They're second only to particle accelerators as an example of expensive blue-sky research.
Better for black-sky research, wouldn't you say?
As for revenue stream, what about all the research grants? Don't astronomers already pay for telescope time from their grant money, to defray operating costs?
Besides, Palomar, Wilson and a number of other observatories and telescopes were originally built by private institutions (money for Palomar's 200-inch telescope came from the Rockefeller Foundation, for instance) and were originally operated by them. Don't know if any are still non-government (assuming we're paranoid enough to call CalTech a government institution), but there's history for private endowment of big astronomy.
Bravo for Brazil! Unfortunately (?) no-one's life is at stake over one-click ordering (or any of the other stupid algorithm and business method patents). It's unlikely this will have any impact on US/European patent law.
All of this explains why my last two clients (think big, big organizations) have
turfed Java and standardized on COM+/VB.
Is this why I'm always getting ODBC errors at Microsoft and other sites, when I try to send in comments? Thanks for explaining how 10x savings makes unreliable systems desirable!
The issue of COM hopefully won't be a problem. C# claims it will make COM programming easier, but it's not the only way;
Machine code doesn't remember which language it used to be.
Don't be too sure -- p-code can remember what language it used to be. I wouldn't put it past MS to phase out the machine-code components in favor of.NET components that run under the VM. This for "cross-platform support" of Itanium and/or other processors that might follow the Pentium as the Windows processor.
Can I take GPL'd code and write a couple wrapper functions and release that piece of code
and call it a plugin to my main application which would not be released under the GPL?
If your code does not perform a significant function, it is quite unlikely that it will compete well with the free GPL software that it "wraps".
On the other hand, if you want to turn a GPL program into a photoshop plugin with a couple of wrapper functions, go for it! That's a legitimate use of the GPL -- to make a new facility using shared free code, even if it's for use with non-free code. Because eventually, someone will plug your photoshop plugin into the GIMP....
How many times do we have to go through this. The GPL states that you don't have to give away the code unless the person already has
the binary, it also stats they they don't have to be available together.
Until you get it right, apparently:-)
The GPL states that you must OFFER the source code with the distribution. Your offer may refer to another method of actually obtaining the source code, but the consumer must be appraised of his or her rights at the time of distribution. (term 3b)
In addition, it must be accompanied by a copy of the GPL. (term 1 included by reference in term 3)
Therefore if the operating system comes to you without a copy of the GPL, or without a written offer to provide the source code or notice of where the source code can be obtained, then one of the following two statements must be true:
1) It's not Linux-based, and therefore although it may be distributed legally it's advertised fraudulently.
2) It is Linux-based, but it's being distributed illegally, as the copy of the GPL and offer of source code is required in order to comply with the terms of the GPL, and you may not distribute a derivative work of Linux unless you distribute it according to the terms of the GPL.
Hewlett-Packard Company today announced it has been awarded a key patent that could remove a major obstacle
to making molecular-scale computing a reality.
I can see RMS exploding with rage right now if he's reading this sentence. How exactly did the patent remove the obstacle?
No, it's a misprint. It removes the major obstacle to making molecular-scale computing a royalty.
Personally I'd rather just see some kind of top 10 laws that everybody should learn in school from a very early age.
Maybe sum them up into even less like: 1. Don't kill 2. Don't lie 3. Don't steal
Oh you mean the Ten Commandments?
The ten commandments are proper as religious rules (God's law), but the 1st four commandments are purely religious (and thus not proper subjects for secular law, unless your society represses other religeons).
The three (kill, lie, steal -- not) were suggested, and are arguably required for a functioning society.
Two of the remaining three are versions of lying (adultery -- making your vows false, and false witness, i.e. lying under oath). The last one is an injunction against covetousness, and while that's perfectly good advice, it's not something I want Big Brother to monitor (acting on it is stealing, but wanting it is just thinking).
You might get a dead-meat version of your body there, but chances are it will take too long to build for it to be resessa.. recesa.. resussi.. brought back to life:-)
Yeah, but now when we fix code we'll have a reasonable expectation that the compiler feature that's complaining works (less guesswork as to whose bug it is).
The difficulty of LCD is that you have a transparent substrate, a charge grid below the crystal, a layer of liquid crystal, a ground grid above the crystal and a color grid that changes the white-light display to RGB. Oh, and the glass plates (substrate?) that sandwitch all of this, and the diffuser and backlight. But the big deal is that the glass has to be extremely flat and parallel, and the grids have to all line up, and there can be no closed cells in any of the grids. Fail any one of these and the screen has a dead region (anything visible is death).
Contrast OLED -- light is provided by the element, so the back panel can be opaque (read: not as fragile). The color is in the element too, so one layer of alignment goes. Finally, with proper design the back electrode can provide ground as well, eliminating another alignment headache, and allowing some play (as long as both + and - make it into the pixel, it lights up, and since there's no grid on top you see it whether it's spot-on or not).
So reduction in precision (or better performance for equal precision) and the fact that everything can be layered on a single glass sheet (rather than 2 which must be aligned and mounted parallel) makes the error rate potentially lower, as soon as the new equipment reaches the same level of debugg-ed-ness as the LCD equipment.
Of course, there's still the problem of dye fading and aligning 2 or 3 layers, and that we now want to have larger sheets for larger displays, and that we want to go to plastic to make more durable displays, but hey, researchers gotta eat...
Interesting indeed. If a computer program is not an expressive publication it should not be subject to copyright. If so, DMCA does not apply to computer programs. Patent might though, if it's a "device".
That could have a chilling effect on the information economy, but it could also have some positive benefits:
- Much of the Art is published already, prior art abounds for a lot of the field.
- Longer cycle times (due to the broader ownership rights of patents) mean more time for debugging and, potentially, application of stronger liabilty standards. Liability for an expression is much more limited than liability for a defective device.
- Therefore, better code and better remedies -- look out Bill:-)
Well, not really -- the attack is on the DMCA's protection of access restrictions. Decryption is now illegal in the US, and copying is irrelevant to the suit.
The region-coding stuff is a restraint-of-trade issue, and perhaps an unlawful consipiracy to reduce competition. Someone else should take it up but again it's irrelevant to this case.
Recipes can be propietary, expensive, and for good reason. Good luck trying to get that famous chef to tell the
secret ingredient.
Not so -- great chefs often give away their recipies. Emeril tells you how to make essence, Paul Proudhomme publishes the Turducken recipie, but McDonalds jealously guards the Secret Sauce. The difference is in the experience and skill of the great chef to improvise, adjust and flow with differences in the raw product, or the patience and precision that brings it together (which is what makes them great).
Sure, there are reasons to be proprietary, but there's a model where you pay the chef for preparing the food well, and there's a model where you pay for the formula regardless of the quality of the product. I prefer the former:-), and I agree that the latter is only sometimes worth while.
I can limit resource usage (read, CPU & memory) and so forth, and pin access plans (to a degree) for the production app, but there's still a couple of bad dynamics going on:
1- The clients get up to their resource caps.
2- There's dozens or hundreds of them.
3- You try logging your boss off after 20 queries, like it or not:-).
If you let the user in with user-defined SQL (like in Access) you're unlikely to get away with limiting his view to small/simple tables, and you can't stop him from trying to join whatever he sees. There are GUI tools for that (Oracle Discoverer leaps to mind) but they cost money, probably as much as using Access on a "reporting" server, so it's a matter of preference. The real problem with them is that the boss is already tool-focussed, and is unlikely to accept learning a new tool.
Better to make offline Access or Excel presentations of the data. They can be quite interesting and sufficient in many cases, and permit the boss to (ab)use his own computer rather than mine. I find bosses usually understand spreadsheets better than databases, but there are some functional users that get into Access.
Not congruent with my experience, but we haven't tried it again with Access 95 or later. Of course, I'm limited in my experience to using Access with Oracle. YMMV with another DBMS (especially SQL Server).
Ok, I'll bite, wheres your point'n'click query solution?
As a DBA, you shouldn't provide one that uses the database. I'm serious -- a production database has no business running SQL that hasn't been vetted by the DBA and run in a test database. The wrong kind of query can bring production to a halt for a manager's whim, and get you in the hot seat to boot.
That said, it's probably better to provide Microsoft Query and Excel rather than Access for "adh hoc query", if your needs are sufficiently simple. Unless you know how to enter "passthrough" queries, Access insists on reading entire tables and doing the filter/join itself. There is no better formula for poor performance.
If data need is not simple enough for MS Query, you need to design a data mart and produce preloaded spreadsheets, 'cause the boss won't understand how to get the data to come out. Give the boss a spreadsheet he understands and he'll be 10 times happier than with a point-and-click he doesn't.
I think it's quite clear that it would help Ogg Vorbis if, to pick and example at random, Microsoft stuck a closed-source.OGG
replay codec into Windows.
Except that if you use a non-Microsoft encoder, their codec will only play back a flattened monaural version of the sound:-). That way it's "supported".
Build a bridge. I know the ferry is leaky, and they won't let me bring some of my stuff without paying exorbitant fees for use of their luggage rack (not required for passage, unless you want to make your own luggage). Oh, and I can't get off the ferry at the other side, just "upgrade" to their rickety bus.
Slashdot Mirror
Better for black-sky research, wouldn't you say?
As for revenue stream, what about all the research grants? Don't astronomers already pay for telescope time from their grant money, to defray operating costs?
Besides, Palomar, Wilson and a number of other observatories and telescopes were originally built by private institutions (money for Palomar's 200-inch telescope came from the Rockefeller Foundation, for instance) and were originally operated by them. Don't know if any are still non-government (assuming we're paranoid enough to call CalTech a government institution), but there's history for private endowment of big astronomy.
Bravo for Brazil! Unfortunately (?) no-one's life is at stake over one-click ordering (or any of the other stupid algorithm and business method patents). It's unlikely this will have any impact on US/European patent law.
What I want to know is if you perk up when you close the <tired/> tag.
Hey, it works!
Is this why I'm always getting ODBC errors at Microsoft and other sites, when I try to send in comments? Thanks for explaining how 10x savings makes unreliable systems desirable!
You'll need a big ol' sword if you talk like that in a bar.
Don't be too sure -- p-code can remember what language it used to be. I wouldn't put it past MS to phase out the machine-code components in favor of .NET components that run under the VM. This for "cross-platform support" of Itanium and/or other processors that might follow the Pentium as the Windows processor.
If your code does not perform a significant function, it is quite unlikely that it will compete well with the free GPL software that it "wraps".
On the other hand, if you want to turn a GPL program into a photoshop plugin with a couple of wrapper functions, go for it! That's a legitimate use of the GPL -- to make a new facility using shared free code, even if it's for use with non-free code. Because eventually, someone will plug your photoshop plugin into the GIMP....
Until you get it right, apparently :-)
The GPL states that you must OFFER the source code with the distribution. Your offer may refer to another method of actually obtaining the source code, but the consumer must be appraised of his or her rights at the time of distribution. (term 3b)
In addition, it must be accompanied by a copy of the GPL. (term 1 included by reference in term 3)
Therefore if the operating system comes to you without a copy of the GPL, or without a written offer to provide the source code or notice of where the source code can be obtained, then one of the following two statements must be true:
1) It's not Linux-based, and therefore although it may be distributed legally it's advertised fraudulently.
2) It is Linux-based, but it's being distributed illegally, as the copy of the GPL and offer of source code is required in order to comply with the terms of the GPL, and you may not distribute a derivative work of Linux unless you distribute it according to the terms of the GPL.
I can see RMS exploding with rage right now if he's reading this sentence. How exactly did the patent remove the obstacle?
No, it's a misprint. It removes the major obstacle to making molecular-scale computing a royalty .
Oh you mean the Ten Commandments?
The ten commandments are proper as religious rules (God's law), but the 1st four commandments are purely religious (and thus not proper subjects for secular law, unless your society represses other religeons).
The three (kill, lie, steal -- not) were suggested, and are arguably required for a functioning society.
Two of the remaining three are versions of lying (adultery -- making your vows false, and false witness, i.e. lying under oath). The last one is an injunction against covetousness, and while that's perfectly good advice, it's not something I want Big Brother to monitor (acting on it is stealing, but wanting it is just thinking).
Although, if the IRS is after you...
Yeah, but now when we fix code we'll have a reasonable expectation that the compiler feature that's complaining works (less guesswork as to whose bug it is).
The difficulty of LCD is that you have a transparent substrate, a charge grid below the crystal, a layer of liquid crystal, a ground grid above the crystal and a color grid that changes the white-light display to RGB. Oh, and the glass plates (substrate?) that sandwitch all of this, and the diffuser and backlight. But the big deal is that the glass has to be extremely flat and parallel, and the grids have to all line up, and there can be no closed cells in any of the grids. Fail any one of these and the screen has a dead region (anything visible is death).
Contrast OLED -- light is provided by the element, so the back panel can be opaque (read: not as fragile). The color is in the element too, so one layer of alignment goes. Finally, with proper design the back electrode can provide ground as well, eliminating another alignment headache, and allowing some play (as long as both + and - make it into the pixel, it lights up, and since there's no grid on top you see it whether it's spot-on or not).
So reduction in precision (or better performance for equal precision) and the fact that everything can be layered on a single glass sheet (rather than 2 which must be aligned and mounted parallel) makes the error rate potentially lower, as soon as the new equipment reaches the same level of debugg-ed-ness as the LCD equipment.
Of course, there's still the problem of dye fading and aligning 2 or 3 layers, and that we now want to have larger sheets for larger displays, and that we want to go to plastic to make more durable displays, but hey, researchers gotta eat...
Oh yeah -- I got interrupted composing and forgot the best part -- patents don't last as long as copyright.
That could have a chilling effect on the information economy, but it could also have some positive benefits:
- Much of the Art is published already, prior art abounds for a lot of the field.
- Longer cycle times (due to the broader ownership rights of patents) mean more time for debugging and, potentially, application of stronger liabilty standards. Liability for an expression is much more limited than liability for a defective device.
- Therefore, better code and better remedies -- look out Bill :-)
Yeah, it's utopian, but it's fun.
The region-coding stuff is a restraint-of-trade issue, and perhaps an unlawful consipiracy to reduce competition. Someone else should take it up but again it's irrelevant to this case.
Not so -- great chefs often give away their recipies. Emeril tells you how to make essence, Paul Proudhomme publishes the Turducken recipie, but McDonalds jealously guards the Secret Sauce. The difference is in the experience and skill of the great chef to improvise, adjust and flow with differences in the raw product, or the patience and precision that brings it together (which is what makes them great).
Sure, there are reasons to be proprietary, but there's a model where you pay the chef for preparing the food well, and there's a model where you pay for the formula regardless of the quality of the product. I prefer the former :-), and I agree that the latter is only sometimes worth while.
Who says it had to be an intelligent species?
Actually I hope it's not a 4/1, but time will tell..
The number you have dialed is imaginary -- please rotate your telephone 90 degrees and try again.
1- The clients get up to their resource caps.
2- There's dozens or hundreds of them.
3- You try logging your boss off after 20 queries, like it or not :-).
If you let the user in with user-defined SQL (like in Access) you're unlikely to get away with limiting his view to small/simple tables, and you can't stop him from trying to join whatever he sees. There are GUI tools for that (Oracle Discoverer leaps to mind) but they cost money, probably as much as using Access on a "reporting" server, so it's a matter of preference. The real problem with them is that the boss is already tool-focussed, and is unlikely to accept learning a new tool.
Better to make offline Access or Excel presentations of the data. They can be quite interesting and sufficient in many cases, and permit the boss to (ab)use his own computer rather than mine. I find bosses usually understand spreadsheets better than databases, but there are some functional users that get into Access.
Not congruent with my experience, but we haven't tried it again with Access 95 or later. Of course, I'm limited in my experience to using Access with Oracle. YMMV with another DBMS (especially SQL Server).
As a DBA, you shouldn't provide one that uses the database. I'm serious -- a production database has no business running SQL that hasn't been vetted by the DBA and run in a test database. The wrong kind of query can bring production to a halt for a manager's whim, and get you in the hot seat to boot.
That said, it's probably better to provide Microsoft Query and Excel rather than Access for "adh hoc query", if your needs are sufficiently simple. Unless you know how to enter "passthrough" queries, Access insists on reading entire tables and doing the filter/join itself. There is no better formula for poor performance.
If data need is not simple enough for MS Query, you need to design a data mart and produce preloaded spreadsheets, 'cause the boss won't understand how to get the data to come out. Give the boss a spreadsheet he understands and he'll be 10 times happier than with a point-and-click he doesn't.
Yes, I am a DBA...
What the heck, get both -- then you can program microcode...
Except that if you use a non-Microsoft encoder, their codec will only play back a flattened monaural version of the sound :-). That way it's "supported".
It all depends on the details ;-)