Looking at the summary of changes, I suspect that they've finally messed up the things that made FORTRAN (or at least F77) great. The addition of pointers stands an excellent chance of rendering code un-optimizable, and I fear that adding OO features is an even bigger mistake.
FORTRAN went OO during with Fortran90 and pointers were added with Fortran95, although for anyone used to C pointers, Fortran95 pointers are (erm) pretty wierd...
Back in my grad school days (5 yrs ago) everyone who had supercomputer time was writing in FORTRAN.
Amoungst physicists and astronomers, anyone with supercomputer time today is still writing in FORTRAN. I don't see that changing in the near future. Graduate students are starting to write in different languages, but to talk to their superivsors (the guys with large existing codebases) they have to be able to talk and write FORTRAN.
Most of the time its not worth their time to port a couple of hundred thousand lines of modelling code to some other language before they can add layers on top, its unlikely it'll ever be worth anyones time to do somethnig like that.
A lot of the Java, Perl/Tk and Python/Tkinter stuff I write these days is sitting ontop of a whole bunch of FORTRAN that it would take the next twenty years to rewrite. If it works, whats the point of doing it all again? Thats no fun...
I think language interoperability is becoming more and more important now the industry is (sort of) maturing. This is where code reuse in the real world is going to come from, rather than the current fads in code design which keep on promising more code reuse.
Good luck with that. It seems to me this "feature", just like text messaging, is being pushed by a culture that just doesn't understand Americans. Just because SMS is all the rage in Norway and the Japan...
SMS has has a major cultural impact in the UK, over 750 million messages are now sent every month in the UK. There are lots of circumstances where sending a quick text message is alot more convenient that trying to call someone who may be busy doing other things. Like email the asynchronous nature of text messaging has advantages at times.
Appart from the fact the that the US mobile infrastructure is still pretty much in the stone age, I'd love to know why SMS hasn't taken off in the States? Pretty much everyone else around the world thinks they're a good idea. What's different about American culture that you guys don't?
I never, ever want to see anything like the registry pop up in Linux. One central place for storing config info may sound nice, but what happens when it gets corrupt? Bye bye to everything!
Err, there is a central place for storing configuration files, its called/etc
From what I have read, the HURD tries to go to a direction that no other O/S has gone before.
Perhaps, but I remember hearing about HURD which was "coming real soon now" around the same time I first heard about Linux back in early 1992. They may have been going in a direction that other OS has gone, but they've being doing that for such a long time you shouldn't be able to see them anymore.
Instead, (some) people are still talking about HURD as the "next big thing". They were doing that back in 1992, HURD has been the vapourware of the Open Source community, only briefly eclipsed in recent years by Mozilla, who at least have finally turned out a decent product. A lot of people have lost faith that the HURD project will ever produce something you can run on a production system.
This is good because innovation is needed, but they should get rid of all the notions of the past: processes, filesystems, users, groups etc. all these things are for O/Ses of the past.
Perhaps, but they're notions that have served us well, we're used to them and we can actually get some work done by using them. I've yet to find anybody with a decent user interface design for an OS based (purely) on an object model. The underlying technologies may be OO, but as soon as you start talking about the user interface the old metaphors start to show up again.
I'll be conviced that (pure) object model operating systems are a good thing when someone can give me an idea of what sort interface you're actually going to present to the user, and when that interface actually offers significant advantages over the more mature systems we have now.
...being able to move the mouse pointer between computers (assuming these are separate computers, not just multiple monitors); I assume this indicates some sort of network-transpart clipboard (and that the user is signed onto both computers). cool, that.
You can already do this in X (and have been able to do it since 1996) using X2X, see the Freshmeat page for details, basically it means that the keyboard and mouse on one machine can be used to control a bunch of others (just so long as you can see the screens). No big deal...
In the UK this is turns out not to be the case, for instance all those "you can't sue us if you die" disclaimers that people make you sign for just about everything you do in the States are totally invalid in the UK. They have no weight in law, if people are negligent there isn't any way for them to get you to sign your legal rights to sue them away.
Have anyone noticed this buzzword used by every Microsoft lobbying effort after 9/11 just to trying to give the probably fake impression of Microsoft being "patriotic"?
Its not exactly going down well in the UK and (the rest of) Europe....
Have a look at the recent BBC coverage of Microsoft; here,
here,
here,
here and
here for instance.
India is planning an unmanned mission to Luna in 2007. The US, Russia (when it was the USSR), and Japan are the only nations to have done so, or so they say. For some reason, I thought that ESA, the European Space Agency, had sent one also.
Nope, the ESA has never had a lunar mission, however we are currently building SMART-1 which is due to be ready in early 2003, and will be launched as an Ariane-5 auxiliary payload.
SMART-1 is actually a testbed platform for the ESA's Solar Electric Propulsion system, with the primary goal being to test this and other technologies that will be useful for deep space missions, however the craft will carry both X-ray and IR spectrometers for inspecting the lunar surface.
Al.
Re:Design Problem? Here's the design problem:
on
Going Up?
·
· Score: 3, Informative
I don't understand these people who think you can build an elevator into space. Can't anybody understand that you cannot just "tie" a cable from Earth to something in orbit in space?
I really hope you're deliberately trolling, but just in case...
The only possibility of maintaining an actual elevator cable is if it is hooked onto something in geosyncronous orbit with the Earth. The only problem there is that the object would have to be 40,000 miles away from the Earth to maintain constant orbit with a fixed position on Earth. Good luck.
Err, yes? Thats exactly what people are proposing, in fact people have been proposing this for many years. See this NASA Summary for details for the current ideas. You'll notice that they specifically say that the elevator will be to geo-stationary Earth orbit (GEO) in the first sentence.
Most shops in europe sell players regionfree, and most are also willing to chip them. Over here, it is completly legal. Therefore, you should be able to find a LOT of british sites that do it. A search on google will give you hundreds of them...
Heck, you can buy multi-region DVD players that will do DTS, PAL/NTSC out of the box for less that £100 (thats about $150) on amazon.co.uk, you don't even have to leave your computer.
Experience suggests that the scans will be of mediocre quality (missing some pages, missing parts of other pages, frequently having insufficient contrast to be legible, and losing any colour or greyscale information present in the originals).
Really? I think most astronomers would disagree, the ADS has scanned a significant fraction of the back catlogue of astronomical journals. While most of the journals now publish electronic editions as well as paper most of these only go back to the mid to late 90's, online access to the back catalogue is amazingly useful for research. A full list of back catalogue of journals they provide access to is here as you can see they've scanned some of them all the way back to the 1880's or 90's. Scan quality is uniformly good, I've yet to find a badly scanned journal article and I' use this service every day.
That publicly held companies have a legal obligation to "benefit sotckholders" or "maximize profits." Is there really a legal basis for this? Are there civil statutes that say companies must do whatever it takes to make money for stockholders? Or is this legal obligation based in contract law where the stockholder will/can sue if the company makes decisions that appear to adversley affect them?
IANAL, however my understanding is, yes, stock holders can sue the Board/CEO if they believe that that they are not working to maximise profits, and therefore stockholder value.
This is interesting. I may be the only one, but I have recently received a lot of spam (via SMS messaging and email) on my cell phone.
I don't know where you're located, but this is actually getting fairly common in the UK. Rest easy that its costing the 5p (or more) per message and so (economically) it's probably going to go away eventually...
In the UK this is actually now fairly common, you see lots of Tesco Online vans running around if you're out and about during their "peak" delivery hours (just after people get home from work).
...it is becoming harder for users of Microsoft-free systems and browsers to view the web.
Not any bits of the web I actually want to use, I haven't come across anything I want to see that isn't still Netscape 4.x compatible, let alone compatible with Mozilla 1.0. As far as I'm concerned the web is still working just fine...
The SI second is defined as the period of time that it takes a specific number of cesium isotope radiation emissions to occur such that it is as close to a mean ABT second (1/86400 day) as feasible given the variance of the earth's rotation. To redefine the SI second to be equal to a MT second would mean redefining it to be equal to whatever number of cesium-133 emissions are close to 10-5 mean day given variation.
Err, you just can't do this, whatever else you do to time keeping you cannot get rid of the SI second. The amount of science and applied maths that this would affect would be mind blowing, you'd be fiddling around adding constants into equations almost everywhere. The entire concept sends a shiver down my spine...
Okay, I've got karma to burn, may as well use it...
This is a great time for me to mention this. Linux will NEVER make the inroads it wants to user desktops until it becomes more like Windows. That's right! I said it! I've been in computer tech support for the last 5 years at an advanced level...
Frankly, who cares? Why should I care if Linux never makes major inroads into the user desktop? Why does this affect me? I'm still going to be writing software for UNIX, and probably whatever replaces UNIX on the academic desktop, which definately won't be Windows for a whole bunch of reasons, mostly money... I'm currently nudging my boss to buy me a MacOS X box to see how easy our legacy software is going to port to that, although my guess our current Linux port won't need that much tweaking to run on MacOS X. On the flip side most of our next generation of software is in cross platform languages (Perl and Java) so its a declining problem.
Why should I care if some random joe off the street can run the stuff I crank out? What is this obsession everyone seems to have these days about getting Linux adopted by the mass market?
It took several phone calls -- escalating to the point of calling a Lindows.com executive, an option not available to non-journalists -- before I discovered LindowsOS doesn't yet have a feature for setting the refresh rate. Instead, the company sent me complicated and potentially risky instructions for changing the refresh rate by entering obscure Unix commands such as "xf86cfg."
I did finally manage to set a high refresh rate, but only after taking a technical step that's equivalent to jumping off a cliff without knowing for sure if there's a safety net below.
I mean come on guys xf86cfg isn't exactly rocket science, it no harder to use than playing with the control panel in Windows.
I'll be sure to remind you of that should they have "good reason" to think you are a "bad guy" and arrest you when you come back from your vacation in Egypt.
I travel to Egypt fairly often on holiday to dive in the Red Sea, I also travel to the States a couple of times a year. Considering the rising tide of paranoia in the States I'm rather glad I've just gotten a new passport, and the only visa stamps in there so far are for Canada.
Fine, okay, Spetember the 11th was a tragedy, lots of people died. But your governments violation of peoples civil liberties is (to me) far scarier. Depressingly the UK government is also using the events of last September to widen their powers at the expense of our privacy (the RIPA stands out as a glaring example, but there are others).
Slashdot Mirror
Looking at the summary of changes, I suspect that they've finally messed up the things that made FORTRAN (or at least F77) great. The addition of pointers stands an excellent chance of rendering code un-optimizable, and I fear that adding OO features is an even bigger mistake.
FORTRAN went OO during with Fortran90 and pointers were added with Fortran95, although for anyone used to C pointers, Fortran95 pointers are (erm) pretty wierd...
Al.Back in my grad school days (5 yrs ago) everyone who had supercomputer time was writing in FORTRAN.
Amoungst physicists and astronomers, anyone with supercomputer time today is still writing in FORTRAN. I don't see that changing in the near future. Graduate students are starting to write in different languages, but to talk to their superivsors (the guys with large existing codebases) they have to be able to talk and write FORTRAN.
Most of the time its not worth their time to port a couple of hundred thousand lines of modelling code to some other language before they can add layers on top, its unlikely it'll ever be worth anyones time to do somethnig like that.
A lot of the Java, Perl/Tk and Python/Tkinter stuff I write these days is sitting ontop of a whole bunch of FORTRAN that it would take the next twenty years to rewrite. If it works, whats the point of doing it all again? Thats no fun...
I think language interoperability is becoming more and more important now the industry is (sort of) maturing. This is where code reuse in the real world is going to come from, rather than the current fads in code design which keep on promising more code reuse.
Al.Last time I checked, you could install SuSE that way...
...and RedHat, you just need the right boot floppy.
Al.Good luck with that. It seems to me this "feature", just like text messaging, is being pushed by a culture that just doesn't understand Americans. Just because SMS is all the rage in Norway and the Japan...
SMS has has a major cultural impact in the UK, over 750 million messages are now sent every month in the UK. There are lots of circumstances where sending a quick text message is alot more convenient that trying to call someone who may be busy doing other things. Like email the asynchronous nature of text messaging has advantages at times.
Appart from the fact the that the US mobile infrastructure is still pretty much in the stone age, I'd love to know why SMS hasn't taken off in the States? Pretty much everyone else around the world thinks they're a good idea. What's different about American culture that you guys don't?
Al.I never, ever want to see anything like the registry pop up in Linux. One central place for storing config info may sound nice, but what happens when it gets corrupt? Bye bye to everything!
Err, there is a central place for storing configuration files, its called /etc
Al.Which of course leads us to the question, what is an SAT?
Al.Now this post comes along and it becomes obvious to me that POSIX is a big deal in the UNIX-like-OS world.
Pretty much, if it doesn't implement POSIX, it isn't really UNIX.
Did MS just screw up their implementation or is it something potentially useful that nobody happens to use?
They screwed up...
Al.From what I have read, the HURD tries to go to a direction that no other O/S has gone before.
Perhaps, but I remember hearing about HURD which was "coming real soon now" around the same time I first heard about Linux back in early 1992. They may have been going in a direction that other OS has gone, but they've being doing that for such a long time you shouldn't be able to see them anymore.
Instead, (some) people are still talking about HURD as the "next big thing". They were doing that back in 1992, HURD has been the vapourware of the Open Source community, only briefly eclipsed in recent years by Mozilla, who at least have finally turned out a decent product. A lot of people have lost faith that the HURD project will ever produce something you can run on a production system.
This is good because innovation is needed, but they should get rid of all the notions of the past: processes, filesystems, users, groups etc. all these things are for O/Ses of the past.
Perhaps, but they're notions that have served us well, we're used to them and we can actually get some work done by using them. I've yet to find anybody with a decent user interface design for an OS based (purely) on an object model. The underlying technologies may be OO, but as soon as you start talking about the user interface the old metaphors start to show up again.
I'll be conviced that (pure) object model operating systems are a good thing when someone can give me an idea of what sort interface you're actually going to present to the user, and when that interface actually offers significant advantages over the more mature systems we have now.
Al.Amazon.com could fix this problem easily by having the affiliate enter their URL when they sign up...
You do enter the URL of your site when you sign up for affiliate status.
...check the URL against the ID and see if they're the legit owner of that site.
I'd always assumed that they did this already, obviously not...
Al....being able to move the mouse pointer between computers (assuming these are separate computers, not just multiple monitors); I assume this indicates some sort of network-transpart clipboard (and that the user is signed onto both computers). cool, that.
You can already do this in X (and have been able to do it since 1996) using X2X, see the Freshmeat page for details, basically it means that the keyboard and mouse on one machine can be used to control a bunch of others (just so long as you can see the screens). No big deal...
Al.Of course, you can sign away legal rights...
In the UK this is turns out not to be the case, for instance all those "you can't sue us if you die" disclaimers that people make you sign for just about everything you do in the States are totally invalid in the UK. They have no weight in law, if people are negligent there isn't any way for them to get you to sign your legal rights to sue them away.
Al.Have anyone noticed this buzzword used by every Microsoft lobbying effort after 9/11 just to trying to give the probably fake impression of Microsoft being "patriotic"?
Its not exactly going down well in the UK and (the rest of) Europe....
Have a look at the recent BBC coverage of Microsoft; here, here, here, here and here for instance.
Compare this with recent Linux coverage; here, here, here, here, here and here.
Is it just me, or is there a strong pro-Linux, anti-MS bias creeping in there?
Al.India is planning an unmanned mission to Luna in 2007. The US, Russia (when it was the USSR), and Japan are the only nations to have done so, or so they say. For some reason, I thought that ESA, the European Space Agency, had sent one also.
Nope, the ESA has never had a lunar mission, however we are currently building SMART-1 which is due to be ready in early 2003, and will be launched as an Ariane-5 auxiliary payload.
SMART-1 is actually a testbed platform for the ESA's Solar Electric Propulsion system, with the primary goal being to test this and other technologies that will be useful for deep space missions, however the craft will carry both X-ray and IR spectrometers for inspecting the lunar surface.
Al.I don't understand these people who think you can build an elevator into space. Can't anybody understand that you cannot just "tie" a cable from Earth to something in orbit in space?
I really hope you're deliberately trolling, but just in case...
The only possibility of maintaining an actual elevator cable is if it is hooked onto something in geosyncronous orbit with the Earth. The only problem there is that the object would have to be 40,000 miles away from the Earth to maintain constant orbit with a fixed position on Earth. Good luck.
Err, yes? Thats exactly what people are proposing, in fact people have been proposing this for many years. See this NASA Summary for details for the current ideas. You'll notice that they specifically say that the elevator will be to geo-stationary Earth orbit (GEO) in the first sentence.
Al.Most shops in europe sell players regionfree, and most are also willing to chip them. Over here, it is completly legal. Therefore, you should be able to find a LOT of british sites that do it. A search on google will give you hundreds of them...
Heck, you can buy multi-region DVD players that will do DTS, PAL/NTSC out of the box for less that £100 (thats about $150) on amazon.co.uk, you don't even have to leave your computer.
Al.Experience suggests that the scans will be of mediocre quality (missing some pages, missing parts of other pages, frequently having insufficient contrast to be legible, and losing any colour or greyscale information present in the originals).
Really? I think most astronomers would disagree, the ADS has scanned a significant fraction of the back catlogue of astronomical journals. While most of the journals now publish electronic editions as well as paper most of these only go back to the mid to late 90's, online access to the back catalogue is amazingly useful for research. A full list of back catalogue of journals they provide access to is here as you can see they've scanned some of them all the way back to the 1880's or 90's. Scan quality is uniformly good, I've yet to find a badly scanned journal article and I' use this service every day.
Al.That publicly held companies have a legal obligation to "benefit sotckholders" or "maximize profits." Is there really a legal basis for this? Are there civil statutes that say companies must do whatever it takes to make money for stockholders? Or is this legal obligation based in contract law where the stockholder will/can sue if the company makes decisions that appear to adversley affect them?
IANAL, however my understanding is, yes, stock holders can sue the Board/CEO if they believe that that they are not working to maximise profits, and therefore stockholder value.
Al.In the UK all the networks are "sender pays" so it doesn't really come up...
Al.This is interesting. I may be the only one, but I have recently received a lot of spam (via SMS messaging and email) on my cell phone.
I don't know where you're located, but this is actually getting fairly common in the UK. Rest easy that its costing the 5p (or more) per message and so (economically) it's probably going to go away eventually...
Al.You actually buy your groceries online?
In the UK this is actually now fairly common, you see lots of Tesco Online vans running around if you're out and about during their "peak" delivery hours (just after people get home from work).
Al.Not any bits of the web I actually want to use, I haven't come across anything I want to see that isn't still Netscape 4.x compatible, let alone compatible with Mozilla 1.0. As far as I'm concerned the web is still working just fine...
Al.The SI second is defined as the period of time that it takes a specific number of cesium isotope radiation emissions to occur such that it is as close to a mean ABT second (1/86400 day) as feasible given the variance of the earth's rotation. To redefine the SI second to be equal to a MT second would mean redefining it to be equal to whatever number of cesium-133 emissions are close to 10-5 mean day given variation.
Err, you just can't do this, whatever else you do to time keeping you cannot get rid of the SI second. The amount of science and applied maths that this would affect would be mind blowing, you'd be fiddling around adding constants into equations almost everywhere. The entire concept sends a shiver down my spine...
Al.Okay, I've got karma to burn, may as well use it...
This is a great time for me to mention this. Linux will NEVER make the inroads it wants to user desktops until it becomes more like Windows. That's right! I said it! I've been in computer tech support for the last 5 years at an advanced level...
Frankly, who cares? Why should I care if Linux never makes major inroads into the user desktop? Why does this affect me? I'm still going to be writing software for UNIX, and probably whatever replaces UNIX on the academic desktop, which definately won't be Windows for a whole bunch of reasons, mostly money... I'm currently nudging my boss to buy me a MacOS X box to see how easy our legacy software is going to port to that, although my guess our current Linux port won't need that much tweaking to run on MacOS X. On the flip side most of our next generation of software is in cross platform languages (Perl and Java) so its a declining problem.
Why should I care if some random joe off the street can run the stuff I crank out? What is this obsession everyone seems to have these days about getting Linux adopted by the mass market?
Al.I was a bit shocked by...
I mean come on guys xf86cfg isn't exactly rocket science, it no harder to use than playing with the control panel in Windows.
Al.I'll be sure to remind you of that should they have "good reason" to think you are a "bad guy" and arrest you when you come back from your vacation in Egypt.
I travel to Egypt fairly often on holiday to dive in the Red Sea, I also travel to the States a couple of times a year. Considering the rising tide of paranoia in the States I'm rather glad I've just gotten a new passport, and the only visa stamps in there so far are for Canada.
Fine, okay, Spetember the 11th was a tragedy, lots of people died. But your governments violation of peoples civil liberties is (to me) far scarier. Depressingly the UK government is also using the events of last September to widen their powers at the expense of our privacy (the RIPA stands out as a glaring example, but there are others).
Al.