i.e. program it for the deadline and then draw lots of sticks and boxes to justify your design
Um, not to sound harsh, but it sounds pretty darned idiotic to justify a design AFTER you finish implementing it. The point of the design is so that it can be reviewed to make sure it it meets the requirements (otherwise, you won't even get funding for the implementation phase) and to keep programmers on the same page during implementation so that the project can be effectively divided amonst a team of programmers without the communications overhead getting out of control.
The point of using 'patterns' is so that people who design systems for a living can speak the same lingo during design, independant of the language being used for implementation. Same with UML. Their complexity characteristics are already well understood and the metaphors being used ("observer," "factory", etc) are also high-level enough so that the overall design of a large component can be held-in-the-head all at once. (you know, that 7+/-2 thing?).
If you design large projects using an object oriented approach (which doesn't exclude using C or even Perl), I'd think that you'd find patterns essential. And yes, people had been doing OOD without using "patterns" as such for quite a while; but more often than not, the same general ideas were still being used, except that any metaphors used to explain them were basically ad-hoc.
QT and GTK both lack good software for GUI building and prototyping. When you're trying out different looks for an application, it's REALLY annoying to have write code to have to specify the look'n'feel of an app when it makes much more sense to use your visual intuition to 'draw' the interface.
So what your saying is that the spirit of the speech not the speech itself is interesting and makes it worth while? Really? Well in that case I guess I can passionately just spout jibberish to people talking about magic and little gnomes and such and you will buy it right? After all it's tought provoking.
Do you read much fiction? It doesn't need to be accurate in any sense of the word, and yet it's still valuable to consider those "what if" worlds. And of course, no one "buys it" or actually believe it's 'true', and yet Orwell's "1984" is probably the largest reason why people fear government propaganda and are fanatical about the evils of software the track people. Strange, that something completely made up could have such an enormous effect just because it provoked people to think about it...
I completely disagreed with the original poster's view -- I think it's a fairly deluded view of things. But it's still interesting. People read Descartes over and over again, it's taught to students, and yet due to the numerous flaws in his arguments there are probably no real Cartesians left in this world. But again, especially in the context of his times, it's still interesting.
By the way what you are describing is what made men like Hitler and Lenin so popular. Tell the people what they want to hear (because it sounds good) and lie, lie, lie. Am I the only person who thinks so?
And, just as much as Hitler and Lenin, Descartes sought to tell people exactly what they wanted to hear (that they have an immortal soul, that God exists, that the Skeptics were all wrong). So yes, I agree with you that targeted rhetoric like that can be extremely manipulative. But, if you are the type of person who is inclined (provoked) to think about it, you don't allow yourself to be manipulated so easily. Hiding your head in the sand like an ostrich doesn't really help.
...basic conventions that have stood the test of time are suddently coming under fire...
You're not really saying that people shouldn't challenge convention, right? If so, we'd still believe the world was flat, that the universe revolves around the earth, etc...
And incidentally, I enjoy Tolkien just as much as the next person -- if you are good at spouting jibberish about magic and little gnomes, go ahead and write to your heart's content (but probably not in this thread, for obvious reasons).
This is a forum for conversation, not a magazine or journal. The post is highly rated not because it's necessarily true or even historically accurate but because it's thought provoking -- much more so than the average Slashdot post. If you find that you are provoked to disagree with the content of the article, why not point out where it goes wrong?
The reason for security problems with NFS on Linux is because NFS is inherently an insecure protocol for file sharing. Add-ons to make NFS secure, like using Kerberos for authentication, are NOT part of the NFS standard. In general, you can always spoof UIDs to get around file perms on any standard NFS implementation.
Oh, and the main reason to use Samba rather than an alternative NFS or Coda is because files are typically served to Windows clients, which automatically support SMB.
If they want to throw more people onto a late project, don't just tell them it won't work. Tell them *why*, quote Fred Brookes, *reason* with them.
This statement definitely goes along with somoene else's observation that a company's technical staff doesn't get enough respect to be promoted to managerial positions. Probably the MAIN reason people don't reason with their managers nor get promoted to management positions (even if they really do know what's up) is because so many of them are incapable of communicating effectively. Even if you know how to do something, you're still useless as a manager unless you can communicate your ideas to everyone involved. And still yet, you've got to be able to develop an organizational infrastructure to support it -- which is NOT the same thing as just proposing a vague idea.
If you are afraid that using a very unrestrictive liscense will result in people modifying and distributing versions that you don't like, just PGP sign the "official version." That way, you can allow people to modify and distribute versions of their own, only they won't be directly associated with you since they're not signed by you.
As for the liscense you pick, thats up to you. Just don't be afraid to choose an extremely liberal liscense out of fear that you will completely loose control of the document. If your version is better than everyone else's, people will make sure to get the version you PGP-signed and not another. If a competing version eventually becomes better or more popular, so be it. That's the whole point anyway, right?
So, we know by now that Crusoe only requires around 1 watt of power to operate, and that this results in a maximum temperature of 48C (thus, a fan isn't required to cool the thing). But, if you don't really care about power consumption and you installed a fan over the heatsink, one has to wonder how much faster these things can be clocked before they start showing glitches. The only problem I can see is the LongRun software which will automatically reduce power consumption if it's not necessarily needed -- this might mean that the only way to overclock the chip would be to modify the LongRun code (stored in FlashROM).
Any guesses as to how long until someone figures out how to patch the FlashROM so to allow overclocking? I give it about 6 months after Crusoe based systems hit the shelves.
I suppose I'm inviting a horde of flames for saying this, but if Via's liscense agreement for the P6 bus specifically prohibits Via from running it at speeds faster than 100Mhz, then Intel is certainly within its rights to revoke the liscense. Of course, having not seen the liscense, I don't know for sure whether such a clause exists... But if so, Intel has a responsibility to its shareholders to protect market share and ought to pursue this lawsuit... unless their lawyers believe that the liscense agreement will be viewed as anti-competitive in nature and will get them back in hot water with the FTC. Of course, whether or not the liscense agreement will hold up in court is a different matter entirely.
Business is about money, not innovation; remember that.
AFAIK, Oracle8i's approach is to abandon the idea of running a database application on top of a "general purpose operating system" since, typically, a database server only runs one application -- the database itself. So, a Oracle8i can be thought of as a "specific purpose OS" optimized for database queries with a database app on top of it. Much faster.
Can someone who knows something about the GNU Hurd kernel comment on how it manages network connections? I remember reading something about how just about everything can be managed by a user-level process in Hurd -- does this include the NIC?
There are compilers available for linux (TowerJ being one) but their primary benifit is for server-side code; it'd be much more difficult (but not entirely impossible -- proof-carrying-code would work) to ensure safety if you distribute binaries to clients. Indeed, the whole point of using platform independent byte-codes is so that the JVM can ensure saftey. Platform-specific machine code running on a server will probably coexist with platform-independent java byte-codes for client applications.
Umm, don't people realize that the "networked toaster" idea was just a joke? Someone was just playing around with the concept of "internet appliances" and it wasn't really meant to be taken seriously. Even the (previously somewhat common) internet coke machines and coffee makers were intended only as cheesy hacks... Does anyone even maintain those things anymore?
Methinks that the misunderstanding here involves confusing publishing with merely posting, as it applies on the web. If someone sent a letter to the editor of a print publication using the phrase "..For Dummies", the editor would probably insert the necessary '(tm)' or whatnot. Clearly, a mailing list moderator or a webmaster of a site like slashdot can't easily do this....
Considering that such a high percentage of us (relative to most web sites anyway) run Linux, I think the management should avoid posting links to videos using the new (Sorensen?) Quicktime 4.0 codec.
How do you think superscalar processors work? They analyze dependancies between instructions -- if two (or more) instructions are independent (one instruction doesn't depend on the result of another) then they can be executed simultaneously.
BTW, you should read about things like partial evaluation and specify to see what's meant by code optimized for a given application. For instance, suppose I have a function:
foo(a,b) = a + b;
If I specify that the param 'a' always equals 0:
foo2 = foo(0,_);
then foo2(b) = b. This is all that specification does. Clearly, the compiled code for foo2() is far simpler than that of foo(). (IOW, calling foo2(x) runs faster than calling foo(0,x)).
Modern functional languages do this stuff pretty well, and there's been some research on applications in operating systems (for instance, optimizing code paths for specific devices).
BTW, when people are discussing compilers, you should realize that the word "optimizing" is meant as a funny joke, nothing to take seriously. An "optimizing compiler" would be more accurately called a "compiler which usually generates better code than a naive approach," but for marketing reasons, the former sounds better.
In general, defacing a web site should be considered MUCH worse than mearly spray-painting a wall. Consider for a minute how much money a popular site like CNN.com could lose if their Ad banners weren't displayed... Cracking a website is more akin to editing an episode of "Friends" before it airs on television so that Chandler appears to say "Suck my #$^@ FBI!" instead of his normal one liners.
I say give him a $1000 fine and a crapload of community service. If he's as stupid as Mitnick and keeps doing this shit, give him a year in prison (along with fining him for any damages). And beat his ass with a cane on national television while you're at it.
Huh!?!?!?!?!? There are LOTS of incredible role models for aspiring computer scientists. Linux is cool, but it's definitely NOT groundbreaking in any technical sense. It's mostly a rehash of older *nix systems. In fact, the only really novel thing about Linux is it's development model (open source) which has more to do with software engineering, management, and perhaps social theory than with computer science.
If you want to find some REAL computer science role models, I suggest you start by looking in the back of your text books for widely and recently (since 1960 or so) referenced computer scientists/mathematicians. Then track down that authors web page at whatever university s/he is at and take a gander at their "Family Tree" (of doctoral students). And while you're at it, download a few of their papers (if available) or at least find a few of their tech-reports in your library. If you want to see some impressive work, this the the only way to go. You may be able to find some papers written by people in industry, but so much of it tends to be proprietary. Fortunately, this seems to be changing (IBM research and Bell Labs/Lucent publish a lot of their material online also).
The alpha hack^H^H^H^Hport was done after the processor was already available. Seeing as Intel's implementation of IA-64 isn't widely known (it requires an NDA for the info) this can't be a project that a few hackers work on in their spare time. Besides, IA-64 is a MUCH different architecture than the Alpha. I'd be much more interested to hear about the progress of compilers (which seems to be the most difficult part).
Incidentally, does anyone know if gcc will be able to easily compile on EPIC architectures? This seems like a pretty non-trivial change...
Slashdot Mirror
The point of using 'patterns' is so that people who design systems for a living can speak the same lingo during design, independant of the language being used for implementation. Same with UML. Their complexity characteristics are already well understood and the metaphors being used ("observer," "factory", etc) are also high-level enough so that the overall design of a large component can be held-in-the-head all at once. (you know, that 7+/-2 thing?).
If you design large projects using an object oriented approach (which doesn't exclude using C or even Perl), I'd think that you'd find patterns essential. And yes, people had been doing OOD without using "patterns" as such for quite a while; but more often than not, the same general ideas were still being used, except that any metaphors used to explain them were basically ad-hoc.
-NooM
Lots of Iritating, Superfluous Parenthesis!
QT and GTK both lack good software for GUI building and prototyping. When you're trying out different looks for an application, it's REALLY annoying to have write code to have to specify the look'n'feel of an app when it makes much more sense to use your visual intuition to 'draw' the interface.
I'd go ahead and write one, but I don't want to.
-NooM
I completely disagreed with the original poster's view -- I think it's a fairly deluded view of things. But it's still interesting. People read Descartes over and over again, it's taught to students, and yet due to the numerous flaws in his arguments there are probably no real Cartesians left in this world. But again, especially in the context of his times, it's still interesting. And, just as much as Hitler and Lenin, Descartes sought to tell people exactly what they wanted to hear (that they have an immortal soul, that God exists, that the Skeptics were all wrong). So yes, I agree with you that targeted rhetoric like that can be extremely manipulative. But, if you are the type of person who is inclined (provoked) to think about it, you don't allow yourself to be manipulated so easily. Hiding your head in the sand like an ostrich doesn't really help. You're not really saying that people shouldn't challenge convention, right? If so, we'd still believe the world was flat, that the universe revolves around the earth, etc...
And incidentally, I enjoy Tolkien just as much as the next person -- if you are good at spouting jibberish about magic and little gnomes, go ahead and write to your heart's content (but probably not in this thread, for obvious reasons).
-NooM
This is a forum for conversation, not a magazine or journal. The post is highly rated not because it's necessarily true or even historically accurate but because it's thought provoking -- much more so than the average Slashdot post. If you find that you are provoked to disagree with the content of the article, why not point out where it goes wrong?
-NooM
The reason for security problems with NFS on Linux is because NFS is inherently an insecure protocol for file sharing. Add-ons to make NFS secure, like using Kerberos for authentication, are NOT part of the NFS standard. In general, you can always spoof UIDs to get around file perms on any standard NFS implementation.
Oh, and the main reason to use Samba rather than an alternative NFS or Coda is because files are typically served to Windows clients, which automatically support SMB.
-NooM
This statement definitely goes along with somoene else's observation that a company's technical staff doesn't get enough respect to be promoted to managerial positions. Probably the MAIN reason people don't reason with their managers nor get promoted to management positions (even if they really do know what's up) is because so many of them are incapable of communicating effectively. Even if you know how to do something, you're still useless as a manager unless you can communicate your ideas to everyone involved. And still yet, you've got to be able to develop an organizational infrastructure to support it -- which is NOT the same thing as just proposing a vague idea.
-NooM
If you are afraid that using a very unrestrictive liscense will result in people modifying and distributing versions that you don't like, just PGP sign the "official version." That way, you can allow people to modify and distribute versions of their own, only they won't be directly associated with you since they're not signed by you.
As for the liscense you pick, thats up to you. Just don't be afraid to choose an extremely liberal liscense out of fear that you will completely loose control of the document. If your version is better than everyone else's, people will make sure to get the version you PGP-signed and not another. If a competing version eventually becomes better or more popular, so be it. That's the whole point anyway, right?
-Noom
So, we know by now that Crusoe only requires around 1 watt of power to operate, and that this results in a maximum temperature of 48C (thus, a fan isn't required to cool the thing). But, if you don't really care about power consumption and you installed a fan over the heatsink, one has to wonder how much faster these things can be clocked before they start showing glitches. The only problem I can see is the LongRun software which will automatically reduce power consumption if it's not necessarily needed -- this might mean that the only way to overclock the chip would be to modify the LongRun code (stored in FlashROM).
Any guesses as to how long until someone figures out how to patch the FlashROM so to allow overclocking? I give it about 6 months after Crusoe based systems hit the shelves.
-NooM
I suppose I'm inviting a horde of flames for saying this, but if Via's liscense agreement for the P6 bus specifically prohibits Via from running it at speeds faster than 100Mhz, then Intel is certainly within its rights to revoke the liscense. Of course, having not seen the liscense, I don't know for sure whether such a clause exists... But if so, Intel has a responsibility to its shareholders to protect market share and ought to pursue this lawsuit... unless their lawyers believe that the liscense agreement will be viewed as anti-competitive in nature and will get them back in hot water with the FTC. Of course, whether or not the liscense agreement will hold up in court is a different matter entirely.
Business is about money, not innovation; remember that.
-NooM
AFAIK, Oracle8i's approach is to abandon the idea of running a database application on top of a "general purpose operating system" since, typically, a database server only runs one application -- the database itself. So, a Oracle8i can be thought of as a "specific purpose OS" optimized for database queries with a database app on top of it. Much faster.
Can someone who knows something about the GNU Hurd kernel comment on how it manages network connections? I remember reading something about how just about everything can be managed by a user-level process in Hurd -- does this include the NIC?
-NooM
There are compilers available for linux (TowerJ being one) but their primary benifit is for server-side code; it'd be much more difficult (but not entirely impossible -- proof-carrying-code would work) to ensure safety if you distribute binaries to clients. Indeed, the whole point of using platform independent byte-codes is so that the JVM can ensure saftey. Platform-specific machine code running on a server will probably coexist with platform-independent java byte-codes for client applications.
-NooM
Umm, don't people realize that the "networked toaster" idea was just a joke? Someone was just playing around with the concept of "internet appliances" and it wasn't really meant to be taken seriously. Even the (previously somewhat common) internet coke machines and coffee makers were intended only as cheesy hacks... Does anyone even maintain those things anymore?
You forgot to critisize the acting -- Keanu Reaves really sucks. Although I agree with you on all other counts.
And yet, somehow, I still enjoyed the movie a lot. Go figure.
Hmmm... Gold, Gould.... any connection here?
Methinks that the misunderstanding here involves confusing publishing with merely posting, as it applies on the web. If someone sent a letter to the editor of a print publication using the phrase "..For Dummies", the editor would probably insert the necessary '(tm)' or whatnot. Clearly, a mailing list moderator or a webmaster of a site like slashdot can't easily do this....
... Unless(tm) You(tm) Write(tm) A(tm) Simple(tm) Perl(tm) Script(tm) That(tm) Adds(tm) These(tm) Marks(tm) Everywhere(tm).
(on your end, you'd want to do the reverse translation)
Considering that such a high percentage of us (relative to most web sites anyway) run Linux, I think the management should avoid posting links to videos using the new (Sorensen?) Quicktime 4.0 codec.
Perhaps a boycott is in order?
I think Cantor might disagree with that assesment [:>).
... but, if you could do it, the size of 'compressed' kernels would definitely drop a lot! Not that it would matter, given the transmission rate...
Wouldn't Firewire be considered a "cheaper-than-scsi yet very fast transfer method?"
How do you think superscalar processors work? They analyze dependancies between instructions -- if two (or more) instructions are independent (one instruction doesn't depend on the result of another) then they can be executed simultaneously.
BTW, you should read about things like partial evaluation and specify to see what's meant by code optimized for a given application. For instance, suppose I have a function:
foo(a,b) = a + b;
If I specify that the param 'a' always equals 0:
foo2 = foo(0,_);
then foo2(b) = b. This is all that specification does. Clearly, the compiled code for foo2() is far simpler than that of foo(). (IOW, calling foo2(x) runs faster than calling foo(0,x)).
Modern functional languages do this stuff pretty well, and there's been some research on applications in operating systems (for instance, optimizing code paths for specific devices).
BTW, when people are discussing compilers, you should realize that the word "optimizing" is meant as a funny joke, nothing to take seriously. An "optimizing compiler" would be more accurately called a "compiler which usually generates better code than a naive approach," but for marketing reasons, the former sounds better.
-NooM
In general, defacing a web site should be considered MUCH worse than mearly spray-painting a wall. Consider for a minute how much money a popular site like CNN.com could lose if their Ad banners weren't displayed... Cracking a website is more akin to editing an episode of "Friends" before it airs on television so that Chandler appears to say "Suck my #$^@ FBI!" instead of his normal one liners.
I say give him a $1000 fine and a crapload of community service. If he's as stupid as Mitnick and keeps doing this shit, give him a year in prison (along with fining him for any damages). And beat his ass with a cane on national television while you're at it.
-NooM
Huh!?!?!?!?!? There are LOTS of incredible role models for aspiring computer scientists. Linux is cool, but it's definitely NOT groundbreaking in any technical sense. It's mostly a rehash of older *nix systems. In fact, the only really novel thing about Linux is it's development model (open source) which has more to do with software engineering, management, and perhaps social theory than with computer science.
If you want to find some REAL computer science role models, I suggest you start by looking in the back of your text books for widely and recently (since 1960 or so) referenced computer scientists/mathematicians. Then track down that authors web page at whatever university s/he is at and take a gander at their "Family Tree" (of doctoral students). And while you're at it, download a few of their papers (if available) or at least find a few of their tech-reports in your library. If you want to see some impressive work, this the the only way to go. You may be able to find some papers written by people in industry, but so much of it tends to be proprietary. Fortunately, this seems to be changing (IBM research and Bell Labs/Lucent publish a lot of their material online also).
-NooM
Methinks this was probably a typo or a severe misunderstanding. If outsourcers have a community, I'd be pretty surprised.
-NooM
The alpha hack^H^H^H^Hport was done after the processor was already available. Seeing as Intel's implementation of IA-64 isn't widely known (it requires an NDA for the info) this can't be a project that a few hackers work on in their spare time.
Besides, IA-64 is a MUCH different architecture than the Alpha. I'd be much more interested to hear about the progress of compilers (which seems to be the most difficult part).
Incidentally, does anyone know if gcc will be able to easily compile on EPIC architectures? This seems like a pretty non-trivial change...
-NooM