Exactly. "Thin client" apps are a pretty accurate analog to the ubiquitous CICS business apps of the 1970's that are still floating around today. Fill out a screen. Click a button. Wait for another screen. Lather, rinse, repeat.
Some business apps do very well in such a framework. Others do not. Spreadsheets, for example, don't (there have been "networked" spreadsheet apps for years now, but what is their penetration? Zilch). Yes, sigh, we can code the entire app in Javascript and hide transfers in the background, and "simulate" a desktop app blah blah blah Ajax blah blah -- but that's not workable (I invite you to try; have fun).
Good God, just got curious and queried Wikipedia: "IBM began shipping the latest release, CICS Transaction Server Version 3.1 for z/OS, in early 2005." Somebody pinch me, I'm in a time warp!
"...all use 2 significant figures for their data, yet quote the answer around 10 significant figures!"
Which is why slide rules ought to be used in introductory science classes, not calculators. Slide rule use does two things: 1) enforces a basic understanding of significant figures, and 2) creates an ability to quickly calculate order of magnitude and therefore to quickly dismiss solutions that cannot possibly be correct.
However, I suspect that the parent posters know all of the above, and that this is just "cut and paste from the calculator" laziness. We're all guilty of it.
SCO are not irrelevant, not yet. They need to be stamped into the ground with a boot heel, every ounce of life ground out of them, every molecule disassociated. Next, their principals need to be sued into oblivion, and their demonic attorneys censured for their unbelievably atrocious behavior. A message needs to be sent to IP trolls and their minions everywhere.
Even though we've centered the SCO trolls in the gun sights, there's still plenty of time to enjoy watching them try to slither away before their component atoms are blasted back to the alternate universe they came from. The longer and more painful this process is for them, the better. Where's the popcorn? Bring on the show.
Mr. Hirai's company is not selling web site content services, they're selling "small business web site construction for dummies," and they claim their tools have built an impressive 170,000 of them. "Citymax" is only one of 5 or 6 of the brands they control -- check out their "master" website, which is mezine.html. I only saw one typo on the mezine site, and it's a missing space.
What's your beef with "$20 bucks?" Sure, it's odd, but heck, they're Canadians, their heads are split in two anyway. Cut 'em a break.
Finally, he says he's grown the thing from zero to $25M since 1999. And if we believe what he says about how happy everyone is, we should listen very carefully to him. Very carefully indeed.
Re:The failure of the PS/2 killed OS/2
on
In Search of Stupidity
·
· Score: 2, Interesting
I totally agree that the PS/2 killed IBM's hardware business. Before the PS/2, IT departments buying PC's said, "Why not IBM?" After the PS/2, it was "Why IBM?" IBM pissed everyone off, including me. I probably told hundreds of clients not to buy PS/2's, and everyone I know who was in a position to recommend PC hardware did the same. Everyone flushed the toilet at once. It was amazing to watch.
However, this has nothing to do with the software business. The point is, and this is made by other commentators later in this thread, none of the clone manufacturers would source an IBM operating system if they could avoid it. Buying an OS from your competitor? What idiot would do that?
The clones were climbing up IBM's ass at the time, and would have eventually won on price/performance anyway. So OS/2 was doomed, since the clones would have won huge market share, and the clones would have avoided OS/2 like the plague.
I've been on the parent's site, and on a few of its child companies. Lots of big numbers on the books and vaporous revenue. However, their electric motor company site does have an interesting flash presentation on the technology, plus what looks like a dribble of real income, and perhaps even a real application in which the product has actually worked (one must sign an NDA to see a video of it in action). Unfortunately, I don't have the background to evaluate it.
Re:Does any major site use pure CSS?
on
CSS Cookbook
·
· Score: 1
Thank you for writing the comment I would have written. Probably better than I'd have written it, too.
You write well and it is enjoyable to read your posts.
However, in a Libertarian world, contrary to Libertarian claims, there is little mercy for the unfortunate. India is a perfect example of this, where one sees desperately poor people living in shanty towns next to glittering high tech campuses.
British professor Alexander Fraser Tyler wrote:
"A democracy cannot exist as a permanent form of government. It can exist only until the voters discover they can vote themselves largesse (defined as a liberal gift) out of the public treasury. From that moment on, the majority always votes for the candidate promising the most benefits from the public treasury, with the result that democracy always collapses over a loose fiscal policy, always to be followed by a dictatorship."
I don't agree with "always," but it is certainly inevitable that there will be very serious social unrest in India, if not now, soon enough.
This is not the age of the British Raj, or the age of the individual balkanized Raja-states, when social uprisings were brutally crushed. Therefore it seems to me that the situation is not sustainable, and that the current freewheeling capitalism that the upper social strata has exploited will inevitably be disrupted.
I know how to read the original thread. The point is, I don't have time to read threads I don't care about, so asking me to moderate on them is pointless. I can, however, moderate on what I've already read, or am reading currently.
OK, I have mod points at the moment but am posting instead... This is why meta-moderation exists and why it is important. Meta-mod frequently and the bad moderators won't get points as often.
If I could meta-moderate ON A SPECIFIC ARTICLE, then I'd agree. But instead I get a random sampling of shit that I'm not qualified to meta-moderate on, because I didn't read the thread originally. In order to do a good job meta-moderating, you have to be into the thread in a big way. How can you decide whether something is "insightful" or not if you haven't read the thread? Or the original article?
It should be possible to penalize idiot moderators on the spot. Oh, I'm sure someone will say, "well, then trolls can meta-moderate blah blah blah." Well, then pass out meta-moderation points like moderation points. Currently the only way to control a moderation troll is to waste mod points to "undo" the damage. Far better to hammer him with meta-moderation badness. Nail the fucker to the wall.
It wasn't a simulation because you then couldn't make a neat demonstration and show it in real time to the numb-nuts military dweeb in charge of your DoD grant.
Or, if you work for the Media Lab, you couldn't have made a cute little video for the Discovery Channel to drool over.
It's all about the marketing, not the science. OF COURSE it would have made more sense to have simulated the behavior. And it could have been done at a fraction of the cost.
Sure, here's your moral
on
IT and Divorce?
·
· Score: 4, Insightful
Your wife is an asshole. That's the moral.
The way I look at it is this. Either you are flying signals ("I'm single! I'm looking! I'm available! Look at me!") or you are not. I'm going to bet that your wife was flying signals. If you look back at her behavior totally objectively, you'll probably see it. If you do, internalize this fact -- she was looking to get out.
The best advice I can offer you is from my own experience.
1) Don't hang around. She says she still loves you? Bullshit. She doesn't. She's not even THINKING about you. Move on. 2) Don't believe her if she comes back. She says she wants to try again? She says she wants to "Renew your vows" (that's a rich one)? Move on. 3) Don't let her set your agenda. Guilt trip about "staying with the kids while she goes away?" Sure, if you want to. If you don't, tell her to drop them off at your place, you'll deal with them.
And don't -- absolutely don't -- let yourself be angry, bitter, unhappy. This is not your fault. It's not about your job or how many hours you put in. It's not that you didn't "work hard enough" at the relationship. Baloney. I know plenty of people who have ridiculous schedules and see each other rarely, but they're doing just fine.
Remember: it's about her wanting to leave. It's not about you "failing" in some respect. Dry your eyes, forget your guilt, and find someone who wants to be with you. And if you notice her "flying signals," put her on the "temporary fling" list and start flying your own signals, fast.
I thought it was funny then, and I still think it's funny, although the image ought to be re-worked (it's a bit rough). It's not "belittling," it's very appropriate -- the Borg is a perfect analogy for Microsoft's anti-competitive maneuvering. And, like it or not, Gates is emblematic of Microsoft, so the image is absolutely perfect.
As far as "growing up" is concerned, I guess I'll "grow up" when the process dispatcher inside Microsoft Windows is as good as the process dispatcher inside the Compatible Time Sharing System (Corbato, et al,1962). Yes, you read that correctly: 1962. It is ridiculous that one can (essentially) hang a single-CPU copy of Windows XP with a spinning application. We solved that problem 55 (that would be FIFTY FIVE) years ago.
So, please keep the belittling image. Keep emphasizing the fact that we are living with crap technology because we're under the thumb of a monopoly that should have been broken up years ago.
Ah, but it is possible to deliver quality software in any language, even VB. It is all about the programmer, not the tools. I've seen absolutely wonderfully-written Assembler H code (in the CP/CMS kernel) that (to paraphrase Dr. McCoy) "a child could understand." Contrariwise, I've seen OO code that is totally incomprehensible and unmaintainable (to be fair, I've seen OO code that's wonderful, too).
1968. Dijkstra's letter was 1968. So much for looking it up and believing the first Google reference. OK, so I had been programming for -2 years in 1968. That counts, right?
Humor, I hope. I enjoy poking fun at sweeping generalizations.
As far as "structured programming" is concerned, I had been programming for 17 years when Dijkstra's "GOTO Considered Harmful" letter was published in the CACM in March 1987. It was an oversimplification then, and it's an oversimplification now. If I'm eight levels deep in crap, and I need to get the hell out cleanly, is my code more readable/maintainable with a zillion IF tests on an exit flag, or with a simple GOTO and an expressive comment?
And please, don't get all Tony Hoare on me about GOTOs ruining proofs of correctness. Come to think of it, that proof of correctness idea really had some practical legs, eh? Can't swing a dead cat around my company without hitting a project that owes its success to a proof of correctness.
Which pretty much means that all device drivers, almost any kind of firmware or assembly language code, OS code, networking code, etc., all suck -- by definition. Thanks for clearing that up.
I kinda figured that was true, 'cause when my computer blue screens and stuff I see a lot of hex numbers and weird crap like that and so that's probably machine language or something so that's why it probably sucked and died and stuff.
Right. The top-down specification fanatic (the opposite of the XP advocate) secretly believes that if his spec is tight enough, the code really will "magically write itself." He lives in a fantasy world, of course, because his spec will not survive its first user experience.
With XP, the pendulum has swung rather "extremely" in the other direction. Unfortunately, most customers don't know what they want, and don't know how to tell you, even if they had time to sit with you (which they typically don't). Showing them interim versions of complex software will scare the living crap out of them.
XP does well in a rapid prototyping environment and on simple applications (like most Web apps), where the feedback is immediate and the discussion isn't about deep concepts or functions, but rather about UI issues like "where would you like this button" or "should this be one screen or two screens."
If you let yourself go to hell, things seem to start breaking down at around 40-45. I've seen guys hit 45 and look 70. Sad. On the other hand, it doesn't take very much effort to stay in shape, and the guys that make some reasonable effort along those lines (food intake, exercise) seem to do just fine -- that is, if they can avoid serious illness, which is a curve ball that can take any of us down.
Slashdot Mirror
Exactly. "Thin client" apps are a pretty accurate analog to the ubiquitous CICS business apps of the 1970's that are still floating around today. Fill out a screen. Click a button. Wait for another screen. Lather, rinse, repeat.
Some business apps do very well in such a framework. Others do not. Spreadsheets, for example, don't (there have been "networked" spreadsheet apps for years now, but what is their penetration? Zilch). Yes, sigh, we can code the entire app in Javascript and hide transfers in the background, and "simulate" a desktop app blah blah blah Ajax blah blah -- but that's not workable (I invite you to try; have fun).
Good God, just got curious and queried Wikipedia: "IBM began shipping the latest release, CICS Transaction Server Version 3.1 for z/OS, in early 2005." Somebody pinch me, I'm in a time warp!
"...all use 2 significant figures for their data, yet quote the answer around 10 significant figures!"
Which is why slide rules ought to be used in introductory science classes, not calculators. Slide rule use does two things: 1) enforces a basic understanding of significant figures, and 2) creates an ability to quickly calculate order of magnitude and therefore to quickly dismiss solutions that cannot possibly be correct.
However, I suspect that the parent posters know all of the above, and that this is just "cut and paste from the calculator" laziness. We're all guilty of it.
Now THAT's funny!
SCO are not irrelevant, not yet. They need to be stamped into the ground with a boot heel, every ounce of life ground out of them, every molecule disassociated. Next, their principals need to be sued into oblivion, and their demonic attorneys censured for their unbelievably atrocious behavior. A message needs to be sent to IP trolls and their minions everywhere.
Even though we've centered the SCO trolls in the gun sights, there's still plenty of time to enjoy watching them try to slither away before their component atoms are blasted back to the alternate universe they came from. The longer and more painful this process is for them, the better. Where's the popcorn? Bring on the show.
Mr. Hirai's company is not selling web site content services, they're selling "small business web site construction for dummies," and they claim their tools have built an impressive 170,000 of them. "Citymax" is only one of 5 or 6 of the brands they control -- check out their "master" website, which is mezine.html. I only saw one typo on the mezine site, and it's a missing space.
What's your beef with "$20 bucks?" Sure, it's odd, but heck, they're Canadians, their heads are split in two anyway. Cut 'em a break.
Finally, he says he's grown the thing from zero to $25M since 1999. And if we believe what he says about how happy everyone is, we should listen very carefully to him. Very carefully indeed.
I totally agree that the PS/2 killed IBM's hardware business. Before the PS/2, IT departments buying PC's said, "Why not IBM?" After the PS/2, it was "Why IBM?" IBM pissed everyone off, including me. I probably told hundreds of clients not to buy PS/2's, and everyone I know who was in a position to recommend PC hardware did the same. Everyone flushed the toilet at once. It was amazing to watch.
However, this has nothing to do with the software business. The point is, and this is made by other commentators later in this thread, none of the clone manufacturers would source an IBM operating system if they could avoid it. Buying an OS from your competitor? What idiot would do that?
The clones were climbing up IBM's ass at the time, and would have eventually won on price/performance anyway. So OS/2 was doomed, since the clones would have won huge market share, and the clones would have avoided OS/2 like the plague.
I've been on the parent's site, and on a few of its child companies. Lots of big numbers on the books and vaporous revenue. However, their electric motor company site does have an interesting flash presentation on the technology, plus what looks like a dribble of real income, and perhaps even a real application in which the product has actually worked (one must sign an NDA to see a video of it in action). Unfortunately, I don't have the background to evaluate it.
Thank you for writing the comment I would have written. Probably better than I'd have written it, too.
Nice Larry Niven reference to the General Products hull.
You write well and it is enjoyable to read your posts.
However, in a Libertarian world, contrary to Libertarian claims, there is little mercy for the unfortunate. India is a perfect example of this, where one sees desperately poor people living in shanty towns next to glittering high tech campuses.
British professor Alexander Fraser Tyler wrote:
"A democracy cannot exist as a permanent form of government. It can exist only until the voters discover they can vote themselves largesse (defined as a liberal gift) out of the public treasury. From that moment on, the majority always votes for the candidate promising the most benefits from the public treasury, with the result that democracy always collapses over a loose fiscal policy, always to be followed by a dictatorship."
I don't agree with "always," but it is certainly inevitable that there will be very serious social unrest in India, if not now, soon enough.
This is not the age of the British Raj, or the age of the individual balkanized Raja-states, when social uprisings were brutally crushed. Therefore it seems to me that the situation is not sustainable, and that the current freewheeling capitalism that the upper social strata has exploited will inevitably be disrupted.
I know how to read the original thread. The point is, I don't have time to read threads I don't care about, so asking me to moderate on them is pointless. I can, however, moderate on what I've already read, or am reading currently.
OK, I have mod points at the moment but am posting instead... This is why meta-moderation exists and why it is important. Meta-mod frequently and the bad moderators won't get points as often.
If I could meta-moderate ON A SPECIFIC ARTICLE, then I'd agree. But instead I get a random sampling of shit that I'm not qualified to meta-moderate on, because I didn't read the thread originally. In order to do a good job meta-moderating, you have to be into the thread in a big way. How can you decide whether something is "insightful" or not if you haven't read the thread? Or the original article?
It should be possible to penalize idiot moderators on the spot. Oh, I'm sure someone will say, "well, then trolls can meta-moderate blah blah blah." Well, then pass out meta-moderation points like moderation points. Currently the only way to control a moderation troll is to waste mod points to "undo" the damage. Far better to hammer him with meta-moderation badness. Nail the fucker to the wall.
It wasn't a simulation because you then couldn't make a neat demonstration and show it in real time to the numb-nuts military dweeb in charge of your DoD grant.
Or, if you work for the Media Lab, you couldn't have made a cute little video for the Discovery Channel to drool over.
It's all about the marketing, not the science. OF COURSE it would have made more sense to have simulated the behavior. And it could have been done at a fraction of the cost.
Your wife is an asshole. That's the moral.
The way I look at it is this. Either you are flying signals ("I'm single! I'm looking! I'm available! Look at me!") or you are not. I'm going to bet that your wife was flying signals. If you look back at her behavior totally objectively, you'll probably see it. If you do, internalize this fact -- she was looking to get out.
The best advice I can offer you is from my own experience.
1) Don't hang around. She says she still loves you? Bullshit. She doesn't. She's not even THINKING about you. Move on.
2) Don't believe her if she comes back. She says she wants to try again? She says she wants to "Renew your vows" (that's a rich one)? Move on.
3) Don't let her set your agenda. Guilt trip about "staying with the kids while she goes away?" Sure, if you want to. If you don't, tell her to drop them off at your place, you'll deal with them.
And don't -- absolutely don't -- let yourself be angry, bitter, unhappy. This is not your fault. It's not about your job or how many hours you put in. It's not that you didn't "work hard enough" at the relationship. Baloney. I know plenty of people who have ridiculous schedules and see each other rarely, but they're doing just fine.
Remember: it's about her wanting to leave. It's not about you "failing" in some respect. Dry your eyes, forget your guilt, and find someone who wants to be with you. And if you notice her "flying signals," put her on the "temporary fling" list and start flying your own signals, fast.
I thought it was funny then, and I still think it's funny, although the image ought to be re-worked (it's a bit rough). It's not "belittling," it's very appropriate -- the Borg is a perfect analogy for Microsoft's anti-competitive maneuvering. And, like it or not, Gates is emblematic of Microsoft, so the image is absolutely perfect.
As far as "growing up" is concerned, I guess I'll "grow up" when the process dispatcher inside Microsoft Windows is as good as the process dispatcher inside the Compatible Time Sharing System (Corbato, et al,1962). Yes, you read that correctly: 1962. It is ridiculous that one can (essentially) hang a single-CPU copy of Windows XP with a spinning application. We solved that problem 55 (that would be FIFTY FIVE) years ago.
So, please keep the belittling image. Keep emphasizing the fact that we are living with crap technology because we're under the thumb of a monopoly that should have been broken up years ago.
God help us all.
Ah, but it is possible to deliver quality software in any language, even VB. It is all about the programmer, not the tools. I've seen absolutely wonderfully-written Assembler H code (in the CP/CMS kernel) that (to paraphrase Dr. McCoy) "a child could understand." Contrariwise, I've seen OO code that is totally incomprehensible and unmaintainable (to be fair, I've seen OO code that's wonderful, too).
1968. Dijkstra's letter was 1968. So much for looking it up and believing the first Google reference. OK, so I had been programming for -2 years in 1968. That counts, right?
...what have you added to the conversation?
Humor, I hope. I enjoy poking fun at sweeping generalizations.
As far as "structured programming" is concerned, I had been programming for 17 years when Dijkstra's "GOTO Considered Harmful" letter was published in the CACM in March 1987. It was an oversimplification then, and it's an oversimplification now. If I'm eight levels deep in crap, and I need to get the hell out cleanly, is my code more readable/maintainable with a zillion IF tests on an exit flag, or with a simple GOTO and an expressive comment?
And please, don't get all Tony Hoare on me about GOTOs ruining proofs of correctness. Come to think of it, that proof of correctness idea really had some practical legs, eh? Can't swing a dead cat around my company without hitting a project that owes its success to a proof of correctness.
Which pretty much means that all device drivers, almost any kind of firmware or assembly language code, OS code, networking code, etc., all suck -- by definition. Thanks for clearing that up.
I kinda figured that was true, 'cause when my computer blue screens and stuff I see a lot of hex numbers and weird crap like that and so that's probably machine language or something so that's why it probably sucked and died and stuff.
I completely agreed. I didn't even bother reading the text and was never going to (except for picture descriptions).
In that case, according to Slashdot dicta, you are supremely qualified to comment on the article!
Thanks, I almost believed parent. And I'm not even new here. No excuse.
Mod this guy up. Still chuckling.
Bzzzt! He got that wrong.
Right. The top-down specification fanatic (the opposite of the XP advocate) secretly believes that if his spec is tight enough, the code really will "magically write itself." He lives in a fantasy world, of course, because his spec will not survive its first user experience.
With XP, the pendulum has swung rather "extremely" in the other direction. Unfortunately, most customers don't know what they want, and don't know how to tell you, even if they had time to sit with you (which they typically don't). Showing them interim versions of complex software will scare the living crap out of them.
XP does well in a rapid prototyping environment and on simple applications (like most Web apps), where the feedback is immediate and the discussion isn't about deep concepts or functions, but rather about UI issues like "where would you like this button" or "should this be one screen or two screens."
If you let yourself go to hell, things seem to start breaking down at around 40-45. I've seen guys hit 45 and look 70. Sad. On the other hand, it doesn't take very much effort to stay in shape, and the guys that make some reasonable effort along those lines (food intake, exercise) seem to do just fine -- that is, if they can avoid serious illness, which is a curve ball that can take any of us down.