell's Connection Magazine (and, as you can see, this article is about 1 1/2 years old):
From Architecture to Secure Identity Management (SIM), Analytics to exteNd, Novell employees are putting their heads together in Communities of Practice. At Novell, Communities are more than just a group of like-minded individuals talking shop. They provide a primary information source for members, while promoting networking and fostering a culture that values and encourages knowledge sharing and collaboration.
While our communities aren't entirely self-governing, this doesn't seem to matter much in practice. Participation in them is entirely optional. Being a co-leader of one of these communities, I can tell you Novell greatly recognizes their value...
The licenses for this programs apply if you are obtaining them as part of SUSE. You could take a look at/usr/share/doc/packages/acroread/LICREAD.TXT and/usr/share/doc/packages/java2-jre/LICENSE. They seem to be the same as the files I liked to; at least, both do include the portion I quoted.
SuSE's License, which you agreed to, explicitly states that "Your license rights with respect to individual components accompanied by separate license terms are defined by those terms; nothing in this Agreement (including, for example, the "Other License Terms and Restrictions," below) shall [...] otherwise affect any rights or obligations You may have, or conditions to which You may be subject, under such license terms."
The fact that Novell doesn't show you the programs' licenses is, in my opinion, not relevant. It is clear that you are *not* allowed to use or redistribute software unless you are licensed to do so by the copyright owner. And the only way you're allowed seems to be those licenses, which don't allow making copies and giving them out.
So it's pretty clear: if you copy SuSE 9.3, you are copying some of the "individual component accompanied by separate license terms", that set explicit rules governing the act of redistributing them.
Moderators: parent post is spreading false information, please don't moderate as Informative. Here is why.
That Novell allows you to redistribute it does *not* mean that you can actually redistribute it.
The text you quote says: "Your license rights with respect to individual components accompanied by separate license terms are defined by those terms; nothing in this Agreement (including, for example, the "Other License Terms and Restrictions," below) shall [...] otherwise affect any rights or obligations You may have, or conditions to which You may be subject, under such license terms."
There are some proprietary packages included in SuSE Professional that you are not allowed to redistribute (consult the list of packages). Here are two examples:
Sun's Java JRE's license specifies that redistribution of this software is only allowed if it is "bundled as part of, and for the sole purpose of running, your Programs". Making a copy of the JRE as part of an entire copy of a GNU/Linux distribution seems to serve a different purpose.
Acrobat Reader's license allows you to redistribute the software but requires you to keep no copies: "You may not, rent, lease, sublicense, assign or transfer your rights in the Software, or authorize all or any portion of the Software to be copied onto another user's computer except as may be expressly permitted herein. You may, however, transfer all your rights to Use the Software to another person or legal entity provided that: [...] (b) you retain no copies, including backups and copies stored on a computer [...]".
I haven't looked at their licenses but I suspect there could be additional problems with Opera, RealPlayer, TextMaker, PlanMaker, all included in SuSE Professional 9.3.
As a consequence, if you give copies away, even if you don't charge for them, you'd be violating these packages' licenses.
So, in short, no, you can not give it away.
If you want more details, please do read the post in my weblog I mentioned in the grandparent post. Feel free to point out errors after you've read it and I'll update it.
Slashdot Mirror
Ugh.
:-(
One of the reasons I'm so worried to see the downward trend towards fascism in the United States is that in many ways Europe is not going in the opposite direction, it is simply lagging behind. Sure, I came to live in Switzerland, but I'm always seeing the same political abuses start to happen here just a few years after they start to happen in the United States, the same pro-corporations laws like the DMCA and the same trampling on people's rights, just a bit delayed.
Somehow this happening in the EU does not really surprise me.
I think, you shouldn't, use so, many commas, as it, makes it hard to, understand what you're, saying.
A quick search reveals that this Roy Schestowitz is the same guy behind boycottnovell.com (though the domain's registers don't seem to match those for schestowitz.com).
I don't know about you but for me this removes a lot of credibility from his claims (which, in good Slashdot fashion, I haven't read). I always found boycottnovell.com extremely FUDish (even though I, for many reasons, didn't like the Novell-MS agreement). The whole Novell-MS agreement taught me that some people from the Free Software or Open Source community will spread as much FUD as MS, or try to.
I'm not trying to make an ad hominem argument, just thought I'd mention who the guy was --as even if you disagree with my opinion of boycottnovell.com, I suppose you'll find this info useful.
Disclaimer: I used to work at Novell at the time but now I don't. I've been critical of many decisions by Novell (including the agreement with MS). Nowadays I work for the main competitor the eventual Microhoo would have.
How is that any different from there only being two main search engines out there, with the next largest competitor MSN at a paltry 5.33%?
Oh, come on, if people were infalible, we would never get to use it!
Ugh.
Lets email them and see what happens. Lousy website.
Well, I think GP is right in claiming that you don't get a nice aliased circle but "you'll get a badly aliased stroke that's fuzzy" instead. It really looks bad, even though it is aliased.
But then again, I'll give you that his "keep in mind here that I've developed software plugins for the gimp since before version 2.0" is entirely irrelevant.
GP: What about "Select > Border" (optionally followed by "Select > Feather") and then just Bucket Fill?
That's a very interesting perspective. Thanks for posting it. :-)
I absolutely agree with you.
I will very likely move bogowiki.org, freaks-unidos.net, bachue.com and robertoforero.com to some other registrar soon.
I installed it on my 1st generation nano two weeks ago. Not only you get drag and drop support but lots of other interesting features. An important one for me is support for OGG and FLAC files. They even have a project going to get Wikipedia on it.
> What happens when an advertiser notices that the related article contains material that has a different spin from its marketing communications? [...] I don't see how anyone can ever build a "Chinese wall" between advertising and editorial when any advertiser can be an editor.
I fail to see why this would be any different from now, when a would-be-advertiser (ie. someone that would advertise on wikipedia if this were allowed) notices that an article contains material that has a different spin from what it would communicate in wikipedia (again, if advertisements were allowed). This would-be-advertiser can, right now, become an editor just as well.
I don't see how being able to pay to place an ad on wikipedia would make an advertiser more prone to place non-neutral information in an article about its products, I'd even be inclined to believe that the current situation of being unable to pay to place ads would make them even more prone to place non-neutral information, if any.
Ballmer is going to be making the announcement at 2 p.m. PST. It's going to be transmited as a webcast in online (probably in a crippled patent-encoumbered file format). It seems a transcript will be posted 2 hours after it ends.
Both Bayesian algorithms as well as sbl-xbl checks are just methods to flag a message/SMTP-transaction as possible spam. What you do after those checks is (eg. drop the message altogether, flag it somehow so the MUA shows it differently) could be claimed to be the difference between "stopping" and "flagging" spam. But make no mistake, sbl-xbl and Bayesian methods are the same kind of things: methods to check a message/SMTP-transaction and see if it is likely spam. How would, for example, automatically dropping messages with a 80% probability of being spam not be "stopping spam"?
That said, I think the best approach is the one used by SpamAssassin, of associating different scores to each of the tests that can be used to distinguish spam from ham. The scores are found using neural networks. While this has the minor disadvantage of requiring the message to be transported to your server (instead of being able to reject it as part of the SMTP transaction), it should have the highest reliability as far as picking spam from ham does. Usually I drop messages with a really high score, store messages with a high score in a separate per-user "Spam" folder and store the rest in the users' regular inbox.
Hmm, in case anyone is interested, I'm starting to write down a draft about how I setup my servers to do this using Debian and free software.
Agreed.
Even Turing failed miserably, predicting machines would be passing his Turing test around 2000.
It doesn't sound like we've progressed that much from the time when he made his prediction;
machines keep getting faster but as far as mathematics and theory is concerned, it doesn't seem we've come that far from where we were 50 years ago.
Uh, I thought that, explaining what you want to a computer, is precissely what programming is all about. Isn't source code a program's best specification? What are programmers doing if not explaining what they want from the computer?
When someone says "I want a programming language in which I need only say what I wish done," give him a lollipop.
Felix, the author of Chicken Scheme, a fairly popular Scheme implementation, doesn't seem very fond of R6RS.
:-)
I just thought my fellow Slashdot readers could want to hear his informed opinion.
The way the Netware stuff is currently being run in GNU/Linux is *not* using XEN. Novell has ported most of the user-space applications that make part of Netware (eg. VirtualOffice, eDirectory, NCP/NSS) to GNU/Linux. These are distributed as Open Enterprise Server, which is, basically, SUSE + Netware Applications. So I don't see how having Netware would make Novell more motivated to support XEN.
I would suggest using svnwiki, a wiki system that stores its whole contents in a Subversion repository (Disclaimer: I am the main author of svnwiki). That allows you to use the usual svn commands (svn diff, svn log, svn update, etc.) to work with your wiki as well as using the web interface.
You can see an example wiki (in spanish) and its associated svn repository (login as anonymous, password is the empty string; Slashdot seems to strip out this auth information from my URL) to get an idea of what the repository looks like.
These are examples of some of its features:
I just want to point out that's something to like about NetPBM as well. You'd use anytopnm.
Note that your bar.tif file in your example is in PNM format (not TIFF, which is probably what you expect, seeing that you're using bar.tif rather than bar.pnm). PNM files can include multiple images. You probably want this:
:)
tifftopnm <foo.tif | pnmscale -xyfit 640 480 | pnmtotiff > bar.tif
According to the tifftopnm and pnmtotiff man pages, support for reading/writing multi-image TIF/PNM files was added in version 10.27 from March 2005. So either identify is not recognizing that your PNM file or you could use a newer version (or, perhaps, gasp, pamscale does not support multi-image PNM files?).
So yes, you can probably give it a try.
Hmm, I tend to find the approach of NetPBM easier to work with: having lots of separate utilities each doing a single simple thing and making it easy to use them together by piping the output of one to the next. Besides, reading/writing PBM files is trivial so you can very easily use these tools from your programs (by piping your image through them) or you can very easily create new filters that integrate well with the rest. I recommend you check out NetPBM.
If you need automatic processing of many images (the sort of thing ImageMagick is being praised for), I recommend you check it out.
Pretty raw, eh?
While our communities aren't entirely self-governing, this doesn't seem to matter much in practice. Participation in them is entirely optional. Being a co-leader of one of these communities, I can tell you Novell greatly recognizes their value...
No.
/usr/share/doc/packages/acroread/LICREAD.TXT and /usr/share/doc/packages/java2-jre/LICENSE. They seem to be the same as the files I liked to; at least, both do include the portion I quoted.
The licenses for this programs apply if you are obtaining them as part of SUSE. You could take a look at
SuSE's License, which you agreed to, explicitly states that "Your license rights with respect to individual components accompanied by separate license terms are defined by those terms; nothing in this Agreement (including, for example, the "Other License Terms and Restrictions," below) shall [...] otherwise affect any rights or obligations You may have, or conditions to which You may be subject, under such license terms."
The fact that Novell doesn't show you the programs' licenses is, in my opinion, not relevant. It is clear that you are *not* allowed to use or redistribute software unless you are licensed to do so by the copyright owner. And the only way you're allowed seems to be those licenses, which don't allow making copies and giving them out.
So it's pretty clear: if you copy SuSE 9.3, you are copying some of the "individual component accompanied by separate license terms", that set explicit rules governing the act of redistributing them.
That Novell allows you to redistribute it does *not* mean that you can actually redistribute it.
The text you quote says: "Your license rights with respect to individual components accompanied by separate license terms are defined by those terms; nothing in this Agreement (including, for example, the "Other License Terms and Restrictions," below) shall [...] otherwise affect any rights or obligations You may have, or conditions to which You may be subject, under such license terms."
There are some proprietary packages included in SuSE Professional that you are not allowed to redistribute (consult the list of packages). Here are two examples:
I haven't looked at their licenses but I suspect there could be additional problems with Opera, RealPlayer, TextMaker, PlanMaker, all included in SuSE Professional 9.3.
As a consequence, if you give copies away, even if you don't charge for them, you'd be violating these packages' licenses.
So, in short, no, you can not give it away.
If you want more details, please do read the post in my weblog I mentioned in the grandparent post. Feel free to point out errors after you've read it and I'll update it.