SASOL is still going strong here in South Africa. We have lots of coal, and it's quite expensive to import oil, so the whole process is still economically viable without the boycott. As a result, though, Sasolburg is one of the more polluted cities in the country (see amongst others the groundwork's 2002 annual report on this site.)
It's perfectly possible to get very good results on comparatively small databases (a few 100 faces). See the comprative tests conducted on the Surrey XM2VTS database, or the FERET test runs. These tests don't adress the disguise issues, but do address natural variation over time (breads, hair-style, presence or absence of glasses). Partly because computer vision is still a comparatively young field, there's very little work on performance of long time periods. Most databases cover only two or three years
It's hard to scale to significantly larger databases, and it's not an uncommon opinion that face recognition is fundamentally unsuited to very large scale problems. Indeed, some studies suggest that faces are inherently not distinct enough to be a reliable measure when the database size gets very large.
GPL-incompatible means GPL incompatible, not non-free. This is really not hard to understand.
Combing GPL code with a GPL-incompatible license produces code that cannot be distributed. The GPL v2 specifies, you cannot add further restrictions, so if I combine this with code with a license that adds further restrictions, the code can no longer be distributed under the GPL. If I don't have permission from all the GPL contributers to relicense their code, I cannot legally redistribute the combined work. This is pretty much the entire point of copyleft.
Since the latest cdrtools packages look to be a combination of GPL'd code and incompatibly licensed code, Debian is removing crtools (not shunting it to non-free), because they feel they can no longer distribute the work.
They are not claiming it can't be stolen. They claiming it can't be stolen by starting the engine without a copy of the key. There are numerous other ways a car can be stolen. While this is provably incorrect, their position is at least logically consistent.
Does anyone even know that the FAQ was written by a lawyer or was it written by an intern in the PR dept?
Yes, it was written by their lawyers, and was intended to clarify those aspects about the license the ftp-masters found troubling. Read the various posts about how java was accepted for details.
A major problem was the large disclaimer declaring the FAQ not legally binding. This appears to be a goof on the part of SUN, and the latest version clarifies this disclaimer considerably (see DLJ FAQ 1.2).
I don't see how the claim that other OS's have their share of assholes makes the article flamebait
The article doesn't claim that Linux is particularly unusual in the number of assholes out there.
The article's premise is that, because some people's first attempt to engage the Linux community means they encounter an asshole, it hurts Linux adoption. There is sufficient ancedotal evidence (just look at the discussion here) of this.
This is also of course hardly news. The Linux Advocacy mini-HOWTO, which includes a lot of "don't be an asshole" advice, has a copyright dating back to 1996. Which also highlights the weakness of all these complaints - the people who cause the problem aren't going to change. We're always going to have some assholes in our community, and, as long as Linux is not an obvious choice for a given application, they have the potential to hurt adoption. Which sucks, but so be it.
Ubuntu breezy:
xmms (from main), includes libmpg123 for mp3 playback
mpg321 & lame (mp3 encoder) are available from universe
Doesn't look like Ubuntu is particularly worried by the mp3 patents
This is probably partly because Debian has never been particularly worried by the mp3 patents - see the (many) discussions on the issues in debian-legal for example.
Xgl -> an Xserver running ontop of OpenGL. i.e. all the X11 drawing calls are converted to OpenGl calls executed on the underlying 3D hardware of the display. It's still just an xserver, so any remote client will still be able to connect and run.
Based on the very little information available in the article, this sounds like an offshoot of the work on interactive proofs, and the UCLA professor quoted does indeed seem to have done some work in the field - see http://www.cs.ucla.edu/~rafail/PUBLIC/index.html for his publications.
A glance at the paper titles suggests "Private Searching on Streaming Data" as being the closest to the original article.
Isn't being fast and effective the whole point of usability? Shouldn't we be focussing on getting people using setups that are useful to what they're trying to do, rather than trying to create some magical "works for everyone" setup.
The commandline is usable, and extremely usable in positions where a GUI is impossible (wrong end of a slow network connection, a situation all to common in remote adminstration), but, granted, it is not particularly useful to someone working in graphics design, for instance.
It's not yet at "my lawyer can beat up your lawyer". Based on the article, the original letter sounds like "You may be doing something illegal. Can you cross the aprpopriate t's and dot the i's so that everyone's happy?". The story will only become interesting if either the Oninion refuses to do anything (and it sounds as if they don't need to do much to be OK), or, if they do something reasonably, the adminstration still pursues them.
Is it any more credible than MS studies, no. However, in certain management circles, the MS studies are considered very credible precisley because they're backed by MS.
This study will be very useful as a counterbalance to the MS-funded studies, andgiven that it's backed by IBM, it can't be as easily ignored by management as some of the other, recent refutations of MS's results.
From the article, they pump the CO2 into a old oil-field, making it easier to pump out oil, and, as a bonus, burying the CO2. Quite a nifty trick, but does require a nearby underground oil-field to work. This also means that you're pumping the gas into a pretty well-sealed enviroment, so leakage is not much of a factor.
I hardly see the need for a lawyer - finding the license for a library is usually as trivial as less/usr/share/doc//copyright (on debian) or COPYING on most other systems. Then the rules are simple - BSD or LGPL, can link, GPL, can't. In fact, if finding the library license is not trivial, then there's a serious error in the packaging and installation of the library.
Furthermore, the only major library on my system which is GPL is Qt, and that's GPL'd for well-known public reasons. The other GPL'd libraries are small, specific ones that I can't see anyone linking to without looking at the documentation, and then not knowing that they're GPL'd is hardly an acceptable excuse. Which libraries did you link to to cause the problem?
That's really up to the search engine implementation, isn't it.
Anyway, a brief look at the proposed format gives very little scope for abuse - you can specify location, change frequency, last modified and a priority, and that's it. The priority is specified as only applying to urls from the same site, so what you can do with it is fairly limited. Overall, it looks written as a set of additional hints to spiders crawling the site.
You're missing the whole point of peer-review. The classical model of scientific jounrals is that the edito removes obvious junk and hands the interesting articles off to the reviewers, If the reviewer, who are people in similiar fields, come back wiith "It basically looks alright", then it gets publsihed. The editor should only really be involved when the reviewers disagree. Its a simple filter to basically catch errors in the science (depending on the filed, the filter can be failry lack, or very tight).
The article is about somebody protesting the editortial decision not to have the paper reviewed. It's an interesting, although probably expensive, way of protesting editorial policy.
Privacy freaks are anti RFID (and any similiar distance tagging method) for precisely two reasons:
It's passive (minimal activity required by anyone to get something scanned) and it's long range. While the ability to link identity to purchases (assuming no cash transactions) exists with bar-code readers, it's a much more active system, and the user has much more control over when and where this information is collected.
If with a few minutes thought, you can't construct a worst case scenerio for long-range (where long range is further than about 20cm) bar-codes, there no hope for you as a privacy freak:).
If you look at the history of the office suite contenders, it was only with Win95, and the simulatenous release of MS Office 95, with 32bit features wel ahead of the competition, that MS was really able to kickstart its monopoly.
Win95 only happened, however, because, Windows 3 had a reasonable market share. The article argues, that, without protected mode support coming when it did, Microsoft would have abadoned that line and gone with OS/2.
The articles more than a little light on details, but digital watermarking is usually used to refer to an optical technique. Usually the idea is to embed some identifying characteristic in the image without visually impacting on the quality of the image. Digital watermarking is aimed at having some method for identifying the oprigin of the image. They seem to be combining this with stenography and embedding additional information in the image.
Thus, if the article is using convential terminology, this is will require an optical scanner, and probaly one that scans the card at quite a high resolution. In this case, there should be minimal risk of remote information leakage. The dangers of any watermarking scheme is that there is a limit to how robust you can make it against changes to the image. If poorly implemented, this scheme could quite easily fail if the license is dirty or damaged.
It is of course, completely possible that they're using a technology like RFID and simply using the phrase digital watermark to confuse the issue, but let us wait for more evidence before making that assumption.
The GIMP: a program written by coders and used by artists. The number of artists who are also coders is quite small. The number of artists who use the GIMP is quite large. This looks like a nice way for the community of users to get involved.
Actually, its very easy to prove human influence on climate, at least locally. Cities are warmer than surrounding rural areas, dams seriously change downriver flood-cycles, acid rain, etc.
Global cliamte changes are harder because of one simple fact. The climate is changing anyway.
This means that, much of the work in modelling climate change is trying to seperate change due to natural processes from change due to human action. Thus, while there are many emimently plausible theories as to how our actions are affecting the climate, and there is general consensus that we are having an impact, the open questions are a) How significant is this impact? and b) if we change our behaviour to X, how will this impact on climate change?
Unfortunatly, curent evidence suggests that the answer to a) is quite significant and the answer to b) is moot as there seems little prospect of any significant changes in pollution behaviour in the forseeable future.
Yes there are many different options open to the world, some of which may actaully make a difference.
Unfortunately, as far as the politicans are concerned, all the options that have any chance of actually being implemented seem to be merely different ways of doing nothing.
Umm - E17 has been checked into public cvs, which means it's at the "code good enough for other people to see, hack on and improve" stage, not the "release this to non-developers of the package" stage. Trying to get it into sarge at this point might be considered a tad premature and would be a great disservice to both the users (stuck with buggy windowmanager until next release) and E (bad reputation becuase debian version is buggy and can't be fixed).
Of course, E17 might stablise and release well before sarge does:).
Firefox 1.0 is in. Gimp 2.0 is in, Gimp 2.2 is unlikely.
Slashdot Mirror
SASOL is still going strong here in South Africa. We have lots of coal, and it's quite expensive to import oil, so the whole process is still economically viable without the boycott. As a result, though, Sasolburg is one of the more polluted cities in the country (see amongst others the groundwork's 2002 annual report on this site.)
It's hard to scale to significantly larger databases, and it's not an uncommon opinion that face recognition is fundamentally unsuited to very large scale problems. Indeed, some studies suggest that faces are inherently not distinct enough to be a reliable measure when the database size gets very large.
Combing GPL code with a GPL-incompatible license produces code that cannot be distributed. The GPL v2 specifies, you cannot add further restrictions, so if I combine this with code with a license that adds further restrictions, the code can no longer be distributed under the GPL. If I don't have permission from all the GPL contributers to relicense their code, I cannot legally redistribute the combined work. This is pretty much the entire point of copyleft.
Since the latest cdrtools packages look to be a combination of GPL'd code and incompatibly licensed code, Debian is removing crtools (not shunting it to non-free), because they feel they can no longer distribute the work.
They are not claiming it can't be stolen. They claiming it can't be stolen by starting the engine without a copy of the key. There are numerous other ways a car can be stolen. While this is provably incorrect, their position is at least logically consistent.
Yes, it was written by their lawyers, and was intended to clarify those aspects about the license the ftp-masters found troubling. Read the various posts about how java was accepted for details.
A major problem was the large disclaimer declaring the FAQ not legally binding. This appears to be a goof on the part of SUN, and the latest version clarifies this disclaimer considerably (see DLJ FAQ 1.2).
The article doesn't claim that Linux is particularly unusual in the number of assholes out there. The article's premise is that, because some people's first attempt to engage the Linux community means they encounter an asshole, it hurts Linux adoption. There is sufficient ancedotal evidence (just look at the discussion here) of this.
This is also of course hardly news. The Linux Advocacy mini-HOWTO, which includes a lot of "don't be an asshole" advice, has a copyright dating back to 1996. Which also highlights the weakness of all these complaints - the people who cause the problem aren't going to change. We're always going to have some assholes in our community, and, as long as Linux is not an obvious choice for a given application, they have the potential to hurt adoption. Which sucks, but so be it.
mpg321 & lame (mp3 encoder) are available from universe
Doesn't look like Ubuntu is particularly worried by the mp3 patents
This is probably partly because Debian has never been particularly worried by the mp3 patents - see the (many) discussions on the issues in debian-legal for example.
Xgl -> an Xserver running ontop of OpenGL. i.e. all the X11 drawing calls are converted to OpenGl calls executed on the underlying 3D hardware of the display. It's still just an xserver, so any remote client will still be able to connect and run.
A glance at the paper titles suggests "Private Searching on Streaming Data" as being the closest to the original article.
The commandline is usable, and extremely usable in positions where a GUI is impossible (wrong end of a slow network connection, a situation all to common in remote adminstration), but, granted, it is not particularly useful to someone working in graphics design, for instance.
It's not yet at "my lawyer can beat up your lawyer". Based on the article, the original letter sounds like "You may be doing something illegal. Can you cross the aprpopriate t's and dot the i's so that everyone's happy?". The story will only become interesting if either the Oninion refuses to do anything (and it sounds as if they don't need to do much to be OK), or, if they do something reasonably, the adminstration still pursues them.
This study will be very useful as a counterbalance to the MS-funded studies, andgiven that it's backed by IBM, it can't be as easily ignored by management as some of the other, recent refutations of MS's results.
News, no. Good PR, most definately.
From the article, they pump the CO2 into a old oil-field, making it easier to pump out oil, and, as a bonus, burying the CO2. Quite a nifty trick, but does require a nearby underground oil-field to work. This also means that you're pumping the gas into a pretty well-sealed enviroment, so leakage is not much of a factor.
I hardly see the need for a lawyer - finding the license for a library is usually as trivial as less /usr/share/doc//copyright (on debian) or COPYING on most other systems. Then the rules are simple - BSD or LGPL, can link, GPL, can't. In fact, if finding the library license is not trivial, then there's a serious error in the packaging and installation of the library.
Furthermore, the only major library on my system which is GPL is Qt, and that's GPL'd for well-known public reasons. The other GPL'd libraries are small, specific ones that I can't see anyone linking to without looking at the documentation, and then not knowing that they're GPL'd is hardly an acceptable excuse. Which libraries did you link to to cause the problem?
Anyway, a brief look at the proposed format gives very little scope for abuse - you can specify location, change frequency, last modified and a priority, and that's it. The priority is specified as only applying to urls from the same site, so what you can do with it is fairly limited. Overall, it looks written as a set of additional hints to spiders crawling the site.
The article is about somebody protesting the editortial decision not to have the paper reviewed. It's an interesting, although probably expensive, way of protesting editorial policy.
Of course, as linuxense offer a security service, this smells like a "to be spun into marketing copy" stunt.
It's passive (minimal activity required by anyone to get something scanned) and it's long range. While the ability to link identity to purchases (assuming no cash transactions) exists with bar-code readers, it's a much more active system, and the user has much more control over when and where this information is collected.
If with a few minutes thought, you can't construct a worst case scenerio for long-range (where long range is further than about 20cm) bar-codes, there no hope for you as a privacy freak :).
Luc Saillard picked up the Phillips driver awhile ago now. See his PWC page for details.
Win95 only happened, however, because, Windows 3 had a reasonable market share. The article argues, that, without protected mode support coming when it did, Microsoft would have abadoned that line and gone with OS/2.
Thus, if the article is using convential terminology, this is will require an optical scanner, and probaly one that scans the card at quite a high resolution. In this case, there should be minimal risk of remote information leakage. The dangers of any watermarking scheme is that there is a limit to how robust you can make it against changes to the image. If poorly implemented, this scheme could quite easily fail if the license is dirty or damaged.
It is of course, completely possible that they're using a technology like RFID and simply using the phrase digital watermark to confuse the issue, but let us wait for more evidence before making that assumption.
Global cliamte changes are harder because of one simple fact. The climate is changing anyway.
This means that, much of the work in modelling climate change is trying to seperate change due to natural processes from change due to human action. Thus, while there are many emimently plausible theories as to how our actions are affecting the climate, and there is general consensus that we are having an impact, the open questions are a) How significant is this impact? and b) if we change our behaviour to X, how will this impact on climate change?
Unfortunatly, curent evidence suggests that the answer to a) is quite significant and the answer to b) is moot as there seems little prospect of any significant changes in pollution behaviour in the forseeable future.
Unfortunately, as far as the politicans are concerned, all the options that have any chance of actually being implemented seem to be merely different ways of doing nothing.
Umm - E17 has been checked into public cvs, which means it's at the "code good enough for other people to see, hack on and improve" stage, not the "release this to non-developers of the package" stage. Trying to get it into sarge at this point might be considered a tad premature and would be a great disservice to both the users (stuck with buggy windowmanager until next release) and E (bad reputation becuase debian version is buggy and can't be fixed). Of course, E17 might stablise and release well before sarge does :).
Firefox 1.0 is in. Gimp 2.0 is in, Gimp 2.2 is unlikely.