My grocery store has one aisle that is entirely cookies/crackers on one side and pudding cups/candy/etc on the other side. Another aisle is entirely soda pop.
So, yeah, bascially my market has _two_ Twinkie aisles.
There are different ways of being overweight. You know those guys with the hard round gut? That's bad. You know those soft flabby guys (but not too flabby), well that's not quite as bad (IIRC).
What really bothers me is I'm starting to see teenagers who have the physique of a 45 year old man with the stereotypical beer gut. That's not just bad, it's really really sad. Their parents should know better.
That's a slightly loaded question, don't you think?;-)
Yes, but I think it is a valid one. At this stage of software technology, I just don't think there are _any_ good solutions for writing large, portable, comprehensible, etc. programs. Even the beauty of Lisp comes with baggage that keeps it from being used more widely. C++ is no panacea, either.
So wouldn't it make any difference if, say, Mozilla was in assembly?
No one really considers assembly an applications programming language, any more, except perhaps HPC/graphics developers.
Of the four you mentioned, three (OO.org, KDE, mozilla) are in C++. GNOME is in C...
Yes, but they are all difficult to work with. That's just a fact of all projects that large, IMO. If it isn't a problem with C kludges, then its a problem with choice of abstraction or a problem with 100 shared libraries or a problem with poor performance or a problem with compiler interoperability, etc. C++ doesn't make these problems go away.
With well-organized header files and a little documentation, I've had no problems with C up to 100K lines. The main advantage, IMO, is that C is just so damn simple to "execute" in my mind, and it is very intuitive to step through in a debugger. The only real catch for me is tracking down memory leaks at times.
Now, for programs with millions of LOC, I don't think any language will be easier than another, just because the program's own complexity overwhelms everything else. Just ask people how quickly they can get up to speed on OO.org, KDE, Mozilla, or GNOME (they all suck for casual developers).
Here's a question for everyone: Which is better? An expertly designed and elegant C++ program with all the proper abstractions, frameworks, and methodologies that 1% of software developers can comprehend quickly, or a C program with a flat set of files, simple structs, and a few shell scripts to build it, which everyone understands (and can even debug)?
Trying to answering this question tears me apart, because I so much like the _idea_ of elegant OO software design. However, I generally find that I can work my way through a C program much more quickly than a comparable-complexity C++ program. Perhaps it's just me.
Uh oh, what about when I say the CPUs in a lot of expensive computers don't even have fans on the CPU heatsinks in addition to no disk coolers? (they just use the case fans for everything)
That's probably the foolish bullshit that adults are telling kids these days. The baby boom is so obvious to everyone that everyone will think it's the next big thing. It'll be like the internet boom all over again, with all the consequences.
For example, if I find out about a 'hot' stock, I pretty much understand that I'm the last person on earth who would really have any inside information. I ignore all 'hot' stocks. Why is it that people are so eager to _not_ ignore the 'hot' job professions? They both suffer from the same problems of quick ramps followed by crushing defeat.
Were you using gigabit networking? There are a few articles now about how NFS from a good server over a good network, for example, can be as fast or faster than some local filesystems.
If a 22-year-old with a CS degree is qualified to architect major projects all I can say is run hard and fast to get another degree
You'd be suprised how often a 22y.o. is stuck doing just that. I worked on one project where no one on the project had more than two years work experience (I had recently gradutated). It was hell.
The saddest thing was seeing something on the cover of Java Pro one week, and have the "project lead" talking about it the following week. The whole team was clueless.
In engineering, the higher paying jobs seem to go to EE majors working on difficult integrated circuits. At least that's what the job ads indicate. Programming jobs earn 1/2 to 2/3 as much.
I agree about forgetting grad school. Grad school is for people who really care, not for people who just want another line on their resume.
You know, even $35,000 UNIX workstations don't have "disk coolers". They just mount the drives with air space all around them (about 1/2 inch top and bottom) and allow the regular case fan to do all the cooling work. This is adequate even for super-fast 10KRPM and 15KRPM SCSI drives.
The super-expensive RAID arrays do the same. They just allow space around the drives and have regular fans pull air through the case.
Sun is definitely a bargain, now, but they have to overcome the baggage of having the UltraSPARC II stuck at 480MHz while the UltraSPARC III was being delayed. That is the source of most of the "but my Pentium is five times faster for 1/20 the cost" trolls. Fortunately, that is _not_ a problem, anymore.
you can love your offspring while still denying them things, despite what your idiot neighbor claims.
Denying children from being overwhelmed by abundance _is_ the responsibility of loving parents. It's the only way for children to develop a perspective on how the real world operates (most people don't get things simply by pouting about it--they have to work for it).
You should be glad you posted AC, because your ignorance is embarrasing. SMF parallelizes the boot process, enforces dependencies among services, restarts crashed services automatically, and gives a really efficient CLI for management. It is everything the old init system isn't.
if people can't get a hold of it and can't figure out how to use the features, why bother
Interesting, considering Solaris is one of the best documented systems I've ever used. http://docs.sun.com is light years ahead of the available Linux documentation.
Sun definitely has made some big mistakes. However, they admit to them. I saw a Sun news video with Schwartz doing a presentation to Wall Street. The whole presentation was admitting that Sun had nothing to sell them a few years ago, but that with Solaris 10, OpenSolaris, Opteron servers, etc. etc. etc. that Sun's 'shelves are now fully stocked' for Wall Street. Another presentation by a Sun rep had a whole slide of Sun missteps and how their new strategies address them.
As far as Java revenue goes, I wonder how much money they get off of those 750 million or so cell phones they license J2ME for.
Actually, the Java thing brings up another point: a lot of Sun's business exists outside of the visible desktop PC universe. Slashdotters might wonder what Sun is doing, but consider those 750 million cell phones. That's a _huge_ user base (bigger than all PC users in the world).
The features in Linux do not compare to those in Solaris. The reason: the ones in Solaris work out of the box with minimal system administration. Dtrace just works. Containers just need to be activated. There is no kernel configuring or re-compiling or patching or finger crossing.
Any sysadmin worth anything wouldn't put an important system on a patched re-compiled kernel. If it doesn't already come pre-configured from Red Hat or SuSE, it's worthless. If they do come preconfigured from Red Hat and SuSE, then mark me corrected.
I don't know...I kinda like having a nice system that is based on open file formats and protocols. Just in the past year or two, I think, has the notion of a "Linux desktop" ("GNOME/OO.org/Mozilla desktop" is more accurate, IMO) matured to a point to really being a true Microsoft replacement. The only times I boot Windows are to view a retarded website (not often), test a new web page in IE (the responsible thing to do), or to play a dinky game.
Slashdot Mirror
My grocery store has one aisle that is entirely cookies/crackers on one side and pudding cups/candy/etc on the other side. Another aisle is entirely soda pop.
So, yeah, bascially my market has _two_ Twinkie aisles.
There are different ways of being overweight. You know those guys with the hard round gut? That's bad. You know those soft flabby guys (but not too flabby), well that's not quite as bad (IIRC).
What really bothers me is I'm starting to see teenagers who have the physique of a 45 year old man with the stereotypical beer gut. That's not just bad, it's really really sad. Their parents should know better.
That's a slightly loaded question, don't you think? ;-)
Yes, but I think it is a valid one. At this stage of software technology, I just don't think there are _any_ good solutions for writing large, portable, comprehensible, etc. programs. Even the beauty of Lisp comes with baggage that keeps it from being used more widely. C++ is no panacea, either.
So wouldn't it make any difference if, say, Mozilla was in assembly?
No one really considers assembly an applications programming language, any more, except perhaps HPC/graphics developers.
Of the four you mentioned, three (OO.org, KDE, mozilla) are in C++. GNOME is in C...
Yes, but they are all difficult to work with. That's just a fact of all projects that large, IMO. If it isn't a problem with C kludges, then its a problem with choice of abstraction or a problem with 100 shared libraries or a problem with poor performance or a problem with compiler interoperability, etc. C++ doesn't make these problems go away.
With well-organized header files and a little documentation, I've had no problems with C up to 100K lines. The main advantage, IMO, is that C is just so damn simple to "execute" in my mind, and it is very intuitive to step through in a debugger. The only real catch for me is tracking down memory leaks at times.
Now, for programs with millions of LOC, I don't think any language will be easier than another, just because the program's own complexity overwhelms everything else. Just ask people how quickly they can get up to speed on OO.org, KDE, Mozilla, or GNOME (they all suck for casual developers).
Here's a question for everyone: Which is better? An expertly designed and elegant C++ program with all the proper abstractions, frameworks, and methodologies that 1% of software developers can comprehend quickly, or a C program with a flat set of files, simple structs, and a few shell scripts to build it, which everyone understands (and can even debug)?
Trying to answering this question tears me apart, because I so much like the _idea_ of elegant OO software design. However, I generally find that I can work my way through a C program much more quickly than a comparable-complexity C++ program.
Perhaps it's just me.
Uh oh, what about when I say the CPUs in a lot of expensive computers don't even have fans on the CPU heatsinks in addition to no disk coolers? (they just use the case fans for everything)
That's probably the foolish bullshit that adults are telling kids these days. The baby boom is so obvious to everyone that everyone will think it's the next big thing. It'll be like the internet boom all over again, with all the consequences.
For example, if I find out about a 'hot' stock, I pretty much understand that I'm the last person on earth who would really have any inside information. I ignore all 'hot' stocks. Why is it that people are so eager to _not_ ignore the 'hot' job professions? They both suffer from the same problems of quick ramps followed by crushing defeat.
Were you using gigabit networking? There are a few articles now about how NFS from a good server over a good network, for example, can be as fast or faster than some local filesystems.
If a 22-year-old with a CS degree is qualified to architect major projects all I can say is run hard and fast to get another degree
You'd be suprised how often a 22y.o. is stuck doing just that. I worked on one project where no one on the project had more than two years work experience (I had recently gradutated). It was hell.
The saddest thing was seeing something on the cover of Java Pro one week, and have the "project lead" talking about it the following week. The whole team was clueless.
In engineering, the higher paying jobs seem to go to EE majors working on difficult integrated circuits. At least that's what the job ads indicate. Programming jobs earn 1/2 to 2/3 as much.
I agree about forgetting grad school. Grad school is for people who really care, not for people who just want another line on their resume.
You know, even $35,000 UNIX workstations don't have "disk coolers". They just mount the drives with air space all around them (about 1/2 inch top and bottom) and allow the regular case fan to do all the cooling work. This is adequate even for super-fast 10KRPM and 15KRPM SCSI drives.
The super-expensive RAID arrays do the same. They just allow space around the drives and have regular fans pull air through the case.
It is fairly common knowledge (I hope!) that complexity increases exponentially with LOC for software in general. UNIX/Linux kernels are no exception.
SunRay software web page.
It looks like it's currently Solaris-SPARC and Linux-x86 as the supported configurations (yes, Slashdot, Sun _does_ sell Linux thin clients).
people will soon see that Sun is a bargain.
Sun is definitely a bargain, now, but they have to overcome the baggage of having the UltraSPARC II stuck at 480MHz while the UltraSPARC III was being delayed. That is the source of most of the "but my Pentium is five times faster for 1/20 the cost" trolls. Fortunately, that is _not_ a problem, anymore.
you can love your offspring while still denying them things, despite what your idiot neighbor claims.
Denying children from being overwhelmed by abundance _is_ the responsibility of loving parents. It's the only way for children to develop a perspective on how the real world operates (most people don't get things simply by pouting about it--they have to work for it).
Crashme is obselete. It has been replaced by Slashdot.
You should be glad you posted AC, because your ignorance is embarrasing. SMF parallelizes the boot process, enforces dependencies among services, restarts crashed services automatically, and gives a really efficient CLI for management. It is everything the old init system isn't.
if people can't get a hold of it and can't figure out how to use the features, why bother
Interesting, considering Solaris is one of the best documented systems I've ever used. http://docs.sun.com is light years ahead of the available Linux documentation.
Sun definitely has made some big mistakes. However, they admit to them. I saw a Sun news video with Schwartz doing a presentation to Wall Street. The whole presentation was admitting that Sun had nothing to sell them a few years ago, but that with Solaris 10, OpenSolaris, Opteron servers, etc. etc. etc. that Sun's 'shelves are now fully stocked' for Wall Street. Another presentation by a Sun rep had a whole slide of Sun missteps and how their new strategies address them.
As far as Java revenue goes, I wonder how much money they get off of those 750 million or so cell phones they license J2ME for.
Actually, the Java thing brings up another point: a lot of Sun's business exists outside of the visible desktop PC universe. Slashdotters might wonder what Sun is doing, but consider those 750 million cell phones. That's a _huge_ user base (bigger than all PC users in the world).
The features in Linux do not compare to those in Solaris. The reason: the ones in Solaris work out of the box with minimal system administration. Dtrace just works. Containers just need to be activated. There is no kernel configuring or re-compiling or patching or finger crossing.
Any sysadmin worth anything wouldn't put an important system on a patched re-compiled kernel. If it doesn't already come pre-configured from Red Hat or SuSE, it's worthless. If they do come preconfigured from Red Hat and SuSE, then mark me corrected.
For the "but the V480/V490 can't be ordered for $999 from Dell" trolls, there's also the v40z Opteron server that now sells with 4 dual-core CPUs.
However, for supporting a 100 desktops, something as robust as the V490 might be a good thing.
"More often than not, adding alot of man power to a project will only make it alot worse."
This is because the complexity of human networks scale exponentially. This has been documented for decades (Fred Brooks, e.g.).
Let's add Evolution to the mix for kicks...that makes GNOME/OO.org/Mozilla/Evolution...whose acronym is GNOME!
Obviously this is a sign from the Gods Of Recursive Elegance (GORE? He's in on it too?) that we're on the right track.
Pardon my french but F*CK LINUX ON THE DESKTOP.
I don't know...I kinda like having a nice system that is based on open file formats and protocols. Just in the past year or two, I think, has the notion of a "Linux desktop" ("GNOME/OO.org/Mozilla desktop" is more accurate, IMO) matured to a point to really being a true Microsoft replacement. The only times I boot Windows are to view a retarded website (not often), test a new web page in IE (the responsible thing to do), or to play a dinky game.