I was one of the "ill-dressed" speakers - usual
shirt and jeans. I suspect that if I had turned up in a suit, then (a) I would have not been taken seriously, and (b) I would have been bitch-slapped by my peers. It was a technical conference - I was there to present my work to my peers, not to drum up capital.
Never, at the current rate. The maintainers of Linux (the current standard-bearer of open source) are infighting, bashing, selfish, attention-grabbing individuals. If you look through the last year of Slashdot articles, I'm sure you can find at least 10 or 12 articles relating to infighting on lists.
I wonder if you've ever worked at a major commercial software firm? You will find egos and back-stabbing tactics to make your head spin. IBM has so much management and so many turf wars, it's a wonder they get anything done at all.
There are groups within Microsoft that refuse to release source code to other groups in Microsoft even when those other groups can't figure out the API's without it! "It's our code and you would just fork it and mess it up!" Seriously.
Just because the OSS infighting is public doesn't make it unusual. It's mild compared to some of the in-house stuff I've seen.
As this is my home network, everything is blocked. If it's incoming, then it ain't me and I don't want to know about it. I'm just getting a little tired of watching the cable modem blink, knowing that I'm going to have to pay for this idiot's excursions at the end of the month.
Verisign does provide a semi-useful service, in that it verifies the identiry of the people whose certificates it signs. However, charging the same person $N every single year to change the expiry date in the certificate is not a useful service - it is highway robbery.
Many of the fields in the X.509 certificate, like the expiry date, and the CA signer field, exist solely to create the business model. There's no technical or trust-based justification for it.
Like many of the DRM initiatives right now, certificates are designed to support the business model, rather than being designed to solve the problem.
In the case of Verisign, this was their intent - RSA Data Security, Inc. used the RSA patent as a weapon in the 1990's to ensure that their way of using the algorithm became standardised, locking every other attempt at solving the trust problem (yet another reason to reform the patent system).
One of the founding principles of Internet design is that protocol and business model are two separate issues. We desperately need an alternative to certificates that can provide the necessary trust matrix without nailing all users to one way of providing the service. Unfortunately, it is probably too late to fix it now.
The GPL already contains similar wording with respect to patents. The purpose is to prevent the following situations:
1. A company puts their code out under an OSS license. It becomes really popular, but unbeknownst to the public, the techniques it uses are patented by the compnay. Company turns around and demands royalties from everyone who made it popular.
2. An OSS product is released, and unbeknownst to its author, someone else has a patent. This someone else waits for it to become popular and then demands royalties.
Mutual termination ensures that if anyone asserts patent royalties then everyone loses all rights to the software. This way, none of the users have the right to use the software any more (in the Copyright sense), and so since they can't be using it, they cannot be forced to pay royalties. This keeps the patent owners honest - they can't hijack OSS to further non-OSS goals.
BSD, X11, etc don't have such protections, which makes them a riskier proposition when dealing with patent owners. The way I read it, the Academic Free License is basically BSD, with patent protection built in.
This is not censorship, and it isn't any different from TV stations editing out naughty bits or beeping out cusswords.
One of my favourite all-time movies is The Breakfast Club. I loved that movie, growing up in Australia. It was always shown in its original four-letter form.
One time, when I was visiting the US, I had the unfortunate experience of watching it on TV with all the naughty words altered. It completely ruined the emotional impact of the movie.
Such editing should not be done without the permission of the director. Not by Clean Flicks, and neither by TV stations.
Those who don't understand X are doomed to re-invent it... poorly.
Point 1: Dump X11 entirely. It's a lot easier to write libraries to display X11 apps in a different environment than it is to make X11 into a modern graphics environment. Its development began 18 years ago (released 14 years ago), and frankly, its age shows, both in performance and in functionality.
X11 has great performance. Unfortunately, few toolkits use it well. It works best when you think of it as a stream: you send asynchronous requests to the display server and it handles them, responding with asynchronous events. As soon as you make a request to the display server that requires a synchronous response, performance is gone.
Point 2: OpenGL compositing a la Quartz Extreme. Windows become patterns mapped onto a plane. 3D graphics are tightly integrated into the same screen model.
Point 3: With the exception of bitmaps (which you map as a pattern), draw all the 2d windows using 3d primitives, say as a variant of splines that have thickness, located just in front of a 2d plane.
Berlin was doing this. Ain't exactly taking over the world at the moment. 99% of all apps don't require anything more fancy than 2D drawing primitives and a few icon pixmaps. For every canvas-based, client-side, anti-aliased app I've seen, I've seen 10 boring apps written by people who understand X11 that perform 10 times better.
Point 4: Do not use a client-server model. It made sense in 1984. It doesn't make sense in 2002. Most people don't have graphical terminals connected to big centralized servers these days. A client-server model can easily be grafted onto a local model if it is designed correctly. By contrast, local communication via a client-server model tends to cause a speed penalty.
Client/server is fundamental to the design of both Unix and X11. Try this: administer your parent's Windows or Mac machine from your home 100 miles away, as though you were sitting right there. Can't do it? Now try this: install Linux. Ssh in and type linuxconf. There's a reason why this works in Unix systems: clear separation of client from server.
The speed penalty only happens because people don't use X11 asynchronously, or they try to use it in raw bitmap mode instead of learning what all those pesky XDrawLine, XDrawString, etc functions do.
Most applications "wear their code on their face". That is, a competent programmer can intuit how the code behaves by clicking the buttons and manipulating the widgets. e.g. you can get a pretty good understanding of a spreadsheet's algorithms by playing with it for a few hours.
Not all applications are like this. It isn't possible to intuit how the Discrete Cosine Transform works in JPEG by using an image viewer, for example.
This ruling appears to be setting the (very bad) precedent that intuiting an obvious algorithm from using a piece of software is reverse engineering.
I just can't image Ted "Theodore" Logan as Superman. It just doesn't work.
They should do what they did with Spiderman - find a little known but good actor and have him re-invent the role. That way, the audience won't bring any misconceptions into the cinema, which will ruin the whole experience.
So, get a chipped X-box, download the distribution, and try it yourself. Or wait for a zillion other/.'s to do it and report their experiences. With open source, checking for a fake is trivial.
RedHat does have some leverage here. For example, giving preferential positioning to icons for their own applications in RedHat 27.2. MS got slapped for that just recently.
But there is a difference. MS used contracts and stand-over tactics with OEM's to prevent the icons from being changed. In RedHat's case, the GPL acts as a counter-balancing force.
While they continue to GPL everything they do, the license makes it legal for an OEM to apply a "mod kit RPM" that modifies the RedHat distro however they want.
Also, unlike MS, RedHat cannot say "fine, we will withdraw your license". The minute they tried that, the OEM's would fork the code and tell RedHat to get lost.
RedHat will only survive so long as they provide a useful service. They are dead the moment they stop.
UnitedLinux would be better off copying RedHat than trying to re-invent the glory days of proprietry Unix where vendors lorded it over users and _all_ Unix distributions sucked.
I'm totally opposed to any form of content tagging, whether it be a flag (gee, I wonder how I turn that off? Toggle it?), a digital watermark, or anything else.
The reason is simple: the studios and TV stations will always flip the flag to "not allowed to record". And then sue the pants off anyone who wants to tape the show for later.
Why bother having a tagging system if it is always going to be set to the same value, and that value is incompatible with common sense uses of the work?
I just hope that the EFF can get their FAQ seen by a sufficient number of politicians that lawmakers begin to see the futility and stupidity of access control tagging. A flag that is set to a constant value is no better than no flag at all.
The times should have been "10:00 UTC Saturday 29th July until 22:00 UTC Sunday 30th July". Sorry for any confusion. Please join us on irc.openprojects.net#dotgnu, any time this weekend.
It took Intel 30 years to acheive a chip that has the same cooling requirements:
The 25120 is easily cooled by employment of a six-foot fan, 1/2" from the package. If the device fails, you have exceeded the ratings. In such cases, more air is recommended.
It was on the ABC in Australia last year, starring Michael Gambon and Jeremy Irons. A BBC series originally I believe.
It was one of the best science-related history shows I've ever seen, tracing the story of H1 through H4 in a way anyone could appreciate.
The show also told the story of the WWI soldier, emotionally damaged by his war experiences, who painstakingly restored the clocks in the early part of the 20th century.
It's amazing how much we owe to Harrison. It's a pity that he had to fight so hard for his compensation, because the upper-class science types of the time refused to believe that a cabinet maker could come up with the solution.
Yep, management sucks. Even when done by ex-programmers. Something about stopping coding 20 hrs a day seems to mess with their heads, and disconnect them from the work that needs to be done. It's the same pretty much everywhere.
We've been trying to solve this problem since the days of Fred Brooks, and we probably never will.
A more difficult problem for job seekers is the phase shift that is happening in the industry right now. i.e. the wholesale move from infrastructure (OS's and applications) to services (support, training, script glue work, etc).
I pride myself on being a "Hacker of the Old School" (HOTOS). But there simply aren't any jobs for HOTOS'es any more. It's all gone to MCSE's, sysadmins, and PHP/Perl/SQL script monkeys.
A few days ago, I hired a career consultant to help me find a new career away from programming. It's either that or slowly go mad in the declining IT industry.
I love programming, but the industry doesn't love the kind of programming I do anymore.:-(
The X11 license is a non-copyleft license. As such, it does not guarantee that future versions remain free.
While it is true that there are no Mono components that are currently under a non-free license, Ximian has made no guarantee to the community that this will remain the case. I have.
All that is possible is to release derived work without a Free license. It's not as if the original Free code would go away. You do not lose anything - you just do not get more. That's quite a difference.
The danger is always that the derived work becomes more extensively supported than the free parts of it. Then it doesn't matter how free the free parts are, because they are useless without the non-free parts. GPL prevents this from occurring. X11 does not.
Mono will only be truly Free when you make a public commitment never, ever, to change the license to something that is incompatible with the wishes of the community. Yet, you still maintain the right to change Mono's license any way you see fit at any moment.
Sure, Copyright law may allow you to do this, but that is a technicality. I have made a strong commitment to DotGNU and the GNU Project to *never* change the license in a way which may harm the community's interests. Will you do the same?
Free Software is more than just using a GPL-compatible license. It also entails that all future uses of the software remain Free, in all respects. The X11 license does not preserve this freedom, as it allows components to be proprietrised.
Ximian is putting pragmatism ahead of the community by bowing to Intel and HP in this.
With this decision, all of the Mono components are now non-Free, or can be made non-Free at any moment. In this message, Miguel makes it clear that Ximian wants to own the Copyright to the engine and C# compiler also, so they can change the license on that whenever they see fit.
This leaves DotGNU Portable.NET as the only true Free Software project tackling the implementation of the CLR, C# compiler, C# class library, etc.
We are looking for developers to help us build our system into a truly-Free implementation. Portable.NET has been around longer than Mono, and remains true to the principles of Free Software.
If they need better writers, then how about submitting an article? I have.
Seriously, this is their first issue, and they are still working the bugs out of the editing process.
Most magazines are a little flaky in their first few issues, but then settle down over time as they attract quality writers and columnists, start evolving their own unique style, etc.
Hong Feng, the magazine's founder, is taking a big chance here, and I think he can pull it off. With our help.
Slashdot Mirror
I was one of the "ill-dressed" speakers - usual shirt and jeans. I suspect that if I had turned up in a suit, then (a) I would have not been taken seriously, and (b) I would have been bitch-slapped by my peers. It was a technical conference - I was there to present my work to my peers, not to drum up capital.
Just as well I'm a habitual procrastinator ... I bought an APC UPS about a year ago and never got around to plugging it in! Saved by my own laziness!
I wonder if you've ever worked at a major commercial software firm? You will find egos and back-stabbing tactics to make your head spin. IBM has so much management and so many turf wars, it's a wonder they get anything done at all.
There are groups within Microsoft that refuse to release source code to other groups in Microsoft even when those other groups can't figure out the API's without it! "It's our code and you would just fork it and mess it up!" Seriously.
Just because the OSS infighting is public doesn't make it unusual. It's mild compared to some of the in-house stuff I've seen.
Make all the jokes you like about File Allocation Tables. See if we care. This is a Microsoft-bashing site after all.
As this is my home network, everything is blocked. If it's incoming, then it ain't me and I don't want to know about it. I'm just getting a little tired of watching the cable modem blink, knowing that I'm going to have to pay for this idiot's excursions at the end of the month.
Free Clue: if you didn't get in on the first 2000 tries, go waste someone else's bandwidth!
Many of the fields in the X.509 certificate, like the expiry date, and the CA signer field, exist solely to create the business model. There's no technical or trust-based justification for it. Like many of the DRM initiatives right now, certificates are designed to support the business model, rather than being designed to solve the problem.
In the case of Verisign, this was their intent - RSA Data Security, Inc. used the RSA patent as a weapon in the 1990's to ensure that their way of using the algorithm became standardised, locking every other attempt at solving the trust problem (yet another reason to reform the patent system).
One of the founding principles of Internet design is that protocol and business model are two separate issues. We desperately need an alternative to certificates that can provide the necessary trust matrix without nailing all users to one way of providing the service. Unfortunately, it is probably too late to fix it now.
1. A company puts their code out under an OSS license. It becomes really popular, but unbeknownst to the public, the techniques it uses are patented by the compnay. Company turns around and demands royalties from everyone who made it popular.
2. An OSS product is released, and unbeknownst to its author, someone else has a patent. This someone else waits for it to become popular and then demands royalties.
Mutual termination ensures that if anyone asserts patent royalties then everyone loses all rights to the software. This way, none of the users have the right to use the software any more (in the Copyright sense), and so since they can't be using it, they cannot be forced to pay royalties. This keeps the patent owners honest - they can't hijack OSS to further non-OSS goals.
BSD, X11, etc don't have such protections, which makes them a riskier proposition when dealing with patent owners. The way I read it, the Academic Free License is basically BSD, with patent protection built in.
One time, when I was visiting the US, I had the unfortunate experience of watching it on TV with all the naughty words altered. It completely ruined the emotional impact of the movie.
Such editing should not be done without the permission of the director. Not by Clean Flicks, and neither by TV stations.
The speed penalty only happens because people don't use X11 asynchronously, or they try to use it in raw bitmap mode instead of learning what all those pesky XDrawLine, XDrawString, etc functions do.
Not all applications are like this. It isn't possible to intuit how the Discrete Cosine Transform works in JPEG by using an image viewer, for example.
This ruling appears to be setting the (very bad) precedent that intuiting an obvious algorithm from using a piece of software is reverse engineering.
They should do what they did with Spiderman - find a little known but good actor and have him re-invent the role. That way, the audience won't bring any misconceptions into the cinema, which will ruin the whole experience.
So, get a chipped X-box, download the distribution, and try it yourself. Or wait for a zillion other /.'s to do it and report their experiences. With open source, checking for a fake is trivial.
But there is a difference. MS used contracts and stand-over tactics with OEM's to prevent the icons from being changed. In RedHat's case, the GPL acts as a counter-balancing force.
While they continue to GPL everything they do, the license makes it legal for an OEM to apply a "mod kit RPM" that modifies the RedHat distro however they want.
Also, unlike MS, RedHat cannot say "fine, we will withdraw your license". The minute they tried that, the OEM's would fork the code and tell RedHat to get lost.
RedHat will only survive so long as they provide a useful service. They are dead the moment they stop.
UnitedLinux would be better off copying RedHat than trying to re-invent the glory days of proprietry Unix where vendors lorded it over users and _all_ Unix distributions sucked.
I'm totally opposed to any form of content tagging, whether it be a flag (gee, I wonder how I turn that off? Toggle it?), a digital watermark, or anything else.
The reason is simple: the studios and TV stations will always flip the flag to "not allowed to record". And then sue the pants off anyone who wants to tape the show for later.
Why bother having a tagging system if it is always going to be set to the same value, and that value is incompatible with common sense uses of the work?
I just hope that the EFF can get their FAQ seen by a sufficient number of politicians that lawmakers begin to see the futility and stupidity of access control tagging. A flag that is set to a constant value is no better than no flag at all.
I really should drink caffeine *before* posting to Slashdot at 7:30am. It is this month. Sorry (again) for the confusion.
The times should have been "10:00 UTC Saturday 29th July until 22:00 UTC Sunday 30th July". Sorry for any confusion. Please join us on irc.openprojects.net#dotgnu, any time this weekend.
It was one of the best science-related history shows I've ever seen, tracing the story of H1 through H4 in a way anyone could appreciate.
The show also told the story of the WWI soldier, emotionally damaged by his war experiences, who painstakingly restored the clocks in the early part of the 20th century.
It's amazing how much we owe to Harrison. It's a pity that he had to fight so hard for his compensation, because the upper-class science types of the time refused to believe that a cabinet maker could come up with the solution.
We've been trying to solve this problem since the days of Fred Brooks, and we probably never will.
A more difficult problem for job seekers is the phase shift that is happening in the industry right now. i.e. the wholesale move from infrastructure (OS's and applications) to services (support, training, script glue work, etc).
I pride myself on being a "Hacker of the Old School" (HOTOS). But there simply aren't any jobs for HOTOS'es any more. It's all gone to MCSE's, sysadmins, and PHP/Perl/SQL script monkeys.
A few days ago, I hired a career consultant to help me find a new career away from programming. It's either that or slowly go mad in the declining IT industry.
I love programming, but the industry doesn't love the kind of programming I do anymore. :-(
While it is true that there are no Mono components that are currently under a non-free license, Ximian has made no guarantee to the community that this will remain the case. I have.
Sure, Copyright law may allow you to do this, but that is a technicality. I have made a strong commitment to DotGNU and the GNU Project to *never* change the license in a way which may harm the community's interests. Will you do the same?
Free Software is more than just using a GPL-compatible license. It also entails that all future uses of the software remain Free, in all respects. The X11 license does not preserve this freedom, as it allows components to be proprietrised.
Ximian is putting pragmatism ahead of the community by bowing to Intel and HP in this.
This leaves DotGNU Portable.NET as the only true Free Software project tackling the implementation of the CLR, C# compiler, C# class library, etc.
http://www.southern-storm.com.au/portable_net.html .
We are looking for developers to help us build our system into a truly-Free implementation. Portable.NET has been around longer than Mono, and remains true to the principles of Free Software.
If they need better writers, then how about submitting an article? I have.
Seriously, this is their first issue, and they are still working the bugs out of the editing process.
Most magazines are a little flaky in their first few issues, but then settle down over time as they attract quality writers and columnists, start evolving their own unique style, etc.
Hong Feng, the magazine's founder, is taking a big chance here, and I think he can pull it off. With our help.